Literature review: Impact of Chilean needle grass ... - Weeds Australia

Literature review: 

Impact of Chilean needle grass Nassella 

neesiana on biodiversity in Australian 

indigenous grasslands 

Ian Faithfull 

1

CONTENTS 

Conventions and standards 5 

Abbreviations 5 

INTRODUCTION 6 

THEORETICAL FRAMEWORK 8 

Invasive potential of a species 8 

Enemy release and biotic resistance 10 

Resource-enrichment and fluctuating resources 11 

Propagule pressure 13 

Vacant niches and competitive exclusion 13 

‘Novel weapons’ 14 

Allelopathy 14 

Rapid evolution 15 

Invasibility of grassland communities 16 

Impact 18 

Transformer species 18 

Impact of N. neesiana 19 

NASSELLA NEESIANA 20 

Taxonomy and nomenclature 21 

Stipeae 21 

Synonyms 22 

Vernacular names 23 

Infraspecific Taxa 23 

Misapplied names 24 

Hybridisation etc. 24 

Evolutionary origin 24 

Morphology and anatomy 26 

Phytoliths 31 

Cytology 31 

Genetic variation 32 

Phenology, growth and productivity 33 

Flowering and fruiting 34 

Cleistogene production 35 

Distribution 35 

South America 35 

Introduced range outside Australia 37 

Australia 38 

History of dispersal in Australia 39 

Identification failures and lack of recognition 39 

Origin 40 

Victoria 41 

New South Wales 43 

Australian Capital Territory 44 

South Australia 44 

Queensland 44 

Tasmania 45 

Potential distribution in Australia 45 

Habitat and climatic and biotic tolerances 46 

Altitude 48 

Climate 49 

2

Fire 49 

Other disturbances 50 

Shade 51 

Soils and nutrients 51 

Landform 52 

Water, drainage and flooding 52 

Other plants 52 

Herbivory 53 

Physiology and biochemistry 55 

Breeding system 56 

Seed production 56 

Dispersal mechanisms 57 

Creeping diaspores 57 

Zoochory 58 

Cleistogene dispersal 59 

Wind 60 

Water 60 

Human activities 60 

Pollen 61 

Rates of spread 62 

Soil seed bank 62 

Germination and seedling recruitment 63 

Seed dormancy 63 

Germination 64 

Establishment of seedlings and juvenile plants 64 

Demography, growth, persistence and dominance 65 

Weed status 66 

Noxious weed status 67 

Undesirable characteristics 67 

Control and management 68 

Hand removal 69 

Herbicides 69 

Slashing and mowing 73 

Fire 73 

Cultivation and cropping 73 

Grazing 73 

Shade 74 

Quarantine and restriction of dispersal 74 

Integrated management in native vegetation 74 

Biological control 75 

Predators and pathogens 76 

Predators 76 

Pathogens 79 

Other biotic relationships 80 

BIODIVERSITY 81 

Definitions 81 

Quantification and indices 81 

Impacts of weeds on biodiversity 82 

General considerations 82 

Impact on threatened species and communities 83 

Approaches to impact assessment 84 

Types of impact 84 

Specific threats posed by weeds to biodiversity 84 

3

Invasive grasses - impacts and threats 88 

Impacts of N. neesiana 91 

GRASSLANDS 93 

Definitions of grassland 93 

Evolution of grasslands 93 

Australian grassland formations 94 

Floristic composition, vegetation structure and ecology 97 

Non-vascular plants and fungi 102 

Soil microflora 103 

Exotic plant components 103 

Rare and endangered plants 105 

Ecological history 108 

Aboriginal management 109 

European Management 110 

Mammal grazing 113 

Sheep and cattle 115 

Rabbits and hares 117 

Marsupials 118 

The role and effects of fire 119 

Nutrients and soil factors 124 

Above- and below-ground biomass 124 

Soil nutrient levels 124 

Weed invasion and nutrient enrichment 125 

Nitrogen dynamics of competing C 3 and C 4 grasses 126 

Nutrient enrichment, nutrient reduction and grassland restoration 127 

Soil disturbance by animals 128 

Native temperate grasslands of south-eastern Australia and their conservation status 130 

Victoria 130 

Gippsland Grasslands 131 

Northern Plains Grasslands 134 

Victorian Basalt Plains Grasslands 135 

South Australia 138 

Tasmania 138 

New South Wales and Australian Capital Territory 139 

Southern Tablelands (NSW and ACT) 139 

Northern Tablelands (NSW) 140 

South West Slopes and Riverina (NSW) 141 

Grassland fauna 142 

Vertebrates 142 

Invertebrates 148 

Grass-feeding insects 153 

Australian Grassland Invertebrate Faunas 156 

Conservation of grassland invertebrates 157 

Impact of N. neesiana on invertebrates 162 

Grassland restoration 162 

CONCLUSIONS 163 

LITERATURE REVIEW APPENDIX 166 

Appendix L1. Grass-feeding invertebrates of south-eastern Australian temperate grasslands 166 

Nematodes of grasses and grasslands 177 

REFERENCES 179 

4

Conventions and standards 

Botanical names follow Walsh and Stajsic (2007) or Shepherd et al. (2001), where possible, unless outdated. Use has also been 

made of the International Plant Names Index (2007) [http://www.ipni.org] and ‘World Grass Species: Synonymy’ (Clayton et al. 

2002 onwards). 

Literature references to Danthonia and Australian Stipa have generally been altered to Austrodanthonia and 

Austrostipa/Nassella/Achnatherum unless there were clear reasons to retain the historical useages. Similarly, references to exotic 

stipoids have generally been changed, where necessary, to currently recognised genera, mainly following Barkworth (2006). 

Abbreviations 

ACT Australian Capital Territory 

c. circa, approximately 

C carbon 

Ca calcium 

cm centimetre 

DSE dry stock equivalent – the annual energy requirement of an adult merino wether 

EVC Ecological Vegetation Class 

gen. genus 

ha hectare 

K potassium 

kybp thousand years before present 

m metre 

mybp million years before present 

N nitrogen 

nov. new (as in nov. gen.) 

NSW New South Wales 

P phosphorus 

Qld Queensland 

S sulfur 

SA South Australia 

sp. species 

spp. more than one species 

t tonne 

Tas Tasmania 

Vic Victoria 

WA Western Australia 

ybp years before present 

5

INTRODUCTION 

Globally and locally, alien invasive plants are one of the most significant causes of degradation of natural ecosystems and 

amongst the greatest threats to the conservation of biodiversity (Carr 1993, Adair 1995, Vitousek et al. 1997, Adair and Groves 

1998, Williams and West 2000, Byers et al. 2002, Wang et al. 2009). Approximately 11% of the Australian vascular plant flora 

consists of established exotic species (Vitousek et al. 1997), a 1990 total of c. 2000 species (Adair and Groves 1998). 

The Poaceae (grasses) contains many of the most damaging invasive plants. 22.9% of all grass species are recognised as weeds, 

the highest proportion of any of the major weedy plant families (Witt and McConnachie 2004). Globally, at least 668 genera and 

c. 2176 species of grass weeds have been recognised (Randall 2002). 

Communities subjected to anthropogenic disturbance and close to human development are more prone to invasion by exotic 

plants (Fox and Fox 1986, Hobbs 1991, Adair 1995, Adair and Groves 1998). Temperate grasslands worldwide have been 

conspicuously invaded (Aguiar 2005) and in Australia are one of the ecosystems most severely affected and heavily invaded by a 

wide range of exotic weeds (McIntyre and Lavorel 1994a, Groves and Whalley 2002). Chilean needlegrass, Nassella neesiana 

(Trinius and Ruprecht) Barkworth (Poaceae: Stipeae) is a relatively new threat. 

N. neesiana has many of the characteristics of a successful invasive species. It is a perennial, long-lived (Cook 1999), cool 

season (winter-spring growing), C 3 , South American, monecious, tussock grass, with a high survival rate of all life stages 

(Gardener et al. 1996a 1999 2003a). It is self fertile (Connor et al. 1993) but can cross pollinate and has a flexible reproductive 

strategy involving both chasmogamous and cleistogamous panicle seeds, along with concealed cleistogamous seeds on the stem 

nodes (Connor et al. 1993, Slay 2002c). In Australia it has been identified as an “aggressive” (McDougall and Morgan 2005 p. 

35), highly invasive (Morfe et al. 2003), high impact (Thorp and Lynch 2000) weed, which is rapidly expanding its range (Lunt 

and Morgan 2000). It’s rapid inter-regional spread from initial successful population confirms its among a set of the “most 

worrisome” invaders (Shea and Chesson 2002) and justifies it status as Weed of National Significance (Snell et al. 2007). It is 

both an environmental weed (Carr 1993) and a weed of agriculture (Grech 2007a). 

The invasiveness of N. neesiana in Australian native vegetation seems to have first come to be widely acknowledged as a result 

of Carr et al. (1992 pp.41, 51) who considered it to be a “very serious threat to one or more vegetation formations in Victoria”. 

In native ecosystems it is reportedly able to actively invade grasslands (Hocking 1998 2007) and is potentially able to 

outcompete C 4 (summer growing) grasses such as Themeda triandra Forrsk. (Ens 2002a). Along with Nassella trichotoma 

(Nees) Hack. ex Arechavelata, it is rated as the most significant weed threat to temperate grassland biodiversity in Australia 

(McLaren et al. 1998, Groves and Whalley 2002) and “the worst environmental weed threatening native grasslands” (Snell et al. 

2007). 

According to Kirkpatrick (1995 p. 77) N. neesiana has “the potential to almost totally displace the native flora” in lowland 

temperate grasslands. Kirkpatrick et al. (1995 p. 35) thought that it seemed to be “capable of dominating grasslands across cool 

temperate southeastern Australia”. Puhar and Hocking (1996) considered it “a serious emerging weed threat” and Liebert (1996 

p. 8) a “major threat” to native vegetation. Morgan (1998d) viewed it as one of the “most potentially threatening species” to T. 

triandra grasslands. Lunt and Morgan (2000 p. 98) rated it as “perhaps the most serious environmental weed in remnant native 

grasslands in southern Victoria” and Morgan and Rollason (1995) considered it to pose “by far the greatest threat of any 

potential new invader” at one grassland. Ens (2005) stated that it “swamps all other ground flora and forms expansive 

monocultures”. According to Beames et al. (2005 p. 2) it is “particularly well adapted to the intensively cultivated areas 

surrounding urban areas and poses a significant threat to mismanaged urban grassland remnants”. 

These opinions are based to various extents on supposition, personal observations and scientific study. Gardener and Sindel 

(1998 pp. 76-77) stated that there is “anecdotal evidence” that N. neesiana causes loss of plant biodiversity in grasslands 

“because litter from the tall tussocks accumulates in the inter-tussock spaces and excludes shade intolerant species”. However T. 

triandra, the major dominant grass of temperate Australian grasslands, has a similar inhibitory effect as the time since fire or 

thinning increases (Stuwe and Parsons 1977). Diversity of bryophytes (mosses, liverworts) and lichens reportedly shows similar 

declines following N. neesiana invasion “because the mosaic of substrates such as rocks and bare soil becomes covered with 

litter” (Gardener and Sindel 1998 p. 77, citing V. Stasjic pers. comm.), as also happens in dense T. triandra (Scarlett 1994). A 

single study of N. neesiana impact on insects indicated that diversity declines, although some groups benefit (Ens 2002a 2002b). 

The success of N. neesiana as a weed has been widely attributed to its ability to produce a large, long-lived soil seed bank (e.g. 

Storrie and Lowien 2003, Gardener 1998) and to the widespread dispersal of seeds by human activities (Bourdôt 1988), 

particularly by mowing and slashing, and on livestock (Gardener 1998, Grech 2007a). However recent evidence (Hocking 

2005b) from southern Australia suggests that the seed bank in native grasslands is much lower and more transient than reported 

in New Zealand (Bourdôt and Hurrell 1992) and agricultural grassland on the Northern Tablelands of New South Wales 

(Gardener 1998). 

The success of N. neesiana is also attributable to the widespread availability of suitable climate and habitat that apparently lacks 

biotic resistance to it. N. neesiana has many of the charactersitics possessed by successful invasive species in general (New 

1994, Williamson and Fitter 1996, Cox 2004): a large native range, abundance in its native range, high vagility (via seed), 

dispersal by abiotic, synanthropic processes, short generation time (reputedly capable of seed production in its first year), high 

reproductive rate, ecological flexibility, wide climatic and physical tolerances, reproduction via a single parent and a general 

association with humans. In addition many varieties and forms have been described, suggesting the species has wide genetic 

variability, at least in South America. Studies of invasive plants in native ecosystems has largely focused on”single-factor 

explanations” for their success (Callaway and Maron 2006), but it is clear that N. neesiana invasion in Australia depends on a 

complex combination of the plant’s characteristics and the environments invaded. 

Williamson and Fitter (1996) stressed the importance of distinguishing between species specific to abundant habitats, and species 

that are habitat generalists, and note that a unique combination of factors account for each successful invasive species. N. 

6

neesiana appears to be a habitat generalist and Australia evidently offers abundant suitable habitat, but the reasons for the 

success of this species in Australia remain poorly understood. 

Management measures for N. neesiana in agricultural areas are currently focused on maximising utilisation by livestock and 

minimising seed production using herbicides or grazing (e.g. Grech 2007a), and in natural areas on control of new outbreaks and 

some serious infestations with herbicides, managing the rate of mineralisation of nitrogen to favour C 3 grasses (Groves and 

Whalley 2002, Hocking 2005b 2007). 

Carr (1993 p. 278) observed that it is “a general tenet of Australian weed ecology that disturbance is a prerequisite for invasion”. 

It is also widely suggested that Australia is more subject to invasion than Northern Hemisphere biomes (Crosby 1986, New 1994 

citing Di Castri 1990), because of its relative biogeographical isolation over a long period (McIntyre and Lavorel 1994a). 

Disturbances of various types operate at all spatial and temporal scales and may have individualistic effects on each organism in 

a community and potential new entrants to that community. A community or ecosystem is the consequence of all disturbances 

that have acted over the period in which it has been assembled. Natural disturbances are essential to maintain native vegetation 

(Hobbs 1991). To understand the role of disturbance it is therefore critical to distinguish between perturbations that have been 

formative factors over evolutionary, biogeographical and ecological time (the endogenous disturbances of Fox and Fox 1986), 

and those perturbations that are of new types or are extraordinary and contribute to community destruction (usually exogenous, 

human induced disturbances, but also geological and cosmological) (McIntyre and Lavorel 1994a, Lockwood et al. 2007). 

Distinguishing between such disturbance regimes is difficult in systems, such as Australian temperate grasslands, that are poorly 

understood historically (Adair 1995). Palaeoecological knowledge of ecosystems has a crucial role in understanding biological 

invasions (Froyd and Willis 2008), but appears to be fragmentory, at best, for these grasslands. Fire and grazing are two of the 

most important disturbance factors that have operated both formatively and destructively. Much of the focus of grassland 

management has consisted of attempts to reinstitute or imitate supposed natural disturbance regimes that are poorly understood, 

in a new context in which invasive exotic species and exogenous human disturbances (fragmentation, N enrichment, global 

warming, etc.) have historically had profound influence and continue to have pervasive effects. Thus there is an ongoing 

requirement to assess contemporary, ‘managed’, disturbance regimes in terms of their new effects. 

Disturbance is often necessary in Australian temperate grassland to maintain the canopy gaps between the tussocks of the 

dominant grasses in which most of the plant diversity of the system exists. But the same sorts of disturbances also promote exotic 

plants, which comprise the bulk of the soil seed bank (Lunt 1990 etc.). Colonisation by novel plants is itself a disturbance factor. 

Hobbs (1991) found that disturbances increase invasion if they increase the availability of a resource that limited the invader 

prior to disturbance and are accompanied by propagule pressure. Identification of the characteristic of disturbance regimes that 

favour particular weeds and suites of weeds, and the management of these disturbances is one of the most critical tasks in 

minimising environmental weed impact (Adair 1995). 

Gardener and Sindel (1998) advocated quantitative studies to evaluate the biodiversity impacts of N. neesiana, compare the 

impacts resulting from general degradation, and evaluate the effects of N. neesiana management techniques on the promotion or 

inhibition of biodiversity. Grice (2004a) concurred with the need for such studies, noting that monitoring of biodiversity can be 

an important tool in evaluating a weed management strategy. 

The success of a plant invader depends on its biological attributes, the attributes of the communities and ecosytems that are 

potentially invasible, and the effects of human interference. Bourdôt and Hurrell (1989a p. 415) considered the invasiveness of 

N. neesiana in sheep pastures to be due to “adaptations that enable the plant to survive the hazards of semi-arid, low-fertility 

environments, rather than to high competitive ability”. 

In the review which follows, the invasion of Australian grasslands by N. neesiana is first contextualised within the frameworks 

of current biological invasion theories and hypotheses. The biological and ecological attributes of the plant are then examined in 

detail, along with its history, impacts and management in Australia. Next, the concept of biodiversity is discussed. The 

components of biodiversity and the ways they may be assessed are examined, along with the nature of weed impacts on the 

various components of native ecosystems. In section 4, knowledge of the properties of the invaded grasslands, including their 

components and dynamics are reviewed. Finally an attempt is made to synthesise this knowledge into a complete picture of what 

is and is not known about the impact of N. neesiana on biodiversity in Australia’s indigenous grasslands. 

7

Theoretical framework 

“Apparently the spheres of competitiveness under which the native vegetation had evolved were irreversibly destroyed by alien 

introduction, or at least by the conditions conducive to the alien introduction.” 

Raymond A. Evans and James A. Young (1972), on the success of invasive grasses in the inter-mountain areas of the western 

USA, in ‘Competition within the Grass Community’, In V.B. Youngner and C.M. McKell (Eds.), The Biology and Utilization of 

Grasses. Academic Press, New York. 

A number of competing hypotheses and theories seek to explain exotic plant invasions and their impacts on biodiversity. These 

concentrate on four elements of the systems: 1. the properties of the invasive plant (‘invasion potential’ or invasiveness) (e.g. 

Rejmánek and Richardson 1996, Williamson and Fitter 1996), 2. the properties of the system at risk from invasion (its 

‘invasibility”) (e.g. Londsdale 1999), 3. the role of disturbance (Hobbs 1991, Hobbs and Heunneke 1992, D’Anotonio et al. 

1999), and 4. dispersal mechanisms and factors (‘propagule pressure’) (Williamson and Fitter 1996, Levin 2006). Adequate 

explanation is beset with the same problems faced by ecologists seeking to understand the factors that determine the species 

composition of any space, in particular the difficulties of distinguishing between causal processes, the environmental conditions 

that modulate them, and the patterns that result (Leigh 2007). 

Residence time, the amount of time that an exotic species has spent in its introduced range, is obviously a major determinant of 

the extent and impact of an invasion. The longer the residence time, the higher the likelihood that the invader will become 

widespread (Hamilton et al. 2005). For example, the minimum residence time of 116 exotic grasses in Venezuela (time since the 

first national record of a species) is significantly correlated with the total number of known localities in which each species 

occurs (Rejmánek 2000). Longer residence time at the patch scale may also be expected to increase impact, due to alterations in 

the density and age structure of the invader population in the patch, and the accumulation of feedback and indirect effects in the 

invaded environment. 

The ecological mechanisms that enable environmental weed invasions are in general complex and poorly understood (Prieur- 

Richard and Lavorel 2000, Levine et al. 2003, Hayes and Barry 2008). Less than 5% of studies on invasive plant impacts 

examined by Levine et al. (2003) attempted to determine the processes causing the invasion. Disturbance has “unanimously been 

shown to favour plants invasions” (Prieur-Richard and Lavorel 2000 p. 3) but many species appear to be invasive in the absence 

of significant anthropogenic disturbance, their success being attributed inter alia to inherently faster growth rates, superior 

competitive abilities related to form, phenology, resource exploitation, etc., and the occupation of unfilled structural niches (Carr 

et al. 1986, Carr 1993). 

However, since each succesful invader and invaded system have distinctive characteristics, unique interactions of multiple 

factors are most likely responsible in each case, and single factor explanations are poorly informative (Callaway and Maron 

2006). 

Some of these hypotheses and theories are explored in more detail in following sections, commencing with invasiveness, then the 

enemy release hypothesis and the concept of biotic resistance, the theories of resource enrichment and fluctuating resources with 

their emphases on the importance of disturbance as a precursor to invasion, theories related to rules of community assembly 

including the ‘empty niche’ concept and competitive exclusion, an examination of the possibilities that rapid evolution of the 

invader is a significant contribution to its success, and a discussion of the concept of invasibility of communities and 

geographical areas. Finally the parameters involved in determination of the impact of an invasive plant are briefly discussed. 

Invasive potential of a species 

Many attempts have been made to identify characteristics of ‘weediness’ or what makes some plants more invasive than others 

(Rejmánek 1995, Rejmánek and Richardson 1996, Hayes and Barry 2008) and various suites of charactersitics possessed by 

successful invasive species have been identified (Table 1). 

Each successful invasive species generally possesses a unique subset of these characteristics (Williamson and Fitter 1996). 

According to LeJeune and Seastedt (2001 p. 1572) a particular species is invasive when it “encounters habitats in which its 

particular suite of traits confers competitive advantage over the native dominants”. Thus the two best predictors of invasion 

success are a climate/habitat match of native and exotic range and a history of invasive success elsewhere (Hayes and Barry 

2008). Other characteristics significantly associated with invasion success of plants are the date of introduction (residence time), 

biogeographic origin, brief juvenile period, growth form, asexual or vegetative reproduction, and flowering period and season 

(Hayes and Barry 2008). Environmental weeds (exotic plants invasive in natural ecosystems) tend to possess similar 

characteristics, a subset of the following: high input of viable propagules to the envionment, development time 5 year dormancy, high biomass production, dense canopy, efficient long distance (>1 km) dispersal, 

allelopathic properties, coloniser of disturbed ground, adapted to fire, broad climatic tolerance and resistance to predation 

(Williams and West 2000, after Adair 1995). A similar set of criteria was used to undertake a rigourous weed risk assessment on 

N. neesiana as part of the Victorian ‘pest plant prioritization process’ (Morfe et al. 2003) and N. neesiana was classed as highly 

invasive. 

The invasive potential of a species is often compromised by specialised requirements. These may include the need for particular 

symbionts, pollinaters or scarce, rare habitat features. Invasiveness may also be increased by the possession of special 

competitive mechanisms, which can include allelopathy or a novel growth form (Newsome and Noble 1986). 

R-strategists have a rapid rate of population increase and high mobility, are adapted to unstable environments and early 

successional stages, and are able to quickly build up their numbers in areas with high levels of unused resources and low 

8

population densities of existing species. K-strategists have low mobility and are adapted to late successional stages and stable 

environments where the carrying capacity is approached and competition is high (Matthews 1976, Rejmánek and Richardson 

1996). For perennial plants, the ability to propagate vegetatively has often been considered important (Newsome and Noble 

1986). 

Table 1. General characteristics of successful vs. unsuccessful invasive organisms. Sources: Newsome and Noble 1986, New 

1994, Adair 1995, Rejmánek and Richardson 1996, Williamson and Fitter 1996, Cox 2004, Whitney and Gabler 2008. 

Successful invaders 

large native range 

wide climatic tolerance 

abundant in native range 

high vagility 

high reproductive rate 

short generation time 

reproduction requiring a single parent 

small propagule size 

high ecological flexibility 

wide physical tolerance 

wide genetic variability 

larger than related taxa 

rapid growth 

absence of specialised requirements 

low susceptibility to attack by other organisms 

special competitive mechanisms 

r strategists 

association with humans (commensal) 

Unsuccessful invaders 

small native range 

narrow climate tolerance 

rare in native range 

low vagility 

low reproductive rate 

long generation time 

reproduction requiring two parents 

large propagule size 

low ecological flexibility 

narrow physical tolerance 

narrow genetic variability 

smaller than related taxa 

slow growth 

specialised requirements 

high susceptibility to attack by other organisms 

no special competitive mechanisms 

K strategists 

not associated with humans (not commensal) 

Subsets of inasive characteristics can be combined to define particular ‘strategies’ possessed by invasive plants. Newsome and 

Noble (1986) identified four such strategies, based on the suites of ecophysiological characters possessed by different types of 

weeds: 

1. Gap-grabbers – early germinators with fast initial growth enabling preoccupation of ecological space. 

2. Competitors - taller growing (light) or with deeper or more extensive roots (water and nutrients). 

3. Survivors – longevity due to resistance to mortality factors or clonal growth. 

4. Swampers – mass germinators. 

Rejmánek (2000) found that Eurasian and North African grass species adventive in eastern and western North America are more 

often those with large rather than small native latitudinal ranges and that the latitudinal size of the native range is highly 

correlated with the size of the introduced range. Species with larger ranges may be more successful because of the larger 

absolute size of the propagule pool and because they are more likely to interact with long distance dispersers (Levin 2006). 

Schmidt et al. (2008) highlighted a particular character suite for invasive grasses: 1. smaller seed size than the native species; 2. 

plastic morphological traits that enable the invader to adjust to water and N deficiencies; 3. faster growth to sexual maturity than 

the native species; and 4. ready stem dehiscence at the lower node (i.e. stoloniferous growth form). However Lonsdale (1994) 

found no relationship between weediness and height growth, relative growth rate, time to maturity (annual or perennial) or seed 

weight of exotic grasses introduced into northern Australia. Instead the successful weeds were most likely to be species judged 

to be useful, high performing or persistent in experimental agronomic field trials. 

Numerous studies have investigated particular biological traits and their relationships with the success of invasive plants. 

Hamilton et al. (2005) investigated specific leaf area (the ratio of the light-capturing area per unit dry mass), plant height and 

seed mass and found significant correlations between invasion success and small seed mass at regional and continental scales, 

and between high specific leaf area at the continental scale. Greater environmental heterogeneity at regional levels, with 

consequent increased biotic resistance, was invoked as one cause of the differences across spatial scales. Rejmánek (1995) found 

that invasiveness of Pinus species correlated with small seed weight, short juvenile period, short mean intervals between large 

seed crops and vertebrate dispersal. 

Various attempts have been made to assess the most important invasiveness characters by analysing subsets of regional weed 

floras. For example, Gassó et al. (2009) assessed invasive success on the basis of area occupied in mainland Spain and found that 

wind dispersed species were the most invasive, followed by animal dispersed species, and that residence time, when

The suitability of a new habitat for a particular invader is known as the ‘invasibility’ of the habitat: “the properties of the region 

of introduction that facilitate the survival of non-indigenous species” (Gassó et al. 2009 p. 51). 

Enemy release and biotic resistance 

‘Enemy release’ theory postulates that an exotic plant becomes a successful invader because it lacks co-evolved natural enemies 

and plant competitors in its introduced range: it is ‘released’ from the effects of specialist predators, parasites and plant 

antagonists of its native environment, and is much less subject to herbivory in the introduced range, the native generalist 

predators being better adapted to, and preferentially consuming native plants (Keane and Crawley 2002, Levine et al. 2004, 

Parker et al. 2006a). Reduced control by natural enemies in the invaded areas is believed to enable the plant to be more 

productive and more reproductively successful, and for its populations to expand. 

‘Biotic resistance’ or ‘diversity-resistance’ theory posits that some combination of biological effects of the native organisms 

regulate the success of the exotic plant, although they seldom prevent invasions (Levine et al. 2004). The original idea is 

attributed to Elton (1958) who proposed that the greater the species-richness of a community, the more resistant to invasion it 

should be. More intense competion for resources and consequent fuller resource sequestration in more diverse communities is the 

explanation usually invoked, either due to a combined effect of all the species, or the greater likelihood that a species or 

functional group competitive with the invader is present in diverse communities (Symstad 2000, Dukes 2002, Stohlgren 2007). 

Early attempts to test the theory through analysis of plant diversity at a mix of spatial scales (e.g. the meta-analysis by Lonsdale 

1999) provided a confused picture and mostly showed correlations of high native plant species richness with high exotic 

richness. But recent consensus is that biotic resistance functions very differently at different spatial scales. A majority of plant 

diversity studies and experiments at small spatial scales (patch, plant association) and some modelling support the theory (Prieur- 

Richard and Lavorel 2000, Symstad 2000, Dunstan and Johnson 2006) but at larger spatial scales there is generally a marked 

positive correlation between plant species richness and invasibility that corresponds to increased environmental (landscape, 

regional, continental) heterogeneity (Knight and Reich 2005, Dunstan and Johnson 2006). However Stohlgren (2007) argued that 

even at the small scale (1 m 2 ) contradictory results have been found, suggesting that environmental factors other than the plant 

diversity of an invaded area may frequently be of greater importance in determining invasibility. 

Enemy release theory is focused on plant diseases and the invertebrate consumers of plants, more or less ignoring the possibility 

that native plants in the invader’s area of origin may have similar negative impacts. In that context, to explain a particular 

successful plant invasion, the enemy release hypothesis requires that the plant’s specialist natural enemies in the native range are 

absent from the introduced range, that specialist enemies in the introduced range do not shift onto the potential new host, and that 

the generalist enemies in the introduced range have less impact on the plant than on native plants in the introduced range (Keane 

and Crawley 2002). The hypothesis is supported by numerous studies comparing the invertebrate predators and diseases of plants 

in their native and exotic ranges (e.g. Memmott et al. 2000). Such studies have often been undertaken preparatory to classical 

biological control programs. 

Enemy release theory has been the “conceptual underpinning” of classical biological control since its inception (Callaway and 

Maron 2006 p. 371).The numerous examples of pest suppression resulting from practical biological control (Julien and Griffiths 

1998) and the very limited non-target impacts of these programs (Waterhouse 1998, Wajnberg et al. 2001) provide strong 

evidence that enemy release is an important cause of plant invasions: deliberately introduced parasites and predators directly 

control population growth of some invaders and may have indirect effects on their performance (Prieur-Richard and Lavorel 

2000). However, to prevent damage to non-target species, classical biological control is constrained to concentrate on specialist 

enemies with narrow host ranges. The full trophic complexity of the system in the plant’s native range is extremely difficult to 

assess and highly variable. Furthermore, succesful biological control of a weed does not necessarily demonstrate that the weed 

became a problem because of natural enemy release, since the biological control agent itself has been released from its natural 

enemies. 

A meta-analysis of 13 studies that quantified the impact of native natural enemies on invasive exotic plants (Keane and Crawley 

2002) partially supported the enemy release hypothesis, finding that in every case where native specialist insects were 

differentiated, they attacked the exotic, however their impact was usually negligible. Native generalist enemies were found to 

have greater impact on the exotic in only two cases. More recent and comprehensive meta-analyses contradict the suppositions of 

the enemy release hypothesis (Cox 2004): native generalist predators or “evolutionarily novel enemies” (Parker et al. 2006a) 

preferentially attack exotic prey, which are poorly adapted to resist them or “defensively naive” (Parker and Hay 2005 p. 965), 

having had limited coevolutionary history of association with the predators. The“evolutionary naïveté” of the invasive species in 

the invaded system places it at greater risk of attack by newly encountered generalist herbivores (Parker et al. 2006b). 

Furthermore, generalist herbivores commonly have greater impact on plant community structure in terrestrial ecosystems than 

specialists (Parker and Hay 2005 citing Crawley 1989). 

Parker and Hay (2005) found that native generalist herbivores offered food choices of congeneric or confamilial exotic and 

native species, significantly preferred, with some exceptions, the exotic plants. Following on from this work, Parker et al. 

(2006a) reviewed a large number of manipulative field studies involving over 100 exotic plant species and found that native 

generalist herbivores suppress exotic plants more than native plants, and that exotic herbivores facilitate the abundance and 

species diversity of exotic plants by preferentially consuming native plants. Invertebrate herbivores were found to have only one 

third to one fifth the impact of vertebrate herbivores on survival of exotic plants (Parker et al. 2006a). However in the studies 

examined, the native range of a high proportion of the exotic plants was the same region as that of the exotic herbivore (Parker et 

al. 2006a), suggesting that exotic plants without a common evolutionary history with the exotic predator, and therefore lacking 

defenses against them, may be as susceptible to predation as native plants. The effect of native generalist predators on exotic 

plants is reduced if the plant is closely related (in the same genus) to plants in the introduced range, and six times as strong on 

genera new to the invaded region (Ricciardi and Ward 2006, Parker et al. 2006b). Invertebrate herbivores are usually specialised 

to feed on particular plant parts and to particular plant species, groups of related species, or wide ranges of species, whereas 

vertebrate herbivores are generally adapted to consume plants of a particular life form or plant product (Cox 2004). Closely 

10

elated plants have similar defences against herbivores (see references in Ricciardi and Ward 2006) and are likely to share traits 

that confer resistance to attack, or have similar physiological adaptations that lessen stress and better enable the operation of their 

defences (Parker et al. 2006b). Exotic predators which share little evolutionary history with either the native or exotic plants 

show no feeding preference for native species (Parker and Hay 2005). 

Most studies framed within biotic resistance and/or enemy release hypotheses have largely ignored the role of plant interactions 

with soil microbes and the acceleration of knowledge of soil microecology. There is good evidence that invasive plants in North 

America have escaped from the host-specific soil microbes of their homelands and also formed new relationships with nonspecific 

microbial mutualists in the invaded territory (Callaway and Maron 2006). 

Mack’s (1989) generalisations about the role of exotic livestock in destruction of native caespitose grasslands in western USA, 

the South American pampas and Australia, their facilitation of invasion by exotic grasses with which they coevolved and the 

history of native grassland degradation in Australia (Moore 1973 1993, Groves and Whalley 2002) mesh neatly with the biotic 

resistance hypothesis. In a sense “exotic plants may thrive not by escaping their native enemies, but by following them” (Parker 

et al. 1996a p. 1460). 

N. neesiana may possess pre-adaptations that minimise predation by Australian native herbivores (suggested perhaps by the 

general presence of native Stipeae in invaded areas in Australia), or the native ecosystems invaded now lack their natural 

herbivore assemblage (e.g. kangaroos generally absent, extinct macropods and marsupial megafauna). The exotic herbivore 

assemblage that has invaded native grassland (including an array of invertebrates as well as mammals) may differentially attack 

the native plants (Parker et al. 2006a), or the ecosystem has been otherwise anthropogenically disturbed, or N. neesiana has other 

attributes (fecundity, high growth rate etc.) that enable it to overcome the effects of native generalist herbivores (Parker and Hay 

2005). 

The biotic resistance hypothesis,as it relates to plants, is not challenged by evidence that landscape scale areas with high plant 

species diversity also tend to have higher numbers of exotic plant species. Lonsdale (1999) compared exotic and native plant 

species richness at 184 landscape-scale sites of wide variation in size and found that the number of exotic plant species, but not 

their proportion of the flora, increased with native plant species richness. Similar analyses at large landscape scales show similar 

trends, however the areas invaded are not plant communities - the ecological units that are expected to show biotic resistance - 

and such trends are at best weak in small areas (Cox 2004). 

The biotic resistance hypothesis has been further elaborated under the moniker of functional group theory: the idea that the 

diversity of functional groups rather than species enables resistance (Prieur-Richard and Lavorel 2000). Absence of a particular 

functional group (e.g. grass seed predators) eases constraints on an invasive plant. The presence of a larger number of functional 

groups of plants suggests that a larger proportion of available resources are already efficiently captured, so there should be 

greater resistance to invasion. 

If there is reduced predator pressure in the new environment, a plant has a lesser requirement for defence and an evolutionary 

trade-off may occur in which the ‘freed up’ resources are allocated to other purposes, inclduing reproduction. Trade-offs 

resulting in growth increases and higher investment in reproduction are believed to be common in invasive plants (Cox 2004). 

The rate at which invasive plants acquire new natural enemies is highly variable, but is generally rapid at first, the diversity of 

generalist herbivores reaching a maximum in as little as 100 years, then slows, with host shifting by and evolution of specialist 

herbivores gradually occurring over longer periods, up to 10,000 years (Cox 2004). Major factors influencing these rates are the 

diversity of the native herbivore and pathogen pools and the extent of their adaptation to phylogenetically related native plants, 

the phenological availabilityof the invasive plant to the potential native utilisers, and the innate defences of the plant (Cox 2004). 

The abundance or total area occupied by the invasive plant also appears to be important, e.g. as found in a global study of the 

arthropod pests of Saccharum officinale, which found that the number of pest species had a linear relationship to the area of 

sugar cane under cultivation (Cox 2004 citing Strong et al. 1977). 

It can therefore be hypothesised that N. neesiana is preferentially consumed by native generalist species compared with 

dominant native grasses, and that exotic generalist plant predators will preferentially consume the dominant native grasses. If this 

is the case, grasslands with diverse and abundant populations of exotic plant predators should be more susceptible to invasion 

and more highly invaded by N. neesiana. Grasslands that have inherent biotic resistance should be occupied by native generalist 

mammalian herbivores including kangaroos, and have a greater diversity of native generalist phytophagous invertebrates. 

Grasslands without biotic resistance should lack large native mammal herbivores, have a depauperate guild of native generalist 

grass-feeding invertebrates, and an increased complement of exotic grass-feeding mammals Manipulative field experiments 

involving predator removal or paired feeding assays (e.g. Parker and Hay 2005) are needed to demonstrate such effects. 

Resource-enrichment and fluctuating resources 

A successful invasive plant may simply be a superior competitor for basic resources such as light, water and nutrients. Many 

biological attributes that engender such superior ability have been identified, for example plants with C 3 and C 4 photosynthetic 

pathways are superior convertors of sunlight to sugars in different environments, and any one plant may have higher fecundity or 

a faster growth rate than another. But because resources are ‘locked up’ variably in space and time by the plants in the preexisting 

community, disturbance that kills or inhibits them and frees up resources is generally required for a successful invasion, 

or their must be extrinsic addition of resources at a rate faster than the native plants can use (Herbold and Moyle 1986, Hobbs 

1991, Burke and Grime 1996, Cox 2004). 

The tenet that disturbance is a prerequisite for invasion is implicity based on the notion that in undisturbed, successionally 

mature vegetation, surplus resources are absent (Carr 1993) or minimised or unobtainable by the existing flora. The fluctuating 

resources theory posits that a “plant community becomes more susceptible to invasion whenever there is an increase in the 

amount of unused resources” (Davis et al. 2000). The community becomes more susceptible to invasion by a particular exotic 

plant if the particular resource was previously limiting the growth or survival of that plant (Hobbs 1991). Continuity of the 

invasion requires that gains the invader makes are not lost when resource supply contracts (Melbourne et al. 2007). 

11

Disturbance, defined as any process that creates open ground, changed habitat or altered resource availability (Hobbs 1989 1991, 

Mack and D’Antonio 1998, Lockwood et al. 2007) is universal at a range of spatial and temporal scales, so gaps in vegetation 

and fluctuating resource pools are always available. Under this definition, disturbance includes reduction or lack of normal 

disturbance to which the community is adapted, e.g. reduced fire frequency, removal of grazing, or loss of burrowing mammals, 

can provide fluctuating resources, as native plants senesce or seed banks decay without replacement. Elimination of perturbation 

in disturbance-dependent systems is one of the most serious ‘disturbances’ they can suffer (MacDougall and Turkington 2007). 

Many plant communities and species require disturbance, particularly for regeneration (Hobbs and Huenneke 1992). Whether or 

not the gaps or unused resources created by disturbance can be taken up by an exotic species therefore depends on their size, 

their spatial and temporal availability, the pool of available species and the requirements of the particular species (Prieur-Richard 

and Lavorel 2000). Whether or not a particular species invades is dependent on the disturbance characteristics: its mgnitude and 

severity,duration, predictability,distribution in space and time and synergistic effects on other disturbances (Lockwood et al. 

2007). When the parameters of the disturbance are suitable, exotic species that possess resource utilisation traits not present in 

the native flora often seize the advantage (McIntyre et al. 1995). Invasions of exotic plants can be greatly increased by 

combinations of disturbance such as nutrient addition and soil disturbance (Cale and Hobbs 1991). 

Early successional stages have greater pools of unused resources and less competition, so are more susceptible to invasion (Davis 

et al. 2000). Resource poor environments are not invaded by many exotic species (Cox 2004). Greater species diversity generally 

corresponds with more complete resource usage, so diversity theoretically confers invasion resistance by limiting resource 

fluctuation. 

Weed invasions are commonly associated with extreme resource fluctuations. One example is the invasion of Buffel Grass 

Cenchrus ciliaris L. in arid Australia. Prior to 1974 it was apparently naturalised in a few small areas in the semi-arid inland, but 

in the unusually wet season of 1974-5 it spread along flood plains and “run-on areas” across large areas of the arid zone (Moore 

1993 p. 316). 

Nutrient enrichment is often a significant contributory cause of invasions (Milton 2004). For example, King and Buckney (2002) 

compared total N and P, and concentration of Na, K, Ca and Mg cations in soils of 16 urban bushlands and 8 national parks in 

Sydney and the proportion of exotic plant species present in the vegetation. All soil nutrients were significantly higher in urban 

areas, and gradients from low to high invasion, and low to high concentrations, were correlated. The correlations were best 

explained by a combination of nutrients rather than any single nutrient. Direct manipulative studies have demonstrated that 

nutrient addition without other disturbance can cause weed invasion. 

Other studies show that both nutrient enrichment and other disturbance are required to alter community composition. Hobbs 

(1989) experimented in a range of heathland, shrublands and woodlands by digging the soil, adding fertiliser, both digging and 

fertilising, and adding seeds of Avena fatua L. and Ursinia anthemoides (L.) Poir. Very similar results were obtained in all 

communities. Avena responded to digging with little fertiliser effect. Ursinia showed little response to any treatment. Digging or 

fertiliser alone had little effect on resultant weed biomass, but a strong effect together. These results were attributed to better 

seed survival and more safe germination sites in the dug areas, a probable small increase in nutrient availability in dug treatments 

and extreme nutrient limitation for the weeds. Hobbs (1989) therefore argued that certain types of disturbance do not 

significantly increase resource availability and do not make a community more invasible. In a long-running experiment Davis et 

al. (2000) demonstrated a strong relationship between the levels of disturbance and nutrient enrichment and the mean aggregate 

cover of 54 plant species deliberately sown into a Derbyshire, UK, grassland. 

Fluctuating resources theory posits that resource removal or impoverishment will not cause an invasion. This was tested by 

Kreyling et al. (2008) who examined the effects of drought and heavy rainfall on simple experimental plant communities by 

counting individual plants that invaded from the matrix vegetation. Heavy rainfall increased invasibility while drought decreased 

invasibility. Higher diversity in the experimental plots (4 spp. vs. 2 spp.) decreased invasibility, and the two effects acted 

independently and were additive. Furthermore, several species that invaded were dependent on the functional groups present in 

the artificial communities or the nature of the particular weather event. Thus the predictions of both niche theory and fluctuating 

resources theory were supported. 

Enrichment or limitation of any resource including available space, soil nutrients, water and light may facilitate invasion. 

However the basic ecological processes that enable invasion are no different to those that enable native plants to regenerate or 

occupy new areas (Davis et al. 2000). In ecological time, an organism that is unable to change its distribution in response to 

environmental change must either evolve or become extinct. 

The theory of fluctuating resources predits that greater susceptibility to invasion occurs: 1. immediately after resource 

enrichment or following a decline in rate of resoure usage; 2. when a disturbance increases resource supply or reduces resident 

vegetation sequestration of resources; 3. when the interval between resource enrichment and resident vegetation sequestration is 

long; 4. when grazing is introduced, paticularly in high-nutrient areas; 5. that there is no necessary relationship between 

community plant diversity and invasion resistance and 6. that there is no general relationship between average community 

productivity and invasion resistance (Davis et al. 2000). 

In temperate Australian native grasslands the theory therefore predicts that: 1. areas subject to more intense or frequent resource 

enrichment are more prone to invasion (e.g. areas with soil disturbance, areas where the existing vegetation has died or been 

killed, floodplain areas, areas subject to fertiliser drift, areas more subject to anthropogenic N enrichment such as roadsides, 

urban areas, etc.); 2. fires in autumn create greater susceptibility to invasion than fires in spring, since autumn fires mean a 

longer interval between the growth period of the dominant grass Themeda triandra and the resource enrichment, providing the 

opportunity for winter growing N. neesiana to sequester more nutrients, light and space; 3. more intense fires intensify invasion 

because they create a greater increase in nutrient supply and a longer period of reduced resource capture by native plants; 4. 

drought will facilitate invasion, especially if it breaks in the period preceding the main N. neesiana seedling establishment and 

growth periods; 5. grazing of long-ungrazed areas will facilitate invasion since it makes available nutrients that were previously 

locked up by the native plants, and 6. grasslands with greater native species richness will be more resistant to invasion only if 

they are characterised by more complete and total resource utilisation. 

12

For invasion to occur there must be not only fluctuating resources or resource enrichment, but propagule pressure (Davis et al. 

2000). Disturbance (including the absence of particular disturbances), resource enrichment and propagule dispersal events are 

often correlated and can occur as part of the same event (e.g. intensive grazing by a flock of sheep contaminated with N. 

neesiana seed). Determining the proximate cause of invasions is thus complicated, but it is necessary to investigate the 

contribution of each of these factors to succesful invasions in order to devise optimal management practices. 

Propagule pressure 

Propagule pressure is the number of propagules dispersed into a given area and may be more important than any other factor in 

determining the success of a potential invader (Williamson and Fitter 1996, Lonsdale 1999, Levin 2006). Propagule pressure is 

equivalent to the factorial combination of the number of introduction events and the number of individuals per event (Lockwood 

et al. 2009). Where the availability of propagules is low, recruitment is always limited (Levin 2006) and when populations are 

small, there is a reduced likelihood that they will survive (Lockwood et al. 2009). Rejmánek (2000 p. 498) examined various 

proxy measures for propagule pressure and concluded that, as a general rule, initial population size and the number of 

introduction attempts determined the success of an invasion: the “most robust but ... trivial, generalization in invasion ecology”. 

However a recent metaanalysis (Hayes and Barry 2008) found a significant association between the number of released or 

arriving individuals or the number of release/arrival attempts and establishment success only for animals, and considered the 

proposition to be ‘untested’ for plants. 

Propagule pressure is a complex function, based on fecudity, dependent on propagule dispersal mechanisms and the availability 

and incidence of dispersal agents, and ultimately determined by the ability of the propagules to find suitable habitat and establish 

new populations (Williamson and Fitter 1996). In situations where an invasive species is already present and reproducing, use of 

the term “propagule rain” may be preferable (Lockwood et al. 2009). Where more than one potential invader is being considered, 

e.g. with community-level processes, the combination of propagule pressures is better termed ‘colonisation pressure’ (Lockwood 

et al. 2009). 

Plant taxa that achieved successful ancient long distance (transcontinental) dispersal have similar characteristics to modern 

invasive plants: high propagule dispersability, large propagule production by multiple, large populations, wide geographical 

range and major presence in their communities (Levin 2006). Species with large native ranges tend to be more abundant and 

produce more propagules per unit area so have a greater chance of becoming invasive elsewhere, purely based on the propagule 

pressure they exert (Levin 2006). Substantial propagule pressure is required to overcome the genetic and demographic liabilities 

of small populations (Levin 2006). 

Vacant niches and competitive exclusion 

Under the theory of competitive exclusion and niche displacement, a more competitive invader can occupy the niche previously 

occupied by a native species, and an empty niche is open to invasion. The theoretical underpinnings of this approach are derived 

from community assembly theory, based on island biogeography (Woods 1997, Seabloom et al. 2003, Cox 2004). Shea and 

Chesson (2002) reframed the theory in the context of community ecology. Niche theory and dispersal assembly theories may be 

contrasted with ‘neutral community assembly’, which predicts that the characteristics of a potential entrant to a community have 

a neutral effect on the possibility of it becoming a part of that community, in particular, each species is equally likely to 

reproduce (Leigh 2007). 

The ‘storage effect’ is a related concept that incorporates the temporal and spatial variation of niche elements. “The invader must 

be able to take advantage of times or locations in the landscape where the environment favours its population growth over that of 

the resident species, and store those gains in time or space in such a way that they are not eroded too much in unfavourable times 

or locations” (Melbourne et al. 2007 p. 84). ‘Storage’ can consist of a seed bank, a population of adult plants or a dormant tuber. 

Richer communities supposedly have more niches, both filled and vacant (Prieur-Richard and Lavorel 2000). Resources that are 

unsequestered by existing native species represent elements that could contribute to an ‘empty niche’. An invader might use 

resources in a different way to the native species or at different times, without interfering with other species, and so could 

theoretically occupy a previously empty niche. Or it might, through competitve processes, sequester resources that would 

otherwise by used by the resident species. 

Species that are ecologically similar to an invader are often lacking in successfully invaded communities (Mooney and Drake 

1989, Systad 2000), suggesting that vacant niches are often present and that competitive exclusion often repels invasions. 

However various authors have argued that there is little evidence for the existence of vacant niches (Newsome and Noble 1986, 

Prieur-Richard and Lavorel 2000) and that the competitive superiority of plant invaders to native species has “rarely been tested 

experimentally” (Seabloom et al. 2003 p. 13384). 

Invasive plants that establish in dense populations must displace other plants and alter community composition, unless they 

occupy an unfilled niche, but if the niche was unfilled there should be no displacement, the invader is not a problem, and it 

successfully integrates into the existing community. The invasive species may displace another plant with a similar niche, or may 

be widely competitive and have the potential to completely restructure or replace a community, alter its successional dynamics 

or directly interact with disturbance regimes (Woods 1997, Mack and D’Antonio 1998), and thus alter the niche space of many 

community components. A community or one of its members may repel even a superior invasive competitor “because of the 

priority effect that established residents have over invaders” (Systad 2000 citing Case 1990 1991). Alternatively a native species 

may persist on sites where they have unusual competitive advantages, e.g. native grasses persist on sites with serpentine soils, 

but have been largely replaced by exotic grasses in California (Melbourne et al. 2007 citing Harrison 1999). 

Other author have argued that the niche concept itself is “a circular argument empty of mechanism and process” (Wedin 1999 p. 

193). Niche theory is largely contradictory, since niches are multidimensional hyperspaces defined by their occupation by a 

species population in relationship with all the other organisms in the community (Herbold and Moyle 1986, my emphasis). A 

13

niche cannot be defined if it is infinitely malleable: if the niche is an autecological attribute of a species, the niche moves with 

the species and there are never empty niches, if it is a synecological attribute, the invasive organism must modify the niches of 

other organisms to sequester niche hyperspace, so again there is no preformed or definable vacant niche (Herbold and Moyle 

1986). An invasive species must rearrange the community and sequester flows or resources previously used by other organisms, 

and therefore must effect them negatively. In general all invasive species which have been investigated have some negative 

effects (Herbold and Moyle 1986). The mechanisms by which this happens, what resources are diverted, etc., is the real question 

of interest (Seabloom et al. 2003), and is dependent on particular circumstances. 

Current consensus, at least in some schools, appears to be that niche theory is a sterile framework, lacking explanatory and 

predictive power. However this non-mechanistic approach, seemingly based on the tenet that ‘diversity confers stability’ and the 

false but “commonly accepted ecological truism” that richer communities are less invasible (Lonsdale 1999 p. 1533) continues 

under a new guise, with ‘functional groups’ replacing niches as a focus of investigation. Functional diversity, “the number of 

functional groups with different behaviours for a particular process”, rather than species diversity, supposedly determines “major 

thresholds in ecosystem processes” and the properties of systems that control their invasibility (Prieur-Richard and Lavorel 2000 

p. 5). At least, if the composition of functional groups is based on similarity of resource use, the processes and mechanisms that 

could cause displacement or facilitiate invasion are more explicit and accessible in these functional group approches than in 

simple diversity-resistance approaches. 

Many mechanisms might be involved in competitive superiority and these are mostly addressed in this essay within the 

frameworks of the main theories. Competive superiority could result from possession of a particular unique characteristic (“novel 

weapons” below), general superior adaptation evolved in the native environment (superior ‘invasive potential’, see above), 

release from natural enemies (above) or the ability to better exploit disturbance (‘resource enrichment and fluctuating resources’, 

above). Dominance by the invader in some but not other areas, suggestive of competitive superiority, may actually be just a 

priority effect (Seabloom et al. 2003): the exotic cannot outcompete an equilibrium population of competitors, but is the first to 

invade after strong disturbance, and establishes dominance, leading to “multiple stable equilibria” (Seabloom et al. 2003 p. 

13384). 

‘Novel weapons’ 

Successful invaders may possess ‘novel weapons’ that enable them to kill, suppress or outcompete native species. Novel 

weapons possessed by invasive plants may include the ability to produce chemicals that are toxic to native herbivores (see for 

example McBarron 1976) or to native plants. More broadly, they may consist of “competitively unique traits” (Seabloom et al. 

2003) or functional attributes, not possessed by the native species, that enable access to unexploited resources (Callaway and 

Maron 2006). 

Grasses have many adaptations that deter predation. These include anatomical structures such as narrow leaves, mechanical 

structures such as leaf phytoliths, and chemical toxins. In many cases involving grass poisoning it is not the grass but it’s 

parasites (usually fungi) that are the source of the toxin. If these adaptations are possessed by an exotic species but not the native 

species, and are effective against predators in the invaded system, then the invader has a novel weapon that facilitates its 

invasion. 

Allelopathy 

Chemicals produced by invasive plants while growing or decomposing may also have detrimental effects on other plants – the 

plant possesses allelolpathic properties (Gill and Davidson 2000). The study of chemical interactions between organisms is still 

in its infancy, as has been amply demonstrated by recent major advances in the study of tri-trophic interactions between 

predators, their herbivores and host plants, and chemical communication between individuals within a plant population (Baldwin 

et al. 2002, Reddy and Guerrero 2004). Plants release a complex range of complex organic compounds into both the soil and the 

air that enable above- and below-ground communication between individual plants in a population and probably between 

populations, and that influence a range of other plants and animals in the environment. 

Simple allelopathy between plant species is known to be widespread, but in most cases the precise chemicals are unknown and 

the effects have been determined using plant extracts or residues (Gill and Davidson 2000). Probably all plants are more or less 

allelopathic (Gill and Davidson 2000), and possibly all plants also release chemicals that are beneficial to other plants. 

Centaurea maculosa Lam. (Asteraceae) in North America is probably the best studied example of allelopathy (Ridenour and 

Callaway 2001). C. maculosa root exudates reduce the plant size and root elongation rates of a native tussock grass of invaded 

areas, Festuca idahoensis Elmer by half, and allelopathy accounts for the major proportion of the total interference of C. 

maculosa with the grass. The allelopathic activity of the litter of Vulpia spp. (Poaceae) against crop and pasture plants has been 

demonstrated in laboratory and glasshouse experiments in Australia (Gill and Davidson 2000). Shoot extracts of different 

cultivars of wheat Triticum aestivum L. have been found to have startlingly different impacts on radicle elongation of another 

grass Lolium rigidum Gaud., that is a major weed of wheat crops in Australia (Lemerle and Murphy 2000). The rapid spread of 

the African grass Eragrostis plana Nees in southern Brazil is due in part to its allelopathic effects (Overbeck et al. 2007). 

Similarly, aqueous leachates of seeds, roots and leaves of the African grass Brachiaria decumbens Stapf, invasive in Brazilian 

cerrado, have been found to reduce germination of potentially competing plants (Barbosa et al. 2008). Allelopathic activity of 

Sorghum halepense (L.) Pers. partly explains its success as an invader in the USA (Rout and Chrzanowski 2009). Invasive 

European plants present in North America have greater allelopathic effects on North American native plants than European 

natives, and on the continental scale chemical co-adaptation amongst members of native plant communities may possibly be 

commonly disrupted by alien invaders (Callaway and Maron 2006). 

Allelopathic effects of species in the Stipeae appear to have rarely been investigated. Ruprecht et al. (2008) found that leaf 

leachate of Stipa pulcherrima C. Koch, a dominant species in abandoned continental European grasslands, reduced seed 

germination, radicle elongation and delayed germination of co-occurring species. 

14

As yet there appears to be no evidence that N. neesiana possesses allelopathic properties. However N. neesiana does have a 

unique phytolith profile (see below) that may confer more robust resistance to herbivory. 

Rapid evolution 

Evolutionary aspects of invasive species have been poorly explored and are little understood (Lee 2002, Callaway and Maron 

2006, Whitney and Gabler 2008). Removed from the environment in which they evolved, new immigrant plant populations are 

subject to new selection regimes, founder effects, genetic drift and new hybridisation possibilities, so should be “prime 

candidates for ... evolutionary changes”, and there is now substantial evidence of rapid evolution in populations of a diverse 

array of invasive plants (Whitney and Gabler 2008 p. 570). Many grasses set seed in their first year, and most are capable of 

reproducing when two years old, so invasive grasses are capable of more rapid evolution than many other plants. 

Strong new selection effects have demonstrably led to rapid evolution of weeds (Callaway and Maron 2006). Herbicide 

resistance is one of the most troublesome recent manifestations, and has developed in a large number of grass species (Preston 

2000). Similarly genetically based “adaptive breakthroughs” (Cox 2004 p. 61) have sometimes been responsible for relatively 

inocuous exotic species becoming serious inavaders. Some species have become invasive after rapid evolution of new genotypes 

with altered seed dormancy or earlier reproduction (Cox 2004), by hybridisation and other mechanisms. Apart from alterations 

due to deliberately imposed anthropic selection (weed management, including herbicides and biological control), evolved trait 

changes within 20 or fewer years are documented for several species (Whitney and Gabler 2008). 

Plant invasions typically result from one or few founder events, so the genetic variance of the invasive populations is usually 

surmised to be much reduced compared to populations in the native range. If the population remains small over several 

generations, genetic drift can result in loss of further variation (allelic diversity and heterozygosity), and the population is said to 

pass through a genetic bottleneck. Inbreeding species are likely to lose more genetic variation during a population bottleneck 

than outbreeding species. Inbreeding depression is uncommon in Poacaeae (Groves and Whalley 2002) so genetic bottlenecking 

may rarely have any impact on their invasions. But genetic bottlenecks also may reduce the potential benefits of genetic 

outcrossing and favour self-fertilisation (Cox 2004). Combined with strong selection, the surviving population may become 

highly adapted to the new environment and can strongly differ, genetically and phenotypically, from the conspecific native 

population (Callaway and Maron 2006). Superior competitive abilities, such as larger stature or greater vigour, may be acquired 

via evolutionary ‘trade-offs’ involving the loss of traits, such as predator defences, that no longer provide an advantage 

(Callaway and Maron 2006). 

Characters that increase dispersal ability are likely to be favoured at the expanding periphery of the invaded range (Cox 2004). 

Thus the evolution of higher levels of self-fertilisation, apomixis and vegetative reproduction are frequent in invasive species, 

partly because their small initial populations restrict opportunity for out-breeding (Cox 2004). Watsonia meriana (L.) Mill., var. 

bulbillifera (J.W. Matthews and L. Bolus) D.A. Cooke is rare in its native South Africa, but is by far the dominant form of the 

plant in its introduced range in Australia (Cooke 1998), where the main means of dispersal appears to be movement of the 

vegetative cormils, produced in clusters on culm nodes, along roadsides by graders and other machinery (Parsons and 

Cuthbertson 1992). In this case, a rare native form with high vegetative dispersal ability has become the dominant form in the 

invaded range. In contrast, if the invaded system is severely geographically constrained, an invasive species may evolve reduced 

dispersal ability that minimises the loss of propagules in sink areas (Whitney and Gabler 2008). 

Increased levels of out-breeding are also recorded in invasive species, possibly because selection regimes in the new 

environment favour some of the recombinant genotypes (Cox 2004). Outbreeding species have high levels of cross fertilisation 

between individuals, so have a more diverse array of phenotypes, theoretically capable of occupying a wider range of habitats. 

Invasion by obligatory outbreeder is constrained, however, because a breeding population requires multiple individuals. 

It is widely recognised that invasion success may depend on the genetic substrate of the source populations. Single source 

introductions of a limited number of individuals are often assumed, so founder populations are thought to have passed through 

genetic bottlenecks, but multiple introductions from multiple source populations may be more usual (Petit 2004). For example 

Echium plantagineum L. populations in Australia seem to be the result of the mixing of genetic material from different European 

sources (Petit 2004). Genetic drift acting alone on the founder population can result in successful invasion, but this is probably 

exceptional. 

Broad environmental tolerance and genetic plasticity in the founder populations are often invoked as the mechanisms behind 

successful invasions, but do not stand up to examination, although possession of high levels of additive genetic variance (i.e. 

variance related to phenotype) for invasive traits in source populations has been demonstrated in a number of studies (Lee 2002). 

Simple directional selection on such traits is likely to explain many successful invasions, with genotype x environment 

interactions in the invaded resulting in diversification of the available phenotypes (Lee 2002). Lag phases, which precede the 

expansion phase of a new invasive organism (Shigesada and Kawasaki 1997), are associated with so-called ‘sleeper weeds’ in 

Australia (Groves 1999, Grice and Ainsworth 2003) and may be attributable to the slow accumulation of such variance (new 

mutations, etc.). Epistatic genetic variance (involving interaction between genes) could also generate new phenotypes on which 

selection could act (Lee 2002). 

There is strong evidence that polyploidy increases the colonising ability of plants (De Wet 1986). Many invasive plants are 

allopolyploid (hybrids retaining chromosomes of both parent species), and hybridisation (inter- or intra-specific) can generate 

more highly invasive genotypes (Lee 2002). Grasses in particular display high levels of hybrid and polyploid speciation. The 

invasive grasses Sorghum halepense (L.) and Bromus hordeaceus L. are both the result of hybridisation, the latter with later 

chromosome doubling (Cox 2004). Spartina anglica C.E. Hubbard, a sterile amphidiploid (tetraploid) evolved by chromosome 

doubling from S. X townsendii H. and J. Groves (Cox 2004), and is a more vigourous species that apparently displaces it in 

Victoria (Walsh 1994), as well as native Spartina spp. in many areas of the Northern Hemisphere (Cox 2004, Petit 2004). S. 

anglica is extremely geneticallydepauperate, unlike most allopolyploids which have large diversity because of the multiple 

15

origins of their parents (Petit 2004). The normally strongly outcrossing S. alterniflora Loisel., introduced from the east coast of 

North America to San Francisco, rapidly evolved high rates of self-fertilisation in its new evironment (Cox 2004). Stebbins 

(1972) recorded the invasive nature of an artificial Ehrharta erecta Lam. autopolyploid he created by colchicine treatment and 

released on the Berkeley campus of the Universityof Califorina. Stipeae as a whole apparently consists largely of species 

resulting from frequent and widespread hybridisation of divergent elements (Johnson 1972, Tsvelev 1977). 

Single genes or a few genes might effect invasiveness or weediness traits. Sorghum halepense, one of the world’s worst weeds 

(Parsons and Cuthbertson 1992), has few genes affecting weediness that distinguish it from non-invasive grain Sorghum spp. 

(Lee 2002). When S. halepense (tetraploid) pollinates the diploid Grain Sorghum, S. bicolor (L.) Moench, sterile triploids are 

produced that are weedy in successive crops (Parsons and Cuthbertson 1992). 

Selection pressure after naturalisation can produce phenotypes with altered morphology, physiology and phenology, or with 

greater plasticity in response to environmental variables (Lee 2002) and these changes can occur within periods as short as a few 

plant generations (Cox 2004). For example invasive populations of Echinochloa crus-galli (L.) P. Beauv. in Canada have 

evolved greater catalytic efficiency of some enzymes, which compensates for the poor adaptation of their C 4 photosynthetic 

system to the cold climate (Lee 2002, see her references). Selection can occur in response to environmental gradients, the 

resident biota and control activities (Lee 2002). Weed mimicry of crop species e.g. by E. crus-galli (Lee 2002), is another 

example of relatively rapid evolutionary adaptation, well known in grasses (Barrett 1983). 

Release from predation and competition in the invaded environment removes some selection pressures on the invading plant and 

may release characters associated with defense mechanisms from evolutionary canalisation (phenotype limitation due to the 

constraints imposed by developmental pathways) and result in rapid evolution (Lee 2002). This has occurred with Silene latifolia 

Poir. in North America which has apparently allocated resources, previously used in defence, to enhanced reproduction (Whitney 

and Gabler 2008). Rapid evolution can also occur if plant predators and specialist herbivores expand or shift their host 

preferences to consume invasive plants (Cox 2004). 

Complex patterns of evolutionary change should be expected in each particular invasion, with particular traits favoured at 

different stages of the invasion, and possible reversals of trait changes (Whitney and Gabler 2008). 

Complementary evolutionary changes may also be expected in the invaded community – the more serious the invader, the 

greater the selective pressure it imposes – and those that have been investigated can also occur rapidly, on timescales of

experimental manipulation of species diversity, or by assembling simple experimental communities, have given conflicting 

results (Prieur-Richard and Lavorel 2000). Dukes (2002) assembled combinations of native and naturalised species from four 

functional groups (annual grasses, perennial grasses, early season forbs, late season forbs) in grassland microcosms and seeded 

them with the invasive Centaurea solstitialis. He tested single species, 2 combinations of 4 spp., and one combination each of 8 

and 16 spp., with all combinations containing an equal number of species from each functional group. Above-ground biomass of 

C. solstitialis decreased rapidly with increasing diversity but reached an asymptote of c. 100 g m -2 with only c. 4 species. C. 

solstitialis allocated significantly more of its biomass to reproduction in 1 year-old communities than in newly established 

communities. The resident species produced less biomass in the 1 year old compared to the newly established microcosms, 

corresponding with increased dominance of C. solstitialis. This was contrary to the expectation that the increased resource 

availability due to disturbance in the newly established microcom would favour the weed. The species most effective in 

suppressing C. solstitialis growth was, like the weed, an annual late-season forb. These findings suggest that at the community 

scale, diversity reduces invasibility by increasing competition for limited resources, either because of the presence of individual 

competitive species or as a collective response of the resident species. Dukes calculated “impactibility”, as the percentage 

change in biomass of resident species divided by invader biomass and found that as species richness increased communities 

became more impactible but less invasible. 

More recent investigations of invasibility have usually explicity incorporated extrinsic properties that relate to human 

environmental impact. Gassó et al. (2009) found that invasive plant richness in mainland Spain was significantly positively 

correlated with the proportion of built-up land and the length of roads and railways in an area, and negatively correlated with 

distance from the coast, altitude and annual rainfall. The factors negatively correlated with invasibility reflect in part the level of 

anthropogenic disturbance. Thus land that has been impacted by human activities is generally more highly invaded and more 

invasible than semi-natural or natural areas. 

Melbourne et al. (2007) proposed that invasibility of a community is dependent on its temporal, spatial and invader-driven 

environmental heterogeneity, ‘invader-driven’ heterogeneity encompassing the effects of the invader itself on the environment. 

Thus small, relatively homogeneous areas are more ‘species saturated’ and have lower invasibility than large, more 

heterogeneous areas, which are able to accomodate more invasive species without losses of natives; and a successful invader can 

modify the invaded environment, altering its invasibility by other species. Environmental heterogeneity theory can be viewed as 

a broadening of the theory of fluctuating resources (Kreyling et al. 2008). 

Dunstan and Johnson (2006) argued convincingly that the spatial scale of a community is a critical variable in determining its 

invasion resistance. Where a community occupies a small area, the variability within the area decreases with increasing species 

richness, but when the area of a community exceeds a critical size, increasing richness increases variability. This pattern results 

from the well-known vulnerability of small populations to extinction through stochastic factors. The invasibility of the system is 

strongly dependent on its variability – less variable communities being more invasion resistant. Their approach partially 

reconciles a number of competing theories but also presages “a much larger continuum of possible relationships between 

richness, stability (both persistence and resilience), invasion resistance, species invasion/extinction, and area than have 

previously been explored” (op. cit. p. 2849) 

Nevertheless the invasibility of an area in relation to a particular invasive plant depends on proximity to invasion sources, the 

availability of dispersal mechanisms or vectors that can deliver propagules into the community, and the existence of suitablyresourced 

patches or openings in which the organism can establish, survive and reproduce (Hobbs 1989). Thus invasibility has 

been demonstrated to be influenced by factors such as landscape situation, edge effects and the size and type of community 

(Morgan 1998d), and is increased by disturbance and high resource availability (Hobbs and Heunneke 1992, Levine et al. 2003). 

For example, in studies of urban open forest and woodland in Sydney, King and Buckney (2001) found the highest number of 

exotic species in the soil seed bank and the above-ground vegetation was at the edges, that the vegetation was a very poor 

indicator of seed bank contents, 84% of the exotic species not being present in it, and that lack of suitable conditions (nutrient 

enrichment or other disturbance), rather than lack of propagules, was probably restricting the establishment of the exotics in 

areas away from edges. Structure and density of vegetation may restrict propagule entry, and integrity of the soil crust may 

restrict invasion, despite nutrient addition (Hobbs 1989). Predators, pathogens and competitors in the community may confer 

invasion-resistance, rather than plant species or the vegetation community, while symbionts and mutualists may act as invasion 

faciltators (Davis et al. 2000). 

Increased susceptibility to invasion in general has been found in areas with strong, temporally-varying change that creates 

abundant under-utilised resources, or to which such resources are anthropogenically supplied in short-term fluxes (Rejmánek 

1989, Davis et al. 2000, Cox 2004). Areas without such fluxes appear to be generally less invasible the greater their plant species 

or functional group richness at small spatial scales (Cox 2004, Melbourne et al. 2007). The environments with greatest plant 

diversity are generally the most nutrient impoverished, and may therefore be the most susceptible to anthropogenic nutrient 

enrichment and consequent increase in invasibility: this may in part explain Lonsdale’s (1999) findings that high plant diversity 

is associated with greater invasibility (Davis et al. 2000). Increased plant diversity may confer invasion resistance only in highly 

stable environments subject to very limited disturbance. 

In Australia as elsewhere in the world, temperate native grasslands communities that lacked co-evolved large, herding ungulate 

graziers have proved to be highly invasible by exotic plants when subjected to continuous grazing by introduced livestock 

(Crosby 1986, Mack 1989). The effects of livestock movement and on nitrogen cycling are important factors (Milton 2004). In 

the temperate grasslands of south-eastern Australia, both physical and chemical soil disturbance, and particularly nutrient 

enrichment, can increase invasibility at patch, community and landscape scales (Morgan 1998d). When not overgrazed, speciesrich, 

high quality grassland is in general less weed-invasible (Beames et al.(2005 citing Hector et al. 2001). Appropriately 

managed (frequently burnt or conservation grazed) Themeda triandra grasslands are more resistant to N. neesiana invasion than 

those that are poorly managed or where the dominant Themeda triandra is allowed senesce (Hocking 1998). When T. triandra 

dies, “areas of dead grass are quickly occupied by exotic species, against which a healthy T. triandra sward provides 

considerable defence” (Lunt and Morgan 2002 p. 183). At the patch scale, resistant grasslands tend to have high cover of healthy 

T. triandra. In a T. triandra grassland at Evans St., Sunbury, burnt 9 months before surveying, a high cover of weeds (i.e. >40%) 

17

correlated with significant reduction in native plants, but high native cover had no such relationship (Morgan 1998d). However 

native plant species richness per se does not appear to inhibit exotic plant invasion in these grasslands (Morgan 1998d). Damage 

to the cryptogam crust appears to be one factor that facilitates invasion (Scarlett 1994), presumably because it enables more rapid 

seed burial and therefore decreases exposure of exotic seeds to fire and predators (Morgan 1998d), but possibly also because of 

the nutrient fluxes that result. 

At the landscape scale, high levels of fragmentation by roads and proximity to agricultural land and urban areas makes native 

grassland more invasible (Williams 2007).These factors are important drivers of propagule pressure, that increase movement of 

seeds from established weed infestations into the surrounding landscape (Melbourne et al. 2007). Similarly, the factors with the 

most influence on the extinction of native plants from grassland remnants in the Victorian basalt plains have been found to be the 

road density of the surrounding landscape, and long intervals between fires, which are general indicators of habitat degradation 

(Williams et al. 2006), and probably represent good indicator measures of the invasibility for such grasslands. 

Apart from factors that affect colonisation pressure, the extent and nature of resource fluctuation is probably the main 

determinant of what species invade (Morgan 1998d). In effect, exogenous disturbance is “a special case” of environmental 

heterogenity (Melbourne et al. 2007 p. 78) – the greater the severity and range of such disturbances, the greater the invasibility 

of the area. 

Impact 

Impact has been defined as the effect that an invader has on the invaded system once established (Melbourne et al. 2007), 

however this approach ignores effects associated with the establishment phase, and longer term effects post-establishment. 

Immediate impacts during the establishment phase might be substantial (e.g. lethal toxicity to animals) and long-term effects 

might include changes to the abiotic environment (e.g. changes to soil pH). The approach also distances any analysis from 

whatever management is targetted at the invader, and ignores effects that persist after its eradication or removal (Mgobozi et al. 

2008). Furthermore such an approach glosses over the complexities of evolutionary change that are likely to occur in the invader 

and the invaded system as they interact (Whitney and Gabler 2008). Since very few plant invaders are ever eradicated, ultimately 

both the invader and the invaded system adapt to accomodate each other, and since there is more adaptive potential in the 

community that the single invader, the impact will eventually decay, or, if these adaptive changes are themselves considered to 

be impacts, may continue to slowly increase (Fig. 1). 

Impact 

Time 

Figure 1. Hypothetical impact of an invasive species on the invaded community over time. Overall impact rises at some rate 

during the establishment and spread phases, then the rate of increase of impact declines as the invader occupies all suitable 

habitat and areas. Impact reaches a plateau after which it either slowly declines, as previously hidden effects of adaptation and 

evolutionary compensation in the invaded community gradually reduce the effects, or continues to slowly increase if these 

compensatory changes are themselves considered to be components of impact. 

Impacts of invasive plants can be negative, positive or neutral, and in a particular invasion are typically composed of a mixture 

of such effects on different system components; so perceived impact on biodiversity is highly dependent on the measures of 

biodiversity that are assessed (Groves 2002). Effects can be identified at hierarchical levels from genetic, through individual to 

community and ecosystem, and can include extinction of other species, reduced abundance of native species and facilitation of 

other species. At one extreme, mere presence of an alien plant in a natural ecosystem may not be tolerated by humans (Groves 

2002), even if the plant is accomodated in the system without other apparent negative effects. 

Transformer species 

Plants that cause major and permanent (or difficult to reverse) changes in invaded communities have been called “transformers” 

(Henderson 2001) or “ecosystem engineers” (Byers et al. 2002), although the latter term is usually applied to animals (Cox 

2004). Transformer species often have a habit, life form or phenology not present in the invaded community (Woods 1997). 

Henderson (2001 p. 253) defined a transformer as a plant which can “as monospecies dominate or replace any canopy or 

subcanopy layer of a natural or semi-natural ecosystem, thereby altering its structure, integrity and functioning”. The definition 

is that of Swarbrick (1991), but not the terminology. 

Henderson (2001) contrasted ‘ruderal’ and ‘agrestal’ weeds, which “invade mainly sites of severe human disturbance”, with 

other invasive plant species – the implication being that invasion by transformers does not require such disturbance. Rather, such 

18

species tend to be those which transform the system invaded by rapidly modifying the prevailing disturbance regime (Richardson 

and van Wilgen 2004). Transformer species often have disproportionately large effects compared to other weeds because they 

become numerically dominant in the invaded area or dominate the biomass. Good examples of transformer species are Mimosa 

pigra L. in Melaleuca woodlands in northern Australia (Braithwaite et al. 1989), Tradescantia in New Zealand forests (Kelly 

and Skipworth 1984) and Pinus and Acacia species in South Africa (Versfeld and van Wilgen 1986). Transformer grasses 

include ‘land builders’ such as Spartina (Gray et al. 1997), and Andropogon gayanus (Kunth) in northern Australia, which has 

massive stature in comparison to native grasses and greatly enhances the intensity and frequency of fires (Rossiter et al. 2003, 

Ferdinands et al. 2006). 

Impact of N. neesiana 

N. neesiana does not neatly fit the definition of transformer species in Australian native grasslands. It dominates the canopy in 

many invaded native grasslands, although its morphology, biomass and phenology is similar to some of the major native grasses 

that it replaces. The invaded systems remain as grasslands, and other effects it may cause are poorly known. 

The biodiversity impacts of N. neesiana in temperate Australian natural grasslands is probably critically determined by its 

presence and impact in the cultural steppe that surrounds and fragments the remaining indigenous remnants. The success of N. 

neesiana in the cultural steppe drives propagule pressure, so determines, in part, the susceptibility of grassland remnants to 

invasion. Their invasibility is also strong influenced by their health or degradation state, which is critically influenced by 

management. The prevalence of major anthropogenic disturbance and the disruption of ‘natural’ disturbance patterns appear to 

be central issues. A range of other factors most likely contribute to the success of N. neesiana in native grasslands and its 

biodiversity impact, including release from natural enemies. 

19

Nassella neesiana 

“Over the past few years is has become increasingly clear that the introduced grass Stipa neesiana is a serious threat to remnant 

stands of native grassland ... and can almost completely displace perennial native grasses, including the dominant Themeda 

triandra ... Prior disturbance does not appear to be necessary for invasion ... We request that urgent attention be given to control 

of the plant ...” 

M.J. Bartley, R.F. Parsons and N.H. Scarlett of the Botany Department, La Trobe University, in a letter to the Victorian 

Department of Conservation and Environment, 7 May 1990 

Nassella neesiana is a long-lived, perennial, cool season (winter-spring growing), C 3 , South American, monecious, tussockforming 

grass, with flexible reproductive mechanisms and a high survival rate of all life stages (Cook 1999, Gardener et al. 

1996a 1999, 2003a, Storrie and Lowien 2003, Benson and McDougall 2005). In Australia it is both a serious enivornmental 

weed (Carr et al. 1992, McLaren et al. 1998) and a problem weed of agriculture (Grech 2007a). 

N. neesiana is a widely recognised threat to Australian temperate native grasslands. One of the earliest reports was that of 

McDougall (1987) who recorded it as a weed in native tussock grassland and Eucalyptus camaldulensis woodland in the western 

region of Melbourne. Carr et al. (1992 pp. 41, 51) considered it to be a “very serious threat to one or more vegetation formations 

in Victoria”. It has rapidly expanded its range (Lunt and Morgan 2000), is able to actively invade grasslands (Hocking 1998), 

reportedly without prior disturbance (Bartley et al. 1990), is allegedly potentially able to outcompete C 4 (summer growing) 

grasses such as T. triandra (Ens 2002a), and is rated, along with Nassella trichotoma, as the most significant weed threat to 

grassland biodiversity (McLaren et al. 1998, Groves and Whalley 2002). Trengrove (1997) considered it to be “much more 

invasive” in T. triandra remnants than N. trichotoma. Morgan and Rollason (1995) considered it to pose “by far the greatest 

threat of any potential new invader” at one grassland, Kirkpatrick et al. (1995 p. 36) classed it as “a major threat to one or more 

grassland communities”, Morgan (1998d) called it one of the “most potentially threatening species” to T. triandra grasslands and 

Lunt and Morgan (2000 p. 98) rated it as “perhaps the most serious environmental weed in remnant native grasslands in southern 

Victoria”. Humphries and Webster (1992 and 2003 p. 2) and Webster et al. (2003 p. 3) wrote that “the aggressive invasion” of N. 

neesiana at Derrimut and Laverton North Grasslands in Victoria needed “immediate attention if the values of these grasslands 

[were] to be preserved”. Ens (2005) stated that it “swamps all other ground flora and forms expansive monocultures”. According 

to Beames et al. (2005 p. 2) it is “particularly well adapted to the intensively cultivated areas surrounding urban areas and poses 

a signficant threat to mismanaged urban grassland remnants”. 

N. neesiana is a successful invader because of it wide ecological flexibility, synanthropic dispersal mechanisms, the availability 

of suitable habitats in a suitable climate, and limited biotic resistance in the invaded communities. It has many of the general 

charactersitics possessed by successful invasive species (Cox 2004): a large native range, abundance and sometimes dominance 

in its native range, high vagility (via seed), dispersal by abiotic processes, short generation time, high reproductive rate, 

ecological flexibility, wide climatic and physical tolerances, reproduction involving a single parent individual and association 

with humans. In addition many varieties and forms have been described, suggesting that the species has wide genetic variability, 

at least in South America. Bourdôt and Hurrell (1989a p. 415) attributed its invasiveness in New Zealand sheep pastures not to 

“high competitive ability” but rather to “adaptations that enable the plant to survive the hazards of semi-arid, low-fertility 

environments”. 

It is often generally contended that invasion only occurs as a result of disturbance and that Australia is more subject to invasion 

than northern hemisphere biomes (e.g. New 1994 p. citing Di Castri 1990). Exotic stipoid grasses in Australia “generally invade 

plant communities which are already highly degraded and have a history of disturbance” (Gardener and Sindel 1998 p. 76 citing 

G. Carr pers. comm.), along with “lands with higher fertility soil often previously used for grazing or farming” (op. cit.). Since 

temperate native grasslands often occupy highly fertile soils, and are communities dependent on disturbance for the maintenance 

of their natural diversity, this appears to create a significant dilemma. The disturbance necessary to create gaps in the dominant 

grass canopy, in which most of the native vascular plant diversity can flourish, simultaneously promotes exotic plants, which 

comprise the bulk of the seed bank at least at some sites (Lunt 1990b, Morgan 1998c). 

Gardener and Sindel (1998) advocated quantitative studies to evaluate the biodiversity changes associated with invasion by 

exotic Nassella species, to determine if any of these changes result from general degradation and to evaluate the impact of 

Nassella management techniques on the promotion or inhibition of biodiversity. To date, only a single study of N. neesiana, with 

a narrow focus, has been undertaken, that of Ens (2002a). She found that diversity of insects declined in areas occupied by the 

plant in Sydney woodlands, although some groups benefited. Monitoring of biodiversity is necessary to provide a sound basis for 

evaluating the techniques and strategies used to manage N. neesiana as a weed (Grice 2004a). Mangement measures are 

currently focused on minimising seed production using herbicides or grazing in agricultural (Grech 2007a) and natural areas, and 

managing the rate of mineralisation of nitrogen to favour C 3 grasses in natural ecosystems (Craigie and Hocking 1998, Hocking 

2002 2005b, Groves and Whalley 2002). However there are concerns that some current approaches may be counterproductive, 

producing grasslands that are more impoverished, and more highly invasible. 

The success of a plant invader depends on its biological attributes, the attributes of the community or ecosystem invaded and the 

effects of human activities. These factors are examined in detail in the review below. 

20

Taxonomy and nomenclature 

Stipeae 

Nassella neesiana is a member of the Poaceae: Stipeae, “a cosmopolitan tribe ... widely distributed” with “major centres of 

diversity in South and North America, Australia and Eurasia”, “primarily in temperate or warm-temperate regions” and 

“dominant in many of the arid grasslands of southern Australia, South America and Asia at varying elevations (0-5000 m)” 

(Arriaga and Jacobs 2006). The tribe contains “approximately 500” (Vásquez and Barkworth 2004 p. 484) or “c. 450” (Arriaga 

and Jacobs 2006) species. Simon (1993) erected a new subfamily Stipoideae for the tribe, a classification adopted by Watson and 

Dallwitz (2005), but not followed by others (e.g. Briese and Evans 1998). 

In the past Stipeae has been included with Poeae in a festucoid group, close to Miliceae, Diarrheneae and Nardeae (Tsvelev 

1984). The tribe was sometimes assigned to subfamily Arundinoideae (e.g. Barkworth and Everett 1986), and sometimes to 

Pooidae (Wapshere 1990) e.g. by Edgar et al. (1991), but molecular evidence indicated that it did “not sit comfortably in either 

subfamily” (Briese and Evans 1998 p. 94) and was not clearly delimited (Barkworth and Torres 2001). Zucol (1996) argued that 

phytolith leaf assemblages in Nassella species indicated an affinity with Arundinoideae, while Honaine et al. (2006) considered 

their phytolith study of Nassella and Piptochaetium spp. supported the assignment to Stipoideae. Molecular data currently 

indicates that Stipeae is monophyletic (Jacobs and Everett 1996) and it is now often considered a basal lineage within Pooidae 

(GPWG 2001, Arriaga and Jacobs 2006, USDA ARS 2006), or the largest of six tribes in Stipoideae, along with the monogeneric 

Nardeae, Lygeae, Ampelodesmae, Anisopogoneae and doubtfully Brachyelytreae (Watson and Dallwitz 2005). 

The tribe is best defined by characters of the embryo (Jacobs and Everett 1996) which are correlated with a set of characters not 

exclusive to the tribe: florets with a single spikelet, a coriaceous or firmer lemma with comparatively large unicellular 

macrohairs (Arriaga and Jacobs 2006), disarticulation of the seed above the glumes and absence of a rachilla extension (Jacobs 

and Everett 1996) (i.e. the rachilla is not prolonged (Jacobs et al. 1989)), a well-developed callus, and a terminal, usually 

articulated awn (Barkworth 1993). Stipoids are commonly wiry, “bamboo-like” perennial grasses (Jacobs et al. 1989 p. 570) and 

generally have a leafless, paniculate infloresence and a lemma that is tougher than the glumes (Barkworth and Everett 1986). 

For many years a large proportion of Stipeae were considered to be included in a broadly defined genus Stipa. From the late 

1970s taxonomic work led to the resurrection of old names and reassignment of species to other genera, some of very long 

standing (Barkworth and Everett 1986, Barkworth 1993, Jacobs et al. 2000, Barkworth and Torres 2001, Vásquez and Barkworth 

2004). The concept of Nassella (Trinius) E.Desvaux was expanded to include species with long florets, formerly in Stipa sens. 

lat. Barkworth (1990) recognised nine genera in Stipeae: Achnatherum, Anemanthele, Hesperostipa, Nassella, Oryzopsis, 

Piptatherum, Piptochaetium, Ptilagrostis and Stipa. Several additional genera are now recognised in the tribe: Jacobs and Everett 

(1996) assigned all Australian native species formerly included in Stipa to the new genus Austrostipa and provided a key to the 

then ten genera of Stipeae; Peñailillo (1996) described a new genus Anatherostipa; Jacobs and Everett (1997) resurrected Jarava; 

Torres (1997) described Nicoraella; Vásquez and Barkworth (2004) defined a new genus Celtica for Stipa gigantea Link and 

resurrected the genus Macrochloa Kunth for Stipa tenacissima (Loefl. ex L.) Kunth. Barkworth (2006) included the small or 

monotypic genera Aciachne, Lorenzochloa, Macrochloa, Ortachne, Psammochloa and Trikeraia in her world list of Stipeae. 

Watson and Dallwitz (2005) included these 15 genera plus, tentatively, Danthoniastrum (possibly Aveneae). Tsvelev’s (1977) 

concept of Stipeae also included Orthoraphium Nees, Eriocoma Rydb., Streptachne R.Br., Stephanacne Keng and Pappagrostis 

Roshev. 

Barkworth (1990) reviewed the complex taxonomic history of Nassella. The name was first used by Trinius 1830 at subgeneric 

rank. Barkworth (1990) expanded Nassella from c. 9 spp. to include 79 species, with most of the additions, including N. 

neesiana, being from Stipa sens. lat. Watson and Dallwitz (2005) probably relied on Barkworth (1990), giving a total of “about 

80” spp. for the genus. According to Barkworth and Torres (2001) Nassella included at least 116 spp., but Barkworth (2006) 

listed only 110. Quattrocchi (2006) gave an upper limit of 116. 

Barkworth (1990) considered Nassella to be most closely related to Piptochaetium and Hesperostipa. Analysis of morphological 

and anatomical characters placed Austrostipa close to Achnatherum and Ptilagrostis, but study of rDNA indicated a closer 

relationship to Nassella than to Achnatherum (Jacobs and Everett 1996). Detailed molecular and morphological examination 

(Jacobs et al. 2000) found Nassella and Piptochaetium to be sister groups; likewise for Austrostipa and Achnatherum. 

Austrostipa appears to be the most recently evolved genus in the tribe (Jacobs et al. 2000). 

Nassella is distinguished from other stipoid genera by having 3 stamens (1-3: Quattrocchi 2006); a tough lemma with three or 

more nerves and strongly and tightly overlapping margins, the outer margin extending 1/3 to 2/3 (25-50% Barkworth and Torres 

2001) of the way around the inner, an unridged surface, heavily silicified with round or oval silica bodies in the epidermis, the 

fundamental cells of the epidermis being extremely short and much shorter than wide, a solid apex often developed into a corona 

(crown) at the summit of the lemma, with cilia that often appear to be fused at the base; and a flat, veinless, usually glabrous, 

short, rudimentary palea, less than 30% as long as the lemma, and completely concealed by the lemma (Barkworth 1990, 

Barkworth 1993, Jacobs and Everett 1996, Barkworth and Torres 2001, Watson and Dallwitz 2005, Barkworth 2006; contra 

Stace 1997 p. 840: “palea 3x) as long as lemma”). 

Other common characters in the genus are the presence of both short and long anthers in a species or on the same plant or floret, 

tuberculate lemma, and abundant apical cilia on the long anthers (Barkworth 1990, Barkworth 2006). A many-noded culm with 

frequent branching is usual, but also occurs in Achnatherum and some Austrostipa spp. (Barkworth 1990). All Nassella species 

are caespitose perennials, the glumes are often strongly anthocyanic (anthocyanins = the water soluble glucoside compounds 

forming colouring matter in many flowers and other plant parts). No vegetative characters that distinguish the genus have been 

identified (Barkworth 2006). Nassella species are bisexual, mostly perennial, rarely annual, have hollow internodes and open 

sheaths and “readily deciduous” awns (Quattrocchi 2006 p. 1361). 

In the Americas Nassella species are found from southern South America to southern Canada at altitudes from 0-5,000 m. Two 

areas have particularly high diversity: the altiplano of the central Andes, and the pampas of Uruguay, southern Brazil and eastern 

Argentina (Barkworth and Torres 2001). 

21

The Nassella lineage has diverged to form two morphologically distinct groups (Barkworth 1990): one, to which N. neesiana 

belongs, with long symmetrical florets, long, sharp calluses and long, persistent awns, the other, to which N. trichotoma belongs, 

with short, eccentric florets, short, relatively blunt calluses, and relatively short, deciduous awns. Quattrocchi (2006 p. 1361) 

incorrectly attributed “readily deciduous” awns to the whole genus. 

The assignment of species present in Victoria to Nassella on the basis of three character states - lemma margins strongly 

overlapping, palea membranous and not more than one third the length of the lemma - was questioned by Walsh (1994) who 

considered these characters to be present in some Stipa sensu stricto (in which he included the native Australian stipoids) and not 

present in other species assigned to Nassella by Barkworth (1990). Austrostipa is the only other stipoid genus with strongly 

convolute, coriaceous lemmas (in some spp.), but the palea is not reduced, glabrous and unveined (Barkworth and Torres 2001). 

Six Nassella species are naturalised in Australia: Lobed Needle-grass N. charruana (Arechav.) Barkworth, Cane Needle-grass N. 

hyalina (Nees) Barkworth, Texas Needlegrass N. leucotricha (Trin. and Rupr.) Poly, Short-spined Needle-grass N. megapotamia 

(Spreng. ex Trin.) Barkworth, Serrated Tussock N. trichotoma, and N. neesiana (Jacobs and Everett 1996; McLaren, Stajsic and 

Iaconis 2004). Mexican Feather-grass N. tenuissima (Trinius) Barkworth, an ornamental grass, is present in the nursery trade and 

gardens, and is possibly incipiently naturalised (Jacobs et al. 1998, McLaren et al. 1999, Maguire 2005). 

The name Nassella is derived from the Latin nassa meaning “a fish basket” or “a basket for catching fish” (Quattrocchi 2006). 

The basionym: Stipa neesiana Trinius and Ruprecht 1842 has been variously cited as appearing in: 

Mem. Acad. St. Petersb. Ser. 6 Sc. Nat 5: 27 (Caro 1966); 

Mém. Acad. Imp. Sci. Saint-Pétersbourg, ser. 6, Sci., Math., Seconde Pt. Sci. Nat. 5:17 (Vickery et al. 1986); 

Mém. Acad. Imp. Sci. St. Pétersbourg, sér. 6, Sci. Nat. 5: 27 (Torres 1997); 

Mém. Acad. Sci. St. Pétersb., sér. 6, Sci. math. phys. and nat. 7 2 : Bot. 27 (Willis 1970); 

Mém. Acad. Imp. Sci. St-Petersb., ser. 6, 5: 17 (1842) (Jessop et al. 2006). 

Synonyms 

The taxonomic synonyms of N. neesiana are numerous. The following synyonyms have been listed by Caro (1966), Torres 

(1993), Zuloaga et al. (1994) Barkworth and Torres (2001), Quattrocchi (2006), Barkworth (2006) and Barkworth et al. (2007): 

Stipa barbinodis Philippi (1896) = S. neesiana var. barbinodis (Philippi) Caro (1966) 

Stipa contracta Phil. 

Stipa eminens Nees in Mart. (1829) nom. illeg. non S. eminens Cavanilles (1799) 

S. fernandeziana Phil. (1873) non S. fernandeziana (Trin. and Rupr.) Steudel (1854) 

S. hackeli Arechavelata 

S. hispida Phil. (1896) 

S. longiflora Steudel (1854) 

S. neesiana Trinius and Ruprecht (1842) 

S. neesiana var. chilensis Trin. and Rupr. (1842) 

S. neesiana var. fernandeziana Trin. and Rupr. = S. skottsbergii Pilger (1916) 

S. neesiana var. glabrata Arechavaleta (1896) = S. setigera forma glabrata (Arechav.) Speg. (1901) 

S. neesiana var. hirsuta Arechavaleta (1896) = S. setigera forma hispidula Speg. (1901) = S. neesiana var. hispidula (Speg.) 

Hackel (1911) 

S. neesiana var. hispidula (Spegazzini) Hackel (1911) 

S. neesiana var. longiaristata Arechavaleta (1896) 

S. neesiana var. sublaevis (Spegazzini) Speg. ex Caro 

S. neesiana var. sublaevis (Spegazzini) Speg. (1925) 

S. neesiana var. virescens Hackel (1904) 

S. neesiana forma contorta Hackel (1904) 

S. neesiana forma depauperata Hackel 

S. setigera auct. non. J. Presl 

S. setigerna sensus Spegazzini non Presl var. glabrata (Arechavaleta) Spagazzini 

S. setigera Presl var. glabrata Arechavaleta ex Spegazzini (1901) 

S. setigera forma glabrata (Arechavaleta) Spegazzini 

S. setigera Presl var. hispidula Spegazzini (1901) 

S. setigera forma hispidula Spegazzini 

S. setigera var. hispidula forma pallida Spegazzini (1901) 

S. setigera var. hispidula forma purpurascens Spegazzini (1901) 

S. setigera var. hispidula forma versicolor Spegazzini (1901). 

S. sublaevis Spegazzini (1901) = S. neesiana var. sublaevis (Speg.) Speg. ex Caro (1966) 

S. skottsbergii Pilger 

S. trachysperma Phil. (1864) 

Urachne longiflora Steudel 

Additional infraspecific taxa synonymous with currently recognised varieties are listed below. 

Ens (2005 under “Notes”) erroneously stated that Nassella tenuissima was previously known as Stipa neesiana. 

Many South American authors continue to use “Stipa” (sens. lat.) for most Nassella species (e.g. Honaine et al. 2006, Iriarte 

2006). 

22

Vernacular names 

‘Needlegrass’ is a name that has been used for the genus Stipa sens. lat. in North America (Hitchcock and Chase 1971). The 

single word form of the name, the hyphenated form and the two-word form are all used in Australia. Shepherd et al. (2001), 

perhaps the best standardised source for vernacular plant names in Australia, uses “needlegrass”. “Spear grass” and “corkscrew 

grass” are also applied as broad names for stipoids (Bourdôt and Hurrell 1989b). The term “arrow-grasses” has been applied to 

species of Stipa sens. lat., Piptochaetium and Aristida in Uruguay (Rosengurtt 1946). 

The name “flechilla” (Hayward and Druce 1919, Bourdôt and Ryde 1986, Soriano et al. 1992, Martín Osorio et al. 2000) 

(wrongly “fletchilla” (Slay 2002a)), meaning ‘little arrow’ (Gardener 1998) or ‘little dart’ (Hayward and Druce 1919, Bourdôt 

and Ryde 1986, Soriano et al. 1992), in reference to the characteristics of the seed, is used in Argentina, but also applies to other 

Nassella, Stipa, Piptochaetium and Aristida spp. with piercing seeds (Soriano et al. 1992). 

Additional vernacular names used for S. neesiana include American Needle-grass in the UK (Stace 1997), Uruguayan 

Tussockgrass in the USA (Randall 2002, Quattrocchi 2006, Barkworth 2006, ITIS 2006) and, not in common parlance, Chilean 

Spear-grass, in Victoria (Carr et al. 1992). Spanish names include “Aguja chilena” and “Hierba chilena de agujas” (Martín 

Osorio et al. 2000). 

Infraspecific Taxa 

Some infraspecific names have already been mentioned. A number of varieties have been recognised but their validity and 

usefulness is unclear: 

var. barbinodis (Philippi) Caro (1966); recognised by Caro (1966) and equivalent to 

Stipa barbinodis Philippi 

var. chilensis Trin. and Rupr. (1842) 

var. fernandeziana Trin. and Rupr. (1842) 

var. formicarioides Burkart (1969); recognised by Zuloaga et al. (1994) and Verloove (2005) 

var. glabrata Arechav. (1896) 

var. gracilior Burkart (1969); recognised by Zuloaga et al. (1994) and Verloove (2005) 

var. hirsuta Arechavaleta; recognised by Zuloaga et al. (1994), and Caro (1966) who included within it 

S. setigera sensu Spegazzini non Presl var. hispidula Spegazzini 

S. neesiana forma contorta Hackel 

S. neesiana var. hispidula (Spegazzini) Hackel 

var. longiaristata Arechavaleta (1896), recognised by Burkart (1969), Rosengurtt et al. (1970), Moraldo (1986), Torres (1993) 

and Zuloaga et al. (1994) and synonymous with 

S. sublaevis Spegazzini (1901), and 

S. neesiana var sublaevis (Spegazzini) Spegazinni, comb. superfl. 

S. neesiana var. sublaevis (Speg.) Spegazzini ex Caro (1966) 

var. neesiana; also recognised by Caro (1966) and Zuloaga et al. (1994), and including: 

var. glabrata Arechavaleta 

f. contorta Hackel 

f. depauperata Hackel 

var. sublaevis (Spegazzini) Spegazzini; recognised by Caro (1966) and including: 

S. hackeli Arechavelata 

S. sublaevis Spegazzini 

var. virescens Hackel; recognised by Caro (1966) and Zuloaga et al. (1994), considered to be a synonym of N. argentinensis by 

Barkworth and Torres (2001) but recognised as part of S. neesiana by Barkworth et al. (2007). 

Torres (1993 p. 19) discussed var. hirsuta and was unable confirm it as a variety of N. neesiana, but the synonymy was accepted 

by Barkworth et al. (2007). 

An additional taxon, S. neesiana var. ligularis Grisebach, is considered equivalent to S. ligularis (Griseb.) Speg. (1901) (Torres 

1993, Zuloaga et al. 1994, Barkworth and Torres 2001), while S. neesiana Kuntze is an illegal name for S. tenuis Phil. (Zuloaga 

et al. 1994). 

Variety determination is sometimes difficult (Anderson 2002b) and the varieties may have little or no validity, merely 

representing extreme individuals within variable populations,clinal variation, or the output of taxonomists working with limited 

specimen material. Verloove (2005) considered the infraspecific variability to be of little taxonomic value and noted that some 

supposedly distinguishing subspecific characters varied between leaf surfaces or with age on the same plant. 

Caro (1966) provided a key for the separation of vars. neesiana, hirsuta, virescens and sublaevis. Torres (1993) provided a key 

for vars. neesiana, gracilior and longiaristata. Moraldo (1986) provided a very brief description of var. longiaristata that does 

not enable it to be distinguished from other varieties. The varieties have sometimes been called subspecies in Australia (e.g. Britt 

et al. 2002, Jessop et al. 2006), without justification. 

Variety neesiana is found in Argentina, Bolivia, Brazil, Chile, Ecuador, Peru (Rosengurtt et al. 1970) and New Zealand (Jacobs 

et al. 1989). Variety longiaristata is found in Argentina and Uruguay (Rosengurtt et al. 1970, Torres 1993, Barkworth et al. 

2007) and according to Moraldo (1986) in Italy. Variety gracilior is found in Argentina (Torres 1993), as are vars. hirsuta, 

virescens and sublaevis (Caro 1966). The type of var. chilensis is from Chile, that of var. fernandeziana from the Juan Fernandez 

Islands (Chile), and those of vars. hirsuta and sublaevis from Uruguay (Barkworth et al. 2007). Var. glabrata is based on 

Uruguayan material (Barkworth et al. 2007). Var. barbinodis is found in Chile (Caro 1966). 

23

Caro (1966) illustrated a plant, ligule, glumes and seed of var. hirsuta. Burkart (1969) illustrated a whole plant and seed of var. 

neesiana, seed of var. longiaristata and the caryopsis of var. neesiana. Torres (1993) illustrated seeds of vars. neesiana, 

longiaristata and gracilior. 

Hocking (2002) suggested that more than one “type” of N. neesiana may be present in Australia. All the Australian specimens 

examined by Vickery et al. (1986) ‘seemed’ to be N. neesiana var. neesiana. Walsh (1994), who examined Victorian material, 

did not contradict this. Britt (2001) and Britt et al. (2002) stated that the varieties present in Australia were “unknown”. Gardener 

et al. (2005) considered the plants they studied on the Northern Tablelands of NSW were var. neesiana and stated that it 

appeared to be the only variety present in Australia. Jessop et al. (2006) classed all South Australian material as subsp. neesiana. 

Nevertheless, the large number of varieties and forms recognised by American taxonomists indicates the existence of 

considerable infraspecific phenotypic variation. Information at hand is inusufficient to indicate whether this variation is merely 

phenotypic plasticity in response to environmental variation or represents distinct genotypes. 

Misapplied names 

In Europe the name Stipa setigera (= Nassella mucronta) has been widely misapplied to N. neesiana and all verifiable records of 

these taxa in Europe are N. neesiana (Verloove 2005). N. neesiana and N. mucronata are easily confused (Vàzquez and Devesa 

1996). Zanin et al. (1992, cited by Gardener 1998) used the name Nassella setigera var. setigera for N. neesiana vars. gracilior, 

virescens and hirsuta in Brazil. Overbeck et al. (2007) also used the name S. setigera for what is apparently N. neesiana. Longhi- 

Wagner and Zanin (1998) used the name Stipa setigera for N. neesiana and recorded it from Paraguay. 

Hybridisation etc. 

Many groups of grasses arose by intergeneric hybidisation (Tsvelev 1984) and interspecific hybridisation is very common in 

Poaceae (Wheeler et al. 1990) and in Stipeae. According to Tsvelev (1977) all extant Stipeae are of hybrid origin and the tribe 

itself may have arisen this way. Grasses in general are highly dispersable and fecund, a characteristic of higher taxa with high 

speciation rates (Levin 2006). Wind pollination and simultaneous flowering of grass species provides much opportunity for 

cross-fertilsation (Groves and Whalley 2002). 

Sterile intergeneric stipoid hybrids are very common (Johnson 1972). Oryzopsis caduca Beal (?name: not in Barkworth 2006) is 

a hybrid between Achnatherum hymenoides (Roem. and Schult.) Barkworth and Nassella viridula (Trin.) Barkworth, and A. 

hymenoides crosses spontaneously with 11 stipoid species in the USA, including Nassella pulchra (Hitchc.) Barkworth, N. 

cernua (Stebbins and Love) Barkworth and species of Stipa, Achnatherum, and Heterostipa, producing plants that have all been 

classified as Oryzopsis bloomeri (Boland) Ricker (Johnson 1972). “In favorable years and under advantageous conditions of 

habitat disurbance hundres [sic] of individuals have been observed in some hybrid swarms” (Johnson 1972 p. 25). Barkworth 

(1990 2006) and Barkworth and Torres (2001) considered N. viridula itself to be most probably an alloploid, with Achnatherum 

and Nassella progenitors, or possibly an autoploid derivative of a common ancestor of Achnatherum and Nassella. According to 

USDA FEIS (2006), the occasional hybrids of N. viridula with A. hymenoides produce the sterile Achnella caduca (Beal) 

Barkworth. Watson and Dallwitz (2005) listed such hybrids as X Stiporyzopsis B.L. Johnson and Rogler and X Achnella 

Barkworth. 

Tsvelev (1984 p. 903) considered intra-sectional hybrids in Stipa to be “probably ... not so rare”, but inter-sectional hybrids to be 

“rarer”. Examples of the latter include S. gregarkunii P. Smirnov, reportedly a result of crossing between S. pulcherrima C. Koch 

(section Stipa) and S. caucasica Schmalh. (Section Smirnovia), and S. kopetdaghensis Czopan. (section Smirnovia), possibly a 

cross between S. caucasica Schmalh. (section Smirnovia) and S. zalesskii Walensky subsp. turcomania (P. Smirn.) Tzvel. 

(section Stipa). The presence of hybrids in S. aggr. capillata L. and in section Stipa was undoubted, and in section Barbatae a 

hybrid S. arabica Trin. and Rupr. subsp. arabica X S. hohenacheriana Trin. and Rupr. subsp. nachiczevanica Tzvel. ( S. 

hohenackerana according to Barkworth 2006) had been recorded (Tsvelev 1984). 

Hybridisation of N. neesiana does not appear to have been reported. However Verloove (2005) examined specimens from South 

America and France with characters intermediate between N. neesiana, N. mucronata (H.B.Kunth) R.W.Pohl and/or N. 

poeppigiana (Trin. and Rupr.) Barkworth. Barkworth and Torres (2001) considered N. mucronata, N. mexicana (Hitchock) R.W. 

Pohl and N. leucotricha (Trin. and Rupr.) R.W. Pohl to comprise a species complex, with intergrades of the former two species 

in northern South America and of N. mucronata with N. leucotricha in northern Mexico. Data in Britt et al. (2002) suggest the 

possibility of some gene flow between N. neesiana and N. leucotricha at Melton, Victoria. 

The new combinations of exotic and native stipoids now occurring in south-eastern Australia would appear to create new 

possibilities for hybridisation. It should be kept in mind that hybrid individuals might be found in Australia, and any suspected 

examples should be collected. Hybrids between N. neesiana races or regional populations may occur as a result of pollen flow, as 

may hybrids with other Nassella spp., or possibly with other introduced Stipeae and native Austrodanthonia spp. 

Evolutionary origin 

The evolutionary origin of the Poaceae is obscure. Grasses are presumed to have existed since the Mid-Cretaceous (c. 120 mybp, 

Mesozoic Era) on the basis of fossil leaves (Tsvelev 1984). Stebbins (1986) suggested their first appearance in the Late- 

Cretaceous, but molecular clock estimates suggest an origin about 83 mybp (Prasad et al. 2005). A minimum age of 90 mybp 

(Cretaceous) for the crown group of Poaceae has recently been suggested (Bouchenak-Khelladi et al. 2009). Thomasson (1986) 

found that the oldest definite grass fossils were from the Oligocene (c. 36-24 mybp, Tertiary Era) of North America, that fossils 

of probable grasses were also known from the Oligocene of Germany, and that Mesozoic leaf impression fossils from Europe, 

North America and Mongolia were only “possible” grasses. Jones (1999a) stated that grass fossils occur in the Eocene (45 mybp) 

in the Americas and Africa. Currently the oldest unequivocal grass macrofossils are recorded from the Paleocene-Eocene 

boundary c. 56 mybp (Piperno and Sues 2005). Bouchenak-Khelladi et al. (2009) considered a spikelet with a minimum age of 

55 mybp to represent a crown node for almost all recognised grass genera. However grass phytoliths representative of a diverse 

range of taxa have been found in 70 mybp (late Cretaceous) fossilised dinosaur dung (Prasad et al. 2005) (65-67 mybp according 

24

to Bouchenak-Khelladi et al. 2009). The earliest fossil grass pollen has been found in the early Tertiary, with doubtful earlier 

records from the Cretaceous (Thomasson 1986), and presumed records (Monoporites) from 70-60 mybp (Prasad et al. 2005). 

Grass pollen first appears in the Australian record in the Paleocene (Macphail et al. 1994), at the end of the Paleocene (c. 58 

mybp) (Keith 2004), or about 50 mybp (Keith 2004). 

Undoubted fossils, with seeds very similar to modern Stipa, Piptochaetium and Phalaris, occur in rocks dated to the mid-Tertiary 

(c. 35 mybp) long after the family first evolved (Stebbins 1972). Bouchenak-Khelladi et al. (2009) indicate that the BEP clade 

(Bambusoideae, Ehrhartoideae and Pooideae), which includes Stipeae, originated in the Paleocene (57 mypb), while the other 

major clade, including most C 4 species, is younger, originating in the late Eocene (40 mybp). C 4 physiology was probably well 

developed by the Miocene (Thomasson 1986), the oldest origin in grasses being approximately 30.9 mybp (Christin et al. 2009, 

Bouchenak-Khelladi et al. 2009). 

Tsvelev (1977 1984) convincingly argued that Poaceae first evolved in response to colder and drier conditions in mountainous 

areas with an increasingly continental climate. Stebbins (1986) argued a probable origin from Joinvillea-like ancestors 

(Joinvilleaceae) and differentiation into major families in lowland, open tropical savannahs with seasonal droughts; the non 

sequitur of a grass-less savannah going unremarked. A South American (Gondwanan) origin for the family has been suggested 

on the basis of the distribution of extant taxa, with diversification of subclades in Gondwana by the late Cretaceous (Prasad et al. 

2005). Although considerable diversification within the family took place in the mid-Miocene (Piperno and Sues 2005), most 

modern tribes probably existed by the Paleocene (early Tertiary, c. 50 mybp), probably along with many modern genera 

(Tsvelev 1984, Jones 1999a). The Stipeae probably arose from primitive Pooideae (Stebbins 1986) which existed at least as early 

as 70 mybp (Prasad et al. 2005). 

Jones (1999a) cited authors who argued the tropical origin of grasslands in areas that were cooling and developing seasonal 

aridity, in forest-savannah ecotone. Poaceae are uncommon in the fossil record until the mid-Miocene (16-11 mybp) (Piperno 

and Sues 2005) and “only became widespread 25 to 15 million years ago when cool, dry conditions kicked in” (O’Donoghue 

2008 p. 39). The unimportance of mammals, except for South American gonwanatherians, with typical grazing adaptations such 

as hypsodont teeth until the Oligocene and Miocene also indicates that grasses were a minor component of vegetation before this 

time (Prasad et al. 2005). 

Barkworth and Everett (1986 p. 261) noted that the character sets that define non-Australian supra-specific stipoid taxa do not 

occur in Australian taxa, even though Australian stipoids in aggregate possess almost all of these characters. They believed that 

Stipeae originated in Gondwana, suggesting a late Jurassic or Cretaceous (c. 135 mybp) origin for the tribe (Jones 1999a). 

Tsvelev (1977) more or less agreed, noting that the impoverished stipoid flora of Africa (excluding the Mediterranean areas) 

probably resulted from repeated long dry periods and the absence of high mountain refugia. Africa began to split from the South 

American section of Gondwana in the mid to late Mesozoic (c. 165-70 mybp) leaving Australia and South America still linked 

through Antarctica by the early Tertiary (c. 65 mybp) (Barlow 1981). Separation of Antarctica and Australia occurred in the 

Paleocene (53-50 mybp) and accelerated in the middle-late Eocene (c. 43-36 mybp) (McGowran et al. 2000). Several authors 

have favoured the Gondwanan centre of origin of Stipa (sens. lat.) including Moraldo (1986 p. 205) who placed it in the ‘suture 

zones between South American, Antarctica and Australia’. 

South America has a very rich stipoid flora, paralleled only by that of Eurasia (Tsvelev 1977). The current centre of diversity of 

Nassella is Argentina with c. 72 species, with greatest diversity in the north west, and 26 indigenous species. Uruguay has 23+ 

spp., and the greatest diversity in Bolivia and Chile is in the central Andes, adjacent to Argentina (Reyna and Barkworth 1994, 

Barkworth and Torres 2001, Barkworth 2006). The pampas or Rio de la Plata grasslands have 25 Nassella species (Gardener et 

al. 1996b). 

Whether the genus evolved in the Pampas region is not known. There are no macrofossil Stipeae known from South America 

(Barkworth 2006). Tsvelev (1977) considered Oligocene (36-25 mybp) fossil panicles from Colorado to be Stipa florissanti 

(Knowlt.) MacGinitie, and noted that the American grass specialist Agnes Chase considered them identical with the extant 

species Stipa mucronata (now Nassella mucronata (Kunth) R.W. Pohl. Barkworth). Everett (1986) accept the Miocene North 

American Berriochloa primaeva Thomasson to be the earliest stipoid fossil. According to Barkworth (1990) “nasselloid” fossils 

are present in Late Miocene-Early Pliocene (c. 13-5 mybp) deposits in the USA. These have Nassella-like lemma epidermal 

patterns, and were considered to be Nassella by Thomasson (1986). Other described fossil stipoids from the USA include 

Oligocene and Miocene Stipidium and Stipa and Oligocene Piptochaetium (Thomasson 1986). Barkworth and Torres (2001) 

appeared to accept four North American fossil Nassella species, but point out that only a single Nassella sp. is today present in 

the areas of Colorado, Kansas and Nebraska where the fossils were found. Tsvelev (1977) accepted lower Miocene fossils as 

Nassella and Piptochaetium spp. along with other grasses, indicating that prairie grasslands existed at that time. Johnson (1972) 

prematurely considered this fossil record provided geological evidence of a stipoid evolutionary ‘hot-spot’ in the high plains of 

Nebraska during the Tertiary and argued for a North American focal point of polyploidisation. 

Barkworth and Everett (1986) considered that American stipoids consisted of groups derived directly from Gondwana, such as 

Nassella (in the narrow sense) and Piptochaetium, and from a separate independent group that initially occupied Eurasia, notably 

Achnatherum. According to Barkworth (1990) citing Tsvelev (1977), the North American Stipa sens. lat. evolved from South 

American taxa, and European taxa from North American, however this is not evident in my reading of Tselev (1977), who 

argued that stipoids in South America evolved in parallel with those in North America and Eurasia over a very long period prior 

to the Pliocene, and that they first evolved around the Tethys sea between Africa and Eurasia when all the continental masses 

were joined in Pangaea (i.e. in the Triassic period c. 200 mybp). Tsvelev (1977) thought that evolution of the Stipeae in all areas 

simultaneously involved elongation of the spikelet and all its parts, and the lengthening of the awns, usually correlated with 

elongation of the glume apices to prevent premature shedding of the floret. Based on anatomical and distributional data, he had 

no doubt that Stipa sens. lat. evolved before the formation of lowland grasslands and considered Achnatherum to be the most 

primitive stipoid genus. 

25

The extant Stipeae are believed to be products of widespread hybridisation and subsequent stabilisation (Tsvelev 1977). There is 

“little doubt that the recombination of genetically divergent evolutionary lines in the Stipeae has been a factor in colonization of 

diverse habitat” (Johnson 1972 p. 26). 

Reconstruction of the eco-geological history of the Pampean stipoid areas of South America is at an early stage but some 

patterns are apparent. Since the mid Tertiary (c. 35 mybp) the Mesopotamian region of Argentina has been subjected to a marine 

transgression that flooded Pampasia, separating the Andean margin from the Brazilian-Uruguayan region, followed by a Pliocene 

(late Tertiary, 2-10 mybp) regression which allowed the development of fluvial plains with a diverse biota, then during the 

Quaternary (c. 2 mybp – 10 kybp), cooling and aridification alternated with warm to temperate humid periods, and finally in the 

Holocene (recent) there was another marine transgression (Aceñolaza 2004). In south-eastern Uruguay the Late Pleistocene (c 

10-15 kybp) sediments show high concentrations of C 3 pooid Poaceae and Asteraceae indicating cool, dry grasslands, similar to 

those at the time in southern Brazil and on the Great Plains of North America; the Early Holocene (c. 7-10 kybp) shows a 

marked transistion to C 4 panicoid grasses with wetlands, indicating warmer and wetter conditions; the Mid Holocene (c. 4-6 

kybp) showed dynamic climate oscillations from wet to dry with freshwater wetlands predominating; the Late Holocene (4 kybp 

to present) showed major increase in wetlands, and panicoid grasses, similar to southern Brazel but unlike conditions in the 

Argentine pampas which became more arid (Iriarte 2006). The cool arid periods in the Tertiary may have favoured stipoid 

speciation, while the cooler times throughout the Caionozoic have probably favoured the proliferation of stipoid dominated 

grasslands. 

The pattern of evolution of the Australian Stipeae is largely unknown. The Australian fossil record of all terrestrial plant taxa is 

absent or very poor over long periods of the Quaternary, although this period probably saw little speciation (McGowran et al. 

2000). Poaceae taxa are difficult to distinguish palynologically, the fossil pollen record is generally poor in Australia (Kershaw 

et al. 2000) and grass macrofossils are rare, although a potential Bambusites has been described from Tertiary stem impressions 

(Thomasson 1986). Phytolith analysis (see below) has potential for greater resolution, but has been little used in Australia 

(Kershaw et al. 2000). 

Morphology and anatomy 

A detailed description of N. neesiana extracted from published works has been compiled. Some descriptions of N. neesiana in 

monographs, floras and other publications are clearly stated to be of material found in the region of naturalisation (e.g. Burbidge 

and Gray 1970, Jacobs et al. 1989, Walsh 1994, Verloove 2005), while others are of material from the native range (e.g. Burkart 

1969). Original or revisionary taxonomic papers generally provide details of the material described and examined. In some other 

descriptions, particularly non-taxonomic and informal publications, it is unclear what material is being described: these often cite 

dimensions, etc. that are clearly derivative of a single earlier author. Often one anatomical description clearly contradicts another 

(notably in dimensions) and it is not clear whether this is due to errors, inherent variation in the plant material or to selective 

assessment of character states within individual specimens and within populations. Where possible, errors have been noted. 

Form and habit: “Erect, strongly caespitose with shoots swollen and close-set at base” (Jacobs et al. 1989, Edgar and Connor 

2000); lightly geniculate (Martín Osorio et al. 2000); but lacking the dense tussock form when growing in association with other 

grasses (Champion 1995). Branching intravaginal (Burkart 1969, Jacobs et al. 1989); short-leaved (Burbidge and Gray 1970); a 

robust tussock when established, “not as clumpy as Poa or Eragrostis” (Duncan 1993), consisting of “a number of independent tufts” 

(ACT Weeds Working Group 2002). Tillers profusely when grazed, forming dense (Bourdôt and Ryde 1986), wide clumps 

(Liebert 1996), that may “form a matt” (Slay 2002c p. 5); grazed tussocks are not large and resemble Festuca arundinacea 

Schreb. (Duncan 1993, Slay 2002c), Austrostipa spp. (Slay 2002c), or in New Zealand at any time of year Rytidosperma spp., in 

winter under hard grazing Sporobolus africanus (Poir.) Robyns and Tournay, and in the flowering and fruiting stage Bromus 

diandrus Roth (Slay 2002c). Areas of mature tussocks in New England Tablelands pastures have basal ground cover of c. 20% 

(Gardener et al. 2003b). Tussocks have an overall “yellowish-green” colour that contrasts with surrounding pasture (Bourdôt and 

Ryde 1986, Liebert 1996), although Snell et al. (2007 p. 10) stated that it “becomes yellow or straw like” in winter in colder 

regions, and “can be a darker green compared to most other pasture species” during early growth stages, and Slay (2002c) stated 

that in continuous pasture it may have narrower leaves and by lighter green in colour, possibly due to soil fertility and 

management practices. Mature tussocks typically have masses of dead leaves in the center and green leaves around the margin 

(Gaur et al. 2005). Heavily grazed tussocks in winter, according to Slay (2002c p. 11) “are ‘mushroom’ shaped ... with’flaggy’ 

20 cms+ long leaves extending horizontally beyond the perimeter ... of the tussock”. The flowering heads are open with drooping 

branches (Weber 2003). Masses of maturing seed heads in dense infestations are “dark green-brown” in colour (Slay 2002c p. 

12). After seed is shed, the stems change to “light green-silver” in colour (Slay 2002c p. 12), but remain green into late summer 

(p.22 and Slay 2001). The green slowly fades, and by autumn the stems are “light brown” (Slay 2002c) or straw-coloured and are 

easily dislodged from the plant (Slay 2002c). 

Like other grasses, the buds (apical meristems) are at or close to ground level and are protected under tightly enclosed leaf 

sheaths, so are more likely to survive fire and mammalian grazing (Wheeler et al. 1999). Young shoots are intravaginal (Watson 

and Dallwitz 2005), a condition that gives rise to the tussock form. In grazed situations in short pastures in the non-reproductive 

phase plants have a rather flat, almost rosette-like form. In the reproductive phase, plants are strongly upright, and consist almost 

totally of reproductive tillers, largely stalks with panicles, dead leaves and a few small living leaves. The contrast between these 

phases is extremely pronounced, and they can easily mistaken for different species. The flat, rosette-like growth stage, with 

smaller leaves, has been called the “sward form” and may be induced by slashing (Bedggood and Moerkerk 2002) and heavy 

grazing. 

Height: culm to 2 m (Jacobs et al. 1989, Edgar and Connor 2000, Slay 2002a), 30-100 cm (Hayward and Druce 1919), 60-200 

(Weber 2003), 30-140 cm (Barkworth 2006), to 140 cm (Moraldo 1986), 40-90 (Baeza et al. 2007), 1 m or more in the absence of 

grazing (Bourdôt and Ryde 1986), up to 1 m (Martín Osorio et al. 2000, Germishuizen and Meyer 2003, Snell et al. 2007 ), to c. 

26

1 m (Walsh 1994), ca. 90 cm (Verloove 2005); to 90 cm (Zanin 1998); 80 cm (Carolin and Tindale 1994), 60 cm (Stace 1997), 30- 

100 cm (Burkart 1969, Barkworth and Torres 2001), 50-120 cm (Muyt 2001). 

Leaves: As with other grasses, the leaves mature and senesce progressively from the tip to the base (Wheeler et al. 1999). Lamina 

mid to dark green (Muyt 2001), flat or loosely inrolled (Jessop et al. 2006), inrolling occurring under stress including drought 

(Snell et al. 2007); flat to convolute (Barkworth 2006, Zanin 2008), or somewhat inrolled (Walsh 1994), plane or involute 

(Verloove 2005), rolled when the plant is under moisture stress (Bourdôt and Ryde 1986), sometimes tightly (Slay 2002a); sharp 

pointed (Martín Osorio et al. 2000); basal leaves up to 40 cm long cauline leaves 20 cm long and 3 mm wide (Martín Osorio et al. 

2000); leaves in general to 30 cm long (Walsh 1994, Slay 2002c), mostly

(Barkworth 2006), 10-25 cm (Barkworth and Torres 2001), 5-30 cm (Burkart 1969), 10-25 cm (Moraldo 1986), 13-28 cm (Baeza 

et al. 2007), 14-35 cm (Zanin 2008) or 20-35 cm long (Martín Osorio et al. 2000) or up to 30 cm long, open, branches drooping, 

flexuous (Jacobs et al. 1989, Edgar and Connor 2000) or erect to nodding (Barkworth 2006), up to 40 cm long, “the branches 

breaking up at maturity” (Carolin and Tindale 1994 p. 772); loose (Walsh 1994, Muyt 2001), lax, nodding (Hayward and Druce 

1919), drooping (Muyt 2001), sometimes interrupted, to 40 cm long (Walsh 1994); rachis smooth to slightly scabrous, branches and 

pedicels with stiff hairs (Jacobs et al. 1989); pedicels 1.5-12 mm (Baeza et al. 2007); branches thin, angular, scabrid (Hayward and 

Druce 1919) or pubsecent (Martín Osorio et al. 2000); overall “very distinctive purplish colour” (Duncan 1993) or “initially ... 

striking violet ... with ....green awns” (Slay 2002c); branches 2.5-8.5 cm with 2-5 spikelets (Barkworth 2006), 3-5 spikelets (Hayward 

and Druce 1919), average of c. 15 spikelets in the aerial section and a progressive reduction in length of infloresence sections and 

number of spikelets at lower nodes (Connor et al. 1993), averages of 16 to 27.4 seeds per panicle (Gardener et al. 2003a). 

Disarticulating above the glumes (Verloove 2005). Published descriptions often lack precise descriptions of the subsidiary 

clandestine panicle sections. According to one description (Jacobs et al. 1989, Edgar and Connor 2000): “Occasionally an aerialtype 

branch may be produced at the uppermost culm node, and so far as can be judged, remains unsheathed”. This may be 

equivalent to what Slay (2002c p. 21) refers to as “another ‘emerged seed head’ [that] may develop from the top node underneath 

the terminal leaf sheath” on larger plants (see his Fig. 26). Slay (2002a) noted that such secondary panicles develop in the leaf 

sheath of all culm nodes, are progressively smaller towards the base of the plant and may occasionally produce exposed, 

strongly-awned seed. The ‘overt’ panicle is purplish in the flowering stage, becoming more silver as the seeds ripen (Bourdôt 

and Ryde 1986). Plants grown from single tillers and from seed produced a mean of 18 and 16 panicles per plant respectively 

after 6-9 months (Hartley 1994). The potential seed yield per panicle of plants in a dense sward in New Zealand was 38 (Slay 

2001). As in other paniculate grasses, the panicle matures basipetally, the uppermost spikelet developing first (Stebbins 1972). 

Flowers: 

hermaphrodite; pedicels 1-8 mm long, angled, scabrous, pubescent (Barkworth 2006), drooping (Slay 2002c); terminal panicle 

spikelets – glumes unequal (Hayward and Druce 1919) or subequal (Walsh 1994, Barkworth 2006); 10-22 mm (Barkworth 2006), 

16-20 mm (Walsh 1994), 15-25 (Moraldo 1986), 16-25 (Walsh 1998), 14-21 mm (Barkworth and Torres 2001), up to c. 2 cm (Weber 

2003), 15-20 mm long (Verloove 2005), shorter than the awn column, the upper to 15 mm, the lower to 20 mm (Jacobs et al. 1989) 

or the lower 17-20 mm long and the upper 2-3 mm shorter (Martín Osorio et al. 2000) or the lower 13-15 mm and the upper 12- 

14 mm (Baeza et al. 2007); the lower 15-17 x 0.4-1 mm, the upper 13-15 x 0.5-1 mm (Zanin 2008); 1.8-2.3 mm wide (Barkworth 

2006), narrowly lanceolate (Martín Osorio et al. 2000, Barkworth 2006), linear lanceolate (Hayward and Druce 1919), acuminate 

(Hayward and Druce 1919, Walsh 1994), produced into awn-like processes to 3 mm (Jacobs et al. 1989); 3-5 veined (Barkworth 

2006), 3-nerved (Hayward and Druce 1919, Jacobs et al. 1989, Zanin 2008), the lower 3-nerved (Martín Osorio et al. 2000), or 5- 

nerved, the upper 3-nerved, scabrous-pubescent on the nerves and/or margins (Verloove 2005), nerves scabrous (Moraldo 1986, 

Jacobs et al. 1989), central nerve prominent with rigid hairs, lateral nerves with rudimenary hairs and lightly rough (Martín 

Osorio et al. 2000); glabrous (Jacobs et al. 1989, Watson and Dallwitz 2005, Barkworth 2006); overall maroon (Cook 1999) or 

purple (Bourdôt and Ryde 1986), the lower violet, the upper clear to violet at anthesis (Slay 2002c); violet below, hyaline above 

(Jacobs et al. 1989), strongly (Walsh 1994) purplish with hyaline apex and margins (Verloove 2005), hyaline or violet (Moraldo 

1986), hyaline and dyed purple in the upper half (Martín Osorio et al. 2000), green-violet (Baeza et al. 2007), brownish with white 

margins (Hayward and Druce 1919), losing colour but retained on plant for some time after seed fall (Muyt 2001), retaining a 

pinkish tinge even after the seeds have dropped (Liebert 1996); florets 6-13 mm long, 1-1.5 mm wide, terete, widest just below 

the crown (Barkworth 2006); anthoecium (including callus, lemma body and crown, plus the concealed palea) cylindrical, 7.5-9 

mm but up to 10 mm long, diameter c. 1.2 mm (Verloove 2005) or ±cylindrical 7-11.5 mm long (Barkworth and Torres 2001) or 

fusiform 6.3-11.5 mm long, c. 1-1.5 mm wide, white or violet in colour (Burkart 1969), purple (Muyt 2001) or corona purple 

(Bourdôt and Ryde 1986), lemma white, corona violet, awn light green at anthesis (Slay 2002c); basal spikelets more or less 

cleistogamous, enclosed by uppermost leaf sheath (Burbidge and Gray 1970), other spikelets chasmogamous. [“In the Australian 

species [of Stipa sens. lat.] there are no externally visible differences between the two, but the cleistogamous spikelets usually 

have shorter stamens. Both types may be found scattered through a panicle” (Vickery at al. 1986 p. 11)]; lemma and palea – see 

description below under ‘Fruit’; ovary glabrous (Jacobs et al. 1989); lodicules (scales below the stamens and ovary, regarded as 

a reduced perianth): 2, c. 1mm (Jacobs et al. 1989), membranous, glabrous (Watson and Dallwitz 2005), hyaline, transparent 

(Baeza et al. 2007), nerveless; styles 2, plumose (Jacobs et al. 1989); ovary 1-1.5 mm, glabrous (Baeza et al. 2007); stigmas 2 

(Watson and Dallwitz 2005), white (Slay 2002c), 2-2.5 mm (Baeza et al. 2007); anthers 3, penicillate,3-3.5 (Barkworth 2006) or 

4-4.5 mm long (Baeza et al. 2007), or up to 3.2 mm long in chasmogamous flowers and in cleistogamous flowers reduced to one 

fertile 0.5-0.7 mm long and 2 sterile, 0.1-0.2 mm long (Edgar and Connor 2000), yellow (Slay 2002c); pollen grains smaller than 

in Aveneae, almost spherical (tribal characters, Tsvelev 1977). 

axillary cleistogamous spikelets - extremely variable (Gardener and Sindel 1998); the various floral parts are progressively 

reduced in number and size from the upper to the basal spikelets (Connor et al. 1993) i.e. loss of glumes and reduction of awn. 

See description under “Cleistogenes”, below. 

Fruit = ‘seed’ (Fig. 2). Caryopsis “a dry monospermic indehiscent fruit” (Sendulsky et al. 1986);( = grain, includes the hilum and 

embryo). 3.5-5 x 0.6-1 mm (Zanin 1998), 4.8-5.5 x 0.9-1.2 mm (Baeza et al. 2007), 3-5 (Barkworth 2006), 4-5 mm (Verloove 

2005) or 6-8 mm (Martín Osorio et al. 2000) long; cylindrical (Burkart 1969), obovoid (Baeza et al. 2007); tightly enclosed by 

the lemma (Bourdôt and Ryde 1986); clear-coffee coloured (Baeza et al. 2007); hilum (the scar left on the caryopsis at the point 

of attachment) linear (Jacobs et al. 1989), 2 mm long (Baeza et al. 2007); embryo small (Jacobs et al. 1989), 1.5-1.6 mm (Baeza et 

al. 2007), festucoid, 1 / 6 - 1 / 3 of the grain (for Stipeae- Tsvelev 1984), with an epiblast and a neglibible mesocotyl internode, 

without a scuteller tail (Watson and Dallwitz 2005); endosperm hard, without lipid, containing compound starch grains (Watson 

and Dallwitz 2005). 

28

Figure 2. Anatomy of the seed of N. neesiana. 

Panicle seed: (aerially exserted) consisting of cayopsis, enclosed by the palea and the lemma, the callus (woody pointed 

extension of the lemma) and the awn; weight c. 6 mg (Gardener et al. 1999; this is the fresh green seed). Lemma hardened 

(Bourdôt and Ryde 1986), conspicuously tuberculate (Verloove 2005), tubercular-scabrous (Jacobs et al. 1989), papillose 

(Barkworth and Torres 2001, Zanin 2008), coriaceous (Martín Osorio et al. 2000), or rugose-papillose, especially near the apex 

(Burkart 1969), or finely rugose-papillose, particularly near the crown (Barkworth 2006), or papillose-scabrid (Walsh 1994), with 5 

(Hayward and Druce 1919, Jacobs et al. 1989) conspicuous (Walsh 1994) or inconspicuous (Martín Osorio et al. 2000) nerves; 

lemma 6-10 (mostly 8-10) mm long excluding the corona (Walsh 1994), 8-14.5 mm long including callus (Martín Osorio et al. 

2000), about 8 mm long including callus (Burbidge and Gray 1970), c. 10 mm (Weber 2003), or 7-10-(13) mm (Moraldo 1986), 5.9- 

10 mm (Zanin 2008), or to 6 mm long (Jacobs et al. 1989); constricted below the crown (Barkworth 2006); glabrous except for the 

lower half on the veins (Verloove 2005), midveins pilose proximally, glabrous between the veins at maturity (Barkworth 2006), 

median nerve with long hairs (Jacobs et al. 1989), glabrous except near the callus and midrib (Walsh 1994), lobes minute (Jacobs 

et al. 1989); often purple (Barkworth 2006), pale brown at maturity (Walsh 1994). 

Corona (called a “crown” by Tsvelev 1977, Barkworth 1990 2006 and Barkworth and Torres 2001; a “membranous cupule” by Stace 

1997 and a “crowned collar” by Walsh 1998), a fusion of the edges of the lemma at its apex resulting in a solid cylinder (Barkworth 

1990), or “a concave platform with the [awn] joint inside” (Tsvelev 1977 p. 8), best developed on mature seed, although apparent on 

immature (Muyt 2001), a prominent ridge (Jacobs et al. 1989), usually wider than long, sides usually flaring somewhat distally 

(Barkworth 2006), 0.8-1.3 x 0.7-1.1 mm (Zanin 2008), c. 1 mm wide and high (Verloove 2005), 0.6-1.6 mm high (Burkart 1969), 1.5 

mm (Baeza et al. 2007), c. 0.7 mm (Moraldo 1986), 0.4-1.6 mm (Barkworth 2006), 0.5-1 mm (Barkworth and Torres 2001), or up to 

1 mm (Jacobs et al. 1989) long, c. 1 mm long excluding the apical spines (Walsh 1994), cupuliform, “bone-like”, constricted 

(Verloove 2005) and narrower (Martín Osorio et al. 2000) at the base, glabrous (Jacobs et al. 1989), with, at the apex, a 

conspicuous ring of cilia (Burbidge and Gray 1970) or denticulate hairs with a clearly broadened base (i.e. “conspicuously triangular” 

(Verloove 2005 p. 107)), or spreading spines 0.2-0.5 (Walsh 1994, 1998), to 0.5 m (Barkworth 2006),0.6-1.6 mm (Martín Osorio et 

al. 2000) or to 1 mm (Jacobs et al. 1989, Edgar and Connor 2000) long, the spines elongating as the seed matures (Walsh 1998). 

Violet or violet-suffused (Jacobs et al. 1989) or dark violet (Martín Osorio et al. 2000). The corona functions to ensure no 

backward movement of the seed once it has lodged in fur or penetrated soil or litter, even after the awn becomes detached (Slay 

2002a). 

Palea oval, much shorter than lemma (Jacobs et al. 1989), up to one third of the lemma in length (Martín Osorio et al. 2000), 1.5- 

2 x 0.5-0.7 mm (Baeza et al. 2007), 1.2-1.4 mm (Burkart 1969), 1.5 mm (Jacobs et al. 1989), 1-2 mm (Walsh 1994, Zanin 2008), 

1-2.5 mm long, membranous (Verloove 2005, Baeza et al. 2007), transparent (Baeza et al. 2007), hyaline (Walsh 1994, Martín 

Osorio et al. 2000, Baeza et al. 2007, Zanin 2008), glabrous (Walsh 1994, Martín Osorio et al. 2000,Verloove 2005, Baeza et al. 

2007), nerveless (Jacobs et al. 1989) but “internerve” glabrous (Jacobs et al. 1989); apex 2-denticulate (Baeza et al. 2007). 

Awn persistent (Edgar et al. 1991); “extremely strong” (Slay 2002c); robust, 60-90 mm (Moraldo 1986, Walsh 1994 1998, 

McLaren, Stajsic and Iaconis. 2004), 50-90 mm (Weber 2003, Jessop et al. 2006), 60-70 mm (Burbidge and Gray 1970), to 70 

mm (Jacobs et al. 1989), 60-80 mm (Bourdôt and Ryde 1986), 60-90 mm (Zanin 2008), up to 70 mm Slay (2002c), 6-9.5 cm 

(Barkworth and Torres 2001, Verloove 2005), 5-12 cm (Burkart 1969, Barkworth 2006), 6-12 cm (Martín Osorio et al. 2000), 44- 

70 mm long (Baeza et al. 2007) [total range of awn lengths cited = 44-120 mm]; bent above the middle (Hayward and Druce 

1919), 15-30 mm to the first bend (Walsh 1994), hygroscopic (Murbach 1900, Bourdôt and Ryde 1986), with 1 (Bourdôt and 

Ryde 1986, Jacobs et al. 1989, Edgar and Connor 2000), 1 or 2 (Slay 2002c), 2 (Burbidge and Gray 1970, Walsh 1994, Barkworth 

and Torres 2001, Martín Osorio et al. 2000, Baeza et al. 2007), often 2 (Storrie and Lowien 2003), or 2-3 (Carolin and Tindale 

1994) bends (“clearly twice-geniculate” Barkworth 2006); initially violet/green in colour, turning brown as seed matures (Slay 

2002a); column stout (Weber 2003), tightly twisted (Jacobs et al. 1989, Edgar and Connor 2000), spirally twisted (Weber 2003), 

appearing like a ‘corkscrew’ (Slay 2002c), long-hairy (Jacobs et al. 1989, Edgar and Connor 2000), pubescent (Baeza et al. 2007), 

or sub-plumose (Burbidge and Gray 1970), or “corto-pilosa” at the base (Zanin 2008 p. 90); 25 mm long, an intermediate section 

about 15 mm long that is shallowly twisted with short stiff hairs (Jacobs et al. 1989, Edgar and Connor 2000) and straight 

(Barkworth and Torres 2001, Barkworth 2006), the lower section, about one third of the lower half of the awn spiralled and 

covered with hairs (Martín Osorio et al. 2000); and a terminal section, the arista (bristle or seta), scabrid to 35 mm long (Jacobs 

et al. 1989, Edgar and Connor 2000), “usually intertwined with awns of adjacent florets” (Jacobs et al. 1989). 

29

The anatomy of the awn of Stipa sens. lat.. (including Nassella) "differs from ... most other grasses ... although there have been 

attempts to explain the twisting and untwisting motion, more work could be usefully done on the subject. The awn in transection 

consists mainly of thickened cells (fibres) with two lateral pockets of chlorenchyma. The chlorenchyma breaks down with age 

and, apparently, the awn does not twist until this breaking-down process has been initiated” (Vickery et al. 1986 p. 11). Murbach 

(1900: studies on the eastern USA sp. Piptochaetium avenaceum (L.) Parodi, formerly Stipa avenacea L.) found that the awn of 

panicle seed has an outer layer of sclerenchyma cells with a spiral structure to the cell wall, a central fibro-vascular bundle and a 

band of chlorphyllous tissue on each side of the bundle. The two outer layers consist of thick-walled mechanical cells with a very 

small lumen oriented on the inner side of the cell and surrounded by spirally arranged cellulosic material. Wetting and drying of 

cells on opposite sides of the awn produce forces that act in opposite directions, producing torsion, and both the outer and middle 

layers are responsible. The awns therefore straighten when wetted and twist as they dry out (Whittet 1969, Groves and Whalley 

2002, personal obs.). The awn twists in an anticlockwise direction (Slay 2002a). 

The awn columns of adjacent seeds often become intertwined (Connor et al. 1993, Edgar and Connor 2000) or twisted together 

at maturity (Walsh 1994) forming a tangled mass (Liebert 1996), and fall as an aggregate, losing the dispersal ability of lone 

seeds (Gardener et al. 2003a). Such seed masses may be retained by the plant long after the seeds mature (Groves and Whalley 

2002), or may be “easily detached en masse attaching themselves to clothes, machinery and animals” (Slay 2002a p. 10). 

Callus - an extension of the lemma formed by the oblique articulation of the lemma with the rachilla (Hitchcock and Chase 

1971), attaching the seed to the stem (Bourdôt and Ryde 1986); sharp (Walsh 1994, Barkworth 2006), particularly sharply 

pointed (Bourdôt and Hurrell 1989b), “extremely sharp ... when dry and mature” (Slay 2002c p. 8); approximately one third the 

length of the seed (Martín Osorio et al. 2000); 2-4 mm (Moraldo 1986, Walsh 1994, McLaren, Stajsic and Iaconis. 2004), 2.5-3.5 

mm (Jessop et al. 2006), 2.8-3.5 mm (Baeza et al. 2007), 3-4 mm (Jacobs et al. 1989), c. 4 mm (Verloove 2005), 2.5.-3.2 mm 

(Burkart 1969), 2-4.5 mm (Barkworth 2006), 3-5.5 mm long (Barkworth and Torres 2001), strigose (Barkworth 2006), oblique 

(Jacobs et al. 1989); with silky white hairs 1-2 mm long (Martín Osorio et al. 2000), or 1.5 mm long (Baeza et al. 2007), bearded 

(Barkworth and Torres 2001), hairy (Bourdôt and Ryde 1986), covered with white appressed hairs (Hayward and Druce 1919), 

villous (Burbidge and Gray 1970) or hidden by a tuft of white hairs (Verloove 2005), the hairs antrorse (Gardener et al. 2003a) and 

up to 4 mm long (measured from their lowest position on the callus to their furthest extent) (Jacobs et al. 1989). 

All the seed hairs are retrorse, i.e. pointed away from the callus tip. 

Cleistogenes: “clandestine seeds” (Bourdôt and Hurrell 1992 p. 102), “floreted cleistogenes” on “clandestine axillary spikelets” 

(Connor et al. 1993); formed at nodes “towards the base of flowering culms” (Jacobs et al. 1989, Edgar and Connor 2000), on all 

(Connor et al. 1993) or able to be produced on all culm nodes (Gardener et al. 2003a); subtended by slender transparent 

prophylls (Connor et al. 1993, Jacobs et al. 1989); 1-3 flowered, with 2 awned glumes about half the length of those in panicle 

florets (Jacobs et al. 1989, Connor et al. 1993), lower glume 1-3 nerved, upper glume with up to 7 nerves; lemma, corona and 

palea as in aerial florets (Jacobs et al. 1989), except reduced in size, awn to 25 mm, caryopsis to 4 mm, callus about 0.5 mm 

(Jacobs et al. 1989, Edgar and Connor 2000); 2.5 x 1.4 mm (Jacobs et al. 1989, Connor et al. 1993); anthers reduced to 1 fertile 

and 2 small sterile anthers, cayopsis round or planoconvex to 4 mm long (Jacobs et al. 1989); stem cleistogenes of two types – 4- 

5 mm, turbinate, long-callused and 1.4-2.5 mm and plump (Slay 2002a); Descriptions rarely have the clarity and detail necessary 

to appropriately describe the variation in cleistogenes. Cleistogenes in general are more rounded than panicle seed (Bourdôt and 

Hurrell 1992); lack a large hygroscopic awn (Gardener et al. 2003a) and have a less well-developed lemma than panicle seed 

(Hurrell et al. 1994). Upper nodes produce larger numbers, a maximum on the third and fourth nodes and an average of c. 7 per 

tiller (Gardener et al. 2003a); up to 5 “unsheathed” cleistogenes on the second and third nodes (Slay 2001 p. 28); each node 

above the basal with the potential to produce a few seeds (Gaur et al. 2005); a potential total of 13 cleistogenes (including the 

basal) per culm (Slay 2001). Thin stems may have more cleistogenes than thick ones, possibly indicating a compensatory effect 

for reduced panicle seed production of thin stems (Julio Bonilla pers. comm.). Mean mass of upper node cleistogenes 2.0 mg, 

maturity on average 4 weeks after maturation of panicle seed (Gardener et al. 2003a). Basal cleistogenes initiated as a bud at the 

base of the culm, present even in small 3-tillered plants (D. McLaren, 26 October 2006, based on observation of Shiv Gaur); 

solitary (Burkart 1969, Gaur et al. 2005), “always singular” on the Northern Tablelands of NSW but “often in multiples” in 

Argentina (Gardener et al. 1996b), successively develop one on top the other, so accumulate upwards (D. McLaren, 26 October 

2006, based on observation of Shiv Gaur); 1-3 (Slay 2001), or 1-2 (Connor et al. 1993), less than half the basal nodes producing 

one cleistogene on average (Gardener et al. 2003a); occurring beneath the soil surface (Gaur et al. 2005) or often below ground 

(Gardener et al. 2003a); nut-like, 2..5 mm long, 1.5 mm wide, yellow to dark brown depending on age (Slay 2002a), light dull 

yellow when newly formed, becoming brown and thin as they mature (Gaur et al. 2005), lemon to fawn colour, plump, with the 

top “‘Turkish roof’ shaped” (Slay 2001 p. 42); mean mass 3.3 mg (Gardener et al. 2003a). 

Roots: fibrous, crown thickened (Muyt 2001), not rhizomatous (Barkworth 2006). 

In the context of the flora of Victoria, identification to species from macroscopic vegetative characters alone is unreliable, but 

can possibly be achieved by microscopic examination of epidermis morphology and distribution of stomata and silica bodies 

(Walsh 1998). The detail provided by Watson and Dallwitz (2005) indicate this is possible, at least for distinguishing the Nassella 

species present in Australia. Identification using leaf phytolith character assemblages also appears possible (see below). 

Evolved changes, which may be reflected in morphology, and perhaps enhanced growth, can be expected in introduced 

populations of exotic plants, but these changes can be difficult to distinguish from genotypic plasticity (Cox 2004). There is a 

need to determine if there are any morphological differences between exotic and native populations and between exotic 

populations in different areas. Possibly some evidence for differences exists in the morphological descriptions above. No 

morphological descriptive publications examined provides sample sizes or standard deviations for any measurements, nor any 

evidence that the material described is truly representative of the populations. Evolved changes are likely to be evident in the 

diaspore and other reproductive characteristics (Cox 2004). For example populations of Pinus concorta ssp. latifolia at the 

exanding edge of their post-glacial invasion front have smaller, better dispersing seed than the core population (Rejmánek and 

Richardson 1996 see their citation). A comparison of the seed morphology of core and fringe populations would be of interest: 

30

are the seeds larger/smaller, longer/shorter awned, less/more hairy, and how might this relate to increased dispersal efficiency? Is 

there evidence of higher cleistogene production in Australia, e.g. does Burkart’s (1969) ‘solitary’ indicate a true difference 

between populations in the core native range and with the Australian and New Zealand populations? 

Phytoliths 

The epidermis of grasses includes specialised small cells known as silica cells which occur in pairs with cork cells amongst the 

unspecialised ground cells, and which manufacture variously-shaped silica (SiO 2 ) bodies (Esau 1977). Silica deposition in these 

cells was thought to occur “by a passive nonmetabolic mechanism” and the silica cells lose their protoplast (Esau 1977 p. 86). 

However Si appears to be actively taken up by plant roots as monosilic acid and concentrated in the shoots where it is 

polymerised (Reynolds et al. 2009). The microscopic particles of hydrated silica, deposited in intracellular and/or intercellular 

spaces are known as opaline phytoliths in the older literature, or simply phytoliths, and are particularly abundant in Poaceae, in 

which a multiplicity of phytolith forms occur (Gallego and Distel 2004, Parr et al. 2009). Phytoliths are found in plants of other 

familes including Equisetaceae, some gymnosperms and dicotyledons, and other plants in the Poales, but they are most abundant 

and diverse in Poaceae (Thomasson 1986). Most phytoliths consist of hydrated SiO 2 (Honaine et al. 2006), but they also occlude 

plant carbon at the time of creation (Iriarte 2006, Parr et al. 2009). The occluded C is probably derived from internal cytoplasmic 

material and is highly resistant to decomposition, in some cases for >10,000 years (Parr et al. 2009). Higher taxa of grasses can 

be readily distinguished by the types of leaf phytoliths they produce (Piperno and Sues 2005) and each grass species has a 

characteristic phytolith assembage, determined by identification of the relative frequency of the phytolith morphotypes (Gallego 

and Distel 2004). These have so far been classified by shape and size into over 50 forms, including dumb-bells, crosses, elongate 

types and hairs (trichomes). 

Zucol (1996) defined leaf phytolith assemblages in eight Argentinean Nassella spp., including N. neesiana, using cluster and 

principal component analyses and 49 morphological characters, and found that N. neesiana was grouped with N. hyalina. 

Gallego and Distel (2004) were able to define a group including N. tenuissima that contained phytoliths with a high frequency of 

dumb-bell shapes with a short central portion, straight ends, rectangular, smooth elongate hooks and and a sharp-pointed apex. 

Species differences within the group could also be identified and the possibility of species level identification. Honaine et al. 

(2006) found that Nassella spp. and Piptochaetium spp. have “abundant Stipa-type dumb-bells” and “great quantities” of rondels 

(truncated cones), and formed a clear group separate from the other grasses examined. They illustrated typical N. neesiana 

phytoliths: a prickle hair, prickle, dumb-bell with a long central portion and convex ends, simple lobate dumb-bell with a short 

central portion and convex ends, Stipa-type dumb-bell with a short central portion and convex ends, and dumb-bell with a spiny 

central portion. Honaine et al. (2009) illustrated N. neesiana rondells. Honaine et al. (2006) provided a chart of the relative 

frequency of phytolith morphotype groups for the species examined. Inter alia, N. neesiana was found to have an exceptionally 

high frequency of dumb-bells with a long central portion and convex ends, and of simple lobate dumb-bells, compared with the 

other stipoids examined. Whether there is a distinctive stipoid phytolith ‘profile’ has yet to be determined, “Stipa-type” dumbbells 

also being present in members of Pooideae and Arundinoideae (Honaine et al. 2006 pp. 1160-1). 

Phytoliths are incorporated into soil after degradation of plant tissue, and because of their resistance to decay can be used to 

reconstruct former plant communities and ecological history (Honaine et al. 2006 2009) and long-term prehistorical grassland 

dynamics (Iriarte 2006). They also survive and are concentrated by animal digestion, so can be extracted from dung to determine 

animal diets (Piperno and Sues 2005, Prasad et al. 2005). As fossils, they are more resistant to oxidisation than pollen (Iriarte 

2006), and more taxonomically informative than grass pollen, which generally has few useful identification characters and is 

very similar to pollen of some other families, including Restionaceae (Thomassen 1986). 

Phytoliths and silica cells undoubtedly play a major role in deterrence of herbivory (Stebbins 1986, Reynolds et al. 2009), both 

vertebrate and invertebrate, and the silica bodies are “among the few substances capable of inducing morphological changes to 

animal mouthparts” (Piperno and Sues 2005 p. 1128). Silicon defenses reduce the palatability and digestibility of plant tissue and 

increase its abrasiveness and hardness (Reynolds et al. 2009). Soluble Si is also involved in induced chemical defences against 

insect herbivore attack (Reynolds et al. 2009). 

The phytolith profile of different plant organs (leaf, culm, root, inflorescence) differs markedly in a species (at least for 

Paspalum quadrifarium) (Honaine et al. 2009), suggesting adaptation to defend against the differing ranges of predators to 

which these organs are susceptible. Since the earliest known grasses contain a range of phytolith morphotypes similar to modern 

taxa, extant grasses produce much larger quantities of phytoliths than other plant taxa (Prasad et al. 2005) and chemical antipredator 

defences are generally uncommon in the family, it is clear that these silica defences have played a very important role in 

the long history of coevolution of grasses and their predators (Piperno and Sues 2005, Reynolds et al. 2009). The relationships 

between particular plant predators and particular silica-based plant defences have been partially explored (Reynolds et al. 2009), 

with some emphasis on the functioning of trichomes in defence against insects. 

Cytology 

Chromosome numbers in grasses range from 2n = 4 to 2n = 263-265 (Hunziker and Stebbins 1986, Groves and Whalley 2002), 

with n = 6 or n = 7 most likely the primitive condition (Stebbins 1972, Tsvelev 1984), probably the former (Hunziker and 

Stebbins 1986). More than 80% of grasses are polyploid in origin, a larger proportion than any other large plant family, and 

polyploid series are common within species (Hunziker and Stebbins 1986, Groves and Whalley 2002). Hybidisation resulting in 

polyploidy has been very important in the diversification of stipoid taxa: Hunziker and Stebbins (1986) calculated that 91% of 

250 Stipa spp. were polyploid, while Barkworth and Everett (1986) and Vásquez and Barkworth (2004) considered all Stipeae to 

be probably of polyploid origin. 

Chromosomes in Stipeae are reportedly “fairly small” and “usually aneuploid” (Tsvelev 1984 p. 848), i.e. the diploid 

chromosome number is usually not an exact multiple of the haploid number (Johnson 1972). N. neesiana material from 

Argentina, Bolivia, Chile and Uruguay was found by Bowden and Senn 1962) to have diploid chromosome numbers of 28 and 

was considered tetraploid, except for one sample from Limache, Chile (Bowden and Senn 1962). The exceptional material, 

31

identified as Stipa. neesiana by Dr. J.R. Swallen but also referred to also as “Stipa sp.?”, had 2n = 60. It differed 

morphologically from the tetraploid specimens, “but it was not possible to clarify the identification” (Bowden and Senn 1962 p. 

1122). According to Moraldo (1986) N. neesiana is tetraploid with 2n = 44. 

The lowest known chromosome count of a stipoid species is n =10 in a Piptatherum sp. and the highest are n = 41 and n = 48 

(Barkworth and Everett 1986). In Stipa (sens. lat.) and Oryzopsis (sens. lat.) the chromosome numbers range from n = 11 to at 

least n = 41 (including n = 12, 16, 17, 18, 20, 21, 22, 23, 24, 32, 33, 34, 35), a situation that can be explained by recurrent 

amphiploidy, i.e. combinations of autopolyploidy (simple doubling of the number of chromosomes in an individual) and 

allopolyploidy (hybridisation of diploid and/or polyploid individuals), begining with n = 6 and n = 5 (Johnson 1972). In most 

European Stipa species that have been studied 2n = 44 (Tsvelev 1977, Moore 1982), derived through aneuploidy from a primary 

basic number of 2n = 22 (De Wet 1986). Of other European species, S. capensis Thunb. 2n = 34, 36, S. parviflora Desf. and S. 

bromoides (L.) Dörfler 2n = 28, and S. gigantea Link 2n = 96 (Moore 1982, Tsvelev 1984, Moraldo 1986). 2n = 24 in many 

Achnatherum spp. (Tsvelev 1977). Among other South American stipoids, 2n = 40 in Jarava plumosa (Spreng.) S.W.L. Jacobs 

and J. Everett, 2n = 42 and 44 in J.ichu Ruiz and Pavon and 2n = 40 and 44 in Amelichloa brachychaeta (Godr.) Arriaga and 

Barkworth (Bowden and Senn 1962). Of the North American species, 2n = 24 in Piptatherum pungens (Torrey) Dorn, 2n = 44 in 

Heterostipa comata (Trin. and Rupr.) Barkworth and 2n = 48 in Oryzopsis asperifolia Michaux and Piptatherum racemosa (Sm.) 

Barkworth (Barkworth 2006). Stipa sens. str. has a base number of 11, while the base number of Acnatherum is unclear but 

frequently cited as 11 or 12 (Vásquez and Barkworth 2004). 

Among the North American Nassella species, n = 11 in N. pungens E.Desv., n = 16 in N. tenuissima, n = 17 in N. lepida 

(Hitchc.) Barkworth, n = 23 in N. tucumana (Parodi) Torres, n = 32 in N. pulchra (Hitchc.) Barkworth, n =35 in N. cernua 

(Stebbins and Love) Barkworth and n = 41 in N. viridula (Trin.) Barkworth (Johnson 1972). However N. tenuissima material 

from Argentina examined by Bowden and Senn (1962) had a diploid number of 40, different from that of Texas plants. Of the 

South American species, 2n = 34 in N. hyalina, 2n = 36 in N. charruana, 2n = 36 and 38 in N. trichotoma, 2n = 42 in N. chilensis 

(Trin.) E. Desv., and N. exserta Phil., 2n = 64 in N. mucronata (Kunth) R.W. Pohl, 2n = 66 in N. lachnophylla (Trin.) Barkworth 

(Bowden and Senn 1962). But according to Watson and Dallwitz (2005) 2n = 38 in N. trichotoma. 

Genetic variation 

Grass species often possess wide racial or ecoytpic variation within their populations, partial discontinuities of phenotypes 

resulting from self-fertilisation, hybrid sterility between phenotypically similar populations, and other features, including 

polyploidy, that tend to obscure species boundaries (Stebbins 1972). Ecotype variation is known for N. trichotoma in Australia 

(Michalk et al. 2002) but has not been recorded for N. neesiana. 

Britt (2001) and Britt et al. (2002) sought evidence of genotypic variation in Australian populations. They compared four 

Victorian (Altona, Melton, Bacchus Marsh and Tarrawingee) and two New South Wales (Armidale and New England 

Tablelands) populations, using three molecular genetic profiling methods: 

1. ITS – RFLP. Internal transcribed spacer - restriction fragment length polymorphism – a method for cutting the DNA at 

particular short sequences of nucleotides using restriction endonucleases – the polymorphisms in the restriction fragments 

are RFLPs and variation in the size of the RFLPs is dependent on the particular sequence recognised by each endonuclease. 

The ITS is a moderately conserved region of nuclear ribosomal DNA, and differences in its length and base pair pattern 

have revealed taxonomic patterns in grasses and enabled accurate scoring of their genotypes (Britt 2001). 

2. ITS – PCR sequencing. Using the polymerase chain reaction (PCR) to generate large quantities of DNA to enable 

sequencing of the ITS region. 

3. RAPD - PCR. Using PCR and random amplified polymorphic DNA (RAPD) analysis on the whole genome. The technique 

allows differentiation of closely related organisms or even single base changes by display of the DNA banding patterns 

using gel electrophoresis. 

DNA was extracted from between 18 and 10 plants grown from seeds collected in each geographical area (Britt 2001 p. 21), but 

no indication was provided of the number of individuals sampled from each population, their spatial arrangement in the source 

area, or the date of sampling (Britt 2001, Britt et al. 2002). The plants from which DNA were extracted were grown from seed 

collected in a particular area, but that seed could have been harvested from a single mother plant, and even if multiple mother 

plants were sampled a variable proportion of the seed from each would be cleistogamous. Thus all the individuals from which 

DNA was extracted for a particular area could have been genetically identical because of inadequate population sampling. 

No variation was found in the size of the ITS region (650 base pairs) or in the ITS-RFLP patterns tested. ITS sequencing 

revealed uniformity within the plants analysed from a particular area, but differences between plants from different areas, 

unrelated to their geographical proximity. All areas had shared homologies above 90%, suggesting conspecificity, except for 

Bacchus Marsh and Tarrawingee populations, but no one else appear to have suggested these populations are not N. neesiana. 

98% homology – suggestive of convarietal status – was shared only by Altona and New England samples (Britt 2001). Jacobs et 

al. (2000) also analysed sequence data of the ITS region of nuclear rDNA and found no variation within the species tested, 

however, again, the number and intra-population source of their samples were not published. Britt’s (2001) RAPDs showed 

inter- but not intra-area variation but no consistent area groupings. Comparison of population differences using the two different 

methods produced two inconsistent ‘distance tree’ structures. ITS data indicated that the Melton population was most similar to 

N. leucotricha (Britt et al. 2002), suggesting the possibility that some hybridisation may be occurring. Jacobs et al. (2000) 

indicated that they are sister species. 

Britt (2001) and Britt et al. (2002) concluded that Australian N. neesiana possesses large genetic variation between areas (a 

number of subspecies/varieties – the terms were used interchangably – are present), that this may have arisen locally, and that 

local populations have no variation, so must have arisen via cleistogenes or seeds fertilised only by local pollen.The latter 

argument is flawed because Britt (2001) failed to demonstrate that the real population variation in each area was sampled. It is 

32

also based on a misunderstanding of the breeding system (see Britt 2001 pp. 86-87): all the panicle seed is not necessarily crossfertilised 

(see the observations of floral anatomy by Burbidge and Gray (1970) and Vickery et al. (1986) above), the basal 

spikelets in a panicle are more or less cleistogamous, while cleistogamous spikelets can be found anywhere in the panicle. In 

order to reach such a concluison a methodology was required that ensured only the sampling of DNA from chasmogamous seeds. 

The local populations could also be the offspring of separate founder events involving different South American source 

populations, with the lack of local polymorphism (if it is not an artefact) resulting from genetic bottlenecking. 

As with most such studies these appear to be rather random probings of the genome, the genetic material examined is not 

necessarily expressed in the phenotype, nor is its functional role known. Future molecular genetical work in this area should start 

with an examination of South American populations ascribed to particular varieties and should also be integrated with a study of 

morphological variation, as suggested by Britt (2001). 

Phenology, growth and productivity 

Within the revised Raunkiaer plant life form spectrum (Mueller-Dombois and Ellenberg 1974), N. neesiana is classified as a 

hemicryptophyte (a perennial herb with periodic shoot reduction), subtype ‘caespitose graminoids’ (bunched or circular shoot 

arrangement with shoots more or less at the soil surface) and probably as 3.103 “sparingly evergreen during unfavourable 

season”. 

N. neesiana grows predominantly over the coooler months (Muyt 2001) with vegetative growth mainly from autumn to spring 

(Snell et al. 2007). Storrie (2006) considered it “one of the few” major weedy grasses in New South Wales “that produces green 

feed in winter”. In Argentina both agronomists and landholders agreed that it produced a large amount of good livestock feed 

during winter (Gardener et al. 1996b). High rainfall in spring promotes panicle proliferation (Cook 1999). Flowering and fruiting 

occurs from September to March in South America (Zanin 2008). In south-eastern Australia flowering occurs mainly during 

spring and early summer (September to December), but can occur at other times of the year when moisture and temperature 

conditions are suitable (Snell et al. 2007). 

At Inverleigh, Victoria, Gaur et al. (2005) recorded plants in the vegetative phase to 3 October 2003 and 1 October 2004, flag 

leaf swelling over the developing panicle on 13 October 2003, spiky stems on 18 October 2004 and full panicle emergence on 27 

October 2003 and 28 October 2004. Pritchard (2002) recorded that plants at Laverton North had panicles still concealed in the 

sheath on 3 October 2000 and all the foliage was green, and that on 15 November there was a dense covering of emerged 

panicles. In Italy N. neesiana flowers and fruits from May to July (Moraldo 1986) approximately 6 months out of phase with 

Australia, where bolting generally begins in mid October and panicle seed drops in mid December (D. McLaren in Iaconis 

2006b). Slay (2002c) reported a similar phenology in New Zealand: elongation of reproductive tillers in spring with the main 

flush of tiller production from mid-September to mid-October, flag leaf swelling in mid-October, then 24 days between the first 

emergence of the panicle and anthesis. Slay (2001) found the period from the boot stage to anthesis was 36 days and from 

anthesis to 100% viability of seed was 33 days. 

Slay (2001) identified: three phases of seeding: 1. basal cleistogene production during vegetative growth of the tiller, initiated in 

autumn and completed in spring before anthesis of panicle flowers; 2. cleistogamous and chasmogamous seed production in the 

panicle; 3. cleistogene production on the stems, initiated before anthesis of panicle flowers and completed after panicle seed 

maturation. 

In Victoria stems are brown and breaking down and cleistogenes form in mid to late summer, and by the end of February most 

stems have cleistogenes (McLaren in Iaconis 2006b). In New Zealand old culms are easily broken by late March (Slay 2001). 

The inactive period from mid summer to the time of autumn rain has been inadequately documented and summer dormancy 

appears to have been widely assumed. However flowering is indeterminate (Grech 2007a) and plants are known to flower in 

response to summer rain (Bedggood and Moerkerk 2002). Senescence of foliage and cessation of leaf growth occurs in all 

grasses in response to drought, but truly summer-dormant grasses display these traits despite summer irrigation (Norton et al. 

2008). Across the range of pasture grass taxa there is a continuum of responses from full dormancy to non-dormancy, and the 

intensity of summer dormancy can be assessed by simulating a mid summer storm in the midst of drought, and measuring 

subsequent herbage production or senescence (Norton et al. 2008). Full dormancy is supposedly characterised by complete 

cessation of growth, full herbage senescence and dehydration of young leaf bases, whereas senescence and partial growth 

cessation may be just a dehydration avoidance strategy which can be expressed in any season under conditions of soil moisture 

stress. The possession of true summer dormancy is supposedly critical for pasture grass persistence in the drought prevalent 

pastures of south-eastern Australia, so a number of commonly utilised exotic grass cultivars have been assessed for this 

characteristic (Norton et al. 2008). Assessments of weedy and native Poaceae, including N. neesiana, would be informative. 

Average production in New England pasture was 2.3 t ha -1 y -1 (Gardener et al. 2005). Slay (2001) recorded production of 5.5±0.5 

kg ha -1 of dry matter per day during the first 48 days after mowing in November in New Zealand. Grech (2004) found that plants 

that were regularly clipped to simulate grazing produced more digestible growth (significantly more crude protein, metabolisable 

energy and digestible dry matter) than unclipped plants. N fertiliser (100 kg ha -1 in two applications) increased the feed value 

only at the seedhead stage. 

Crude protein levels of green leaves (in South America) were 6.3-18.3% (Gardener et al. 1996b). Crude protein levels of winter 

foliage regrowth after clipping in Australia were 12.7-16.6% and digestible dry matter of green leaf material was c. 60% (range 

58-66%), compared to Festuca arundinacea with crude protein of 13.0-18.8% and digestible dry matter 62-69% (Gardener, 

Storrie and Lowien 2003, Gardener et al. 2005). Culm material had a crude protein content of 4.5%. New Zealand data indicates 

a crude protein level of green leaves of plants in the vegetative phase of 14.5%, while leaves of plants early in the flowering 

stage had a crude protein level of 6.4%. The metabolisable energy value of vegetative phase leaves was 7.7 in early summer and 

of 7.5 in early flowering. Comparable crude protein and metabolisable energy (ME) levels were 9.9% and 11 for Lolium perenne 

L., while ME values in summer 1991 for D. glomerata, L. perenne, Phalaris aquatica L. and Festuca arundinacea were 8.3, 7.5, 

33

8.4 and 9.3 respectively (Slay 2001, Slay 2002a). The minimum maintenance level of crude protein for livestock is 7% (Slay 

2001). Gardener et al. (1999 p. 10) compared N. neesiana productivity and feed analysis with Festuca arundinacea, and Grech 

(2007a) compared it with Dactylis glomerata. 

Flowering and fruiting 

According to Barkworth and Torres (2001 p. 462) and Barkworth (2006) Australian populations “do not appear to set seed in the 

exposed inflorescences”. These erroneous statements are possibly based on a few herbarium specimens collected during drought 

conditions. 

Plants in their first year of growth do not flower (Sethu Ramasamy pers. comm. 13 June 2007). Benson and McDougall (2005) 

provided the primary juvenile periods of many other grasses but failed to provide one for N. neesiana. No definitive published 

information appears to be available on the juvenile period. 

The factors that trigger a shift from vegetative to reproductive growth, are partially known. Low winter temperatures are 

necessary for vernalization (Gardener et al. 1996b). (See Sinclair 2002 for the variety of factors involved in floral inducation in 

grasses). Rainfall appears to play a role. In southern Victoria bolting generally commences in mid October (D. McLaren, 26 

October 2006). 

According to Gardener et al. (2003a) the main ‘flowering’ period is November to February in Australia and October to January 

in South America (based on herbrium specimens) with some flowering as late as May. Near Guyra on the Northern Tablelands of 

NSW ‘flowering’ commenced between early October and early November and finished between early March and late April 

(1994 and 1995, Gardener et al. 2003a). Two separate flowering periods were observed near Guyra in 1995, but not in 1996 or 

1997. However Gardener et al. (2003a, etc.) had a very broad, inaccurate definition of ‘flowering time’ covering the period from 

the first emergence of the panicle from the leaf sheath to the time when the last mature seeds were dropped. 

Burkart (1969) may be more accurate, giving the flowering period in Argentina as October and November. In south western 

Europe, flowering occurs from May to July (Verloove 2005). Slay (2002c) is pehaps most specific, stating that anthesis 

(flowering) occurs around mid-November in New Zealand, while Slay (2001) observed flowering over a period of approximately 

17 days from mid to late November in an infestation at Waipawa, New Zealand. In Australia, according to Cook (1999), panicle 

production occurs from late spring to early summer, sometimes with a secondary, much less prolific, flush in autumn. Snell et al. 

(2007 p. 8) noted that there is “a second panicle seeding period during autumn” in northern NSW. Bedggood and Moerkerk 

(2002) state that flowering occurs from early October through to March or April and that panicle seed are produced in autumn in 

northern areas. In the ACT N. neesiana usually flowers from November to February (ACT Weeds Working Group 2002). In 

South Australia flowering or fruiting inflorescences are present in November and December (Jessop et al. 2006). Flowering in 

Victoria is mostly October to February (Walsh 1994) and in Australia generally from spring to early summer with occasional 

plants to April (McLaren, Stajsic and Iaconis. 2004). 

According to the ACT Weeds Working Group (2002) and Ens (2005 citing Ens 2002a) N. neesiana “ has the ability to flower all 

year round” given favourable environmental conditions. Bedggood and Moerkerk (2002) recorded that it will flower in response 

to summer rain. High density of tiller production (and thus seed production) appears to require consistent rainfall during the 

reproductive period (Gardener et al. 2003a). Slay (2002c) noted that a second flush of panicles can be produced if the normal 

late spring panicles are removed by mowing or grazing, and that isolated plants are more likely to have a secondary flowering 

period than plants in dense populations. 

The flowering periods recorded mostly appear to apply to large populations in broad areas. No information seems to be available 

on site specific aspects of flowering, the proportion of the population that has a particular floral phenology, or the individual 

variation of a single plant. There is no indication of the duration of flowering on a particular panicle or the sequence and timing 

of individual flower opening and closing on a panicle. The time of day that the flowers open and their period of opening do not 

appear to have been recorded. Information about the timing of fertilisation, the period of receptivity of the stigma, the period 

between pollination and fertilisation and between fertilisation and embro development have rarely been determined for any 

plants (de Tirquell 1986) and are unknown for N. neesiana. 

The period from flowering to seed maturation is about 7 weeks (Liebert 1996). According to Slay (2002c), seed became viable 

within 8 days of flowering, 76% of seed was still ‘soft’ after 14 days, with 60% viable after 22 days. In New Zealand most seed 

reaches full maturity between late December and early January and is quickly shed, but seed produced on sub-panicles that 

emerge from stem nodes matures later (Slay 2002c). 

The seed is a velate caryopsis, i.e. the caryopsis falls free of other spikelet parts (Sendulsky et al. 1986). Panicle seed matures 

and is shed from mid to late summer (Gaur et al. 2005), generally in mid December in southern Victoria (D. McLaren, 26 

October 2006). Near Guyra on the Northern Tablelands of NSW most seeds matured between late December and mid-February 

(1994 and 1995, Gardener et al. 2003a), and was generally shed in December-January (Gardener et al. 2003ba). Slay (2001) 

noted that panicle seed at one site was almost all shed simultaneously in early January 2001. Most panicle seed has been shed by 

late February in NSW (Duncan 1993) and this is the case in Victoria (Liebert 1996). Stem cleistogene seed from ensheathed 

inflorescences matures later than panicle seed, in February (Slay 2002a). In the Melbourne region in 2006 under drought 

conditions, most panicle seed was shed in November-December, largely in late November-early December (pers. obs.). Puhar 

(1996) noted that seed began to drop at St Albans, Victoria, in early January 2006. 

Slay (2002c) modified the Feekes Scale used to define the growth stages of cereals to classify the reproductive growth stages of 

N. neesiana (Table 2). 

34

Table 2. Modified Feekes Scale for Nassella neesiana panicle seed (adapted from Slay 2002c). Point 10 on the scale was called 

the ‘boot stage’ by Pritchard (2002). 

Feekes 

Scale 

Description of stage for N. neesiana 

10 Flag leaf completely grown out, sheath swollen [boot stage] 

10.1 Tips of awns emerged from swollen sheath 

10.2 25% of heads emerged 



10.5 100% of heads emerged; panicle free of sheath 

10.5.1 Beginning of anthesis 

10.5.2 Anthesis complete to top of panicle 

10.5.3 Anthesis finished at base of panicle 

10.5.4 No unopened flowers, stamens dropped, some seed developing 

11.1 Milky ripe (milky dough stage) 

11.2 Mealy ripe, seed contents soft but dry 

11.3 Seed firm, chalky 

11.4 Seed shedding 

Tillers usually die after flowering and fruiting (as for grasses in general - Wheeler et al. 1999) usually within about 2 months 

(personal observations). In southern Victoria the stems have usually browned off by mid January and begin to break down (D. 

McLaren, 26 October 2006). The dry culms can remain standing for up to 6 months (Gardener et al. 1999) and generally persist 

for several months before falling over and forming a dense mat that shades underlying vegetation (Gardener et al. 2005). 

Cleistogene production 

Separate hormonal influences appear to independently control panicle seed and cleistogene production (Connor et al. 1993). 

Stem cleistogenes occur at two different periods:1. during the vegetative growth stage of the tiller - initiated in spring as single 

spikelets under the leaf sheaths of young shoots, reaching maturity by the time of emergence of the aerial infloresence, and 

released when the leaf sheaths weaken or rupture; and 2. during the flowering period, resulting in cleistogenes tightly enclosed 

by the sheath and prophyllum (Connor et al. 1993). According to Slay (2001 2002c) basal cleistogenes are initiated after 

seedlings establish and on older plants in autumn, and are maturing or mature by the time that panicle production and flowering 

commences in mid spring. Stem cleistogenes are initiated as the stem elongates, and ripen by February. In southern Victoria, 

non-basal stem cleistogenes have formed by the time that panicle seed is dropped and are possibly stimulated to mature by 

panicle seedfall, and most stems have mature cleistogenes by the end of February (D. McLaren, 26 October 2006, based on 

observations of Shiv Gaur). Slay (2001) observed that the non-basal stem cleistogenes are immature when panicle seed is 

dropped, especially at the top node. Gaur et al. (2005) found that non-basal cleistogenes mature, and are shed, after the fall of 

panicle seed. 

Dyksterhuis (1945) found a similar phenology with N. leucotricha: axillary cleistogenes with endosperm in the dough stage were 

present before anthesis of panicle flowers or before any panicle production. Seedlings were able to produce cleistogenes in the 

dough stage by the age of 6 months, and cleistogene seedlings 9-10 months old, that were clipped to a height of 38 mm every 

two weeks, produced very few tillers and on average less than half the cleistogenes (range of 0-3 cleistogenes per seedling) of 

unclipped seedlings (range of 2-5) (Dyksterhuis 1945). 

Slay (2002c) noted that the ability of an autumn germinating seedling to produce cleistogenes within 6 months means that the 

species can survive as an annual. 

Distribution 

South America 

Map: Gardener et al. (1996b) and Gardener (1998); records between 26º and 40º S from two herbaria and literature. 

Argentina, Bolivia, Brazil, Chile, Ecuador, Peru, Uruguay (Torres 1997, Barkworth and Torres 2001, Martín Osorio et al. 2000, 

Barkworth 2006). It is also known from Paraguay (Ramirez 1951,USDA ARS 2006, Zanin 2008) although presence in that 

country was considered doubtful by Barkworth and Torres (2001) and Barkworth (2006). Longhi-Wagner and Zanin (1998) used 

the name Stipa setigera C. Presl. for N. neesiana (“syn. S. neesiana Trin. and Rupr.”) and recorded it from Paraguay and 

Columbia in addition to the countries already cited. 

Considerable uncertainty remains in the understanding of the South American distribution due to confusing use of specific names 

and differences of opinion about the actual identity of specimen material. Barkworth and Torres (2001), who did not refer to 

Longhi-Wagner and Zanin (1998), considered S. setigera to be a synonym of N. mucronata (Kunth.) R.W. Pohl, and the two S. 

setigera varieties hispidula and glabrata to be synonyms of N. neesiana, the latter synonymies established by Hitchock in 1925 

(Torres 1993). They listed N. mucronata as present in Columbia, and did not list N. neesiana for that country. Zanin (2008) 

35

affirmed the opinion of Longhi-Wagner and Zanin (1998) that N. neesiana is present in Paraguay, but cited a single specimen 

that Barkworth and Torres (2001) considered to be N. argentinensis (Speg.) Peñail. Furthermore, Barkworth and Torres (2001) 

did not list Brazil, Paraguay or Uruguay as part of the range of N. mucronata, so it appears likely that the entity Longhi-Wagner 

and Zanin (1998) treated as S. setigera and considered to be the same as N. neesiana was not the same entity as the S. setigera 

assigned to N. mucronata by Barkworth and Torres, but was equivalent to the N. neesiana of Barkworth and Torres. N. 

mucronata is not recorded from Brazil (Torres 1997, Barkworth and Torres 2001), but it remains unclear how Longhi-Wagner 

and Zanin (1998) treated extra-Brazilian literature records of S. setigera sens. lat. and whether they considered which of them 

were Nassella mucronata. Further complicating interpretation, Barkworth and Torres (2001) listed Portugal, Spain, France and 

Italy as countries to which N. mucronata had been introduced, but Verloove (2005) found that the name S. setigera had been 

misapplied to these European records, i.e. the specimens were misidentified, and were in fact N. neesiana, not N. mucronata, and 

that this probably also applied to Mexican S. setigera, which was also N. neesiana. Given this confusing situation, the entities 

actually present in Columbia, and Mexico, and perhaps Brazil appear to require clarification. 

Gardener et al. (1996b) stated that in South America as a whole N. neesiana had a latitudinal range of at least 22-51º S. Gardener 

(1998), corrected this to 26-40º, stating that the published record from Chile at 50º 53’ S by Soto (1984) was most likely a 

misidentification of Stipa brevipes Desvaux. However even the wider latitudinal range fails to cover either Ecuador or Peru, and 

excludes Bolivia except for a small area in the extreme south.The narrower range excludes all of Bolivia, most of Paraguay and 

indicates the absence of herbarium and literature records from most of southern Argentina. In southern South America it has a 

continuous distribution from near the coast of Chile, across the Andean zone and the pampas of Argentina through Uruguay to 

the Paranaense (south east Brazil) region (Longhi-Wagner and Zanin 1998, as Stipa setigera). In tropical and subtropical South 

America it occurs in highland enviroments (Martín Osorio et al. 2000) north along the Andes to Ecuador, to c. 1.5°N. Columbia 

and Venezuela are probably the only South American countries in which Stipa sens. lat. has been recorded (Columbia has 3 spp. 

and Venezuela 5: Longhi-Wagner and Zanin 1998) but in which S. neesiana has not. However its presence in the province of 

Carchi of Ecuador suggests that it may well occur in south-western mountains of Columbia. 

In Argentina it is found in the following provinces - in the east (Misiones, Corrientes, Entre Ríos), the centro del pais (Buenos 

Aires, Córdoba, San Luis, La Pampa, Santa Fé), the west (Mendoza, San Juan), the north (Chaco, Formosa), the northwest 

(Jujuy) and the northeast (Catamarca, La Rioja, Salta and Tucumán) (Caro 1966, Gardener et al. 1996b, Torres 1997, Gardener 

1998). It is present in the Ventania land system in the south-west of Buenos Aires Province, north of Bahía Blanca (Amiotti et al. 

2007). Large populations are common in the central west around the Sierra de Córdoba (Sierras Pampeanas) (Anderson et al. 

2002), where the grasslands are at moderately high altitudes (Soriano et al. 1992). It is a common species in the pampas, a vast, 

humid, fertile plain stretching from the Rio de La Plata and the Atlantic coast west towards the Andes, occupying the provinces 

of Buenos Aires, parts of Entre Rios and southern Santa Fé, into western San Luis, northern La Pampa and southern Córdoba 

provinces (Soriano et al. 1992). In Santa Fe province it is a characteristic species of the Pampean phytogeographical province 

flechillar Stipeae community in Rosario County in the south of the province, and in San Cristóbal County in the central west of 

the province is the dominant species of another flechillar grassland community of the Espinal phytogeographical province 

(Feldman et al. 2008). In Entre Ríos it was recorded by Zucol (1996) from Santo Ana, Dpto. Federación; Puerto Yerúa, Dpto. 

Concordia; Dpto. Villaguay; San José, Dpto. Depto. Colón; Colonia Elia, Dpto. Uruguay; Dpto. Gualeguay; and Camino a 

Puerto Unzué, Dpto. Gualeguaychú. It is not a dominant grass in the semiarid southern Caldén District near Gaviotas, west of 

Bahía Blanca (Distel 2008), nor is it one of the major grasses in the whole Caldenal, a 10 million ha semi-arid region extending 

from the Atlantic coast south of Bahía Blanca north west to the Sierra de Córdoba, to the south and west of the Humid Pampa 

(Fernández et al. 2009). Martín Osorio et al. (2000) mention its presence in Patagonia (apparently citing Rivas Martínez) but 

other references to its presence there have not been found. In Uruguay it is common in pastures (Gardener et al. 1996b citing 

Rosengurtt et al. 1970) with records in the south, north-west and north-east (Gardener 1998) including Florida (Bowden and 

Senn 1962) and Montevideo (Barkworth et al. 2007). In Paraguay it was collected in 1989 at Chacoí in the far south of the 

province of Presidente Hayes, near the border with Central province, in the vicinity of Asuncion (Zanin 2008), east of Formosa 

province in Argentina. But it possibly occurs also in southern Paraguay, being known from close to the border in the states of 

Misiones and Corrientes in Argentina (Gardener et al. 1996b, Gardener 1998). Temperate grassland is present in Paraguay only 

in small area of the far south-east (Overbeck and Pfadenhauer 2007) in the vicinity of Encarnacion, immediately to the northwest 

of Misiones. In Brazil it is found in the south (McLaren, Stajsic and Iaconis. 2004), in the states of Rio Grande do Sul and 

Santa Catarina, where it is the most common species of Stipa sens. lat. (Longhi-Wagner and Zanin 1998). In Uruguay and Brazil 

it is mostly found in the campos, a temperate subhumid grassland formation similar to the Argentine pampas, covering most of 

Uruguay and the southern Rio Grande do Sul of Brazil (Soriano et al. 1992). In Brazil it is considered a characteristic species in 

the Pampa biome of the southern Rio Grande do Sul (Overbeck et al. 2007 – as “Stipa setigera C.Presl.”). It was not recorded 

from a humid subtropical grassland site in the Atlantic Forest biome on Morro Santana (30°03’S) by Overbeck et al. (2006). The 

most southerly distribution on the continent is in Chile, where it is found over a wide latitudinal range, at least from c. 26º 30’ to 

40º S (Gardener et al. 1996b, Gardener 1998). In continental Chile it is found in the Metropolitan Region (Santiago 33-34°S) and 

Regions IV (29-32°S), V (32-33°S), VIII-IX (35-39°S) and X (39-44°S ) (Baeza et al. 2007). It was collected at Concepcion 

(36°50’S) by Senn (Bowden and Senn 1962) and has been recorded on Robinson Crusoe Island (Más á Tierra) (33º 38’ S 76º 52’ 

W) (Nelis 2006, Baeza et al. 2007) and Alejandro Selkirk Island (Más Afuera), in the Juan Fernández group, c. 650 km west of 

mainland Chile, where it is considered to have been introduced (Baeza et al. 2007). It is known from Valparaíso (the types of S. 

longiflora) and near Santiago (the type of S. trachysperma) (Barkworth et al. 2007). In the northern part of its South American 

range its is found at high altitude. In Bolivia it is known from La Paz (Martín Osorio et al. 2000) and was collected at 

Cochabamba (17°26’S 66°10’W) by Senn (Bowden and Senn 1962). In Ecuador it is also native to the Andean region, at 

altitudes of 3000-3500 m in the Provinces of Carchi (c. 1.5° north of the equator near the Columbian border) and Canar (Clark et 

al. 1999-2008). 

The type of var. fernandeziana is from the Juan Fernández Islands of Chile (Barkworth et al. 2007). Vars. formicarioides, 

gracilior and virescens are endemic to Argentina, vars. hirsuta and longiaristata are recorded from Argentina and Uruguay, and 

var. neesiana is recorded from Argentina, Bolivia, Brazil, Chile and Uruguay (Zuloaga et al. 1994). Populations in north-eastern 

36

Argentina, in the provinces of Chaco, Corrientes and Misiones, are possibly not var. neesiana (Gardener 1998). The specimen 

from Paraguay was not assigned to a variety by Zanin (2008). 

Introduced range outside Australia 

Europe, first found in Europe in France by Touchy in 1847 at Port Juvenal, Montpellier; early records from imports on hides and 

wool (Hayward and Druce 1919). Mediterranean (unspecified, Jessop et al. 2006); occurs “from time to time, particularly in the 

Mediterranean region” (Martinovský 1980); believed to be an unintentional introduction to south-western Europe “apparently ... 

absent as naturalized ... from the rest of Europe” (Verloove 2005). Not recognised as present in Europe outside the British Isles 

and the Canary Islands and Madiera by Weber (2003). 

British Isles: Weber (2003) considered it to be not invasive in natural areas nor solely a weed of agroecosystems. Scotland: a 

wool-alien first found on the banks of the River Gala below the town of Galashiels, County Selkirk by Ida Hayward in 1916 

(Hayward and Druce 1919); now extinct in that area (Vines 2006); recorded near wool factories (Bourdôt and Hurrell 1987a). 

England: found on a rubbish heap at Mortlake, Surrey, in 1916 (Hayward and Druce 1919); sometimes more or less naturalised 

in the south east, scattered in England (Stace 1997); “certainly” naturalised (Gardener 1998); first recorded at Mort Lake (sic) in 

1916, Kirkheaton in 1960, Mossley and Mauldenin 1965, Flitton in 1969 and Ware in 1988 (Gardener 1998). 

Portugal: including Azores (Madiera) (Martín Osorio et al. 2000, Verloove 2005), first recorded on Madiera in 1970 (Gardener 

1998) and known from Coimbra on the Iberian Peninsula (Vàzquez and Devesa 1996 as N. mucronata). 

Spain: including Alt Empordà, Rossellò near Lleida (Font et al. 2001) and Gerona in Catalonia (Verloove 2005), Madrid 

(Vàzquez and Devesa 1996 as N. mucronata) and the Canary Islands, specifically Gran Canaria, Gomera and Tenerife (Martín 

Osorio et al. 2000, Sans-Elorza et al. 2005, Verloove 2005); “becoming naturalised” (Scholz and Krigas 2004, p. 78); not 

considered an “established alien” in continental Spain (Gassó et al. 2009). First detected in the Canaries in May 1964 by J. Lid in 

the dominion of Monte Verde de Anaga on the island of Tenerife (Martín Osorio et al. 2000). Distribution maps for the Canary 

Islands and the Parque Rural de Anaga on Tenerife were provided by Martín Osorio et al. (2000). 

France: including Corsica (Martín Osorio et al. 2000, Verloove 2005), where it rapidly proliferated (Font et al. 2001); 

“introduced into southern France” (Barkworth 2006); “becoming naturalised” (Scholz and Krigas 2004 p. 78); recorded near 

wool factories (Bourdôt and Hurrell 1987a); adventive in the Montpellier region, introduced with wool and “plus et moins 

naturalisées ... dans des stations naturelles” (Thellung 1912 p. 654), including Port Juvénal 1847-1877, Montplaisir 1877, 

Lodève 1877 and Bèdarieux 1894 (op. cit. p. 94), introduced to Lodève in wool (op. cit. p. 614); present in the neighbourhood of 

Port Juvénal “for a long period of years in the [wool] drying yards at Montplaisir near Lodève and Bèdarieux on the river Orb, 

both in the Hérault” in the 18th and 19th centuries, and in 1909 on beaches away from factories below St Hélène and Caras at 

Nice (Hayward and Druce 1919 p. 228). First recorded at Lodeve in 1847 (Gardener 1998) or 1877 (Thellung 1912), Montpellier 

in 1894 (Gardener 1998) and Nice in 1909 (Thellung 1912, Gardener 1998). 

Germany: adventive in Berlin and Anhalt (Thellung 1912); recorded at Rodleben wool factory at Rosslau, Anhalt, in 1910 

(Hayward and Druce 1919); recorded near wool factories (Bourdôt and Hurrell 1987a). 

Italy: Map: Moraldo (1986 p. 237). ‘Adventitious naturalised’ in grassy areas (Moraldo 1986 p. 217); “becoming naturalised” 

(Scholz and Krigas 2004 p. 78). First reported on the banks of the Polcervera near Genoa, 1904, at a tan works that had hides 

from Argentina (Hayward and Druce 1919, Moraldo 1986), then in other localities in Liguria (Moraldo 1986); first recorded at 

Bordighera in 1910 (Gardener 1998). ‘Most recently’ in Rome at Villa Ada, first collected in 1970 (Moraldo 1986 p. 217). Rome 

and Liguria (Verloove 2005); recorded near wool factories (Bourdôt and Hurrell 1987a). 

Greece: East Macedonia, Nomos and Eparchia of Thessaloniki, Thessaloniki city “mowed and watered lawn of a traffic island 

within the University campus, about 100 individuals in total, fragmented in patches of 10-20 individuals each”, herbarium 

specimen 22 May 2002 (Scholz and Krigas 2004 p. 78). 

South Africa (Gibbs Russell et al. 1985, Wells et al. 1986), “threatens to invade disturbed grassland areas from the Cape into the 

Transvaal” (Wells and Stirton 1982); first found in Barkley East in 1941, “emerging as serious weed” and common in the 

Eastern Cape province (Gardener 1998 p. 12), also in Free State province, at altitudes of 600-1700 m (Germishuizen and Meyer 

2003). First recorded from Grahamstown and Bosberg in 1968, Adelaide in 1977, Ladybrand in 1988 and Sterkstroom at an 

unknown date (Gardener 1998). 

New Zealand: Jacobs et al. (1989 Fig. 2) and Slay (2002a) provided country maps. First recorded in 1940 by H.H. Allan, with 

the earliest specimen undated, but probably collected in the late 1920s in Auckland (Jacobs et al. 1989, Edgar et al. 1991). A 

very restricted, discrete distribution. Limited to about 1500 ha of pasture in Marlborough and smaller areas in Hawkes Bay and 

Auckland (Bourdôt and Hurrell 1989a); 3000 ha in Marlborough by 2001, 600 ha infested in Hawkes Bay by 2002 (Slay2002). 

North Island: Auckland (first recorded during the late 1930s, “a few plants still occur today in a public domain at Western 

Springs” (Bourdôt and Ryde 1986) a “picnic area” and “a railway enthusiasts station” (Slay 2002a p. 11)), Waitakere Ranges, 

Waipawa (central Hawkes Bay) (Jacobs et al. 1989, Edgar et al. 1991, Edgar and Connor 2000); first collected probably in the 

late 1920s in the Waitakare Ranges, and c. 1962 at Hawkes Bay (Connor et al. 1993), although the Waipawa infestation possibly 

may have arisen from contaminated seed from Marlborough sown in the early 1950s (Slay 2002a); an estimated total of 600 ha 

infested in Hawkes Bay (Slay 2002a). South Island: Marlborough (roadsides and pastures near Blind River, Seddon and Lake 

Grassmere) (Jacobs et al. 1989, Edgar and Connor 2000), near Blenheim Airport and farms at Renwick (Connor et al. 1993), 

with first occurrence anecdotally dated to about 1930 (Bourdôt and Hurrell 1989a) but said to be first found there about 1945 

(Bourdôt and Ryde 1986). 1n 1986 it was present on 7 farms and about 30 ha at Waipawa, with isolated plants over up to 1 km 

along the Waipawa River, and at Blind River it was present on at least 15 farms, with c. 100 ha densely infested (Bourdôt and 

Ryde 1986); recently located in the Awatere valley 33 km from Blind River (Slay 2002a). The Auckland and Marlborough- 

Hawkes Bay populations are distinct forms, differing in a few characters, and represent separate provenances (Connor et al. 

1993). The Auckland material has long hairy laminae and lemma nerves, with hairs on the main lemma nerve almost reaching 

the corona (Jacobs et al. 1989). Weber (2003) considered it to be not invasive in natural areas or solely a weed of 

agroecosystems. 

37

USA: once found on ballast dumps in Mobile, Alabama (Hitchcock and Chase 1971, McLaren et al. 1998, McLaren, Stajsic and 

Iaconis 2004), first recorded in 1935 (Gardener 1998), but “no recent collections” (Barkworth 1993) and “has not persisted” 

(Barkworth 2006), although it is currently mapped as naturalised in Mobile County (USDA NRCS 2006, Zipcode Zoo 2006). 

Mexico: (Zipcode Zoo 2006); several records, earliest 1896 (Verloove 2005). Verloove noted the widespread misapplication of 

the name Stipa setigera (= Nassella mucronta) to N. neesiana in Europe and apparently considered Mexican material had also 

been misidentifed. N. mucronata was recorded as widespread in Mexico by Reyna and Barkworth (1994) and present in Mexico 

by Barkworth and Torres (2001), but these authors did not recognise the presence of N. neesiana in that country. 

South America: Introduced prior to 1921 to the Juan Fernandez Islands of Chile, where well established (Baeza et al. 2007). 

Probably introduced elsewhere in South America, but the native range is difficult to determine. 

Australia 

Maps: Walsh (1994 - Victoria), Liebert (1996 - Victoria), McLaren et al. (1998 - Australia), Gardener (1998 all known 

Australian records, plus dates of important regional distribution points), Thorp and Lynch (2000 - Australia), ARMCANZ et al. 

(2001, Australia, with distribution points for each decade from the 1930s), Bruce (2001- mainly natural temperate grassland sites 

in the ACT), Mallett and Orchard (2002), Frederick (2002 - North Central Region, Vic.), Ens (2002a - Cumberland Plain, 

Sydney), McLaren et al. (2002b – south-eastern Australia, based on a survey of land owners and managers in areas thought 

likely to be infested and including absence data), Jim Backholer, DPI Victoria (2006 – Australia, reproduced in Snell et al. 

2007and below as Fig. 3). The grid map of Thorp and Lynch (2000) allowed for display of areas of low, medium and high 

density but no high density areas were shown; the distribution displayed included only Victoria and a small area of south-eastern 

South Australia and was based only on records provided by Primary Industries and Resources South Australia and the Victorian 

Department of Natural Resources and Environment. Data consisted of 0.5º cells for South Australia and 0.25º cells for Victoria. 

The most recent and most comprehensive national map is that produced by Backholer in September 2006 for the National 

Chilean Needle Grass Taskforce (Fig. 3). 

Anderson et al. (2002) noted that there were no published estimates of the area of Australia infested. McLaren, Weiss and 

Faithfull (2004) noted that there were also no published records of the area infested for the States in which it occurred. By 

surveying landholders in areas known to have N. neesiana populations in Victoria, New South Wales and the Australian Capital 

Territory, McLaren et al. (2002b) determined that infestations were dispersed over an area of over 4 million ha and that the plant 

was still actively dispersing in Victoria. 

Distribution in the States and Territories of Australia is summarised below, and more detailed records are provided in the 

following section on the history of the plant’s dispersal in Australia. N. neesiana was not recorded from the Australian Alps by 

McDougall and Walsh (2007). 

Victoria: Northcote - first Australian record 1934 (McLaren, Stajsic and Iaconis. 2004); widespread in the Port Phillip, 

Corangamite, Glenelg-Hopkins, North Central, Goulburn and North East Catchment and Land Protection Regions with scattered 

records in West Gippsland and Wimmera Regions. 

New South Wales: Glen Innes - first NSW record 1944 (McLaren, Stajsic and Iaconis. 2004); Central Coast, Northern 

Tablelands including Glen Innes, Guyra (Duncan 1993), Tenterfiled, Emmaville (Gardener 1998), Southern Tablelands (Wheeler 

et al. 1990) including Bungendore (Eddy et al. 1998), North West Slopes (Storrie and Lowien 2003), Cenral West Slopes (data 

points in map in McLaren et al. (2002b); widespread on roadsides in the Sydney region (bounded approximately by Rylstone, 

Singleton, Nowra and Taralga) (Carolin and Tindale 1994). 16 sites on the Cumberland Plain, Sydney region (Ens 2002a), first 

recorded in the Cumberland Plains at Mount Annan Botanic Garden in 1989 (Benson and von Richter 2009); within the Sydney 

region mainly in Western Sydney (Benson and McDougall 2005); Balranald (Sheehan 2008). 

Australian Capital Territory: near Burbong “from where it may spread downstream along the Molonglo” River (Burbidge and 

Gray 1970). Found at 82% of sites investigated in the Canberra area in 2000-01(Bruce 2001); extensive additional sites detected in 

2002 (Sharp 2002). Currently very widespread and a major component of suburban ‘nature strips’ and lawns (S. Sharp and J. 

Connelly pers. comms. 2006, personal observations). 

South Australia: first recorded in 1988 (Jessop et al. 2006) at Lucindale (McLaren et al. 1998), considered naturalised (Storrie 

and Gardener 1998); South East (Thorp and Lynch 2000); Northern Lofty region (near Bundaleer), Southern Lofty region, South 

East (Lucindale) (Jessop et al. 2006). ) Okagparinga Valley by late 2000 (Obst and How 2004). 53 infestations in the Mt. Lofty 

Ranges, Fleurieu Peninsula and greater Adelaide regions totalling 14.0 hectares recognised up to December 2003, including 

Modbury (moderate to heavy, 0.07 ha), Adelaide Parklands (5 plants removed by hand), Clarendon (one site, 0.02 ha, low 

density, grazed pasture) and Wirrina (several sites, 13.77 ha) (Obst and How 2004). 

Queensland: southern (Mallet and Orchard 2002, Michael Hansford in Iaconis 2003), in the Shires of Clifton, Warwick and 

Cambooya (Phil Maher in Iaconis 2006b). 

Tasmania: Hobart area (Mallett and Orchard 2002, Hocking 2005b). 

38

Figure 3. Recorded distribution of N. neesiana in Australia, September 2006. Source: J. Backholer, Department of Primary 

Industries Victoria; data from the Integrated Pest Management System (used in Snell et al. 2007). This map fails to include some 

significant outliers in South Australia including Lucindale in the South East, and the Bundaleer area in the Northern Lofty 

region. 

N.neesiana has a very wide latitudinal range. In South America from approximately the equator (Columbia) to c. 40º S, in 

Europe from c. 28-51º N, and in Australia approximately 28-43º S. 

There may have been more than one introduction of N. neesiana to Australia (Hocking 2002). Wide separation of initial 

populations in each State suggest that this is likely. The geographical origin of Australian populations is unknown (but see 

discussion of this question in the section, below). 

N. neesiana has a syanthropic distribution (New 1994), strongly associated with towns and cities and within the web of major 

roads around Melbourne (Hocking 2007). It is currently most frequent in urban and urban fringe areas (in public open spaces, 

vacant land etc.), roadsides, agricultural pastures, and the cultural steppe, and its range expansion in Australia appears to be 

dependent to a large extent on human dispersal (Snell et al. 2007). Like many other weeds, its dispersal can be characteristed as 

stratified diffusion (Shigesada and Kawasaki 1997), with long distance propagule movement giving rise to isolated new 

populations and expansion at the edges of existing populations The extent to which it intrudes into natural habitats without 

facilitation by humans is one of the main focuses of this investigation. 

History of dispersal in Australia 

Identification failures and lack of recognition 

The history of dispersal of N. neesiana in Australia is poorly documented because of a widespread lack of recognition of its 

presence and failure to accurately identify it. This appears to be a generalised problem with invasive grasses in native grasslands 

on a world basis, at least for members of the general public (Witt and McConnachie 2004). In southern Europe N. neesiana has 

been widely confused until recently with the closely related N. mucronata (Kunth) R.W. Pohl (Verloove 2005). In New Zealand, 

an infestation established at Hawkes Bay in the 1950s was not identified until 1982 (Slay 2001). In Australia, an infestation 

discovered at Tamworth, NSW, in 1996 had an estimated age of 30 years (Cook 1999). In discussing whether or not N. neesiana 

was spreading in the ACT, the ACT Weeds Working Group (2002 p. 2) suggested that “an increase in ability to identify the 

species” may have been the reason for its “presence/abundance ... now being noted”: until awareness campaigns were 

implemented, few rural landholders were aware of its existence and even fewer could identify it. 

According to Walsh (1998) N. neesiana has probably often been mistaken for native spear grasses Austrostipa spp. Such 

mistakes have been made “even by herbarium botanists” (McLaren et al. 1998). The first known Australian collection, made in 

October 1934 at Northcote (Melbourne) was originally mis-determined as the native Stipa elatior (Benth.) Hughes (a synonym 

of S. scabra var. elatior Benth., now Austrostipa flavescens (Labill.) S.W.L. Jacobs and J. Everett) (McLaren et al. 1998, 

39

McLaren, Weiss and Faithfull 2004), a common native species with a very wide distribution across southern Australia (Vickery 

et al. 1986). 

Lack of recognition is widely acknowledged: N. neesiana “does not stand out in the landscape” (Frederick 2002) and is “difficult 

to identify” (Auckland Regional Council 2002). Gardener et al. (1996a) state that its spread “went unnoticed” until about 1981, 

while Virtue et al. (2004 p. 85) refer to the “relatively unnoticed spread” of Nassella species. Duncan (1993) noted that plants in 

the vegetative state “can be mistaken for many other winter green species, chiefly Danthonia and Fescue”. Even with intensive 

public education and awareness raising, public reporting rates have been very low, with most new reporting by people with 

botanical interests or previous familiarity with the plant (Frederick 2002). The date of arrival in a particular area or site is rarely 

known, so the rates at which invasion is occurring can generally only be approximately calculated. 

Origin 

The geographical origin of N. neesiana populations in Australia and the means by which propagules first entered the country 

remain mysterious. The means of introduction has been said to be “unknown” (Grice 2004b), and a common assumption has 

been that the species is an accidental introduction. No published records of quarantine interceptions appear to exist. Carr (1993 p. 

291) listed both South and North America as possible origins of Victorian populations and was non-committal on the possibility 

of deliberate introduction. Kirkpatrick et al. (1995) suggested that New Zealand was “perhaps” the source. Slay (2002a) stated 

that N. neesiana was accidentally introduced into Australia in the late 1920s, but that is just speculation: the possibility of 

deliberate introduction has not been ruled out and the dates at which the species was first recorded are not necessarily a good 

guide to the actual dates of introduction. Caley et al. (2008) found in an analysis of herbaceous perennials naturalised in South 

Australia that the time between introduction and recorded naturalisation could extend up to 140 years, although the majority of 

species that naturalised did so within 40 years. A further complication in the Australian history of N. neesiana is the lack of 

certainty that the first herbarium specimens were from truly naturalised populations – current populations may have arisen from 

later introductions rather than being progeny of the earliest specimens. 

Deliberate introductions of the plant to Australia for evaluation as a pasture grass have been made (see below), but some records 

of wild individuals existed before all of the known deliberate importations. 

Agricultural contaminant 

Gardener et al. (2003a), influenced by numerous European records of alien stipoids associated with commerce, including 

Hayward and Druce (1919), suggested the possibility of introduction in wool, on sheep or in fodder. Hayward (Hayward and 

Druce 1919) had found the species downstream of the woollen mills at Galashiels, Scotland, where the adventive flora, totalling 

348 alien species, clearly reflected the naturalisation of species from seeds imported on wool from many parts of the world, 

including Argentina, Chile, Peru and Bolivia (Vines 2006). 

Reflecting the diverse range of possible introduction methods, Muyt (2001) suggested that N. neesiana arrrived “probably as an 

agricultural impurity”, while Benson and McDougall (2005 p. 159) considered it an accidental introduction “probably as an 

agricultural contaminant”. In a newspaper article, quoting no source, Dalton (2000) stated that it was “believed to have 

hitchhiked to Australia on a haybale”. Slay (2001) stated that seeds were distributed to Australia on exported pelts and wool, a 

presumption apparently based on the 19th century European experience. Australia however has long been a major exporter of 

wool and sheep skins: wool exports were initiated in 1807, grew enormously from 1870 onwards and approximately quintupled 

in the following 70 years (Australian Wool Bureau 1963). Australia also appears to have been largely self sufficient in hide 

production for leather over the period when introduction may have occurred, except for a few specialised leather types, and had 

substantial exports of cattle, horse and sheep hides in 1938-39. However New Zealand was the chief source of hides and skins 

imported into Australia in the early 1960s (Anderson 1963). 

Evidence of establishment of exotic (Australian) stipoids in New Zealand indicates that the late 19th century was “a prime time” 

for dispersal (Connor et al. 1993 pp. 303-304), however N. neesiana was probably first collected in that country in the late 1920s 

(Connor et al. 1993). Establishment of Austrostipa spp. and other Australian grasses in the Canterbury area of New Zealand has 

been correlated with the importation of Australian sheep in the 1840s and 1850s (Connor et al. 1993). Slay (2002a p. 4) thought 

it “possible” that New Zealand populations of N. neesiana may have resulted from importation of contaminated heavy machinery 

from Europe for the construction of bridges and railways, but stated (p. 11) that there was anecdotal evidence that the 

infestations in Marlborough had most likely arisen from pasture seed imported from South America. 

Introduction to Australia in pelts or wool, or on sheep or cattle from South America or New Zealand seem unlikely given 

Australia’s long history as a significant exporter of these products, and its general low levels of trade with South America. Wild 

animal furs are a possible source of introduction: as recently as 1980 Argentina exported over 4 million Nutria (Myocastor 

coypus (Molina)) pelts to the USA and Europe (Soriano et al. 1992). Importation of fodder from these countries would seem to 

be similarly improbable, unless accompanying livestock. Introduction as an external contaminant of livestock is a possibility, the 

trade in racing horses perhaps being the most likely means. The possibility of importation as a seed contaminant deserves better 

evaluation, while further investigation of deliberate importation of potential pasture species may prove enlightening. 

Coincidentally, another major pest species native to South America, the Argentine Ant Linepithema humile (Mayr), established 

in Melbourne for the first time in Australia at about the same time as N. neesiana was first detected. The ant first came to official 

attention in 1939, but was originally noticed in Balwyn about 1931 (Clark 1941). Surveys from late 1939 to March 1940 found a 

second major infestation at Yarraville and Footscray, and small infestations in Williamstown, Brunswick and Caulfield (Hogan 

1940). The possibility that the same commercial vectors brought both species to Melbourne is perhaps worthy of further 

investigation. 

Deliberate introduction 

An alternative hypothesis is that of deliberate introduction, however the deliberate importations identified to date by weed 

researchers all post-date the first collection of N. neesiana in Australia. Cultivation of N. neesiana, leading to naturalisation 

would appear to be not unlikely, since “there are few modern examples of accidental first introductions of a weedy species to a 

new range” (Mack and Lonsdale 2001 p. 96) and a high proportion of invasive grasses are deliberately introduced (e.g. Lonsdale 

40

1994). Only 3 of 186 exotic grasses imported for potential use as pasture species in northern Australia between 1947 and 1985 

proved to be solely useful and not weedy, while 17% subsequently became weeds (Lonsdale 1994). 

N. neesiana seed would not have been difficult to obtain. It was, for example, included in a widely circulated list of seed 

available from the 1948 harvest at Uppsala Botanical Garden, Sweden (Nannfeldt 1949). N. neesiana was imported to Australia 

under the Commonwealth Plant Introduction (CPI) program, established in 1930 to introduce exotic forage and pasture plants for 

the ‘improvement’ of grasslands (Cook and Dias 2006). Two CPI importations are currently known, in 1945 (CPI accession 

numbers 9731) and 1951 (CPI accession number 13476) (Cook and Dias 2006). The US Department of Agriculture ran a similar 

introduction program and imported 11 accessions of N. neesiana to the USA between 1945 and 1972 (Cook and Dias 2006). Two 

were still listed as available for distribution in 2006 (USDA ARS 2006): PI 237818 from Spain, donated in 1957 and PI 311713 

from Chile, donated in 1966. Some of the USDA material may may have been exchanged with Australia (Cook and Dias 2006). 

From 1949 to 1952 N. neesiana was also imported to Canada from Argentina, Bolivia, Chile and Uruguay, and grown in 

experimental plots at the Plant Research Institute, Ottawa (Bowden and Senn 1962). 

The full extent to which CPI Nassella accessions were trialled or released in Australia has not yet been adequately investigated. 

The fate of imported forage plant material was “often poorly recorded” (Cook and Dias 2006 p. 610), and the absence of trial 

information does not necessarily indicate a failure to grow the species in the field. Cook and Dias (2006) were unable to list any 

evidence of N. neesiana testing, but the species was trialled in Western Australia (Rogers et al. 1979), a State where it not known 

to be currently established. Testing of the plant was a component of “small sward” “nursery” trials for the Western Australian 

perennial grasses evaluation program during the period 1943-1968. The material grown was found to have ‘little promise’ (low 

to moderate productivity or survival or low leaf/stem ratios) at CSIRO Glen Lossie Field Station, Kojonup (some time between 

1951 and 1968), and ‘some promise’ (fair to moderate production and fair to good survival) at Muresk Agricultural College 

(1943-47). Many other identified and two unidentified “Stipa” and Nassella spp. were also evaluated under this State 

Government program (Rogers et al. 1979). In nothern Australia, trial plots for testing potential pasture species were “often 

simply abandoned after use” leaving the plants free to naturalise and spread (Lonsdale 1994 p. 350). But although other tested 

grasses in the Western Australian trials have subsequently established in the Kojonup and Muresk areas, Nassella spp. have not 

been found (Sandy Lloyd, Agriculture Western Australia, in litt. 18 February 2008). 

Cook and Dias (2006 p. 608) mistakenly claimed that Ratcliffe (1936) discussed the potential use of exotic “Stipa” for arid land 

rehabilitation. However Ratcliffe’s study was influential in the establishment of State government soil conservation authorities in 

Australia (Cook and Dias 2006), which experimented with a wide range of exotic plants. In Victoria, the Soil Conservation 

Service was involved in trial uses of Nassella after 1963 (Cook and Dias 2006), but according to Zallar (1981) these involved 

only N. hyalina (CPI No. 25801 from the USA), although another stipoid, Amelichloa brachychaeta, was also trialled. 

Detailed genetic comparison of world populations is probably capable of narrowing down the probable origin(s) of Australian 

material. 

Victoria 

McLaren, Stajsic and Iaconis (2004 pp. 64-65) suggested that an exotic stipoid grass ‘introduction epicentre’ in the northern 

suburbs of Melbourne was possibly linked to the trotting stables of Edgar Tatlow, east of Darebin Creek in Epping, and the 

possible importation of horses and hay from South America. This suggestion was sourced to a personal communication by P. 

Haberfield, who linked Nassella leucotricha to the Tatlow property, presumably marked today by suburban Tatlow Drive, east of 

Epping Secondary College. Numerous streets to the east including Trotting Place and Derby Drive suggest the probable extent of 

the property. Nassella leucotricha, a Mexican and southern USA species, was locally known as Tatlow grass and had, like N. 

neesiana, first been recorded in Victoria at Northcote in October 1934 (McLaren et al. 1998, McLaren, Stajsic and Iaconis 

2004). 

Tatlow, who died at the Epping property, ‘Derby Lodge’ stud, in March 1968 (Anon. 1968), cannot be blamed for the original 

introductions of N. neesiana or N. leucotricha to Victoria, because these precede the transfer of his business from Tasmania. 

However he may have been responsible for later Nassella introductions. Records indicate that in 1938 he purchased the horse 

‘Raider’ in the USA and imported it to his Tasmanian stud, also called ‘Derby Lodge’, at Hagley (Pedigree Online 2007), near 

Launceston. Tatlow frequently imported horses, including ‘Globe Derby’, imported to Hagley in 1927, ‘Belle Logan’ from New 

Zealand, and ‘Ayr’, bought in Christchurch in 1932. He also regularly visited “America”, where, in 1954, he purchased Stanton 

Hall and Volo Chief (Anon. 1968). Tatlow moved to Epping, some time after 1938. “Most of the broodmares at his studs in both 

Tasmania and Victoria were purchased in NZ, many from Southland, and he was a regular visitor to America, where he 

purchased ... successful sires” (Anon. 1968). N. neesiana might have been imported with horses from New Zealand, where it was 

probably present from the late 1920s, although it has never been recorded from Southland or Christchurch (Slay 2002a), and N. 

leucotricha might have accompanied animals purchased in the USA, where it is native in Texas and Oklahoma (McLaren, Stajsic 

and Iaconis 2004). 

An equine connection otherwise appears to be sound speculation. Horses frequently create areas of bare ground in pastures, are 

commonly provided with supplementary fodder, and pass a high proportion of consumed seed in their dung. Horses readily carry 

weed seeds externally and create ‘windows’ on the ground surface for their germination and establishment, so horse pastures are 

commonly much weedier than those grazed by sheep and cattle (Gurr et al. 1996). Presence of grass seeds such as Bromus 

diandrus in horse dung and other data (Weaver and Adams 1996) suggest the possibility that horses could well have dispersed N. 

neesiana seed to Australia. 

There is substantial evidence for the northern suburbs of Melbourne ‘Nassella epicentre’ hypothesis. McLaren et al. (1998) 

referred also to a [another?] report by P. Haberfield of N. charruana being present near Cooper Street at Epping from before 

1958. The east end of Cooper St is about 900 m south of Tatlow Drive. The first official record of this species in Australia was a 

specimen collected on 21 February 1995 at Thomastown by A. Muir (Hansford 2006). Thomastown is about 3.6 km south of 

Tatlow Drive, downstream along Darebin Creek. Later in 1995 N. charruana was found on a rural property at Epping in an 

infestation believed by the property owner to have been present since the 1950s (Hansford 2006). Mapping of 45 infestations that 

41

were known by April 1996 showed that they were all along the Darebin Creek and its tributaries, within or close to the Epping 

Road corridor (Hansford 2006). At its closest point, Epping Road is 400 m from Tatlow Drive. 

Similarly the first recorded Victorian infestation of N. trichotoma, c. 4 ha, in 1958 at Broadmeadows (McLaren, Stajsic and 

Iaconis. 2004), “where it was believed to have existed for about 20 years” (Parsons 1973), was possibly about 8 km from Epping. 

Searches soon after the first discovery found five other patches “just north and north-east of Melbourne” in areas “apparently ... 

at one time butchers’ holding paddocks ... presumably carrying sheep from infested areas in New South Wales” (Parsons 1973 p. 

152). However both N. neesiana and N. leucotricha were first recorded in 1934 at Northcote, c. 14 km S of Epping (McLaren et 

al. 1998). 

Ian Suitor, a farmer with property off Somerville Road, Greenvale (pers. comm. 22 November 2006) reported that the first place 

N. neesiana appeared in his area was on land managed by Tommy Thomas, a dealer in livestock, on a granitic hill, 

approximately 0.5 km north of Greenvale Reserve. Previously it had been believed (Charles Grech pers. comm. 22 November 

2006) that Suitor’s land had been invaded downstream along Moonee Ponds Creek. 

N. neesiana was not recorded as a component of the basalt plains flora by Willis (1964), and was not found by Groves (1965) at 

St Albans, where it is now common. Its distribution in Victoria was described by Willis (1970 p. 182) as: “locally frequent on 

basaltic grassland north from Melbourne (Fairfield, N. Preston, Broadmeadows etc.) and along a railway embankment at N. 

Brighton.” In 1972 its known distribution was still reportedly restricted to major grid N (the Melbourne region) (Churchill and de 

Corona 1972), however Gardener (1998) recorded a distribution point near Koroit in the Western District with the first collection 

there in 1967, shown in the ARMCANZ et al. (2001) map, which also showed a 1960s locality in the Horsham region. Stuwe 

and Parsons (1977) did not record it at any of 59 T. triandra grassland remnants they surveyed in 1976. 

By 1986 it had been recorded from “Bung Bong”, Yan Yean; Purnim; Woodstock, south east of Whittlesea; and Rosanna 

(Vickery et al. 1986 p. 81, herbarium material examined). Beauglehole (1987) recorded it in sector F of the Wimmera, bounded 

roughly by Horsham, Dimboola, Jeparit, Nyamville, Minyip and Lubeck. In the western region of Melbourne, McDougall (1987) 

recorded populations in remnant native vegetation at Napier Park, Essendon; O’Brien Park, Sunshine and the nearby Braybrook 

Rail Reserve Grassland, and Laverton North Grassland Reserve, Altona North, and recommended control or monitoring at each 

site. Gardener (1998) recorded a distribution point near Geelong with the first collection there in 1984, and another in the North 

East in the Wangaratta region, first recorded in 1989. Bartley et al. (1990) described the plant as a serious threat to remnant 

grasslands in the Melbourne area, including the Laverton North Grassland Reserve, Derrimut Grasslands and railway reserves. 

They summarised the then known Victorian distribution: major grid cells N (Melboure region), K (near Warrnambool) and C (in 

the Wimmera) (see Willis 1970 or Churchill and de Corona 1972 for a grid map), and sectors D, F and O within the Land 

Conservation Council’s Melbourne study area (citing Beauglehole 1983). Staff of the Botany Department, LaTrobe University, 

had also found it at sites in sector B to as far to the south-west as Geelong (Grid P) (Bartley et al. 1990). Gardener (1998) 

recorded a distribution point to the west of Bendigo, with the first collection there in 1990. Carr et al. (1992) summarised the 

distribution as ‘limited’, in medium to large populations (rather than widespread, or rare and localised, or with small populations) 

in lowland grassland and grassy woodland and rock outcrop vegetation. Stuwe (1994 p. 94) warned that it threatened “to 

dominate several grassland remnants near Melbourne”. 

Walsh (1994) recorded the presence of N. neesiana in seventeen 6’ x 6’ grid cells (c. 9 x 10.8 km or c. 10,000 ha) based on 

records in the Flora Information System (managed currently by the Victorian Department of Sustainability and Environment), 

National Herbarium of Victoria records, and other “verifiable” reports. These were concentrated in the area west and north of 

Melbourne (8 grid cells) including Whittlesea, with 4 cells in the Ballarat-Maryborough-Tarnagulla area, 2 cells in the North 

East (Wangaratta-Beechworth area), 2 in the Geelong-Torquay area and 1 in the Warrnambool area. No Wimmera record was 

included. Walsh (1994 p. 378) mentioned also Cressy (65 km W of Geelong) but failed to map such a record. Kirkpatrick et al. 

(1995) noted that it was found at Cressy in 1994. Walsh (1994 p. 378) considered it “Locally common ... mostly on basalt soils 

and often near watercourses, but established and spreading on improved pasture and road verges”. 

By the 1990s N. neesiana distribution was ballooning. Kirkpatrick et al. (1995 p. 35) stated that its “rate of spread in the two 

grassland reserves in Melbourne [Derrimut and Laverton North] [had] shocked botanists”. Intensive publicity and Victorian 

government investigations from c. 1993 greatly increased knowledge of its occurrence. The Victorian Government employed 

four regional facilitators, in the Port Phillip, North East, South West and North West regions, and they undertook mapping of 

infestations, with data being recorded on the Pest Management Information System (later the Integrated Pest Management 

System, IPMS), a Department of Primary Industries database (Iaconis 2006a). Mapping was also undertaken in the Rural City of 

Wangaratta, and of roadside infestations in the Shire of Indigo (Iaconis 2006a). In the North Central Region, Liebert (1996) 

reported 4 urban infestations in Bendigo, 1 infestation at Campbell’s Creek, 1 at Clunes, 3 at Maryborough, 2 at Tarnagulla and 

2 at Tooleen, mostly discovered by local field naturalists, and mostly on roadsides and railway sidings. Precise details were 

provided of the location and extent of each of these infestations along with a description of the habitat and the nature of the 

infestation. Almost all of them were on disturbed soils. An up-to-date Victorian distribution map (attributed to David McLaren) 

was also provided, showing that approximately 23 minor grids (6’ x 6’) contained infestations, including one in the North East 

near Wangaratta and one in the South West near Terang. N. neesiana found in the Sunbury-Bulla areas was believed to have 

originated by dispersal of seed or roadside slashers from the Greenvale district (Nair 1993). Stewart (1996) described extensive 

cover in Broadmeadows Valley Park. 

According to Frederick (2002) the number of infested sites known in Victoria in 1998 was 43, and 338 in November 2001, 

approximately 60% on public land, nearly all on roadsides, along waterways or in flood zones, and mostly small patches or 

isolated plants. Infested areas totalling 350 ha had been discovered in that period with an estimated total ground cover of c. 24 

ha, reduced to c. 15 ha after treatments. Frederick (2002) provided a detailed tabulation of the number of infested sites (land 

parcels or management units) and the year of their discovery at 19 localities in the North Central Region. New infestations were 

still being discovered in well investigated areas at the time of writing. In the Western District the plant had been detected in the 

Moyston area by 2000 (Dalton 2000). 

42

According to Morfe et al. (2003) the reported area of infestations in Victoria by 2002 was 815 ha. Extensive further mapping 

occurred throughout the State in 2003 (Iaconis 2006a). 

Matthews (2006) reported a sharp increase in the spread and occurrence in south-western Victoria. In addition to a large 

established infestation at Lake Hamilton and surrounding urban areas of Hamilton, there were various roadside populations in 

Southern Grampians Shire, and infestations in the general areas of Mt Napier Road, Digby Road, Kirkwood Road, roads between 

Portland and Digby Roads, Murndal Road and Lake Hamilton, with some spread from roadsides onto adjacent private land. 

Infestations have recently been found on the Mornington Peninsula, south of Melbourne, by Gidja Walker (pers. comm. 7 

December 2007). A “small colony of about 20 plants” was found on the roadside at Point Nepean Road next to the Shelley 

Beach turnoff, opposite Campbells Road, between Portsea and Sorrento (Anon. 2008a, Walker pers. comm.). It was also found at 

Rye c. late 2005 associated with the Mobil service station, growing in scoria (Walker pers. comm). 

Major infestations were reported between Bairnsdale and Lindenow (100 and 60 ha) and at Dargo (several properties) in East 

Gippsland by Geoff Harman of the Department of Primary Industries in January 2009, much further east than the previously 

known range in Gippsland (Harman pers. comm. 12 January 2009). 

New South Wales 

The development of the invasion in New South Wales is poorly documented in published sources. A substantial contributing 

factor has been the absence of a State-wide weed mapping system (Linda Iaconis pers. comm. 2006). 

The first known collection of N. neesiana in New South Wales was on the Northern Tablelands at Glen Innes in 1944 (McLaren 

et al. 1998) or 1948 (Gardener 1998). Duncan (1993) recorded that it had been widely present in the Guyra – Glen Innes area, in 

the northern half of the Tablelands from the 1960s. A major infestation, discovered in 1996 in the Reedy Creek catchment near 

Tamworth on the North West Slopes, had an estimated age of 30 years (Cook 1999). Tamworth is off the tablelands, on their 

inland, southwestern side. Slay (2001 p. 21) erroneously mentioned a “continuous sward” infestation at Deniliquin, citing 

Mulham and Moore (1970), but those authors refer only to Austrostipa spp. Duncan (1993) and Storrie and Lowien (2003) noted 

that N. neesiana was thought to have spread very slowly until the late 1970s. According to the map in Gardener (1998) it was 

first recorded in the Guyra district in 1968. A herbarium specimen from Bathurst on the Central Tablelands was collected in 

1972 (Benson and McDougall 2005), shown in the ARMCANZ et al. (2001) map. It was first recorded in the Sydney area in 

1974 (Gardener 1998) a herbarium specimen from Mt Druitt (Benson and McDougall 2005), and at Tenterfield close to the 

Queensland border in 1976 (Gardener 1998), both shown in the ARMCANZ et al. (2001) map. 

In 1986 Vickery et al. (1986 p. 81) examined material collected at Mt Druitt, Tenterfield and Glen Innes and considered it to be 

“now spreading” on the Central Coast and Northern and Southern Tablelands. It was recorded in the Goulburn region in 1985 

(Gardener 1998). Specimens from Pendle Hill, Ingleburn and Brush Farm Park in the Central Coast region were collected in 

1986 (Benson and McDougall 2005). It was recorded in the Armidale region (Northern Tablelands) in 1990 (Gardener 1998). 

Wheeler et al. (1990) added no new information. Jacobs and Everett (1993) reiterated its presence in the four major botanical 

divisions of the State already noted (not including the then unknown Tamworth population) and stated that it “grows along 

roadsides”. It was not recognised as an environmental weed in NSW by Swarbrick and Skarratt (1994). Carolin and Tindale 

(1994) recorded that it was widespread on roadsides in the Sydney region (bounded approximately by Rylstone, Singleton, Nowra 

and Taralga). In 1995 it was found in the Coonabarabran region (Gardener 1998). 

Gardener et al. (1996a) observed that N. neesiana then dominated large areas of pasture on the Northern Tablelands and was 

becoming increasingly common on the Central and Southern Tablelands. Eddy et al. (1998) noted its presence at Bungendore 

(Southern Tablelands). By 1998 it had been declared noxious in the New England County (Northern Tablelands), Severn Shire 

and Glen Innes Shire (McLaren et al. 1998). 

Ens (2002a) collated herbarium and other records and searched for populations on the Cumberland Plain (Sydney region), 

finding 16 infested sites, mostly with

Figure 4. Distribution of N. neesiana in New South 

Wales (National Herbarium of New South Wales 

2009). 

Australian Capital Territory 

Published information about early records of the grass in the ACT are scanty. Gardener (1998) dated the first record to 1960. The 

ACT Weeds Working Group (2002 p. 2) vaguely noted that it was “known to have occurred in the ACT for some time, but was 

not considered a species of concern until the late 1990s”. Vickery et al. (1986 p. 81) considered it was “now spreading” in the 

ACT, and examined material from Burbong, Black Mountain, Commonwealth Gardens in Canberra, and O’Connor. Berry and 

Mulvaney (1995) did not list the species as a significant weed in the ACT: it was not considered a “widespread or dominant” 

environmental weed, nor a “common weed” of any major habitat type except grasslands and road verges, and was not recorded 

from 40 or more 2.5’ x 2.5’ grid cells (c. 3.5 x 4.5 km), nor known to be dominant over an area of > 30 x 30 m. The weed 

database at that time recorded zero locations for it in the ACT. In fact the species was not properly recognised in the ACT in this 

period, and its presence was being overlooked (Sarah Sharp pers. comm. 11 October 2006). Berry and Mulvaney did note 

however (1995 Vol. 2 Appendices p. 261) that it was a weed of grasslands in Canberra, “observed as a dominant in a small patch 

of grassland in the Barton area ... also common along the bicycle path at Yarramundi” (at the western end of Lake Burley 

Griffin). 

Eddy et al. (1998) considered it common in the ACT. Bruce (2001) surveyed 39 sites in the ACT including natural grasslands, 

parks and road verges in rural, urban and periurban areas and found N. neesiana to be widespread. It was most common in urban 

and peri-urban areas subject to mowing and was present at all urban and peri-urban sites investigated, but at only 67% of rural 

sites. Abundance data suggested that the invasion in the older urban areas was of longest standing and at a more advanced stage. 

Lowest levels of abundance were found on agricultural land and grazed areas (whether or not managed for agriculture). 

Further surveys from 2000 to 2002 (ACT Weed Working Group 2002, ?Sharp 2002) revealed expansion of known infestations at 

numerous sites, plus numerous previously unrecorded patches and linear infestations along primary, secondary and semi-rural 

roadsides leading outwards from major infestations. The Tuggeranong Valley contained the most noteworthy severe new 

infestations. N. neesiana was found to be dominant in numerous areas in inner Canberra including most suburban nature strips 

and roadsides, and around Parliament House. ACT Government (2005) recorded that it had spread dramatically in abundance 

and distribution in the previous 10 years. The highest abundances in the ACT were in the central city and at Belconnen, with 

lower abundances in the Jerrabomberra and Majura districts, Gunghalin, but was absent from sites examined in the Tuggeranong, 

Tidbinbilla and Namadgi districts. 

South Australia 

The first known collection in South Australia was at Lucindale in the South East on 18 November 1988, where, according to J.P. 

Jessop of the Adelaide Herbarium, it ‘did not seem to be causing any trouble’ (McLaren et al. 1998). Gardener (1998) mapped a 

1988 record in the vicinity of Mallala (north of Adelaide). Gardener et al. (1996a) mentioned its presence in the Adelaide Hills, 

and Gardener (1998) dated the first Adelaide area record as 1989. By 2000 it was recorded from the South East (Thorp and 

Lynch 2000) and was known from the Okagparinga Valley by late 2000 (Obst and How 2004). Infestations at Belair National 

Park were also identified. Field surveys in spring-summer 2003 enabled the mapping of all known infestations in Adelaide- 

Fleurieu Peninsula (Iaconis 2006a). 53 infestations in the Mt. Lofty Ranges, Fleurieu Peninsula and greater Adelaide regions 

totalling 14.0 hectares had been recognised up to December 2003, including Modbury (moderate to heavy, 0.07 ha), Adelaide 

Parklands (5 plants removed by hand), Clarendon (one site, 0.02 ha, low density, grazed pasture) and Wirrina (several sites, 

13.77 ha) (Obst and How 2004). Further surveys in 2004 located a few additional sites. During 2003-2005 the original survey 

area was extended to include Randall Park and adjoining roads and railway infrastructure to Belair National Park, but no new 

infestations were discovered (Iaconis 2006a). Jessop et al. (2006) recorded its presence in the Northern Lofty region (near 

Bundaleer), Southern Lofty region and the South East. 

Queensland 

The presence of N. neesiana in Queensland may have been first recorded in published literature by Mallet and Orchard (2002). 

Michael Hansford (in Iaconis 2003) noted its presence in “southern” parts of the State, in the Darling Downs. Extensive 

surveying for N. neesiana was undertaken in 2005 and 2006 including roadside mapping in Clifton and Warwick Shires, with a 

delimiting survey planned for the whole of the eastern Darling Downs (Iaconis 2006a). As of October 2006 infestations were 

restricted to Clifton, Warwick and Cambooya Shires, and property management plans had been finalised for all known 

infestations (Phil Maher, Queensland Department of Natural Resources and Mines, in Iaconis (2006b)). Infestations were 

concentrated in the Darling Downs, with some along the Condamine River, and approximately 100 ha were known to be infested 

by 2007 (Snell et al. 2007). The largest infestation was at Clifton Showgrounds and polo field where anecdotal evidence suggests 

it may have been present since at least 1977 and have given rise to the other infestations (Snell et al. 2007). Additional 

44

populations have been found in the City of Toowoomba, probably spread by an energy utility company (Queensland 

representative on the National Chilean Needle Grass Taskforce 28 February 2007). 

The Queensland infestations most likely are the result of seed movement via vehicles or livestock from the Northern Tablelands 

of New South Wales. A plan with the long term objective of eradication from the State is being implemented (Snell et al. 2007). 

Tasmania 

An infestation at Hobart is mentioned by Mallett and Orchard (2002) and Hocking (2005b). This infestation, at the University of 

Hobart, was mapped in 2005 and was the only one then known in Tasmania. The outbreak was sprayed with herbicide in autumn 

2005, with a follow-up spray, mulching and planting of trees in winter. Other potential invasion sites were surveyed in 2006, 

including open space and grasslands at Montague Bay, Rose Bay, Rosny and Mornington (Iaconis 2006b). New infestations 

were discovered in 2005 that prompted further surveying (Iaconis 2006a). However according to the Tasmanian Department of 

Primary Industries and Water (DPIW)(2007 p. 2) N. neesiana had been recorded from only a single site, in Hobart, “from which 

it has ... been eradicated”. 

Hocking (pers. comm. 2006) noted that the recent reports were in central and southern Tasmania. As of July 2009 the core 

infestations were all located in urban settings on the eastern shore of the Derwent estuary in Hobart in the suburbs of Montague 

Bay, Rosny and Bellerive (National Chilean Needle Grass Task Force Agenda, July 2009). The main infestations were along 

walking tracks, road easements and the grounds of a primary school. It was also found west of the Derwent River, in small areas 

of The Domain and near the Technopark (Karen Stewart, DPIW Tasmania, pers. comm. July 2009). DPIW mapped and treated 

all known infestations during 2007-08. Hocking (pers. comm. 2006) noted that bioclimatic modelling indicated that 

establishment was highly likely in areas around Devonport and Launceston where ferries from mainland Australia regularly 

deposit large numbers of motor vehicles. Bioclimatic modelling indicated that much of eastern Tasmania was suitable for its 

establishment (ARMCANZ et al. 2001), although Hobart was outside the predicted area (Hocking 2005b). DPIW (2007) 

prescribed hygiene and quarantine procedures to prevent breaches of the Tasmanian importation prohibition. 

This basic data for the whole of Australia suggests at least four very widely separated areas of introduction, excluding the 

deliberate importations to Western Australia: on the northern outskirts of Melbourne, the New England tablelands of NSW, 

southern NSW or the ACT, and South Australia, however dispersal from the earliest known area of infestation north of 

Melbourne to the other major foci cannot be ruled out. 

It is apparent that the known extent of infestations during the course of the N. neesiana invasion has been closely related to the 

ability to identify the plant and the effort devoted to detecting it. As Grice and Ainsworth (2003) pointed out, the confounding 

‘awareness factor’ means that very little can be concluded about the actual rates of population increase and dispersal. Ongoing 

control activity further confounds any attempts to determine whether N. neesiana might be reaching the limits of its potential 

range and population size. Nevertheless, understanding of the invasion would be improved through retrospective mapping of the 

extent of infestations at appropriate historical intervals and would assist in theoretical discussions of lag phases and the time 

courses of plant invasions, and potentially contribute to improving prediction abilities for weeds in general (Grice and Ainsworth 

2003). 

Potential distribution in Australia 

Various predictions of the potential Australian distribution of N. neesiana based on climate parameters have been made. These 

all suffer from a poor understanding of the range of the grass, both in South America and in countries where N. neesiana has 

been introduced, resulting from misidentifications or uncertainty about the true classifications of material, and inadequate floral 

studies, and in the introduced ranges, of continued colonisation of new areas. 

Gardener (1998) estimated a potential range in Australia of nearly 40 million ha, from Western Australia to south-east 

Queensland, using the climate modelling program CLIMATE, which analysed 16 parameters for each data point, based on 

temperature and rainfall. He made two separate analyses, using firstly the 224 then known locations of infestations in Australia, 

and secondly a sample of locations in the rest of the world, based on herbarium specimens and literature records. 

McLaren et al. (1998 p. 63) used distribution and altitude data from a “representative sample” of known infestations in Australia, 

analysed the data with the BIOCLIM program and used CLIMATE to predict areas with a similar climatic profile. Their map 

showed a potential distribution range covering much of south-east Queensland, most of the eastern half of New South Wales, 

most of Victoria, southern parts of South Australia and Western Australia and parts of north-east Tasmania, covering a total of 

41 million ha. A map provided by D. McLaren in Liebert (1996) derived using BIOCLIM based on the climatic parameters of all 

Victorian infestations then known, showed the potential distribution in Victoria was approximately 3.86 million ha. These 

climate-based predictions ignore other potentially important factors affecting distribution such as land use, tree cover and soil 

type (Liebert 1996). They also rely on geographical proximity of weather stations to assign climate parameters to a record, a 

problem likely to lead to major distortions when records are a long distance from weather stations, at a different altitude, etc. 

Under the State of Victoria’s weed risk assessment process (Weiss 2002, Morfe et al. 2003), the maximum potential distribution 

of N. neesiana was estimated based on world distribution data, climate modelling (rainfall and temperature data) and 

geographical information system layers of susceptible land uses, vegetation classes and soil properties. A map was produced 

showing the likely and unlikely distributions (Morfe et al. 2003p. 879). An additional estimate was made that N. neesiana would 

take about 75 years to occupy its entire potential range in Victoria, or 50 years if the rate of spread was “very high”. Any future 

expansion of range and area occupied was acknowledged to be critically dependent on the extent and effectiveness of 

management activity directed against the weed. In the absence of government-coordinated mangement, the level of infestation in 

Victoria was predicted to expand from less than 1000 ha to c. 1.5 million ha in 30 years (Morfe et al. 2003). 

Thorp and Lynch (2000) provided a grid map generated by Agriculture Western Australia based on the known world distribution 

of N. neesiana, using the CLIMATE system with ‘core’ and ‘high’ predicted densities (± 20% of the average climate required). 

45

This indicated that a small part of south eastern NSW, parts of Tasmania, much of Victoria, south-eastern South Australia and 

parts of the southern coast of Western Australia had high climatic suitability, but ‘core’ areas were restricted to near-coastal 

Victoria and South Australia. 

According to Gardener et al. (1999 p. 8) N. neesiana “appears to have the ability to colonise most grasslands and grassy 

woodlands in temperate areas of Australia with more than 500 mm of rainfall”. The expansion of range of an invasive plant at a 

regional level is generally faster from many, small infestations than from a single large one of similar area (Mack and Londale 

2001). The presence of many small populations is certainly a feature of the Australian invasion, so continued rapid spread can be 

expected. Increases in the frequency and density in landscapes already invaded can be expected if inadequate management is 

undertaken. Lunt and Morgan (1998, 2000) documented an increase in frequency at Derrimut Grassland Reserve from 16% of 

quadrats in 1986 to 42% in 1996, largely driven by senescence of the dominant native grass T. triandra. 

Habitat and climatic and biotic tolerances 

N. neesiana is an “extremely hardy” (Benson and McDougall 2005) grass, adapted to a wide range of conditions (Muyt 2001). It 

reportedly “has the capacity to switch from being stress tolerant to being vigorous” (Bedggood and Moerkerk 2002). In South 

America it is the most widely distributed Nassella species with a distribution from the Pacific coast across the Andes, through 

the Pampean and Paranaense biogeographical provinces to the Atlantic, a range similar to the overall distribution of Stipa sens. 

lat. in that continent, and occupying a variety of habitats (Longhi-Wagner and Zanin 1998), “primarily” steppe (Barkworth 

2006). 

N. neesiana is a particularly prominent species in the 750,000 km 2 Rio de La Plata grasslands, consisting of the Argentine 

pampas and the Uruguayan and southern Brazilian campos (Soriano et al. 1992). Campos and pampas are similar formations and 

the Rio de la Plata is often considered the boundary between them (Overbeck and Pfadenhauer et al. 2007). The “typical” 

landscape of the Pampas south of Buenos Aires in 1960 was described by Durrell (1964) as “flat golden grassland in which the 

cattle grazed knee-deep ... a lush, prosperous and well-fed-looking landscape that only just escaped being monotonous”. Lorentz 

(1876 quoted by Schimper 1903 p. 503) described the pampas grasslands as consisting of “scattered dense tufts of stiff grasses, 

chiefly species of Stipa and Meliea, which rise like islets above the yellowish-brown loam ... between these isolated tufts of grass 

bare loam, which is frequently washed out and carried away by the rain, so that the separate tufts of grass rest on actual mounds; 

but also frequently, especially during the favourable season of the year, it is covered by all kinds of more delicate grasses and 

herbaceous perennials, few in species ... Viewed from a distance, these grasses seem to form a close grassy covering, and the 

pampa presents the appearance of extensive grassy tracts whose colouring varies with the seasons: coal-black in spring, when the 

old grass has been burned; bright bluish-green when the young leaves sprout; later on brownish green, the colour of the mature 

grass; finally – at the flowering time – when the silvery white spikes overtop the grass, over wide tracts its seems like a rolling, 

waving seas of liquid silver ... After the Gramineae ... the greatest number of individuals is that of Compositae; usually twiggy 

under-shrubs with inconspicuous flowers ...Verbena, species of Portulaca, of Malva and a few Papilionaceae are chiefly 

responsible for the meagre floral beauty ...”. A similar picture is presented by Soriano et al. (1992): Poaceae is the predominant 

family, with Stipa sens lat. the best represented genus (25 spp.), followed by Piptochaetium and Poa (each with 8 spp.) and 

Aristida and Melica (each with 6 spp.); Asteraceae are next most abundant, including species of Baccharis, Eupatorium, 

Hypochoeris and Veronia; native Fabaceae, sensitive to cultivation, poorly represented, with Cyperaceae, Solanaceae, 

Brassicaceae, etc, in order of abundance. In the Brazilian Campos region Asteraceaeare is the most diverse family (c. 600 spp.), 

followed by Poaceae (c. 400-500 spp.), Leguminosae (c. 250 spp.) and Cyperaceae (c. 200 spp.) (Overbeck et al. 2007). 

N. neesiana was a dominant species in a climax grassland community that once covered the majority of the formation known as 

Rolling Pampa, extending south and west of the Río Paraná and the Rio de La Plata and encompassing the Río Salado basin. 

Rolling Pampa covered a gently undulating, well drained plain, commonly with very deep soils, and regular drought and 

flooding. In the very fertile soils common in Rolling Pampa, such as in La Plata district of Buenos Aires Province, N. neesiana 

was dominant along with the shortly rhizomed summer-autumn growing Bothriochloa laguroides (DC.) Herter and three smaller, 

tufted grasses Piptochaetium montevidense (Spreng.) Parodi, Aristida murina Cav. and Jarava plumosa (Spreng.) S.W.L. Jacobs 

and J. Everett (Cabrera 1949, Soriano et al. 1992). This type of grassland is called “flechillar”, in reference to the dominance of 

grasses with piercing seeds (Soriano et al. 1992). Shrubs are generally a minor component but include Baccharis spp. 

(Asteraceae), and the inter-tussock spaces are occupied by many species of small herbs and sedges (Soriano et al. 1992), with the 

“original” plant diversity in well-drained soils being c. 222 species (Aguiar 2005). The weed flora has many species in common 

with the more mesic grasslands of south-eastern Australia. In other areas of Rolling Pampa Nassella charruana and Amelichloa 

brachychaeta were the dominant grasses, the former in parts of the eastern zone and the latter in the north west (Sante Fe 

province) (Soriano et al. 1992). Distribution of N. neesiana in Rolling Pampa has probably been radically altered by 

development. Six million ha of the formation had been cultivated by c. 1900 but by 1984 this had increased to 26 m ha (Aguiar 

2005). 

N. neesiana was a dominant species in thhe natural vegetation of the Southern Pampa, covering much of southern Buenos Aires 

province, along with N. clarazii (Ball) Barkworth, N. trichotoma, N. tenuis (Phil.) Barkworth, Piptochaetium napostaense 

(Spegazzini) Hackel, P. leiopodum (Spegazzini) Henrard and Poa ligularis Nees ex Steud. with many other grasses also 

abundant and very similar non-grass components to Rolling Pampa (Soriano et al. 1992). Some of the stipoid species in Southern 

Pampa “often ... form pure stands in areas that have never been cultivated” (Soriano et al. 1992 p. 385). The stipoids are largely 

displaced by shrubby vegetation on hills, rocky sites and wet areas in this region. 

N. neesiana is one of the most abundant species in mixed flechilla-Paspalum quadrifarium Lam. grassland in the Tandilia Range 

system of the southern Pampas of Argentina, where the stipoid dominants include Piptochaetium spp. and Nassella trichotoma 

46

(Honaine et al. 2006). The flechillar (Nassella-Piptochaetium) community in this area has especially high biodiversity and is 

ecologically significant as a faunal refuge (Honaine et al. 2009). 

To the north in Argentina, in the department of Gualeguay in the south of Entre Rios Province, N. neesiana occurred in 

undulating treeless plains covered in grasses (Marco 1950), an area classed as Mesopotamic Pampa and characterised by a more 

subtropical climate (annual rainfall c. 900-1200 mm), and a network of water courses lined with gallery forest (Soriano et al. 

1992). Stipoids generally are of lesser importance in this formation than in the southern pampas, but N. neesiana was one of the 

dominants along with N. tenuissima, Eragrostis cilianensis (All.) Janchen. and species of Axonopus, Paspalum, Digitaria, 

Schizachyrium and Bothriochloa (Soriano et al. 1992). 

The Flooding Pampa, an extremely flat coastal plain covering 9000 km 2 (Perelman et al. 2001) in eastern Buenos Aires province 

that includes the Laprida and Río Salada basins and extends in tongues into the interior (Soriano et al. 1992), has some dominant 

grasses in common with Rolling Pampa but N. neesiana is not a major species (Soriano et al. 1992). Nevertheless it occurs 

through most of the region across its latitudinal range (c. 35-38°S) and with a frequency of ≥20% (in 25 m 2 samples) in 8 of the 

11 Flooding Pampa grassland community types identified by Perelman et al. (2001), reaching a frequency of 87% in one 

community. In the Flooding Pampa it has been classified in a three-species floristic group with Jarava plumosa (Spreng.) S.W.L. 

Jacobs and J. Everett (as Stipa papposa) and Bothriochloa laguroides (Perelman et al. 2001). 

N. neesiana is of minor importance in most of the Inland Pampa, a drier, more open grassland, lacking a river and stream 

network, found to the west and south-west of Rolling Pampa. However it is one of a number of grasses of lesser importance in a 

community dominated by Poa ligularis, Nassella tenuissima, N. trichotoma, N. filiculmis (Delile) Barkworth and Panicum 

urvilleanum Kunth on the border between the provinces of Buenos Aires and La Pampa (Soriano et al. 1992). Its minor 

importance in the Caldenal, an ecotonal region between the Humid Pampa and the Monte desert, is reflective of the regional 

climate with an annual rainfall of 300-400 mm concentrated from spring to autumn, annual potential evaporation of 800 mm and 

cooler temperatures than the pampas (Fernández et al. 2009). 

Little native grassland remains in Uruguay, and protected areas of natural grassland landscape are non existent (Cosse et al. 

2009). In the Juan Jackson region of south-western Uruguay N. neesiana was considered to be the third most important grass in 

“virgin and regenerated” caespitose grasslands on granitic soils, after Piptochaetium stipoides (Trin. and Rupr.) Hackel ex 

Arech. and P. montevidense (Spreng.) Parodi (Gallinal Heber et al. 1946). Soriano et al. (1992) considered it to be the most 

frequent grass in the Southern Campos, a formation extending over most of southern part of the country, along with other 

Nassella spp., Paspalum spp., Poa spp. and other grasses. On deep fertile soils the natural vegetation was probably dominated by 

Nassella charruana. In the Atlantic coastal south-eastern plain Stipeae are less abundant, being displaced by an abundance of 

subtropical genera, and are dominant only in relatively restricted “upland prairies” located on hills and knolls with shallow soils, 

dry in the summer (Iriarte 2006). 

In Brazil N.neesiana is most common in ‘cleared and treed fields’ (rough translation of Longhi-Wagner and Zanin 1998). It is 

not an important component of the natural vegetation of the Northern Campos of northern Uruguay and the campos of the 

southern Brazilian State of Rio Grande du Sol, which have fewer Stipeae and more numerous representatives of Paniceae and 

Andropogoneae (Soriano et al. 1992). N. neesiana was not found by Overbeck and Pfadenhauer (2007) in surveys on Morro 

Santana, near Porto Alegre in Rio Grande do Sul, near the eastern edge of Brazilian campos, an area with rainfall exceeding 

1200 mm per annum and no dry season, but Overbeck et al. (2007 p. 106) list it (as Stipa setigera C. Presl) as a subtropical area 

“characteristic” species of grasslands in the southern Rio Grande do Sul of Brazil, along with N. megapotamia, N. nutans (Hack.) 

Barkworth and N. philippii (Steud.) Barkworth. The northern boundary of the Pampean province of southern Brazil is at about 

30°S (Overbeck and Pfadenhauer 2007). 

On the Juan Fernández Islands of Chile it is found on rocky slopes near the sea, in dry or humid gorges, rocky cliffs, on and 

between rocks and in open places (translation from Baeza et al. 2007). 

In south-western Europe most exotic Stipeae occur on disturbed land or higly modified areas and have failed so far to occupy 

more natural areas, with the exception, to date, of N. neesiana, which “penetrates into natural, protected areas ... and behaves like 

an aggressive invader” in the Canary Islands (Verloove 2005) e.g. in the Garajonay National Park, Anaga Rural Park and Osorio 

Reserve (Sanz-Elorza et al. 2005). Since the mid 1960s it gradually spread along roads and through gaps and sparse vegetation 

on the island of Tenerife in nitrogen enriched environments (Martín Osorio et al. 2000). The larger Tenerife infestations occur on 

hard, basaltic substrates (Martín Osorio et al. 2000). South-western European N. neesiana “inhabits a wide range of usually 

anthropogenic biotopes varying from road-verges, abandoned vineyards, urban parks ... more or less ruderalized pastures, etc.” 

but also occupying rocky slopes in river valleys in France and Italy (Verloove 2005 p. 108). In Catalonia (Spain) small 

populations have been found on path margins and in abandoned vineyards (Font et al. 2001). 

In New Zealand it occurs mainly in pastures and on roadsides (Edgar and Connor 2000) in dry, low-fertility, open habitats 

(Bourdôt and Ryde 1986) on various soil types, but not normally in fertile pastures where there is good competition from other 

plants (Bourdôt and Ryde 1986, Auckland Regional Council 2002). It occurs on “rolling hills and flats” at Waipawa (Bourdôt 

and Ryde 1987a), and often occurs on “steep inaccessible eroded land” (Bourdôt 1988). It invades unimproved pastures, 

infesting two of seven paddocks dominated by native tussock grasses in the Lake Grassmere area (Bourdôt and Hurrell 1989a). 

Its invasiveness in New Zealand sheep pastures was attributed by Bourdôt and Hurrell (1989a) to adaptations enabling survival 

in semi-arid, poorly fertile environments, rather that high competitive ability. 

The little information available on the native vegetation formations invaded by N. neesiana in areas of its exotic range outside 

Australia is summarised in Table 3. 

47

Table 3. Native vegetation formations invaded by N. neesiana outside Australia. 

Vegetation class Locations Country Major associated spp. Reference 

Bidenti pilosae-Ageratinetum 

adenophorae community 

Teneriffe, Canary Islands Spain Bidens pilosa 

Ageratina adenophora 

Myrica faya, Salix canariensis 

Martín Osorio et al. 2000 

(Erica platycodon): Ilici Teneriffe, Canary Islands Spain Martín Osorio et al. 2000 

canariensis-Ericeto playcodonis 

Aeonietum cuneati Teneriffe, Canary Islands Spain Martín Osorio et al. 2000 

Periploco-Phoenico canariensis Teneriffe, Canary Islands Spain Phoenix canariensis Martín Osorio et al. 2000 

Within Australia, occurrence in disturbed anthropogenic habitats is usual: e.g. North Central Region of Victoria: roadsides, 

waterways, small neglected semi-urban allotments (Frederick 2002), Sydney region “roadsides, pastures, grassy areas, along 

creeks and rivers” (Benson and McDougall 2005). In the ACT it occurs under a wide range of conditions “from heavily mown or 

grazed sites, from damp gullies and depressions out to drier slopes, to sites with low to medium levels of disturbance and in 

shaded and unshaded locations” (Bruce 2001). Areas in which it is most likely to occur include roadsides, tracks and paths 

(vehicle, walking, animal), mown areas, waterways and drainage lines (Bruce 2001). It is generally more abundant in grasslands 

that are small in size (

In Australia the altitudinal range is from near sea level (e.g. in the Altona area near Melbourne) to c. 1200-1400 m at Guyra on 

the Northern Tablelands of NSW (Gardener et al. 2005). In the Sydney region it is found from 0-600 m (Benson and McDougall 

2005). 

Climate 

Stipeae are generally classic xeromorphs or mesic, Austrostipa aristiglumis (F.Muell.) S.W.L. Jacobs and J. Everett and 

Trikeraia being exceptions (Arriaga and Jacobs 2006). In south-eastern South America members of the tribe increase with 

increasing latitude, being absent in the megathermic zones, a major vegetation component in mesothermic zones and dominant in 

Patagonia (Perelman et al. 2001). Indeed, increases of c. 15 to 23% in the average relative cover of Stipeae compared to other 

grass tribes has been measured across the narrow latitudinal range (c. 35-38°) of the Flooding Pampa, in Argentina (Perelman et 

al. 2001). The distribution of particular Stipeae, including several Nassella species, in this region shows strong latitudinal 

correlations, with a group including N. hyalina and N. charruana found only in the north, and others found only in the south 

(Perelman et al. 2001). 

N. neesiana is unusual in that it has a very wide climatic tolerance. It has invaded areas outside the climatic range it occupies in 

South America (Gardener 1998). The stronghold of N. neesiana, the pampas region, has warm, often humid summers, mild, 

usually drier winters with temperatures uncommonly below freezing, annual rainfall of 600-1000 mm, and little drought (Mack 

1989). In the pampas, “grass is driven out only where water is very abundant in the soil, as ... along the banks of rivers ... the 

climate is a perfect grassland-climate, with ... rainfall not more than moderate but well distributed ... humid, moderately warm 

vegetative season [and] strong winds ... with moderate atmospheric humidity ... hostile indeed to woodland” (Schimper 1903 p. 

459). Summer rainfall is prevalent in eastern South America south of 30° including Rio Grande do Sul, eastern Uruguay and 

Argentina (Schimper 1903). Restricted occurences in southern Brazil, where the grasslands are considered to be relicts of drier, 

cooler conditions, may be determined by the more subtropical, humid climate (Overbeck and Pfadenhauer 2007). N. neesiana is 

one of the dominant grasses in La Plata district of Buenos Aires where the climate is warm temperate and humid, with a mean 

annual temperature of 16.5 ºC and mean annual rainfall of 992 mm (Cabrera 1949). It occurs across the Flooding Pampa of 

north-eastern Argentina which has annual rainfalls of 850-900 mm, average annual temperatures of 13.8-15.9°C and average 

minimum temperatures in July of 1.8-6°C (Perelman et al. 2001). It is a major grass in the Tandilia Range of south-east Buenos 

Aires province where the annual rainfall is about 800 mm and the mean monthly temperatures for the coldest and warmest 

months are 13 and 23°C (Honaine et al. 2009), and in parts of south-west Buenos Aires Province, where the climate is humid 

temperate with a mean annual temperature of 14.5ºC and annual average rainfall of 850 mm (Amiotti et al. 2007). It occurs in 

the littoral areas and lower parts of the La Plata basin where the climate lacks a marked dry season (Schimper 1903). Large tracts 

of northern Argentina, the base of the Andes and the Provinces of Entre Rios and Corrientes have annual rainfall of 1000-1200 

mm (Schimper 2003) and in those areas N. neesiana loses its dominance in grassland to species better adapted to a wetter, more 

subtropical climate (Soriano et al. 1992). In Uruguay N. neesiana is supposedly “resistant to adverse climates” (Gardener et al. 

1996b p. 3, citing Rosengurtt et al. 1970). In the Pampean province of southern Brazil the annual rainfall is in the range of c. 

1200-1600 mm and mean annual temperatures are 13-17°C (Overbeck et al. 2007). On the Juan Fernández Islands of Chile it 

grows in dry or humid areas (Baeza et al. 2007). 

In continental South America as a whole N. neesiana is found in the 500-1500 mm annual rainfall zones, with some extreme 

outliers, and its northern distribution is “limited by lack of low winter temperatures necessary for vernalization” (Gardener et al. 

1996b), so is restricted to higher altitudes in the tropics. It occurs in southern Chile near Valdiva where the average annual 

rainfall exceeds 2600 mm and maximum/minimum temperatures are 22.8/11.1 in January and 11.1/5.0ºC in July (Gardener et al. 

1996b). Similar seasonal temperatures occur in montane-alpine areas of south-eastern Australia, although few such areas have 

rainfall exceeding 1600 mm. It occurs in areas of far north-east Argentina near Posadas, where the rainfall was over 1900 mm 

per annum, with January and July maximum/minimum temperatures of 33.2/21.8ºC and 21.9/11.4ºC (Gardener et al. 1996b). 

The closest approximations to such a climate in Australia occur in the wet tropics. In western Argentina it occurs at Medoza, 

where the climate is very dry (223 mm) with January/July maximum/minimum temperatures of 32.0/19.4ºC and 14.7/2.4ºC 

(Gardener et al. 1996b), similar to some arid areas of inland southern Australia. In arid regions of South America it may only 

occur where the microclimate is more mesic, e.g. along rivers (Gardener et al. 1996b). 

In South Africa N. neesiana has been recorded in temperate summer rainfall areas (Wells et al. 1986). The climate parameters of 

one of its stronghold areas in New Zealand are an annual average rainfall of 650-900 mm, warm summers with frequent droughts 

and moderate winters, conditions said to be very similar to the climate in its Argentinian range (Bourdôt and Hurrell 1989a). The 

900 mm isohyet has been quoted as a limiting factor in north-facing hill slopes in the Hawkes Bay area of New Zealand (Slay 

2002) where its potential range was considered to be “northerly facing hill country in low to medium rainfall (700-900 mm)” 

(Slay 2001 p. 5). 

In south eastern Australia N. neesiana occurs primarily in areas with 500-800 mm average annual rainfall (Muyt 2001) in a range 

of climates including the warm wet summer/cold dry winter of the Northern Tablelands of NSW, and hot dry summer/cold wet 

winter of southern Victoria (Gardener et al. 1999). In the Sydney region it is found in areas with 700-1000 mm mean annual 

rainfall (Benson and McDougall 2005). 

N. neesiana appears to be well adapted to seasonal dryness (Bourdôt and Hurrell 1987b) and is drought tolerant (Muyt 2001, 

McLaren et al. 2002b, Slay 2002c, Storrie and Lowien 2003). Tsvelev (1977 p. 2) thought that the small chromosomes and small 

pollen grains (general stipoid characters), lemma and palea structure protecting the flowers, a dense clumping habit and the 

presence of intravaginal shoots (as in N. neesiana) all indicated specialisation for xerophytic climates. Slay (2002a p. 10) thought 

that the bare ground resulting from drought was “the catalyst for seedling establishment”. 

Fire 

Fire is a feature of the campos and pampas grasslands of South America, but as in other areas of the world, the ancient and pre- 

European fire regimes are poorly understood. Darwin (1845 p. 114) reported that the plains from Bahía Blanca to Buenos Aires 

were commonly deliberately burnt, “chiefly for improving the pasture”: “it seems necessary to remove the superfluous 

49

vegetation by fire, so as to render the new year’s growth serviceable”. Great fires across the pampas were reported in 1853 

(Soriano et al. 1992). 

The fire history of Southern Brazil is “poorly known” but palynological studies indicate that fires were rare before c. 7400 years 

ago and then became frequent, possibly as a result of deliberate burning for hunting (Overbeck and Pfadenhauer 2007 p. 28). In 

southern Brazilian grasslands, where N. neesiana is native, fire became more frequent from the beginning of the Holocene, 

proabably as a result of the advent of human populations, and currently grazed native grasslands are burned approximely 

biannually, usually in August (Overbeck et al. 2007). Most of the grassland species of southern Brazil “seem to be adapted to 

frequent ... burning” (Overbeck et al. 2007 p. 107). 

Honaine et al. (2009) stated that winter burning, along with grazing, in the flechillar community of the Tandilia Range of Buenos 

Aires province, in which N. neesiana is a major component, led to the dominance of Achillea millefolium L. and Carduus 

acanthoides L. (Asteraceae). Historical observations suggest that fire occurred at intervals of 5 years in the stipoid dominated 

savannah grasslands of the more arid Caldenal, to the south and west of the Pampean region, before the introduction of livestock 

(Fernández et al. 2009). 

Bourdôt (1989) noted that in New Zealand “regular burning promotes the grass” and that after fire “tussocks quickly begin to reestablish 

from clandestine seeds present in old tiller bases”. Liebert (1996 p. 15, quoting Craig Bay on the plant’s behaviour at 

Organ Pipes National Park, Victoria) stated that it is “usually the first grass to resprout after fire”. Stewart (1996) recorded that 

dense N. neesiana quickly regained high cover after a wildfire at Broadmeadows Valley Park, Victoria, in December 1994, 

reaching 50% cover after approximately 6 months and 70% cover after 7 months, and reaching a height of over 50 cms after 11 

months. Kirkpatrick et al. (1995 p. 35) considered that it “generally ... recovers more quickly than other species” and is “as well 

adapted to fire as the native dominant” T. triandra (op. cit. p. 82). However the precise timing of fire in relation to rainfall likely 

has a large influence on which plant species recover first. 

Young seedlings and small plants are possibly killed by fire, but larger plants lose most of their above-ground biomass and 

resprout from protected buds. In the classification used by Overbeck and Pfadenhauer (2007) the seedlings are “non-sprouters”, 

while tussucks are fire “resistors”, retaining some dead stem biomass after burning. The above-ground meristems are protected 

by the densely packed basal leaf sheaths (Overbeck and Pfadenhauer 2007), some of which burn, but many of which resist 

burning because of their high moisture content and tight packing and merely char on the outside. Thus, older tussocks are best 

adapted to frequent burns. 

Britt (2001) found that burning, after application of 1 litre of methylated spirits to square metre areas surrounded by a metal 

frame, eradicated adult plants in an infested pasture. Hocking (2005b) reported on small scale experimental burning of thickets at 

the Iramoo native grassland, Victoria, in late spring and summer 2002. Both early and late spring burning resulted in less than 

10% of mature tussocks resprouting after the autumn break, much lower than unburnt treatments, a reduction of more than 75% 

in mature tussocks, but an increase in the number of small tussocks and very immature tussocks, likely resulting from 

fragmentation of what had previously been large plants. Late spring burning removed all viable seed from the site, reduced seed 

production in the subsequent fruiting period by c. 50% and appeared not to lead to any major seedling recruitment, up to and 

including the following spring. Early spring fire had no effect on subsequent seed production. 

As with other grasses, the specific effects of fires are likely to be dependent on the intensity, severity and precise timing of the 

fire. Seed that has found its way into the soil is likely to survive, presumably in a similar way to that of Austrostipa. 

Other disturbances 

Weed invasion of natural communities is often facilitated by disturbance, and the more frequent, intense or prolonged that 

disturbance the greater the invasion is likely to be (Fox and Fox 1986, Carr et al. 1992, D’Antonio et al. 1999). According to 

Weber (2003 p. 280) in a survey of environmental weeds of the world, based on a very limited literature review, N. neesiana 

“invades mainly degraded and disturbed plant communities”. Bartley et al. (1990), commenting on the situation in Victoria, 

wrote that “prior disturbance does not appear to be necessary for invasion” for N. neesiana invasion of native grasslands. This 

was echoed by Kirkpatrick et al. (1995 p. 35), who nevertheless added that “its spread is certainly facilitated by soil 

disturbance”. According to Slay (2002a p. 4) N. neesiana “evolved under conditions of low disturbance”, however Gardener 

(Gardener 1998, Gardener et al. 1996, 1999, 2003b) found that N. neesiana seeds only germinate on bare ground, in gaps or 

areas bared by herbicides and other disturbances, and that seedlings only survive in bare areas. The plant survives well in the Rio 

de Plata grasslands where trampling by cattle “is the main anthropic disturbance” (Soriano et al. 1992). Bedggood and Moerkerk 

(2002) recorded regrowth of N. neesiana in wheel tracks after treatment of infestations with glyphosate using vehicle mounted 

wick wipers. Liebert (1996) recommended the avoidance of unnecessary soil disturbance to minimalise N. neesiana invasion. 

Bruce (2001) attempted to determine the importance of different types of disturbance in determining the level of N. neesiana 

infestation in native grasslands in the ACT. She assessed the current extent of four ‘disturbance types’ on a four point scale from 

absent (0) to high (5): bare ground, other exotic weed populations, soil disturbance (earthworks, erosion, cultivation, animal 

diggings etc.) and the amount of refuse dumping (garden waste, soil, rubbish). The literature contains numerous mentions of the 

importance of ‘bare ground’ to enable seedling establishment. The usual meaning appears to be soil lacking other vascular plants 

above ground and devoid of litter. Bruce (2001) found that N. neesiana had wide levels of abundance, from absent to dominant, 

where bare ground was present at low and medium levels. It had the largest range of abundances (0-5) at high weed levels, and 

the lowest (0-3) at low weed levels (perhaps suggesting that N. neesiana facilitates invasional meltdown). The single site with no 

soil disturbance had occasional patches of the weed, sites with low soil disturbance had a range of infestation levels (0-5) and 

sites with medium soil disturbance were overall more highly invaded (1-5). No sites had high soil disturbance. Infestation levels 

varied widely for all levels of dumping, but N. neesiana was observed to have established and be spreading from scattered piles 

of lawn clippings at one site. 

50

Proximity to urban development appeared to be the most important predisposing factor for N. neesiana invasion in the ACT, and 

use of the land as urban open space appeared to almost guarantee that it would become infested, probably as a result of seed 

dispersal by mowing (Bruce 2001). 

No-one appears to have comprehensively evaluated past disturbance in relation to present N. neesiana infestations. Historical 

disturbance patterns are more difficult to determine and there are more difficulties in assessing their intensity, frequency and 

duration. Consensus opinion appears to be that disturbances of various types can facilitate invasions and it appears likely that in 

those cases where invasion has been recorded without disturbance, there has been inadequate appreciation of the effects of prior 

disturbances. The effects of various disturbances including fire, grazing and herbicides on N. neesiana establishment and 

survival are discussed in more detail below. 

Shade 

N. neesiana has been called a sun loving (“heliófilas”) species (Martín Osorio et al. 2000 p.39) but is adapted to moderate shade 

(Muyt 2001). It has considerable shade tolerance, up to medium level tree densities, e.g. in wooded areas around Bendigo, 

Victoria (Hocking 1998). Bruce (2001) found it was commonly present in shaded areas in the ACT , e.g. in windbreaks, very 

open woodlands, under oak trees, pines, eucalypts including Eucalyptus melliodora A. Cunn. Ex Schauer, Acacia spp. and 

shrubs. It grew in both shaded and unshaded locations at 59% of sites investigated (Bruce 2001). In the Sydney region it occurs 

in woodlands with Eucalyptus moluccana Roxb. and E. tereticornis Sm. (Benson and McDougall 2005). In New Zealand, N. 

neesiana plants growing under 25 year old Pinus radiata are generally weaker than plants growing in the open and have reduced 

seeding potential (Slay 2002a). N. neesiana was present in 3 of 6 treed paddocks examined by Bourdôt and Hurrell (1989a) in 

New Zealand. In the Canary Islands it “thrives” under the evergreen Laurus azorica (Seub.) Franco and the deciduous Castanea 

sativa Mill. (Verloove 2005). 

The literature appears to contain no precise informat.ion on the effects of solar radiation levels. N. neesiana plants growing under 

Eucalyptus in the ACT in early-mid summer were much greener and in an earlier reproductive stage than those in the open 

(Bruce 2001). Plantations with 2500 trees per ha have been established in New Zealand to examine the effects of shading as a 

possible control option (Slay 2002a 2002c). 

Soils and nutrients 

N. neesiana occupies areas on a wide range of geological substrates with a diversity of soil types. The soils may heavy or light 

textured (Cook 1999) and have low or high fertility (Muyt 2001, Slay 2002c). 

In Argentina it is found on well-drained, drought-prone soils (Bourdôt and Hurrell 1989b citing Cabrera and Zardini 1978, Lewis 

et al. 1985). In the Argentine pampas it is found on deep, generally fertile loessic silts and clays (Gardener 1998). These soils 

were formed on very deep (>c. 300 m) deposits of silt or loess, rich in swelling clays, and are mainly mollisols (commonly very 

deep, high fertility argiudolls) or frequently vertisols, of very young age (Soriano et al. 1992). In the Flooding Pampa t is not a 

component of the floristic groups characteristic of deep, non-saline, well-drained soils, nor the group that characterises areas 

subject to long flooding, nor in areas with saline alkaline soils (Perelman et al. 2001). It is most frequent in areas with higher 

topographic position, being scarce in depressions and lower lying areas, and occurs over a pH range of c. 6-8.2, disappearing in 

the more alkaline soils that occur in wetter depressions and are correlated with salinity (Perelman et al. 2001). In the Ventania 

land system in south-west Buenos Aires Province the soils are formed on a more complex mosaic of Palaeozoic sedimentary 

rocks (Soriano et al. 1992) and are“Typic and Lithic Argiudolls and Hapludolls developed from pure loess sediments or mixed 

with rock detritus”, slightly acid (pH 6.4-6.8), a high saturation with bases, organic carbon levels > 5.5. g kg -1 and total N > 0.4 g 

kg -1 (Amiotti et al. 2007 p. 535). In the Gualeguay area of Entre Rios province it is found on light loam, with much organic 

matter to a depth of 40-60 and pH in the range 5-6.8 (Marco 1950). In the Tandilia Range it occurs on soils derived from loess 

over quartzite, that are typic Haludolls 10-25 cm deep and Argiudolls 20-70+ cm deep with organic horizons less than 5 cm deep 

(Honaine et al. 2009). Although widely dominant in the Rolling Pampas, it is replaced by Sporobolus spp., Jarava plumosa and 

other plants where the soils are slightly alkaline (Soriano et al. 1992). Mollisols and vertisols predominate also in the southern 

campos (Soriano et al. 1992). Soils of the Southern Pampa and Flooding Pampa are noteably deficient in P (Soriano et al. 1992). 

In Europe it is generally found on well-drained, “sometimes slightly eutrophic” soils and on rocky slopes (Verloove 2005 p. 

108). On Tenerife, Canary Islands, it prefers the nitrogen-enriched ruderal areas of fringe roads (Martín Osorio et al. 2000). 

In New Zealand the substrate rock types are loess, mudstones, and sandstones on slopes commonly 16-25° (Bourdôt and Hurrell 

1989a), and it prefers acidic soils, low in phosphate, calcium and summer moisture (Esler et al. 1993, Champion 1995). It is 

recorded from Yellow Grey Earths, particularly when acidic or low in P, Ca or summer moisture (Bourdôt and Hurrell 1989b). 

Slay (2002a p. 4) considered it “apparently better adaptated to low fertility sites” in that country. 

Gardener (1998 p. 10) considered it was then generally found in Australia on “more fertile ... predominantly volcanic” soils, but 

noted its presence on granitic soils at Tenterfiled and Emmaville, NSW, on rich clay at Lucindale, South Australia, and alluvial 

soils near Melbourne. Infestations are widespread on the primarily heavy clay soils derived from basalts on the Victorian 

Volcanic Plains. Liebert (1996) found infestations on granitic and sedimentary soils in North Central Victoria. In the ACT it does 

not tend to establish on slopes with westerly aspects and shallow soils (Bedggood and Moerkerk 2002). In the Sydney region it 

occurs on clays on shale substrates with medium nutrient levels (Benson and McDougall 2005). 

According to Bedggood and Moerkerk (2002 p. 6) N. neesiana “does not establish well where nutrient levels are really low such 

as hillsides”, but “does better on flats where nutrient levels are moderate”. When P is added to pastures, desirable pasture grasses 

respond better than N. neesiana (Bedggood and Moerkerk 2002) and increase their basal cover at its expense, with the effect 

apparent with or without strategic grazing at 300 DSE ha -1 (McWhirter et al. 2006). Grech (2007a 2007b) found that N. neesiana 

responds to phosphorus fertilisation. 

51

Landform 

N. neesiana is widespread in the pampas, which generally consists of completely flat plains alternating with gently undulating 

landscapes (Soriano et al. 1992), although it appears to be restricted to slightly elevated areas, i.e. away from minor depressions 

in Flooding Pampa (Perelman et al. 2001). N. neesiana occurs in the vegetation of other landforms in Argentina including rocky 

terrain in the low hills of the Tandilia Range of south-east Buenos Aires provine (Honaine et al. 2009). 

Bruce (2001) found that N. neesiana occurred mostly in the ACT on slopes (rather than valleys, watercourses, gullies or flats), 

where it had a wide range of abundances, and at sites with combinations of these landforms. Slay (2002c) considered that it 

colonised the dry northerly faces of hills in New Zealand. 

Water, drainage and flooding 

N. neesiana reportedly tolerates seasonal water-logging (McLaren, Stajsic and Iaconis. 2004) but the tolerable frequency and 

duration of such stress are not precisely known. In Argentina, Gardener et al. (1996b) found that it did not occur on wetter, 

often-inundated low ground, although such depressions were often saline, or on deep clays. Stipoids do not occur in the saline 

and alkaline soils of halophytic steppes of shallow, wet depressions of Flooding Pampa (Perelman et al. 2001) and appear to be 

entirely absent from saline soils in the pampas (Soriano et al. 1992). N. neesiana is not an important species in the Flooding 

Pampa of Argentina, an area subject to drought and very dry summers, and flooding “almost every year”, mainly lasting c. two 

weeks and 5-10 cm deep, with “extensive and lenthy flooding ... 3 to 6 times per century” (Soriano et al. 1992 pp. 394 and 374). 

In Australia N. neesiana was found on alluvial flats subject to seasonal waterlogging at Wattle Park, Burwood, Victoria by Geoff 

Carr (McLaren et al. 1998), often establishes in damp depressions (Liebert 1996) and on land subject to periodic inundation 

(Muyt 2001), and is “more common in low-lying, wetter areas” (Hocking 1998 p. 90). Swarbrick and Skarratt (1994) erroneously 

isted a single habitat known to be invaded, saline wetlands (atttributed to Carr and Yugovic 1989). Bruce (2001) found that it 

occurs in wetter areas in the ACT, as well as on slopes, and that an association with drainage lines, streams and damp 

depressions was apparent at most sites surveyed, with an infestation in one case radiating up-slope from a creekline. She 

illustrated a plant overhanging a drain, downstream of which patches of further plants were found. Kirkpatrick et al. (1995 p. 35) 

stated that it “can grow in a large range of soil moisture regimes”. Slay (2002c) noted that it thrives under moderate to severe 

stress due to low soil moisture. 

High rainfall in spring promotes panicle proliferation (Cook 1999) and plants flower in response to summer rain (Bedggood and 

Moerkerk 2002). N. neesiana prospers on roadsides where there is run-off and good soil moisture (Bedggood and Moerkerk 

2002). F. Overmars (in Iaconis 2006b) reported that it was growing and setting seed in drains in the Melbourne area in October 

2006 during severe drought. 

Other plants 

Other plants or plant associations in areas invaded by N. neesiana might provide biotic resistance to invasion, or may alone or in 

combination faciltate or prevent N. neesiana establishment or survival. 

The pampas grasslands, which have probably persisted without major change since the late Miocene (c.14 mybp ) (Webb 1978), 

are treeless and dominated by caespitose grasses with a wide variety of smaller herbs. In the Argentine pampas N. neesiana 

occurs as one element of a diverse grassland flora that normally includes summer growing grasses, including Bothriochloa, 

Panicum and Paspalum as common components (Gardener et al. 1996b). In the Flooding Pampa it is closely associated with 

Jarava plumosa and Bothriochloa laguroides, and is commonly found in communities with high frequencies of Paspalum 

dilatatum Poir., Piptochaetium montevidense, P. bicolor (Vahl) Desvaux, Vulpia sp., Bromus catharticus J. Vahl and other 

grasses and with a rich array of forbs (Perelman et al. 2001). It is “one of the dominant grassland species” in the pastoral zone 

around Buenos Aires (Gardener et al. 1996b). In the department of Gualeguay in south-west Entre Rios (Argentina) it is of 

secondary importance as a winter grass in pastures of the high campos (better drained areas with grasses maintaining a height of 

20-40 cm), along with N. hyalina, the more important winter-spring forage plants being Bromus catharticus, Lolium multiflorum 

Lam. and Medicago spp. (Marco 1950). The co-occuring spring-summer grasses of most importance include Paspalum spp., 

Axonopus compressus (Sw.) P. Beauv., Setaria spp., Andropogon saccharoides Sw. and Eleusine tristachya (Lam.) Lam. (Marco 

1950). In the Ventania land system in south-west Buenos Aires Province it naturally occurs in association with Stipa ambigua 

Spegazzini and Amelichloa caudata (Trin.) Arriaga and Barkworth in grassland with isolated trees of Prunus mahaleb L. 

(Amiotti et al. 2007). Genera that include dominant or subdominant species south of Buenos Aires include Agropyron, Aristida, 

Briza and Piptochaetium. Other Nassella species are widespread components with N. hyalina and Jarava plumosa common 

around Buenos Aires. In drier western areas Jarava ichu Ruiz and Pavon, N. tenuissima and N. trichotoma are more prominent, 

along with Poa ligularis (Mack 1989). 

In southern Santa Fe province (Pampean phytogeographical province) it is a characteristic species of a flechillar community 

along with N. hyalina and Jarava plumosa, with Bothriochloa laguroides, Sporobolus africanus and Carex bonariensis Desf.ex 

Poir. (Feldman et al. 2008). The Espinal phytogeographical province occurs in dier areas to the north, west and south-west of the 

Pampean phytogeographical province. Espinal vegetation is woodland or savannah with Prosopis, Acacia and Celtis as the main 

(woody) dominants but in San Cristóbal county of Santa Fe province includes a flechillar grassland community over 1 m tall 

strongly dominated by N. neesiana (Feldman et al. 2008). 

In the Tandilia Range of south-east Buenos Aires province it is a major species in a community of diverse grasses, the most 

abundant of which include Piptochaetium biocolor (Vahl) Desv., P. medium (Speg.) M.A. Torres, P. hackelii (Arech.) Parodi, 

Stipa bonariensis Henr. and Parodi, Briza spp. and Bothriochloa laguroides (DC.) Pilger, and some shrubs, but not in the 

monospecific pajonal community which is dominated by Paspalum quadrifarium, a grass that produces large accumulations of 

dead standing biomass in its tussocks (Honaine et al. 2009). 

N. neesiana is not a grass of major importance in the semi-arid Caldenal of Argentina, where the vegetation was orginally grassy 

steppe, but which is now largely savannah grassland with scattered trees, mainly Prosopis caldenia Burkart, known as caldén, 

and shrubland with rich shrub diversity. The main grasses are other species of Stipeae and Poa ligularis (Fernández et al. 2009). 

52

In the southern Brazilian campos of Rio Grande do Sul, N. neesiana occurs along with a wide range of both C 3 and C 4 grasses 

including Aristida spp., Paspalum spp., Piptochaetium spp. and other Nassella spp. (Overbeck et al. 2007). 

On Tenerife, Canary Islands, N. neesiana occurs in environments characteristic of the Bidenti pilosae-Ageratinetum 

adenophorae community, especially in areas cleared of vegetation along margins of roads and gutters and is especially prevalent 

relatively humid areas in gorge bottom inhabited by such species as Myrica faya and Salix canariensis (Martín Osorio et al. 

2000). The vegetation invaded was characterised in detail by Martín Osorio et al. (2000) (see Table 3). 

In Villa Ada, Rome, Italy, it has ‘colonised some hectares of hedges and lawns’ and it is found in grassy areas in Italy generally 

(Moraldo 1986 p. 217). 

These records in the native and invaded habitats indicate that N. neesiana coexists with a diverse array of other dominant and 

subsidiary grasses and a wide variety of forbs in natural grasslands but is rarely associated with trees and shrubs. 

In southern New South Wales N. neeisana forms dense monocultures that can dominate pastures (Verbeek 2006). In New 

Zealand pastures, dense clumps of N. neesiana also exclude other pasture species (Bourdôt and Ryde 1986) and replace more 

desirable grasses, particularly Lolium perenne (Bourdôt and Hurrell 1989b). Bourdôt and Hurrell (1989b) sprayed out a dense 

infestation of N. neesiana on low-fertility soil, rotary hoed the area and sowed plots of Lolium perenne, Dactylis glomerata 

L.and Phalaris aquatica. Plots were fertilised with superphosphate and lime or unfertilsed, and N. neesiana germinated 

uniformly across the area. 13 months later, N. neesiana ground cover in unsown areas was greatest in fertilised plots (69%) than 

unfertilised (53%) and its dry mass production in fertilised plots was also greater (3.43 t/ha vs. 2.72 t/ha). In Lolium plots N. 

neesiana cover was only 1% in fertilised plots and 11% in unfertilised plots. In Phalaris plots N. neesiana cover was 

approximately equal in fertilised and unfertilised treatments. In Dactylis plots N. neesiana cover was much higher in unfertilised 

plots (39%) than fertilised plots (19%). Plots were fertilsed again in year two. Over three years the dry mass of N. neesiana in 

both fertilised and unfertilised unsown plots as a proportion of total dry mass in the plot fell significantly from c. 95% to c. 75%, 

the balance being “litter and other species”, but mainly litter. In competition with the other grasses, dry mass production of N. 

neesiana as a proportion of total plot biomass was greatest with Phalaris. N. neesiana was less productive in competition with all 

three pasture grasses in fertilised plots, with the effect most pronounced for Lolium and least with Phalaris. Seeding of 

N.neesiana was considerably reduced in the sown plots compared to the unsown and most reduced in competition with Dactylis. 

In unfertilised plots with Lolium, N. neesiana became the dominant biomass component after three years, with Dactylis remained 

at a constant proportion over the period and with Phalaris declined as a proportion of total biomass. Fertiliser treatment induced 

temporal stability in terms of the proportion of contributions of the sown grasses and N. neesiana to total biomass production. In 

conditions of low fertility S.neesiana appeared to suppress Lolium, but under high fertility the dominance was reversed. S. 

neesiana was considered to be a stress-tolerant competitor, “evolved under conditions of low disturbance but moderate-severe 

stress from low soil moisture and probably low fertility” (Bourdôt and Hurrell 1989b p. 324). 

During the early period of invasion at Derrimut Grassland, Victoria, N. neesiana occurred occasionally in a Vulpia association 

often with Austrostipa bigeniculata, along drainage lines and in areas ploughed during the 19th century or subsequently heavily 

grazed (Lunt 1990a). This formation was considered to be occupying areas probably previously dominated by T. triandra. 

N. neesiana has poor competitive abilities with grasses and clovers that respond to high soil fertility (Connor et al. 1993, Liebert 

1996). Vigorous pastures can resist invasion including Phalaris, although Phalaris pastures are sometimes invaded (Bedggood 

and Moerkerk 2002). Grech (2007a 2007b) found little difference between Phalaris aquatica and N. neesiana responses to 

increased phosphorus. Lunt and Morgan (2000) found a negative relationship between T. trianda and N. neesiana at Derrimut 

Grassland Reserve which indicates that this competitive summer-growing C 4 grass can resist invasion. There is evidence for 

similar resistance to N. trichotoma by T. triandra (Hocking 1998) and Bothriochloa macra (Steud.) S.T. Blake when maintained 

in a healthy condition (Michalk et al. 1999), but distribution surveys suggest no such resistance is provided by C 3 wintergrowing 

native grasses (Badgery et al. 2002). N. neesiana is reported to have “choked out” N. trichotoma in trial plots (Hunt 

1996, McLaren et al. 1998) and to have invaded infestations of this grass (Liebert 1996 citing David Boyle). Bruce (2001) 

compared the level of N. neesiana invasion to the “botanical significance” rating of 39 grassland sites in the ACT and found no 

clear trends,both rich and poor sites having both zero and high level infestations. 

Such data indicate that N. neesiana can be excluded from vegetation that is dominated by healthy growth of other perennial 

grasses. Management practices that reduce competition by other plants, such as slashing of N. neesiana on roadsides, might 

therefore be counterproductive (Bedggood and Moerkerk 2002). 

Herbivory 

Grasses have coevolved with large grazing mammals and have a wide array of adaptations to grazing. One of the most important 

is the presence of intercalary meristems at the base of the leaves, rather than on the plant apices, a defence that probably played 

an important role in the evolution of the family (Stebbins 1986) and enable a plant to more readily regenerate after it is grazed. 

As in some other grasses (de Triquell 1986), the presence of multiple inflorescences on the panicle of N. neesiana allows the first 

developed, upper panicle to be sacrificed to herbivores, while the inflorescences closer to the ground and concealed beneath leaf 

sheaths, remain protected. The basal cleistogenes of N. neesiana are well protected from grazing mammals and develop even 

under conditions of heavy grazing (Dyksterhuis 1945, Gardener and Sindel 1998). N. neesiana thus has major advantages 

compared to Australian native grasses which lack cleistogenes. 

Little information appears to be available about the native mammalian herbivores that utilise or once utilised N. neesiana in 

South America. Before human occupation the pampean region was occupied by“outlandish humpless camels and giant flightless 

birds” (Crosby 1986 p. 159). Large grazing animals went extinct at about the same time as indigenous human occupation of the 

southern Brazilian grasslands in the early-mid Holocene (Overbeck et al. 2007). According to Overbeck and Pfadenhauer (2007) 

large native herbivores became extinct in the pampas c. 8000 years ago, at the end of the last glacial period. The Pleistocene 

megafauna and other large grazing mammals through Argentina, Uruguay and southern Brazil included such species as Toxodon 

platensis (Notoungulata), Glyptodon spp., Megatherium, Pampatherium sp., Stegomastodon platensis (Xenarthra), Equus sp.and 

53

Hispidon sp. (Perissodactyla: Equidae), Hemiauchenia sp., Lama gracilis, Lama guanacoe (Artiodacytla: Camelidae) and 

various Cervidae (Carlini et al. 2004, Norriega et al. 2004). The herbivorous megafauna of South America was comparatively 

large compared to other continents, and went extinct from the late Pleistocene into the Holocene (Johnson 2009). It appears 

probable that these animals strongly influened the distribution of grasslands and had strong selective effects on the grassland 

flora, although knowledge of these palaeoecological relationships is currently very deficient (Johnson 2009).. 

Its ancestors may have subject to predation by dinosaurs and probably Gondwanatheria, about which little is known, but which 

had teeth which indicate that grasses may have been major components of their diets (Prasad et al. 2005). 

Before Spanish occupation in the 14th century, the pampas lacked bovids (cattle and sheep) (Mack 1989) and large herbivores 

were scarce from the end of the Pleistocene until European colonisation (Fernández et al. 2009). An array of medium sized 

mammals was present, of which the the largest grazers were deer (Darwin 1845), and the camelids, the most important of which 

was the guanaco Lama guanicoe (Müller). Large herds of grazing animals were absent at least from the Tertiary until the arrival 

of Europeans (Coupland 1992). 

The camelids have broad, spreading, soft-padded feet that do not compact the soil (Castillo-Ruiz and Lundrigan 2007) and a 

digestive system with foregut fermentation, similar to ruminants (Siebert and Newman 1989). The Guanaco is a moderate sized 

herbivore about 1 m tall at the shoulders, which commonly forms herds of 4-18 with males troops of up to 200, and is a grazer 

and browser, consuming grasses, lichens, forbs, seeds, fruits, rushes and sedges. It has narrow foot pads and a more hoof-like 

nails than camels. The Alpaca Lama pacos (L.) and the Llama L. glama (L.) are domesticated forms, probably of the Vicuna L. 

vicugna (Molina) and the guanaco respectively, and all four of these taxa will interbreed. Recent evidence indicates that 

domestication of the wild species occurred 6000-7000 years ago. There were probably 10 million alpacas and several million 

llamas on the continent when the Spanish arrived in the 14th century, along with similar numbers of the undomesticed species, 

but within 100 years the populations collapsed, probably largely as a result of competition with domestic livestock (Long 2003), 

but no doubt greatly propelled by the destruction of the indigenous human cultures. 

The Vicuna is now restricted to the central and southern Andes in arid and montane grasslands at altitudes of 3500-4800 m, but it 

formerly had a more extensive range now occupied by exotic sheep and goats (Long 2003, Castillo-Ruiz and Lundrigan 2007). It 

too feeds on grasses and forbs (Long 2003). The reported diet of the alpaca in highland Chile is dominated by grasses including 

Festuca spp., Deschampia caespitosa and Agrostis tolucensis (Castillo-Ruiz and Lundrigan 2007). 

Darwin (1845) reported that the only large indigenous mammal still common in the extensive grasslands of the Maldonada 

region of Uruguay (near Montevideo) by 1852 was the Pampas Deer Ozotoceros bezoarticus L., but rheas Rhea americana (L.) 

were also common. The Pampas Deer is now extinct throughout the Rio de la Plata grasslands except in a few small reserves and 

ranches (Soriano et al. 1992, Cosse et al. 2009) while rhea populations have sharply decreased due to uncontrolled hunting and 

wire fences (Soriano et al. 1992). Cosse et al. (2009) investigated the diet of Pampas Deer on a ranch in south-eastern Uruguay 

and found that cultivated rice and Lolium sp. were the most important food plants in this agricultural environment, that the 

dietary overlap with sheep was complete, but that there was limited competition for food with cattle. Various grasses were 

recorded in the diet, not including Nassella or other Stipeae, although unidentified monocots generally comprised about one third 

of material found in faeces. 

Wild cattle Bos taurus L. and horses Equus caballus L. became established in Argentina after the abandonment of Buenos Aires 

by the Spanish in 1537 (Long 2003). 80,000 horses were present around Buenos Aires by 1585, large numbers of cattle were 

present by 1609 (Soriano et al. 1992) and millions of cattle and horses were present on the pampas by 1699 (Long 2003). 

According to Darwin (1845 p. 120): “The countless herds of horses, cattle and sheep, not only have altered the whole aspect of 

the vegetation, but ... have almost banished the guanaco, deer, and ostrich” (Rhea). Horses and cattle were introduced east of the 

Uruguay River by Jesuit missionaries in the 17th century and rapidly proliferated in the feral state (Overbeck at al. 2007). The 

whole of the campos and pampas have been grazed by sheep, cattle or horses “generally for centuries” (Soriano et al. 1992 p. 

380). Cattle grazing and other agricultural pursuits have greatly altered the landscape, with agricultural ecosystems replacing 

native grasslands in most of the region, and large areas of native grassland surviving only in Buenos Aires province due to 

wetter, more saline soils (Pedrano et al. 2008). As with other major grasslands lacking a recent history of grazing by hardhooved 

livestock (Mack 1989) the pampas grasslands were invaded by a large suite of exotic species after European 

colonisation. Livestock grazing in the Flooding Pampas has actively promoted these invaders, 75% of which are annuals and 

74% forbs (Perelman et al. 2001). 

Livestock grazing of natural grasslands and improved pastures is a major economic activity in the Rio de la Plata grasslands 

(Soriano et al. 1992) as is cattle ranching in the natural grasslands of southern Brazil, where 13.2 million head were present in 

Rio Grande do Sul state in 1996 (Overbeck at al. 2007). In Brazilian campos, grazing is “often considered the prinicpal factor 

maintaining the ecological properties and physiognomic characteristics of the grasslands” (Overbeck at al. 2007 p. 104). 

N. neesiana may have existed through much of the Quaternary period and into early historical times under grazing pressure from 

large mammals that was relatively light, and was then subjected in much of its core range to unprecedented grazing by feral 

livestock. Limited evidence indicates that N. neesiana declines under livestock grazing in its native range in Argentina. After 

exclusion of large ungulates from Flooding Pampa grasslands the relative basal cover of the major grass species (Briza 

subaristata Lam., Danthonia montevidensis (DC.) et Lam., Sporobolus indicus (L.) R.Br. and N. neesiana) more than doubled to 

97% after 4 years, while annual and broad-leaved herbs disappeared (Soriano et al. 1992). Under a management regime of 

grazing and winter burning in the Tandilia Range of Buenos Aires province, the major grass components of the flechillar 

grassland (including N. neesiana) decreased and the broadleaved Achillea millefolium. and Carduus acanthoides began to 

dominate, while under grazing alone C. acanthoides, Cirsium vulgare, Dactylis glomerata L., Vulpia dertonensis (All.) Gola. and 

Bothriochloa laguroides became more dominant (Honaine et al. 2009) 

The situation appears to be different in Australia where N. neesiana is generally considered to be tolerant of heavy grazing by 

sheep and cattle (Storrie and Lowien 2003, McLaren, Stajsic and Iaconis. 2004, Grech 2007a), in part because of its relatively 

low palatability to these animals compared with desirable pastures grasses (Bourdôt and Hurrell 1989a, Grech 2007a). It has 

relatively low herbage production compared to some such desirable pasture grasses used in Australia, notably Festuca 

54

arundinacea (Gardener et al. 2005). N. neesiana tillers profusely when grazed (McLaren, Stajsic and Iaconis. 2004) and 

produces large amounts of unpalatable flower stalks and very little leaf material during flowering and fruiting (Gardener et al. 

1996b, Gardener 1998, Grech et al. 2004, Grech 2007a). Stock avoid eating it in the reproductive stage even at stocking rates of 

300 DSE ha -1 (Grech 2007a). However plants are palatable to livestock during much of the year (Grech et al. 2004, Grech 

2007a). Slay (2002a) reported observations that ewes grazed N. neesiana in preference to Dactylis glomerata and Lolium 

perenne in winter near Napier, New Zealand. It can be reduced by appropriately timed grazing at high intensity, but the stocking 

rates and practices required are generally unrealistic for Australian conditions (Grech 2007a). 

The impact of goats on N. neesiana is unknown, but goats are able to substantially reduce infestations of N. trichotoma 

(McGregor 2003). Goats and alpacas have been considered as potential control agents in Australia (Bedggood and Moerkerk 

2002) but no trials appear to have been undertaken and no observations of N. neesiana herbivory by these species in Australia 

appear to be on record. Studies of biotic resistance (e.g. Parker et al. 2006a) suggest that alpacas would preferentially utilise 

Australian native plants or European grasses rather than N. neesiana or other South American grasses with which they may have 

coevolved. 

No studies of marsupial grazing of N. neesiana appear to have been undertaken. It is not recorded whether it is eaten by 

kangaroos and whether they prefer it to native grasses, as predicted by the biotic resistance hypothesis. Studies of the feeding 

preferences of Eastern Grey Kangaroos by Robertson (1985) indicate that species of Austrostipa and other grasses with sharp and 

long-awned seeds are avoided when their inflorescences are present, and that narrow-leaved species that develop substantial 

standing biomass of dead foliage that resticts selective foraging for green shoots suffer less damage. Thus macropod grazing of 

N. neesiana is probably similar to that by domestic livestock, in that it is avoided when in flower and fruit and has partial 

protection of green shoots resulting from retention of dead biomass, but will be selectively consumed when it is comparatively 

more palatable and accessible. 

Mammal grazing selects for prostrate or low-growing ecotypes of Stipa (Peterson 1962) and many other perennial grasses, but 

the low-growing forms of many grasses are the result of phenotypic plasticity (Cox 2004). The so called ‘sward form’ of N. 

neesiana in Australia (Randall Robinson pers. comm.) characterised by a large number of semi-prostrate stems and more limited 

development of an uptright tussock form, appears to occur where mowing is frequent as well as in heavily grazed situations, but 

whether the plants resume a more normal form if these pressures are relaxed is unclear. 

Grazing of various European pasture grasses has been found to lead to the development of ecotypes characteristed by small size 

and increased silica concentrations (Tscharntke and Greiler 1995). The relationships between the silica defences of grasses and 

grazing are as yet poorly understood, but it may be presumed that the altered grazing pressures in the introduced environment 

will select for different leaf hairiness and phytolith profiles characters. The possibility that such selection has altered the 

morphology or other characteristics of N. neesiana in Australia has not been investigated. 

Physiology and biochemistry 

The dominant native grasses in Australian temperate lowland grasslands are all C 3 species except for the major dominant 

Themeda triandra. Various other C 4 species commonly occur (e.g. Aristida spp., Dichanthium sericeum, Panicum spp.) and 

some form dense subdominant populations (e.g. Bothriochloa macra ) but none are dominant over wide areas. N. neesiana is a 

C 3 grass. C 3 and C 4 plants possess different biochemical pathways of photosynthetic carbon acquistion and consequently have 

markedly different leaf anatomy, different ratios of carbon isotopes in their tissues ( 13 C/ 12 C ratios or δ 13 C) and different 

responses to climatic factors (Hattersley 1986). The biochemistry of these differences is complicated and will not be explained 

here. In the simplest terms, the species that produce acids with four C atoms as the main initial stable product of CO 2 fixation are 

C 4 species, while those that produce the three-C acid 3-phosphoglyceric acid as the primary product are C 3 species (Salisbury 

and Ross 1992). There are three biochemical variants of C 4 photosynthesis in grasses, that differ in the main enzymes that 

catalyse the decaroboxylation of the four-C acids, the output products of that process and, with some additional compications, 

the leaf anatomy that supports the biochemistry (Hattersley 1986). Those that form malate utilise PEP carboxykinase and are 

called PCK species, while the two types that form aspartate utilise either NADP malic enzyme or NAD malic enzyme and are 

NADP-ME and NAD-ME species respectively (Hattersley 1986). Eleven differing biochemical/structural combinations have 

been identified. Andropogonanae have the classic structural NADP-ME type (Hattersley 1986), possessed also by T. triandra. 

C 3 grasses generally have a lower optimum temperature for photosynthesis, and grow better in cooler environments with less 

light than C 4 grasses, which often have temperature optima in the 30-40ºC range and are mostly native to unshaded environments 

(Monson 1989, Salisbury and Ross 1992, Sinclair 2002, Jessop et al. 2006), although some have a large ecological range 

(Christin et al. 2009). C 3 species are most numerous in areas with a cool, wet spring and become dormant in summer (Sinclair 

2002). In contrast C 4 grasses grow mostly in summer and are more efficient users of water and N, in terms of CO 2 assimilation, 

in high-light environments (Monson 1989). C 4 species have a competitive advantage when photorespiration costs become 

important, and are more efficeint in high temperatures and arid and saline environments (Christin et al. 2009) and much more 

efficient at producing biomass in warm environments with high light intensity (Salisbury and Ross 1992). The variations in the 

evolved biochemical-structural types of C 4 grasses likely have other ecophysiological and ecological implications, for instance 

differential leaf digestibility (Hattersely 1986). 

C 3 grasses tend to have shallower root systems than C 4 grasses, which require water for growth in summer when it is generally 

less available, so widespread replacement of C 4 with C 3 species may increase deep drainage to water tables, and contribute to 

dryland salinity (Sinclair 2002). 

Reactions of C 3 and C 4 species to increasing atmospheric CO 2 are complex, with neither group necessarily advantaged (Sinclair 

2002). The C 4 species grow more rapidly under elevated CO 2 levels and it is likely that they will expand southwards, but this is 

dependant on the N usage of C 3 spccies under the changed conditions (Keith 2004). 

Little is known of the specific biochemistry of N. neesiana, however a photosystem IIA protein, A9YFV6, has been described 

from N. neesiana chloroplasts (Boeckmann et al. 2003). Yeoh and Watson (1986) included N. neesiana in a study of the 

55

taxonomic patterns of protein amino acids in leaves and caryopses of grasses, but provided only aggregrate data at the tribal 

level. Seeds of Stipeae are high in Asx (aspartic acid or asparagine), glycine, methionine and arginine and low in glutamic acid, 

proline, alanine and leucine. They are nutritionally richer than seeds of Danthonieae, Aristida, Chloridoideae and Panicoideae, 

have protein contents equivalent to cultivated cereals and superior nutritional value to wheat, sorghum and maize varieties. The 

“Stipa” spp. analysed (including Nassella spp.) in aggregate are very similar to Oryza in leaf protein pattern except that Stipeae 

are lower in alanine and higher in valine. Stipeae, Oryza and Ehrharteae have the lowest free Glx (glutamic acid or glutamine) 

and the highest free Asx levels in the Poaceae. The variation in leaf amino acids are “unlikely” to be of any nutritional 

importance for mammals but may influence insect herbivory. 

Investigations have recently been undertaken by the Victorian Department of Primary Industries in an attempt to identify 

molecular fingerprinting techniques for identification of Nassella spp. (David McLaren, DPI, pers. comm.) but results have not 

been published. 

Breeding system 

N. neesiana is self fertile (Connor et al. 1993) but can cross pollinate. It produces both chasmogamous and cleistogamous panicle 

seeds, along with concealed cleistogamous seeds on the stem nodes (Slay 2002c) (see discussions above on floral morphology). 

Such a breeding system is well suited to colonising grassland habitats (Evans and Young 1972). 

Cleistogamy is more common in grasses than any other plant family and occurs in c. 19% of genera and 5% of species (Groves 

and Whalley 2002). Culley and Klooster (2007) found that it had been reported in 326 grass species, including 41 spp. of Stipa 

and 5 of Nassella. Two types of cleistogamy occur in Nassella, complete and dimorphic. Plants that are completely 

cleistogamous produce only cleistogamous flowers. Plants with dimorphic cleistogamy have both chasmogamous and 

cleistogamous flowers, the latter characterised by prominent differences in floral morphology including reduction in size or 

number of floral parts. In the cleistogamous flowers of grasses, the anthers, pollen sacs, stamens, stigmas, and lodicules are much 

reduced in size and the flowers develop no further than the bud stage (Brown 1952). The different flower types can be separated 

spatially on the plant, or temporally (often seasonally), or both, and in some plant species the ratio of flower types can vary 

between individuals and populations (Culley and Klooster 2007). The proportion of cleistogamous and chasmogamous florets in 

grass species is often under environmental control, e.g. Amphicarpum purshii produces few cleistogenous seeds after fire, but 

larger numbers as the likelihood of fire becomes high (Groves and Whalley 2002, citing Quinn 1998). Brown (1952) found that 

such facultative or ecological cleistogamy occurred in Nassella leucotricha in response to low availability of soil water. The 

distribution of cleistogamous and chasmogmaous fruits in the panicle depended on the soil water potential at the time of floral 

initiation. 

Chasmogamy allows for gene exchange between individuals via pollen, and has population and evolutionary advantages where 

there is greater environmental variability, or the prevailing genotypes produce phenotypes that are poorly adapted. Cleistogamy, 

on the other hand, ensures self fertilisation with consquent high uniformity of genotype frequencies, so maintains the existing 

frequencies of locally adapted phenotypes and genotypes (Groves and Whalley 2002, Culley and Klooster 2007). Meiosis occurs 

in both gametes of the cleistogamous flower, so gene resorting does occur (Groves and Whalley 2002), enabling continued 

“repatterning of the gene pool, differing only in degree from [that in] outbreeding populations” (Evans and Young 1972 p. 235). 

Cleistogamous seeds may be energetically less expensive to produce because of their greater rate of fertilisation and savings in 

pollen (Connor 1986), so the seeds can be larger (Culley and Klooster 2007). However in N. neesiana they are generally 

smaller,with much reduction in the size of the appenedages. Cleistogamy has disadvantages, including increased inbreeding 

depression and the aforementioneddecreased genetic variation (Culley and Klooster 2007). Dimorphic cleistogamy requires that 

each flower/seed type offers specific selective advantages. In grasses, the most important of these may be the differential 

dispersability of the seed types. Variation in the incidence of cleistogamy in the panicle of N. neesiana does not appear to have 

been studied, nor have potential environmental influences of facultative cleistogamy. 

Reproduction in N. neesiana is probably entirely sexual; apomixis, often evident by the production of twin seedlings from a 

single seed (Groves and Whalley 2002), does not seem to have been reported (e.g. Puhar 1996). 

Seed production 

In comparison to most other grasses the panicle seed of N. neesiana is large, and relatively few are produced per plant. In 

pastures at Waipawa, New Zealand, 793 ±128 culms m -2 were present in a pure ungrazed sward in pasture, and potential seed 

yield per panicle was 38, indicating a potential annual panicle seed yield of c. 30,000 m -2 (Slay 2001). In dense, ungrazed 

infestations on the Northern Tablelands of NSW Gardener et al. (1999 2003a) recorded maximum production of 1,584 panicle 

seed m -2 in 1995 (the preceding year and spring being as dry) and 22,203 panicle seed m -2 in 1996 (above average rainfall), 

determined from the number of infloresences m -2 and the number of glume pairs per inflorescence. The variation between the 

drought year and the wet year resulted from changes in the number of panicles m -2 . 

Gardener’s (1998) figure of 22,000 has subsequently been widely quoted (e.g. “20,000” in Snell et al. 2007). However the spring 

of 1995 was actually exceptionally wet (see Gardener 1998 Fig. 3.2 on p. 22) and may have resulted in a mast seeding event, 

accounting for the c. 14 fold increase in the estimated seed production over the previous year, and a c. 3.7 fold higher production 

than 1997. Masting has been defined as “synchronous highly variable seed production among years by a population” (Kelly et al. 

2008) and may have evolved as a a seed-predator satiation strategy (Kelly et al. 1992). The prime characteristic of masting 

events is that a very high proportion of the population reproduce massively in mast years and poorly in non-mast years, cued by 

weather events (Kelly et al. 2008). Masting has been described in a number of long-lived tussock grasses including Stipa 

tenacissima, and appears to result from the build up of stored reserves and rapid responses to high water availability during the 

growing season (Haase et al. 1995). Conversely, poor seed set is a feature of dry years and drought periods. Bountiful weather is 

required for mass-seeding, but predator satiation, in which a much larger proportion of the seed crop survives in mast years than 

non-mast years, accounts for the evolutionary benefits of the phenomenon, at least in New Zealand Chionchloa grasses (Kelly et 

56

al. 2008). Cues for masting may be temperature based, since temperatures can be more effective in synchronising dispersed 

populations than rainfall, and masting may involve plant anticipatory responses that associate weather events with future 

conditions that would improve recruitment (Kelly et al. 2008). 

In a dense pastures sward in New Zealand potential N. neesiana cleistogene production was 10,300 m -2 if all culms produced the 

maximum of 13 seeds (Slay 2001). Potential cleistogene numbers represented about 25% of total potential seed production. 

Under drought conditions, Slay (2002a) found more basal cleistogenes per vegetative tiller on autumn-germinated juvenile plants 

in December than on culm-bearing tillers of mature plants. In Northern NSW, an average of 7.2 cleistogenes per flowering tiller 

was produced, and a dense ungrazed infestation in 1996 produced an estimated 6,100 m -2 (Gardener et al. 1999 2003a); but note 

again that this was probably a mast year. Cleistogenes represented an estimated 21.5% and 26.1% of total seed production in 

1996 and 1997. Clipping of flowering tillers above the basal node, and above and below the top node, had no effect on the 

number of cleistogenes produced on the nodes below the clipping point (Gardener et al. (1999 2003a). 

The numbers of panicle seeds produced per tiller on the Northern Tablelands of NSW (38) and Argentina (27 – small sample 

size) are similar and the number of cleistogenes per tiller identical (Gardener et al. 1996b). But there is wide variation in plant 

density in the pampas, tussocks are much smaller, and the maximum % basal ground cover is less than half the maximum on the 

Northern Tablelands. The number of inflorescences m -2 in the Argentinian sites examined varied from 0 to 200, giving a 

maximum seed production of c. 8000 seeds m -2 (Gardener et al. 1996b). 

Awns of several seeds often twist together while still attached to the plant, forming a tangled mass that usually includes 

infloresence branches, etc. (McLaren, Stajsic and Iaconis. 2004; illustrated by Frederick 2002). Seeds that become entangled are 

retained on the plant for a longer period than seeds that do not. According to Groves and Whalley (2002 p. 158) the “ecological 

implications of [such] retention are obscure”; however potential advantages of retaining seed in the canopy include protection 

from predation and decay processes on the ground surface. 

The morphology of the seed and its presentation on the plant are evidently defenses against mammalian herbivores. Sheep and 

cattle stop eating the plant as soon as the flowering stalks are produced (Grech 2007a). 

Dispersal mechanisms 

Poaceae in general have very effective dispersal mechanisms: they comprise approximately 4% of world Angiosperm genera but 

account for 13% of cosmopolitan genera (Wheeler et al. 1990). The proportion of introduced grasses in a regional flora is usually 

much higher than for the flora as a whole, for instance in the Juan Fernández Islands of Chile 81% of the 53 grass species are 

adventive compared to 70% of the whole flora (Baeza et al. 2007). Alien grasses account for a large proportion of the grass flora 

in many regions, particularly in “livestock-based economies” (Milton 2004 p. 69). Grazing has long been a backbone of the 

economy in Victoria, for example, and 62% of the vascular plant genera in that State and 44% of the species are exotic, with 

another 10% of the genera having both native and exotic species (data from Ross and Walsh 2003). In comparison, in southern 

Africa 15% of the genera and 12% of the species are exotic (Milton 2004). In mainland Spain Poaceae account for a larger 

proportion (16 of 106 spp.) of the etablished alien plant flora than all other families except Asteraceae (20 spp.) (Gassó et al. 

2009). More species of Poaceae are considered to be environmental weeds in New Zealand than any other plant families, but in 

Australia more Asteraceae are environmental weeds (Williams and West 2000). 

Stipeae are commonly adventive species (Watson and Dallwitz 2005) and numerous species have dispersed to remote islands and 

intercontinentally. According to Tsvelev (1977 p. 7) “it can hardly be doubted” that Stipa capensis, described from and very 

common in South Africa, was carried there by the first colonists. Amelichloa brachychaeta was described in 1853 from French 

material by Godron “who at that time was unaware of its native home” (Hayward and Druce 1919 p. 226). Hayward and Druce 

(1919) recorded seven South American species (N. neesiana, N. poeppigiana (Trin. and Rupr.) Barkworth, N. pubiflora (Trin. 

and Rupr.) E. Desv., N. caespitosa Griseb., N. leptothera (Speg.) Torres, Amelichloa caudata and A. brachychaeta) in the 

adventive flora of Tweedside, Scotland, on wool refuse heaps or otherwise associated with wool factories. Nine (Connor et al. 

1993) or 12 (Edgar et al. 1991) stipoids have been recorded as naturalised in New Zealand including Nassella spp. and 

Austrostipa spp. The Argentinian and Uruguayan Nassella manicata (E. Desv.) Barkworth, established in California,was 

probably introduced in the 19th or early 20th centuries (Barkworth 1993 2006) and “apparently hitched a ride ... with South 

American vaqueros and their livestock looking for greener pastures” (Amme 2003). California also has introduced populations of 

N. tenuissima derived from horticultural plantings (Amme 2003). Five of the 25 “Stipa” species recorded in Italy are exotic 

species, all from South America: N. neesiana, N. hyalina, N. trichotoma, N. formicarum (Delile) Barkworth and A. caudata 

(Moraldo 1986).The two Nassella species (N. neesiana and N. laevissima (Phil.) Barkworth) found on the Juan Fernández 

Islands of Chile are both introduced (Baeza et al. 2007). On a world basis, at least 12 Nassella species have been reported 

growing outside their native range (Randall 2002, Barkworth 2006, Baeza et al. 2007), c. 10% of the species. 

There is general consensus that human activities are the major cause of N. neesiana seed dispersal in Australia (Bedggood and 

Moerkerk 2002, Snell et al. 2007). The panicle seed, like that of stipoids in general, has many adaptations that enable it to attach 

to a wide range of objects: according to Slay (2002c p. 23), it “attaches to almost everything”. 

However records of actual seed dispersal are very limited and the conclusion that anthropic factors account for current 

distributions is surmise based on patterns of infestations, seed biology, and general observations of the carriage of seed on 

machinery, vehicles and livestock. 

Based on evidence of exotic stipoid dispersal to New Zealand and within that country, Connor et al. (1993) suggested that 

stipoids with falcate awns may be more highly dipersible than those (such as N. neesiana) with geniculate awns. 

Creeping diaspores 

The panicle seeds of N. neesiana are classed as creeping diaspores (Davidse 1986, Connor et al. 1993) that are able to move 

along the ground and position themselves in microsites favourable for germination (Gardener and Sindel 1998, Sinclair 2002). 

Creeping diaspores of grasses generally result in little actual dispersal via ‘creeping’, the adaptations being more important in 

57

enabling microsite lodgement (Peart 1979, Davidse 1986). The awn provides the hygroscopic torsion mechanism by which the 

seed is supposed to drill itself into the soil (Murbach 1900, Davidse 1986). Peart (1979) however argued that horizontal rather 

than vertical propulsion by awn torsion is usually of greater importance, even in those species such as N. neesiana with a sharp 

callus and long, stout active awns that appear best adapted to actually drill into the soil. Alternate wetting and drying of the awn 

twists the column, driving the seed forward, the artista provides a brace for leverage, and the retrorse spines on the corona, and 

the hairs on the awn, callus and lemma-body restrict backward movement (Bourdôt and Ryde 1986). The awn of Piptochaetium 

avenaceum increases in length by 20% when wet, assisting propulsion of the seed (Murbach 1900) and this effect likely occurs 

also with N. neesiana. As with similar species (Murbach 1900), the sharp N. neesiana callus leads the seed into the ground and 

the callus hairs hold it in place once ground penetration has started and anchor the seed after germination, countering the 

opposing force provided by the radicle. The depth to which a stipoid seed is buried is related to the length of the awn: “species 

growing in areas where seeds need to be buried to ensure adequate amounts of soil moisture for germination tend to have a long 

callus, a long, narrowly cylindrical floret, and long, persistent awns” (Barkworth and Everett 1986 p. 254). 

Awns of awned grasses are able to drill seeds into cracks and crevices in the soil but there is little evidence that penetration of an 

unbroken soil crust occurs (Peart 1979, Bourdôt and Ryde 1986, Sinclair 2002) although stipoid seeds are often credited with the 

ability to ‘bury themselves’ in the soil by this mechanism (e.g. Whittet 1969 p. 129). 

Non-hyroscopic straight awns have another function immediately after seed shedding: to rotate the seed while falling in such a 

way that the seed lands vertically on the callus (Peart 1984, Sinclair 2002). This is also appears to happen with N. neesiana 

seeds, even though their awns are strongly twice-bent and hygroscopic (personal observations). 

The dispersal ability of lone panicle seeds of N. neesiana is lost by a significant proportion of seeds, which aggregate in the 

panicle when the awns twist together, forming a tangled mass that usually includes infloresence branches (Connor et al. 1993, 

McLaren, Stajsic and Iaconis. 2004; illustrated by Frederick 2002). The aggregation often ultimately falls to the ground as a unit 

(Gardener et al. 2003a) or may become attached to a vector enabling seed transport en masse (Slay 2002c). These seed clusters 

frequently hold seed in the canopy for a longer period than would otherwise be the case. The adaptive significance and 

ecological implications of such seed retention is obscure (Groves and Whalley 2002). However seed held in the panicle over a 

longer period will be accessible to a greater variety and intensity of dispersal factors, so may have adaptive advantages in areas 

where the potential range of the plant has not been reached. Similar seed aggregrates formed by twisting together of awns occur 

in a range of other grass tribes, e.g. the East African Acritochaete (Paniceae), in which the whole mass, or parts of it, may be 

dispersed by attachment of the exposed calluses to passing animals (Davidse 1986). 

Zoochory 

Seed dispersal by animals (zoochory) occurs in more than half of all plant species, most commonly by ingestion (endozoochory) 

and external attachment (exozoochory) (Stanton 2006), and also by deliberate animal carriage. N. neesiana seeds can be 

dispersed via all of these processes. 

Slay (2002a p. 15) recorded a report by C. Lee in New Zealand that unspecified birds “use awn/seed clusters for the building of 

nests”. Such deliberate dispersal is probably of little significance, but may enable the plant to cross otherwise impenetrable 

barriers. Dispersal of N. neesiana seeds by accidental attachment to birds does not seem to have been reported. Conole (1994) 

observed Red-rumped Parrots, Psephotus haematonotus, “wading up to their bellies” in drifts of seed-bearing panicles of 

Nassella trichotoma in southern Victoria, and suggested that they were highly likely to be exozoochorous dispersal agent of this 

much smaller seeded Nassella species. 

Endozoochory is certainly much rarer in birds than in mammals but in general has been very little studied (Whelan et al. 2008). 

It may occur via faeces, regurgitated pellets, or secondarily via predator consumption of the bird (Twigg et al. 2009). Conole 

(1994) also observed N. trichotoma seed consumption by Red-rumped Parrots and argued that a proportion of the seed ingested 

could survive and be dispersed. Twigg et al. (2009) found that seed fed to finches, pigeons and ducks was generally passed in the 

faeces within 0.3-5.0 hours, with longer passage times in the larger species, and that very few whole seeds survived the digestion 

process. Most of the seeds tested were not grasses, but gut passage reduced the viability of wheat seeds by 33%, and significantly 

reduced the viability of millet seeds. Finches and parrots are probably less likely to disperse seeds than pigeons because of their 

efficient digestive systems, while generalist omnivorous/herbivorous birds are probably the most likely to dispese viable seed 

endozoochorously (Twigg et al. 2009). 

Some grasses in the Paniceae, Andropogoneae and Olyreae have evolved presumed elaiosomes (lipid-cotaining diaspore 

appendages) that may attract ant seed dispersers (Davidse 1986). The structures, containing stable oils, occur in the rachilla, 

pedicel, glume base or lemma. They are difficult to identify on herbarium specimens and no field observations that confirm their 

function have been identified (Davidse 1986). In other plant families the eliasome is removed by the ants after carriage of the 

diaspore to the nest, and the seed itself may often not be damaged. There appears to be no evidence of such eliasomes in Stipeae. 

The panicle seed of N. neesiana is able to attach to a wide range of materials. Seeds attach to the coats of livestock and clothing, 

and lodge on machinery (Gardener et al. 1999, Slay 2002a). The hairs and corona of the seed, and the twisting together of awns 

of adjacent seeds that come into contact, can enhance the adhesion of seeds to objects which the callus cannot penetrate. 

Transport of N. neesiana seeds on livestock has been recorded in New Zealand (Bourdôt and Ryde 1986) and Australia 

(Gardener 1998). Grazing of sheep is the probable cause of spread in the Hawkes Bay area of New Zealand (Slay 2002c). Panicle 

seeds are carried in the fleece of sheep (Connor et al. 1993) and can remain there for at least 166 days (Gardener and Sindel 

1998). Larger grass seeds with long appendages are generally retained for longer periods in long pellage than in short, and 

retention time is probably little affected by environmental factors (Stanton 2006). Gardener et al. (2003a) found that 25% of N. 

neesiana seed naturally lodged in the wool of sheep remained after 5 months, and that shearing prior to seed production reduced 

the lodgement rate. However lodged seed often lost their awns, so would have reduced dispersal, and probably survival ability 

when shed from the fleece. A high proportion of lodged seed was subsequently shed (Gardener 1998), but details of natural seed 

shedding from fleece are scanty, so sheep may not be particularly effective dispersal agents (Connor et al. 1993). 

58

Approximately 200 alien grass species introduced with imported wool have been recorded in the British Isles (Hubbard 1968) 

and 25 species of Austrostipa, Stipa and Nassella have been recorded there as casuals regarded as mainly “wool-aliens” that have 

entered the country as seed contaminants of raw wool (Stace 1997). Some exotic stipoid populations in France may also have 

this origin (Verloove 2005). Noting European records near wool factories and tanneries, Bourdôt and Hurrell (1987a 1989a) 

suggested that that N. neesiana was likely to have reached New Zealand from South America in the wool and hides of grazing 

animals. In 19th century Europe the treatments applied in wool processing, including scouring in alkali and acid baths, dry 

heating and crushing through rollers (Vines 2006), were intended to remove all such contaminants, which were nevertheless 

dispersed in waste streams from the treatment plants, including water discharges and in wool waste or “shoddy” which was used 

as garden fertiliser in Scotland (Hayward and Druce 1919, Vines 2006). Installation of a sewage treatment plant at Galashiels, 

Scotland, soon eliminated viable seed discharge to waterways (Vines 2006). Dispersal in wool has been a major method of 

introduction of weeds to Australia, and has possibly been the most important overall dispersal method within the country, after 

their arrival (Carr 1993). Movement of wool in bales after shearing in Australia is possibly responsible for some N. neesiana 

spread. Seed was probably able to penetrate and move through hessian or jute bale bags (which probably went out of use in 

Australia in the early 1990s), but appear to be unable to penetrate the densely woven high density polyethylene or nylon fabrics 

used in modern bales bags (personal observations). 

Movement of livestock between farms is the most likely cause of intermediate range expansion in New Zealand (Connor et al. 

1993). Slay (2002c) stated that stems bearing seed can be walked across tracks by livestock. Seeds are unlikely to attach to the 

pelts of cattle but could be moved in mud on hooves (Gardener et al. 2003a). They may also adhere to other animals including 

kangaroos and rabbits (Gardener and Sindel 1998). Ens (2002a) suggested dog and rabbit dispersal as possibilities at a number of 

Sydney sites. Bruce (2001) found patches of the plant in Macropus giganteus Shaw daytime rest areas under trees, and suggested 

that kangaroos may disperse seed. Liebert (1996 p. 8) implied that dispersal by native animals was unknown, however Bedggood 

and Moerkerk (2002 p. 6) stated that “dogs, humans and wild animals such as kangaroos and rabbits spread the seed”, 

presumably on the basis of general experience of presence in areas where wild animal movements are concentrated, and other 

informed speculation. Peart (1979 p. 860) however noted the absence of any awned grasses seeds in the fur of “some 100 

carcasses of wild marsupials in Australia”. Slay (2002c p. 16) mentioned “birds” and “vermin” as probable dispersal agents in 

New Zealand and stated that stems bearing seed can be walked across tracks by livestock. No documented records of carriage on 

animals other than sheep in Australia appears to be available. 

Both panicle and stem seeds of N. neesiana can be distributed and remain viable after ingestion by livestock, but usually a high 

proportion of ingested plant seeds are digested, and the viability of the seeds that survive gut passage is significantly reduced 

(Stanton 2006). Gardener et al. (2003a) found that an average of 1.7% of N. neesiana panicle seeds and 5.3% of cleistogenes fed 

to Angus cattle (Bos taurus) were voided in dung within 4 days, mostly within 1-2 days. Less than half the voided seeds 

remained viable and no viable seed was passed after 4 days. Endozoochorous dispersal was considered to be less likely by sheep, 

which digest a higher proportion of seed (Gardener et al. 2003a), probably because they chew their food more thoroughly 

(Stanton 2006). Sheep, but not cattle, fed with a range of pasture seed digest a greater proportion of long seeds than short 

(Stanton 2006). Rabbits also void a wide range of seed in their dung (Bloomfield and McPhee 2006) but would be unlikely to eat 

N. neesiana seed. 

Even at high stocking rate livestock avoid eating N. neesiana once the reproductive stage is reached (Grech 2007), and the 

extremely sharp callus and rough texture of the panicle seed assist in making the panicle unpalatable. These seeds evidently are 

adapted to avoid being eaten. 

Cleistogene dispersal 

Gardener et al. (2003a p. 614) thought that cleistogenes “have no obvious dispersal mechanism” but that ingestion by grazing 

animals was one possibility. Barkworth and Everett (1986) suggested that stipoid seeds adapted to zoochory by animal ingestion 

may have short, deciduous awns, globose florets and obtuse calluses, a set of characteristics possessed increasingly by N. 

neesiana stem cleistogenes from upper to basal. 

Cleistogenes can develop even if the flowering tiller is damaged, and are important in maintaining the species during climatic 

extremes (Gardener and Sindel 1998) and under conditions of heavy grazing or fire (Dyksterhuis 1945). They are better 

protected from some predators than panicle seeds (Gardener and Sindel 1998) being tightly covered by leaf sheaths during their 

formation and after maturity, and remain available for dispersal from the parent for as long as 6 months on standing dry culms 

(Gardener et al. 1999) e.g. in hay, and for much longer if attached on basal nodes. According to Connor et al. (1993), stem 

cleistogenes are released when the leaf sheaths weaken or rupture, so any dispersal of released cleistogenes requires tiller 

breakdown. According to Slay (2001 p. 38) the tiller dies after panicle seed is produced and its roots decay over a period of up to 

two years, “eventually releasing the cleistogenes and or producing the ideal germination conditions for their establishment, as 

evidenced by the seedlings that grow out of decayed clumps”. He found old, rotted root/stem areas “up to three tiers deep, 

suggesting plants die and re-establish ... on top of each other” (p. 42), and that at least the basal cleistogenes are adapted to not 

disperse. 

In terms of large grazing mammals, the tightly attached covering leaf sheath and stem node segment can be conceived of as an 

attractive ‘fruit’ containing the cleistogene ‘seed’, and when ingested this ‘fruit’ may enable better survival of the seed and faster 

passage through the gut (Davidse 1986). Lllamas and alpacas reportedly eat the straw and possibly disperse cleistogenes in their 

guts (Colin Hocking, 26 October 2006). 

Cleistogenes are dispersed by cultivation machinery (Connor et al. 1993) and in decayed tussocks or sods (Slay 2002a). Old and 

broken culms bearing cleistogenes can be carried by livestock (Slay 2001). Bourdôt (1989) noted that basal cleistogenes in 

particular are likely to survive fire in situ, and have probably evolved not to disperse, but rather to replace the parent plant should 

it die. 

Grasses that are ‘herbivore exploiters’ are palatable, recover well after grazing, and have seeds adapted for dispersal by grazing 

mammals (Milton 2004). N. neesiana appears to have a mixed dispersal strategy, probably being an exploiter of large grazers via 

59

endozoochory in terms of non-basal stem cleistogenes, a repeller of grazers at the time of panicle seed production, and a ‘sit and 

wait’ strategist in terms of basal cleistogenes. 

Wind 

Carr (1993) listed wind as a dispersal agent, but the panicle seeds have no particular adaptations for wind dispersal. In windless 

conditions the seed falls vertically (Slay 2002a). Of the 39% of panicle seed recovered in a wind dispersal experiment, Gardener 

et al. (2003a) found none more than 2.8 m from the centre of the source plant, and the majority of seed within 1 m. However 

these findings may create a misleading impression about the frequency of wind dispersal and the distance that wind may carry 

the seed. Small scale, short duration, high intensity atmospheric turbulence events have a very strong impact on aerial transport, 

and strong vertical winds associated with thunderstorms can lift seeds that lack special wind dispersal adaptations (Nathan et al. 

2005). Seeds could certainly be blown along flat surfaces such as roads, possibly assisted by vehicle eddy or suction currents 

(Barwick 1999). Willy willies (small whirlwinds) that carry large amounts of loose plant debris occur frequently in summer in 

Australia, particularly in inland areas. Surprisingly large grass seeds can be lifted to high altitudes by natural processes, e.g. the 

spikelets of Paspalum spp. have been obtained by aircraft sampling at altitudes of up to c. 1500 m in Louisiana, USA (Hitchcock 

and Chase 1971). 

Slay (2001 2002c) noted that panicles and stems can be blown short distances by strong winds, resulting in dispersal of 

cleistogenes. Wind dispersal of late maturing seed on ‘secondary panicles’ could presumably occur more readily since the 

remainder of the panicle, with only glumes attached, could be more readily lifted and carried. 

Water 

Extensive distribution along floodways and watercourses has led to the inference that movement of N. neesiana seeds in flowing 

water is important (Frederick 2002). According to Bourdôt and Ryde (1986) seeds are “carried along water courses, giving rise 

to isolated patches of the plant”. Slay (2002b) stated that running water disperses seed. Hayward (Hayward and Druce 1919) 

found plants along the river bank downstream of wool processing factories in Scotland. Cook (1999 p. 91) stated that N. 

neesiana is a weed of “flood zones” and Iaconis (2003) stated that flood waters are responsible for dispersal. In urban and periurban 

Canberra, seeds have possibly been dispersed widely in the drainage system (Jenny Connolly and Sarah Sharp pers. 

comms. 2006). Bedggood and Moerkerk (2002 p. 6) state that “run-off water can carry seed from one property to another”. There 

is very little detail to support these claims, which mostly appear to be based on the pattern of distribution of infestations in small 

catchment areas. No published information appears to be available on the buoyancy of the seeds, their presence in flood debris, 

etc. Preliminary observations indicate that they remain afloat in still water for at least 4 days. The awns quickly become entirely 

straight while acquiring extreme flexibility, and the seed becomes ‘sperm-like’, a characteristic that would enhance more rapid 

and effective carriage in moving water. 

Human activities 

The consensus view among most commentators is that human activities, little mediated by domestic animals or physical 

environmental factors are the major cause of dispersal in Australia (Snell et al. 2007), but the vector strength and tempo are 

unknown. Roadside management activities including slashing, mowing and grading, particularly when the plants are seeding 

(Frederick 2002), are generally the inferred causes. Roadsides carry some of the densest Australian infestations (Snell et al. 

2007) and infestations in a new area frequently occur first along roads. Transport of seeds in hay has been recorded in New 

Zealand (Bourdôt and Ryde 1986) and dispersal in contaminated fodder has been called “a primary mechanism” of dispersal 

(Frederick 2002 citing Liebert 1996), although there appear to be no specific incidents of such dispersal on the public record in 

Australia. 

Seeds are said to “adhere” to machinery “via [the] ... callus” (Gardener and Sindel 1998 p. 77) but there are few points on 

vehicles and machines which the sharp end of the callus could penetrate, so such adhesion presumably involves the callus hairs 

and a range of leverage options with the callus tip, lemma body and awn. The actual mode of lodgement needs to be far better 

described. For instance, penetration of rubber vehicle tyres has not been reported – piercing by the callus does not seem to be a 

factor. Various characteristics of the whole seed alone and in aggregates are responsible for attachment. 

Linear distribution of infestations along roadsides and vehicle paths is widespread and commonplace (Bruce 2001, Frederick 

2002, ?Sharp 2002) but this could have a variety of causes unrelated to direct movement attached to machinery or vehicles. In 

particular roadsides are subject to a variety of disturbances such as mowing, soil compaction and pollution that can reduce the 

competitive abilities of native vegetation (von der Lippe and Kowarik 2007a). Studies of the seed rain from motor vehicles in 

long road tunnels along a German motorway have demonstrated that roadsides are “preferential migration corridors” for invasive 

plants and that long distance dispersal (> c. 200 m) is routine (Von der Lippe and Kowarik 2007a). Grass species accounted for 

four of the 20 most frequent species in the seed rain (including the two cereal crop species Triticum aestivum, L. and Secale 

cereale L., plus Poa annua L., and Lolium perenne L.), but at least one long-awned grass Bromus tectorum L., highly invasive in 

the USA, was represented. Half of the 204 species detected were not local natives. Seeds of 42% of the regional roadside flora 

were found in samples, 69% of the species sampled in the tunnels were found growing within 100 m of the tunnel entrances and 

and 98.5% of all seeds sampled were from species in the regional roadside flora. Approximately one third of the species found 

were not present in areas near the tunnel entrances. Non-native species were more often subject to long distance dispersal than 

natives. Seeds represented

Dispersal of seed in contaminated soil has been recognised as a dispersal mechanism (Muyt 2001, Snell et al. 2007). Spread of 

seed along roadsides by graders and other earthmoving equipment was mentioned by Bourdôt and Ryde (1986), but no further 

information was provided. 

Liebert (1996) observed that slashing of the plant while seeding was a “primary mechanism” for dispersal. Trengrove (1997) also 

observed that dispersal along roadsides is caused by slashers operating at the time of seed set. Bruce (2001) found that N. 

neesiana was generally more abundant at sites in the ACT that were mown to some extent, that it was never absent from areas 

that were entirely mown, and at infested sites where mowing occurred it was generally spreading from mown into unmown 

areas, in some cases very obviously along mown walking tracks into native grasslands. In fact, the overall distribution of N. 

neesiana in the ACT correlated extremely well with mown areas (Bedggood and Moerkerk 2002). Ens (2002a) suggested seed 

carriage on mowing equipment was the most likely explanation for a number of the infestations she examined in the Sydney 

area. Sharp (2002) noted continued expansion of infestations along roadsides in the ACT “despite efforts to control spread with 

more appropriate management practices”. 

Moerkerk (2005a 2005b 2006a 2006b) analysed the plant propagules found in material manually cleaned from vehicles and 

machinery used in natural resource mangement activities, particularly weed management, in Victoria. The flora of each vehicle 

reflected the flora of the region in which was used. Nearly three times as many species of Poaceae were found in ‘clean-downs’ 

than the next most abundant family, and grass species accounted for 6 of the 10 most frequently detected contaminants. N. 

neesiana was detected on 5 of 106 State government, local government and private contractor vehicles and machines sampled: 4 

passenger vehicles and 1 slasher. A four wheel drive utility vehicle used by the Shire of Hume (an area where N. neesiana is 

common) cleaned in June 2005, yielded 24 weed species including N. neesiana, from 23% of the 810 g of material removed. A 

Hume Shire tractor and slasher cleaned in June 2005 yielded 26 spp., including N. neesiana, from 27% of 178 g of material. N. 

neesiana was found in the cabin, engine bay and tray of vehicles and in the wheel guards and chassis of two vehicles. Noxious 

weeds were mostly frequently found in the cabins and engine bays of passenger vehicles. 39% of passenger vehicles and 29% of 

machinery items were found to be carrying noxious species. Of the vehicle types examined, 4WD utility vehicles had the highest 

rates of contamination, while tractors with attached slashers, and graders had the highest rates of contamination (40%) of the 

machinery types examined. 

Grech (2005b) ‘cleaned-down’ a four-wheel-drive utility vehicle used on a property infested with N. neesiana using brushes, 

manual plucking and high pressure water. The latter method was found to be ineffective and failed to dislodge entangled masses 

of N. neesiana seeds. Weighing of the removed material indicated in excess of 10,000 N. neesiana seeds, “equivalent in volume 

to a medium sized couch cushion”. 

These studies identified important potential vectors but failed to determine under what conditions seed become attached and are 

deposited in potentially suitable sites, and the actual spatial effectiveness of vehicles as seed vectors. Transport vehicles have 

much higher potential for long distance dispersal and are perhaps mainly to blame for inter-regional dispersal,while machinery is 

likely of greater importance at a local scale. 

Slashing machinery actively disperses seed around the slasher. Detailed investigations have been undertaken of one particular 

model of slasher in relation to N. neesiana seed dispersal (Erakovic et al. 2003, Erakovic 2005). Rotating slasher blades caused 

upward air movement within the slasher frame, and high outlet velocities occured at the front of the slasher deck where the 

cutting process actually occured. At the commonly used slash height of 7 cm with this machine, most of the slashed material was 

deposited within the slashed strip. Over 98% of the expelled material (not deposited in the slashed strip) fell within 1 m and the 

remainder within 2 m, with more deposited on the left of the slashed strip than on the right (Erakovic 2005). Sigificant amounts 

of slashed chaff and seeds were deposited on the top of the slasher unit, and significant amounts lodged in the debris ejection 

protector chains (Erakovic et al. 2003). Pronounced accumulation of seeds on the top front of the deck resulted from direct 

dislodgement of seed from the plant that never came into contact with the slasher blades (Erakovic 2005). 

More importantly, slashing was also found to disperse seed over longer distances if the slasher was not kept clean. Seed adhered 

inside the deck of the slasher in crevices and boltholes, and slashed material accumulated around front and rear internal walls, 

the blade shaft and in corners (Erakovic 2005). Decontamination and cleaning was time-consuming and tedious, and it was very 

difficult and unsafe to clean the underside of slasher decks in the field (Erakovic 2005). The use of counter-rotating twin blades 

and other improvements in a new slasher design, plus the development of slasher accessories (covers and shields, flaps to replace 

chains) showed great promise of reducing these problems (Erakovic 2005). 

A range of other dispersal processes mainly related to trade and commerce have been implicated in N. neesiana spread. Seed has 

been spread in hay bales in New Zealand (Slay 2002c) and movement of contaminated fodder has been identified as likely (Muyt 

2001). Some infestations in New Zealand have originated from contaminated pasture seed (Connor et al. 1993, Slay 2002a), 

sown as recently as 1980 (Slay 2002c). Spanish infestations may have originated in cereals imported from Argentina (Verloove 

2005). Introduction in railway traffic may be the origin of a large population at Bédarieux-Nissergues in France (Verloove 2005). 

Cultivation can carry whole plants and seed within paddocks (Slay 2002c). Slay (2002a p. 11) noted that a small infestation at 

Waipawa, New Zealand, was thought to have been “spread by lawn mower/lawn clippings”. 

Like N. neesiana, other Stipeae spp. have been carried internationally in solid ship ballast, e.g. Amelichloa brachychaeta from 

Argentina and N. chilensis (Trin. and Rupr.) E. Desv. from Chile to Portland, Oregon, USA (Hitchcock and Chase 1971), the 

latter species “once collected” and not established (Barkworth 2006). Solid ballast such as beach sand and rocks began to be 

replaced by water ballast in the late 1870s and all large ships now use it (Jones 1991), so this invasion pathway is now probably 

largely obsolete. 

Pollen 

Pollen (the male gametophyte) is also dispersed. Grasses in general are wind pollinated and produce large amounts of pollen, and 

the pollen concentration downwind decreases at a rate inversely proportional to the square of the distance from the source 

(Connor 1986). Pollen may have an important role in gene flow and in promoting plant invasions (Petit 2004). For plants as a 

61

whole, gene flow via pollen is estimated to be on average an order of magnitude greater than gene flow via seeds, and most long 

distance gene flow is via pollen (Petit 2004). 

Anthesis in grass species occurs at particular times of day, often in the morning or the afternoon (Connor 1986). Release of 

pollen occurs at times of high temperatures and low humidity.There appear to be few records of stipoid anthesis times. 

Piptatherum holciforme flowers at night and P. virescens early in the morning (Connor 1986). Ramasamy (2008) observed most 

Nassella trichotoma anthesis in the morning (7-11 am) with some later in the day. The structure of grass inflorescences and 

flowers influence their pollen dispersal and trapping characteristics (Connor 1986). Wind is the main dispersal agent for grass 

pollen, but insects may play some small role (Connor 1986). 

Rates of spread 

Few published records exist of the rate of change in the dimensions of N. neesiana infestations and the rates at which the plant 

spreads (Table 5). In New Zealand the maximum rate of dispersal on a linear front from known sources was 8 km over 59 years 

at Marlborough and 3.5 km over 30 years at Waipawa (Connor et al. 1993). Comparable Australian data does not appear to be 

available. 

N. neesiana was rated by Platt et al. (2005) as having a rapid, rather than moderate or slow rate of dispersal. The ACT Weeds 

Working Group (2002 p. 4) stated that the “rate of spread and establishment is unknown, but believed to be rapid”. However 

perceptions of rapid spread in Australia may be partly false, due to recognition failures (Walsh 1998). 

Table 5. Measured and inferred rates of spread of N. neesiana. 

Locality 

Distance Period Rate Notes 

Reference 

(m) (y) (m y -1 ) 

Marlborouegh, NZ 8000 59 136 District infestation expansion Connor et al. 1993 

Waipawa, NZ 3500 30 117 District infestation expansion Connor et al. 1993 

New Zealand 120-140 With no active management Slay 2002c 

Hawke’s Bay NZ 3-10 5 0.6-2 Patch expansion Slay 2002c 

Areas (ha) Period 

(y) 

Rate 

(ha y -1 ) 

Marlborough, NZ 1555-3000 14-15 101-103 District expansion Slay 2002a, 2002c 

(3071) 

hypothetical 1 5 100 Expansion at 100 m per year Slay 2002a 

hypothetical 1 10 350 Expansion at 100 m per year Slay 2002a 

Rare long-distance dispersal events (e.g. by water or human transport) are thought to contribute to accelerating rates of spread 

that have been recorded as plant invasions proceed (Mack and Lonsdale 2001). This is because the likelihood of successful 

dispersal increases in proportion to the size of the propagule pool. This factor may also be contributing to the Australian 

perception. 

Invasion patterns 

Trengrove (1997) observed that N. neesiana dispersed along roadsides by slashers then invades into adjoining paddocks “in a 

front”. The pattern in the ACT is of movement outwards from a central Canberra source population along urban and periurban 

roadsides mainly via mowing and slashing, with spread outward from the linear corridors, often by the same means (ACT Weeds 

Working Group 2002). Slay (2002c) listed a range of situations where infestations occur in New Zealand that are indicative of 

seed dispersal patterns: “paddocks sown with uncertified seed between 1950 and 1980 ... holding paddocks close to the road ... 

the edges of farm tracks ... 1-3 m away from power poles, along fence lines or other places where stock ‘rub’ ... river banks ... 

around hay barns ... sheep yards”. 

Soil seed bank 

The soil seed bank may be thought of as the consequence of four different processes: dropping of individual panicle seeds, the 

shedding of inter-twined seed masses, the release of stem cleistogenes when the culm decays and the release of basal 

cleistogenes when stem bases decay. 

Cleistogenes enter the seed bank as the culms decompose or after fire (Groves and Whalley 2002). Culms deteriorate slowly 

through summer and autumn and the leaf sheaths rupture in autumn or winter, releasing stem cleistogenes (Slay 2002c). Basal 

cleistogenes are released after the tiller or parent plant dies and decomposes, possibly 12-18 months after mortality (Slay 2002c). 

Testing of panicle seeds of N. neesiana with triphenyl tetrazolium chloride (tetrazolium) reported by Puhar and Hocking (1996) 

indicated 80-95% viability. The seeds in general are reportedly viable for more than 12 years (Benson and McDougall 2005), 

“many years” (Quattrocchi 2006) or “in excess of three years” in the soil (Snell et al. 2007p. 4). Bourdôt and Ryde (1986) stated 

that both panicle and stem seeds have >90% viability and survive for “several years” in the soil. The soil seed bank was large 

and persistent in heavily infested sites investigated by Gardener on the Northern Tablelands of NSW in the 1990s (Gardener et 

al. 2003b). 

Assessments of the seed bank in seven populations in the Argentinian pampas showed it be close to zero (Gardener et al. 1996b, 

Gardener et al. 1997). Possible reasons for this include high levels of seed predation by ants, attack by a seed pathogen after seed 

shed, or rapid microbial decomposition in the soil, however the closely related Nassella clarazii (Ball) Barkworth was found to 

have aseed bank of c. 1200 m -2 (Gardener et al. 1996b). 

62

In New Zealand, Hurrell et al. (1994) measured a soil seed bank (to about 10 cm depth) under old established stands of 2,600- 

35,000 seeds m -2 , with an average density of 10,500 seeds m -2 , of which approximately 70% were cleistogenes, with few or no 

aerial seeds in some samples. Tetrazolium testing showed 58% viability of cleistogenes and 86% of aerial seeds, with a total 

average viability of 67%. In pastures at Waipawa, New Zealand, the seed bank was composed almost entirely of cleistogenes 

(Slay 2002c), with a mean of 6437 ±3437 viable cleistogenes and 660 ±1289 viable panicle seeds m -2 (Slay 2001). Bourdôt and 

Hurrell (1992) found 4000-18,000 viable seeds m -2 under pasture, 99% in the top 25 mm and 0% below 100 mm, on a silt loam 

at Marlborough. Viability (as determined by tetrazolium treatment) of seed buried in polyproylene mesh bags for 6 years in this 

soil increased with increasing depth of burial, 24% of seed remaining viable at 25 cm depth, 17% at 5 cm, 5% at 2.5 cm and 

0.1% for soil at the surface (Bourdôt and Hurrell 1992). Analysis of decay data suggested that an increaing proportion of buried 

seed survived at increasing depth indefinitely in a state of dormancy, whereas no surface seed was viable after 1 year (Bourdôt 

and Hurrell 1992). However there are doubts about the reliability of tetrazolium tests for determination of seed viability (Puhar 

and Hocking 1996). Seeds that are hard when pressed between the thumb nail and forefinger are “almost certainly viable” 

(Bourdôt 1988). 

In pastures in New Zealand in the absence of seed input, annual depletion rates of the soil seed bank were 38% when “regularly” 

mown to prevent panicle seeding, 61% with repeated glyphosate applications, and between 66% and 77% for single to repeated 

annual cultivations (Bourdôt and Hurrell 1992). 

Smaller seed banks have been described in New England Tablelands pastures by Gardener (1998) who made direct counts of 

loose seeds in the soil to 4 cm depth, with all seeds considered to be of panicle origin (Gardener et al. 2003b p. 622). Under 

dense infestations 681-11,307 viable seeds m -2 were found. If basal cleistogenes, contained in tiller bases and not loose in the 

soil, are included, the estimated total seed numbers were increased by 35.5% (Gardener and Sindel 1998) or 20% (Gardener et 

al. 2003b). Within N. neesiana tussocks 44.1% of seeds were panicle seeds and 55.9% basal cleistogenes (Gardener et al. 

2003b). 

In native grasslands and grassy woodlands in the Melbourne area the viable seed bank appears to be substantially smaller. Seed 

banks at Iramoo Wildlife Reserve determined by Bram Mason (unpublished) were up to 7000 m -2 (Robinson 2003 2005). 

Beames et al. (2005) assessed panicle seed banks in areas of high quality native grassland at Laverton and Grey Box woodland at 

Melbourne Airport that had been subject to a range of different management regimes to control N. neesiana. They separated the 

seeds into four categories: filled, unfilled (caryopsis absent), successful germinants (germinating at the time of sampling) and 

unsuccessful germinants. At Laverton a maximum seed bank (seed of all categories) of ca. 2200 m -2 was found, however unfilled 

and unsucessfully germinating seed accounted for ca. 90% of seed. In areas subject to N. neesiana management, seed bank 

numbers did not exceed 1000 m -2 , with similar proportions in each seed category, except for filled seed, which comprised a 

much smaller proportion in two of the managed areas. At Melbourne Airport the maximum seed bank exceeded 7000 m -2. The 

unburnt, unsprayed treatement had a significantly smaller seed bank than most of the managed areas. Significantly more filled 

seed was found outside managed areas, which in most cases had zero filled seed. One major conclusion was that most of these 

differences were the result of reduced seed input due to ongoing herbicide treatment, which possibly reduces the proportion of 

viable seed produced and seed persistence. The study indicates that the seed bank in non-agricultural areas is much less 

persistent and easier to deplete than had previously been assumed. 

Hocking (2005b) reported even lower levels in infestations subject to herbicidal control in southern Victora: 80% unviable seed. He suggested that the size of the seed bank may be widely variable on a 

regional basis and between sites managed for agriculture and conservation, and that agents in the soil at some sites may be 

destroying a high proportion of seed. Hocking (in Iaconis 2006b) reported no differences in the soil seed banks of agricultural 

and natural areas, but high variability between sites, and >50% of the seeds unfilled. 

In New England pastures, dramatically better fruiting in wetter years (Gardener et al. 1996a) led to major addition to the seed 

bank (e.g. 41.6% of seeds incorporated in 1996), but in drier years input was only sufficient to maintain existing seed bank 

numbers or inadquate to maintain pre-existing levels (Gardener et al. 2003b). 

Rates of decline of the seed bank determined for New England pastures without input after 3 y, were 4676 to 1323 seeds m -2 in 

bare plots and 4585 to 1507 seeds m -2 in vegetated plots, with no significant effect of ground cover on decline, and a predicted 

decline without input (based on a fitted exponential decay curve) to 10 seeds m -2 after 12.4 y (Gardener et al. 1999, Gardener et 

al. 2003b). The seed bank longevity was found to be >6 y (Gardener and Sindel 1998) and half-life 1.3 y (Gardener et al. 

2003b). There is anecdotal evidence of gemination from continuously bared ground after 6 y (Gardener and Sindel 1998). 

Where soil cracking occurs in summer, as in the clays of the Victorian basalt plains, seed dropped in late spring and early 

summer will certainly move into the soil to greater depths than was sampled in the two Australian studies, which both assumed a 

similar depth profile of the seed bank to that found on a silt loam in New Zealand by Bourdôt and Hurrell (1992). This may be 

especially significant since viability of seed increases with depth of burial (Bourdôt and Hurrell 1992). Grasses that produce 

relatively few large seeds “often emerge from seed located deeper in the soil …where … water is available for a longer time … 

and many have high seedling growth rates” (Groves and Whalley 2002). 

Germination and seedling recruitment 

Seed dormancy 

The panicle seed have dormancy - the tight lemma may provide a barrier to water and gas exchange and mechanically constrain 

the embryo (Gardener and Sindel 1998) - and appear to have an after-ripening requirement after falling from the plant of 

between 3 and 12 months for maximum germination (Gardener et al. 1999, Gardener et al. 2003b). The overlapping margins of 

the lemma and its toughness make it difficult to break open to expose the caryopsis, a feature also of Austrostipa spp. (Barkworth 

2006). Puhar (1996) found by staining with tetrazolium chloride that 93.5% of N. neesiana seeds collected in the previous 

summer were viable, but in laboratory tests less than 2% germinated under a day/night cycle of 30/20ºC and 14/10 hr light. 

Removal of the lemma enabled 100% germination within three days under the same conditions. 

63

Dormancy mechanisms allow the seeds to persist in the soil for “many years” (Gardener et al. 1999 p. 10). Dormancy can be 

broken by scarification and de-hulling (removal of the lemma) (Puhar and Hocking 1996) but is not broken by stratification 

(chilling) at 3ºC for periods of 24, 72 and 168 hours (Puhar 1996). Heating of seeds to 60 ºC in an oven for 30, 60 or 240 minutes 

signficantly increased germination (Puhar 1996). Gardener et al. (2003b) found that de-hulling of seeds stored for 8 months 

increased germination from 48.5% to 82% at 15-25ºC with a 12 h light/dark cycle. Andersen (1963) found that germination of 

naked caryopses of the closely related N. leucotricha was usually complete after 21 days in light at 15-25 ºC and in darkness at 

15 ºC, and that seed with intact glumes germinated best when sown upright in soil and sand for 28-35 days at 15-25 ºC. Chilling 

had little effect on germination. 

An intact lemma may prevent germination by restricting the embryo or by acting as a barrier to water and oxygen. The lemma 

protects the embryo from dessication and other harsh environmental conditions. Germination occurs only when the lemma is 

broken or degraded by weathering (Puhar 1996, Slay 2002c) or the actions of decomposers. This may require 3-4 months in the 

soil, so germination of panicle seed is delayed until mid-winter (Slay 2002c). Germination rate increases with duration of ground 

burial: 90% of de-awned seeds buried in the ground for 2 y in terylene cloth bags germinated in the laboratory compared with 

48% of laboratory-stored seeds, but seed viability did not vary significantly with burial depth (0-10 mm vs. 10-20 mm) 

(Gardener et al. 2003b). In New England Tablelands pastures, only a small proportion of the seed bank germinates over 2 y 

(Gardener et al. 2003b). 

Dyksterhuis (1945) found that cleistogenes of N. leucotricha commonly failed to germinate within a year of their production and 

were usually not wetted by rains because of the tight wrapping of the leaf sheath. Disintegration of the sheath was required for 

germination. Few cleistogenes appeared to germinate on living plants. Basal cleistogenes reporrtedly gave rise to seedlings 

especially in old, closely grazed, dead tussocks. Here they were protected from frost heave, which caused major mortality of 

seedlings of panicle seed origin that germinated in areas cleared of litter and outside of tussocks. Seedling numbers were much 

higher in bare areas than areas with accumulated litter. 

Germination 

The panicle seeds of N. neesiana, like most of the awned grass species tested by Peart (1984), are probably adapted to germinate 

after lodgement in suitable sites at the surface with their lemmas only partially buried, and, contrary to the opinions above, to 

have no dormancy but to react whenever moisture conditions and temperatures are suitable. In vitro, fresh seed germinates after 

about 10 days at constant temperatures between 18 and 25ºC under a 16 h photoperiod, but not at constant temperatures of 18, 20 

or 25ºC in continuous darkness, although germination is stimulated in darkness by a temperature fluctuation of 10-20ºC, 

“suggesting a mechanism for pasture gap detection” via both light and heat (Bourdôt and Hurrell 1992 p. 101). 

The timing of seed germination is regulated by rainfall (Bourdôt and Hurrell 1992). Germination occurs mainly in spring and 

autumn but can happen at other times of the year if adequate soil moisture is available and temperatures are suitable (Bourdôt 

and Ryde 1986, Duncan 1993, Gardener et al. 1999, Britt 2001, Slay 2001). In New Zealand pastures, there are two distinct 

peaks in autumn or early winter, and in spring or late spring, and high winter rainfall may delay spring germination. Germination 

is probably limited by low winter temperatures, and almost certainly by summer drought (Bourdôt and Hurrell 1992). Similar 

patterns are apparent in Australia. Germination occurs predominantly in autumn and spring on the New England tablelands 

(Gardener et al. 2003b). 

According to Muyt (2005 p. 4), germination is “likely ... in response to the death of adult plants”. Seeds reportedly germinate 

only in bare areas (Gardener et al. 1996a) or when gaps are created in pastures (Gardener et al. 1999). 

Establishment of seedlings and juvenile plants 

Many characters of seedling grasses may be useful taxonomically (Sendulsky et al. 1986), but those of N. neesiana do not appear 

to have been systematically described, nor has the development of seedlings and juvenile plants been adequately documnted. 

Internodes of grass seedlings are initially meristematic but new roots form mainly at the nodes, and the production of more, 

larger, adventitious roots from the increased stem surface is enabled as the stem increases in diameter (Clark and Fisher 1986). 

Later the thickening process of the stem changes to an elongation stage, with nodes forming on the stem at the leaf insertion 

points where there is no elongation. Unlike most other plants, the most proximal nodes and the upper sections of the internodes 

mature first, leaving an intercalary meristem, capable of cell division and growth, at the base of the internode, surrounded by the 

base of the leaf sheath (Clark and Fisher 1986). The seedling gradually transforms into a juvenile plant that produces new leafy 

shoots (tillers) from basal buds, each with their own roots. In tillering grasses the initiation of the inflorescene in the culm 

generally corresponds with cessation of new vegetative tiller production, which may resume after flowering (Clark and Fisher 

1986). 

Peart (1979 1984) established that grasses with hygroscopic awns and a barbed callus like N. neesiana, have a distinct seedling 

recruitment strategy, involving lodgement in an upright position on the soil surface with c. half the lemma exposed and no 

dormancy. Seeds that fail to lodge by the callus produce seedlings in which the radicle is less likely to successfully penetrate the 

soil. These seeds lie on the surface and were found to be destroyed by fire. 

Recruitment of N. neesiana seedlings is often high. Slay (2001) recorded 1108 seedlings m -2 in winter in densely infested pasture 

in New Zealand. Earlier germinating cohorts have better survival rates than later cohorts (Gardener et al. 2003b). On bare 

ground, 78% of seedlings survived over 20 months (Gardener et al. 2003b). Seedling emergence from a natural seed bank on 

bared ground varied from 5-136 m -2 in any 6 month period (Gardener et al. 2003b). Seedlings emerging from de-awned panicle 

seeds deliberately placed, callus-downwards, 2 cm apart, at shallow depth (0-10 mm) had lower survival rates than seedlings 

from seeds buried at 10-20 mm depth (Gardener et al. 2003b). 

Where there is a seed bank, areas bared with herbicide “generally produce a large germination ... within 12 months” (Duncan 

1993) and are ‘quickly reinvaded’ (Bourdôt and Ryde 1986). Cover and abundance data from surveys at Derrimut Grassland 

Reserve, Victoria, suggested that seedling establishment is uncommon in areas of dense T. triandra (Lunt and Morgan 2000). In 

experimentally bared ground (glyphosate application) emergence ceased after the regrowth of surrounding vegetation (Gardener 

et al. 2003b), suggesting that disturbance that creates bare ground and sunlight are germination triggers (Gardener et al. 1996a). 

64

No emergence was observed in undisturbed vegetated areas (Gardener et al. 1999), however on vegetated ground it is difficult to 

determine if recruitment is from new tillers of existing plants or from seed (Gardener et al. 2003b). According to Bourdôt and 

Hurrell (1987a), after herbicide treatment seedlings “may establish within the tiller bases of the dead tussock”. Pritchard (2002) 

recorded widespread seedling presence after herbicide treatment at Laverton North, Victoria, in April 2001, “often growing from 

within dead tussocks”. 

Approximately 50% of seedlings were derived from cleistogenes in a Marlborough, New Zealand pasture (Bourdôt and Hurrell 

1992). When plants were treated with herbicide giving total kill and preventing seed head production, seedling establishment in 

the following autumn and winter was from cleistogenes (Slay 2001). Slay (2002c) reported that cleistogenes in the soil seed bank 

germinated on bare soils (resulting from herbicidal control during spring) in autumn, before panicle seed. Earlier germination of 

these seeds was considered likely to be due to the much reduced toughness of the cleistogene seed coat. 

The soil surface conditions needed for germination do not appear to be adequately known. The necessity of bare ground is 

generally recognised as a requirement (Gardener et al. 1996a, Gardener 1998, Slay 2001) and this has been attributed to a light 

requirement (Slay 2002a). Awns are shed when the seed has penetrated 10-30 mm into the soil and remain on the soil surface 

(Slay 2002a). 

In summary, Bedggood and Moerkerk (2002) wrote that seedlings establish best in the open but can also establish in shaded 

situations under the canopy of other plants. However the evidence for establishment in shade is weak and uncertain. 

The seedling “does not appear ... very vigorous” (Duncan 1993) and “grows quite slowly” (Storrie and Lowien 2003). 

Demography, growth, persistence and dominance 

N. neesiana tussocks are long-lived and “very hardy” (Storrie and Lowien 2003). 73% of plants survived over 3 years (Gardener 

et al. 1999) and individual plants have a longevity of over 20 years (Benson and McDougall 2005). 

The physical structural characteristics of infestations, their spatial arrangement, patchiness and dynamics have been poorly 

described. Bourdôt and Hurrell (1987a) reported that “pure stands” occupying areas from several hectares to several square 

metres were common in pastures near Lake Grassmere in New Zealand. When fruiting, such infestations can look like cereal 

crops (Slay 2002c). Slay (2001 p. 11) reported that herbicide treatment that reduces competition from other pasture species can 

eventually result in “a dense mat, and total cover”. Kirkpatrick et al. (1995 p. 35) stated that N. neesiana in native grasslands 

“generally grows to the exclusion of all other species”. 

Bourdôt (1988 p. 1) described its areal pattern in New Zealand as “scattered ... clumps and small patches”. Slay (2002c p. 11) 

recorded that plants “are generally found in circular patches” in pasture in New Zealand and that infestations may be 3-10 m in 

diameter 5 years after establishment, with individual tussocks 5-12 cm in diameter. Slay (2002c fig. 28) illustrated a pasture with 

an array of scattered, irregularly rounded patches in New Zealand. Pritchard (2002) recorded a mean tussock density of 20.5 m -2 

(“relatively dense”) at Laverton North Grassland, Victoria, in October 2000, in a stand selected because it was “almost pure”; the 

tussocks mainly being “large, mature” and “up to 30 cm high”. 

Hurrell and Bourdôt (1988 p. 237) stated that N. neesiana “often does not have a dense and well-defined tussock form when in 

association with other grasses”, while the ACT Weeds Working Group (2002) noted that it forms “dense thickets”. Gardener 

(1998 p. 4) found that it “can completely overrun pastures resulting in canopy ocover of up to 60%”. Slay (2002c) noted that it 

can from “continuous pasture” (in contrast to discrete tussocks). Stewart (1996) recorded cover in two 5 x 5 m quadrats in 

Broadmeadows Valley Park, Victoria, which ranged from 50-70% in one quadrat and 5-50% in the other over a 6 month period. 

Grech et al. (2005) reported that its canopy cover can exceed 60% in invaded pastures. Gardener et al. (2005) stated that it can 

can have high basal cover of 20%. Muyt (2005) estimated foliar cover across 102 quadrats each 25 x 25 m (= 25.5 ha) at 

ungrazed and not recently burnt ACT natural grassland (Yarramundi Reach) and found cover in excess of 75% in 2 quadrats, 50- 

75% in 1 quadrat, 25-50% in 21 quadrats, 5-25% cover in 30 quadrats, many individuals and up to 5% cover in 24 quadrats, and 

3-20 individuals in 9 quadrats. Bourdôt and Hurrell (1989a) assessed cover in 161 paddocks in New Zealand and found that most 

had ≤5% cover with plants present mainly as clumps or dense groups. Where cover was >25% the plants were present mainly in 

pure stands. Plants had persisted at the probable first introduction point in the area for c. 60 years and in 1988 that infestation 

consisted of a pure stand of several hundred m 2 (Bourdôt and Hurrell 1989a). Gardener et al. (2005) described an infested 

paddock near Guyra, NSW, as consisting of two communities. N. neesiana was dominant in the slightly better drained areas 

while the second community on lower, poorly drained areas was dominated by Festuca arundinacea with little N. neesiana. 

Bruce (2001) found it was dominant at 8% of sites investigated in the ACT, subdominant at 13% and common at 22%. Liebert 

(1996 p. 8) stated that it can “almost completely displace perennial native grasses” including T. triandra and had “destroyed” 

two wetlands at Laverton, Victoria, by excluding all other plants, in less than 10 years. In the Geelong area isolated plants on 

roadsides had become monocultures within 3 years (Liebert 1996 citing David Boyle). Slay (2002c) noted that it persisted when 

other pasture grasses fail. 

McDougall and Morgan (2005) measured the cover and frequency of N. neesiana on native grassland re-establishment areas on 

former agricultural land at Organ Pipes National Park from 1989 to 2003. The site was burnt in autumn 1993, 1995 and 1997 and 

there was a severe drought from March 1997. From initial values close to zero, both % cover and % frequency varied markedly. 

Percentage cover never exceeded c. 5% and was much

Campbell (1977, quoted by Gardener 1998) observed that Nassella trichotoma first appears in an area as widely sccattered 

tussocks, seven to twelve years later as patches, with scattered plants between them, and four to six years later as complete 

infestations. It appears that the development of N. neesiana infestations may show a similar pattern. 

Weed status 

Nassella species appear to become a problem wherever they naturalise. Verloove (2005, p. 112) that the complex of exotic 

stipoids including Jarava and Nassella spp. naturalised in Europe were“a serious threat for native vegetation” and “worldwide 

among the most troublesome weeds” the control and eradication of which was “very time consuming and expensive”. 

In New Zealand N. neesiana has been rated as extremely undesirabable (Bourdôt and Ryde 1986). It was declared a Class B 

noxious plant under the Noxious Plants Act 1978 by the Marlborough District Noxious Plants Authority in 1979 (Bourdôt and 

Hurrell 1987a) and later Class B in the whole country(Bourdôt 1988). However some have questioned its potential for major 

impact. Jacobs et al. (1989 p. 569) considered it “a localised troublesome weed of pastures”, Connor et al. (1993 p. 301) thought 

it has achieved only ‘modest success’, and that there were “no serious grounds” for predicting it would become a widespread 

problem, while Edgar and Connor (2000) considered to be only. “locally troublesome”. As of 2002 it was classed under the 

Biosecurity Act 1993 as requiring “Progressive Control” in the Marlborough region , and “Total Control” with the aim of 

eventual eradication in Hawkes Bay region (Slay 2002a). 

Although native to Chile, it was classified as a weed there because of the effects of the seeds on livestock and livestock produce 

(Gardener et al. 1996b - see their citation). 

N. neesiana was listed as one of 20 Weeds of National Significance (WONS) in Australia in 1999 (Iaconis 2003). The process of 

determining WONS was “the first attempt to prioritise weeds over a range of land uses at the national level” and was “not a 

purely scientific process” (Thorp and Lynch 2000 p. v). Of 71 weeds nominated by States and Territories N. neesiana was 

ranked 12th, based on evaluation by technical experts on six invasiveness questions, seven impact questions, potential for spread, 

and documentation of socioeconomic and environmental impacts (Thorp and Lynch 2000, McLaren et al. 2002a). The 

environmental impact assessment was rudimentary and “could be achieved only by taking a few pertinent environmental 

indicators and combining them into a ranking” (Thorp and Lynch 2000 p. 6). These were: 1. presence in a biogeographical 

region (each region being assigned equal value); 2. monoculture potential (ability to form pure stands giving a high score); 3. 

biodiversity indicator - based on the number of threatened species and special conservation areas (Thorp and Lynch 2000). N. 

neesiana was rated medium impact in respect of impact on threatened species, low impact in terms of threatened communities, 

“minimal national relevance” based on presence in less than 25% of biogeographic regions and low monoculture potential 

(Thorp and Lynch 2000 pp.15-16). Despite this, the species given a national listing. Recognition as a WONS resulted in a 

National Strategic Plan (ARMCANZ et al. 2001), increased mapping and recording, codes of practice to prevent spread and the 

development of better management methods (McLaren et al. 2002a). 

Groves et al. (2003b) developed a weed rating system for invasive plants in Australia and came up with the following 

ratings for N. neesiana: 

Australia 

5S 

New South Wales 

5nceS 

Victoria 5S, 

Tasmania 

OXS 


OXS 

Queensland, Western Australia, Northern Territory no rating 

where: 

5 = naturalised and known to be a major problem at 4 or more locations within a State or Territory 

0 = naturalised but only known naturalised population now removed or thought to be removed 

S = potential to spread further 

n = naturalised in part of a State 

c = under active control in part of a State 

e = eradication being attempted in part of a State 

X = potentially a greater agricultural problem than the rating shown 

The weed risk assessment process in Victoria, known as the the Victorian Pest Plant Prioritisation Process (Weiss et al. 1999, 

Weiss and McLaren 2002) enables the relative importance and potential impact of a species to be determined by scoring weeds 

on their invasiveness characteristics, current and potential distribution, and impact criteria. On a scale of 0-1, an invasiveness 

score of 0.72 was determined for N. neesiana, slightly less than that for N. trichotoma and much higher than Eragrostis curvula 

(0.50) (McLaren, Weiss and Faithfull 2004). 

There is little dissent from the view that N. neesiana is a serious pasture and environmental weed in south-eastern Australia 

(McLaren et al. 1998), however Grice (2004b) did not rate it as a threat in terms of non-pastoral agriculture, forestry, fire or 

amenity. Carr et al. (1992 pp. 41, 51) considered it to be a “very serious threat to one or more vegetation formations in Victoria”. 

McLaren et al. (1998) considered it to be potentially the worst environmental weed of indigenous grassland in Victoria, while 

McLaren, Stajsic and Iaconis (2004) considered it to be rapidly degrading critically endangered native grassland remnants in 

Victoria. It was listed as a perennial grass ‘weed of concern’ for South Australia (Virtue et al. 2004). Groves et al. (2003b) 

determined that it was having a direct impact on rare or threatened native plant species. In Victoria it has been portrayed as a 

strong resource competitor, “even choking out Nassella trichotoma ... in indigenous grasslands” (Iaconis 2003 p. 6). 

McLaren et al. (2002b) undertook a survey of land owners and managers in Victoria, the ACT and NSW and found that 5% of 

respondents, all from NSW, considered it a beneficial plant, while 86% did not. Even in the Angahook-Otway region of Victoria, 

an area largely occupied by forest, heathy woodland or heathland it has been rated by expert opinion as a weed of importance 

66

and high impact (“ability to cause actute disruption to ecological processes, dominae vegetation strata, cause severe loss of 

biodiversity”), with a “moderate” area that could be occupied in a “low” range of habitat types (Platt et al. 2005). 

Noxious weed status 

The noxious weed status of N. neesiana in Australia States and Territories has developed as follows: 

ACT: Declared Pest Plant under the Land (Planning and Environment) Act 1991 (Bruce 2001, ACT Weeds Working Group 

2002), then from 12 November 2005 the Pest Plants and Animals Act 2005 (Australian Weeds Committe 2007). It is classified as 

a C3/C4 pest plant: control is required (Glanznig and Kessal 2004), infestations must be contained and propagation and supply is 

prohibited. 

Victoria: First declared noxious on 27 October 2005, as a Restricted Weed under the Catchment and Land Protection Act 1994 

(Anon. 2005). It is an offence to trade and transport the plant and to deposit it on land, but there is no legal requirement for 

landowners to control infestations on their land. Listed as a high priority ‘New and Emerging Weed’ in the North Central Region 

Weed Action Plan 2001-2005 (Frederick 2002) and as a ‘Priority weed’ in the Regional Weed Action Plans for Port Phillip, 

North East, Goulburn-Broken, North Central, Wimmera,Corangamite andGlenelg-Hopkins regions (McLaren et al. 2002b). 

New South Wales: Declared noxious in 11 local government areas (Benson and McDougall 2005). In 2004 regional control and 

declaration was applied under the Noxious Weeds Act 1993,with prohibition of trade and eradication or control required 

(Glanznig and Kessal 2004). Declared a weed statewide in 2006 (Iaconis 2006a) and listed as a Class 3 (Regionally Controlled) 

weed for 25 Local Control Authority (LCA) areas and a Class 4 (Locally Controlled) weed for 103 LCA areas as of March 2007 

(Australian Weeds Committee 2007). According to Snell et al. (2007) it is Class 3 species in 30 LCA areas and a Class 4 species 

in all other areas. Both Classes cover weeds considered to be a serious threat to agriculture of the environment in the area. Class 

3 weeds are not, and Class 4 weeds are widely distributed in the area in which they are declared. Declarations of a Class 3 weed 

must include a minimum of three adjoining LCAs. LCAs may be either a local municipal government or a special purpose 

county council. N. neesiana is not prohibited from trade since Class 3 and 4 weeds are not “notifiable” (Australian Weeds 

Committee 2007). 

Tasmania: Under the Weed Management Act 1999, as of June 2004, prohibited in trade, prohibited import and control required 

(Glanznig and Kessal 2004). It is a declared plant under the Act and the details on actual restrictions imposed are contained in the 

weed management plan (Australian Weeds Committee 2007). These include prohibitions of the importation of seed, the 

importation of contaminated livestock, sale and distribution of the plant and measures to manage infestations and quarantine 

items contaminated with seed (DPIW 2007). Importation is also restricted under the Plant Quarantine Act 1997 (DPIW 2007). 

South Australia: Proclaimed under the former Animal and Plant Control (Agricultural Protection and Other Purposes) Act 

1986 for control in the whole State in 2001, and after repeal of that Act declared under matching sections of the Natural 

Resource Management Act 2004 (Iaconis 2006a). A Class 2 weed, requiring to notification by land owners to the relevant 

authority and control throughout the State (Australian Weeds Committee 2007). Prohibited in trade, prohibited import (Glanznig 

and Kessal 2004). 

Queensland: Under the Land Protection (Pest and Stock Route Management) Act 2002, as of 2004, prohibited in trade and for 

importation, eradication required (Glanznig and Kessal 2004). Declared a Class 1 weed throughout the State in 2005 (Iaconis 

2006b). Class 1 weeds are not commonly present in the State are prohibited in trade, and where established are subject to 

eradication (Australian Weeds Committee 2007). 

Western Australia: Classed as ‘unassesed’: not risk-assessed under Western Australian protocols, so not included in the 

‘Permitted and Prohibited List’ under the Plant Diseases Act 1989, but recognised as a declared pest plant in other States and 

therefore prohibited from importation (Glanznig and Kessal 2004, Australian Weeds Committee 2007). According to Snell et al. 

(2007) prohibited under Plant Diseases Act and listed as a P1 plant under the Agricultural and Related Resources Protection Act 

1976 which prohibits movement and trade in the species. 

Northern Territory: Under the Weeds Management Act 2001 not declared as of June 2004 (Glanznig and Kessal 2004), but 

declared by March 2007 as a Class A and Class C noxious weed under the Weed Management Act 2001: “to be eradicated”, 

prohibited from sale, and prohibited from introduction to the Territory (Australian Weeds Committee 2007, Snell et al. 2007). 

Undesirable characteristics 

In its introduced range including south-western Europe (Verloove 2005), New Zealand and Australia, N. neesiana forms dense 

stands in invaded vegetation, including bushland and parkland (Liebert1996), and in pastures, where it can form canopy cover of 

up to 60% (Gardener et al. 2003a). N. neesiana possesses many environmental traits that allow it to outcompete native 

vegetation (Gardener and Sindel 1998), being competitive for space, light, water and nutrients (Wells et al. 1986), and very 

competitive when mature (Cook 1999). 

N. neesiana is resilient after cutting and grazing. Like other grasses, the intercalary meristems (at the base of each leaf) are 

stimulated to grow by removal of the upper stems or leaves (Wheeler et al. 1999), an adaptation to vertebrate grazing, so plants 

usually produce new flowering stems after mowing (Muyt 2001). Grass leaves also continue to elongate after being cut, and their 

stems often return to an upright position after being flattened by wind or water (Wheeler et al. 1999). 

N. neesiana has the capability of dispersing rapidly e.g. south western Europe (Verloove 2005) over long distances. 

Its recorded impacts on biodiversity includes the ability to replace preferred indigenous vegetation (Wells et al. 1986), to cause 

loss of species, loss of food sources for wildlife and seed damage to wildlife (Liebert 1996). Kirkpatrick et al. (1995) considered 

it a threat to the Kangaroo grass – Blue Devil – Common Bog Rush Basalt Plains Grassland community, the main grassland type 

of the Victorian basalt plains. 

The seeds of various species of Stipeae have long been recognised as harmful to animal husbandry. Stipa capillata L. seeds 

penetrate the coats and flesh of cattle, “often leading to mortality or seriously injuring the oral cavity” (Tsvelev 1984 p. 859). N. 

67

neesiana seeds are similar to those of some of the more robust seeded Austrostipa spp. which McBarron (1976 p. 135) 

considered to be “undoubtedly ... the major cause of seed troubles for livestock, especially sheep in New South Wales”. These 

Austrostipa spp. are commonly cited as one of the main undesirable attributes of native pastures, requiring timely destocking to 

avoid problems (Garden et al. 2000). Seeds of Austrostipa spp. penetrate the eyes, mouthparts (Whittet 1969), skin and flesh 

(Mulham and Moore 1970) of sheep. “Wrinkled, long-woolled and young sheep are particularly susceptible ... the wrinkles and 

long wool collecting more of the seed, and the softer skin of young sheep allowing easier and deeper penetration. Severe damage 

to the eyes, jaws and feet can be caused by the seeds, which have also been known to penetrate the abdomen and internal organs 

... in extreme cases blindness, lameness, fever and death can result” (Mulham and Moore 1970 p. 105). Austrostipa and Aristida 

spp. are the most common contributors to ‘vegetable fault’ (plant contamination) of wool in Australia (Grice 1993). 

N. neesiana seeds are an irritant of skin (Wells et al. 1986) and “readily bore into the skins of animals, causing painful wounds” 

(Hayward and Druce 1919). Irritation of livestock by attached seeds causes discomfort and loss of condition (Wheeler et al. 

1990). Lambs appear to be particularly susceptible to eye injury (Bourdôt and Ryde 1986).. Infestation of livestock with seeds 

may be exacerbated by rain, as happens with Austrostipa (McBarron 1976). The hides of cattle are too thick for the seed to 

penetrate (Gardener et al. 1996b) but cattle may suffer injuries to the mouth and intestinal tract. The seeds “cause discomfort” 

for dogs and humans (Liebert 1996), and can injure pet animals (Snell et al. 2007) and could be expected to cause a range of 

serious medical problems based on their similarity to other stipoid seeds (see McBarron 1976). Awned seeds in general can 

readily penetrate the soft tissue of the buccal and gastrointestinal tracts, producing inflammation, abcesseses and tooth and gum 

disease (McBarron 1976). They may pass through the skin into muscle and can occasionally penetrate internal organs, 

potentially causing fatal injuries (McBarron 1976). However penetration of skin, carcases and eyes by N. neesiana seeds is rare 

on the northern tablelands of NSW (Cook 1999). 

The seeds are a contaminant of wool (Hayward and Druce 1919, Wells et al. 1986, Auckland Regional Council 2002). The halflife 

of seeds in the coats of sheep exposed to seeding plants for 17 days and then removed from exposure was measured at 7.5 

days, with nearly half the seed remaining embedded after 100 days and only very slow subsequent seed loss (Gardener et al. 

2003a). Upon removal of exposure, half the seeds on sheep had the callus embedded in the skin, but this reduced to 5% after 35 

days, and few seed penetrated through the skin into flesh (Gardener et al. 2003a). The seeds damage pelts, and reduce the quality 

of carcases and hides (Bourdôt and Ryde 1986, Bourdôt and Hurrell 1992, Slay 2002, Auckland Regional Council 2002). 

In the context of livestock grazing N. neesiana is a ‘conflict of interest’ species (Grice 2004b) because it is a valuable fodder for 

much of the year (Gardener 1998, Grech et al. 2004, Grech 2007a). Although it has harsh, often hairy leaves and tall course 

culms (Connor et al. 1993), it is considered to produce moderate quality, palatable forage during winter and early spring (Slay 

2002c) and to be of “modest” grazing value (Connor et al. 1993). In Argentina “it produces fairly good fodder” (Hayward and 

Druce 1919 p. 228) and is considered one of the most important winter grazing species, valued because of its perenniality, 

persistence, long life and good quality feed with relatively high crude protein levels in the young foliage (Gardener 1996, 

Gardener et al. 1999). Its undesirability as a pasture species results not only from the problems caused by the seeds but from the 

rapid reduction in foliage that accompanies the production of large numbers of unpalatable flowering stems in late spring and 

summer, which results in a large seasonal reduction in carrying capacity at a critical time of year. Livestock avoid the plant in its 

reproductive phase, so it gradually displaces more valuable pasture grasses (McWhirter et al. 2006). Its feed value (crude protein 

and digestibility) is less than that of deliberately grown pasture grasses at the same stage of growth (Gardener et al. 1996b, 

Gardener 1998, Cook 1999). It generally has a lower feed value than the widely cultivated, moderate feed value Dactylis 

glomerata and is less reponsive to applications of N fertiliser (Grech et al. 2004, Gaur et al. 2005). But it responds well to 

clipping (as a simulation of grazing), the regrowth sward after clipping having significantly higher crude protein, metabolisable 

energy and digestible dry matter contents than growth in unclipped swards (Grech et al. 2004). Fertilisation and clipping can be 

used to improve its usefulness as fodder (Grech et al. 2004) but grazing can promote its dominance when pasturage consists of 

more palatable species (Liebert 1996, Gardener 1998). 

N. neesiana is undesirable also because it can contaminate other agricultural produce, including hay (Frederick 2002). 

Increased fire risk has rarely been seen as a problem. Bartley et al. (1990) argued that “the greater height and density” of N. 

neesiana swards at Laverton North Grassland Reserve presented “a much greater fire hazard than native grasses”. According to 

Liebert (1996 p. 9): “Regional fire authorities recognise the fire risk ... and consequently slash swards from November to 

December”. However, comparative biomass production and breakdown assessments appear to be lacking and there appears to 

have been no proper evaluation of fire risk, which should involve comparisons with alternative vegetation states. 

All plants deplete soil moisture and the amounts of water used at particular times may have implications for co-occuring species 

or have a wider ecological impact. Slay (2001) observed that soil moisture in early January under a dense ungrazed sward was 

20.8%, while where the sward had been sprayed with glyphosate at flowering time it was 26.6% due to reduced transpiration and 

the reduction of evaporation due to dead thatch. Infestations in T. triandra grasslands presumably deplete soil moisture in spring 

and early summer, at the same time as the inter-tussock species are growing and before the main growing period of T. triandra. 

The overall effect could be a premature drying-out of the grassland landscape. 

Like other weeds N. neesiana can have beneficial impacts, although apart from its fodder value, these have hardly ever been 

recorded in its invasive range. Slay (2002a) noted that well stablished populations can provide erosion control on steep land. 

Control and management 

N. neesiana is difficult to control and according to Gardener and Sindel (1998 p. 78) there is “overwhelming evidence” that it is 

“almost impossible to eradicate” because of the difficulty of killing mature plants, the size and longevity of the soil seed bank 

and the production of basal cleistogenes. Gardener et al. (1996a p. 243) considered there then existed “no widely successful 

management techniques which result in the eradication or long term reduction”, while Gardener et al. (2003a p. 613) judged that 

“chemical and mechanical control have had little success to date, at best temporarily slowing its spread”. Slay (2002c p. 24) 

considered that the “overall tenacity” of Chilean needle grass made it “an extremely stubborn weed to manage and control”. He 

68

also noted that despite a wide range of control measures applied over a long period, land managers in New Zealand had reported 

“no success in terms of eradication” (Slay 2002a p. 7). Kirkpatrick et al. (1995 p. 35) claimed that it “seems impossible to 

control in its early invasive stage without causing great damage to native vegetation”. Liebert (1996) also considered it difficult 

to manage. Recent overviews of control techniques include Slay (2002a), Michelmore (2003) and Snell et al. (2007) the latter 

providing the most comprehensive details. 

An initial requirement in all management plans, but one often neglected, is to obtain a good representation of the plant’s 

distribution and the status of populations. Thus the ACT Weeds Working Group (2002) listed survey and monitoring as the first 

priority in their management plan, and Muyt (2005), for example, undertook one such survey. Site assessment should include 

mapping and density assessment (Snell et al. 2007). Targeted surveys, a public reporting mechanism and mapping of infestations 

have been identified as important elements in a regional management approach (ACT Weeds Working Group 2002). 

The critical foci of management activity is on techniques for depleting the soil seed bank by controlling seed input through 

mowing, herbicide application or cultivation/cropping (Bourdôt 1988, Bourdôt and Hurrell 1992, Slay 1992, Duncan 1993), 

slashing, burning or grazing (Slay 2002c, Beames et al. 2005, Grech 2007a, Snell et al. 2007), on reducing recruitment and 

density of established plants (Beames et al. 2005), along with prevention of spread (Bourdôt 1988, Frederick 2002, DPIW 2007, 

Snell et al. 2007). In pasture situations where eradication is unlikely, the emphasis has been on the development of methods for 

better utilisation, including strategic stocking (Duncan 1993, Gardener et al. 1999, McWhirter et al. 2006, Grech 2007a, Snell et 

al. 2007), and on agronomic techniques to maximise production of palatable foliage and minimise production of flowering culms 

and seed, including spray-topping or wick wiping, fertiliser application and intensive grazing (Slay 2002c, Grech et al. 2004, 

Grech 2007a). Long term management requires replacement by competitive species in pastures (Slay 2002c, Grech 2007a) and 

non-agricultural areas including native grasslands (Mason and Hocking 2002, Hocking 2005b) or conversion to another land use 

such as cropping or forestry (Slay 2002c). 

Morfe et al. (2003) presented a cost-benefit analysis of alternative management strategies for three different rates-of-spread 

scenarios. They considered it realistic to reduce infestations by 99% over 10 years only for infestations of less than 10 ha, and to 

reduce infestations by 90% for areas up to 500 ha. 

The vast majority of information on N. neesiana control and management relates to agricultural areas (e.g. Bourdôt 1988, 

Bourdôt and Ryde 1986 1987a 1987b, Duncan 1993, Gardener 1998, Gardener et al. 1996a 1996b 2005, Grech 2007a, Grech et 

al. 2004 2005 2007, McWhirter et al. 2006, Slay 2001 2002b 2002c, Snell et al. 2007) and the National Strategic Plan for N. 

neesiana (ARMCANZ et al. 2001) did not detail an integrated management approach for native grasslands (Downey and Cherry 

2005). To date the best guide for N. neesiana management in grasslands and grassy woodlands is that of Beames et al. (2005), 

although Snell et al. (2007) provided useful guidance. Various options for conservation areas have been suggested, with the aim 

of generally reducing populations while maintaining existing native species and managing to advantage the native plants 

(Bedggood and Moerkerk 2002). Spot spraying is the favoured method of eradication, using non-persistent herbicides (Beames et 

al. 2005). Snell et al. (2007) recommended use of flupropanate, but warned that it can be very damaging to many native grass 

species. Regular biomass reduction by burning or grazing is widely used in the south-eastern Australian grasslands (see below 

under grassland management) to maintain plant diversity and has been suggested as useful in reducing N. neesiana biomass and 

fuel load (Bedggood and Moerkerk 2002). Use of fire to reduce seeding and destroy fallen seed has also been recommended 

(Snell et al. 2007). Resowing with native grasses is recommended when more than three tussocks are removed or an area >0.5 m 

diameter is treated, either with the ‘spray and hay’ method (see below) or other techniques (Bedggood and Moerkerk 2002, Snell 

et al. 2007). Annual grazing and burning of native pasture was listed as a successful management option by Bedggood and 

Moerkerk (2002). The pasture is grazed heavily until N. neesiana enters the reproductive phase, stock are then removed and the 

area is burnt late, i.e. in autumn or winter, so as not to kill T. triandra. 

Hand removal 

Isolated plants and small infestations can be grubbed out (Bourdôt 1988), although Slay (2002a) considered this impractical. 

Liebert (1996) recommended using a mattock. Slay (2002c) recommended digging to at least 15 cm depth and 40 cm diameter 

and disposal of bagged material by incineration or deep burial. Snell et al. (2007) considered manual chipping a preferred 

method because the potential for regeneration from basal cleistogenes is eliminated if chipped plants are removed from the site. 

Herbicides 

According to Slay (2002a), who reviewed herbicidal management of N. neesiana, twenty five herbicides have been used in 

documented control attempts. The most frequently used in Australia have been glyphosate, flupropanate, 2,2-DPA, atrazine, 

hexazinone and simazine; the first two being the main chemicals used in Australian native grasslands (Lunt and Morgan 2000). 

The herbicides with soil residual properties, particularly flupropanate, are valued for the ability to prevent seedling emergence. 

Use of herbicides on N. neesiana in Australia was severely inhibited for many years by the failure of herbicide manufacturers 

and retailers to include the plant on herbicide labels. None of the chemicals mentioned by Duncan (1993) were at the time 

registered for use against N. neesiana in Australia, nor were those mentionded by Liebert (1996). This situation remained 

unchanged in October 2004, although State Government agencies, funded by the Australian Government, were undertaking 

detailed trials in order to achieve appropriate herbicide registrations (Iaconis 2004). Thus chemical control relied on ‘off-label’ 

use for a very long period. Lack of label recommendations in some cases meant that State Government officers could provide no 

recommendations or advice on N. neesiana herbicidal management (Iaconis 2004). In Victoria, a government ‘Code of Practice 

for Provision of Chemical Advice to Clients’ (DNRE 2000) prohibited any mention of off-label uses by government officers 

involved in recommending herbicides for weed control. Interpretation of the Victorian Code was complicated by the listing of 

broad weed categories such as ‘perennial grasses’ on some herbicide labels. In New South Wales and the Australian Capital 

Territory label directions were able to be overridden byAustralian Pesticides and Veterinary Medicines Authority (APVMA) 

permits (Michelmore 2003). NSW recommendations relied on a series of temporary APVMA permits, for pasture spraying with 

glyphosate and flupropanate, spray-topping with glyphosate, and use of fluazifop-P in legume pasture and lucerne (listed in 

69

Storrie and Lowien 2003). However under Victorian legislation governing pesticide use these permits did not legalise use: offlabel 

use, with certain provisions, not being illegal. 

Label registration for all Nassella species was finally achieved for a flupropanate product in 2004 (McLaren et al. 2005). Label 

recommendations were for application to actively growing plants, by boom or spot spraying, once per year, in urban open space, 

woodlands, roadsides, nature reserves and pastures. Other products with on-label uses soon followed. As of October 2007 the 

only other herbicide with specific label recommendatons for N. neesiana was glyphosate, registered only for spot spraying, 

although a number of minor use permits were in operation (Snell et al. 2007). 

Glyphosate 

Glyphosate prevents synthesis of essential aromatic amino acids. It is probably the world’s most widely used herbicide. It is a 

non-selective, systemic, water soluble herbicide, absorbed by foliage, rapidly translocated throughout the plant and non-residual, 

being inactivated on contact with soil (Tom.lin 1997). It has relatively rapid activity, kills all plants, and creates bare ground, so 

when there is a soil seed bank of N. neesiana the plant quickly re-establishes (Liebert 1996). Pritchard (2002) found it gave 

excellent initial control, preventing seed formation when applied at the boot stage, but that 10 months after application many 

tussocks showed new growth. 

Off-target damage from glyphosate can be high, but depends on the selectivity of the application method. There is no 

withholding period for livestock grazing or other agricultural activity although stock grazing should not be undertaken for 7 days 

after application (Snell et al. 2007).. 

Flupropanate 

Flupropanate (flupropanate-sodium, formerly called tetrapion) is an halogenated (fluorine) alkanoic acid which functions as an 

inhibitor of fat sysnthesis. It is a water soluble, systemic herbicide with low contact activity, taken up mainly by roots, and used 

only for control of perennial and annual grasses in pastures and uncultivated areas (Tomlin 1997). It is moderately selective and 

slow acting, with soil residual effects that kill emerging seedlings (Parsons and Cuthbertson 1992, McLaren et al. 2005). 

Recommended rates for control of mature N. neesiana are 1.5-3 L ha -1 (Grech et al. 2009a). Effects on target plants may be 

noticeable after 3-5 months but it may take upt to 12 months for plants to die (Snell et al. 2007). Pritchard (2002) found only 

slight foliage death 15 weeks after treatment but recorded 92-95% kill of tussocks after 45 weeks. The residual effects are 

increased at higher application rates and their duration is dependent on rainfall, which leaches the herbicide from the soil 

(Michelmore 2003). Fluporanate activity and behaviour is dependent on soil type – higher rates are used on fertile basalt soils 

and in higher rainfall areas, lower rates on infertile soils derived from sedimentary and granitic rocks and in lower rainfall areas 

(Snell et al. 2007). Flupropanate provides some selectivity: Austrodanthonia, Austrostipa spp. and Microlaena are killed, 

especially at higher applications rates, while T. triandra, Bothriochloa macra and Poa labillardieri Steud. are generally tolerant 

at label rates. Phalaris aquatica L., Dactylis glomerata L. and Festuca arundinacea are affected but recover, while young 

Trifolium subterraneum L. are severely affected (Michelmore 2003). In Australia, a withholding period for grazing stock or 

cutting for stock food of 4 months minimum after blanket application is mandatory, and 14 days for spot application, along with 

some other restrictive provisions for production agriculture (McLaren et al. 2005). 

Flupropanate has been the preferred herbicide for spot spraying in agricultural areas and on land lacking significant native cover 

(Liebert 1996). An infestation in Geelong, originally with 80% cover, was reduced to 10% cover after spraying in 3 consecutive 

years and the broadleaf weeds that initially colonised were short-lived and some native grasses began to establish (Liebert 1996). 

Grech (2007a) found that set-stocking combined with flupropanate spraying resulted in more bare ground and an increase in 

infestations. Of the range of herbicide and stocking options examined, flupropanate combined with strategic grazing and pasture 

rehabilitation provided the best control. Grech et al. (2009b) found that low rate treatments with flupropanate (0.25 and 0.75 L 

Ha -1 ) we ineffective in controlling N. neesiana seedlings. Grech et al. (2009a) found that flupropanate at 2 L Ha -1 effectively 

controlled N. trichotoma in pot trials, but not N. neesiana, and that a range of native and exotic grasses was unaffected apart 

from a transient decline in growth. 

A major disadvantage of flupropanate is the creation of bare ground that persists for long periods (Snell et al. 2007). These areas 

may be prone to subsequent reinvasion. 

2,2-DPA 

2,2-DPA is an halogenated alkanoic acid which acts by precipitation of protein (Tomlin 1997). It is a systemic herbicide 

absorbed by leaves and roots, used mostly for control of annual and perennial grasses (Tomlin 1997). It is selective for some 

grasses at low rates, and at high rates provides non-selective control of monocots (Parsons and Cuthbertson 1992). It is generally 

considered to be less effective and useful than the two most favoured herbicides, and does considerably more damage to nontarget 

species than flupropanate (Michelmore 2003). Pritchard (2002) found it quickly killed foliage, but 45 weeks after 

application had only reduced the number of tussocks by 40%. Hartley (1994) found that 2,2-DPA reduced seed production by 

90-99%, but detected resistance to 2,2-DPA in New Zealand after 2 years of annual or biannual application. 

Hexazinone 

Hexazinone is an inhibitor of photosynthesis and is a non-selective contact herbicide, absorbed by leaves and roots (Tomlin 

1997), mainly the latter (Parsons and Cuthbertson 1992). It has a long residual life in soil (Parsons and Cuthbertson 1992) and 

maintains bare ground for long periods, and is available commonly in granular preparations. 

Triazines 

Atrazine is a triazine chemical which acts by inhibiting photosynthesis. It is a water soluble, selective systemic herbicide, 

absorbed mainly through the roots but also through foliage, mainly used to control broadleaf weeds and annual grasses (Tomlin 

1997). It is relatively rapid acting and has pronounced residual effects for 6 months or more after application (Parsons and 

Cuthbertson 1992). Simazine is another triazine compound that acts by inhibiting photosynthesis. It is mainly root-absorbed and 

provides selective pre-emergence control of grasses and broad-leaved weeds, with long residual control at high application rates 

(Parsons and Cuthbertson 1992). Pritchard (2002) found that both atrazine and simazine application at the boot stage resulted in 

70

a small reduction in seedhead production but no other useful control. Atrazine was used against Nassella spp. by Philips and 

Hocking (1996) and Mason and Hocking (2002). Mason (2005) reported reductions from c. 20 to 7 plants m -2 two months after 

application of atrazine at an unstated rate. 

Other herbicides 

Paraquat dichloride (“Gramoxone”), a non-selective contact herbicide, has some activity against N. neesiana but is not as good 

as glyphosate in limiting seed production (Bedggood and Moerkerk 2002). Grass selective herbicides such as fluazifop-P are 

reportedly effective on N. neesiana seedlings but require excessively costly high application rates to control larger plants 

(Bedggood and Moerkerk 2002). Fluazifop-P is a relatively fast acting and selective systemic herbicide absorbed by foliage, is 

not residual and usually kills N. neesiana in 3-5 weeks (Snell et al. 2007). Slay (2001, 2002c) found in spray topping treatments 

that haloxyflop was superior to glyphosate in reducing seed production and subsequent seedling emergence but was more than 

ten times as costly. 

Mason (2005) reported that acetic acid (4% acetic acid vinegar) and surfactant solution applied at 0.5 L m -2 gave close to 100% 

control of above-ground N. neesiana two months after application and reduced plant density from c. 20 to 8.6 plants m -2 . 

Soil fumigants 

Hurrell et al. (1994) examined the efficacy of three soil fumigants to kill buried seed. They found that dazomet and methyl 

bromide were highly effective, killing 98% of viable seed, while metam-fluid was less effective (83% of viable seed killed). 

Cleistogenes and panicle seed appeared to be similarly susceptible. Soil treated with these fumigants can be resown within 7 

days of application, in contrast to residual herbicides such as hexazinone that create bare ground for long periods. However soil 

fumigation is expensive, and may be hazardous and difficult. Small patches have been successfully treated with Dazomet (Slay 

2002c). 

Wick wiping 

Hocking (2009) demonstrated that mechanised wick wiping using glyphosate prior to flowering achieved major reductions in 

density and cover of N. neesiana, but was less effective when the height differential between with other plants was small and 

when active N. neesiana growth was limited. Mature tussocks density was reduced by up to 95% when wiping was undertaken in 

two successive years. Wick wiping was viewed as a valuable approach to integrate with mowing to reduce roadside infestations 

to a density at which subsequent spot spraying could be cost-effective (Hocking 2007 2009), but the technique may have wider 

applicability. 

Grech et al. (2009c) examined the effects of wick wiping with glyphosate, flupropanate and a tank mix of the two herbicides in 

exotic pasture grassland and found that wiping was no more effective in achieving control than boom spraying, although it used 

less herbicide. The tank mix was ineffective, indicating antagonistic responses of the two herbicides. 

Agricultural areas 

Early recommendations for control in agricultural areas were for spraying of glyphosate to kill established plants, and of 2,2- 

DPA (dalapon) to control seedlings in pasture (Bourdôt and Ryde 1986 1987a 1987b, Bourdôt 1988) and to prevent flowering 

and reduce plant density (Hartley 1994). Duncan (1993) recommended glyphosate application in autumn and flupropanate in 

spring or autumn and noted that paraquat could be mixed with the flupropanate “to provide a quicker dessication and eliminate 

seed production”. Storrie and Lowien (2003) recommended a glyphosate/flupropanate mix for this latter purpose, but research by 

NSW Agriculture found that glyphosate reduces the effects of flupropanate and it is better not to mix them, except when a quick 

marker is required for the treated area. 

Bourdôt and Hurrell (1987a) found that hand-held wiper application of glyphosate in spring lead to higher rates of tussock 

survival (56% of plants with some surviving tillers) than boom spray application, which killed all plants (as assessed 9-18 

months post-treatment), except when applied under drought conditions. 2,2-DPA was less effective, but suppressed the plant 

except when applied in winter. Other herbicides tested were ineffective. Wick wiper application was suggested but not 

experimentally examined. Storrie and Lowien (2003) reported that wiping with glyphosate between flowering and the milkydough 

stage of the grain prevented panicle seed set, but killed

Bourdôt and Hurrell (1987b) tested a range of herbicides for control in lucerne. Hurrell and Bourdôt (1988) tested granular 

formulations of three herbicides and found that hexazinone at 16 kg ha -1 killed plants and prevented seedling establishment for at 

least 7 months. Bourdôt (1988) recommended hexazinone for infestations on roadsides and waste places. 

Establishment of competitive plants (pasture species or fodder crops) is generally recommended after herbicidal control on 

agricultural land (e.g. Bourdôt 1988, Liebert 1996). 

Natural areas 

Herbicidal treatment of Nassella spp. in native vegetation has often had devastating effects on native vegetation, killing native 

plants or whole populations and facilitating irreversible weed invasions (Hocking 1998). Non-selective herbicides and nonselective 

application measures constitute one aspect of the problem. Uncritical adoption of recommended agricultural 

applications is another. A consistent complicating factor is the low level of knowledge of the resistance and susceptibility of 

native plants to particular herbicides across a range of seasons. Little detailed information appears to be available about offtarget 

effects in native grasslands, perhaps because there is reluctance to admit that good intentions can produce bad results. 

McDougall (1989) established that atrazine had “no direct effect” on Themeda triandra establishment and growth and suggested 

it could be used to reduce weed competition in T. triandra grassland sites lacking native species that are sensitive to it. Harris 

(1990) recorded that this herbicide would be a major component of N. neesiana control programs in native grasslands in the 

Melbourne area, including Derrimut and Laverton North, and that trials were to be undertaken comparing high volume spot 

spraying with low volume application methods such as ‘Micro-Herbi’, ‘Splatter Gun’ or ‘Gas Gun’. 

Flupropanate at low rates (1.5-3 kg active ingredient ha -1 ) can be used to selectively remove Nassella trichotoma from pastures 

with little effect on native grasses including T. triandra, Botriocholoa macra and Poa labillardieri, but Austrodanthonia spp. and 

Microlaena stipoides are killed at less than half this application rate, and legumes may be damaged (Campbell 1997 1998). 

Hocking (1998) warned that atrazine, simazine and flupropanate were likely to be highly toxic to native grassland forbs. 

Bedggood and Moerkerk (2002) stated that atrazine killed N. neesiana in native grasslands with little effect on T. triandra. 

Dare and Hocking (1997) tested glyphosate, atrazine, flupropanate and simazine at unspecified rates for control of N. neesiana in 

Melbourne area native grasslands and found that atrazine and flupropanate provided effective kill when applied in winter, as 

assessed 11 months later. Atrazine was found to act fast, providing effective kill within 10 weeks, and treated plots remained 

weed free. Glyphosate and simazine were slower acting (12 weeks) and glyphosate plots were reinvaded by weeds. 

Brereton and Backhouse (2003 p. 3) observed that herbicides used to control grass growth in native grasslands to reduce fire fuel 

loads “usually” promoted “the establishment of exotic grasses to the detriment of the native flora”, and Williams (2007) found 

that herbicide application as an alternative to burning for fuel reduction has wiped out a number of remnant native grasslands in 

western Victoria. Significant herbicide damage to desirable native species has been noted in the ACT. To address this the ACT 

Weeds Working Group (2002) recommended promotion of glyphosate spraying at the correct time and use of low rates of 

flupropanate for seedling control, as well as the use of alternative control techniques and greater concentration on control in 

buffer zones to restrict spread to conservation areas. Muyt (2005) suggested the use of either grass-selective or non-selective 

herbicides from autumn to late spring in one ACT grassland. McDougall (1989) argued that fluazifop would have advantages 

since is has no direct effects on establishment and growth of Themeda triandra, although it inhibits flowering, and is ineffective 

at reducing competition by annual forbs. 

Puhar (1996) tested the impact of glyphosate, flupropanate and atrazine on N. neesiana by measuring the root and shoot lengths 

of seedlings germinated on agar plates impregnated with herbicide at various rates. Germination of dehulled seeds (i.e. lemma 

removed) was unaffected by any of the herbicides at any concentration. Glyphosate had significant negatives effect on root and 

shoot elongation. Atrazine had a lesser, but still significant effect, while flupropanate also caused significant negative impact, but 

with higher root growth rates than for glyphosate. All four herbicides were judged to be able to kill emerging seedlings at 0.25 of 

the label rate for grassland spraying, but the rapidly acting glyphosate was more damaging. Similar tests were undertaken for T. 

triandra, but the susceptibility to off-target damage of the remainder of the grassland flora remains largely unknown. 

Disadvantages of herbicidal management 

Whatever the herbicide used, cleistogenes that have already matured but remain attached to the plant, concealed beneath leaf 

sheaths are not killed (Hurrell et al. 1994). Furthermore, when a large N. neesiana seed bank is present, baring the ground with 

herbicides encourages seedling recruitment and may lead to rapid re-establishment, and an ultimate increase in density and cover 

(Hartley 1994, Gardener et al. 1996b, Gardener et al. 1999, Lunt and Morgan 2000, Slay 2002a, Storrie and Lowien 2003). 

Herbicidal management has often resulted in the expansion of populations and exacerbation of spread due to the elimination of 

competition (Slay 2001 2002a 2002c). When herbicides are applied in spring, for example, the basal cleistogenes are already 

mature, there is a greater effect on potential competitor plants and the bare areas created allow for more N. neesiana seed to 

germinate (Slay 2002c). 

However if the seed bank is small, herbicial control may not simultaneously provide conditions for seedling establishment. Britt 

(2001) found no seedling growth 6 months post-treatment. Hocking (2005b) reported on the seed bank at infestations managed 

by repeated herbicide applications, mainly of glyphosate, in south-central Victoria and Hobart. In all cases the density of filled 

seeds (i.e. seeds containing a grain and embryo) in the seed bank was

Slashing and mowing 

Slashing can reduce the production of panicle seed and stem cleistogenes but has little other benefit, apart from stimulating 

regrowth that is more palatable to grazing livestock (Snell et al. 2007). Britt (2001) found that plants slashed to 1 cm in early 

May at Greenvale had failed to produce flowering stems by late October, whereas 60% of unslashed plants had produced 

flowering heads by that time. Hocking (2005b) found that mowing in spring had no major effect on the number of mature 

tussocks and failed to significantly reduce tussock size. Slay (2001) incorrectly attributed to N. neesiana the findings of Mulham 

and Moore (1970) about mowing Austrostipa swards. Those authors found that mowing in late September prevented seed 

maturation and that late mowing could allow seed maturation on cut stems. Slay (2001) found that mowing twice at early seed 

head emergence (Feekes stages 10-10.4) totally prevented the formation of viable panicle seed and stem cleistogenes and 

resulted in regrowth acceptable for livestock grazing. Grech (2007a) also studied slashing impact in pastures and quantified some 

benefits in terms of palatable regrowth that could be exploited by grazing. Mowing before emergence of the panicle may result 

in the growth of more reproductive tillers, so repeated treatments are generally required. The best time to slash or mow is before 

flowering when 25-75% of the heads have emerged from the flag leaf sheath (Slay 2002c), or at the flowering stage (Grech 

2007a, Snell et al. 2007). Slay (2002c) noted that the resultant reduction in shade may facilitate the survival of seedlings. 

Mowing can encourage the formation of prostrate swards, so may also reduce the production of stem cleistogenes (Snell et al. 

2007). 

Fire 

Published information on the value of fire as a managment method are equivocal. According to Bourdôt (1989) the New Zealand 

experience was that repeated fires favour the grass, and when tried as a means of control in Marlborough appeared to have 

“eliminated other species resulting in pure stands”. Muyt (2001 p. 73) stated that fire “stimulates vigorous regrowth”, often 

promotes spread, but is useful for improving access and effectiveness of herbicide application by removal of litter and dead 

material. However he also noted that fire “generates bare ground and reduces immediate competion, conditions that are ideal for 

seed germination” (Muyt 2001 p. 73) so should therefore should be followed by herbicide application (Muyt 2005). In pastures, 

Slay (2002c) recommended spraying with glyphosate to dry out the grass before burning and follow-up herbicidal control to kill 

seedlings that grow from cleistogenes in burnt tussocks. 

Snell et al. (2007) advocated the use of fire to prevent seed set, burn off standing seed and to stimulate growth of seed from the 

soil seed bank, but advised integration with other selective control techniques. They also noted its usefulness in enabling a 

clearer indication of intensity and pattern of infestations. 

In T. triandra grasslands burning in spring is recommended, so competing cover establishes quickly (Muyt 2001). Hocking 

(2005b) found that burning in spring in a native grassland under drought conditions resulted in major reduction of large, mature 

tussocks and proliferation of small tussocks, probably derived from the large. The total area occupied by tussocks was decreased 

by >75% without decreasing the density of the population. Late spring burns reduced seed production by half, and did not lead to 

major seedling recruitment. He suggested that fire was therefore useful to weaken the stand, making it more susceptible to 

subsequent control activities of a different type and in limiting seed production, and could play a role in a containment strategy 

and integrated management. Hocking (2007) noted that fires kill few plants and that seed production resumes within a year. Snell 

et al. (2007) agreed that burning usually does not result in much kill of mature plants. However reductions in density of c. 90% 

have been reported after annual November burning for 5 years at Plenty Gorge Parklands, Victoria, however N. neesiana was 

replaced by Phalaris aquatica (Snell et al. 2007 p. 34), which may have even worse biodiversity impacts. 

The effects of repeated periodical burning and of fires in autumn do not appear to have been adequately assessed by scientific 

tests. In particular better knowledge of the effects of fire on the litter and near-surface soil seed banks would be desirable. 

Cultivation and cropping 

The objective of cultivation and cropping is too reduce the soil seed bank before re-establishment of a vigorous pasture that can 

suppress any seedlings that may subsequently appear. Cultivation stimulates seed germination and shallow cultivations are 

supposedly superior in depleting the seed bank because most of the seed is close to the surface (Snell et al. 2007). Herbicides 

applied by boom spray are recommended as the intitial treatment, followed by a series of shallow cultivations interspersed with 

dense, uniform sowing of annual fodder crops (Bourdôt 1988, Duncan 1993, Slay 2002a 2002c). Shallow cultivation (no greater 

than 5 cm) is recommended so that seed will not buried and the maximum amount will germinate. Slay (2002a) warned that the 

cultivation technique and timing were critical to minimise dispersal of cleistogenes. Duncan (1993) recommended winter 

cropping for two years followed by pasture establishment, and in less arable areas herbicide spraying, direct drilling of fodder 

crops, and fertiliser application. Bourdôt (1988) recommended three years of such management before sowing of ‘permanent’ 

pasture. Storrie and Lowien (2003) broadened the range of recommended crops to include summer forage species and summer 

grains, and the aerial application of herbicide, seed and fertiliser to treat steep, rocky, inaccessible areas. Establishment of 

lucerne after glyphosate spraying, with residual grass herbicide application and winter spraying with grass-selective paraquat and 

metribuzin is an alternative approach (Bourdôt 1988).The best methods, including crop and pastures to sow, depend on the 

specific characteristics of the infestation, regional agronomic factors, etc. (Slay 2002c). 

Grazing 

Overgrazing is likely to favour N. neesiana, due to its lower palatability (Bourdôt 1988) at least during its reproductive period 

(Grech 2007a). Crude protein, metabolisable energy and digestible dry matter contents of N. neesiana peak in winter and early 

spring and decline markedly from September to December, corresponding with the onset of reproductive phase (Grech 2007a). 

Strategic grazing in pastoral ecosystems to best utilise the green feed produced by N. neesiana in winter and spring has been 

intensively studied (Gardener 1998, Grech 2005a 2007a). Intense grazing pressure over a short period is required to suppress 

seed production (Storrie and Lowien 2003) but in practice is very difficult to achieve and probably not feasible in most grazing 

enterprises (Slay 2002a, Grech 2007a). In addition there is a risk that overgrazing will encourage further N. neesiana 

establishment (Slay 2002a). Set stocking with sheep exacerbates the problem by reducing the cover of desirable pasture grasses 

and creating more bare ground suitable for further N. neesiana establishment (Grech 2007a, Snell et al. 2007). Preliminary 

73

esults of grazing trails reported by Grech (2005a) at Greenvale, Victoria, showed that cattle reduced the amount of panicle seed 

produced in comparison to ungrazed paddocks by 95%, and in comparison to sheep grazed paddocks by 77%, when stocking at a 

normal rate for the district, at 12 Dry Sheep Equivalent ha -1 . Both sheep and cattle continued to gain weight over the spring. 

Strategic grazing can be combined with spray-topping and/or wiping before flowering. Grazing management is predicated on 

acceptance of N. neesiana as an ineradicable pasture component and making best use of the feed it offers. In pasture situations it 

is a weed one can ‘learn to love’ (Storrie 2006). 

Shade 

Establishment of Pinus radiata to shade out N. neesiana is being investigated by Hawkes Bay Regional Council in New Zealand 

(Slay 2002a). Five years after plantation establishment, N. neesiana was “weaker, rotting and forming a dense thatch” (Slay 

2002a p. 33). 

Quarantine and restriction of dispersal 

Bourdôt (1988) recommended a range of measures to restrict seed dispersal: restricting livestock access to seeding plants, 

appropriate mangement for contaminated stock, not harvesting fodder from infested land, cleaning of contaminated vehicles, 

machinery and clothing, and eradication from flood-prone land. Additionally Liebert (1996) mentioned the use of only weed-free 

fodder and seed, and not moving contaminated soil. These measures can be implemented by strategically timed slashing and 

mowing, restricted grazing (including droving of livestock along infested roadsides), fodder harvesting (Liebert 1996, Frederick 

2002), and vehicle movement, signage to alert people to N. neesiana presence, particularly on roadsides (Liebert 1996, Frederick 

2002), machinery hygiene programs (Frederick 2002, Moerkerk 2006a), State-wide quarantine (DPIW 2007) and other measures. 

The presence of stem cleistogenes requires that fodder harvested from infested swards at any time should not be fed out in 

uninfested areas. 

To minimise seed movement on livestock the Tasmanian Department of Primary Industries and Water has prescribed measures 

in Regulations under the Weed Management Act 1999. The length of hairs on the coat is not to exceed 25 mm, seeds are not to be 

adhering to the animal, a permit for importation is required and the animals must be imported to an approved facility or 

slaughterhouse. Suggested actions include liaison with suppliers and confinement of the animals in holding pens until they have 

been thoroughly inspected and have completed “bowel evacuation” (DPIW 2007 p. 3). Similar hygiene activities are 

recommended for clothing, machinery, soil and other materials. Persons wishing to dispose of contaminated materials must 

contact a government officer who shall determine whether removal to a quarantine place or destruction in situ is most 

appropriate (DPIW 2007). 

Integrated management in native vegetation 

Spot spraying with glyphosate is the usual management method in natural or semi-natural areas (e.g. at Organ Pipes National 

Park, McDougall and Morgan 2005). Repeated herbicide treatments are commonly required to kill mature plants (Muyt 2001) 

and several cycles of treatment and monitoring are required for long-term control (Liebert 1996, Frederick 2002). Liebert (1996) 

considered the residual activity of flupropanate made it unsuitable for use in native vegetation. Bedggood and Moerkerk (2002) 

noted that investigations were underway in the ACT to determine spraying times when the native plants were dormant and thus 

less likely to be impacted. However native C 3 grasses probably generally have similar growth periods to N. neesiana, so the 

scope for such temporal selectivity appears very limited. Victorian experience is that native species are always damaged 

(Bedggood and Moerkerk 2002, citing C. Hocking). 

Liebert (1996) suggested that burning before November could help to kill seeds and promote germination of cleistogenes, which 

could be sprayed with glyphosate the following autumn. 

Lunt and Morgan (2000) recommended maintenance of dense swards of the dominant grass as the most efficient means of 

limiting the establishment and density of N.neesiana in T. triandra grasslands, and re-establishment of dense T. triandra after 

herbicidal control of N. neesiana. They reported slower and comparatively little invasion where T. triandra cover exceeded 50%, 

and in one invaded grassland, proximity to large infestations did not appear to be a significant factor in determining the presence 

of the weed in particular quadrats (Lunt and Morgan 2000). Oversowing of T. triandra does not appear to reduce the soil seed 

bank of N. neesiana, possibly because the growing periods of the two species has little overlap, however Austrostipa or 

Austrodanthonia spp. may be more suitable (Beames et al. 2005). Direct drilling of T. triandra seed after spraying has been 

found to provide effective control (Liebert 1996 citing Craig Bray). 

Current best practice management in invaded Themeda-dominated grasslands has been detailed by Beames et al. (2005) and 

Hocking (2005b) and involves finely targetted biannual or more frequent glyphosate spraying before flowering, followed by 

establishment of T. triandra. Fire can be integrated into these programs both to improve germination of native and N. neesiana 

seeds and open up the landscape, and the broadcasting of native seeds is recomended. A brief case study of such an approach 

was provided by Snell et al. (2007). 

Muyt (2005) recommended selective mowing with catcher mowers at the edges of dense stands, but the scale and irregular 

distribution of patches in most infestations make this approach impractibable. 

In uninvaded grasslands, best practice management is focused on minimisation of major disturbance, and the maintenance of 

existing native grass sward density and cover. Many herbicides used to control Nassella spp. have severe impacts on native 

vegetation and can result in major weed invasion similar to those which occur after ploughing (Hocking 1998). As noted in more 

detail above, when the dominant native grass dies or is killed by disturbance, the N and P held in the plant is released into the 

soil, creating a nutrient flush which enables successful establishment of Nassella spp. (Henderson and Hocking 1997, Wijesuriya 

and Hocking 1997, Hocking 1998). 

A range of techniques have been developed for re-establishment of native grasses and replacement of N. neesiana in native 

grasslands (McDougall 1989, Stafford 1991, Hocking 2005b) and have been recommended for use for paticular areas (e.g. Muyt 

2005). However the development of effective techniques requires much improved understanding of the underlying ecological 

processes (Hocking and Mason 2001). 

74

Themeda re-establishment 

McDougall (1989) reported the results of detailed experiments to determine suitable methods for the re-establishment of 

Themeda triandra. The most effective method involved application of T. triandra thatch immediately after harvest in January, 

followed by burning of the thatch in September when dry enough. This provided seedbed conditions favourable for T. triandra 

germination and seedling establishment. A mulch of brown coal greatly improved seedling establishment from surface-sown 

seed. Hebicide spraying prior to T. triandra seed germination indicated that establishment was not inhibited by low-growing 

weeds. Spring burning promoted germination of T. triandra seed in the soil seed bank. 

Stafford (1991) reported on long-term investigations of techniques to establish T. triandra to restrict weed invasion in secondary 

grassland and woodlands. The ability to burn established T. triandra swards in spring/early summer and its tolerance to 

herbicides used to control woody weeds provided the weed management advantages. His technique can be characterised as 

‘hay/spray/burn’: haying with T. triandra culm thatch, spraying to suppress weeds and burning to remove the thatch and dry 

weeds. Herbicidal control of weeds was critical to success, but the effective herbicides are widely active against a range of 

plants, both exotic and native. High labour inputs to obtain sufficient seed were identified as a major problem. A vehiclemounted 

reel-stripper was developed, but proved to have limited efficiency in gathering undamaged, chaff-free seed. It also 

damaged the sward. Modification of the stripper, including addition of wire flails to the rotor, enabled harvesting of the panicles 

at high efficiency. However germination of the material harvested was only 10% of that with hand-cut material. Harvesting 

sufficient seed for the areas requiring rehabilitation remains an ongoing problem. 

‘Spray and hay’ techniques have been developed and modified to a point where N. neesiana swards in Themeda grasslands can 

be selectively removed and replaced with T. triandra when seasonal rainfall is adequate (Dare and Hocking 1997, Hocking 1998, 

Mason 1998, Mason and Hocking 2002, Mason 2004, Hocking 2005b). The method, first developed for N. trichotoma (Phillips 

and Hocking 1996, Mason 2004) involves mowing in late spring to reduce N. neesiana biomass, spraying of N. neesiana plants 

with glyphosate the following autumn, application of seed-bearing T. triandra grass hay in winter and allowing the seeds to bury 

themselves in the ground, then removal of the hay thatch by hand or by burning in early spring (Hocking 2005b). Tested 

variations on the technique have involved tilling of the soil and use of other herbicides (Mason 1998 2004). Dense patches of N. 

neesiana have been controlled in this way in dry and average rainfall years, with N. neesiana cover reduction from >65% to c. 

10% and T. triandra cover increase from c. 5% to >85% over 3 y with c. 17 T. triandra and c. 6 N. neesiana plants m -2 

established in the most successful trials (Mason and Hocking 2002, Hocking 2005b). A very similar ratio of T. triandra to N. 

neesiana tussocks was found in these trial plots 5 years later, adding weight to the finding of Lunt and Morgan (2002) that 

invasion of N. neesiana is restricted when there is a healthy cover of T. triandra (Hocking 2005b). 

Atrazine has also been used as the herbicide in the ‘spray and hay’ method (Mason 1998). Both atrazine and acetic acid vinegar 

were tested in a small trial by Mason (2005) but acetic acid treatments resulted in much reduced T. triandra seedling 

establishment and control of Nassella spp. that was less than, or approximately equivalent to that provided by atrazine, 10 

months after treatment. 

Dare and Hocking (1997) reported an unsuccesful trial of the method, blamed on very low rainfall during summer. Appropriate 

timing of the stages of the treatment is critical, and abundant, high quality seed is required (Mason and Hocking 2002). The best 

time to spray when using the spray and hay method was determined by Phillips and Hocking (1996), who found that late autumn 

to winter herbicide application resulted in a 3 month weed-free window, whereas February and September spraying resulted in 

rapid invasions by a suite of weeds. 

‘Spray and hay’ methods have generally had poor effectiveness under drought conditions (Dare and Hocking 1997, Hocking 

1997, Hocking pers. comm.) and are subject to the same general constraints on rehabilitation of native grasslands as other 

methods, notably the difficulty of harvesting and very limited supplies and low quality of native seed or thatch (McDougall 

1989, Bedggood and Moerkerk 2002, Muyt 2005) and high labour inputs. 

To circumvent some of these problems Muyt (2005) recommended sowing of other native perennial grasses at the same time as 

T. triandra, including Austrodanthonia, Austrostipa, Microlaena stipoides and Elymus scabra. 

Biological control 

No succesful biological control programs against grass species have been undertaken and programs targetting grasses are all of 

recent origin (Witt and McConnachie 2004). Grass species were once thought to lack specific arthropod herbivores because they 

very rarely produce toxic secondary metabolites as defenses, and are too simple and similar in structure, physiology and ecology 

for specific herbivores to have evolved (Evans 1991). However secondary metabolites that function as defensive compounds are 

common in many cereals and grasses, including hordenine in barley (McDonald 1991), a specific invertebrate herbivores have 

gradually begun to be discovered. Gross similarities and close taxonomic relationships between grass weeds and valuable crop 

and pasture species was also thought to provide little scope for development of biological control (Witt and McConnachie 2004). 

The potential for biological control of Nassella spp. in Australia was also initially considered to be low due to their close 

relationship to Australian Stipeae (Wapshere 1990 1993). 

An Australian project with Argentine and New Zealand collaborators to biologically control Nassella spp. with fungi has been 

pursued since 1999: see below under “Pathogens”. Anderson et al. (2006) presented a recent update on progress in host 

specificity testing and production of potential agents. One reason biological control remains difficult is because the geographical 

source of Australian populations has not been identified: effective predators and parasites in the area of origin cannot therefore 

be identifed. Another reason for slow progress is the need to demonstrate that the large number of endemic Australian Stipeae 

will not be affected. Biological control using insects may be worth further consideration in the future, given increasing 

recognition of the existence of host specificity amongst invertebrate grass herbivores and overemphasis of the role of plant 

chemical defenses as an evolutionary driver for monophagy (Witt and McConnachie 2004) (see discussion below under 

‘Predators and Pathogens’). 

75

Predators and pathogens 

Evans (1991 p. 60) considered that the natural enemy complexes (invertebrates and fungal diseases) of all the world’s worst 

grassy weeds to be “largely unknown”. This paucity of information extended to grasses in general, in part as the result of failures 

to consistently identify grasses found to be under attack (Evans 1991 p. 53). 

Wapshere (1990) found natural enemy species-specificity to be rare amongst grasses, compared with genus-specificity. He 

considered it likely that Australian grass genera with many species would have large groups of specific or near-specific predators 

and parasites. The corollary of this is that Nassella, with many South American species, should have a large herbivore fauna and 

a a wide variety of diseases. 

Wapshere (1990) determined from published records the main invertebrates and fungi that attack grasses in Europe: Noctuidae 

(cutworms, armyworms, etc., Lepidoptera), microlepidoptera (various families), Cecidomyiidae (gall midges, Diptera), 

Brachycera (flies, Diptera), Aphididae (aphids, Hemiptera), Ustilagines (smuts) and Uredinales (rusts). As an indication of host 

specificity, he also determined the number of species in each group recorded from a single grass species, a single grass genus, 2- 

3 genera and 4 or more genera. Cecidomyiidae and Ustilagines showed the highest levels of species- and genus-specificity 

amongst these groups. High levels of specificity was also apparent with leaf-miners (insects) and particularly with gall makers 

(arthropods, nematodes and fungi). 

Predators 

Very little appears to be known about the animals that attack N. neesiana, a similar situation to that for N. trichotoma, for which 

Wapshere (1990 p. 71) noted an absence of “any readily available knowledge concerning the arthropods … in its home range” 

and (1993 p. 344) no arthropods recorded from it in Australia. Herbaceous monocots have often been viewed as having relatively 

impoverished invertebrate faunas compared to other groups of plants. Their simple architecture and decreased structural 

complexity (hence lower niche diversity) is often cited as the cause (e.g. Lawton and Schroder 1977). However, as Waterhouse 

(1998 p. vi) noted in relation to the paucity of grasses as targets for classical biological control, “it would be surprising if coevolution 

and co-adaptation have never led to effective and highly specific natural enemies of at least some individual species in 

the Poaceae, as it has in members of other plant families”. 

Molluscs 

Published information about slug and snail utilisation of grasses indicates they may often be avoided in preference for other 

plants, but the data is equivocal and it is clear that some species are recognised pests of cereals and grass fodder, and that grasses 

in general are not unpalatable (Barker 2008). Holland et al. (2007) found that Milax gagates Draparnaud (Milacidae) consumed 

two native grasses including T. triandra in laboratory tests and that the palatibility of the grasses was within the palatibility range 

of native dicot species tested. Newly emerged seedlings appear to be at most risk of damage by molluscs. No information 

appears to be available on mollusc attack on N. neesiana. 

Insects 

Evans (1991 p. 52) referred to a seeming “general absence of host specific insects associated with grassy weeds”. He found from 

a literature survey that insects specific to grass species were unknown, and that grass-feeders were generally polyphagous, 

attacking a range of grasses, often including cereals. The relatively uniform structure of Poaceae supposedly “promotes 

polyphagy” (Evans 1991 p. 53) and reduces the evolutionary pressures for monophagy (Briese and Evans 1998). 

Evans’ (1991) generalisation was largely, but not entirely correct, as presaged by Waterhouse (1998) and demonstrated by the 

studies of Wapshere (1990), Witt and McConnachie (2004) and others. Wapshere (1990) found host-specificity at species or 

genus level for grasses in Europe to be particularly common in Elachistidae (Lepidoptera), Chloropidae and Cecidomyiidae 

(Diptera), and Tetramesa (Hymenoptera: Eurytomidae). He noted that all Tetramesa spp. in the USA are relatively 

monophagous, being recorded only from single grass genera. In Australia the genus is represented by three species that are 

phytophagous in grass seeds or internodes, according to Naumann (1991), although Boucek (1988) stated that no Australian host 

records were known. Tetramesa spp. “develop as phytophages feeding on the inner tissues of the internodes or of the seeds and 

places attacked, specific for each species, swell some to some degree, sometimes considerably, so that a characteristic gall is 

formed” (Boucek 1988 p. 95). De Santis and Loiácono de Silva (1981 1983) found that the stem-boring Tetramesa adrianae De 

Santis was the main natural enemy of the stipoid Amelichloa brachychaeta in the provinces of Buenos Aires, La Pampa and 

Entre Rios, Argentina, and discussed programs to breed and disperse it more widely as part of an integrated control program. 

This species causes gall-like stem deformations that affect seed production (De Santis and Loiácono de Silva 1983). Possibly this 

wasp also affects N. neesiana: Gardener et al. (1996b) reported that botanical specimens in the Instituto Darwinion Herbarium 

had galls of undetermined origin on the flowering stems which appeared to prevent flowering. The Palaearctic species T. 

cylindrica (Schlechtendal) and T. punctata Zerova attack the flowers of Stipa capillata L. and S. lessingiana Trin. and Rupr. (De 

Santis and Loiácono de Silva 1983). A. brachychaeta is also attacked by a cecidomyiid, Contarinia sp. (De Santis and Loiácono 

de Silva 1983). 

Some Eurytoma spp. “can complete their development feeding solely on the plant tissues in stems of grasses, but because their 

eggs are laid only in places where a genuine phytophagous eurytomid of the genus Tetramesa is developing (and causing growth 

of plant tissues), they normally devour the larva of Tetramesa before reaching maturity” (Boucek 1988 p. 107). Australian 

Eurytoma are poorly known and no grass hosts were mentioned by Boucek (1988). The Australian eurytomid genus Giraultoma 

Boucek is also associated with grasses, while the biology of some other Australian genera is unknown (Boucek 1988, Systematic 

Entomology Laboratory USDA 2001). 

Because Australia has a wide diversity of Austrostipa species, closely related to Achnatherum, Australian species from these 

insect taxa may be the most likely candidates amongst the host-specific or narrowly-oligophagous species to ‘host shift’ on to N. 

neesiana in Australia. Exchange of insect species (“acquisition of a herbivore guild on an evolutionary timescale”) between 

closely related plants is hypothetically more likely because of closer biochemical and structural similarities between related 

76

species (Lawton and Schroder 1977 p. 138). Lawton and Schroder (1977) found evidencefor this relationship for monocot 

species (not including grasses) but not other plant groups they analysed in Britain. 

No published records have been located of invertebrates feeding on N. neesiana in Australia although a few species are known to 

feed on Nassella trichotoma. A most interesting recent development is the finding by Braby and Dunford (2006) of empty pupal 

cases of the endangered Golden Sun Moth Synemon plana Walker (Lepidoptera: Castniidae) protruding from N. neesiana 

tussocks, from which it was inferred to be a probable larval food plant (see further discussion below). 

Zimmerman (1993) recorded N. trichotoma and “some other grasses” as hostplants of the ground weevil Cubicorhynchus 

sordidus Ferguson (Coleoptera: Curculionidae: Amycterinae: Amycterini), evidently an identification of Howden’s (1986) 

“Cubicorrhynchus sp.” observed near Yass, NSW (see also May 1994), and noted that B.P. Moore had found both larvae and 

adults of the NSW species Phalidura abnormis (Macleay) (Amycterini) feeding on N. trichotoma, the native host plants being 

unknown. Larvae of P. elongata (Macleay) feed on underground parts of N. trichotoma, and other grasses (Zimmerman 1993), 

while adults consume N. trichotoma and pupae are also found in association with it (Howden 1986). Howden’s (1986) listing of 

Phalidura assimilis Ferguson feeding on N. trichotoma near Yass are treated by Zimmerman as P. abnormis and May (1994) 

listed P. abnormis as the only known Phalidura N. trichotoma feeder. Adults and a larva of Cubicorhynchus calcaratus Macleay 

of eastern and southern Australia have been found in a clump of “Stipa” in South Australia, while Austrostipa nitida and A. 

nodosa along with other grasses are hosts of another eastern and southern species C. taurus Blackburn, with larvae found in the 

crowns and root masses (Howden 1986, Zimmerman 1993, May 1994). Other species in the genus also have grass hosts 

including Microlaena stipoides and “Stipa” for the Western Australian C. bohemani (Boheman) (Zimmerman 1993) and 

unidentifed grass for C. crenicollis (Waterhouse) (May 1994). Howden (1986 p. 100) noted that all Cubicorhynchus species 

“collected to date have been associated with either native or introduced species of Poaceae”. Larvae “collected from the crowns 

of grass plants often regurgitated green material, indicating that they fed on underground stems and not the roots” (Howden 1986 

p. 100). Sclerorinus spp. (Amycterini) have been recorded from undetermined “Stipa sp.” (May 1994 p. 495). Amycterini are 

flightless ground-dwellers and most are confined to grasses, or other monocots, most larvae living underground in the root 

crowns and the adults eating leaves, evidently including, unusually, dry grass (Zimmerman 1993). Larvae are free living in the 

soil and eggs are deposited directly into the substrate, rather than a prepared site (Howden 1986, May 1994). Adults of the 

species that feed on wiry stems have stout, blunt mandibles and gular roll (‘lip’), while species that feed on soft tissues have a 

different mouthpart morphology (Howden 1986). Themeda triandra and Austrodanthonia are not known hosts (not mentioned in 

Zimmerman 1993 or May 1994). Zimmerman (1993) considered the tribe to be Gondwanaland relics with no known closely 

related group in South America. “Over vast areas of the country the Amycterinae have been nearly exterminated by the clearing 

of vegetation, cultivation and the grazing of livestock, especially by the destructive activities of millions of sheep” (Zimmerman 

1993 p. 176). 

Gardener et al. (1996a) suggested that N. neesiana seed predation by ants appeared to be lacking, possibly due to the 

impenetrability of the lemma providing good protection to the edible caryopsis. However Gardener (1998) set up experiments in 

pasture in which de-awned seeds were placed on the soil surface or buried in soil at 1.5 cm depth and exposed for six weeks from 

late January to early March, or offerred, along with de-awned Themeda triandra seed, in dishes arranged to prevent access by 

larger animals for 4 weeks in May. In the first experiment 99.5% of buried N. neesiana seeds were recovered but significantly 

fewer (71%) of the unburied seeds. In the second experiment 97.5% of the seeds of both species were recovered (“no seeds ... 

taken” Gardener 1998 p. 53). The experimental results were considered inconclusive: seeds in the first experiment may have 

been removed by other organisms, ants may have been inactive in the second experiment or not interested in the seeds. No 

identification was suggested for any ant species that may have been responsible. 

Absence of ant predation appears to be unlikely. On a world basis, Formicidae are amongst the most important granivores in 

desert ecosystems (Saba and Toyos 2003). Ants are the dominant seed predators in the Sonoita Plains grassland of Arizona, 

where they selectively remove seeds with awns, projections or significant pubescence (Pulliam and Brand 1975). The other main 

seed predators, sparrows and rodents, consume smaller amounts, selectively favour ‘smooth’ seeds, and have different seasonal 

foraging patterns, so have little dietary overlap with the ants. In the Monte Desert of northern Patagonia, Argentina, where the 

vegetation is a steppe dominated by Larrea divaricata Cav. and stipoid grasses, ants are the most important granivores in spring 

and summer, but remove little seed in other seasons (Saba and Toyos 2003). Ants are the dominant granivores in arid areas of 

Australia (Morton 1985), although the dominant grasses in these areas rarely include stipoids. The Australian fauna of seedharvesting 

ants is rich and the species are capable of causing severe seed losses: up to 90% of weed seeds in crops can be 

removed (Vitou et al. 2004). Ants are the dominant weed seed predators in agricultural landscapes in Western Australia and have 

preferences for particular weed species (Minkey and Spafford Jacob 2004). The few studies of ants in south-eastern Australian 

native grasslands show that seed-harvesters are generally present (Coulson 1990 appendix 3, Miller and New 1997), one of them, 

Pheidole sp., being amongst the most abundant ants in Victorian Basalt Plains grasslands (Yen et al. 1994a). The seeds of N. 

neesiana are very similar to those of some Austrostipa spp., so it appears highly likely that they would be harvested by ants as a 

matter of course. Ant seed herbivory has been found to be higher for African grasses in Brazil compared to native savannah 

species (Klink 1996), so in Australia N. neesiana seeds may even by preferred. 

Apart from ants, Gryllidae and granivorous carabid ground beetles are the most important post-dispersal seed predators in 

temperate agro-ecosystems (Lundgren and Rosentrater 2007) and no doubt the seeds of N. neesiana must be destroyed by speices 

in these groups, although only one example seems to be on record. Slay (2001 p. 33) recorded that “field crickets” consumed 

shed N. neesiana seeds in a New Zealand pasture. The seeds were “hollowed out” and the insect “might well be implicated in 

reducing the numbers of recently shed seeds”. This is almost certainly the Black Field Cricket, Teleogryllus commodus (Walker) 

(Gryllidae), a recognised pasture pest in New Zealand (Heath 1968) and Australia, and a common native insect in south-eastern 

Australian grasslands. 

A wide range of non-specific grass feeders including Orthoptera, Thysanoptera (thrips), Hemiptera (true bugs) and Noctuidae 

(Lepidoptera) are likely to be found to utilise N. neeiana in Australia (see the Appendix to this Literature Review). 

77

Vertebrates 

Many stipoid grasses are readily eaten by livestock, including N. neesiana. In the Great Basin region of the USA Stipa spp. sens. 

lat. are considered “for the most part valuable forage plants” (Hitchcock and Chase 1971 p. 445). N. neesiana is readily eaten by 

sheep, cattle and horses in Argentine pastures in winter and spring, but consumption is much reduced when the plant is flowering 

and seeding, except under drought conditions or when stocking rates are high (Gardener et al. 1996b, Gardener 1998). It has 

been rated as one of the most valuable winter pasture species in the pampas (Gardener et al. 1996b). 

Grasses defend themselves against grazing mammals in a variety of ways, including adaptations of form, habit and phenology, 

presence of indigestible structural compounds and presence of toxic chemicals. Very few grass species contain toxic secondary 

metabolites that deter grazing and herbivory: less than 0.2% contain alkaloids, and the presence of noxious terpenoids and 

cyanogenetic compounds is rare (Tscharntke and Greiler 1995, Witt and McConnachie 2004), although McDonald (1991) 

claimed that defensive metabolites are common in cereal and other grasses. Records of toxic effects of stipoids on livestock are 

scarce. According to Quattrocchi (2006 p. 1361) “Some” Nassella species “have caused poisoning to mammals”. Randall (2002) 

stated that N. neesiana has been recorded as toxic, and the US Food and Drug Admimistration Poisonous Plants Database 

(USFDA 2006) lists N. neesiana, citing Kellerman et al. (1988). 

Probably most cases of toxicity attributed to grasses are the result of grass fungi. Possibly the best known example is ergot, 

Claviceps purpurea, which infects cultivated Secale cereale L., other cereals, and other grasses, the toxic effects of which have 

resulted in mass human poisonings (Gair et al. 1983, Wink and Van Wyk 2008). However none of the many Claviceps species 

affect Stipeae (Wink and Van Wyk 2008). The Mexican and southwestern USA Achnatherum robustum (Vasey) Barkworth (= 

Stipa vaseyi Scribn. = Stipa robusta (Vasey) Scribn.), known as Sleepy Grass, allegedly acts as a “narcotic”, especially on horses 

(Hitchcock and Chase 1971) due to infection with endophytic Acremonium fungi. The active principles were identified by 

Petroski et al. (1992), the dominant one, lysergic acid amide, likely being responsible for a reportedly extreme “sedative” effect 

on animals. However lysergic acid and its derivatives are generally considered to be hallucingens, producing delerium but not 

sedation (Wink and Van Wyk 2008). Achnatherum inebrians (Hance) Keng of China and Mongolia, known as Drunken Horse 

Grass, probably affects animals by ergot alkaloids produced by the endophytic Neotyphodium gansuense Li and Nan 

(Clavicipitaceae) (Moon et al. 2007). The ergot alkaloids consist of two main series, clavine alkaloids and the lysergic acid 

amides (Wink and Van Wyk 2008). Such fungal endophytes occur widely in grasses (see section on “Other biotic relationships” 

below) and produce a variety of poisonous secondary metabolites that effectively deter pathogen attack and grazing, and 

sometimes poison mammals (Tscharntke and Greiler 1995, Jallow et al. 2008), and are thus generally considered to be 

symbionts, rather than pathogens (Wink and Van Wyk 2008). 

Grass poisoning can also be caused by organic acids, which irritate skin and mucous membranes (Wink and Van Wyk 2008). 

Stipa capensis contains glycosides producing a strong acid that can harm cattle (Tsvelev 1984). 

Specialist granivorous birds, of which the most important are finches, parrots and pigeons, probably commonly consume stipoid 

seeds, however there appear to be few relevant records. Twigg et al. (2009) found from anyalysis of stomach contents that seeds 

in the diets of Australian finches (two Emblema spp.) and Alcedinidae (three dove and pigeon spp.) were predominantly

y increased importance of ant granivory. Specific information on mammal granivory in areas where N. neesiana is native 

appears to be lacking and no information is availabe about rodent granivory in Australian native grasslands. 

Pathogens 

A wide range of pathogenic fungi attack grasses. These are often highly host-specific, notably head smuts and rusts, with 

specificity often being confined to particular host biotypes (Witt and McConnachie 2004). Wapshere (1990) mentions, in 

addition, the genera Phyllochora (Ascomycetes), Cercospora (Hyphomycetes) and Stagonospora (Coelomycetes) as having, on 

a world basis, a high proportion of their species known from only a single grass genus. 

N. neesiana has a rich pathogenic fungal flora in South America (Briese et al. 2000, Anderson et al. 2004), including the 

following taxa: 

Powdery mildew (Anderson 2002b) 

Septoria leaf spot (Anderson 2002b) 

Teliomycetes (Rusts) 

Puccinia digna Arth. and Holw. (Greene and Cummins 1958) 

Puccinia graminella Diet. and Holw.- damaging infestations in central Argentina (Briese et al. 2000) 

Puccinia nassellae Arth. and Holw. var. platensis Lindquist (Briese and Evans 1998, Briese et al. 2000, Anderson et al. 2004 

2008 ) 

Puccinia saltensis var. saltensis (Briese et al. 2000) 

Puccinia aff. avocensis (Anderson et al. 2002) 

Uromyces pencanus Arth. and Holw. (Greene and Cummins 1958, Anderson 2002a, Anderson et al. 2006 2008) 

Uredo sp. (Briese et al. 2000) 

Ustomycetes (Smuts) 

Tranzscheliella hypodytes (Schltdl.) Vánky & McKenzie (= Ustilago hypodytes (Schlecht.) Fr., sensu lato (Briese et al. 2000, 

Anderson 2002a, Anderson et al. 2004), listed as Ustilago sp. by Anderson et al. (2002). 

However some populations in Argentina are free of pathogens including Entre Rios Province where in one survey “many huge 

and dense populations ... were completely devoid of disease” (Anderson 2002b). Knowledge of these pathogens has been greatly 

enhanced by studies undertaken for an Australian biological control program for Nassella spp., initiated in 1999 (Anderson et al. 

2008). 

Puccinia graminella is known from western and southern South America and California, and also attacks N. hyalina (Greene and 

Cummins 1958). It was found to be damaging N. neesiana populations in central Argentina by Briese et al. (2000). During one 

survey P. graminella was found at 12 of 14 sites and was killing N. neesiana leaves at 4 sites (Anderson et al. 2002). It was 

initially thought to cause little damage to N. neesiana (Anderson 2002b) but can cause mortality under wet conditions (Anderson 

et al. 2004) and severe damage in the field (Anderson et al. 2008). It appears to have restricted host-specificity, is autoecious and 

completes its lifecyle on N. neesiana, with both uredina and telia having been recorded on the plant (Anderson et al. 2004 2006). 

However N. neesiana appears to have qualitative resistance to infection and a pure culture has not been established (Anderson et 

al. 2008). 

Puccinia nassellae is known from Argentina, Bolivia and Chile (Greene and Cummins 1958). Two strains have been 

investigated as potential biological control agents, specific to N. neesiana and N. trichotoma. The strain on N. neesiana 

commonly forms telia and produces pustules that are plain to see on open leaf lamina. It is virulent (Anderson et al. 2008), easy 

to rear, probably host-specific, hemicyclic (urediniospores and teliospores on N. neesiana, but possible other stages unknown), 

but probably not autoecious (Anderson 2002a) with Clematis montevidensis, Solidago chilensis, Cestrum parqui, Verbesina sp. 

and Morrenia sp. found with aecial rust infections, although none appear to be solid candidates as alternate hosts (Anderson 

2002b). It has not infected Austrostipa spp. and Stipa spp. that have been tested, but of the strains collected in the wild, only one 

has been able to infect 3 of 6 Australian N. neesiana accessions, so there is high specificity at the subspecific level (Anderson 

2002a, Anderson et al. 2002 2006). Trap plants of N. neesiana obtained in the Australian Capital Territory became infected with 

strain NT27 in Argentina (Anderson 2002a). 

Puccinia digna is known from Bolivia and Chile (Greene and Cummins 1958) and has not been studied for biological control 

purposes. 

Uromyces pencanus is known from Argentina and Chile (Greene and Cummins 1958), appears to be host-specific and can be 

very damaging to N. neesiana in Argentina, but its lifecyle on the plant is incompletely understood (Anderson et al. 2006 2008). 

It is easy to rear and has infected 5 of 6 Australian accessions of N. neesiana. Isolate Up27, virulent against most Australian 

accessions of N. neesiana, has been found not to infect N. hyalina, Austrostipa aristiglumis (F. Muell.) S.W.L. Jacobs & J. 

Everett and a range of economically important agricultural grasses (Anderson et al. 2008). 

Mixed species rust infections are not uncommon in Argentina and appear to be particularly damaging to the plant (Anderson et 

al. 2006). 

The smut Tranzscheliella hypodytes is a cosmopolitan species (Vánky and Shivas 2008) and infects a number of Austrostipa 

species (Briese and Evans 1998) in south-eastern Australia and Dichelachne crinita (L.f.) Hook. in NSW and Victoria, but does 

not appear to have been recorded on Australian N. neesiana (Vánky and Shivas 2008). It infests upper culm internodes of N. 

neesiana in Argentina, preventing most seed production when plants are severely attacked, and infection occurs at germination 

(Anderson et al. 2002). It is not known if strains found on N. neesiana can infect N. trichotoma and vice versa (Anderson 2002a). 

Conditions for infection appear to be uncommon in nature (Anderson et al. 2004) and the species has been found only at very 

low incidence in Argentina (Anderson 2002b). 

Knowledge of the fungal flora of N. neesiana in its introduced ranges is limited. Slay (2002a) summarised results of a Landcare 

Research survey of N. neesiana fungi at five sites in New Zealand. Fourteen species were identified, of which six were 

79

potentially pathogenic: Alternaria sp., Colletotrichum sp., Drechlera spp., Phoma leveilli Boerma and Bollen and Phoma sp. 

Britt (2001) conducted a fungal pathogen survey at eight sites in Victoria and found four possible species of Alternaria, and one 

each of Aspergillus, Fusarium and Epicoccum on leaves, none of which was precisely identified. Leaf discoloration, spotting and 

necrosis was common in the field. Cultures on agar plates were distinguished by colour and form 

(downy/woolly/cottony/powdery). Testing showed that the fungi aided seed germination of N. neesiana but only Epicoccum sp. 

had a statistically significant effect, probably due to the small sample sizes. No seeds germinated in the absence of fungi. The 

effect could be due to fungal production of plant hormones, such as gibberellin, that break dormancy or stimulate germination, or 

to fungal digestion of the lemma. Growth substrates other than N. neesiana seed may be preferred by the fungi,so the effect may 

not occur under field conditions. 

As of 1998, none of the 27 species of fungi recorded from Austrostipa in Australia were recorded on Nassella spp. in Australia, 

nor do South American and Australian Stipeae have any rusts in common (Briese and Evans 1998). 

Greene and Cummins (1958) recorded a single rust species on Austrostipa, Puccinia flavescens McAlp., with two known hosts, 

A. flavescens (Labill.) S.W.L. Jacobs and J. Everett, A. semibarbata (R.Br.) S.W.L. Jacobs and J. Everett. Vánky and Shivas 

(2008) recorded three smuts from Austrostipa in south-eastern Australia in addition to T. hypodytes: Fulvisporium restifaciens 

(D.E. Shaw) Vànky, Tranzscheliella minima (Arthur) Vànky and Urocystis stipeae McAlpine. The latter also occurs on 

Achnatherum spp. in south and east Asia (Vánky and Shivas 2008). 

Other biotic relationships 

Fungal and bacterial symbionts of Nassella spp. appear to be poorly known, apart from the observations of Britt (2001) (see 

“Pathogens” section, above). Higher plants in general have fungal endophytes that live within their tissues without causing 

damage, and plant roots are always inhabited by mutualistic fungi, usually classed as arbuscular mycorrhizal fungi, 

ectomycorrhizae or dark septate fungi (Khidir et al. 2009). Presence of these organisms can greatly affect the invasiveness of a 

plant and influence its ability to modify soil properties (Rout and Chrzanowski 2009). 

Grasses harbour large and diverse communities of root-associated fungi, including arbuscular mycorrhizal fungi (AMF), the 

colonisation of which appears to be strongly effected by climatic conditions and nutrient availability, and dark septate fungi, 

which are the main colonisers of grasses in semiarid environments (Khidir et al. 2009). No mycorrhizal relationships have been 

recorded for N. neesiana but vesicular arbuscular mycorrhiza have been reported for N. leucotricha and other stipoid grasses 

(Clark and Fisher 1986, Wang and Qiu 2006) and a dark septate fungus of the genus Paraphaeospheria sp. has been recorded 

from Achnatherum hymenoides (Roemer & J.A. Schultes.) Barkworth (Khidir et al. 2009). 80% of surveyed land plant species 

are mycorrhizal (Wang and Qiu 1996) so the possibility that N. neesiana lacks these root fungi seems remote. 

Root associated fungi appear to be generalist species that inhabit a range of species in a community and across large areas 

(Khidir et al. 2009). Khidir et al. (2009) found that co-occuring grasses had a common flora of non-AMF groups from the 

Pleosporales, Agaricales and Sordariales, with Paraphaeospheria spp. (Pleosporales), Moniliophthora spp. (Agaricales) and 

Fusarium spp. (Hypocreales) most common, but AMF fungi (Glomus sp.) were also present. The dark septate fungi may enable 

the plant host to access nutrients, including N and P, and may play a role in drought and heat tolerance (Khidir et al. 2009). 

N fixation by free-living bacteria associated with grass roots has been recorded (Clark and Fisher 1986) and several African 

grasses are known to fix significant levels of N in their native habitats (Rossiter et al. 2003). Rout and Chrzanowski (2009) 

found that Sorghum halepense harboured a range of bacteria in its rhizomes known to fix N, and almost certainly able also to 

chelate iron and mobilise phosphorus. 

Species of Neotyphodium Glenn and their teleomorphic relatives Epichloë Tul. (Balansieae, Clavicipitaceae) are endosymbiotic 

non-pathogenic fungi found in an estimated 20-30% of graminoid species, mainly in grasses of the subfamily Pooidae (Moon et 

al. 2007, Rudgers et al. 2009). They are closely related to the ergot fungi, Claviceps spp., and are commonly called ‘grass 

endophytes’ (Aldous et al. 1999). Each sexual species of Epichloë is associated with a particular grass tribe in North America or 

Europe, but the many asexual species, transmitted via seeds, are hybrids resulting from crosses between Epichloë species or 

between Epichloë and Neotyphodium species, that bear no such direct relationship with their range of hosts and may have 

‘jumped’ between tribes (Moon et al. 2007). Endophytes transmitted vertically (inherited) are more likely to evolve to be 

symbiotic than those that spread vertically by contagious spread, which are more likely to retain pathogenicity (Rudgers et al. 

2009). Some grass endophyte species, including some affecting Achnatherum species, cause poisoning in livestock, and many 

produce chemicals that deter attack by insects (Moon et al. 2007). Tests with the mollusc Deroceras reticulatum Muller indicate 

that the different secondary metabolites produced by Neotyphidium can deter or enhance feeding (Barker 2008). Other 

documented benefits from endophyte infection include drought tolerance, increased vigour and higher nutrient content. 

Within Stipeae, species of Neotyphodium/Epichloë are known to infect only Achnatherum species (Moon et al. 2007, Rudgers et 

al. 2009) except for an unknown species infecting Nassella viridula. Those currently known are Neotyphodium chisosum (White 

and Morgan-Jones) Glenn et al. and probable Epichloë amarillans from A. sibiricum (L.) Keng (Wei et al. 2007, Ren et al. 2008, 

Moon et al. 2007), N. chisosum from A. eminens (Cav.) Barkworth, N. gansuense Li and Nan and its morphologically and 

geographically distinct variety inebrians C.D. Moon and C.L. Schardl from A. inebrians (Hance) Keng, N. funkii K.D. Craven 

and C.L. Schardl from A. robustum and an undescribed species from A. sibiricum (Moon et al. 2007). Undetermined species are 

also known from A. lobatum (Swallen) Barkworth, A. purpurascens (Hitchcock) Keng, A. splendens (Trinius) Nevski and A. 

viridula (Rudgers et al. 2009). The last has been classified as Nassella viridula by Barkworth (2006) who suggested it may be an 

alloploid between Nassella and Achnatherum. 

No evidence appears to be available on the presence or absence of endophytes in N. neesiana. But given their importance to 

plant fitness, endophyte presence should be investigated. In the USA endophyte infection of the non-native Festuca arundinacea 

Schreb. increases its invasiveness and impact on biodiversity (Rudgers et al. 2009). 

80

BIODIVERSITY 

“Biodiversity ... one of the best descriptors of ecosystem condition” 

(Aguiar 2005 p. 262) 

This section explores the concept of biodiversity and aspects of its assessment, discusses the range of impacts that invasive 

plants, and grasses in particular can have on biodiversity, and evaluates existing knowledge of the impact of N. neesiana on 

biodiversity in temperate Australian grasslands. 

Definitions 

The United Nations Convention on Biodiversity concluded at Rio de Janeiro on 5 June 1992 defines biological diversity as “the 

variability among living organisms from all sources including, inter alia, ... ecosystems and the ecological complexes of which 

they are a part: this includes diversity within species, between species and of ecosystems.” Biodiversity has elsewhere been 

defined as “the number, variety and variability of living organisms at genetic, population, species, community and ecosystem 

levels” (Giles 1994). It is present at every heirarchical level within the purview of biology (Mayr 1982): molecular, genetic, 

chromosome, organelle, cellular, tissue, organ, organism, taxon, association, etc. At each level, diversity varies spatially and 

temporally, in historical origin, functional role and evolutionary significance (Mayr 1982). In broader terms, biodiversity 

encompasses not just the biological taxa, but the processes and functions in which the organisms participate (Saunders 2000). It 

therefore includes the range of interactions organisms have with one another and the physical environment, and the associations 

they form, including mutualisms, competitive relationships, guilds, functional groups, successional dynamics and patterns, 

trophic relationships and foodwebs (Woods 1997, Saunders 2000). Indeed, “there is hardly any biological process or 

phenomenon where diversity is not involved” (Mayr 1982 p. 133), so understanding the impact of an invasive species on 

biodiversity requires understanding this broader context. Since all individuals differ in their history and precise chemical 

makeup, and in sexually reproducing populations in many other ways, every individual of every population is a unique part of 

biodiversity (Mayr 1982). 

Biodiversity enables ecosystem services, provides direct economic benefits and creates the distinctive milieu in which human 

cultures flourish (Saunders 2000, Mansergh et al. 2006b). The concept of ecosystems services provides a framework for the 

economic quantification of chemical and biological reserves and cycles in areas including soil stabilisation and fertility, water 

quality and quantity, biological production, etc. (Mansergh et al. 2006b). Biodiversity can also create ecosystem “dis-services”, 

including, in general, exotic invasive species (Mansergh et al. 2006b p. 300). However many processes alter ecosystem 

functioning, not just alterations to biodiversity, and its contribution to ecosystem services has not been adequately resolved 

(Aguiar 2005). Less diverse anthropogenic systems may in some circumstances provide similar levels of service to those 

provided by diverse natural ecosystems. 

Contracting parties to the Convention on Biodiversity, which include Australia, are required to identify components of 

biodiversity important for conservation and sustainable use, monitor them through sampling and other techniques, identify 

processes and categories of activities that have or are likely to have significant adverse impacts on the conservation and 

sustainable use of biological diversity and monitor their effects. Parties are also required to prevent the introduction of, control or 

eradicate those alien species which threaten ecosystems, habitats or species (United Nations 1992). 

Quantification and indices 

Full quantification of biodiversity requires enumeration of diversity at each heirarchical level, both in terms of taxonomy (classes 

to subspecies) and levels of biological organisation (molecular to ecosystem). Understanding of the processes and mechanisms 

that alter or maintain biodiversity requires study and measurement of the interactions on and between each level (Giles 1994). 

Numerous authors caution against the over-reliance on species richness as an index of biodiversity (e.g. Aguiar 2005), but 

biodiversity assessment must start somewhere, so there has been long historical emphasis on species, and their intrinsic worth, 

novelty, or ‘uniqueness’. The dependence of eccological processes on biodiversity is a more recent concern, that has come to be 

included under the moniker of “sustainability”. Fuller capturing of these multidimensional attributes of biodiversity requires a 

series of indicators (Aguiar 2005). Biodiversity attributes for which such indicators exist or can be measured include composition 

(the identity and variety of the component elements at each heirarchial level from gene to landscape), structure (physical, 

chemical, biological and geographical organisation of these elements) and function (ecological and evolutionary processes that 

organise the systems) (Aguiar 2005). Consideration of the biodiversity attributes within such a framework enables a much fuller 

appreciation and understanding of system functioning. 

As a general rule, the number of species present increases as the total area under consideration is increased (Londsdale 1999). 

Assessment of biodiversity must therefore standardise for spatial scale. Exotic fraction, the ratio of exotic species to native 

species, has been widely used as an indicator of invasibility, but does not control for scale (Lonsdale 1999). 

The most basic measure of diversity is species richness, a simple count of the number of species present. Species richness indices 

can be compiled from species richness data. All more complex diversity measures rely on determining the number of individuals 

of each species. When population numbers are known the heterogeneity of the community can be determined, and a community 

with only two species that are equally abundant is supposedly more heterogeneous than when one species is more abundant than 

the other (Krebs 1985). 

Species abundance models often indicate that there are few common species and many rare ones, i.e. the relationship between 

the abundance of individuals of the species in a sample and the number of species in a sample is logarithmically related. This 

relationship results in the alpha diversity index, which gives an indication of diversity that is independent of sample size: 

81

S = α log e (1+N/α) 

where S = the number of species in the sample and N = the number of individuals. 

However the logarithmic model is applicable only to a limited set of communities with few species. A larger number of 

communities are better represented by a log normal distribution, which is the generally expected statistical distribution (Krebs 

1985). Determining the total number of species present in a community takes a large amount of sampling effort. But if such a 

distribution is assumed, it is possible to predict the total number of species present without having to sample intensively to detect 

the rare species. 

The Shannon-Wiener index is based on information theory and incorporates the concept of evenness in the population size of the 

species. The more even the numbers of the species the higher the diversity. Other indices include Simpson’s index and 

Margalef’s index (Krebs 1985). 

The usefulness of various indices is debatable (Adair and Groves 1998). They often rely on assumptions that remain untested and 

provide few benefits in terms of enabling ready interpretation of comparitive data. In practice the basic information on the 

number of species in a sample and abundance of each summarises most of the diversity information (Krebs 1985). Indices 

assume fixed relationships between numbers of individuals and numbers of species, but the number of insect individuals is so 

temporally and spatially variable that the indices may only become meaningful after prohibitively extended sampling efforts 

(Farrow 1999). The same arguments apply to plant sampling, where the number of individuals may be prohibitively large when 

the species is of small size, or the determination of what constitutes a single individual may often be difficult. Farrow (1999) 

therefore argued that simply enumerating the species present by extending sampling over a longer period was a superior 

approach for grassland invertebrate biodiversity assessment because of the major effort involved in counting what may often be 

superabundant organisms, and that simple presence/absence assessments are appropriate in some circumstances. 

Diversity indices also tend to perpetuate the emphasis on species richness, which they all require for their computation, and 

divert attention from the structural and functional attributes of biodiversity that may be more important and valuable, and from 

the compositional attributes at other heirarchical levels – genetic, association, community, etc. 

Diversity studies also require that important and unimportant species be identified, that exotic species are distinguished from 

natives (Greenslade 1994), pest from beneficials, etc., i.e. the identification and appropriate weighting of desirable and 

undesirable elements (Driscoll 1994). Proper contextualisation also requires precise identification of all taxa. 

Assessments of overall biodiversity of a diverse ecosystem or community is always difficult, and beset with temporal and spatial 

scaling problems. Simple species richness assessments based on very few higher taxa (e.g. mammals or vascular plants) are of 

little value because no high level taxon appears to adequately indicate biodiversity in any other high level taxon (Melbourne 

1993). Furthermore, the biodiveristy significance of a particular area can only be adequately assessed in the context of other 

similar areas and overall regional biodiversity (Melbourne 1993). 

Impacts of weeds on biodiversity 

General considerations 

An impact is “a disruption to a particular set of ecosystem services or functions” (Mathison 2004), or more simply, any change in 

the diversity or abundance of one organism that is caused by another. Impacts of invasive species may be predicted by introvert 

and extrovert measures (Williamson 2001). Introvert assessments are made from study of the invasive organism, its range, 

abundance and ecology, to predict likely effects. Gardener and Sindel (1998) predicted impact on Button Wrinklewort Rutidosis 

leptorrynchoides F. Muell. and Kirkpatrick et al. (1995) on Sunshine Diuris Diuris fragrantissima D.L. Jones and M.A. 

Clements using this approach (Ens 2002a). Extrovert measures involve direct quantification of impact on affected organisms, 

processes or communities. 

In the trivial sense, any invasive species initially increases the biodiversity of the area it invades. Many invasive plants “integrate 

smoothly” (Woods 1997) into the invaded ecosystem and are recognised as having minimal impact (Kirkpatrick et al. 1995, 

Woods 1997, Grice 2006). Most Australian temperate grasslands have large inventories of alien vascular plant species and all 

areas have at least some exotics (Kirkpatrick et al. 1995). Invasion by multiple weed species, together or sucessively, is usual, 

particularly in southern Australia (Adair 1995). A relatively small number of native plant species have largely disappeared, but 

on the broad scale the overall vascular plant species diversity is much higher than before European occupation. ’Xenodiversity’ 

is the richness of a community in exotic species and of new communities dominated by, or assembled from alien species (Cox 

2004). Xenodiversity of plants in general increases total species biodiversity, and on a world basis, the rate of new aliens 

entering communities much exceeds the rate of extinction of native species. A central problem is that similar sets of alien species 

are entering all the world’s biogeographical regions, so the world flora is being homogenised (Cox 2004). Invasions are “blurring 

the regional distinctiveness of Earth’s biota” (Vitousek et al. 1997 p. 6) and grasslands everywhere are being invaded by similar 

sets of species. 

The minimum impact of an exotic plant that integrates smoothly into a native community might possibly be very small: 

conceivably it might use resources that would not otherwise be used by the native plants and space that they would not occupy. 

In general however resource must be used and displacement is usual. Larger impacts involve displacement of more species over 

wider areas. Major impact may involve preemption of the niche of a community dominant. The highest levels of impact involve 

alteration to the community properties – the invader is a so-called ‘transformer species’ (Henderson 2001). Typically these 

dominate by forming a high proportion of the biomass in the community or stratum or have disproportionate influences on 

ecosystem function (Grice 2006). In order to measure the impact of an invasive species on biodiversity it is necessary to examine 

the effects on native biota and ecosystem functioning, determine any threshold below which impact is minimal and determine the 

management factors that influence the degree of impact (Adair and Groves 1998). The interactions between the invader and the 

invaded system are complex, and of many types, and are often indirect (Groves 2002). 

82

According to Woods (1997 p. 61) “there have been few cases where competition from invaders has been shown unambiguously 

to be responsible for significant alteration of communities. Most of the extensive literature suggesting such effects is based on 

correlative studies, historical records, or anecdotes”. Byers et al. (2002) considered that much of the research purportedly 

demonstrating detrimental impacts fails to clearly demonstrate that the invasive organism is the cause of supposed effects. The 

presence of weeds where native plants once grew may be due to their ability to invade without disturbance, or be a consequence 

of damage to the native species by disturbance. Correlations between weed density and reduction in cover and abundance of a 

native plant implies a direct negative interaction. However the affected species could be reacting in an opposite way to some 

independent environmental factor such as an altered disturbance regime resulting from human activity. It is generally difficult to 

determine if the invading species or the altered conditions are the cause of such changes (Weiss and Noble 1984, Huenneke et al. 

1990, Woods 1997). If anthopogenic disturbance is the cause, management should address the disturbance, rather than the weed. 

Despite such difficulties there is wide consensus that “introduced alien species are the most rapidly growing cause of extinction 

and extirpation of endemic, native species” worldwide (Cox 2004 p. 220) and currently “a significant cause of global 

biodiversity decline” (Downey and Coutts-Smith 2006 p. 803). Environmental weeds apparently cause fragmentation of habitat, 

disintegration of plant communities and extinctions, but the details of how this occurs and what is impacted have been scanty 

(Adair 1995) and little quantitative information of effects on native species and ecosystem functioning has been published (Byers 

et al. 2002). The severity of impact generally increases with the extent of weed cover (Carter et al. 2003). 

In Australia, even the simplest data on the proportion of the landscape or habitat invaded and the relationship of weed density to 

impact has been lacking for most weeds (Adair 1995). “Remarkably few” studies have attempted to quantify the impact of weeds 

on biodiversity, in part because the effects are viewed as “obvious” (Adair and Groves 1998 p. 3). Groves (2002 p. 18) 

considered it “surprising” that the impacts of some of Australia’s worst invasive plants on species richness was “still unknown” 

and noted only “few well-documented” studies of biodiversity impact. Most published statements about impact “are based on 

more or less casual observations” (Grice 2006 p. 28) and this reliance on anecdotal and subjective information is a worldwide 

problem (Byers et al. 2002). 

Overall in Australia the mechanisms by which weeds impact on ecosystem structure and function - “how” weeds affect 

biodiversity - have not been widely quantified (Grice 2004a, Grice et al. 2004), although a few studies have examined land use 

and environmental factors associated with invasion (Adair and Groves 1998). There is little data specifically dealing with the 

effects of weeds on the biodiversity of Australian rangelands (Grice 2004a 2006). Most studies have focused on vascular plants 

and some on vertebrates (Grice 2006) but there are much fewer studies of the effects on animals than on plants (Grice 2004a 

2006) and very few on invertebrates and soil biota (Grice 2006). However most fauna studies indicate marked reductions in 

diversity and abundance of vertebrates and invertebrates (Adair and Groves 1998). 

In general, information has not been readily available on the species and communities acually impacted in Australia, nor has 

there been an adequate compilation of the weeds causing the impacts (Downey and Cherry 2005, Downey and Coutts-Smith 

2006, Downey 2008). As with the worldwide information ( Byers et al. 2002), causal relationships between invasion and impact 

have “generally [been] implied but not demonstrated” (Grice 2004a p. 54). For example, Chejara et al. (2006) claimed that 

Hyparrhenia hirta “greatly reduced the species richness of native flora”, although they only compared sites where the grass was 

present or absent, and undertook no manipulative experimentation. McArdle et al. (2004 p.50), studied the same species using 

“matched plots” with “similar ... apparent disturbance history”, and reached a similar conclusion: “demonstrated impacts on plant 

diversity” (p. 54), despite little attempt to determine the mechanisms that enabled the invasion or its history at invaded study site. 

Impact on threatened species and communities 

Leigh and Briggs (1992) compiled data on the proportion of threatened plants in Australia that were threatened by “weed 

competition” and found that alien plants had been dentified as a threat to 69 species. Adair and Groves (1998) found that weeds 

had been cited as a major cause of extinction of four plant species and an endangering process for 57 nationally endangered plant 

species. They also found that 23 entities listed under the Victorian Flora and Fauna Guarantee Act 1998 were at risk from 22 

exotic weed species. Groves and Willis (1999) found that environmental weeds have been implicated in the extinction of four 

native species. 

In terms of detailed data on the impact of weeds on threatened species or communities in Australia 19 of the 20 studies of 

environmental weed impact on plant communities in Australia examined by Adair and Groves (1998 p.7) “demonstrated a 

decline in either species richness, canopy cover or frequency of native species”. A brief review by Vidler (2004 p. 652) found a 

general absence of quantitative information, but concluded that weeds are “a major threat to at least 41 threatened plant and 

animal species”. A much more thorough review (Coutts-Smith and Downey, 2006, updated by Downey and Coutts-Smith 2006) 

found negative impacts on 283 plant species (including algae and fungi), 63 animal species, 15 threatened populations and 71 

endangered ecological communities (90% of all officially recognised endangered communities) in NSW alone. Weeds threatened 

45% (or 44% according to Downey 2008) of the 970 entities listed under the NSW Threatened Species Conservation Act 1995. 

127 weed species were on record as biodiversity threats. Unspecified weed species accounted for 51% of the threats and 43% of 

threats comprised multiple weed species. Of the major animal groupings, invertebrates had the highest proportion of listed 

‘threatened entitites’ at risk from weed invasions. Competition (as opposed to habitat degradation by weeds and weed control 

activity) was determined to be the main threatening factor, and accounted for 81% of the threats. The number of native species at 

risk from alien plants in one State alone was found to be an order of magnitude larger than previous estimates for the whole of 

Australia (Downey 2008). Furthermore, these assessments only considered threats to species formally listed by the State, so 

considerably underestimated the real threat (Downey 2008). 

Downey(2008) applied similar methodology as used in NSW by Coutts-Smith and Downey (2006) and by Downey and Coutts- 

Smith (2006) to the biodiversity listed under the national Environment Protection and Biodiversity Conservation Act 1999, and 

found that alien plant invasions were a threat to 291 theatened species. Specific weed species were identified for only 33% of the 

threats, and these totalled 57 species. 

Invasion of native vegetation by environmental weeds is recognised as a threatening process under the Victorian Flora and 

Fauna Guarantee Act 1988 (Department of Sustainability and Environment 2009b) but no action plan has been developed. The 

83

negative impacts mentioned include limitation or prevention of recruitment of native taxa, alteration to fire regimes, hydrological 

cycles, nutrient cycling and other processes, increased soil erosion, genetic pollution, alterations to structure and floristics of 

native vegetation communities, competition, and niche modification. 

Approaches to impact assessment 

Assessments of the biodiversity impacts of weed are of four main types (Downey and Cherry 2005): 1. scientific studies of 

individual weed species and the systems they have invaded; 2. reviews or meta-analyses of such studies; 3. reviews of threatened 

species databases, and 4. detailed consultation with biodiversity stakeholders as part of a process of threat assessment and 

abatement planning. The former two methods approach the problem primarily from the individual weed perspective, the latter 

two more from the perspective of the impacted biodiversity. 

In terms of scientific studies, knowledge is particularly poor about how the biodiversity impacts of weeds vary in space and time 

(Grice 2004a). Assessment is complicated by communities and systems being in disequilibrium or being dependent in their 

evolution or dynamics or historical factors, rather than having a single stable state or successional pathway (Woods 1997). In 

general very little is known of the impact of invading species at different stages of community succession, or of the permanence 

of the occupation, i.e. the potential for community recovery if the factors initially permitting invasion are mitigated (Woods 

1997). Furthermore, the balance of negative and positive impacts can shift dramatically over time and across habitats (Groves 

2004). Knowledge about how weeds alter fire regimes and other ecological processes, and scientific understanding of the 

responses of a range of different taxa in the same area to a weed invasion is also very limited (Grice 2004a). Impacts on fauna 

are more complex than those on vascular plants and are therefore more difficult to determine (Grice 2006). In addition, the 

impact of measures to control the weed upon biodiversity is rarely known or investigated, although Coutts-Smith and Downey 

(2006) found that 7% of species threatened by alien plants were at risk from inappropriate control measures. 

A simple example representing a combination of the metanalysis/review approach to impact assessment is that of Carr et al. 

(1992), who identified 166 taxa threatened by environmental weeds in Victoria, including many found in grasslands. A slightly 

different approach identified the communities at risk: FFG SAC (1996) listed numerous communities threatened by weed 

invasion in Victoria, including Northern Plains Grassland, Plains Grassland (South Gippsland) and Western Basalt Plains 

Grassland. 

Much information about specific impacts is available but has never been adequately compiled and mobilised. Coutts-Smith and 

Downey (2006) demonstrated the great utility of a more comprehensive review of such existing threat information. The authors 

found that weeds posed a threat to 45% of threatened species, populations and communities in New South Wales and were the 

most important single threat after land clearing. However details of the specific biodiversity threatened by particular weeds, 

including Weeds of National Significance, in Australia was found to be almost entirely lacking, and only a very small proportion 

of the threat information obtained came from scientific studies. 

Downey and Cherry (2005) demonstrated the utility of the consultation approach in assessing weed impact for the coastal dune 

weed Chrysanthemoides monilifera subsp. rotundata (erroneously called subsp. monilifera by the authors). They found the 

number of species threatened by it to be 25 times higher than previously suggested. 

Types of impact 

Weed impacts can be harmful or beneficial (Adair and Groves 1998, Williams and West 2000, Low 2003, Richardson and van 

Wilgen 2004). Weeds can provide food, fodder, building materials, nectar, shade and numerous other benefits (Richardson and 

van Wilgen 2004). Weeds can contribute to conservation of biodiversity, for example by protecting other plants from herbivores 

and acting as refuges. Invasive plants may become food for native fauna, which ‘host-shift’ to feed on them, or already have 

wide host preferences. The possibility of host range expansion is one of the most important hazards in classical biological control 

of weeds (Hopper 2001): the deliberately introduced invader may prefer a non-target plant. Shapiro (2002) documented the case 

of the city of Davis, California, where a diverse, highly valued urban butterfly fauna is largely dependent on naturalised and 

cultivated alien plants, and where, in consequence, efforts to control the alien species conflict with biodiversity goals. Low 

(2002) provided numerous Australian examples of native animals, including endangered species, benefitting from alien plant 

invasions. 

In evolutionary time, the interactions of invasive species with other species in the invaded community changes selection 

pressures and ultimately results in evolutionary change, with new species arising (Cox 2004). Thus invasive species eventually 

tend to “become integrated into the new biotic community in such a way that their initial impacts are softened. Integration occurs 

through the processes of coevolution and counteradaptation” with the ecological adjustments tending to precede the evolutionary 

(Cox 2004 pp. 246-247). 

Food webs are one conceptual basis for comprehending the interactions of invasive species on the invaded community (Strong 

and Pemberton 2002 p. 59). Those that develop around animals introduced for biological control “are simpler than in natural 

communities” (Strong and Pemberton 2002 p. 57) and similar simplified systems may be expected around invasive plants. 

Unfortunately there is a general lack of detailed information about the food webs of even the most abundant native plants in 

Australia and the complexities of such interactions greatly complicate scientific assessment. 

Further complications are usually provided by weed management activities, since the weeds most worthy of study for 

biodiversity impact are generally those that are subject to control activities. Weed control activities themselves may impact 

negatively on biodiversity. In native grasslands herbicidal control in particular can have detrimental effects on native flora and 

lead to proliferation of non-target weeds (Lunt 1991, Slay 2002c, Brereton and Backhouse 2003). Such impacts are, by default, 

attributable to the particular weed that is being targetted, but little quantitative information on off-target damage is available. The 

effects of weed management activities on native invertebrates is unkown (Yen 1999). 

Specific threats posed by weeds to biodiversity 

Invasive plants potentially influence the structure, function and composition of ecosystems by impacting on growth, recruitment 

and survival (Grice 2004a Vidler 2004). These impacts are “ovewhelmingly negative”, but positive impacts also occur (Groves 

84

2004, Richardson and van Wilgen 2004). Complex, simultaneous negative and positive effects are probably usual. For example 

Lenz et al. (2003) found that the presence of annual exotic grasses on a hillside in one South Australian grassland facilitated 

native perennial grass growth on upper slopes but impeded it at the lowest elevations. 

Feedback processes, in which the invasive plant modifies the invaded environment or habitat for other organisms are doubtless 

frequently important. The invader may increase temporal or spatial resource fluctuations and may increase the heterogeneity or 

homogeneity of the area invaded in a wide variety of ways (Melbourne et al. 2007). 

Impact on the invaded systems may include changes to: 

1. Competitive interactions with other plants for light, nutrients, water, pollinators and other resources - resulting in changes in 

species composition, niche displacement, or replacement of another species (Weiss and Noble 1984, Adair 1995, FFG SAC 

1996, Woods 1997, Prieur-Richard and Lavorel 2000, Williams and West 2000, Levine et al. 2003, Vidler 2004). 

2. Species richness or dominance patterns (Adair 1995, FFG SAC 1996,Woods 1997). 

3. Physical structure and chemistry of the habitat (Adair 1995, FFG SAC 1996, Woods 1997, Williams and West 2000). 

4. Alterations to animal health, habitat, food chains and trophic structure of communities (Williams and West 2000, Groves 

2002, Low 2002, Levine et al. 2003). 

5. Phenology of native species (Woods 1997). 

6. Facilitating or allowing invasion of other species, including other plant or animal pests, or pathogens (Groves 2002). 

7. Genetic changes, including rates and details of evolutionary interactions, introduction of foreign genes, hybridisation and 

gene swamping (Carr 1993, FFG SAC 1996, Williams and West 2000, Cox 2004). 

8. Disturbance regimes and successional pathways (Woods 1997, Vitousek et al. 1997, Mack and D’Antonio 1998, D’Antonio 

et al. 1999, Prieur-Richard and Lavorel 2000). 

9. Ecosystem function and ecosytem services (Versfeld and Van Wilgen 1986, Adair 1995, FFG SAC 1996, Prieur-Richard 

and Lavorel 2000, Levine et al. 2003, Richardson and van Wilgen 2004) including nutrient cycling (Vitousek et al. 1997, 

Rossiter et al. 2006), hydrological processes (Vitousek et al. 1997, Versfeld et al. 1998, Williams and West 2000), 

geomorphological processes including soil erosion and landform (Adair and Groves 1998, Williams and West 2000), fire 

cycles (D’Antonio and Vitousek 1992) and C storage (Seabloom et al. 2003). 

10. Management regimes, resulting from altered management directed against the weed (Groves 2002). 

More detailed explanations and examples of each of these rather arbitrary categories are provided below. 

Competitive interactions 

Competition by invasive plants is by far the most frequently invoked cause of invasive plant impact on biodiversity (Levine et al. 

2003). ‘Weed competition’ with native plants had been identified as by far the most important cause in NSW, however in 

approximately half of the instances the competing weeds were not identified, and only rarely have competitive mechanisms been 

investigated (Downey and Coutts-Smith 2006). Mechanisms of competition between plants include sequestering of resources 

(space, water, nutrients, light), and alterations to the pathways and rates of cycling of energy, water and nutrients (Levine et al. 

2003, Grice 2004a, Vidler 2004). In general, the more resources an invasive plant obtains, the fewer are available for the invaded 

community, and invaders that dominate resources can reasonably be expected to have high biodiversity impacts (Grice 2006). 

The foliar cover achieved by successful invasive plants is a useful general indicator of their potential impact. Canopy dominance 

reduces light availability to subsidiary species and clearly reflects biomass dominance, which ultimately indicates the extent to 

which the plant monopolises available resources. A ‘canopy dominant’ environmental weed can totally or largely alter the nature 

and functioning of an ecosystem by dominating, overtopping or replacing the natural canopy, while a subcanopy dominant can 

have similar effects in a lower stratum (Swarbrick 1991). Woods (1997) argued that few experimental studies had demonstrated 

that competition for light by invasive plants is a causal factor of community change and there was similarly little support for the 

contention that competition for other limiting resources is important. This may largely be due to the lack and difficulty of 

adequate study, rather than the unimportance of such effects. Multi-factor competition may well be usual (Levine et al. 2003). 

Hautier et al. (2009) clearly demonstrated that competition for light causes losses of understorey species in experimental 

grassland communities. 

Asparagus asparagoides L. Druce (Liliaceae) has a strong negative impact on the endangered Pterostylis arenicola M. Clements 

and J. Stewart (Orchidaceae) probably because both grow from tuberous roots in autumn and winter and senesce in spring and 

summer (Groves 2002 2004). Root competition of this species has also been demonstrated to significantly reduce germination of 

the endangered Pimelea spicata R.Br. (Thymelaeaceae) (Groves 2002 2004). 

Competitive superiority of the invader to an ecologically similar native has been partially demonstrated for Chrysanthemoides 

monilifera subsp. rotundata (DC.) Norl. which directly displaces Acacia sophorae (Labill.) R.Br. on coastal dunes in New South 

Wales (Weiss and Noble 1984). Similarly a species may replace an ecological guild - Woods (1997, citing Chilvers and Burdon 

1983, etc.) cites the example of Eucalyptus spp. replacement by Pinus radiata D. Don. in Australia. 

Successful invasion resulting from superior competitive abilities may not result in any functional changes in ecosystem 

properties, the invasive species essentially functioning like the displaced native (Adair and Groves 1998), but the examples cited 

have, or will likely lead to more profound shifts in dominance patterns and the conversion of ecosystems to new types. 

Species richness and dominance patterns 

A consistent negative relationship has been found between the abundance or presence of invasive plants in Australian and that of 

native species (Grice 2006). Invasive plants are recognised threats to whole communities, e.g. the aforementioned 

Chrysanthemoides monilifera, which ultimately suppresses seedling establishment by most native species in some of the 

ecosystems it occupies, including canopy trees (Groves 2004). In Australia, Mimosa pigra has replaced native sedgelands with 

tall shurbland, Annona glabra L. has replaced wet grassland with closed forest, Acacia nilotica (L.) Willd.ex Del. has replaced 

dry grassland with tall shrubland, rainforest has been converted to vine thickets by Thunbergia grandiflora Roxb., Mcfadyena 

unguis-cati (L.) A. Gentry and Andredera cordifolia (Ten.) Steenis in Queensland (Panetta and Lane 1996), and Tamarix aphylla 

(L.) H. Karst. Had replaces Eucalyptus camaldulensis woodland in the inland (Groves 2002). Amongst the invasive Poaceae, 

85

Glyceria maxima (Hartman) Holmb and Brachiaria mutica (Forrskal) Stapf convert shallow water aquatic systems to wet 

grassland (Panetta and Lane 1996). Weeds that convert one ecosystem to another, changing its major functional characteristics 

have been termed transformers (see above) and are usually ‘canopy dominants’ (Panetta and Lane 1996). 

Most invasive plants are not transformer species but are are functional analogues of native species in the invaded systems. Few 

invasive grasses have the potential to change a grassland into any other ecosystem, but they frequently appear to simplify the 

system by reducing species richness and becoming dominant. For instance Hyparrhenia hirta was found to dominate areas it 

occupied, which also had reduced native plant species richness (McArdle et al. 2004). 

Weed invasion “can affect invertebrates adversely by elimination of native plant species, habitat alteration and ... the spread of 

exotic invertebrates” (Yen 1999 p. 63). 

Physical structure and chemistry 

Alterations to the biotic structure and composition of a community by an invasive plant affects the spatial and temporal patterns 

of resource flow within that community (Grice 2006). Altered soil chemistry can result from allelopathy, pH changes or changes 

in the availability of minerals, particularly major nutrients and salts (Woods 1997, Levine et al. 2003). 

Legumes (Fabaceae, Caesalpineaceae, Mimosaceae) such as Lupinus arboreus Sims, may increase soil N levels (Adair and 

Groves 1998) and have a direct effect on soil fertility. On young volcanic soils in Hawaii, the invasive tree Morella faya (Aiton) 

Wilbur (Myricaceae) and its microbial symbionts increased soil N fixation to levels 90 times that of all native plants combined 

and increased the rates of N mineralisation, which created a cascade effect through the system and altered its subsequent 

development (Vitousek et al. 1987, Vitousek and Walker 1989). Bacterial symbionts in Sorghum halepense enable it to invade 

N-poor soils, partly explain its ability to form dense near-monocultures that exclude other plants, and are largely responsible for 

its ability to significantly alter soil biogeochemistry (Rout and Chrzanowski 2009). 

Other plants can deposit salt on the surface, e.g. Mesembryanthemum crystallinum L. (Aizoaceae) (Woods 1997). In Argentine 

grasslands Pinus spp. alter the pH and other properties of soils to the detriment of some native grasses (Amiotti et al. 2007). 

Alterations to the soil chemistry in turn commonly result in alteration to nutrient cycles. 

Many weeds were introduced to combat soil erosion. Chrysanthemoides monilifera was widely planted and became a major 

weed because it is an efficient sand binder (Groves 2004). McIlroy et al. (1938) advocated the use of three now severely invasive 

grasses Paspalum dilatatum, Pennisetum clandestinum Hochst. ex Chiov. and Cynodon dactylon L. for erosion control in 

Victoria. 

The many physical, chemical and biotic effects of the production and deposition of litter by invasive plants can impact markedly 

on biodiversity. Presence of litter can reduce seedling establishment (Lenz et al. 2003). Litter alters the microclimate, surface 

conditions and soil properties including temperature, water infiltration, retention and evaporation, and may produce chemical 

leachates. These changes in turn can modify competitive interactions between organisms, alter rates of seed and seedling 

predation, favour proliferation of fungal pathogens, etc. (Lenz et al. 2003). Avena litter at 400 g m -2 was found to significantly 

reduce maximum soil temperature (by c. 3ºC in early November) but not minimum temperature or soil moisture (Lenz et al. 

2003). Ens (2002a) found that dense litter mats produced by N.neesiana altered the species composition and activity levels of 

invertebrate communities with flow-on effects through the system. Dense grass litter in grasslands generally results in reduced 

plant biodiversity (Lenz et al. 2003). Buildup of perennial native grass litter in the higher productivity temperate south-eastern 

Australian grasslands, paticularly T. triandra grasslands, has the same effect, so regular biomass reduction through fire, grazing 

or other management is necessary to maintain vascular plant diversity (Stuwe and Parsons 1977, McIntyre 1993, Morgan 1997, 

Henderson 1999, Lunt and Morgan 2002). For example Morgan (1995a 1995b) found that seedling establishment of Rutidosis 

leptorrhynchoides in dense T. triandra grassland required large canopy gaps, mortality of young seedlings in smaller gaps being 

due to shading and increased herbivory. In contrast, litter experiments in Dry Themeda grasslands in the ACT by Sharp (1997) 

found that retention of the T. triandra litter resulted in higher native forb richness and cover than when litter was removed, with 

the opposite effect for exotic forbs. The effects of grass accumulation on plant productivity differs seasonally and from site to 

site and species to species (Lenz et al. 2003). Generally plants with smaller seeds are inhibited more by litter because the 

germinants have inadequate energy reserves (Lenz et al. 2003). Annuals are thus more likely to be negatively impacted. 

In pot experiments using soils from native grassland in South Australia Lenz et al. (2003) found that Avena arbata litter at 

different densities had complex, time-dependent effects on the emergence and biomass of seedlings, that varied between taxa. 

The establishment or growth of some exotic dicots was reduced by dense litter, but after 4 months the biomass of annual grasses 

was positively affected , and total biomass fell markedly in the dense litter treatment, mainly due to suppression of Trifolium 

spp.. Medium and high litter levels had little effect on the only native plants that emerged in sufficient numbers to assess, 

Austrodanthonia spp., but after 4 months all had died due to fungal disease. Field experiments, in which all standing biomass 

was removed and dry Avena litter added, demonstrated similar complex effects. In winter, c. 4 months after treatment, growth of 

annual grasses was increased by dense litter and there were significant suppressive effects on Austrodanthonia spp., and some 

exotic herbs, but no effect on Austrostipa eremophila (Reader) S.W.L. Jacobs and J. Everett. After c. 9 months, the biomass of 

exotic annual grasses was slightly increased by high litter levels, there was no siginficant effect on native perennial grasses, but 

the biomass of all other species combined (mostly exotic forbs) was significantly decreased. 

Invasive plants can severely modify hydrological cycles (Vidler 2004). Tamarix in inland Australia lowers the water table, alters 

stream flow and flooding regimes and ultimately salinity levels (Griffen et al. 1989). Andropogon virginicus L., develops a high 

biomass of dead shoots that reduce evaporation rates from the soil and also passes through an inactive senescent phase during 

which transpiration is reduced (Mueller Dombois 1973). In rainforest communities this results in excess water in the soil, 

increased runoff and accelerated erosion. 

Altered microclimates (Vidler 2004) are probably commonplace, particularly through increased shading. 

Alterations to animal health, habitat, food chains and community trophic structure 

Invasive plants frequently have impacts on consumers and decomposers, including their community composition, diversity and 

behaviour (Levine et al. 2003). The impacts of invasive plants on fauna are more complex than on plants, and can be positive or 

86

negative depending on the particular animal group or species (Grice 2006). Invasive plants may be utilised by animals in the 

same wide range of ways as native plants and can degrade or enhance the habitats of animals (Low 2002, Vidler 2004). Weeds 

provide food and shelter for vertebrates (Low 2002) including vermin such as rabbits and foxes (Parsons and Cuthbertson 1992) 

and native and introduced birds (Loyn and French 1991). Peter (2000) for example provided details of shelter and nest site 

provision by Lycium ferocissimum Miers, and utilisation of its fruit by native and exotic birds. Fruit and seeds are widely eaten, 

along with foliage, while nectar, pollen, roots and other plant parts may be exploited. Invasive plants may be used when nothing 

else is available. Loyn and French (1991 p. 138) noted that weeds used as food by birds “may be better than nothing, but not as 

good as the native plants they displace”. Grice (2004a) noted that existing studies provide little indication of the importance of 

the plant for the diet of the animal utilising it, nor the impact of the feeding on the plant. 

Invaded habits may be less suitable for animals in many ways. Habitat degradation in invaded areas may mean loss of food or 

shelter or alteration to physical conditions that make the site unsuitable for habitation. Much of the literature has focused on how 

the behaviour and ecology of individual native animals is altered by different characteristics of the invasive and native plants 

(Levine et al. 2003), Valentine (2006) found that areas invaded by Cryptostegia grandiflora Roxb. ex R.Br. (Asclepiadaceae) 

were less suitable for lizards. Invasive plants may simply out-compete food plants of endangered animal species (Vidler 2004). 

In general they can be expected to support a different suite of primary consumers to the plants they displace (Grice 2004a). 

Weeds may be toxic to animal species. Groves (2002) discussed impacts on animal health, but mentioned only livestock related 

cases. Little appears to be known about poisoning of native animals by invasive plants in Australia but some examples are on 

record. The decline of the Richmond Birdwing, Ornithoptera richmondia (Gray) (Lepidoptera: Papilionidae) in Australia is 

partly due to the introduced Dutchman’s Pipe vine Aristolochia elegans Mast. (Aristolochiaceae), a close relative of the native 

food plants. A. elegans is highly attractive to ovipositing female butterflies but the young larvae are poisoned and do not survive 

(Sands and Scott 1996, Braby 2000). Another vine, Cryptostegia grandiflora is possibly toxic to reptiles (Valentine 2006). 

Environmental weeds can also reduce access to breeding, nesting or feeding sites (Vidler 2004). Vertebrates that consume the 

fruit and seeds of exotic plants can in turn can become important dispersers of propagules. Dispersal of fleshy-fruited weeds by 

birds is particularly important in this respect (Loyn and French 1991). Carr (1993) provided a list for Victoria of some 

naturalised plants and the indigenous and exotic animals which disperse their seed following ingestion of fruit or seed. The 

complexities of impact may take many years to run their course. Synergistic effects, e.g. when a bird dispersed weed facilitates 

invasion by other bird dispersed weeds (Grice 2004a) may continue almost indefinitely in environments where new adventives 

are always appearing. 

Impacts on detritivores and decomposers have been much less investigated (Levine et al. 2003). 

Phenology of native species 

Phenology is the study of the relationship between climate and the temporal variation of the lifecycle of an organsim. Any 

modification of microclimate caused by an invasive plant may affect a range of other species. Increased shade due to riparian 

invasions of the exotic Siam weed Chromolaena odorata (L.) R.M. King and H. Rob. (Asteraceae) in South Africa have reduced 

soil temperatures and altered the sex ratio of locally breeding Nile crocodiles (Leslie and Spotila 2001). Any plant examples? 

Facilitating other invasions 

Invasive plants can facilitate the invasion of other plants. Fixation of N by legume weeds can profoundly alter soil conditions to 

the detriment of native species and potential benefit of new invaders (Levine et al. 2003). Australian Acacia species invasive in 

South African fynbos make the environment far less suitable for fynbos plants (Versfeld and van Wilgen 1986) and thus more 

suitable for other species. As previously mentioned the attraction of furit eating animals to a weed food source may facilitate 

consumption of the fruit of other weed species and dispersal of their seed. 

Genetic changes 

Loss of genetic diversity in particular plant species or populations resulting from weed invasion are probably widespread, but 

have rarely been investigated, in part due to the lack of or difficulty of appropriate techniques (Adair and Groves 1998). 

Probably the most common impacts occur where weed invasion destroys small, isolated or remnant populations that often 

possess more extreme geneotypes than core populations. Invasive perennial grasses have played a role in reducing the genetic 

variance of Rutidosis leptorhynchoides in temperate Australian grasslands by contributing to the extinction of local populations 

(Groves 2004). 

On a world basis genetic diversity information for grasslands is entirely inadequate and the genetic composition of only a very 

small proporrtion of species has been investigated. Aspects of genetic diversity requiring investigation include its spatial 

distribution, structural and functional attributes and processes under various disturbance regimes (Aguiar 2005). Inadequate 

baseline data makes assessment of genetic biodiversity impacts very difficult. 

Hybridisation is another threat. Hybridisation of indigenous species with exotic garden escapes, accidentally introduced exotics 

and other indigenous species established outside their native range have all been reported in Australia, along with exotic-exotic 

hybrids (Carr 1993). For example hybrids or possible hybrids between the Argentinian Nicotiana glauca Graham (Solanaceae) 

and the natives H. suaveolens Lehm. and N. velutina H.-M. Wheeler have been found in Victoria (Carr 1993, Jeanes 1999a). The 

likelihood of hybrid vigour and the possibility of hybridogenic speciation are particular concerns (Carr 1993). The presence in 

Australia of a set of Nassella species from different areas of the Americas and their introduction into an environment inhabited 

by a large set of native stipoids provides unique circumstances that may allow novel gene flows, with unpredictable effects. 

Disturbance regimes and successional pathways 

Invasive plants may act directly as disturbance agents but they can also modify the response of the community to disturbance. 

The literature survey of Mack and D’Antonio (1998) found numerous studies in which plant invasions led to subsequent 

alteration of disturbance regimes. Such changes included alteration of physical or biological attributes of the disturbance (e.g. 

enhancement or suppression of fire and erosion) and changes in the responses of other plants. Invasion by species that interact 

strongly on disturbance regimes can often produce “discrete state changes in ecosystem structure and function” (Mack and 

87

D’Antonio 1998 p. 195). Alterations to fire regimes are the best documented (Versfeld and van Wilgen 1986, D’Antonio and 

Vitousek 1992, Grice 2006), along with invasions by woody weeds (Adair and Groves 1998). 

Altered disturbance regimes are less likely when the invasive species differs little from the natives, but subtly different 

organisms can produce subtle changes (Mack and D’Antonio 1998). Permanent changes to disturbance regimes and successional 

pathways eventually result in conversion of the system to a new state. 

Ecosystem function and ecosytem services 

Weeds that significantly modify ecosystem function affect the living conditions of all the other species in the system, and thus 

have the most profound effects on biodiversity (Adair and Groves 1998). Most studies of these impacts have involved 

comparison of invaded and uninvaded areas and require cautious interpretation because the mechanisms that enable one system 

to be invaded and not the other are usually poorly understood (Levine et al. 2003). 

Woody weed invasions of South African fynbos have resulted in major declines in stream flows and water yield (Versfeld and 

van Wilgen 1986, Versfeld et al. 1998). Mechanisms of impact include increased interception of rainfall, increased transpiration 

and changes in infiltration and erosion rates (Versfeld and van Wilgen 1986). Invasive species impacts can also result in 

reductions in water use (Levine et al. 2003) and thus increased flooding and habitat change. The replacement of native perennial 

bunchgrasses by annual exotic grasses in California grasslands has decreased the amount of C they store (see references in 

Seabloom et al. 2003), and consequently the amount of atmospheric CO 2 and the extent of global warming. Impacts on nutrient 

cycling have been widely investigated with a focus on nitrogen (particularly in grasslands) and leguminous plants (Levine et al. 

2003). 

In the most extreme cases weeds threaten whole ecosystems or ecosystem units, either by creating a new vegetation stratum or 

by altering major ecosystem properties (Adair and Groves 1998). Acacia nilotica (L.) Delile thickets threaten Mitchell Grass 

grasslands in the Northern Territory and northern Queensland by creating a dense overstorey, with over 1 million ha infested by 

1992 (Parsons and Cuthbertson 1992). The grass-fire feedback cycle (D’Antonio and Vitousek 1992), where the biomass of an 

invasive grass enhances fire (see below) is transforming ecosystems around the world. 

Impacts on the delivery of ecosystem goods and services may be evident, but few studies have quantified the effects at regional 

or wider scales (Richardson and van Wilgen 2004). 

Management regimes 

Management activities are targetted at significant weeds and often have unintended consequences. Attempts to reduce the 

prevalence of Nassella trichotoma in Victoria by establishment of trees have attracted criticism that they will also eliminate the 

native grassland remnants in which the weed is growing. Similarly, herbicidal control of exotic stipoid grasses in Melbourne area 

grasslands has been criticised for its severe impact on native plants. 

Invasive grasses - impacts and threats 

Useful grasses have been widely introduced as forage plants for livestock and for other purposes, and many grasses of less forage 

value have dispersed widely without deliberate human intervention. In many cases they have replaced native grasses, being the 

agents, the beneficiaries, or both agents and beneficiaries of ecosystem transformation. These transformations have taken place 

in many of the major temperate grassland regions of the world. North American grasslands have been subjected to massive 

changes through the deliberate introduction of a wide range of grasses considered superior for livestock production including 

Bothriochloa spp. (Schmidt et al. 2008). The grasslands of the Llanos, Venezuela, and savannah-forest in the Cerrado, Brazil, 

both supported cattle grazing, but the Spanish and Portuguese immigrants considered the native grasses inferior and by the late 

18th century had intoduced African species such as Brachiaria mutica (Forrsk.) Stapf, Melinis minutiflora P. Beauv. and 

Panicum maximum Jacq., which are now the dominant species in huge areas of tropical and subtropical Latin America (Mack 

and Lonsdale 2001). Similar transformations have occurred in Australia. 

Poaceae is one of few plant families that consistently provides a high proportion of invasive species relative to its total taxa 

(Rejmánek 2000). Poaceae is probably the dominant weed family in Australia in terms of areas occupied and species diversity. 

Poaceae represent about 14% (375 spp.) of all naturalised vascular plant species in Australia, and 141 spp. (37.6%) are 

considered major weeds (Grice 2004b). There are more species and infraspecific taxa of Poaceae in the exotic flora of Victoria 

than any other family (Carr 1993). 

Invasion of native plant communities by exotic perennial grasses is listed as a key threatening process under the NSW 

Threatened Species Conservation Act 1995, based on the impact of five species, Hyparrhenia hirta, Cenchrus ciliaris, Eragrostic 

curvula, N. trichotoma and N. neesiana (NSW Scientific Committee 2003, Downey in Virtue et al. 2004). Perennial grasses are 

one of the major groups threatening biodiversity in NSW (Downey and Coutts-Smith 2006). N. neesiana, N. trichotoma and E. 

curvula are among the major species recognised as threats in temperate grasslands (Kirkpatrick et al. 1995, Groves 2004). More 

cautiously, Adair and Groves (1998 p. 9) suggested that N. neesiana invasion of temperate Themeda grasslands is “perhaps” an 

example of simple species displacement, causing no significant functional changes in the ecosystem. 

McArdle et al. (2004) investigated the impact of H. hirta by comparison of the botanical composition of matched invaded and 

uninvaded areas in Kwiambal National Park, northern NSW, and demonstrated reduced native plant richness and projective 

foliar cover in the ground strata of invaded areas. Exotic components of the system were not affected. The tendency of this plant 

to dominate was quantified, with infested sites being more homogeneous. Chejara et al. (2006) further investigated the impact on 

vascular plant diversity of this species on a travelling stock route near Manilla NSW. Where it was present it was the dominant 

species. Invaded areas had native plant species richness significantly reduced by half or more, as determined by spring and 

autumn surveys in 2003 and 2005, and native cover was significantly less in infested plots. A major fault with this study was that 

the areas lacking H. hirta had been spot sprayed with glyphosate to control the plant from 2001 to 2004. These plots were found 

to contain significantly greater numbers of exotic weed species in 2005. Another problem was that the invaded and uninvaded 

areas were 1 km apart and were “similar in respect to soil, landform, drainage and apparent disturbance history”, but there was 

no way to tell whether the vegetation prior to invasion had been similar (Chejara et al. 2006 p. 208) 

88

Competition with native plants 

Competition with other plants is a widespread general expectation for invasive grasses (Evans and Young 1972, Newsome and 

Noble 1986, D’Antonio and Vitousek 1992, Lonsdale 1999, Grice 2004). ‘Unequal competition’ is widely assumed where 

grasses are invasive, but the mechanisms by which it operates are rarely demonstrated (Seabloom et al. 2003). Release from 

natural enemies is one contributing factor often suggested (e.g. Schmidt et al. 2008). Phalaris aquatica and Ehrharta calycina 

J.E. Sm., for example, are associated with major habitat degradation on roadsides on Kangaroo Island, threatening a range of 

endangered plants, imputedly the result of their superior competitive abilities (Vidler 2004, citing Davies 1996 and Taylor 2003). 

Competitive superiority of invasive species was demonstrated for Old World Bothriochloa species over native Kansas 

bunchgrasses by Schmidt et al. (2008). In pot experiments, the exotics reduced one or more of three productivity attributes of 

native species, either vegetative tiller height or above or below ground biomass, while two of three native species tested failed to 

inhibit growth of the exotics. 

Demonstable competition has been recorded where the invasive grass has higher rates of N uptake and higher N use efficiency 

(Rossiter et al. 2006). This is thought to be the case with exotic Bothriochloa inavsion of native tallgrass prairie in the USA 

(Schmidt et al. 2008). Increasing N levels in one American grassland created dramatic shifts in grass species dominance, with 

Bromus hordeaceus becoming a superior competitor to grasses including Aira caryophyllea and Briza minor, in the presence of 

adequate P. B. mollis has also been demonstrated to be an inferior competitor to Erodium botrys (Cav.) Bertol. (Geraniaceae) 

when the S status of soil is low, E. botrys having more rapid root growth. But when nutritional conditions were favourable, B. 

mollis outcompeted the herb for light because of its greater size and more erect habit (Evans and Young 1972). Competitive 

shifts in the flora resulting from changed nutrient status of the soil may be altered by a range of other environmental factors, 

including weather, grazing regime and the interaction between rainfall, temperature and grazing (Evans and Young 1972). N 

enrichment of soils in Californian grassland in dry years resulted in the almost complete elimination of the native bunchgrass 

Agropyron intermedium (Host) Beauv. by Bromus tectorum L., depletion of soil moisture by the superior competitor being 

thought to be the main mechanism (Evans and Young 1972). 

Monocultures of invasive Bothriochloa species in the USA are more structurally homogeneous than native grassland and have 

reduced forb species richness (Schmidt et al. 2008). 

There are relatively few studies that document the effects of exotic grasses on species diversity in Australia in any detail 

(Chejara et al. 2006), but numerous introduced species are implicated in decline of native vegetation via competition. These 

include Cortaderia spp. (Harradine 1991). Threats to small populations of endangered plants by competition from introduced 

grasses, include Nassella spp. on Amphibromus pithogastrus S.W.L. Jacobs and Lapinpuro “by reducing potential bare areas for 

establishment of seedlings” (Ashton and Morcom 2004 p. 2), and Phalaris aquatica L. on Prasophyllum fosteri D.L.Jones 

(Coates 2003a), P. sp. aff. suaveolens (Western Basalt Plains) (Coates 2003b) and Thelymitra gregaria D.L. Jones and M.A. 

Clem. (Coates 2003c). 

In Queensland, Brachiaria mutica Para grass and H.amplexicaulis are a threat to the aquatic Aponogeton queenslandicus 

H.Bruggen (Vidler 2004 citing Williams pers. comm.). Kikuyu, Pennisetum clandestinum, is a threat to Pimelea spicata R.Br. in 

NSW (Vidler 2004 citing Groves and Willis 1999). Ehrharta calycina is a threat to Blue Gum woodlands (Eucalyptus leucoxylon 

F. Muell.) and Metallic Sun Orchid Thelymitra epipactoides F. Muell. in South Australia (Vidler 2004 citing Mercer pers. 

comm.) and to Eucalyptus incrassata and E. fasiculosa woodland associations and the Sandhill Greenhood Pterostylis arenicola 

(Virtue and Melland 2003). E. calycina frequently establishes on bare ground. Native vegetation “subject to disturbances such as 

livestock grazing, fire or soil movement [is] particularly prone to invasion” although “certain ‘naturally open’ vegetation types 

on sandy soils appear susceptible to invasion in the absence of major disturbance” (Virtue and Melland 2003 p. 111, citing pers. 

comms. of B.Bartel and D. Ancell) or the plant “may be mainly establishing in gaps (e.g., on lichen crusts) where there is no 

competing vegetation” (Virtue and Melland 2003 p. 112). E. calycina “can have a major effect on the diversity and regeneration 

of native plants, particularly understorey species” (Virtue and Melland 2003 p. 112) and can form 100% groundcover (G.Carr 

pers. comm. cited by Virtue and Melland 2003). 

Grasses collectively, or particular categories of grasses, have also been regularly listed as threats to particular native plants. 

Introduced grasses are a threat to Shiny Peppercress Lepidium aschersonii Thell. in NSW (Vidler 2004 citing Ayers et al. 1996). 

In NSW exotic annual grasses are a threat to Red Darling Pea Swainsona plagiotropis F. Muell. while exotic grasses are a threat 

to Swainsona recta A.T. Lee (Vidler 2004 citing Ayers et al. 1996). Grasses are a threat to Ironstone Grevillea Grevillea 

elongata Olde and Marriott in WA (Vidler 2004 citing Stack and English 2003a). Annual grasses are a threat to Acacia aprica 

Maslin and A.R.O. Chapm. blunt wattle in WA (Vidler 2004 citing Bayliss 2003). Wild Oats Avena fatua and other “grasses” are 

a threat to Pinnate-leaved Eremophila Eremophila pinnatifida Chinnock MS in WA (Vidler 2004 citing Stack and Brown 2003) 

and introduced grasses are a threat to Spreading Grevillea Grevillea humifusa Benth. in WA (Vidler 2004 citing Stack and 

English 2003b). African Lovegrass Eragrostis curvula (Schrad.) Nees is a threat to Narrow-petalled Featherflower Verticordia 

plumosa (Desf.) Druce var. pleiobotrya A.S. George in WA (Vidler 2004 citing Phillimore and Evans 2003). Weeds associated 

with irrigation crops are a threat to Menindee Nightshade Solanum karsense Symon, in NSW (Vidler 2004 citing Ayers et al. 

1996). Quaking Grass. Tall wheatgrass Thinopyrum ponticum (Podp.) Z.-W. Liu and R.R.-C. Wang “has been observed invading 

native Themeda triandra, Austrostipa and Austrodanthonia grasslands” in Victoria (Virtue and Melland 2003 p. 127) but is most 

successful at sites with high soil moisture levels and water tables and saline or alkaline soils, particularly winter-wet, saline sites 

(Virtue and Melland 2003). 

Competition from grasses have been listed as a threat to a wide range of plants in basalt plains grasslands, including, introduced 

grasses on Carex tasmanica Kuk. (Morcom 2004) and Comesperma polygaloides F. Muell. (McIntyre et al. 2004), annual 

grasses on Senecio macrocarpus Belcher (Hills and Boekel 1996 2003) and ‘dense grasses’ on Rutidosis leptorrhynchoides 

(Humphries and Webster 2003), although this probably refers to T. triandra more than exotic species (cf. Morgan 1995a). 

89

In comparison to annual grasses, perennial grasses have relatively low seedling vigour and slow early growth (Evans and Young 

1972). The outcome of competition between perennial grass species is therefore highly dependent on the timing of germination 

and differential rates of early growth. 

Impacts on animals 

Invasive grasses may provide or fail to provide resources for a wide range of other organisms. They are used by birds in nest 

construction e.g. Nassella trichotoma by Yellow-rumped Thornbill Acanthiza chrysorrhoa (Quoy and Gaimard) (Peters 2000). 

Naturalised grasses provide food for various animals. Many birds eat the seeds of exotic grasses in Australia, including the 

Stubble Quail Coturnix pectoralis Gould and Plains Wanderer Pedionomus torquatus Gould in native grasslands (Loyn and 

French 1991). Most cockatoos and parrots consume exotic grass seeds and some species have become largely dependent on the 

resource, paticularly in cereal growing areas (Cole 1975, Loyn and French 1991, Barker and Vestjens 1989). In Victoria 

kangaroos eat and disperse the seed of Aira elegantissima Schur, Briza minor L., Hordeum marinum Huds., Lolium rigidum 

Gaudin and Vulpia bromoides, while feral horses and sambar eat and disperse the seed of Anthoxanthum odoratum and Holcus 

lanatus L. (Carr 1993). 

Significant negative impacts on native fauna appear to be commonplace where extensive, dense infestations of exotic grasses 

occur, but have rarely been investigated in any detail in Australia. Compared to habitats dominated by native grasses, 

monocultures of invasive Bothriochloa species in central Texas have been found to have reduced rodent species richness and in 

Kansas had significantly reduced bird species richness, and reduced abundance and biomass of arthropods (Schmidt et al. 2008). 

The suppression of native psammophilous grasses by invasive Cynodon dactylon (L.) Pers. in Germany has impacted on the 

native leafhopper fauna of these grasses, and two exotic leafhoppers associated with C. dactylon now occur (Biedermann et al. 

2005). The advent of New World-Old World hybrid Spartina in Europe appears to have resulted in the establishment of an 

American planthopper that attacks the European native Spartina maritima (Curtis) Fern. (Biedermann et al. 2005). 

Some recorded negative impacts on fauna in Australia include those of Aleman Grass Echinocloa polystachya (Kunth) A.S. 

Hitchc., and Olive Hymenachne Hymenachne amplexicaulis (Rudge) Nees, which “can choke out waterways used by ... pygmy 

geese” (Vidler 2004, citing Garnett 2003). In Queensland, Brachiaria mutica Para grass and H.amplexicaulis are a threat to the 

Jabiru Ephippiorhynchus asiasticus Latham (Vidler 2004 citing Williams pers. comm.). Briza maxima L., is considered a threat 

to the Eltham Copper butterfly Paralucia pyrodiscus lucida Crosby (Vidler 2004 citing DPI/DSE 2003b). 

Hydrology 

Invasive grasses can modify hydrological cycles in a range of simple and complex ways. They can modify the rate and timing of 

evapotranspiration, infiltration and overland flow of water, and of the nutrients, minerals and soil particles in the water (Levine 

et al. 2003, Grice 2004). 

Alterations of soil water usage, in total, seasonally and at different levels in the soil may occur. Replacement of native summergrowing 

grasses with annual spring-growing grasses results in wetter autumn soils, higher water tables and increased drainage 

flows (Sinclair 2002). Replacement of deep-rooted perennials by shallow rooted annuals may reduce water use and concentrate 

water use to a particular season (Levine et al. 2003). In Californian native grassland, annual grasses reduced the reproduction 

and seedling growth of native perennial Poaceae through competition for soil moisture (Lenz et al. 2003). Absence of summer 

growth due to prior depletion of soil water may also result in higher erosion during intense summer rainfall events (Sinclair 

2002). Markedly increased rates of soil drying by high densities of the annual invasive Bromus tectorum adversely affect 

seedlings of a native perennial grass when the two were germinated simultaneously (Evans and Young 1972). 

Disturbance regimes 

Exotic grasses can change the nature and timing of disturbances through feedback effects (Woods 1997). Alterations to fire 

regimes and the frequency and intensity of flooding, erosion or herbivory may occur. 

The most dramatic and best documented of feedback disturbance effects involve increases in fire (D’Antonio and Vitousek 1992, 

Woods 1997, Mack and D’Antonio 1998, Levine et al. 2003, Vidler 2004). Changes to evapotranspiration may lead to altered 

soil and vegetation dryness patterns, while changes in timing and density of biomass production can alter the fuel load and its 

temporal and spatial (vertical and areal) distribution (including continuity and curing rates), leading to changes in the the 

frequency, severity and timing of fires. Grass invasions commonly result in increased grass biomass production (Rossiter et al. 

2006). Grass leaves have a high surface area: volume ratio and grasses commonly accumulate large amounts of dead biomass 

(Mack and D’Antonio 1998) with some species producing much more poorly biodegradable, inflammable bulk than others. But 

in most cases where increased biomass production occurs, the specific reasons are unknown (Levine et al. 2003). More frequent 

and hotter fires result in higher rates of nutrient loss and alterations in microclimate, and may stymie succession processes; thus 

altered fire regimes can result in major shifts in the composition and functioning of ecosystems, including dramatic alterations to 

biodiversity. In northern Australia, Gamba grass, Andropogon gayanus (Kunth), a South African species, produces up to seven 

times as much fuel as native grasses, resulting in a fire regime that is more frequent and much more intense (Rossiter et al. 2003, 

Ferdinands et al. 2006). If the changed fire regime leads to greater abundance of the responsible grass a ‘grass-fire cycle’ is 

initiated that reinforces the impact of the invasion (D’Antonio and Vitousek 1992, Hobbs and Heunneke 1992). Invasion by A. 

gayanus has led to reduced tree cover in open woodlands by this mechanism (Ferdinands et al. 2006). 

Mission grass Pennisetum polystachion (L.) Schultes and buffel grass Cenchrus ciliaris L., two other high biomass invasive 

species in northern Australia also effect fire regimes (Rossiter et al. 2003, Grice 2004). Fire enhances growth of C. ciliaris which 

in turn enables more intense fires (Puckey and Albrecht 2004). A large range of short lived native grasses and forbs disappear 

when the density of C. ciliaris reaches a certain threshold. The number of native ground cover species declines significantly, 

there is very little germination of native seed and total invertebrate diversity and abundance of most inverterate groups is reduced 

(Puckey and Albrecht 2004). It is a specifically identified as a threat to the skipper butterfly Croitana aestiva E.D.Edwards 

(Vidler 2004 citing Wilson and Pavey 2002). Increased cover of C. ciliaris has been correlated with a decline in the numbers of 

Carnaby’s skink Cryptoblephrus carnabyi and delicate mouse Pseuodmys delicatulus in central Queensland (Puckey and 

Albrecht 2004 - see their references). C. ciliaris cultivar populations have low genetic diversity because of the dominance of 

90

asexual seed production, so local fungal epidemics can damage whole populations, increasing the likelihood of further ecological 

damage (Puckey and Albrecht 2004). 

Temperate Australia appears to be less susceptible to an invasive species grass-fire cycle, in part because climatic conditions 

mitigate against very high biomass production. Milberg and Lamont (1995) inferred increased fire susceptibility due to invasion 

by Ehrharta calycina and Eragrostis curvula on roadsides in Western Australia. E. calycina was also implicated as a cause of 

more frequent fire by Virtue and Melland (2003), as were Cortaderia spp. in Tasmania by Harradine (1991). McArdle et al. 

(2004) suggested that Hyparrhenia hirta has the potential to induce a positive feedback fire cycle because of its dense tussock 

form that may protect the growing points from fire damage. Stoner et al. (2004) demonstrated that invasive Phalaris aquatica 

produced approximately three times the fine fuel biomass of T. triandra, the grass it replaced in their study area of southern 

Victoria, and argued that the increased fire intensity and flame residency and burnout times would be more likely to irreversibly 

damage native plant communities. 

Invasive plants may also decrease the intensity or frequency of fire. Succulent plants or mesic species can have this effect (Carr 

1993). However, as with Pittosporum undulatum in south-eastern Australian it may be difficult to tell whether the plant is 

reducing the fire-proneness of the vegetation or invading as a result of a pre-existing reduction of burning (Carr 1993). N. 

neesiana might reduce the incidence or severity of fire in spring in Themeda triandra grassland by increasing the ratio of green 

to dry vegetation in the standing crop (N. neesiana being a spring grower and T. triandra a summer grower), or it might possibly 

reduce fire in general by producing a smaller amount of flammable material than the plants it displaces. 

Impacts on nutrient cycling 

Several African grasses are known to fix significant levels of N in their native habitats (Rossiter et al. 2003). Invasive grasses 

can also alter N fixation rates by displacing legumes or by reducing the litter of other plants that support non-symbiotic N fixers 

(Rossiter et al. 2003). Invasive grasses may produce litter with different physical and chemical properties which accumulates and 

decays at altered rates and seasons (Grice 2004a). Higher C:N and lignin:N ratios in the foliage and litter may reduce nitrogen 

mineralisation rates (Levine et al. 2003). The decomposition rates of invasive grass litter was lower than that of native grasses in 

three of six cases reviewed by Rossiter et al. (2003). 

Where invasive grasses displace summer growing species there is reduced uptake N mineralised in summer, so more is lost by 

leaching after autumn and winter rains (Sinclair 2002). 

Other effects 

Another class of feedback effects include erosion and soil stabilisation. According to Heyligers (1986) the introduced coastal 

dune grasses Ammophila arenaria and Thinopyrum junceiforme are more efficient at trapping sand and better colonisers of the 

backshore zone than native dune grasses. The dunes they build are larger and have a different shape. They also build foredunes 

in areas where the native grasses would be ineffective sand stabilisers. Changes in erosion patterns resulting from substrate 

stabilisation are also caused by Spartina (Gray et al. 1997) and Cynodon dactylon (Mack and D’Antonio 1998). 

More complex alterations to disturbance regimes occur with grazing. Caldwell et al.(1981) found that invasive Agropyron repens 

(L.) Beauv. had greater photosynthetic capacity in its new growth and recovered more quickly after grazing than a dominant 

native species Agropyron spicatum (Pursh) Scribn. and J.G. Sm., and that these factors were driving species replacement over 

large areas. Thus invasive grasses have the potential to alter successional dynamics (Grice 2004). 

Impacts of N. neesiana 

Hocking (1998 p. 86) argued that the biodiversity impact of N. neesiana in Australia was “likely to be major” in part because it 

was known to be “actively invading high quality grassland remnants at much higher rates than serrated tussock and to have a 

greater potential for invasion of grassy woodlands, over a wide range of climatic conditions” (Hocking 1998 p. 89). Earlier, 

Morgan (1994 p. 88) considered it to be“one of the most troublesome grassy weeds of grasslands”. However major biodiversity 

impacts are more likely to arise from weeds with “growth forms that are novel to the invaded ecosystem [rather] than growth 

forms for which there is a native ecological analogue” (Grice 2004a p. 55). Such weeds are more likely to be ‘transformer 

species’. N. neesiana has a growth form similar to a number of native species that are commonly dominant or subdominant in 

temperate native grasslands in south eastern Australia. Various Austrostipa and Austrodanthonia species have similar tussock 

forming habits and stature, have similar cool-season growth periods and probably a markedly similar phenology. Hocking (1998 

p. 86) also observed that “some well-managed” native grassland remnants have shown resistance to invasion “but further 

documentation is needed”. 

Exotic stipoid grasses including N. neesiana have been identified as “one of the most significant issues ... threatening nationally 

important remnant grasslands in Australia” (McLaren, Stajsic and Iaconis. 2004). N. neesiana has been identified as a particular 

threat to numerous grasslands, e.g. notably by Craigie (1993) as a “very serious threat to the integrity” of the Laverton North 

Grassland, because few native plant species survive beneath dense infestations. According to Craigie (1993) “prior disturbance” 

did “not seem to be necessary” for N. neesiana invasion and it was “invading the margins of swamp depressions and spreading 

out from those” with most infestations in areas where Themeda cover was sparse. She observed that “It grows back more quickly 

than other perennials after burning and [cleistogenes] ... may partially escape burning”, that it “aggressively colonise[d] the intertussock 

spaces” and initially “grows faster than native species”. Liebert (1996 p. 8) noted that it “quickly invades disturbed 

soils” resulting from revegetaion programs, while a report by Bob Bates (Jessop et al. 2006 p. 108) observed that it wa: “able to 

become established on even the hardest bare sites on disturbed ground”. 

Despite inclusion amongst the few exotic perennial grasses listed as a key threatening process in NSW, N.neesiana is not listed 

by Coutts-Smith and Downey (2006) as posing a threat to threatened biodiversity in NSW. However this indicates a failure both 

of their literature review technique (for Ens (2005) had previously stated that N. neesiana “threatens the ecological integrity of 

affected natural ecosystems”) and the administrative process of threat identification in NSW, and also reflects the general lack of 

integration of weed impact literature. A poor historical linkage between biodiversity conservation and invasive species 

management is also to blame (Downey and Cherry 2005). 

91

The greater height and density of N. neesiana swards than those of native grasses has been considered to make it a fire hazard 

(Liebert 1996), although no directly comparative quantitative data appears to have been published. Grasslands dominated by T. 

triandra are green in summer, while N. neesiana ceases to grow in summer after producing a large quantity of stems in spring 

and early summer. There are therefore good a priori reasons to suggest that N. neesiana might alter fire regimes by changing the 

seasonal distribution of fuel. Comparisons of total biomass of invaded and natural T. triandra grasslands during a range of 

seasons could easily determine this. 

Four threatened plant species are listed by ARMCANZ et al. (2001) as threatened by N. neesiana: Sunshine Diuris Diuris 

fragrantissima D.L. Jones and M.A. Clem., Small Milkwort Comesperma polygaloides, Plains Riceflower Pimelea spinescens 

and Button Wrinklewort Rutidosis leptorhynchoides. More details for D. fragrantissima are provided by Webster et al. (2004) 

and Vidler (2004) and for R. leptorynchoides by Humphries and Webster (2003). However the national strategic plan 

(ARMCANZ et al. 2001) fails to “identify the biodiversity at risk in a manner that can be used to deliver effective management” 

and contains no specific section on impact minimisation (Downey and Cherry 2005 p. 42). 

In summary, very little is known about the impact of N. neesiana on biodiversity and what little is ‘known’ appears to be largely 

based on simple correlative observations without adequate scientific study. Suggestions that the impacts are major or 

catastrophic appear to be founded on the rapid proliferation and high cover of the plant in native grasslands under conditions that 

have been poorly documented and in which the supposed impacts may be due to disturbance. Possibly N. neesiana is basically 

similar to a native grass and replacement of native grasses by N. neesiana may have little biodiversity impact. In the following 

section the atttributes of the temperate native grasslands of south-eastern Australia are examined and the effects of a range of 

disturbances on N. neesiana and the native grasses are discussed in more detail. 

92

GRASSLANDS 

Grasses: “... the most important single family of ogranisms in the world of life ...” 

G. Ledyard Stebbins (1986 p. 360). 

This section provides general descriptions of Australian temperate grasslands and their biodiversity, including their 

palaeoecology and historical and current disturbance and management regimes, and the features and attributes that make them 

vulnerable to N. neesiana invasion 

Definitions of grassland 

Grasslands are plant communities structurally dominated by grasses and without trees (Mott and Groves 1984, McDougall 1999). 

One criterion for a grassland that is widely used in Australia is

Oligocene-Miocene boundary, c. 36 mybp, while in the eastern Murray-Darling basin fire appears to have been present through 

much of the Tertiary although markedly increasing in the late Miocene (Kershaw et al. 1994). In the south-east, grassland 

became more dominant during the Pleistocene (3 mybp +), and may have been as widely developed as today by the late 

Pleistocene (Jones 1999a). 

Fossil grass pollen has been identified only to family level (Martin 2004). The fossil record in Australia for the period of greatest 

interest from the Miocene through the Pleisocene is “very fragmentory” and often difficult to interpret (Kershaw et al. 1994 p. 

299). 

The Quaternary period 

Grassland has dominated much of Australia during much of the last 2 million years (the Quaternary period), expanding during 

long, cooler, relatively dry periods and contracting during warmer, wetter interglacial periods (Kirkpatrick et al. 1995, Benson 

and Redpath 1997, Keith 2004). On the western plains of Victoria more substantial rainforest occurred during interglacials in the 

mid-late Pleistocene than in the early Pleistocene, the formation expanding from refugia such as the Otway Ranges (Kershaw et 

al. 2000). Glacial-interglacial oscillations occurred throughout the Quaternary and selected for taxa tolerant of drier, cooler 

climates and repeated climatic change, “linked to disturbances including wildfires prior to and coincident with the arrival of 

humans” (McGowran et al. 2000 p 449). Little speciation of all terrestrial taxa is believed to have occurred in the Quaternary – 

instead species shifted their distribution in response to the dramatic climate changes (Kershaw et al. 2000), and grasslands as a 

whole would have undergone wide geographical shifts in their extent and location (Jones 1999a), and probably in their 

composition, since “it is very clear” from the Tertiary fossil record “that taxa, not communities, migrate” (Martin 1994). Hope 

(1994) however argued that the repeated, rapid destruction of ecosystems due to climate shifts caused extinctions, range 

fragmentation and rapid speciation, e.g. of Acacia spp. 

An “extensive grassland steppe vegetation” with Casuarinaceae as the tree component occurred from the late Pliocene-early 

Pleistocene (c. 2.5 mybp +) and has “no identified modern analogue” (Kershaw et al. 2000 p. 494). This inland vegetation had 

high levels of Poaceae and Asteraceae (Hope 1994) and a major Asteraceae component, with the form taxon name 

Tubulifloridites pleistocenicus, does not correspond with any extant Australia daisy and may have been similar to North 

American Ambrosia and African Stoebe (Kershaw et al. 1994). Evidence of the dominance of Asteraceae, probably woody, c. 1 

mybp has been obtained from a site on the edge of Port Phillip Bay, Victoria, and pollen data from the Pejark Marsh volcano in 

western Victoria covering the period 1-0.7 mybp indicates Poaceae and Asteraceae co-dominance with Casuarinaceae as the 

main trees (Kershaw et al. 2000). Extensive grasslands existed in north-western Australia and possibly in central Australia in the 

early Pleistocene (Kershaw et al. 2000). Late Pleistocene data from north-eastern Queensland indicates a major increase in 

Poaceae c. 175 kybp indicates “a major expansion of grassland prior to substantial increases in eucalypts and charcoal" (Kershaw 

et al. 2000 p. 500). A number of currently dominant taxa, notably Pooidae (Poa spp.) and T. triandra entered Australia from the 

north during the Pleistocene, but the number of recent migrant taxa is small (Jones 1999b). 

The Holocene (Recent) 

Lower rainfall and temperatures associated with the last ice age, which ended about 10 kybp, probably saw much of southeastern 

Australia dominated by grasslands and grassy woodlands (Jones 1999a).The many vegetation histories that have been 

established from the fossil record indicate that at the peak of the last ice age, c. 18 kybp, most of south-eastern Australia was 

dominated by largely treeless vegetation, a very open, cold, dry steppe of Poaceae and Asteraceae with annuals, perennial 

geophytes and shrubs, and that open eucalypt woodlands were widespread across the Bassian plain (Hope 1994). Montane 

grasslands, such as those in the Monaro region of New South Wales and the midlands of Tasmania are the best current analogue 

for these formations (Hope 1994). The current interglacial is unusally warm compared to 85% of the Quaternary, and the 

lowland grassland communities present have therefore formed largely from the species set that survived the cooler, drier glacials 

(Hope 1994). 

Australian grassland formations 

Four main types of grassland occur in Australia: 1. tropical summer rainfall coastal grasslands dominated by Sporobolus and 

Xerochloa spp., mainly in the Northern Territory and north-western Queensland, 2. Triodia arid hummock grassland of the 

continental interior; 3. Astrebla (Mitchell grasss) grasslands in areas with 200-500 mm average annual rainfall, mainly in 

summer, in western Queensland, inland northern New South Wales, the Northern Territory and northern Western Australia; and 

4. subhumid grasslands of eastern Australia (Specht 1970, Groves 1979, Mott and Groves 1984, Groves and Whalley 2002, 

Benson 2004). The latter has been subdivided into three types: 4a. tropical subhumid grasslands of eastern and northern 

Queensland, dominated by Dichanthium and Eulalia and sometimes by Bothriochloa and Heteropogon; 4b. temperate grasslands 

of New South Wales, Victoria and South Australia, dominated by Themeda triandra, Poa spp., Austrodanthonia and Austrostipa 

and 4c. subalpine tussock grasslands of wet tablelands and montane areas of the south east, dominated by Poa spp. and 

Austrodanthonia (Groves 1979, Groves and Whalley 2002). 

Moore (1993) described a substantially different set of subhumid grassland types present in south-eastern Australia: Temperate 

Tallgrass dominated by Themeda, Poa and Dichelachne, Temperate Shortgrass dominated by Austrodanthonia, Austrostipa and 

Enneapogon, Subalpine Sod Tussockgrass dominated by Poa, Themeda and Austrodanthonia, Xerophytic Midgrass (Southern) 

dominated by Austrostipa, Chloris and Aristida, and Saltbush Xerophytic Midgrass dominated by Atriplex, Maireana and 

Austrostipa. The Temperate Shortgrass communities are largely derived from woodlands and consisted mainly of taller warm 

season grasses: in the wetter areas Themeda triandra, Austrostipa bigeniculata (Hughes) S.W.L. Jacobs and J. Everett and Poa 

labillardieri Steud. and in the drier areas Austrostipa aristiglumis and Themeda avenacea (F.Muell.) Maiden and Betche. 

Temperate Tallgrass is a formation corresponding with disturbed forests and heaths and defined by T. triandra, P. labillardieri 

and Dichelachne spp. (Moore 1993). 

Small areas of other grassland types occur including maritime grasslands (on beaches, headlands etc.) and grasslands associated 

with river margins and freshwater wetlands (reed beds, meadows, cane grass swamps, etc.) (Moore 1993, Carter et al. 2003, 

Benson 2004). Only the humid, temperate, non-alpine grasslands are considered here, since the others are thought to be less 

94

susceptible to N. neesiana invasion. For example “few weed species have successfully established and persisted” in the 

Australian Alps, the number declining with increasing altitude (McDougall and Walsh 2007 p. 44). Excluded formations include 

aforementioned wetland formations and maritime grasslands, the Western Slopes Grasslands of NSW including Moore’s (1993) 

Xerophytic Midgrass (Southern) and Saltbush Xerophytic Midgrass, Eastern Victorian highlands grasslands (e.g. Lake Omeo), 

high altitude grasslands of the Monaro region of NSW, and alpine grasslands and meadows. The grasslands that once existed 

around the shores of Lake Omeo (Benambra area) were similar to those of the Southern Tablelands and were dominated by T. 

triandra (Lunt et al. 1998) or Austrodanthonia (Kirkpatrick et al. 1995), but show clear affinities with high altitude grasslands 

(Carter et al. 2003). According to Kirkpatrick et al. (1995) the Lake Omeo grasslands are probably derived from Eucalyptus 

pauciflora Sieber ex Spreng. woodland. 

The natural occurrence of temperate grasslands in south-eastern Australia was determined by relatively low rainfall (in the range 

of 350-1000 mm mean annual, mainly 500-600 mm), flat to undulating topography, and soils that were poorly drained, heavytextured 

and moderately to highly fertile (Mott and Groves 1994, Sharp 1997, Jones 1999b, Lunt and Morgan 2002). The 

geological parent materials produce soil with a high clay content (Jones 1999b). Heavy clay or clay loam soils, and a relatively 

dry, cold climate “are prerequisites for grassland vegetation worldwide” (Benson and Redpath 1997 p. 307). South-eastern 

Australian grasslands generally occur on younger soils that have not been heavily leached or become lateritic or infertile, and 

thus tend to occur where the parent rocks have weathered in situ and there is little erosion, and the soils are relatively fertile 

(Jones 1999b). Most grassland soils have a high water content but low water availability because the clay minerals bind water, 

and there is little pore space (Jones 1999b). 

Some temperate Australian grasslands occurred in the drier areas down to c. 250 mm annual rainfall on the arid margins of 

chenopod shrublands (Mack 1989) although these are probably better categorised as semi-arid formations (Carter et al. 2003). In 

the Australian Alps, grasslands are generally found on deep humus soils in valley bottoms subject to cold air accumulation and 

frosts (McDougall andWalsh 2007). 

Causes of treelessness 

Heavy-textured clay substrates dry out and crack deeply in summer (c. 13% on an areal basis according to Patton 1935) and this 

may prevent the establishment of trees and shrubs, as may fire, which occurs ubiquitously (Groves and Whalley 2002). Moore 

(1993 p 353) explained treelessness to be the result of “shallow penetration of water in environments with relatively low rainfall 

and high evaporation, where tree seedlings would be subjected to intense competition for surface-rooted grasses and other 

herbaceous species” and in wetter areas “poor aeration following temporary waterlogging after winter rains”. Patton (1935 p. 

175) argued that trees can be established artificially on the Victorian basalt plains by “opening up the ground and ... destroying 

the native vegetation”, so concluded that the native vegetation itself prevented the establishment of tree seedlings. Elsewhere 

however he argued that heavy textured soils inhibit free entry of water, have slow percolation and bad aeration, that summer 

cracking leads to deep drying, and that deep rooting is difficult, so it is the physical characteristics of the soil, accentuated by the 

evenness of contour in the plains that determines their treelessness (Patton 1930). Barlow and Ross (2001) argue that multiple, 

confounded, variable factors are responsible, with soil charcteristics, particularly drainage, the most important and subsidiary 

influences from climate and fire. In a trial in the South American pampas competition from tussock grasses prevented 

establishment of half the tree species tested (Aguiar 2005). Kirkpatrick et al. (1995) more or less concurred: tree seedlings are 

largely excluded by competition in dense grassland swards, which may use all the available soil moisture, and severe 

disturbances such a soil digging are required for trees to establish. 

In the Volcanic Plains Grasslands of western Victoria, where ever any other geological formation abuts basalt, trees occur. 

Eucalyptus camaldulensis grows, for example, where granodiorite is exposed or eroded out of the basalt (Patton 1930). The 

granite of the You Yangs provides another notable example. Soil chemistry, as well as structure can determine treelessness: soils 

dominated by sodium, common in northern Victoria, usually lack trees (Jones 1999b). 

Distribution of grasslands in the ACT is determined by accumulation of cold air pockets in valley floors, creating conditions that 

are too cold to permit growth of trees and shrubs (Chan 1980), but low rainfall, heavy-textured soils and the legacy of aboriginal 

fire regimes are also important (Sharp 1997). Very low temperatures associated with nocturnal temperature inversions have been 

invoked as the cause of treelessness in subalpine valleys (Moore 1993). 

Lowland grasslands 

The term “lowland” grasslands, considered to include those formations at altitudes below 1000 m, has been widely used (Lunt 

1991, McDougall and Kirkpatrick 1993, Kirkpatrick et al. 1995, Sharp 1994, Lunt et al. 1998). Prior to European occupation 

they probably covered c. 2 million ha, of which “perhaps” 10,000 ha survived in “more or less natural” condition by 1992 

(Kirkpatrick et al. 1995 p. 8). Classifications of lowland grasslands have varied between authors and government authorities and 

have not always coincided. The New South Wales classification of Benson (2004) has four classes covering sub-humid 

grasslands, including the Temperate Montane Grasslands, found in the south-east, which include high altitude formations 

dominated by Poa spp. Such grasslands, including those on the Monaro Plains of NSW, are generally excluded from the 

‘lowland’ category by most authors. The change from lowland to montane and alpine grasslands is gradual and clinal, so 1000 m 

is an arbitrary delimination (see discussion in Carter et al. 2003). 

Secondary and derived grasslands 

‘Secondary grassland’ is derived from other vegetation formations, often in Australia from grassy woodland in which trees have 

been cleared and on which livestock are grazed (Moore 1993, Mott and Groves 1994, Groves and Whalley 2002). Temperate 

grassy woodlands are also a threatened vegetation type (Benson 2004, Keith 2004, McIntyre and Lavorel 2007) and their grassy 

stratum may be indistinguishable, for practical conservation purposes, from that of adjoining natural grassland (Carter et al. 

2003, Keith 2004). Indeed Moore (1993 p. 343) considered that most of his Temperate Shortgrass communities, including most 

of the Western Basaltic Plains of Victoria, the NSW Riverina and Gippsland Plains, to be “mostly the understoreys of temperate 

woodlands modified by clearing and grazing by livestock and rabbits”. The grasslands derived from temperate grassy woodlands 

are frequently also very similar floristically to natural grasslands ( Mott and Groves 1994) and can have the same ecological 

95

functioning as natural grassland (Groves and Whalley 2002). In some areas derived grasslands are the only grassland remnants 

(Kirkpatrick et al. 1995). 

In the absence of grazing, secondary grassland can revert back to woodland or shrubland (Groves and Whalley 2002, Benson 

2004). O’Dwyer (1999 p. 324) suggested that livestock grazing had “prevented regeneration of a shrubby woodland” in many 

derived Victorian grasslands. Distinguishing anthopogenic grasslands is difficult or sometimes impossible because of continuous 

human influences over long periods (Wheeler et al. 1999, Carter et al. 2003), as well as long term climatic fluctuation and large 

scale natural disturbances which can alter plant dominance patterns. Many grasslands in South Australia are derived formations 

(Davies 1997), and the Riverine Plains Grasslands of south-west NSW were once dominated by Acacia pendula A.Cunn. ex 

Don. and Atriplex spp. (Benson 2004, Keith 2004). 

The term ‘secondary grassland’ is generally used to designate formations that are a product of post-colonisation management, or 

the lack of it. The extent to which grasslands are cultural landscapes resulting from aboriginal land management has generally 

been treated as a separate issue in Australia, despite overlap between pre- and post colonisation management regimes and species 

invasions, and despite the “persistent legacies of past human impact on species composition ... structure, disturbance regimes and 

soil conditions” (Froyd and Willis 2008 p. 1729). 

Grassland distribution 

Australian grasslands have mostly been well described botanically, but poorly mapped, in particular in terms of their historical 

distribution (Groves 1979). Moore’s (1993, Fig. 13.1) map, which supposedly shows the distribution of “herbaceous 

communities”, derived from other vegetation communities or not, that were then “used for livestock production” and “composed 

essentially of native species” (Moore 1993 p. 315), seems particularly defective for south-eastern Australia. For example large 

areas of continuously forested land in eastern Victoria that have never been managed as grazing land are depicted as “Temperate 

Tallgrass” grassland. His 1993 and earlier maps have probably caused much confusion over a long period. 

Remnants continue to be discovered in areas thought previously to have had no grasslands (Cook and Yugovic 2003, Sinclair 

2007). Major vegetation mapping in the last 15 years has greatly improved the situation for current vegetation. The pre-European 

distribution of temperate grassland in south-eastern Australia was recently mapped by Lunt and Morgan (2002), although 

Kirkpatrick et al. (1995 p. 15) thought it was “no longer possible to map much of the pre-European grassland distribution with 

any accuracy” because so much was rapidly and completely destroyed. Lunt and Morgan (2002) found that temperate grassland 

was one of the dominant vegetation classes, covering extensive areas, but noted that the ‘grassland’ status of some areas, 

particularly in lower rainfall regions, was in dispute (e.g. the NSW Riverina), mainly on the basis that they were recently derived 

from other vegetation types as a result of land use. The extent to which these grasslands are ‘natural’ or derived has been an 

ongoing area of argument. Kirkpatrick et al. (1995) considered that the riverine plains grasslands are more correctly called 

herblands, being often not dominated by grasses, and Groves (1979) considered that temperate grassland “was never very 

widespread”. Benson and Redpath (1997) argue that too much emphasis on the records of early explorers may have resulted in 

an exaggerated conception of the extent of open grassy vegetation at the time of European settlement, in part because the 

explorers were typically tasked with finding new grazing lands and preferentially travelled in country that was easier to traverse 

on horseback. Carter et al. (2003) avoid the problems of interpretation of historical information and speculative argument that 

characterise the debate, by defining a data cut-off of 1982, which excluded only the most recently derived grassland formations. 

The somewhat legalistic definition of Natural Temperate Grassland provided by Carter et al. (2003), adapted from McDougall 

and Kirkpatrick (1993), also overcomes some of the other terminological difficulties and areas of scientific dispute. It is a broad 

vegetation class defined inter alia as being dominated by tussock grasses of the genera Austrodanthonia, Austrostipa, 

Bothriochloa, Chloris, Enteropogon, Poa or Themeda or by Lomandra (Xanthorrhoeaceae), with

Floristic composition, vegetation structure and ecology 

“Unfortunately, there exists no really satisfactory description of the vegetation when the grazier first began to exploit the 

country” (Wadham and Wood 1950 p. 87). Eighty years after exploitation commenced, Schimper (1903 p. 503) noted that there 

were “no available descriptions of the extensive savannahs and steppes of the interior of New South Wales and Victoria”. Some 

decades later, McTaggart (1936), as part of an Australia-wide ‘survey’ of pastures, provided a description of the open grasslands 

(in which he included “savannah”, now grassy woodland) of south-eastern Australia, but it is evident from his text and the listed 

information sources that no detailed compositional and structural knowledge then existed for most areas. 

The earliest detailed studies of Australian grassland vegetation, made in the decades preceding 1950, provide “no guide to the 

proportions of the various species, or the composition of the vegetation from the standpoint of grazing value” of the areas before 

grazing commenced, but it is known that “profound” changes had occurred, particularly where rainfall was low or highly 

variable (Wadham and Wood 1950 p. 87). By 1875, for example, the grasslands of South Australia had “mostly disappeared” by 

conversion to cereal growing (Schimper 1903 quoting Schomburgk 1875). The “lack of adequate description before alteration 

occurred” (Jones 1999b p. 29) is an ongoing problem. Lunt (1990a p. 47) thought that an accurate representation was “almost 

impossible”, while Lunt et al. (1998) considered “any reconstruction of the original vegetation must be somewhat speculative, 

particularly at a detailed level”. The pre-European composition is still considered to be “poorly understood” (McIntyre and 

Lavorel 2007) but pre-European grassland was probably more diverse than current reference areas, and many species have 

presumably been eliminated or greatly depleted in most areas (Sharp 1997). 

A similar situation exists with the temperate grasslands of South America. One of the first floristic surveys of pampas vegetation 

was made by L.R. Parodi in 1930 by which time so-called “virgin grasslands” were rare (Soriano et al. 1992 p. 381) and after 

which agricultural development greatly intensified. 

Early historical descriptions 

One of the earliest Australian descriptions is Schomburgk’s (1875, quoted by Schimper 1903 pp. 504-505) portrait of South 

Australian grasslands, in which the grasses mentioned are already a mixture of native and introduced species, although the forbs 

appear to be entirely native: “The plains near the coast ... the soil mostly fertile, extending often to the sea ... The grasses consist 

of more nourishing kinds ... Poa, Panicum, Festuca, Agrostis, Aira, Andropogon, Cynodon, Stipa, Pennisetum, Bromus, 

Eriachne, Anthistiria [Themeda], Hordeum ... The banks of the rivers and creeks, which mostly cease running druing the 

summer, are lined with majestic gum-trees ... the shrubs extending more or less on the plains, according to the nature of the soil 

... the appearance ... has during the dry season, the ... sunburnt yellow character ... destitute of all green herbage ... In the month 

of May the rainy season generally commences ... a few showers change the aspect ...into a green and beautiful carpet ... in a few 

days the plains appear clothed with luxiarant verdure ... With the grasses are also recalled to new life Ranunculus ... Oxalis ... 

Hypoxis glabella ... Drosera ...Wahlenbergia gracilis ... Anguillaria [Wurmbea]... Stackhousia ... Every week adds new colours 

to the beautiful carpet ... Kennedya [sic] prostrata...Swainsona procumbens ... and S. lessertiaefolia [sic] ... Thysanotus ... 

climbing up the dry grass stalks ... Compositae are seen blooming over the plains in all colours ... But by the middle of 

November the number of flowering plants already lessens considerably, the annual grasses and other herbaceous plants begin to 

dry up, droop and disappear, and in January the grass land resembles a ripe thinly-sown cornfield, and we find only ... a few 

plants of Convolvulus erubescens, Lobelia gibbosa ... and Mesembryanthemum australe [?Disphyma crassifolium] ... The seeds 

of the annual plants have been scattered, perennial herbage returned to its dormant state ... and the plains have during the summer 

months a dismal, dried-up appearance”. 

The earliest study of note for Victoria is Sutton’s (1916-1917) “Sketch of the Keilor Plains flora” at the drier, eastern end of the 

Victorian volcanic plains. He provided a valuable census of the vascular flora that included species from neighbouring 

vegetation associations, but deliberately excluded all of the exotic species. Patton (1935) provided a list of vascular plants for 

Victorian basalt plains grasslands, discussed their structure, composition and seasonal dynamics and the climate and soils. 

Contemporary studies 

A much greater range of more detailed studies have taken place in the last fifty years, commencing with investigations such as 

those of Willis (1964) and Groves (1965). These describe grassland generally dominated by perennial tussock grasses including 

T. triandra, Poa, Austrodanthonia and Austrostipa species, with the latter two genera being more dominant in drier areas (NSW 

Riverina, Northern Plains of Victoria and Northern Lofty region of South Australia) (Kirkpatrick et al. 1995, Sharp 1997, Lunt 

and Morgan 2002). Bothriochloa, Enteropogon and Chloris may also be dominant grasses, while Lomandra is the major 

dominant in some South Australian formations (Carter et al. 2003). 

Based on readings of historical records T. triandra is thought to have been the dominant species across much of the plains before 

European occupation, although this is based on the assumption that the vernacular term “kangaroo grass” was applied only to T. 

triandra (Lunt et al. 1998). T. triandra was almost certainly the most widespread and dominant grass (Groves 1965, Mack 1989, 

Moore 1993, Kirkpatrick et al. 1995) and is considered the major dominant in Victoria, but is replaced by Poa spp.,usually Poa 

labillardieri Steud. on wet sites and in higher rainfall areas (Willis 1964, Moore 1993, Entwisle et al. 1994). Austrodanthonia 

and Austrostipa species were the dominant grasses in drier areas including the Murray River plains grasslands and in parts of 

South Australia (Kirkpatrick et al. 1995). Moore (1993) considered Austrostipa bigeniculata was the principal T. triandra 

associate in drier sites. Sutton (1916-1917) found Austrodanthonia penicillata (Labill.) H.P. Linder to often be the dominant in 

drier areas of the Keilor Plains with Poa, Austrostipa setacea (R.Br.) Jacobs and Everett, A. semibarbata (R.Br.) Jacobs and 

Everett and Dichelachne crinita (L.f.) Hook. f. more prominent in moister areas along with “Panicum” spp. Willis (1964) found 

that Austrodanthonia spp. became less prevalent in wetter areas. A large proportion of these grasslands have been modified by 

grazing and became dominated by shorter grasses such as Austrodanthonia auriculata (J.M. Black) H.P. Linder, A. carphoides 

(Benth.) H.P. Linder and Austrostipa scabra (Lindl.) S.W.L. Jacobs and J. Everett (Moore 1993). Shrubs are rare and 

substantially absent (Entwisle et al. 1994, Lunt et al. 1998). Except for T. triandra, the dominant grasses have the C 3 

photosynthetic pathway and grow mostly in spring and autumn (Groves and Whalley 2002). Most have seeds with an ‘after 

ripening’ period (Groves and Whalley 2002) which presumably acts to prevent germination in the dry season (summer) and 

delay it until adequate soil moisture is more likely to be available for seedling establishment. Annual native grasses are rare. 

97

Groves (1965) found that a very high proportion of the biomass (62-90% varying seasonally), in a T. triandra grassland at St 

Albans consisted of the dominant grass. Usually the Themeda tussocks are mostly widely spaced (10 cm or more apart) and grass 

cover may be only 30-50% (Lunt et al. 1998, Lunt and Morgan 2002), but they can be more closely spaced with a canopy of 

loosely interlaced leaves (Carter et al. 2003). Such cover values of dominant grasses appear typical for mid-latitude grassland: 

for example McArdle et al. (2004) found that the invasive grass Hyparrhenia hirta had an average cover of 65.5% in areas where 

it dominated in open woodlands on the North West Slopes of New South Wales, and caespitose grasses and Cypreacace 

accounted for >50% of cover in one southern Brazilian grassland (Overbeck and Pfadenhauer 2007). Projective foliar cover of 

grasses in fire adapted grasslands is obviously highly dependent on the time since fire. One year after fire in T. trianda grassland 

at Derrimut, bare ground was c. 40%; after 2 years this had reduced to c. 25%; and after 3 years had fallen to

proportion of the flora then present on the Victorian basalt plains (including non-grassland formations) had distributions outside 

the plains, many being ecological ‘wides’, with only c. 10 spp. restricted to the region. Similarly many plants of the New 

England Tablelands grasslands have very wide ranges, with the flora as a whole having low habitat specificity (McIntyre and 

Lavorel 1994a), and it is generally agreed that only a very small number of vascular species are restricted to Australian 

temperate grasslands (Kirkpatrick et al. 1995). The few native annuals present are often very small plants which tend to occur in 

wet depressions (Lunt et al. 1998) and are ephemeral, growing in early spring (Sharp 1997). 

Phenology 

Most of the plant species in south-eastern temperate grasslands flower mainly in spring, and a substantial proportion also in 

autumn, “presumably” responding to day length and day temperatures of c. 20ºC (Groves and Whalley 2002), along with soil 

moisture. A high proportion of the perennials flower and fruit in spring-early summer and have no living parts above ground for 

much of the year (Lunt et al. 1998). All the Orchidaceae fit this pattern, with no above-ground living material during the warmer 

months (Smith et al. 2009). Patton (1935) considered October-November to be the peak flowering period for grasses and other 

plants in the Coburg to Melton area of the Victorian basalt plains. He and Willis (1964) noted that the major flowering period 

corresponds with the time that evaporation begins to exceed precipitation. Sutton (1916-1917) observed that many species were 

in flower long before the end of winter (31 August). Groves (1965) found that nearly half the species present at St Albans 

flowered in October and November, with similar patterns for exotic and native species. Most of the grasses in the Victorian 

basalt plains flower in November (Willis 1964). Yellow-flowered daisies were a prime feature in spring (Sutton 1916-1917) but 

are now greatly depleted. Chan (1980) recorded the flowering periods of 61 native species at Yarrumundi Reach, ACT, and 

found that 33 species flowered in September, 51 in October, 49 in November, 32 in December, 28 in January and 7 or fewer 

species in every other month. Exotic species also showed a marked flowering peak in October and November. 

As moisture levels decline the flowering period rapidly ends and “most of the vegetation passes into a resting stage until the 

following autumn” (Patton 1935 p. 172), being practically dormant from December to April (Willis 1964), with vegetative 

growth recommencing after autumn rains and continuing slowly during the winter. Summer rains may stimulate growth of many 

species. Morgan (1995b) for example observed rapid growth of juvenile Rutidosis leptorhynchoides after summer rains. 

Information on the germination and establishment periods of most species appears to be lacking , possibly because recruitment 

events are rare. Morgan (1995b) found that Rutidosis leptorhynchoides seeds germinated 8-12 days after the first major autumn 

rains and continued to germinate through until early July. 

Plant species richness 

The richest terrestrial plant communities in Australia include the kwongan of south-western Australia with up to 103 vascular 

species in 0.1 ha and the herb-rich woodlands of western Victoria with up to 96 spp. in 0.1 ha, 93 species in 128 m 2 , and 45 spp. 

m -2 (Lunt 1990d). The richest communities in the world have traditionally been considered to be the European chalk grasslands, 

with a maximum of 54 spp. m -2 (Lunt 1990d). One recently burnt grassland in the vicinity of Porto Alegre, southern Brazil, had 

maxima of 28 vascular spp. per 0.25 m -2 , 34 per 0.75 m -2 and approximately 450 spp. in 220 ha (Overbeck et al. 2007). 

Vascular plant species richness in Australian temperate grasslands shows considerable variation across the range of spatial 

scales, but native richness is strongly related to the historical disturbance regime, particularly burning and grazing (Kirkpatrick et 

al. 1995, Dorrough et al. 2004). At the regional scale, Willis (1964) considered the Victorian Basalt Plains flora to be 

floristically ‘deficient’ in comparison with other regions of the State. However communities in this region can nevertheless be 

species rich at a small scales (Morgan 1998e). Sutton (1916-1917 p. 117) considered that the vegetation of typical Keilor Plains 

grasslands contained “some hundred or more species”. 

Patton (1935) calculated a species/area curve for basalt plains grassland and found an average of c. 8-9 species present in 1 m 2 , 

rising to c. 14 spp. in 5 m 2 , and to c. 17 spp. in 10 m 2 with a total of 45 spp. in all quadrats, and 73 species comprising the whole 

flora. A relatively high proportion of species occurred infrequently, or were only very sparsely present. Stuwe and Parsons 

(1977) found species richness of c. 12-18 m -2 in Victorian basalt plains grasslands. Kirkpatrick et al. (1995) considered it 

common for more than 40 native species to occur in an area of 100 m 2 . Morgan (1998e) recorded 22 spp. in 0.25 m -2 and 27 spp. 

m -2 in some species-rich Victorian basalt plains remnants and total species data for two ungrazed roadside sites – 71 native and 

22 exotic species in annually burnt site and 67 native and 19 exotic spp. in a bienially burnt site. Other data includes Groves 

(1965) study along the railway line at St Albans - 101 spp. including 64 natives; Evans St., Sunbury, c. 11-17 species m -2 

(Morgan 1998d) and c. 103 species present in quadrats with 48 additional spp. recorded in 3.5 ha (Morgan and Rollason 1995); 

Derrimut Grassland Reserve, a 154 ha “species-poor” remnant with “a long history of domestic stock grazing and little or no 

recent burning” (Morgan 1998c) 102 native and 78 exotic species in the above-ground vegetation (Lunt 1990a) with 1 additional 

native species and 3 exotics found only in the soil seed bank (Lunt 1990b); five Victorian volcanic plains T. triandra grasslands 

with a range of historical fire frequencies (1 y, 3 y, >10y) and a total area of 32 ha surveyed in 1993-95 (Morgan 1998c) – 61 

native and 30 exotic spp. in the vegetation, 32 native and 28 exotic in the soil seed bank of which 11 were not present in the 

vegetation; the same five sites plus an additonal one, all surveyed in August 1998 (Morgan 2004) – a total of 69 spp. from a 

single 150 m 2 quadrat at each site; New England tablelands an average of 19.9 native species in 30 m 2 , with a maximum of 28 

and minimum of 1 (McIntyre 1993); native pasture on the New England Tablelands (Chiswick) 124 species in 2.4 ha of which 26 

were grasses, 11 Cyperaceae + Juncaceae and 87 other forbs, with a range of 4-26 spp. and means of 11-16 spp /0.25 m 2 in 

grazed areas and range of 1-9 and means of 3-4 spp./0.25 m 2 in grazed areas (Trémont 1994); 39 natural grasslands in the ACT - 

191 spp., 56% forbs, including 10 native grass spp., with a range of 23-56 spp. in ten 1m 2 quadrats (Sharp 1997). On the Monaro 

Tablelands of New South Wales Dorrough et al. (2004) found mean plant species richness of 11.7, 12.6 and 15.7 species 0.25 m - 

2 in native pastures, on roadsides and in travelling stock routes respectively, with the richness of native species being 5.0, 7.8 and 

8.6 and of exotic species 6.7, 4.8 and 7.1 respectively, with a total of 120 species in 100 m 2 . Kirkpatrick et al. (1995) calculated 

a total of 771 vascular species in 77 families in the south-eastern lowland grasslands. In high-richness sites natives grasses are 

always the dominant species, whereas low-richness sites may be dominated by exotic or native Poaceae (New England 

tablelands, McIntyre 1993). The grasslands of the Victorian Volcanic Plains were generally floristically rich (DNRE 1997) but 

99

considerably less diverse than those of herb-rich grassy woodlands in the Grampians and Langi Ghiran areas of western Victoria 

studied by Lunt (1990d). 

Areas of grasslands on sedimentary soils and soils derived from granites, which are generally less fertile, are less productive so 

grow much smaller grass tussocks and consequently tend to have greater floristic richness (Stuwe 1994). 

Productivity 

Little attempt appears to have been made to quantify the productivity of the temperate native grasslands of south-eastern 

Australia, although there has been considerable interest in the amount of above-ground biomass accumulated by dominant 

grasses, particularly T. triandra. In minimally disturbed (ungrazed and unburnt) communities the matrix species (dominant and 

subdominant grasses) grow large, accumulate abundant litter and suppress the intertitial forbs (Trémont and McIntyre 1994). 

Groves (1965) undertook a seminal study of a T. triandra grassland at St Albans. He determined the above-ground biomass of T. 

triandra versus all other species and the root biomass for all species combined at 3-6 weekly intervals for 15 months from 1962 

to 1964. The grassland had been burnt in the summer of 1961-62. T. triandra constituted the major component of above-ground 

biomass througout the period, with a mimimum of 62% and a maximum of 90%. The standing biomass one year after fire was 

about two thirds of that two years after the fire, but nevertheless fell to approximately the same level in summer in successive 

years. Standing biomass fell rapidly to very low levels in mid summer after a late spring- early summer peaks. Maximum levels 

exceeded 3000 kg ha -1 . Much dead biomass was apparently broken down or moved off site. Only a small fraction of root biomass 

penetrated below 15 cm. The below-ground biomass followed similar trends to that above ground in one summer and not the 

other, but consistently peaked in late spring and early summer. The maximum was c. 8000 kg ha -1 . In autumn, rapid growth 

resumed, many seedlings appeared and a carpet of moss developed. Standing biomass in April was etimated at 1025 kg ha -1 . 

During winter, growth was inhibited and many shoots of T. triandra died, but Austrodanthonia spp., Dichelachne crinita and the 

exotic annual grass Rostraria cristata (L.) Tzvelev grew vigourously. In spring rapid growth of T. triandra resumed and the 

native forbs Eryngium ovinum, Plantago varia and Wahlenbergia stricta grew. Growth ceased when soil moisture fell to the 

permanent wiliting point. T. triandra growth rate peaks from October to early December and growth continued into summer if 

there was adequate soil moisture, and there was a minor peak in autumn. 

Morgan (1998e) calculated annual peak standing biomass for two T. triandra grasslands in the Victorian Volcanic Plains: 1300 

kg ha -1 in December for a grassland at Derrinallum burnt annnually in February, and 2600 kg ha -1 for a grassland at Karrabeal, 

burnt biennially in summer, two years after burning. Annual dry matter production in improved pastures in the New England 

region of New South Wales varies from c. 8000 kg ha -1 in drought years to 16,000 kg ha -1 (Davidson 1982). Estimates for net 

above-ground primary productivity of Flooding Pampa grasslands include 5320 kg ha -1 , with green standing crops of 1550-2220 

kg ha -1 (Soriano et al. 1992). 

Dynamics 

The dynamics of temperate Australian grasslands cannot be adequately explained within the classical botanical framework of 

succession, and there is no climax formation (Mott and Groves 1994). If the exotic components are disregarded, the composition 

varies little over time, but widely on a patch scale in otherwise uniform areas (Mott and Groves 1994). Morgan (1998e) 

investigated patch-scale (0.01 and 1 m 2 ) dynamics of exotic and native vascular species in Victorian volcanic plains grasslands 

and found a 50% increase in cumulative species richness over 4 years, involving high turnover rates and high spatial mobility of 

species, but little variation in mean species richness. Life-form characteristics were the main determinants of the patterns of plant 

movement: annuals and geophytes tended to have higher turnover and mobility while hemicryptophytes often had low turnover. 

The few species with large, persistent seed banks had high turnover, including exotic annual grasses. High turnover of geophytes 

was somewhat illusory, being explained by their frequent dormancy and failure to produce above-ground parts, but this ‘pseudoturnover’ 

was displayed by c. 30% of species (Morgan 1998e). Such pseudo-turnover might be particularly significant with 

Orchidaceae, which may remain dormant for several years (Smith et al. 2009). About 40% of species had low mobility at the 1 

m 2 scale (Morgan 1998e). 

‘Dispersal limitation’ at a range of scales is a feature of many native grassland systems (MacDougall andTurkington 2007) and 

may in part result from loss of native animals that created safe sites for seedlings and dispersed seed, and low fecundity due to 

loss of native pollinators. Frequent fire has been suggested to be the most important cause of the mobility patterns in Victorian 

volcanic plains grasslands, but climatic variation may be important for some species (Morgan 1998e). Management history and 

‘chance’ appear to be more important determinants of the status of a particular remnant than recurrent major disturbance (Mott 

and Groves 1994). The absence of a successional climax means that the dynamics of Australian temperate grasslands are best 

described by ‘state and transition’ models, with disturbance and management regimes determining the dynamics of the plant 

components. These are discussed in detail below. 

Areas of bare ground at sites with low vascular plant richness are mostly due to exogenous disturbance, while at sites with high 

richness, they are more often the result of constrained production (McIntyre 1993) due to near-complete resource utilisation. 

Sharp (1997) created 1 m 2 areas of bare ground experimentally using glyphosate herbicide in Dry T. triandra and 

Austrodanthonia grasslands and studied colonisation of the gaps for 18 months. Native grass cover had not recovered to pretreatment 

levels after 18 months, while exotic grass cover and richness initially increased, but after 18 months decreased to levels 

similar to those prior to treatment. Native and exotic forb richness and cover was increased. Sharp (1997) also experimentally 

removed litter in areas dominated by native grasses and tested combined treatments of gap formation by herbicides, litter 

removal/retention, and soil disturbance (scarification to 2 cm depth). When plant litter was retained, native forb richness and 

cover were higher than when litter was removed. The opposite response was found for exotic forbs. Soil scarification had no 

significant effects on recruitment of species. 

The dominant native grass can reduce species richness in intertussock spaces (McIntyre 1993). This is a common feature of 

temperate grasslands around the world: in the absence of biomass reduction by fire or grazing, the highly productive dominant 

casepitose grasses accumulate dead leaves and litter and gradually exclude forbs, “irrespective of their habitus” (Overbeck and 

Pfadenhauer 2007). Intertussock spaces disappear in T. triandra grasslands not subject to regular biomass reduction because the 

T. triandra plants accumulate large canopies of dead leaves, which reduce and eventually eliminate bare ground and inter- 

100

tussock space (Stuwe and Parsons 1977, Morgan 1995b, Morgan 1997, Morgan 1998b, Henderson 1999). T. triandra is generally 

the only native plant with high cover and frequency and most native species have low cover and low frequency (Morgan and 

Rollason 1995). Germination of a high proportion of species requires exposure to light, but otherwise there are no specialised 

germination requirements in the majority of species (Robinson 2003). Regular opening of canopy gaps, e.g. by burning, is 

required for forb seedling recruitment (Morgan 1995b, Sharp 1997), but actual removal of vegetation, with associated soil 

disturbance is probably even more effective (Robinson 2003). The dominant tussock grasses may not only command most of the 

space and light but may also starve the intertussock species of moisture and nutrients (Keith 2004). T. triandra has been said not 

to compete directly for resources with the annual (exotic) grasses (Morgan 1994), but agricultural experience suggests that the 

exotic annuals deplete soil moisture prior to the main T. triandra growth period. Germination and establishment of T. triandra 

has also been reported to be unaffected by weeds (McDougall 1989, Morgan 1994 citing Hagon 1977) but there is likely to be a 

threshold above which it will be affected (Morgan 1994). 

None of the native perennial intertussock species in existing native temperate grasslands are obligate seed regenerators, almost 

all being obligate resprouters, or resprouting and with limited seedling production, and mostly able to set, and actually setting, 

seed within 12 months of regeneration (Lunt 1990c, Morgan 1996, Lunt and Morgan 2002). Austral Toadflax Thesium australe 

R.Br. (Santalaceae) a peculiar semi-parasitic semi-shrub, once common in T. triandra grasslands, has a ruderal strategy, is shortlived 

(Keith 2004) and apparently dependent on annual seedling recruitment, but populations no longer exist at lowland 

grassland sites (Scarlett and Parsons 1993). Hosts of this species include T. triandra and a range of other native and introduced 

herbs and grasses (Keith 2004). At sites investigated by Morgan (1996) all perennial species but one had flowered 12 months 

after a late-spring fire. Only 19% of native perennials regenerated from seed after a fire at Derrimut (Lunt 1990c) and only one 

native species was present in the seed bank that was not found in the standing vegetation (Lunt 1990b). In a study of five western 

Victorian grasslands Morgan (1998c) found that only 12% of species, all annuals, formed large, persistent seed banks and that 

most native hemicryptophytes and perennials in general had a transient seed bank. T. triandra and perennial native forbs were 

present in the seed bank at “exceedingly low densities at all times” (Morgan 1998c p. 150). 

Little is known about the persistence of native forbs in temperate Australian grasslands (Lunt 1996). Morgan (1995) found viable 

seeds of Rutidosis leptorrhynchoides in the seed bank only immediately after seed shed. All the seed of this species germinated 

within 4 months of shedding, immediately after autumns rains, and any seed that had not germinated by this time was either dead 

or had been “eaten by soil invertebrates” (op. cit. p. 5). Lunt (1996) found that no viable seeds of Microseris scapigera (G. 

Forst.) Sch. Bip. (Asteraceae) buried or placed in mesh bags under the canopy in a long unburnt T. triandra grassland in 

Canberra remained after 3 months and that virtually all germinated rapidly. Lunt (1995a) tested seeds of three native lilies and 

three other native daisies in the same way and found that >90% of the seeds of all but one species germinated or were unviable 

after 12 months, although greater longevity was recorded for some species if the seed was buried or on the surface. The daisy 

Chrysocephalum apiculatum (Labill.) Steetz. appeared to have the highest potential to develop a persistent seed bank, due to 

small seed size, inhibition of germination in darkness and an ability to remain viable when buried. The perennial native plant 

seed bank as a whole appears to usually be small and highly transient (Lunt and Morgan 2002), seedlings are generally 

uncommon (e.g. Morgan 1998d), most seed germinates or dies within 12 months, and only perennial forbs with small seeds (e.g. 

Hypericum gramineum, Juncus spp., Wahlenbergia spp.) have large, persistent seed banks (Lunt and Morgan 2002). The longterm 

native seed bank was considered to probably have “little functional importance” and “contribute little to seedling 

regeneration processes following disturbance” (Morgan 1998c). Scarlett (1994) reported a similar problem with plantings and 

direct seeding of native dicotyledonous grassland herbs: they set abundant seed, but had very low germination and seedling 

survival. 

Species with transient seed banks are a common feature of world grasslands and “are normally able to exploit gaps in grassland 

canopies ... created by seasonally-predictable disturbances such as fire, drought and grazing”: their seeds usually germinate 

whether or not suitable seedling establishment sites are available (Morgan 1995a p. 7). 

In basalt plains grassland sites with a history of frequent fire the hemicryptophyte (‘resprouter’) rosetted forbs, including many 

of the common native forbs, had low turnover and mobility (Morgan 1997), as would be expected with a very low seed bank 

(Morgan 1998c). In total, 30% of species (notably geophytes) appeared to remain dormant underground in some years, despite a 

stable management regime (Morgan 1997). A high proportion of the seed bank in these grasslands consists of annuals which are 

largely exotic species (Morgan 1998c 1998d, Lunt and Morgan 2002). The annual monocots have a high turnover and high 

mobility (Morgan 1997). Native hemicryptophytes, chamaephytes and geophytes are largely absent from the seed bank (Morgan 

1998c). Exotic species overwhelming dominate the soil seed bank in grazed grasslands in Victoria and Tasmania, with annual 

grasses, legumes and R. rosea being major components (Lunt 1995a). Dodd et al.(2007) found that native species comprised 

only 13% of the soil seed bank of six roadside grasslands along a 200 km urban-rural gradient west from Melbourne and that 

similarities between the species composition of the seed bank and the above ground flora was low and declined in more rural 

sites. Exotic annual graminoids dominated the seed banks (Dodd et al. 2007) with Juncus capitatus, Isolepis sp. and Aira spp. 

being the most abundant (Aaron Dodd pers. comm. October 2007). 

Lack of recruitment of inter-tussock species is a widespread problem in native grasslands: inadequate seed availability appears to 

be the dominant cause, rather than weed competition or climatic stress (McDougall and Morgan 2005). Many species of native 

forbs may be physiologically unable to produce seeds with innate dormancy (Morgan 1998c). Instead the plants are long lived 

and rarely recruit from seed. According to Benson (1994) the major transformation of native grassland have resulted in 

widespread loss of long-lived, deep rooted perennials and their replacement with more ephemeral exotics. 

The native grassland flora is also notable for the absence of myrmechorous (ant dispersed) species, in contrast to Australian 

forests, woodlands and heaths (Berg 1975). Very few of the 87 myrmecochorous genera mentioned by Berg (1975) are present in 

these grasslands and these may be viewed usually as ‘trespassers’ from grassy woodlands. In general however seed dispersal 

processes in these grasslands have not been studied. 

101

Based on a study of a three Asteraceae in New England Tablelands grasslands, a rare native Microseris sp., a rare to common 

native Cymbonotus lawsonianus Gaudich. and an abundant exotic Hypochoeris radicata L., McIntyre (1995) argued that many 

native grassland forbs must have highly specific ecological requirements for regeneration that are rarely being met under current 

mangement regimes, and that many of these requirements might possibly be met by the types of exogenous disturbances often 

considered to be harmful management practices. For example, a suite of endangered grassland forbs appear to require soil 

disturbance, although the general result of soil disturbance is a decline of rare native species and enhanced cover of exotics 

(McIntyre 1995). This apparent paradox may be best resolved by determination of the type and nature of ‘natural’, endogenous 

grassland disturbances that have been lost, or now occur in few areas or at a different frequency (e.g. soil disturbance by 

marsupial foraging or burrowing), and by more detailed examination of processes and disturbances that occur in a range of 

marginal habitats in which rare and uncommon species persist or from which they have disappeared. 

Morgan (1998) investigated germination responses of many of the forbs, none of which are promoted by darkness (i.e. seed 

burial) (Groves and Whalley 2002). 

Robinson (2003) compared the survival of planted seedlings of Podolepis sp. 1 and Bulbine semibarbata in disturbed and 

undisturbed soil of a Victorian basalt plains grassland. Much greater survival of seedlings occurred on the disturbed soil. In the 

same grassland, Reynolds (2006) compared the establishment of seedlings of five native forb species (Podolepis sp. 1, Ptilotus 

spathulatus, Velleia paradoxa, Rutidosis leptorhynchoides and Pycnosurus chrysanthes) after surface sowing of seed and 

planting of seedlings on disturbed and undisturbed soil. Plots were treated with mowing, raking, repeated herbicide treatments 

over a period of a few months and fire to remove all above ground plant biomass prior to sowing. The disturbed soil was dug 

with a spade to a depth of at least 20-30 cm. Six months after sowing the proportion of seed that produced established seedlings 

on undisturbed soil did not exceed 2%, while in disturbed soil the proportion that established was 2-65%. Four species grown on 

disturbed soil flowered, but all seedlings on undisturbed soil remained very small. Mortality of seedlings after six months was 

greater on undisturbed soil (54% survival) than disturbed soil (80%) and significantly different for three of the four species for 

which there was adequate data. All species grew significantly taller and wider in disturbed soil. Soil digging created an 

abundance of suitable microsites for seed germination, whereas the smooth compacted surface of the undisturbed soil provided 

very few. Seedlings of exotic weeds originating in the soil seed bank had to be continuously removed from the disturbed plots 

during the period of the study but no data about them was provided. Reynolds concluded that the germination of both native 

forbs and exotic weeds would be promoted by such soil disturbance. 

Soil disturbance affects the population dynamics of the native forbs in a range of ways that remain poorly understood. Soil 

disturbances increase seed incorporation into the soil, which advantages some species and not others. Lunt (1995a) found that 

buried seed had significantly greater survival than unburied seed for three of the species he tested. Some species exhibited major 

differences in the timing of germination between buried and surface seeds, while others did not. 

Particular disturbances or disturbance regimes always favour a subset of native species and disadvantage some species. A 

changed disturbance regime is small remnants normally results in the disappearance of some species which are unable to 

recolonise because of the now highly fragmented geographical distribution of remnants (Kirkpatrick et al. 1995). 

A vegetation assessment protocol that enables the classification of native grasslands on the basis of their botanical signficance 

has been devised for use in the Australian Capital Territory (Sharp 2006). A grassland is characterised as native if >50% cover 

consists of native species. The botanical significance rating (BSR) depends on the level of vascular plant species richness (very 

low, low, moderate, high, very high) and the presence of very common, relatively common, less common and uncommon native 

species, which in turn are very tolerant, moderately tolerant, sensitive or highly sensitive to anthropogenic disturbances. 

Grasslands with the highest BSR have several to many representatives from each of these native species groups including the 

most disturbance-sensitive taxa. The systems has been used for conservation planning in the ACT. 

Non-vascular plants and fungi 

Bryophyta (mosses), Hepatophyta (liverworts) and Anthocerophyta (hornworts), collectively known as bryophytes (Meagher and 

Fuhrer 2003) provide another component of biodiversity in south-eastern temperate grasslands. Bryophytes of Australian native 

grasslands are relatively poorly studied, e.g. Meagher and Fuhrer (2003) provided lists of genera for numerous habitats and 

vegetation formations, but not for native grasslands. The bryophytes along with lichens (symbiotic associations of algae or 

cyanobacteria with fungi) and algae, collectively known as cryptogams, are found on exposed rock surfaces but are more 

prevalent on the soil surface, forming what is known as the ‘cryptogam crust’ (Scarlett 1994), or where lichens are largely 

lacking, the ‘bryophytic mat’ Morgan 2004). This acts to restrict soil erosion and increase water infiltration (Lunt et al. 1998 p. 

11). Many of the lichens fix atmospheric N (Lunt et al. 1998) and biological soil crusts also contribute significantly to C fixation 

(Morgan 2004). The lichen crust reportedly becomes apparent after the inter-tussock forbs die back (Kirkpatrick et al. 1995). 

Effects of the soil crust on shed seed and seedling germination differ between plant species (Morgan 2004). It can provide “a 

foothold for germinating seedlings” (Lunt et al. 1998 p. 11) but commonly restricts seedling establishment by limiting contact of 

the radicle with mineral soil (Mack 1989). 

Cryptogam diversity in native grasslands has rarely been studied, but where it has, it has it has been found to be highly variable, 

e.g. there were at least 32 species at Evans St., Sunbury, but only 3 at Derrimut (Morgan and Rollason 1995) perhaps in part due 

to the ‘overdominance’ of T. triandra at the latter site (Lunt 1990a) and its long history of grazing. Soils crusts are poorly 

developed also at other previously grazed grasslands – Organ Pipes National Park and Laverton North Grassland Reserve 

(Scarlett 1994). Willis (1964) found a total of about 85 species in the Victorian basalt plains flora, markedly less than in nongrassland 

areas of the State. Scarlett (1994) found that railway reserve grasslands were dominated by mosses and liverworts, 

rather than lichens. Morgan (2004) surveyed six T. triandra grassland remnants on the Victorian volcanic plains and found a 

total of 19 mosses and 8 liverworts of which 9 were present only at single sites. The richest site has 10 mosses and 7 liverworts 

in 150 m 2 , the poorest had 6 mosses and 3 liverworts in the same area. Again lichens were uncommon. 

102

Like vascular plant diversity, community richness appears to be highly dependent on past management. Morgan (2004) found 

that sites burnt at 1-2 yr intervals had lower diversity in the bryophyte mat than those burnt at 4-20+ year intervals, apparently 

due to loss of mosses. The liverworts Fossombronia intestinalis Taylor and Lethocolea pansa (Taylor) G.A.M.Scott and 

K.Beckmann, and the moss Rosulabryum billardieri (Schwaegr.) Spence occurred at all sites and were considered to be adapted 

to fire at all frequencies. Morgan (2004) found a strongly significant positive correlation between the cover of T. triandra and the 

species richness of the bryophyte mat, but no correlation with vascular plant species richness. 

Based on observations of ungrazed and grazed native grassland reserves and experimental studies by other workers Scarlett 

(1994 p. 127) determined that the cryptogam crust in T. triandra grasslands “delays the establishment of some alien weeds and 

minimises their cover/abundance when they are already established”. In south-eastern Australia, invasions of Holcus lanatus, 

Briza maxima, Bromus hordeaceus and Vulpia spp. are facilitated by crust damage. Soil crusts dominated by brypophytes have 

been found to delay or inhibit germination of some plants by affecting the penetration of light and its spectral characteristics, and 

through leachates, and the crusts also maintain a more humid environment at ground level. They increase the time a seed spends 

above ground, increasing its probability of predation, dessication or destruction by fire, although awned seeds, including those of 

N. neesiana, are less affected. They also could be expected to restrict root growth of any seedlings that do germinate (Scarlett 

1994). Davies (1997) noted that seeds of native plants are generally better adapted to penetrating cryptogam crusts than those of 

many grassland weeds. 

The commonest components of the soil crust in Victorian volcanic plains grasslands include the prostrate leafy and thallose 

liverworts Riccia spp., F. intestinalis, F. pusilla and L. pansa, the mosses R. billardieri, Fissidens spp., Tortella calycina, and the 

squamulose lichen Cladia sp. (Scarlett 1994, Morgan and Rollason 1995, Morgan 2004). Large mosses including Bruetelia 

affinis, Triquetrella papilata and Campylopus clavatus occur mainly around the bases of T. triandra tussocks and create denser 

cover, while Polytrichum juniperinum and various liverworts occur on stony rises. The basalt rocks are occupied by thallose and 

crustose lichens. Drier sites tend to have a crust dominated by crustose lichens and algae (Scarlett 1994). Thick moss mats are 

relatively rare and there is relatively little variation in composition of the crust over the 500-600 mm rainfall zone (Scarlett 

1994). Non-vascular plants accounted for 25% of plant diversity at Evans St., Sunbury (Morgan and Rollason 1995), 28% at six 

sites surveyed by Morgan (2004) and 13.5% for the Victorian basalt plains flora as a whole (Willis 1964). 

The non-lichenised fungi constitute another diverse element of the biota but little specific information related to Australian 

native grasslands appears to be on record. Fuhrer (1993) noted that some species are restricted to grasslands and recorded 

Lycoperdon spp. and Xerula australis (H. Dorfelt) R.H. Peterson from grasslands. Slime moulds (Protocista: Myxomycota) are 

similarly poorly known. Most fungi species are very small, a high proportion are undescribed and identification is difficult. Öster 

(2008) concluded from a study of semi-natural grasslands in Sweden that there was probably low congruence bewteen vascular 

plants and grassland fungi and that some grasslands with low plant richness can have high macrofungi richness. Many plant 

pathogenic fungi occur on grassland plants, notably on grasses, which are infected by a range of smuts, rusts and endophytic 

fungi. 

Endophytic grass fungi 

Grass endophytes, Neotyphodim spp., have not been found in 13 genera of Australian native grasses investigated, including T. 

triandra, Microlaena stipoides, Austrodanthonia spp., Chloris spp., Poa spp. and Bothriochloa macra, except for an unknown 

species in Echinopogon ovatus (G. Forst.) P. Beauv. and Neotyphodium-like hyphae in herbarium species of other Echinopogon 

species (Aldous et al. 1999). 

A Tilletiopsis Derx. fungus, related to smut fungi, has been isolated from seeds of Austrodanthonia pilosa (R.Br.) H.P. Linder; 

an Acrodontium De Hoog. sp. from seeds of Chloris ventricosa R.Br., a species similar to Neotyphodium from seed of 

Austrodanthonia racomosa (R.Br.) H.P. Linder, and the seed-transmitted Atkinonella hypoxylon (Peck) Dell is common on a 

range of Austrodanthonia spp. (Aldous et al. 1999). 

Soil microflora 

The soil microflora of the temperate native grasslands of south-eastern Australia also appears to have been little investigated. 

Such floras consist mainly of bacteria and fungi that decompose organic matter and those that form symbiotic relationships with 

plants (Keane 1994). The vast majority of plants form sysmbioses with soil fungi via their roots, known as mycorrhizae. Many 

grasses and herbs form vesicular-arbuscular mycorrhizae with zygomycetes, that are characterised by “little bush-like growths 

inside the root cells and large, swollen vesicles within the roots”(Keane 1994 p. 132). The fungi have never been grown in pure 

culture and they benefit by gaining sugars from the plant while assisting plant P uptake. Orchids form more complicated 

symbioses with Rhizoctonia fungi. Legumes form associations with Rhizobium bacteria that are important N fixers. Some herbs 

and liverworts form associations with cyanobacteria that are capable of fixing atmospheric N (Keane 1994). The soil also 

contains parasitic microbes which can cause new plant disease problems when the soil is disturbed. Manipulation of the soil 

microflora is widely practiced in agriculture, but has been little investigated in Australian native vegetation, except in regard to 

orchids (Keane 1994). Exotic soil microbes may be important in south-eastern grasslands, but the best known species in 

Australia Phytophthora cinnamomi does not appear to cause problems. 

Exotic plant components 

The criterion of

ecorded in them. Lowland grassland and grassy woodland were ranked highest with 344 taxa of which 87 were considered very 

serious. Between one quarter and one third of the flora in each of the main grassland regions consists of exotics (Kirkpatrick et 

al. 1995). Weed invasion is a major problem for survival of the native flora. Dicotyledonous herbs are the most threatened group 

(Groves 2004). Weed diversity and dominance at grassland sites is often high. For example McIntyre (1993) found that 97% of 

samples in New England tablelands native grasslands in 1990-91contained exotic species. Trémont (1994) found that secondary 

grasslands in that area were composed of 27% exotics in areas ungrazed for 16 years and 32% in grazed areas, with exotics 

comprising 50% and 43% of the grasses in the two treatments respectively. In the Monaro region exotics accounted for an 

average of 35% of species, and sites on the Southern Tablelands generally had >20% exotic cover in spring (Sharp 1997). 

According to Kirkpatrick et al. (1995), the high invasibility of grasslands is probably related to their high soil fertility. McIntyre 

and Lavorel (1994a 1994b) reported declines in native species richness and increases in exotic richness on sites of increasing 

natural substrate fertility (granite < sediment < basalt) on the New England Tablelands and a similar trend for water enrichment. 

Exotic invasion occurred simultaneously with the introduction of livestock and resulted from the carriage of seed in their coats or 

digestive tracts, or by movement of fodder, and the superior adaptations for survival that these plants possessed under the new 

grazing regimes (Kirkpatrick et al. 1995). Invasions may have been facilitated by the disappearance or dysfunction of the C 4 

grass, the C 3 grasses, or the intertussock forbs (Groves and Whalley 2002) resulting from various forms of disturbance. Soil 

disturbance that raised the level of available nutrients in situ has had particularly insidious effects (Wijesuriya and Hocking 

1999). Extraneous nutrient addition also commonly results in damage. Roadside grasslands in western Victoria were rapidly 

invaded by Holcus lanatus in 1983 after nutrient rich soil drifted from drought-affected agricultural land, a number of significant 

remnants being destroyed (Kirkpatrick et al. 1995). Competition from annual grasses may significantly impact on native forbs 

with similar cool season growth patterns (Morgan 1994) but invasion by perennial grasses including N. neesiana has resulted in 

much greater community change (Hocking 1998, Lunt and Morgan 2000). 

In the New England grasslands low native diversity occurred in areas with greater frequency of soil disturbance and water 

enrichment, and was associated with thick litter cover and >5% bare ground (McIntyre 1993). Exotic species were rarely 

dominant, but Hypochoeris spp., Plantago lanceolata L., Vulpia spp., Trifolium arvense L., Bromus spp. and Paspalum 

dilatatum were dominants in a small number of 30 m 2 quadrats. The six sites with the lowest richness all had high soil 

disturbance and litter >5 cm deep, and were dominated by perennial exotic grasses. Severe soil disturbance in these grasslands 

eliminated a large number of native taxa, which were replaced largely by exotic species (McIntyre and Lavorel 1994a 1994b). 

The disturbance-intolerant species included many rare taxa which apparently were unable to recolonise disturbed ground or were 

uncompetitive in situations of high productivity (McIntyre and Lavorel 1994a). A group of disturbance-tolerant species were 

found over the range of disturbance states and included natives such as Acaena spp. and Asperula conferta Hook. f. and the 

exotics Hypochaeris radicata L. and Centaurium erythraea Rafn. A further group of “disturbance specialists” exploited areas 

that had been heavily grazed or where the soil had been disturbed, and were mainly exotics, including Paspalum dilatatum and 

Plantago lanceolata L., but included a few native herbs. Disturbance specialists tended to dominate the sward where disturbance 

was intense (McIntyre and Lavorel 1994a). 

T. triandra grasslands of the Victorian volcanic plains in general have a large exotic vascular plant component, even those that 

have been minimally disturbed and are rich in native species (Morgan 1998c 1998d). For example 41% of the 102 species found 

in quadrats at Evans Street, Sunbury grassland in 1993-4 were exotics, 52% being annuals, and 77% of them having

Opinions differ on the nature and intensity of disturbance required for weed invasion. Patton (1935 p. 175) considered intact 

grasslands on the Victorian basalt plains to be highly resistant: “So long as the natural vegetation covering, open though it be, is 

maintained, entrance to new-comers is denied.” Lunt (1990a) found that areas of Derrimut Grassland ploughed during the 19th 

century were amongst the most diverse, with an average of 17 native and 13 exotic vascular plant species per 15 m 2 , probably 

illustrating one of the dilemmas of native grassland management, that certain types of disturbance favour both weeds and the 

native flora (Trémont and McIntyre 1994). Some exotics allegedly invade without major prior disturbance, by virtue of their 

superior competive abilities (Carr 1993), but prior disturbance may often be difficult to detect, and effects of historical 

disturbance may resonate long into the future. Disturbance that destroys native species certainly facilitates invasion by weeds 

both in terms of species and population sizes, and their occupation reduces recolonisation by natives. Nutrient and water 

enrichment often favours the exotics (McIntyre 1993). Soil disturbance was found to be the most important factor determining 

variation in plant species composition in grasslands of the New England Tablelands, more important than altitude, topographic 

position, lithology or water enrichment, but the greatest floristic effects resulted from soil disturbance accompanied by grazing 

(McIntyre and Lavorel 1994a). Dorrough et al. (2004) found that the intensity of grazing explained significant increases in the 

exotic richness of Monaro Tablelands grasslands, but that exotic status itself did not significantly explain plant responses in 

matched areas with contrasting grazing history. Exotic plant diversity was high even at the lowest grazing frequency. In many 

areas the local species pool is more likely to contain exotics that are well adapted to wetter conditions than natives (Kirkpatrick 

et al. 1995). The annual grass Bromus hordeaceus can be abundant where fire has been absent for a long period but is quickly 

eliminated by burning (Kirkpatrick et al. 1995). 

The consensus position of most grassland specialists appears to be well represented by McIntyre and Lavorel (1994a p. 381): 

“With little exogenous disturbance, the native grassland consists of a matrix of dominant perennial tussock grasses (e.g. Poa, 

Themeda) as well as smaller statured interstitial species that form the bulk of of species richness... including some exotic 

‘tolerant’ species. With increasing disturbance, the tolerant species will persist and many ‘intolerant’ species will decline, to be 

partially replaced by disturbance specialists. At the highest levels of disturbance, the matrix of native perennial grasses is usually 

replaced by large statured exotic grasses or forbs ... species richness is very low and a only a few tolerant native species persist.” 

The presence of weeds in native grasslands may impact on native plants not only through direct competition, but also by 

ramifiying effects through the food chain. Foraging by cockatoos for the bulbs of Romulea has been recognised as a theat to the 

integrity of one northern Victorian grassland (Kirkpatrick et al. 1995). 

Prevention of exotic weed invasion in grasslands has two main components, minimisation of disturbance to maintain an intact 

ground vegetation stratum, and hygiene measures to prevent entry of propagules (Davies 1997). Edges of grassland remnants 

tend to be the most highly invaded e.g. at Evans St. the road edges have significantly greater exotic plant richness and 

significantly lower native plant cover and richness than in core areas (Morgan and Rollason 1995, Morgan 1998d). Morgan 

(1998d) found no correlation between exotic plant richness and the amount of ‘bare’ (cryptogam encrusted) ground, a weak 

negative correlation with T. triandra cover and a stronger negative correlation with native plant richness. 

Grasslands that are floristically rich with native species are considered of high quality and have high diversity of forbs, an open 

structure with a high proportion of intertussock space, few weed species and management regimes involving regular biomass 

reduction by grazing or burning (Henderson 1999). Floristically poor (low quality) grasslands have low forb diversity, a canopy 

of dominant grasses that is often closed, abundant weeds and often a management regime lacking regular biomass reduction 

(Henderson 1999). 

Rare and endangered plants 

“The most imminent threat to the biological diversity of grassland is the extinction of rare and threatened species and genotypes. 

These cannot be resurrected once lost, whereas the grassland communities could conceivably be re-established ...” (Kirkpatrick 

et al. 1995 p. 87). 

A large number of rare and threatened plant species occur in native temperate grasslands of south-eastern Australia (Tables 6 and 

7). Robinson (2005) stated that approximately 200 native forb species that occur in these grasslands are rare and endangered, 

many of which were formerly common and widespread. Scarlett et al. (1992) provided lists of depleted, rare, vulnerable, 

endangered and presumed extinct grassland plants in Victoria. Scarlett and Parsons (1993) compared the numbers of threatened 

vascular plants across major landscape types in Victoria, including a category consisting of grasslands, grassy woodlands and 

fertile lowland open forests (Table 6). The highest concentration of endangered species occurred in this ‘grassy’category, 

whether assessed in the Victorian context or in terms of species populations throughout Australia. There was no doubt that 

undescribed taxa were under threat including species in Senecio, Craspedia, Podolepis, Leptorhynchos and Microseris (all 

Asteraceae), although some have subsequently been described. Herbs and low semi-shrubs comprised the majority of rare and 

threatened taxa. Kirkpatrick et al. (1995) listed 24 lowland temperate grassland taxa considered to be nationally rare or 

threatened, and calculated that 11% (79 species) of the total lowland grassland flora was rare or threatened. These data can now 

be considered to be poor indicators of current circumstances: “Even the most common species of lowland grassland forbs are 

now comparatively rare ...” (Robinson 2005). 

Table 6. The number of extinct, endangered, vulnerable, rare and insufficiently known vascular plant taxa of Victorian 

grasslands, grassy woodlands and fertile lowland open forests in 1993. Source: Scarlett and Parsons (1993). 

ConservationStatus Extinct Endangered Vulnerable Rare Insufficiently 

known 

Total 

Australia-wide 3 16 19 5 5 48 

Victoria 9 33 54 43 3 142 

105

Table 7. Some extinct, endangered, vulnerable and rare vascular plant taxa of south-eastern Australian native temperate grasslands. See Scarlett and Parsons (1993) and Ross and Walsh (2003) for 

definitions of the categories. R = rare, U = unlisted, K = poorly known, X = extinct, E = endangered, V = vulnerable, - = not present, ? = status not determined 

Species Common Name Family Aust ACT NSW Vic Tas References 

Ammobium craspediodes Yass Daisy Asteraceae U V - - Eddy et al. 1998 

Amphibromus pithogastrus Plump Swamp Poaceae K ? ? E - Ross and Walsh 2003, Ashton and Morcom 2004, DSE 2009a 

Wallaby-grass 

Calotis glandulosa Mauve Burr-daisy Asteraceae U V - - Eddy et al. 1998 

Carex tasmanica Curly Sedge Cyperaceae V - - V V Kirkpatrick et al. 1995, Morcom 1999, Ross and Walsh 2003 

Colobanthus curtisiae Colobanth Caryophyllaceae - - - - E Kirkpatrick et al. 1995 

Comesperma polygaloides Small Milkwort Polygalaceae - - - V Ross and Walsh 2003, McIntyre et al. 2004 

Cullen parvum Small Psoralea Fabaceae E - X E - Kirkpatrick et al. 1995, Ross and Walsh 2003, Muir 2003, DSE 2009a 

Discaria pubescens Anchor Plant Rhamnaceae R R - Kirkpatrick et al. 1995, Ross and Walsh 2003, DSE 2009a 

Diuris fragrantissima Sunshine Diuris Orchidaceae E - - E - Kirkpatrick et al. 1995, Ross and Walsh 2003, Webster et al. 2004, Smith et al. 2009, DSE 2009a 

Dodonaea procumbens Creeping Hopbush Sapindaceae V U V V Eddy et al. 1998, Ross and Walsh 2003 

Glycine latrobeana Clover Glycine Fabaceae V V V Kirkpatrick et al. 1995, Ross and Walsh 2003, DSE 2009a 

Lachnagrostis adamsonii blown grass Poaceae V - - V V Kirkpatrick et al. 1995, Ross and Walsh 2003, DSE 2009a 

Leucochrysum albicans subsp. Paper Daisy Asteraceae E ? ? E V? Kirkpatrick et al. 1995, Ross and Walsh 2003 

albicans var. tricolor 

Maireana cheelii Chariot Wheels Chenopodiaceae V V - Kirkpatrick et al. 1995, Ross and Walsh 2003, DSE 2009a 

Microseris lanceolata Yam Daisy Asteraceae - - - - - Sharp and Shorthouse 1996: regionally uncommon in the ACT, 

Pimelea spinescens subsp. Spiny Rice-flower Thymelaeaceae V - - V/E - Kirkpatrick et al. 1995, Ross and Walsh 2003, Tumino 2004, DSE 2009a 

spinescens 

Podolepis sp. 1 Basalt Podolepis Asteraceae - - - E - Kirkpatrick et al. 1995, Ross and Walsh 2003, Robinson 2005 

Prasophyllum diversiflorum Gorae Leek-orchid Orchidaceae E - - E Ross and Walsh 2003, Pritchard and Ingeme 2003, DSE 2009a 

Prasophyllum fosteri 

Shelford Leekorchid 

Orchidaceae E - - E Ross and Walsh 2003, Coates 2003a, DSE 2009a 

Prasophyllum petilum leek orchid Orchidaceae E E - Eddy et al. 1998 

Prasophyllum suaveolens Fragrant Leekorchid 

Orchidaceae E - - E - Kirkpatrick et al. 1995, Ross and Walsh 2003, Coates 2003b, DSE 2009a 

P.sp. aff. suaveolens (Western 

Basalt Plains) 

Psoralea tenax Emu Foot Fabaceae - - - - - Sharp and Shorthouse 1996: regionally uncommon in the ACT, 

Pterostylis basaltica Basalt Greenhood Orchidaceae E - - E - Kirkpatrick et al. 1995, Ross and Walsh 2003, Ingeme 2003, DSE 2009a 

Pterostylis truncata Brittle Greenhood Orchidaceae - E Ross and Walsh 2003, Bramwells 2003, DSE 2009a 

Rutidosis leiolepis 

Monaro Golden Asteraceae U V - Eddy et al. 1998 

Daisy 

Rutidosis leptorhynchoides Button Wrinklewort Asteraceae E E E E - Morgan 1995a 1995b, Kirkpatrick et al. 1995, Sharp and Shorthouse 1996, Eddy et al. 1998, 

Scarlett and Parsons 1993, Ross and Walsh 2003, Humphries and Webster 2003, DSE 2009a 

Schoenus absconditus Obscure Bog-rush Cyperaceae - - - - R Kirkpatrick et al. 1995 

Sclerolaena napiformis Turnip Bassia Chenopodiaceae E E - Kirkpatrick et al. 1995, Ross and Walsh 2003 

Senecio behrianus - Asteraceae E ? X? E - Walsh 1999, Ross and Walsh 2003, DSE 2009a 

Senecio macrocarpus 

Large-headed 

Groundsel 

Senecio georgianus Asteraceae X - X - X Kirkpatrick et al. 1995, Ross and Walsh 2003 

Spyridium obcordatum Dusty Miller Rhamnaceae - - - - V Kirkpatrick et al. 1995 

Swainsona adenophylla Violet Swainson-pea Fabaceae - E Ross and Walsh 2003, Earl et al. 2003, DSE 2009a 

Table 7 (continued). 

Asteraceae V - - E - Scarlett and Parsons 1993, Kirkpatrick et al. 1995, Hills and Boekel 1996 2003, Walsh 1999, 

Ross and Walsh 2003, DSE 2009a 

106

Species Common Name Family Aust ACT NSW Vic Tas References 

Swainsona monticola Fabaceae - - - - - Sharp and Shorthouse 1996: regionally uncommon in the ACT, 

Swainsona murrayana Slender Darling-pea Fabaceae V E - Kirkpatrick et al. 1995, Ross and Walsh 2003, Earl et al. 2003, DSE 2009a 

Swainsona plagiotropis Red Swainson-pea Fabaceae V V V E - Kirkpatrick et al. 1995, Ross and Walsh 2003, Earl et al. 2003, DSE 2009a 

Swainsona recta Small Purple Pea Fabaceae E E E E - Kirkpatrick et al. 1995, Sharp and Shorthouse 1996, Eddy et al. 1998, Ross and Walsh 2003, DSE 

2009a 

Swainsona sericea Silky Swainson-Pea Fabaceae - - V Sharp and Shorthouse 1996: regionally uncommon in the ACT, Ross and Walsh 2003, Earl et al. 

2003, DSE 2009a 

Thelymitra gregaria Basalt Sun-orchid Orchidaceae E - - E - Ross and Walsh 2003, Coates 2003c, DSE 2009a 

Thesium australe Austral Toadflax Santalaceae V U V V V Kirkpatrick et al. 1995, Eddy et al. 1998, Scarlett and Parsons 1993, Scarlett et al. 2003, Ross and 

Walsh 2003, DSE 2009a 

107

Much less is known about the conservation status of plants at the intraspecific level. Groves and Whalley (2002) emphasised the 

need to conserve the widespread intraspecific variation in Australian grasses (polyploid complexes, ploidy levels, breeding 

systems). T. triandra for example, is widely variable, consisting of a polyploid complex, mainly of diploid and tetraploid 

populations in southern Australia (Hayman 1960) and different populations have varying flowering responses to photoperiod and 

temperature (Groves 1975). 

Many of the rare and endangered plants have become scarce as a result of the same processes that have led to the loss and 

degradation of grassland communities. Continuous livestock grazing is considered to be a major cause of decline of such species 

as Swainsona recta A.T. Lee and S. plagiotropis F. Muell. (Fabaceae), Rutidosis leptorhynchoides, Senecio macrocarpus 

(Asteraceae) and Thesium australe R.Br. (Santalaceae), and, except for S. plagiotropis and T. australe, fire is probably, or should 

be, important in maitaining their habitat. Competition in dense T. triandra swards affects S. recta, R. leptorhynchoides, S. 

macrocarpus and T. australe (Scarlett and Parsons 1993). Colobanthus curtisiae J.G. West and Stakhousia gunnii Schltdl. 

disappear from grasslands when the perennial grass sward covers all the bare ground (Kirkpatrick 2007). Rutidosis 

leptorhynchoides survived in areas subject to frequent fires, but disappeared from two railway reserves in the Melbourne area 

when regular burning ceased in the 1980s (Kirkpatrick et al. 1995). Three sites where it existed, at Canberra, Queanbeyan and 

near Melbourne, were allowed to be developed on the condition that populations were transplanted, but the transplanting failed 

(Kirkpatrick et al. 1995). Morgan (1995b) considered it was restricted to remnant grasslands never subjected to grazing, 

ploughing and fertiliser application. Senecio behrianus Sond. and F. Muell. was formerly known from the Western District of 

Victoria and the Northern Plains and was apparently restricted to heavy clay, winter-wet soils (Walsh 1999). The one known site 

where it continued to exist appeared to have once been Eucalyptus camaldulensis woodland but was highly modified and 

difficult to botanically categorise (Scarlett and Parsons 1993). 

Taxa endangered in the Riverina include two chenopods, Maireana cheelii (R.H. Anderson) Paul G. Wilson and Sclerolaena 

napiformis Paul G. Wilson, along with several from other families that are also found in the more mesic grasslands (Kirkpatrick 

et al. (1995). 

Senecio georgianus DC “known ... probably from grassland” (Kirkpatrick et al. 1995 p 30) was recorded from Lake Omeo and 

the Macalister River in Victoria and from Western Australia, South Australia and Tasmania (Walsh 1999) but is now extinct 

(Walsh and Stajsic 2007). Stemmacantha australis (Gaudich.) Dittrich, the only Australian native thistle, known in Victoria from 

Lake Omeo and Murrindal, “probably” from grasslands, is now extinct in Victoria (Jeanes 1999b p. 677) and NSW, but survives 

in Queensland (Kirkpatrick et al. 1995). 

Diuris fragrantissima, Sunshine Diuris or Fragrant Doubletail (Orchidaceae) is endemic to Victoria (Walsh and Stajsic 2007), in 

particular to an area with a 25 km radius now part of western urban Melbourne (Smith et al. 2009). It and was “once widespread 

on the Keilor Plains” (SGAP 1991 p. 206), and was so common at the time of European settlement that it was referred to as 

‘Snow-in-the-Paddocks’ (Webster et al. 2004). It is now “exceedingly rare, restricted to remnant dry grassland on the basalt 

plains near Sunshine” (Entwisle 1994 p. 858) at two sites – a railway reserve at Tottenham and Laverton North Grassland where 

it has been planted (Webster et al. 2004, Smith et al. 2009). The site at Sunshine/Tottenham has few remaining plants, and the 

plantings at Laverton North have progressively declined and not resulted in a viable population (Smith et al. 2009). N. neesiana 

is a “very serious” direct threat to the small population at the Tottenham site, and the immediate surrounding area is extensively 

invaded (Webster et al. 2004). 

The threatened Curly Sedge, Carex tasmanica Kuk grows in wetlands and at grassland edges, often in areas too wet to support 

tussock grassland (Morcom 2004). In Victoria it was known to exist at only nine sites, “in remnant grasslands” (Morcom 2004). 

Leucochrysum albicans has no persistent seed bank, but disperses widely on the wind, and in part of its range is dependent on 

heavy sheep grazing to create patches of bare ground every few years in which seedlings can establish (Kirkpatrick 2007). 

Many of the rare and endangered forbs produce large amounts of seed and are long-lived, and some reproduce vegetatively, but 

recruitment of new individuals is usually rare, for reasons that are poorly understood (Robinson 2005). Rare and threatened 

vascular plants of grasslands often survive in areas with ususual disturbance regimes, which also usually favour weeds and few 

other natives, so tend to be degraded, and to be assigned little conservation value (Kirkpatrick et al. 1995, Kirkpatrick 2007). 

These species usually have poor competitive abilities and are disturbance-dependent (Kirkpatrick 2007). Robinson (2003 2005) 

demonstrated that many species that recruit poorly in grassland remnants lack specialised germination requirements and can be 

propagated and grown relatively easily under nursery conditions, but that a range of Apiaceae, Fabaceae, Ranunculaceae and 

some Asteraceae do have specialised requirements such as cold stratification and after-ripening requirements that may rarely be 

met in the wild. 

Factors that are probably important in limiting forb recruitment include the presence of weeds, particularly Nassella spp., 

herbivory by exotic invertebrates, grazing by livestock and rabbits, and the loss of small-scale soil disturbance formerly achieved 

by native vertebrates (Robinson 2003 2005). Reynolds (2006) demonstrated that soil digging is indeed an important factor, 

critical for recruitment of several species. Soil disturbance may have a significant role in increasing seed contact with the soil or 

actual seed burial, or may be more important in enabling penetration of the radical (Robinson 2005). Iincreased survival and 

flowering of Diuris fragrantissima established using soil digging might be due to enhanced functioning of its fungal symbiont 

due to improved aeration (Smith et al. 2009). The scale and intensity of the disturbance is of obvious importance in determining 

the effects on competitors, nutrient levels and other factors. 

Ecological history 

The pre-European disturbance regimes of temperate Australian grasslands are poorly understood (McIntyre and Lavorel 2007). 

Fire is universally recognised as an important factor, both pre-aboriginal and ‘managed’ aborignal burning (Kershaw et al. 1994, 

Hope 1994, Kershaw et al. 2000 ). Marsupial grazing must have had a large influence: Australian grasslands are presumed to 

have once been grazed by a now extinct marsupial megafauna, and known to have been occupied by numerous smaller 

herbivorous mammals, most of which are now extinct or have relictual distributions. Climatic variation has also been important 

108

in shifting the distribution, extent and composition of grasslands. All these major factors are compounded, their individual 

influences are difficult to determine from the limited palaeecological records (Hope 1994) and existing remnants may have little 

resemblance to their pre-European condition (Trémont and McIntyre 1994). The palaeontolgy of temperate Australian grasslands 

is discussed in the section on grasslands origins (above), while the pre-European influences of aboriginal grassland management, 

grazing and fire, and the post European disturbance and management factors are discussed in sections on these topics below. 

Aboriginal management 

In south-eastern Australia, apart from the coast, grassland was the main habitat occupied by aborigines (Kirkpatrick et al. 1995). 

The earliest undisputed evidence for aboriginal occupation of Australia remains at around 40,000 ypb in the Kimberley of 

Western Australia, close to the limit of resolution of radiocarbon dating, but evidence using other techniques indicates first 

occupation may have occurred c. 60 kybp (Kershaw et al. 2000). Coutts (1982) claimed that aboriginal people were present in 

Victoria at least as long ago as 40 kybp, but the oldest south-eastern mainland site known, at Willandra Lakes, is dated at c. 36 

kybp (Kershaw et al. 2000). Most of Australia, including Tasmania was probably occupied by 35 (Hope 1994) or 32 kybp, all 

major environments were certainly occupied by 22 kybp, and occupational intensities increased after c. 5 kybp (Kershaw et al. 

2000). Aborigines coexisted for many thousands of years with the extinct marsupial megafauna in the early prehistoric period 

(Coutts 1982, Kershaw et al. 2000). 

Aboriginal people managed the land over many thousands of years (Zola and Got 1992) and would have witnessed the assembly 

of many Australian grasslands, notably on the Victorian Volcanic Plains, where the most recent terrain was formed several 

thousand years ago. The total aboriginal population of Victoria in 1788 has been estimated at 15,000 (Coutts 1982) but actual 

population is very uncertain. Mulvaney (1964) suggested a population density of one person per 13 km 2 in the western district of 

Victoria prior to European settlement, with a total population of 1,800. Mulvaney’s estimate is probably much too low, since 

800-1,000 western district aborigines were known to gather annually at Lake Bolac (Jones 1999b). The South East of South 

Australia supported an “unusually large population, perhaps numbering 2000” (Pretty et al. 1983 p. 116). The population in that 

region is estimated to have declined by half every five years, from the beginning of settlement in 1840 (Pretty et al. 1983). 

Numerous archaeological sites are recorded on the Victorian Basalt Plains including campsites, earthen mounds (probably long 

occupied campsites), burials, canals and weirs of basalt rocks used as fish traps, basalt block walls probably roofed with timber 

and used as huts, etc. (Coutts 1982). Indications are that some populations were more or less sedentary, rather than nomadic. 

Digging and burning of the vegetation were probably the most important aboriginal activities in terms of grassland ecology, and 

areas of grassland appear to have been extended by aboriginal activities (Kirkpatrick et al. 1995). However in the mid to late 

Holocene some societies in more arid areas developed economies based on systematic harvesting and processing of grass seed 

(Kershaw et al. 2000, Gammage 2009). One species exploited was Barley Mitchell Grass, Astrebla pectinata (Lindl.) F. Muell., 

a common and very widespread species of the summer rainfall semi-arid and arid zones. It has “relatively large seeds that 

separate easily from the chaff” and “at one time provided an important part of the diet of the Aborigines” (Cribb and Cribb 1974 

p. 101). Another grain, with distribution extending into more temperate winter rainfall areas of south-eastern Australia, was 

Native Millet or Windmill Grass, Panicum decompositum R.Br., a species “often associated with temporarily wet places such as 

creek beds and flood plains” (Jessop et al. 2006 p. 461). It was a major foodplant (Jessop et al. 2006), extensively cultivated 

(Gammage 2009), the seeds being ground into a paste and baked to form bread (Cribb and Cribb 1974). Another arid zone 

species harvested was Woollybutt, Eragrostis eriopoda Benth. (Jessop et al. 2006), which has abundant, readily husked, soft, 

easily-ground seed that is held on the plant for months (Low 1989). Other Panicum spp. were also used (Low 1989). Perhaps 

Hairy Panic, Panicum effusum R.Br., a common associate of T. triandra in temperate south-eastern grasslands, was used in a 

similar manner. Eragrostis tef (Zucc.) Trotter and Panicum miliaceum L. were amongst the grasses cultivated in pre-Islamic 

civilisations in Arabia (Pohl 1986). Current consensus appears to be that grasses were not important in the diets of aborigines 

living outside the dry interior, and Gammage (2009) has convincingly argued that they were little used in areas with a reliable 

supply of edible tubers. However the possibility that aboriginal management of cereal grasses influenced the structure and 

biodiversity of the temperate grasslands of south-eastern Australia should not be ignored. 

Intensive digging took place in the temperate grasslands to harvest roots for food, particularly Murnong, Microseris spp., which 

was stockpiled and traded (Gott 1983), and Turrac (probably Pelargonium rodneyanum) (Lunt et al. 1998). Numerous other 

tuberous or bulbous grassland species were eaten (Gott 1999) including Vanilla and Chocolate Lilies Arthropodium spp. 

(Wigney 1994, Zola and Gott 1992), Bulbine Lily Bulbine bulbosa (R.Br.) Haw., Milkmaids Burchardia umbellata R.Br. and a 

range of other lilies and orchids (Zola and Gott 1992). Approximately one quarter of the vascular plant species recorded in the 

Victorian Basalt Plains were used by aboriginals, of which approximately 20% were used as food (Gott 1999). Over 100 plant 

species found in the grasslands and grassy woodlands of Tasmania are known to have been used by aborigines elsewhere in 

Australia (Kirkpatrick et al. 1995). Harvesting of roots resulted in improved aeration and water infiltration into the soil and 

nutrient incorporation from litter and ash, and would therefore have enabled better plant regeneration. Dug areas would have 

increased the availability of regeneration niches for many plants (Gott 1999). Recent studies have confirmed that this occurs with 

the orchid Diuris fragrantissima, where soil aeration, consisting of tilling to 20 cm depth, significantly increased the proportion 

of spring-planted plants that emerged and flowered (Smith et al. 2009). Aborigine thinned patches of the tuberous and bulbous 

food plants, deliberately replanted them and traded valued plant production between clans and tribes (Gott 1999). 

Burning in the dry season in Victoria may have enhanced the abundance and distribution of some plants e.g. Microseris spp. 

(Gott 1983). According to Flannery (1994) fire was critical in releasing nutrients into the system to enable regeneration of plant 

foods, however this may have been of limited importance in grasslands because of the limited nutrient reserves in above ground 

vegetation during a significant proportion of the fire season. Burning was probably more critical to keep grass biomass low: the 

tuber feeding people burnt “to expose the tubers, to improve their taste, and to keep grass sparse and give tubers and herbs 

space” (Gammage 2009 p. 286). Burning was frequent on the Victorian Volcanic Plain prior to European occupation, being used 

by aborigines in hunting and to encourage new growth, and probably assisted in maintaining treelessness (Stuwe 1994, DNRE 

1997). Fires were reported by early maritime travellers in Port Phillip region in all months from spring to autumn (Jones 1999b). 

Aborigines would have increased fire frequency above the background rate as a result of deliberate burning, accidental escapes 

from camp fires (Stuwe 1994) and possibily the use of fire as a weapon against invading non-indigenous people (Flannery 1994). 

109

But little is known anywhere in the world about the motives, scale and ecological significance of aboriginal fire management 

(Murphy and Bowman 2007). The palynological records of grasses in lake and swamp cores indicate that south-eastern 

Australian grasslands existed long before aboriginal occupation and are not of anthopogenic origin (Jones 1999b, Kershaw 

2000); rather, aboriginal activities modified existing grassy ecosystems and shifted their boundaries (Jones 1999b). 

The European explorer Thomas Mitchell thought that burning was critical to aboriginal management of country: “fire, grass, 

kangaroos, and human inhabitants seem all dependent upon each other for existence” (Mitchell 1848, cited by Murphy and 

Bowman 2007). Fire use in hunting had two aspects: as a direct tool to flush the animals or drive them towards waiting hunters, 

and habitat manipulation, in which the young green growth attracted the animals, increased the carrying capacity and made the 

prey easier to locate and kill (Murphy and Bowman 2007). Numerous ecological studies show that kangaroos are attracted to 

burnt areas but there has been little investigation of the mechanisms that cause this response. In studies in northern Australian 

savannah and seasonal wetland Murphy and Bowman (2007) found that the abundance of Macropus spp. scats was much greater 

in burnt than unburnt moist areas, and in unburnt than burnt dry rocky areas, from 4 weeks to 1 year post-fire. Kangaroos moved 

into the burnt moist areas away from burnt, dry rocky habitats. The resprouting grass foliage contained higher N concentrations 

than senescent foliage and may have provided macropods with better quality nutrition. Aboriginal burning may indeed have 

created a self-reinforcing cycle by creating a mosaic of habitat patches, very different to that subsequently developed by 

Europeans. 

Pastoral development had an immediate and devastating effect on the aboriginal population. Introduced livestock destroyed their 

prime feeding grounds and muddied and destroyed the waterholes and soaks (Zola and Gott 1992). Considerable numbers of 

aborigines were shot by European occupiers. Smallpox had already decimated the Victorian aboriginal population by 1835 and 

populations collapsed to an estimated 1,700 in 1871 and 850 in 1901 (Coutts 1982). The cessation of aboriginal fire regimes 

resulted in well-documented substantial change in vegetational structure, particularly involving increases in tree cover (Hope 

1994). 

European Management 

European occupation of Australia brought a novel range of exogenous disturbances that resulted in rapid, abrupt changes to 

Australian grasslands (McIntyre and Lavorel 1994). Aside from the changes in land use that must have occurred as imported 

diseases decimated aboriginal populations, sheep and cattle grazing was the first major European impact and the shift to frequent 

livestock grazing is generally recognised as the major cause of vegetation change (Dorrough et al. 2004). Grasslands were 

preferentially occupied by squatters and their livestock very early in the colonial period. Pastoral settlement commenced in the 

1820s in the Southern Tablelands of NSW and the ACT (Sharp 1997) and in the early 1830s on the Northern Tablelands 

(Johnson and Jarman 1975). In Victoria occupation commenced in the mid 1830s and there were 25,000 sheep in the colony 

before legal settlement commenced in 1836 (Mansergh et al. 2006a), over 41, 000 by May of the following year, along with 155 

cattle and 75 horses (Jones 1999b). Sheep numbers reached 700,000 by 1841 and doubled by 1843 (Gott 1983), by which time 

most of the grasslands in Victoria outside of Gippsland had been occupied (Jones 1999b). There were over 6 million sheep in the 

1850s, along with about 1 million cattle (Mansergh et al. 2006a). On the New England tablelands of NSW 66 stations occupied 

all of the best grazing land by 1840 and only a few more were added in rougher country by 1848 (Johnson and Jarman 1975). 

Thirty million sheep had been introduced to the grassy plains of Victoria and NSW by 1851 along with 1.7 million cattle and 

32,000 horses (Lunt et al. 1998). European occupation, at least in Victoria, probably coincided with a major climate shift to drier 

and hotter conditions (Jones 1999b) which probably exacerbated the impact of livestock. However the impact on native 

grasslands up until the 1860s were relatively mild compared with what followed (Scarlett and Parsons 1993). 

In Victoria the Selection Acts of the 1860s led to the alienation of large areas of Crown land and the advent of major cultivation, 

particularly cereal growing (Scarlett and Parsons 1993). In NSW the Land Act of 1861 allowed occupation of areas up to 259 ha 

before government surveys, and fencing began to be used instead of shepherding (Johnson and Jarman 1975). Land grazing 

under licence became legitimate in Victoria in the period to 1900. By 1891 livestock farming on large freehold properties was 

well established, particularly in the Western District (Powell and Duncan 1982). Landscape change was intensified by the 

extension and intensification of more efficient types of fencing (Powell and Duncan 1982). By 1910 pasture occupied c. 12 

million ha in Victoria, of which

of these systems is largely explained by the form and growth patterns of caespitose grasses, which develop by intravaginal 

tillering, the emerging tillers remaining erect inside the leaf sheaths. The bunching, erect form makes the plant more susceptible 

to ungulate grazers than the contrasting rhizomatous grasses, which have extravaginal tillering and axillary bud production. 

Ungulate grazers are able to reduce the reproductive potential of tussock grasses to a much larger extent than rhizomatous forms 

by grazing the emerging flower heads (Mack 1989). Another contributing factor was the perenniality of the dominant species, 

which “made annual re-stablishment unnecessary” (Evans and Young 1972 p. 231). For example in the intermountain west of the 

USA the once dominant Agropyron spicatum (Pursh) Scribn. required above average rainfall to establish, so significant 

recruitment events were relatively uncommon. 

In Australia the syndrome of decay had the following course. Continuous grazing by hard-hooved livestock preferentially 

removed the more palatable and sensitive intertussock herbs and the tall C 4 grass (T. triandra); fire exacerbated these losses; T. 

triandra was replaced by cool-season C 3 native grasses (such as Austrodanthonia spp.). Further grazing favoured short coolseason 

grasses and eliminated or greatly reduced the remaining palatable forb components, and loss of both these functional 

groups led to nutrient enrichment of the soil, particularly with N, which in turn allowed invasion by alien forbs and annual 

grasses (e.g. Vulpia spp.) of European origin, etc. (Moore 1973, Mack 1989, Moore 1993, Groves and Whalley 2002, Groves et 

al. 2003a). In temperate Australian grasslands, the main trends in plant composition have been from summer to winter-growing 

grasses, from perennials to annuals and from native to introduced species (Moore 1973, Stuwe and Parsons 1977, Mack 1989, 

Moore 1993, Groves and Whalley 2002). 

Intense grazing of T. triandra during its reproductive phase when it is mobilising nutrients from leaves to storage organs may 

have been the critical factor in its extensive demise (Dunin 1999). Greater palatibility compared with other native grasses, 

trampling damage to surface roots, reduced seed production and seedling establishment were probably additional important 

factors (Groves et al. 1973, Chan 1980). Alterations to drainage patterns and soil disturbance have also contributed to losses: 

Lunt (1990a) noted that T. triandra occurred at Derrimut Grassland only in well-drained areas that had not been ploughed, as 

well as areas subjected to no more than brief periods of heavy grazing. A similar mechanism to that reported by Grice (1993) for 

the proliferation of Austrostipa and Aristida spp. in the semi-arid woodlands of western New South Wales may be partly 

responsible: grazing did not cause greater mortality of the more desirable and long-lived grasses, rather, the plants avoided by 

sheep (Austrostipa and Aristida spp.) produced much more seed in grazed areas and so proliferated, while the more palatable 

native pasture grasses, were more fecund when ungrazed. T. triandra decline in inland New South Wales has been blamed on 

overstocking and low seed production (Whittet 1969). 

Native pasture in turn was developed by addition of fertilisers, and the sowing of exotic grasses and herbaceous legumes (Moore 

1973 1993). Naturalisation of Trifolium and Medicago species after accidental introduction and spread was important (Moore 

1993 p. 345) probably from the time of first settlement onwards. Addition of nutrients as a feature of grassland ‘improvement’ 

for agricultural grazing became widespread after 1929 when the Australian Government introduced a superphosphate subsidy 

(Mansergh et al. 2006a). The advent of cheap superphosphate coincided with government promotion of introduced C 3 perennial 

grasses and legumes (Trifolium spp. particularly T. subterraneum, Medicago and Lotus spp.), varieties of which were bred by 

State Departments of Agriculture for use in ‘pasture improvement’ programs, which usually involved cultivation (Groves and 

Whalley 2002). Seed of T. subterraneum first became commercially available in the early 1920s and that of T. fragiferum L. in 

1938, while cultivars of T. repens were first registered in the mid-1930s (Oram 1990). Use of Trifolium spp. and superphosphate 

increased rapidly in some areas in the mid 1930s (Browning 1954) and became commonplace during the 1940s and 1950s 

(Mansergh et al. 2006a). These changes raised the P and N status of the land to high, facilitating the invasion of new suites of 

weeds. The improved pastures, including a legume component, were able to support high intensity grazing, but required ongoing 

fertilisation and periodical resowing, and lifted productivity “in the short term” (Keith 2004 p. 105). Spread of the exotic grasses 

was the “desired outcome” sought by agronomists” (Cook and Dias 2006 p. 617) and some of these grasses escaped from the 

paddock and started to become major weeds of roadsides and eventually natural areas, including remnants of the natural 

grasslands - the weed potential of a species for a natural ecosystem being more or less equivalent to its hardiness, persistence and 

productivity values as a new pasture grass. 

These changes led to the eventual disappearance of the native grasses, particularly T. triandra, in many areas, although some 

Austrodanthonia spp. can re-occupy high-nutrient, grazed sites (Groves and Whalley 2002). Groves et al. (1973, echoed by Chan 

1980) thought that the mechanisms causing the loss of T. triandra remained to be properly identified, but listed a number of 

probable reasons including greater palatability to livestock than other native grasses, susceptibility of the adventitious roots to 

grazing damage at the soil surface, gradual exhaustion of underground reserves due to continuous shoot removal, low seed 

production and poor seed seedling establishment, and poor competitive abilities for light and nutrients. Chan (1980) 

demonstrated that repeated close (2 cm above ground) mowing at intervals of ≤3 months reduced yields and reproductive fitness 

of T. triandra, Austrostipa bigeniculata and Austrodanthonia spp., with the least affected of the native grasses examined being 

Bothriochloa macra because of its low habit and prostrate tillers. Similar results were obtained by Nie et al. (2009) on a range of 

native grasses cut at 3-5 week intervals to a height of 2, 5 or 10 cm. All species tested had reduced survivorship when cut to 2 cm 

height, but plant survival was least with the two C 4 grasses, T. triandra (c. 51%) and B. macra (c. 57%). Cutting to 5cm 

increased survivorship to c. 85% with T. triandra and >95% with B. macra. Cutting at 5 and 10 cm enabled T. triandra to 

increase its shoot biomass compared to the 2 cm cut far more than the other species. Furthermore, most of the species tested had 

little or no response to P fertilisation (Nie et al. 2009) so would be outcompeted by exotic pasture species when superphosphate 

was applied. 

The ecological and evolutionary circumstances that led to the dominance of summer-growing T. triandra in south-eastern 

Australian temperate grasslands prior to European occupation have not been adequately explained (but see Bond et al. 2008). 

Ostensibly the species appears to be poorly adapted as a dominant in grasslands that have winter rainfall maxima, spring growing 

periods and dry summers. This peculiarity of “a system growing partially out of phase with the rainfall regime” may explain why 

sheep and rabbit grazing led to rapid decline of T. triandra in south-eastern Australia and complete disappearance in south-west 

Western Australia (Moore 1993 p. 351). T. triandra may have been at a disadvantage compared with spring-growing exotic 

grasses because its demands for water are highest during the driest time of the year (Groves 1965, Mack 1989). However swards 

111

of T. triandra are able to more effectively trap rainfall and reduce runoff than swards of C 3 grasses, so soil moisture from spring 

rainfall appears to be effectively preserved for later use (Dunin and Reyenga 1978, Dunin 1999, Singh et al. 2003). Grazing may 

have resulted to general dessication of the system: reduced rainfall infiltration due to soil compaction, general opening of the 

sward and destruction of deep rooted perennials with access to subsurface soil moisture. 

Currently most of the pasture legumes and some of the deliberately introduced grasses have poor persistence under livestock 

grazing, due in part to the prevailing high variability and extremes of climate, difficulty in managing grazing intensity, and to 

their near-monoculture nature (pastures with one grass + one legume, or just a single legume species, with volunteer broadleaf 

species) (e.g. Madin 1993) which makes them more prone to pest and disease attack. Continued increased nitrification by 

legumes has enabled invasion by nitrophilous species such as thistles, while continued use of superphosphate has resulted in 

widespread soil acidification (Moore 1993). These ‘improved’ pastures are unstable, requiring ongoing inputs or renovation 

(McIntyre and Lavorel 2007). 

McIntyre and Lavorel (2007) presented a unifying conceptual model of the states and the transition processes for the grasslands 

of south eastern Australia that exemplifies historical chain of developments (Fig. 5). Transitions from one state to another occur 

as a result of specific managment activity (or the lack of it). Invaded grasslands with moderate or high exotic components have 

been permanently altered (Mack 1989) and appear to represent new metastable states that do not revert to their former status 

(Sharp 1997). Native pasture may revert back to native grassland, but its properties and composition will have been more or less 

altered. McIntyre and Lavorel (2007) introduced the concept of “enriched grassland”, to include those areas no longer cultivated 

or managed for grazing, but which remain nutrient enriched. These are largely dominated by robust exotic perennial pasture 

grasses including Paspalum dilatatum, Dactylis glomerata and Phalaris aquatica, along with exotic rosette-forming herbs such 

as Plantago lanceolata and Hypochoeris radicata. Much of the enriched grassland is currently in shelterbelts and other areas of 

recent tree and shrub planting (McIntyre and Lavorel 2007) although many former pastures on the edges of urban areas are in 

this state, and some disused pastures dominated by Nassella spp. can be included in this category. Little is known about the 

floristics, functioning and successional dynamics of enriched grasslands but they appears to be “an alternative stable state 

[requiring] a very high level of management to shift” (McIntyre and Lavorel 2007p. 15). Management inputs required include 

nutrient depletion, weed control (including the dominant grasses) and reintroduction of native species, particularly forbs. A 

similar set of problems is faced in extensification of former pastures to species rich grassland in Europe (Eschen et al. 2007). 

Figure 5. State and transition model for temperate grassy woodlands of south-eastern Australia. Each state is characterised by a 

level of livestock grazing, soil disturbance and soil fertility, straight arrows indicate mangagement factors associated with 

transitions between state and ciruclar arrows indicate management associated with maintenance of the state. Source: McIntyre 

and Lavorel 2007, p. 14. 

Fertilised and sown pastures are considered to be the most unstable states because of their ongoing requirements for inputs 

(McIntyre and Lavorel 2007). Removal of inputs results in transition to enriched grassland or to a native pasture. Transition from 

reference grassland to enriched grassland generally occurs accidentally through nutrient enrichment from fertiliser drift or water 

movement. Changes in composition and function of areas that have ‘regressed’ to native pasture are probable, but currently 

unknown. 

There are pronounced differences in scientific understanding of the states and processes. This reflects the historical agronomic 

approach to the study of agricultural land, and the use of floristic ecological techniques in natural grasslands. The overall 

floristics of sown pastures and of enriched grassland are very poorly known, while there is little data for native forbs in fertilised 

112

pasture. The least understood transitions are those that may be classed as restoration: from developed pastures through enriched 

grassland back to native grassland (McIntyre and Lavorel 2007). 

McIntyre and Lavorel (2007) tabulated known and hypothesised leaf, broad morphology and regeneration traits of the grasses 

associated with each of the vegetation states. The traits considered were for leaves: specific leaf area, N content, dry matter 

content, toughness and life-span; for morphology: size, form, annual/perennial; and for regneration: seed size (large, medium or 

small), flowering period (early-late) and presence or absence of vegetative regeneration. European experience has demonstrated 

the utility of such simple biological modelling in predicting and managing transitions between states. The traits are poorly 

known for the Australian grasses, but if found to be universally applicable they should enable prediction of the effects of land 

use change on ecosystem services. 

ACT Government (2005) provided a useful tabulation of the degree of disturbance corresponding with various levels of botanical 

significance in temperate grasslands of the ACT. Very low disturbance levels correspond with the highest botanical significance 

level. The ground layer includes the most sensitive species including orchids (Diuris spp. Caladenia spp., Thelymitra spp.) and 

lilies. Under low levels of disturbance the most sensitive species disappear but forbs such as Dichopogon spp., Bulbine bulbosa, 

Pimelea spp. and Wurmbea diocea survive, along with T. triandra. At moderate disturbance levels, sensitive species are rarely 

present, and the native herbs are generally disturbance tolerant, including Chrysocephalum apiculatum, Convolvulus erubescens, 

Plantago varia and Asperula conferta. High disturbance levels may contain a range of native grasses but T. trianda and most 

native forbs disappear. Very high disturbance sites are dominated by perennial exotic species or a low cover and diversity of 

native species, mostly grasses. These categories have reasonable correspondence with the states in the McIntyre and Lavorel 

model. Vey low and low disturbance = reference grassland. Moderate disturbance = native pasture. High disturbance = possibly 

improved native pasture. Very high disturbance = enriched grassland. 

Most of the native grasslands of south-eastern Australia have been irreparably altered by pastoralism and agricultural 

intensification, but large areas have been destroyed by other developments including urbanisation (Sharp 1997). Much of the 

remaining native grassland and native pasture is currently managed by grazing of livestock. The ecological effects of grazing 

will next be examined in more detail. 

Mammal grazing 

“The countless herds of horses, cattle, and sheep, not only have altered the whole aspect of the vegetation, but they have almost 

banished the gunaco, deer, and ostrich.” Charles Darwin, The Voyage of the ‘Beagle’, 19 September 1833, on the Argentine 

pampas in the Buenos Aires region. 

The grazing of domestic livestock has been more important than any other exogenous disturbance in altering the composition 

and structure of Australian temperate grasslands (Trémont and McIntyre 1994). “Grazing has a detrimental impact on 

communities with little history of grazing, but is necessary to maintain communities with a long history of grazing” (van Andel 

and van den Bergh 1987 p. 11). Sheep and cattle grazing initially caused increases in plant diversity in south-eastern Australia, 

resulting from the decline of the dominant native grasses, and invasion by exotics and native species from adjacent drier 

communities (Moore 1993). But ultimately under the nearly geographically uniform regimes of continuous grazing at set 

stocking rates, vascular plant diversity declined across large areas (Moore 1993). The irregular occurrence of severe drought, 

combined with plagues of introduced grazers, particularly rabbits, led to to episodes of intense overgrazing which have severely 

altered the natural grasslands. Overgrazed areas largely became “synthetic communities” (Trémont 1994 p. 511), lacking most of 

the orginal native plants and with a high proportion of exotic species. 

The long term detrimental effects of overgrazing by livestock on south-eastern Australian temperate grasslands need to be kept 

in perspective. Grazing reduces the competitive effects of dominant species and creates open ground suitable for plant 

colonisation, so in general results in increased diversity of plant species and functional groups, with annuals particularly 

favoured (Trémont 1994, Trémont and McIntyre 1994, Lunt 1995c, Overbeck et al. 2007). Grazing at low intensities has 

maintained vascular plant diversity over large areas of so-called native pasture. Trémont (1994) compared native pastures 

intermittently grazed by sheep from the time of European tree clearing in the mid 1800s to 1976, and then either left ungrazed 

for 16 years or grazed at a stocking rate of 6.7 sheep ha -1 over the same period. Both areas were dominated by native grasses and 

most species present were native perennial forbs. The grazed treatment had a more open canopy, more bare ground, much greater 

species richness including small forbs and grasses and more common exotic annuals, whereas the ungrazed treatment had dense 

grass and litter cover, was species poor, but had a higher proportion of native perennials and more vegetatively reproducing forbs 

species. Dorrough et al. (2004) compared areas that were frequently grazed (native pastures), infrequently grazed (travelling 

stock reserves) and minimally grazed (roadsides) on the Monaro Plains and found that infrequently grazed areas had the highest 

native and exotic richness 

The maintenance of high plant diversity in many grasslands is dependent on continued grazing, but negative biodiversity 

impacts, particularly on native species, may occur if grasslands have a limited evolutionary history of grazing (Hobbs and 

Heunneke 1992). Any change in the grazing regime may constitute a disturbance, and the impact of the changes on diversity and 

invasive species depends on their nature in relation to the historical regime (Hobbs and Heunneke 1992). As cogently observed 

by Trémont and McIntyre (1994 p. 646), “the absence of grazing stock from grassy communities is as important, in terms of 

community structure, diversity and composition, as its presence”. 

Large grazing mammals coevolved with grasslands in the Americas (Webb 1977 1978), Africa and Eurasia and probably 

Australia (Jones 1999b). According to Bouchenak-Khelladi et al. (2009 p. 2398) a “major and rapid radiation of vertebrate 

herbivores ... occurred between 20 and 10 million years ago ... along with a near simultaneous rise to ecological dominance of 

grasses ... suggesting that grasses coevolved tightly with vertebrate herbivores”. Occupation of C 4 -dominated grasslands by 

ungulates is inferred to have commenced c. 26 mybp, with occupation by Bovideae and Cervideae occurring in the early to late 

Miocene (23-5 mybp). Significant increases in silica density in C4 lineages, believed to be anti-herbivore defences that decrease 

113

palatability, digestibility and nutritional value, coincided with increases in ungulates with hypsodont characters, i.e. highcrowned, 

longer lasting teeth (Bouchenak-Khelladi et al. 2009). 

During the late Tertiary and the Pleistocene epoch of the Quaternary in Australia a wide range of herbivorous birds and 

mammals developed in association with the wide occurrence of grassland. These included the 2-3 tonne Diprotodon optatum, 

giant macropods Sthenurus, Procoptodon and Protemnodon, the small grazing diprotodon Palorchestes azael and open country 

flightless birds (Frith 1973, Hope 1994, Johnson 2009). But in all continents except Africa the open country megafauna largely 

became extinct during the late Pleistocene, and the grazing regimes were therefore completely changed, with major ecological 

repercussions (Johnson 2009). The Australian herbivorous mammalian fauna was more severely affected than the Americas and 

Africa, losing almost all species over 100 kg in weight (Johnson 2009). Some 20 of the 50 Australian species that disappeared 

were grazers, with Macropus spp. being the largest survivors (Jones 1999b). Loss of the megafauna due to aboriginal overkill, as 

argued by Flannery (1994) remains contentious, but the major role of hunting in the extinctions throughout the world now 

appears to be widely accepted (Johnson 2009). The extinctions appear to have occurred over a long period from 50-30 kybp for 

the larger species, through to c. 10 kybp for the smaller (Hope 1994). Kershaw et al. (2000) argued that people coexisted with 

the megafauna for many thousands of years, but conceded (pp. 502-503) that recent redating of significant fossil megafauna sites 

largely reduced the overlap, making it unlikely that climate or habitat change were to blame. Loss of the megafaunal grazers 

would have resulted in major change in the vegetation (Flannery 1994), particularly by reducing tree and shrub herbivory, as in 

African savannas (Hope 1994). Knowledge of megaherbivores indicate that they effectively acted as ecological engineers. Their 

extinction has likely led to the replacement of large areas of open vegetation with closed, woody formations, a decline in the 

heterogeneity of vegetation across a range of scales, increased fire, primarily due to increased fuel loads and possibly resulting in 

new areas of uniform grassland, and the decline of coevolved plants, including the species they dispersed or which had evolved 

defences against them (Johnson 2009). 

Grazing of sheep and cattle is now the major form of management in a large range of native grasslands, notably in the Australian 

Capital Territory (ACT Government 2005) and in many privately owned native pastures (particularly in southern NSW), and is a 

substitute for burning to reduce the biomass of dominant grasses (Stuwe 1994). Moderate or more intense grazing pressure can 

be more effective than fire in maintaining bare ground and the low biomass of dominant grasses required to maintain high 

diversity of indigenous annuals and prostrate species, but disadvantages the taller, upright herbs (Kirkpatrick et al. 1995). Native 

pastures near Armidale NSW ungrazed for 16 years had a species richness of only 20 species per 30 m 2 because of large grass 

tussocks, dense grass litter and little bare ground, while areas grazed continuously by sheep at moderate to low stocking rates 

were much shorter, with small tussocks, little litter and abundant small bare areas, and had a diversity of 36 species (Trémont 

and McIntyre 1994). Based on experimental results from disturbance treatments Sharp (1997) recommended retention of 

livestock grazing as a management regime in the ACT, particularly at sites with high diversity of small intertussock forbs. Thus 

it appears that sites that have been managed by livestock grazing for long periods will degrade if grazing is removed, but can be 

managed without significant biodiversity loss by continuing the established grazing regime. 

The effects of introduced livestock grazing on native grasslands are complex, being dependent, inter alia, on the grazer species, 

the intensity and duration of the grazing pressure, and interactions with other environmental and management factors 

(Kirkpatrick et al. 1995, Lunt 1995c, McIntyre et al. 1995, Aguiar 2005). Some of the most basic components of grazing impact 

in Australia have been poorly understood and await adequate investigation (Johnston et al. 1999). For instance conceptions of 

what plants are important fodder have significantly changed historically. White (in the foreword in Leigh and Mulham 1965, p. 

v), commenting on the findings of pastoral research in the NSW Riverina, wrote that the pastoral research group had investigated 

“the question of which species are important at various seasons of the year. Often thirty or more species are available for 

selection and the extent to which the sheep exercises this opportunity is quite surprising. In many instances, in spite of 

appearances, inconspicuous plants or rather unattractive shrubs are more important to sheep than was formerly believed”. Lack 

of understanding of appropriate grazing regimes for natural grassland management continued into the period when grassland 

preservation became a cause célèbre: “Currently it is not known if it is preferable to maintain a continuous, light grazing regime 

or to graze intermittently, either lightly or heavily ... avoiding peak flowering and seeding times for native plants ... It may be 

that particular sites require different regimes, on the basis of the different species mix present, type of exotic infestation, drainage 

or other factors” (Sharp and Shorthouse 1996). 

One effect that was believed to result in the deterioration of native grassland as a productive pasture was depletion of soil P 

through the harvest of livestock and from continual removal of wool (Wadham and Wood 1950). However Groves et al. (2003a) 

considered nutrient addition from continuous grazing to be a major factor in historical changes in grassland species composition, 

initially due to increased nutrient mobilisation through soil disturbance and death of native plants, and increased returns from 

faeces and urine, later through the addition of fertiliser and the sowing of legumes. Grazing adds highly labile mineral N direct to 

the soil, leads to higher average plant tissue N concentrations, decreased root: shoot ratios and sometimes less efficient N-uptake 

(Wedin 1999). Deposition of urine and faeces also alters the cycling rates of other nutrients. Nutrient enrichment and soil 

compaction associated with livestock camps in native grasslands favours weeds (Kirkpatrick et al. 1995). 

Mammalian herbivore grazing impacts on patterns of recruitment, production and mortality of grazed and ungrazed species, on 

plant resource use and on resource availability. The most direct effect is removal of and damage to the above-ground parts that 

are consumed or trampled (Kirkpatrick et al. 1995), but the effects on plants that are not eaten may be among the largest impacts. 

Interactions between defoliation and release from competition can result in complex changes in structure and composition 

(McIntyre et al. 1995). Grazing alters the composition of the community by differentially altering the population density, genetic 

composition and structure of the component plants (Aguiar 2005) and by creating bare patches suitable for colonisation 

(McIntyre et al. 1995). It may also affect the structural diversity at the community level, although this possibility has been 

widely ignored (Aguiar 2005). Status as a grassland may be dependent on grazers that differentially destroy juvenile trees and 

shrubs, preventing reversion to woodland or shrubland (Hobbs and Heunneke 1992). The accessibility of perennating buds to the 

grazing animals is an important determinant of survival, so suvival of hemicryptophytes (with buds close to ground level) tend to 

decline only when grazing is intense (McIntyre et al. 1995). One immediate structural effect is that green leaf material becomes 

more concentrated close to the soil surface (Soriano et al. 1992). Plants that avoid grazing because of their small size or height 

114

tend to benefit more from relaxed competition under grazing pressure (McIntyre et al. 1995). In pampas grasslands, livestock 

grazing markedly reduces the average distance between plants by dividing large tussocks into multiple smaller plants (Soriano et 

al. 1992). Grazing may affect larger-scale patterns in vegetation including the ‘dual-phase mosaic’ or ‘banded vegetation’ of 

semi-arid rangelands, which consist of patches or bands of high cover vegetation in a matrix of low cover with different 

dominant plants; and since the matrix controls the hydrology and water availability in the patches, any disruption to it can impact 

the whole system (Aguiar 2005). At the small scale, grazing tends to reduce the biomass of dominant grasses, enabling higher 

vascular plant diversity. It commonly enables continued existence of the range of native grassland species, but allows the entry 

of a pool of exotics (e.g. Soriano et al. 1992 p. 392). Grazing in flooding pampas grasslands has increased plant diversity at the 

stand scale in this way, but has decreased it at the landscape scale because the exotics are mainly generalists with wide 

environmental tolerances (Perelman et al. 2001). 

Grazing can exert strong selective pressure by altering the population structure of grazed and ungrazed species. Populations that 

are reduced in size can lose significant genetic diversity through genetic drift and inbreeding depression, while palatable species 

may evolve traits that provide grazing resistance or enable grazing avoidance (Aguiar 2005). Development of pastoral agriculture 

in the South American pampas has led to an increase of species that produce toxic secondary metabolites such as alkaloids and 

terpenoids (Aguiar 2005), and similar concerns have long been a focus of pastoral weed research in Australia. 

Grazing directly affects litter degradation rates, and, by altering the floristic and growth form composition of the vegetation by 

removal of palatable and less grazing-tolerant species, alters litter makeup, decay rates and nutrient turnover (Villarreal et al. 

2008). Reduction of litter by grazing reduces fuel loads and the probability of wild fire (Aguiar 2005). 

Trampling can create openings for seedling establishment, reduce the dominance of tall growing species or directly reduce 

fragile species (Hobbs and Heunneke 1992). Direct evidence of trampling damage has been noted for example by Archer (1984) 

who recorded cattle and feral horse damage to colonies of Thesium australe (although not in temperate grassland). 

Grazing animals also disperse seeds of native plants and exotic weeds, in their dung and externally (Hobbs and Heunneke 1992), 

discussed in more detail below. Other less direct effects include soil disturbance and compaction, destruction of the cryptogam 

crust, redistribution of nutrients via urine and dung, and the creation of bare ground (Mack 1989, Kirkpatrick et al. 1995. Sharp 

1997). All of these changes can facilitate exotic grass invasion. The effects can be highly concentrated by congregation of 

herding livestock, resulting in wide variability in the spatial arrangement of damaged patches (Hobbs and Heunneke 1992). 

Studies in arid Australia have shown that sometimes >50% of the grazing in a paddock occurs in 30% of annual production is 

removed by grazing animals (Mott and Groves 1994). Drought periods are common in many Australia grassland areas and 

throughout the historical record, stocking rates have commonly not been altered to correspond with the accompanying reduction 

in forage production (Mott and Groves 1994). Continued grazing at accustomed levels during periods of climatic stress and 

alterations in other disturbance regimes can result in transformative shifts in grassland state (Aguiar 2005). 

Grazing reduces the insect biodiversity of grasslands by simplifying the structural complexity of the plant components, with 

different major taxonomic groups affected in different ways (Tscharntke and Greiler 1995). Invertebrates can suffer particularly 

negative impacts from trampling (Hobbs and Heunneke 1992). Greenslade (1994) found indications that grazing has a strong 

impact on the composition of the Collembola fauna of ACT grasslands, one of the important detritivore groups, reflecting a 

general trophic cascade through the system. 

Sheep and cattle 

Audas (1950 p. 472) considered many Australian native grasses to be “excellent pasture or fodder plants ... equal, and in some 

cases superior to, the cultivated exotic kinds”, including T. triandra, Austrodanthonia penicillata, Microlaena stipoides, eight 

species of Adropogon, and fifteen species of Panicum (Mitchell Grass and Umbrella Grass) (“splendid fodder” op. cit. p. 472). 

He noted the “rich, succulent and varied character” of indigenous pasture during spring and summer but the paucity of green 

foliage in winter. However quality is mainly determined by environmental conditions, and all species, native or exotic, have 

periods in which their forage quality is low (Johnston et al. 1999). The lack of cool season feed provided by native grasses 

contributed to a strong focus on the introduction, breedings and widespread planting of exotic C 3 grasses for pastoral use in south 

eastern Australia. A widespread perception developed from the late 1950s, based on unreplicated and otherwise biased studies, 

that native species were of little pastoral value (Johnston et al. 1999). 

Long term continuous grazing by introduced bovid livestock, primarily by sheep Ovis aries and cattle Bos taurus, has resulted in 

major changes in the vascular plant species composition of grasslands throughout Australia (Moore 1993, Trémont 1994, 

Kirkpatrick et al. 1995, Groves and Whalley 2002). The most immediate effect of continual grazing, particularly by sheep, was 

to suppress or eliminate the most palatable and most easily damaged plants, which were eaten, trampled or failed to regenerate. 

Intertussock herbs were the first to disappear (Stuwe and Parsons 1977, Groves and Whalley 2002, Groves et al. 2003a). 

Murnong, Microseris spp., a staple food of aborigines, and once very abundant on the open plains, was greatly diminished or 

eliminated in some areas by 1846 due to depradation by sheep, which learnt to root up the whole plant, or continually defoliated 

it (Gott 1983). This species was also highly palatable to rabbits (Lunt 1996). Of 59 Victorian grassland sites sampled by Stuwe 

and Parsons (1977) Microseris was present at only one, an ungrazed railway grassland. Other Asteraceae including Senecio 

macrocarpus Belcher and Rutidosis leptorrhynchoides were unable to tolerate heavy grazing and are now severely depleted 

(Morgan 1995a, Humphries and Webster 2003, Hills and Boekel 1996). The grossly contracted range of R. leptorrhynchoides 

known in the mid 1990s consisted entirely of areas protected from domestic livestock (Morgan 1995a). Native legumes were 

probably also heavily affected: e.g. grazing and trampling by cattle is considered an important current threat to Psoralea parva 

(Muir 2003). By the 1990s, all remnants of Victorian basalt plains grassland with high vascular plant richness had been protected 

from livestock grazing for decades (Morgan 1998c). The fragile cryptogam crust also sufferred major early impact. 

115

The grass component was more resilient than the forbs because grasses are less palatable, grass leaves continue to elongate from 

the base, dormant buds that develop lateral shoots are rapidly induced after destruction of apical meristems, and much of the 

meristematic tissue is located in the crown and is protected from grazing damage (Tscharntke and Greiler 1995). In general, 

rhizomatous and prostrate grasses are the least affected by bovid livestock (Mack 1989) or are favoured, so that grazed 

grasslands tend to become two-layered – a short layer of grazed species and a discontinuous taller stratum often dominated by 

caespitose Poaceae and other unpalatable species (Overbeck et al. 2007). In south-eastern Australian native pastures the 

‘layering’ in the more heavily grazed grasslands tends to have a temporal rather than spatial dimension, with small short lived 

annual grasses such as Vulpia forming a large component of the living biomass in spring. 

Most of the historical grazing regimes in Australian temperate grasslands have resulted in major alteration to the grass 

components. Throughout the world, selective herbivory of palatable grasses by domestic livestock is a major cause of species 

replacement (Mack 1989, Moretto and Distel 1998). T. triandra is regarded as a “good forage” in native pastures, although 

“mature plants are neglected” (Chan 1980 p. 22), and Audas (1950) warned of the danger of T. triandra being “eaten out” if 

grazed in spring and summer. However T. triandra has a high C:N ratio, which makes it less palatable than many other grasses 

(Moretto and Distel 2002). It is lost along with other native perennial grasses when there is continuous heavy grazing by sheep 

(Stuwe and Parsons 1977, Chan 1980) and it is more susceptible to elimination when grazing follows fire (Groves and Whalley 

2002). Heavy grazing by sheep or cattle causes more damage to T. triandra than that by macropods or horses (Kirkpatrick et al. 

1995). In the Australian Alps, cattle “prefer inter-tussock herbs (such as Craspedia spp., Celmisia spp. and Leptorynchos 

squamatus subsp. alpinus) but make up bulk in their diet with tussock grasses (Poa spp.)”, so they spend more time in grassland 

communities than other vegetation formations (McDougall and Walsh 2007 p.6). Discontinuous bovid grazing however enables a 

high proportion of native species to survive (Kirkpatrick et al. 1995). Losses of native grasses can continue even after pastures 

become highly degraded by invasion of exotic perennial grasses: data of Badgery et al. (2002) indicated declines of (un-named) 

perennial C 4 grasses under sheep grazing at 4.5 DSE ha -1 in areas with up to 50% coverage of Nassella trichotoma, along with 

declines of about 30% with grazing plus fertiliser (120 kg N and30 kg P ha -1 y -1 ). Stafford (1991) found that remnant T. triandra 

in the East Torrens region of South Australia was generally found at sites with a long history of stock exclusion. However T. 

triandra, Poa and Austrodanthonia may survive long after the intertussock herbs have been eliminated (Groves et al. 2003a). 

T. triandra is eventually replaced by exotic winter-growing species adapted to trampling and close grazing (Moore 1973, Gott 

1983, Moore 1993, Kirkpatrick et al. 1995). At Derrimut, Victoria, N. neesiana presence and density is strongly negatively 

correlated with that of the dominant grass (T. triandra) and is probably a long-term result of previous heavy grazing and 

ploughing (Lunt and Morgan 2000). Grazing exclusion can lead to the dominance of tall caespitose grasses in the pampas 

(Overbeck et al. 2007). Long term grazing results in forb loss also in areas that are both burnt and grazed (Lunt 1990b 1997, 

Morgan 1997). Intermittent grazing has less severe effects on species composition and may alter or reverse the changes in the 

general degradation syndrome (Groves et al. 2003a). Hadden (1997) suggested that suitably modest regimes were 2DSE ha -1 

during summer and autumn in Western Plains Grasslands, and 1DSE ha -1 during winter and summer and possibly parts of 

autumn in the drier Northern Plains. 

Loss of intertussock plant species diversity also occurs in T. triandra grasslands that are protected from ungulate and rabbit 

grazing, and not frequently burnt. In the Western Basalt Plains, Hadden (1998) and Hadden and Westbrooke (1999) found a 

significant decline in herb species in plots protected from sheep and rabbit grazing in native pasture but an increase in cryptogam 

cover. Decline of herbs was due to T. triandra canopy thickening and closure over 2 years. Increase in cryptogams was attributed 

to absence of soil trampling. Native grass cover of ungrazed areas reached c. 80%, compared to 35% in grazed areas, while total 

plant biomass reached 3,575 kg ha -1 compared to 718 kg ha -1 in grazed areas. Trémont (1994) found that grazed grasslands in 

northern NSW contained greater native plant richness than ungrazed grasslands, probably as a result of suppression of dominant 

grasses (Sharp 1997). Hadden (1998) found no significant changes in botanical composition between grazed and ungrazed plots 

in a Victorian Northern Plains grassland when grazing was excluded. 

Bovid grazing continues to threaten grassland remnants, notably along roadsides used for droving and grazing in drought 

conditions, e.g. to Comesperma polygaloides (McIntyre et al. 2004). 

Gap creation intensifies with increased grazing pressure and pasture gaps are subject to less root and shoot competition and thus 

favour the survival of grass seedlings (Moretto and Distel 1998). In native grassland in Argentina Moretto and Distel (1998) 

found that gaps in the vegetation dominated by Nassella clarazii (Ball) Barkworth, characterised by low competitive pressure, 

enabled seedling establishment of the unpalatable stipoid grasses Jarava ichu Ruiz and Pav and Nassella tenuissima. However 

they did not compare regeneration of the palatable species, N. clarazii, so failed to test their stated hypothesis, that such gaps 

favour the unpalatable species. Creation of such gaps by overgrazing of the palatable species was nevertheless suggested as the 

mechanism enabling establishment of the undesirable grasses. 

Bovid livestock are hard-hooved and weigh from 40 kg (Capra hircus) to nearly 1 tonne (Bos taurus) (Groves 1989) Their 

activity thus causes soil compaction, exacerbated with proximity to watering points, which they must visit regularly (Moore 

1993). Physical effects include increases in bulk density and bearing capacity and decreases in hydraulic conductivity (a measure 

of infiltration) that are directly related to stocking rate (Willatt and Pullar 1983). Hoof pressures of >160 kPa for cattle and 64- 

100 kPa for sheep have been calculated for animals standing flat on four feet, and higher pressures result when animals are 

moving. Comparable pressures from tractors are 30-150 kPa (Willatt and Pullar 1983). Biological effects of compaction include 

significant decreases of arthropods and reduction of earthworm body weight and numbers (Brown 1987). In the early days of 

ungulate grazing, the basalt soils at Sunbury, Victoria, were changed from loose to hard by continuous trampling (Gott 1983). 

Thus Microseris spp., which preferentially germinates in loose soil (Gott 1983), was affected. Soil compaction makes harder 

soils, and many studies have demonstrated that root growth is negatively affected as penetration resistance increases (Willatt and 

Pullar 1983). 

Disturbance of the soil and damage to the cryptogam crust favours exotic species over native (Stuwe 1994). With high levels of 

grazing the crust is often completely eliminated except where protected from trampling (e.g. along fence lines), but with sheep 

grazing at or below about 2.2. sheep ha -1 the soil crust is discontinuous and still noticeable (Scarlett 1994). 

116

Bovids crop their food between the incisor-canine row and a hard pre-maxillary pad, with sheep cropping directly, close to the 

ground through a cleft in the upper lip, or after the fodder has been pulled into the mouth by the tongue (Groves 1989). They 

graze with a tearing motion, as opposed to the shearing, scissor-like motion of the incisors of marsupials and thus tend to pull 

whole plants from the ground, so have a more detrimental impact on perennial grassland species like Psoralea parva than 

macropods (Muir 2003). Sheep graze more closely to the ground than cattle and are more selective so can be more damaging to 

natural grassland (Moore 1993). They can prevent regeneration of chenopod dominants in drier Atriplex-Maireana grasslands 

(Moore 1993). Carr and Turner (1959) found that on the Bogong High Plains of Victoria Poa inflorescences were highly 

palatable to cattle but the mature leaves were not. Cattle feeding on snowgrasses generally pulled parts of tussocks out of the 

ground, causing damage to swards. Archer (1984 p. 85) found that cattle and horses not only ate the endangered Thesium 

australe “but tended in the process to uproot them or break them off at ground level”. 

The impact of feral sheep on vegetation and landscape in the absence of other livestock has been well documented for the 

Mexican island of Socorro by Walter and Levin (2008). Sheep were introduced in 1869 and have been the key cause of 

ecosystem degradation. Sheep trampled and pulverised the soil to dust, ate even tiny seedlings in overgrazed areas and 

transformed the native forest and woodland into open habitats with a mix of native and exotic vegetation. The half of the island 

without sheep was found to contain only one exotic vascular plant species, compared to 44 spp. in the sheep impacted half which 

contained many hectares of denuded ground and was susceptible to severe erosion. Sheep grazing favoured some poisonous 

unpalatable plants, enabled the rapid spread of some exotic grasses and was a causative factor in the decline and replacement of 

endemic birds. 

Slashing, with removal of slashed material, and fire can be interpreted as grazing surrogates in so far as they result in removal of 

herbage. Slashing of T. triandra grassland once in April or twice in successive Aprils made no significant difference to 

intertussock gap distance at Derrimut compared with similar burning treatments (Henderson 1999). Stuwe and Parsons (1977) 

found that the only plant species more common in grazed sites than frequently burnt (railway) sites or roadsides were exotics: 

two spp. of Briza and Centaurium spp. 

Insect faunas are generally impoverished under increasing intensties of livestock grazing (Samways 2005). Grazing alters 

invertebrate habitat by affecting plant components, vegetation structure, microclimates, litter and soil properties (Yen 1999). On 

the New England tablelands fertilised, improved pastures grazed by sheep have been found to have a higher proportion of exotic 

invertebrates than native pastures (Yen 1999 see his reference). Hadden and Westbrooke (1999) examined the effects of removal 

of sheep and rabbit grazing on the Formicidae, Coleoptera and Araneae fauna of long-grazed native T. triandra pasture near 

Ballarat, Victoria. 6 m x 6 m plots were fenced out of the pasture, grazed by sheep at 3 DSE/ha, for 2 years. Of the 7 most 

abundant ant species, the abundance of 2 species significantly declined, as did the abundance of the hot-climate functional group, 

a result attributed to increased plant cover. Of the most abundant Coleoptera, significant increases were found with 2 spp. and 

significant decreases with 3 spp., due to a complex of effects. The nominal “decomposer” functional group (comprising only 

Anthicidae) decreased. The two most abundant species of Araneae significantly declined in ungrazed plots. Changes in the 

proportions of spiders were related to structural change in the vegetation. As part of the same study Hadden (1997) found no 

significant differences in the richness or abundance of the same taxa individually or when grouped between ungrazed and grazed 

plots on the Northern and Western Plains over the whole two year period. Removal of grazing appeared to offer no short term 

improvements in the biodiversity of these groups. The fauna appeared to be well adapted to the grazing regime. 

Rabbits and hares 

The effects of overstocking with sheep and cattle are difficult to distinguish from those caused by large populations of European 

Rabbits, Oryctolagus cuniculus Lilljeborg, in Australia (Rolls 1984) and it is impossible to retrospectively separate their effects 

for the purposes of determining ecological history (Frith 1973). European Hares Lepus europaeus Pallas, have had a lesser 

impact and their effects are also confounded with those of livestock. Although hares are approximately twice the weight of 

rabbits, the two species are trophic competitors: both are herbage-feeding lagomorphs which include woody twigs in their diet 

under extreme conditions during summer. Both have digestive systems with caecal fermentation of digesta, both re-ingest soft 

faeces (caecotrophs) and daytime hard faeces and both are poor digesters of cellulose (Stott 2007). 

Large populations of rabbits first became widely established in Australia on the Victorian basalt plains near Winchelsea in 1859 

and by 1890 had occupied all suitable habitat in Victoria (Menkhorst 1995d). The first reports of pastures being “eaten out” were 

made in 1868 (Rolls 1984 p. 30). The impact on native grasslands was ecologically disastrous e.g. factories at Colac and 

Camperdown canned half a million rabbits from the stony rises in 1881 (Rolls 1984 p. 53) and a factory was opened in Hamilton 

in 1892 which processed 720,000 animals in 1894 and a million in the first eight months of 1895 (Brown 1987). Rabbits had 

spread widely on the Central Coast of NSW by 1883 and were present on the New England tablelands in 1862 after release in 

1854, but did not become abundant there until the early 20th century (Rolls 1984), with significant impacts in the Armidale 

region from c. 1909 (Johnson and Jarman 1975) and plague proportions in the Glen Innes district by 1920 (Cameron 1975). 

Areas of the NSW western plains and Riverina were occupied in 1879 and suitable areas in the rest of the State were slowly 

occupied by 1925, at a rate of about 15 km per year (Rolls 1984). Rabbits occupied the burrows of wombats in wetter areas, 

bilbies in the inland (Rolls 1984) and burrowing bettongs Bettongia lesueur (Quoy and Gaimard) (Noble 1993, Noble et al. 

2007). 

Rabbits extensively graze grasslands, reducing total herbage yield markedly, altering floristic composition and facilitating weed 

invasion (Long 2003, Bloomfield and McPhee 2006). Cameron (1975 p. 22) considered it “clear” that they “eradicated a number 

of the species of better natural grasses” in the Glen Innes district of NSW. Depradation of rabbit populations following the first 

myxomatosis epidemics resulted in the appearance of grassland on previously bare and stony lands (Frith 1973). 

Rabbits are selective feeders, with a high rate of feed intake, rapid gut passage, selection excretion of fibre and selective 

retention of non-fibre constituents for microbial fermentation in the hind gut (Stott 2007). Rabbits graze very close to the ground 

(Long 2003) and preferentially eat new seedlings during autumn and winter, in spring increasingly consume grass flower heads 

and leaves of broadleaf species and later the inflorescences of other dicots, and in summer the green feed of summer-growing 

native grasses, which may be eaten ‘as fast as they grow’, along with inflorescences and roots of Trifolium spp. (Menkhorst 

117

1995d p. 280). Rabbit grazing preferences reportedly result in increases of less palatable grasses and weeds (Frith 1973). In the 

Riverina the combined effects of rabbit and sheep grazing “practically eliminated” T. triandra (Lunt et al. 1998). Annual grasses, 

such as Hordeum spp. often proliferate where rabbit damage has been severe (Bloomfield and McPhee 2006). 

Rabbit grazing is a recognised threat to a range of endangered vascular plants in temperate grasslands. These include Senecio 

macrocarpus (Hills and Boekel 1996 2003) and Comesperma polygaloides (McIntyre et al. 2004). Grazing by rabbits or 

European hares Lepus capensis L. is a threat to remnant populations of Swainsona spp. in the Wimmera and Northern Plains of 

Victoria (Earl et al. 2003). Thesium australe, once common in grasslands, is also threatened by rabbits (Archer 1984). 

Rabbit grazing, like that of bovid livestock, can also be beneficial by reducing cover of grasses. Lunt (1990d) thought it may 

have contributed to the high richness of vascular plants in herb-rich woodlands he surveyed in western Victoria. 

Rabbits and the rabbit industry had a severe impact on grassland animals. Large number of bandicoots were killed in rabbit traps, 

by poisoning, and the fumigation and ripping of burrows and the spread of rabbits probably assisted the proliferation of foxes, 

which after the introduction of myxomatasis in the early 1950s, had to switch their prey preferences to include a larger 

proportion of native animals (Brown 1987). Paradoxically, for a time, the availability of rabbit burrows for shelter may have 

aided the survival of the Eastern Barred Bandicoot Perameles gunnii Gray after destruction of tussock grass cover (Brown 1987). 

Rabbits “not only till the soil they fertilize it as well” (Bloomfield and McPhee 2006 p. 150): rabbits displace large quantitites of 

soil by digging, create bare areas around their warrens and deposit large quantitites of dung at discrete latrine sites, which 

become nutrient enriched. These areas seasonally support dense patches of weeds and can function as establishment foci for new 

weeds (Hobbs 1989, Bloomfield and McPhee 2006). 

Rabbit plagues accompanied by livestock grazing have severely affected grasslands in the Victorian Wimmera and Murray 

Mallee (DNRE 1997) and overgrazing by rabbits is still a problem in native grasslands west of Melbourne (Brereton and 

Backhouse 2003) and elsewhere. 

Hares select higher quality food than rabbits, consume much more food per unit body weight and produce much more faecal 

material, but have relatively smaller stomachs and caeca and retain a significantly smaller proportion of poorly digestible 

material (Stott 2007). They forage much further from cover than rabbits (Stott 2007). European hares were introduced into 

Victoria in the 1870s and spread through much of south-eastern Australia “particularly in the better-grassed areas” (Frith 1973 p. 

81). In the Glen Innes area of the Northern Tablelands the first sightings occured in 1889, numbers increased steadily through the 

1890s with a bounty first offerred for scalps in 1893, but populations declined and ceased to be problematic by 1920 (Cameron 

1975). A bounty was offerred in the Armidale district by 1902 and scalp returns peaked in the middle of that decade, decling to 

c. 30,000 by 1914 when the bounty was removed due to budgetary problems (Johnson and Jarman 1975). The grazing habits of 

hares in Australia have not been studied (Frith 1973, McDougall and Walsh 2007). 

Marsupials 

South-eastern Australian grasslands were once occupied by a diverse array of marsupial herbivores (see section below on the 

mammalian fauna of grasslands) of which nearly all are now absent. Little is known about the intensity or nature of marsupial 

grazing (Ellis 1975, Morgan 1994). Moore (1993 p. 351) stated that southern Australian T. triandra grasslands “evolved under 

light and intermittent grazing by native marsupials”, but this statement conceals a pronounced level of ignorance and is probably 

a misleading generalisation. Marsupials would have consumed significant quantities of plant material and maintained a much 

more rapid and patchy cycling of nutrients than occurs today in ungrazed grasslands (Flannery 1994). Kangaroos would have 

been very numerous, particularly in areas with adjacent tree cover (Willis 1964, quoting George Russell, an early settler). Early 

historical records in Tasmania describe ‘marsupial lawns’ created by heavy marsupial grazing, and such closely grazed areas 

exist today in areas where large numbers of Macropus spp., Thylogale billardierii (Desmarest) and Vombatus ursinus (Shaw) 

occur, as well as in areas grazed both by sheep and marsupials (Kirkpatrick 2007). 

Little information appears to be available about the ecological functioning of marsupial grazers in native grasslands and their 

impact on exotic grasses. Ellis (1975) mentioned studies in north-eastern NSW that found that nine sympatric species grazed 

both native and introduced plants and maintained niche segregation by differences in habitat utilisation. Cameron (1975 p. 20) 

reported that kangaroos in the Glen Innes area of the Northern Tablelands “seemed to prefer the green shoots on the kangaroo 

grass in the late spring and summer”, but there were “signs that [they like] the imported grasses and ... numbers will increase” (p. 

24). Robertson (1985) investigated the interactions between the most important and widespread of the extant marsupial 

herbivores, the Eastern Grey Kangaroos Macropus giganteus in open grassy woodlands at Gellibrand Hill, Victoria. The diet of 

M. giganteus was found to consist almost entirely of monocots, mainly grasses. Green shoots and new growth were selectively 

eaten and the species selected tended to be those with the highest nutritional value at a particular time. From late spring to 

autumn T. triandra was by far the most important dietary constituent, but other warm season grasses were also eaten. Cool 

season grasses were the main forage during other periods, with Austrodanthonia spp., Microlaena stipoides and a range of exotic 

species being the most important. Grass inflorescences were readily consumed, particularly those of the exotics Lolium rigidum 

and Briza maxima L., but species with long awned and sharp seeds such as Austrostipa spp. and T. triandra were usually 

avoided. Poa sieberiana Spreng. and Austrodanthonia spp. , both large grasses with relatively fine leaves, were largely avoided 

once they developed large amounts of attached dead litter, but T. trianda plants with high standing litter were not. Burning 

provided more palatible and accessible fodder. McIntyre (1995) noted that grassland sites in National Parks on the Northern 

Tablelands of NSW were rich in native vascular plant species, because they had marsupial grazing, an effect resulting from 

release from competition with the dominant grasses (Morgan 1994). Lunt (1990d) thought kangaroo grazing may have 

contributed to the herb richness of woodland vegetation in the Grampians and Langi Ghiran in a similar way. Murphy and 

Bowman (2007) referred to a study in semi-arid western NSW which found that 16 months post-fire, areas burnt but ungrazed by 

kangaroos had four times the abundance of Austrostipa variabilis (Hughes) S.W.L. Jacobs and J. Everett than areas that were 

kangaroo-grazed, suggesting a feedback loop between fire, kangaroos and the dominant grass. 

The toes of macropods are padded and soft in comparison with the feet of bovid livestock, and extant large macropods have 

masses much less than cattle (large male Macropus giganteus 80 kg: Hume et al. 1989) and more comparable with sheep. They 

118

are therefore less likely to disturb the soil and damage the cryptogam crust. The digestive system and digestive physiology of the 

large macropods have many similarities to those of ruminants (Frith 1973), but they are much better able to process high fibre 

grass than ruminants, whose intake increasingly declines as the proportion of cell wall constituents in the feed increases (Hume 

et al. 1989). Large grazing Macropus spp. “appear to be ideally adapted to maintaining their feed intake as grasses mature during 

dry periods and increase in fibre content” (Hume et al. 1989, p. 684) and some species can efficiently use low-quality forage 

(Frith 1973, Ellis 1975). Macropod diets consist mainly of leaves of monocots and dicots, grasses being the most important food 

of the larger species (Ellis 1975, Hume et al. 1989, Morgan 1994, Bennett 1995), with readily digestible green leaves being 

prefered. The diet of M. fuliginosus (Desmarest) includes “many other plants, particularly forbs” and dietary shifts during 

periods when grasses decline commonly occur (Bennett 1995 p. 138). 

Kangaroos eat many plants which livestock usually avoid. They are able to be more selective feeders than the blunt-muzzled 

bovids because they possess narrow muzzles, bearing narrow incisor arrays (Groves 1989). Under good conditions macropods 

can coexist with livestock, particularly cattle and horses, because they select a very different set of food items (Frith 1973, Hume 

et al. 1989). However where weedy dicots are a problem, their preference for grasses may exacerbate invasion (Morgan 1994). 

Other macropods, including the Spectacled Hare-wallaby Lagorchestes conspicillatus and the Bridle Nail-tailed Wallaby 

Onychogalea fraenata have decreased greatly as a result of overgrazing of livestock on Astreleba grassland and chenopod 

shrubland (Ellis 1975). 

Many macropod species, including Macropus spp., cause minor soil disturbance at resting sites, often in the shade, or at basking 

sites (Hume et al. 1989). 

Overgrazing by kangaroos is also a threat to grasslands. The population of Senecio macrocarpus at Yan Yean, Victoria, is 

thought to have dramatically declined rapidly between 1987 and 1992 due to high kangaroo density (Hills and Boekel 1996 

2003). 

No information appears to be available on interactions between N. neesiana and macropods. Existing knowledge of their ecology 

suggests they may be better able to suppress it than domestic livestock, and theoretical considerations based on recent analyses 

of biotic resistance (Parker and Hay 2005) suggest that they might prefer it to native grasses. Macropus spp. are probably poor 

vectors of seed in comparison to sheep, since they have short pelage and more flexible mode of foraging, making them less likely 

to contact seeding culms, and have coexisted with the similarly hazardous seeds of Austrostipa spp. in evolutionary time. 

The role and effects of fire 

“In grassy plains unoccupied by the larger ruminating quadrupeds, it seems necessary to remove the superfluous vegetation by 

fire, so as to render the new years growth serviceable”. Charles Darwin, The Voyage of the ‘Beagle’, 15 September 1833, on the 

Argentine pampas between Bahía Blanca and Buenos Aires. 

Fire is a regular feature of temperate grasslands throughout the world. In Australia fire became an important environmental 

factor in the late Miocene (c. 10 mybp), coincident with the decline of rainforest (Martin 1994). It has been a constant and 

generally increasing feature through the Quaternary period (c. 1 mybp +) but “it has not proved possible as yet to quantify 

vegetation/fire relationships” (Kershaw et al. 2000 p. 482). 

Fire is a critical process in maintaining open patches and the maintenance of high productivity in fire-dependent grasslands 

(Hobbs and Heunneke 1992). Fire reduces the cover of all herbaceous grassland plants to close to zero, with generally only ‘fireresistor’ 

perennial caespitose grasses retaining some standing live biomass, slightly above the soil surface, after burning 

(Overbeck and Pfadenhauer 2007). Post fire, the contribution to total plant cover from caespitose grasses gradually increases 

(Overbeck and Pfadenhauer 2007) and after several years can approach 100%, under suitable climatic conditions, leading to the 

loss of a high proportion of other vascular plants. Thus plant diversity is “directly linked to the fire cycle” (Overbeck and 

Pfadenhauer 2007 p. 36). Biomass accumulation has been well documented in Australian T. triandra grasslands, where Lunt and 

Morgan (1998a) recorded doubling of T. triandra density in the first year post-fire, doubling again in the second year to c. 5 

tonnes ha -1 with c. 50% the biomass dead, and doubling again >6 years post-fire, by which time biomass levels of 8 t ha -1 had 

been reached of which over 5 t ha -1 consisted of dead material. 

The mechanisms by which fire alters the composition and functioning of communites include the removal of the litter layer, 

creation of bare ground for seedling establishment, removal of shade and transpiration and thus alteration of microclimate, 

addition, depletion and creation of nutrients and the formation of ash beds. Fires in grasslands usually result in brief increases in 

soil fertility (Hobbs and Heunneke 1992), with temporary increases in N, K, Ca, Mg and pH in the uppermost layer, but can 

reduce fertility in the long term, depending on frequency, severity and season of burning (Overbeck et al. 2007). Nutrient 

addition from fire could result in increased invasion (Hobbs 1989). Fire results in liberation of plant-stored N to the atmosphere 

and alterations to the fire cycle resulting from grass invasion may thus impact on N fluxes in a grassland (Rossiter et al. 2003). 

Particular plant species may be promoted or disadvantaged depending upon the frequency, seasonal timing and intensity of fires, 

the condition of the vegetation, and the biology of the native and exotic plants in the community (Hobbs and Heunneke 1992, 

Adair 1995, MacDougall and Turkington 2007). Fire probably commonly has differential effects on different life stages of 

particular species (Overbeck and Pfadenhauer 2007) so deliberate grassland management using fire must consider not just the 

seasonal phenologies of the native species (MacDougall andTurkington 2007) but the demographic structure of the species 

populations (Hobbs and Heunneke 1992). 

The effects of fire on the flora of native grasslands is dependent on its thoroughness (patchiness), which is related to fire 

intensity but dependent on landscape features and local site characteristics such as presence of rocks and steep slopes, and on the 

season of burning (Stuwe 1994, Overbeck and Pfadenhauer 2007). Fires during different seasons favour different sets of plants. 

For instance annual species can be lost if fire occurs after full germination of their seed bank but before flowering (Kirkpatrick et 

al. 1995). The patchiness of fires determines the amount of diversity at the community level in grasslands (Lunt and Morgan 

2002). 

119

Soil temperatures below 45-55ºC cause little damage to vascular grassland plants and higher temperatures resulting from fire are 

generally restricted to the top centimetre of soil (Overbeck and Pfadenhauer 2007). 

Fire-suppression can have serious negative consequences for biodiversity in fire-adapted grasslands and may result in their 

transformation into shrub or tree dominated systems (Hobbs and Heunneke 1992). The extent to which native grasslands in 

south-eastern Australia were subject to fire prior to European settlement remains controversial, as does the role of fire, as 

opposed to climatic factors, in the maintenance of grasslands through ecological time (Jones 1999b). Flannery (1994), amongst 

others, marshalled substantial evidence that grassy woodlands and grasslands were the dominant vegetation types on the 

mainland before 1750, or at least much more widespread than contemporary vegetation suggests, and that the land was regularly 

burnt by aborigines. According to Kirkpatrick et al. (1995 p. 25) “all evidence suggests that fire was highly frequent, if not 

annual” before European occupation. Moore (1993 p. 351) stated that southern Australian “Themeda grasslands evolved under ... 

periodic burning”, a clear exaggeration of the extent of scientific understanding. 

Benson and Redpath (1997) examined these views of pre-European fire regimes and found no evidence of large scale annual 

aboriginal burning in south eastern Australia as a whole. They identified misinterpretation and unwarranted extrapolation from 

the records of early explorers and settlers as the basis for the view that ‘fire-stick farming’ was widespread in south-eastern 

Australia. However they appear to be overenthusiastic debunkers: no serious researchers make claims that grassy ecosystems 

were present in many of the areas they discuss. 

Jones (1999b) argued that the persistence of grassland requires fire at a certain frequency, that can burn without impediment 

across the landscape. Pollen in a core from Lake George on the southern tablelands of New South Wales indicates that 

communities with significant grass components were present during periods of glaciation. According to Jones (1999b) a major 

change to the dominance of Poaceae and Eucalyptus occurred during the late Pleistocene (uncertainly dated to c. 130 kybp), but 

Kershaw et al. (2000 p. 501) argued that eucalypt and grass dominated vegetation “with fire as a well-established environmental 

component” existed for most of the Quaternary before this time, possibly largely due to regional warming. Pollen of Poaceae is 

present throughout a swamp core from Lake Terang in western Victorian covering a period to c. 120 kybp, but increases towards 

the Holocene (Jones 1999b). Records from charcoal deposits are complex, but they currently provide no general, strong 

indication of an increase in fire during the periods (30-50 kybp) of first human occupation and the extinction of the megafauna 

(Johnson 2009). It appears likely therefore that any general increase in vegetation biomass expected with the loss of the 

megafauna was mainly of shrubland that was not susceptible to wide scale burning under the prevailing cool, wet conditions 

(Johnson 2009). However, grassy steppe vegetation that occurred in areas now in Bass Strait 25 kybp “shows abundance 

evidence for fire” and carbon particles reach high levels at Lake George at the start of the Holocene (Hope 1994). 

According to Stuwe (1994), natural fires, ignited by lightning were of frequent occurence in the natural temperate grasslands of 

continental south-eastern Australia. Enough fuel accumulates in most grasslands to allow an annual fire (Kirkpatrick et al. 1995). 

Kirkpatrick (2007) argued that lightning has been relatively rare in Tasmania during the last few thousand years but that it 

currently caused fires and must have caused the charcoal deposits present in Tasmanian soils before aboriginal occupation. 

Flannery (1994) argued that there are more than enough natural ignition events in most areas to burn the standing crop of fuel 

before it can decompose, and that deliberately lit fires would therefore have increased fire frequency but decreased fire intensity. 

The extent and temporal variation of natural fires are “largely unknown” though evidence suggests that soil and climate rather 

than fire determined treelessness, and lack of recent evidence of tree invasion of grasslands not managed by fire supports this 

conclusion (Lunt and Morgan 2002). 

Very little is known about the frequency and seasons of aboriginal burning in south-eastern Australian grasslands (Morgan 

1994). Aboriginal activities increased the fire frequency (Stuwe 1994), but opinions differ about the magnitude of the increase 

and the effects. Benson and Repath (1997) considered that “the extent, frequency and season of their use of fire is largely 

unkown”. In “the more developed areas” of Tasmania wide aboriginal use of fire appears to have occurred in the last 10,000 

years (Kirkpatrick 2007). Gott (1999) argued that burning of the temperate grasslands was deliberately controlled and timed to 

provide the most beneficial effects in terms of food production. Lunt et al. (1998) concluded that early historical records indicate 

that indigenous grasslands were frequently burnt. Any plants not adapted to frequent fires must therefore have been eliminated or 

confined to fire-free refugia long ago (Stuwe 1994). 

Grasslands taken up for livestock grazing were not managed with fire. In contrast, non-agricultural grassland remnants were 

frequently burnt in rail reserves (1860s?-1970s, often annually in late spring) and on roadsides (Western Victoria at least 1940s- 

1970s), as a cost-effective form of management (particularly to protect against wildfires) and those areas had the highest vascular 

plant diversity of all grassland remnants (Morgan 1997, Lunt and Morgan 2002), although some species were “presumably... 

eliminated” by the historical management activities (Morgan 1997). Sutton (1916-1917 p. 117) observed that in the Keilor Plains 

the “fires of spring ... never quite die down”. Recent fire regimes involve deliberate small scale management fires and occasional 

wildfires. Many linear remnants (roadside and rail easements) have been regularly burnt in early summer while conservation 

reserves have been subject to irregular (often “erratic”) early autumn burning (Lunt and Morgan 2002 p.180). 

The fire intensity (energy release per unit area) in temperate south-eastern Australian grasslands is generally low (in comparison 

to tropical grassland fires), because of the small quantitites of fuel and the slow rate of spread of managed fires (Lunt and 

Morgan 2002). In T. triandra grasslands above-ground plant biomass is in the range of 7-11 tonnes ha -1 in infrequently burnt 

grasslands (7-11 years since fire) to c. 1-4.8 tonnes ha -1 in grasslands burnt 1-2 years previously (Lunt and Morgan 2002). T. 

triandra stands rarely produce enough litter in the first year after a fire to enable another burn (McDougall 1989). In Poa and T. 

triandra grasslands the dominant grass often accounts for >90% of the biomass (Groves 1965, Lunt and Morgan 2002). 

Nutrients in ash are more readily lost in runoff or by wind erosion after fire (Flannery 1994). Fire also directly releases nutrients 

to the atmosphere, particularly N, and the levels of available soil N are often affected by repeated burning (MacDougall and 

Turkington 2007). An estimated 24 kg ha -1 of NO 2 was lost when tropical grassland at Katherine, Northern Territory was burnt 

(Flannery 1994). MacDougall and Turkington (2007) however found no such effect of N loss in savannahs in British Columbia, 

consistent with other recent studies, and thought that a combination of fire effects on litter, soil temperature, soil microbes, etc. 

could have suppressed N mineralisation. Moore (1993 p 352) stated that fires in T. triandra grassland ungrazed by livestock 

120

esult in a “short-term flush” of N mineralisation in the soil, and suggested that T. triandra is uniquely able to take advantage of 

this resource. 

The critical factor for survival and resprouting of vascular plants of frequently burned grasslands is the degree to which buds are 

protected from fire damage (Overbeck and Pfadenhauer 2007). In general in temperate south-eastern Australian grasslands fire 

promotes vigorous resprouting by native perennials, but generally very little perennial seedling recruitment: fire enhances the 

vigour and flowering of many perennial herbs but results in little change in plant species composition (Lunt and Morgan 2002). 

However buring can lead to major increase in the abundance of annual exotic species on long-unburnt sites (Lunt 1990c). T. 

triandra grassland fires are relatively cool, and when occurring in summer and early autumn have little negative effect on the 

flora, which survives with maximal underground carbohydrate storage, buried buds and buried seed, although seeds on the soil 

surface are destroyed (Lunt and Morgan 2002). The majority of intact T. triandra grasslands on the Victorian volcanic plains are 

burnt at intervals of 1-5 years (Morgan 1998d) a frequency recommended by McDougall (1989). Healthy T. triandra swards are 

usually maintained with fires at this frequency and no grazing (Lunt and Morgan 2000). The time of burning (late summer or 

autumn) is probably not critical for the health of T. triandra populations, but fire greatly reduces the amount of T. triandra seed 

produced in the following autumn (McDougall 1989). Reduced fecundity in the short term is compensated for by increased plant 

surival and vigour. Morgan (1997) predicted that reducing the fire frequency from 1-2 years to 5 or more years would 

substantially alter the dynamics of the population if seedling recruitment was reduced and established plants were incapable of 

adjusting. 

Most of our knowledge relates to the grasslands in which T. triandra is dominant and highly productive, systems that naturally 

promote fire. Much less is known about fire effects in grasslands not dominated by T. triandra. Poa spp. “appear to be able to 

maintain their dominance without disturbance by burning or other forms of biomass removal” while Austrodanthonia and 

Austrostipa grasslands accumulate little biomass and are not thought to need regular biomass reduction to maintain plant 

diversity (Lunt and Morgan 2002 p. 183). Little seems to be known about fire effects in the Riverine Plains grasslands. 

Themeda triandra biomass accumulation and senescence dieback 

The dependence of T. triandra grasslands on fire is almost universally recognised, despite the very limited understanding of their 

evolutionary origin and palaeoecology, particularly in relation to ancient grazing regimes. They appear to be similar to temperate 

fire-adapted grasslands with C 4 dominants worldwide, where the C 4 species promote their own dominance by building high 

levels of biomass and thus promoting fire. T. triandra stands that are not burnt, or otherwise biomass-reduced, gradually develop 

massive quantities of dead leaves and litter, and failure to remove this biomass can cause tiller and plant senescence, attributed to 

“self-shading” (Lunt and Morgan 2000). Major T. triandra mortality occurred at Derrimut and Laverton North grasslands when 

fire frequency exceeded 5 y, and when fire was finally used, plant and tiller densities were much lower than in regularly burnt 

grassland (Morgan and Lunt 1999, Lunt and Morgan 1999a 1999c). Mass dieback of T. triandra resulting from the absence of 

fire or other biomass reduction has been described as “grassland collapse” (C. Hocking pers. comm.). 

According to Lunt and Morgan (1998a p.70) “the grassland becomes increasingly choked up with dead grass material, above 

which the tussocks form a thin mantle of new, green growth”, and eventually living tillers can “no longer poke up through the 

dead grass to reach the sunlight, causing the tussocks to die”. Supposedly, there is “insufficient light penetrating through the 

canopy of old foliage to the young tillers to enable them to photosynthesise sufficient energy” (Lunt et al. 1998). T. triandra 

grassland not burnt for greater than 6 years is now thought to senesce in this way (Morgan and Lunt 1999, Lunt and Morgan 

2002), rather than reach a steady state (Lunt and Morgan 2002). Low soil fertility and moisture levels may sometimes prevent 

such senescence (Kirkpatrick et al. 1995), so on less productive sites T. triandra senescence requires considerably longer periods 

or may never occur. O’Shea (2005 p. 161) stated that on productive sites, the phenomenon requires 10-11 years: “the canopy 

collapses upon itself and forms a thick layer of dead thatch over the soil surface, allowing only minimal seedling recruitment and 

preventing new tiller initiation”. According to Muyt (2005 p. 3), the dense T. triandra thatch “undermines the growth of [the 

grass] itself; plants become increasingly brittle and subject to collapse”. After burning T. triandra usually regains high cover 

quickly, returning to pre-fire biomass levels in 2-4 years (Morgan 1994, McDougall and Morgan 2005) and can form a complete 

dense canopy in as little as 3 years. Death of the dominant grass results in a major nutrient pulse in the soil, resulting primarily 

from decay of T. triandra crowns and roots, and this enables invasion by exotic plants (Wijesuriya 1999, Wijesuriya and 

Hocking 1999). Weedy exotics are now generally pervasive in these systems, so what would happen in the absence of exotic 

seed sources after T. triandra die-off is not clear. 

The senescence dieback phenomenon has been widely reported worldwide for other dominant temperate and subtropical C 4 

tussock grasses adpated to fire (Bond et al. 2008). Although often described under different rubricks (e.g “detritus accumulation” 

– Knapp and Seastedt 1986) the effect is the same: accumulation of standing dead litter shades out and kills the shade-intolerant 

new tillers (Knapp and Seastedt 1986; Everson et al. 1988, Uys et al. 2004). Fire frequency is thus one of the most important 

determinants of the grass species composition in C 4 grasslands: “fire-dependent grass species, typically members of the 

Andropogoneae ... are fire-dependent in the sense that they decrease in, or disappear from, a sward in the absence of frequent 

burning” (Bond et al. 2008 p. 1747). Lunt and Morgan (1999a) reported a 70% decrease in T. triandra tussock density and a 

58% decrease in live tiller density in rarely burnt areas. Similar changes are indicated by Uys et al. (2004) who found that the 

cover of T. triandra in a mesic (735 mm per annum) South African grassland decreased from c. 70% when annually burnt to < 

5% after fire exclusion for 4 or more years, and the plant effectively disappeared from mesic and montane sites that were 

unburnt.. 

In southern Brazilian mixed C 3 and C 4 tussock grasslands dominated by C 4 grasses, Overbeck and Pfadenhauer (2007 p. 35) 

observed that “without periodic removal of biomass ... shading by dead [grass] biomass inhibits survival and tillering and higher 

humidity under the litter may cause death and decay of underground plant parts within a few years”. Post-fire effects that 

increase the vigour of tussocks have been recorded for a number of dominant species in other parts of the world, including 

increased photosynthetic activity, growth rates and sexual reproduction (Overbeck and Pfadenhauer 2007). 

The T. triandra senescence phenomenon may be compared with ‘normal’ grass senescence, which occurs in all Poaceae in 

response to drought (Norton et al. 2008) and is a mechanism to reduce plant mortality from water stress, by the gradual 

121

‘abandonment’ of foliage. Senescence dieback of T. triandra swards after extended periods of biomass accumulation might 

possibly be the result of altered water relations, rather than the postulated ‘self-shading’: the underlying mechanisms require 

further investigation. But like other dominant C 4 grasses worldwide, accumulation of large quantitites of dead biomass by T. 

triandra is apparently an adapted strategy that enables it to perpetuate its dominance by providing appropriate conditions for 

frequent burning (Hocking and Mason 2001). The high C:N ratio of C 4 grass foliage means the leaves have low nutritional 

quality and limited palatability to herbivores (Moore 1993, Moretto and Distel 2002) so a higher proportion die without being 

eaten than leaves of C 3 grasses, and the litter is more resistant to microbial breakdown than that of non-C 4 plants (Wedin 1999, 

Groves and Whalley 2002) so more of it can accumulate. 

Senescence dieback of T. triandra increases the soil available nutrients, probably as a result of increased rates of decay of both 

above and below ground vegetation, due to increased moisture and temperature under the thatch of dead leaves, and by reduced 

nutrient uptake by living biomass (Hocking and Mason 2001). However utilisation of the nutrient pulse by other plants is not 

possible until the high cover of dead grass decays or or is removed by fire or other biomass reduction. When this occurs major 

weed growth usually follows (Hocking and Mason 2001). 

Effects of T. triandra biomass accumulation on other species 

Commonly, in the absence of of regular biomass reduction by fire, grazing or mowing, litter accumulation by exotic or native 

perennial grasses results in the suppression of the smaller intertussock native vascular plant species (McIntyre 1993, Kirkpatrick 

et al. 1995, Morgan 1998e). Entire populations of perennial forbs can disappear within a short period in the absence of fire 

(Morgan 1998e). In T. triandra grassland unburnt for >5 years the cryptogam crust also degenerates due to litter accumulation, 

shading and increased earthworm activity (Scarlett 1994). However bryophyte diversity in areas burnt at 1-2 year intervals is 

reduced compared to longer unburnt areas (Morgan 2004). 

Periodic biomass reduction is required to maintain the vascular flora, a large proportion of which have soil seed banks which 

disappear after 1 year (Lunt 1990c 1995a, McIntyre 1993, Stuwe 1994). Morgan (1995b) for example found that 90% of 

Rutidosis leptorhynchoides seed germinated within a few weeks of autumn rains. Endangered species that are threatened when 

fire frequency in T. triandra grasslands is too low include Senecio macrocarpus (Hills and Boekel 1996) and Rutidosis 

leptorrhynchoides (Morgan 1995a, Humphries and Webster 2003). Sharp (1997) experimentally confirmed the hypothesis that 

litter removal is required to facilitate establishment and reduce suppression of the smaller native grasses and the low-growing 

and small forb components in ACT grasslands. Lack of fire or some other management regime with similar effects is therefore a 

threat to the continued existence of the more mesic, T. triandra dominated grasslands, both in terms of the keystone species (T. 

triandra) and most of the other plant components. Suppression of other native species by the dominant grasses is not generally a 

problem in grasslands on shallow rocky soils and the more xeric inland plains, and fire is not necessary to maintain their 

indigenous vascular plant diversity (Kirkpatrick et al. 1995). 

Effects of fire on Themeda triandra 

T. triandra is well adapted to survive fire but mortality nevertheless occurs. Stafford (1991) reported that very few four-year-old 

T. triandra plants survived an intense wildfire in Cleland Conservation Park, South Australia, and a fire at Organ Pipes National 

Park, Victoria, in April 1997 before a severe drought resulted in substantial mortality and a dramatic cover decline (McDougall 

and Morgan 2005). However biennial autumn burning at Organ Pipes usually did not inhibit an already established trend of 

increased T. triandra frequency and cover (McDougall and Morgan 2005). Burning generally does not kill T. triandra tussocks 

(Henderson 1999) nor the very high proportion of other grassland plants which are fire-adapted hemicryptophytes, geophytes 

etc., i.e. perennial plants with perennating buds protected underground (Morgan 1996, Lunt 1990a 1990c.). T. triandra usually 

regains high cover quickly, returning to pre-fire biomass levels in 2-4 years (Morgan 1994). At Evans St., Sunbury, cover 

reached 43% after 9 months and was predicted to reach 100% after 2-3 years (Morgan and Rollason 1995). Creation of bare 

ground along with an ash bed probably enhances seedling establishment for most native species (Stuwe and Parsons 1977), but 

these are favourable conditions for most exotic vascular plants as well. 

In South African grassland Uys et al. (2004) recorded declines of T. triandra related to longer fire frequency but continued 

persistence at a semi-arid (550 mm per annum) site 26 years post fire. 

Fire also enables T. triandra regeneration from seed. In T. triandra establishment experiments, Stafford (1991) found that 

burning of areas to which a close thatch of whole T. triandra culms had been applied resulted in immediate seed germination, 

with an estimated seedling density of c. 1000 m -2 . Areas thatched in December and burnt 10 months later produced seedlings at 

the end of October, the most suscessful of which produced seed the following February. However fire kills T. triandra seeds that 

have not worked their way into the soil (Hocking pers. comm.). 

Effects of fire on other vascular plants 

Particular forb species may be negatively effected by regular burning during a particular season. The late-flowering native pea 

Glycine labrobeana (Meisn.) Benth. is very susceptible to regular fires in late spring-early summer, which destroy its flowers 

and seeds (Scarlett and Parsons 1993). Cullen spp., also late flowering peas, may be rare in rail reserves for the same reason 

(Morgan 1994). Thesium australe, once widespread in native temperate T. triandra grasslands, is short-lived and highly 

dependent on annual seedling recruitment, is probably eliminated by annual burning (Scarlett and Parsons 1993), apparently 

germinates well without fire and after fire, but seems to require open conditions for growth (Scarlett et al. 2003). 

As previously noted, Morgan (2004) found that more frequent fires reduced bryophyte diversity in Victorian basalt plains 

grasslands by loss of species, mostly mosses, although none of the 150 m 2 quadrats he surveyed had a richer moss and liverwort 

flora than the total of 990 m 2 surveyed at the frequently burnt Evans St. Sunbury site by Morgan and Rollason (1995). 

Presumably the greater exposure and dessication resulting from frequent fire destroys mosses and removes suitable habitat, 

including the shade and increased humidity of dense cover provided by T. triandra. Slow recovery and recolonisation of mosses 

post-fire has been widely reported in other ecosystems, and, as for vascular plants, frequent burning of native grasslands through 

ecological time has probably eliminated the most fire-sensitive species long ago (Morgan 2004). Fire damage to soil cryptogam 

crusts can also facilitate grass invasion (Milton 2004). 

122

Fire effects on weeds 

Moore (1993 p. 351) argued that Themeda grasslands remained “remarkably stable” under regimes of marsupial grazing and 

periodic burning, and if otherwise undisturbed were “not invaded by introduced species”. Fire results in temporary increases in 

nutrient availability and this fluctuation of resources may enhance invasion by exotic plants (Hobbs 1989). In general burning of 

south-eastern Australian native grasslands promotes post-fire colonising plants, most of which are exotic annuals that recruit 

from a large soil seed bank (Lunt 1990b, Lunt and Morgan 2002). However Stuwe and Parsons (1977) found a smaller 

proportion of the vascular flora consisted of alien species at regularly burnt (railway) sites than in grazed or unmanaged T. 

triandra grasslands, while Dodd et al. (2007) found that sites burnt less frequently had higher exotic seed banks, and considered 

regular burning to be detrimental to the exotics. Burning of some degraded T. triandra grasslands has been reported to lead to 

major increases in cover and density of exotic annuals, which then slowly decline (Lunt 1990b 1990c, Morgan 1998d). Scarlett’s 

(1994) view was that regular burning of sites with a long history of grazing and trampling rarely has any major impact on exotic 

annuals, an effect he linked to the poor re-establishment of the cryptogam crust. 

Fire is often considered to provide advantages in reducing weed populations, but may promote or inhibit particular weed species 

or functional groups. Fire is demonstrably an effective tool to greatly reduce the cover of exotic invasive grasses when the 

invasives are not fire adapted, and to restore native plant cover (e.g. MacDougall and Turkington 2007). In temperate Australian 

grasslands spring fires might reduce or prevent seed set in exotic annual grasses (Stuwe 1986) but late autumn burning promotes 

their regeneration along with Romulea rosea (L.) Eckl. (Iridaceae) (Lunt 1990c). Burning can reduce the density of the exotic 

grasses Briza maxima, Cynosurus echinatus L., Lolium rigidum Gaudin, L. perenne and Bromus hordeaceus in T. triandra 

grasslands but promotes a range of others including Aira and Vulpia spp. (Lunt 1990c, McDougall 1989, Adair 1995). 

As with the native plants, seasonality of the fire may determine the effect on particular exotics (see Morgan 1996), while the 

impacts on propagule production and seed bank levels are obviously important factors. According to Adair (1995) no clear 

evidence was then available that the season of burning has a significant effect on introduced grasses and forbs. Exotics are likely 

to be advantaged if they account for a high proportion of the soil seed bank, which is generally the case (Kirkpatrick et al. 1995, 

Morgan 1998c, Dodd et al. 2007). Robertson (1985) found similar densities of exotic annuals after autumn and spring burns at 

Gellibrand Hill. Stuwe (1994) among others argued that biomass-reduction burning now facilitates the growth of both native and 

exotic plants, while Lunt (1990b), based on seed bank studies at Derrimut, argued that burning is likely to promote exotics more, 

or at least as much as natives. Dodd et al. (2007) argued that exotic dominance of the seed bank is partly due to removal of fire, 

which along with exotic dominance in the surrounding landscape matrix, is likely to result in increased weediness of remnants. 

Decisions about the desirability of burning need to be based on an understanding of the contents of the soil seed bank and the 

phenology of the above ground cover at each site in order to predict whether a burn would be beneficial for the flora. 

Determining an appropriate fire regime even at a single small site can thus becomes prohibitively difficult. Most native 

grasslands contain a mix of weeds with different life strategies, so fire cannot be used as a general tool to favour the native 

species (Lunt 1990c). 

Stuwe (1994) warned that burning of areas with a N. neesiana seed bank should not be undertaken unless follow up herbicide 

treatment could be undertaken. However there appear to have been no studies on the specific effects of fire timing and intensity 

on N. neesiana populations in native grasslands. No information appears to be available about the comparable fuel loads of 

grasslands with and without N. neeesiana, so it is difficult to speculate about whether N. neesiana infestations increase or 

decrease the frequency and intensity of fires. 

In relation to vascular plants of native grasslands, there is a general consensus that the interaction between fire and grazing is the 

major factor involved in the abundance of most native species, including threatened taxa, and that low fire frequency is 

detrimental to most species (Scarlett and Parsons 1993). Frequent burning (a fire every 2-5 years) is necessary to maintain 

grasslands dominated by T. triandra, but other Victorian grassy ecosystems (E. camaldulensis and E. melliodora grassy 

woodlands in the Grampians and Sandplain Grassland in the Mallee) appear not to require burning to maintain their vascular 

plant diversity (Lunt 1991). In T. triandra grasslands, long intervals between fires results in loss of forb diversity (Stuwe and 

Parsons 1977, Lunt 1990c, Lunt and Morgan 1999). Morgan (1998c) found no correlation between seed bank species richness 

and fire history for five grasslands in the Victorian volcanic plains. However fires do not generally lead to significant seedling 

recruitment of native plants (Henderson 1999, Lunt and Morgan 2002), possibly because of severe depletion of the seed bank 

resulting from prior affects of long-term livestock grazing on seed production (Lunt 1990b 1990c). T. triandra is more sensitive 

to grazing shortly after fire (Groves and Whalley 2002). 

Effects of fire on animals 

Combinations of positive and negative effects of fire also occur with other organisms. Fire can change the abundance, range 

(spatial and temporal), fecundity and dietary choice of other organisms by altering the provision of food, shelter and habitat 

(Low 2002). Fire directly results in the elimination of many animal populations, although many sedentary species survive 

underground. 

According to Yen (1995 1999) the effects of fire on temperate Australian grassland invertebrates were unknown, while Driscoll 

(1994) noted that they were ‘yet to be investigated’. Studies from other ecosystems tend to be equivocal, with few lessons for 

grasslands (Driscoll 1994). Edwards (1994) argued that fire often causes local extinction, but recolonisation from unburnt areas 

is usual, so in a situation where an invertebate is dependent on a community that is scarce and highly fragmented, recolonisation 

is often impossible. This appears to be the case with Synemon spp. (Edwards 1994) and with xanthorhoine moths (McQuillan 

1999), although with organisms like Synemon that spend a major part of their lifecycle underground, the precise timing of the 

fire with respect to the above-ground stages would appear to be critical. 

Removal of dense cover by fire and the creation of open ground will radically alter temperatures at the ground surface. Lunt 

(1995a) quantified temperature differences between the canopy surface in T. triandra grassland, the ground surface temperature 

and soil temperatures at 3 cm depth, in areas where the canopy was closed and in open gaps from July to April. The canopy 

provided excellent insulation with temperatures on the surface similar to those in the soil. Temperatures in gaps were generally 

123

5°C warmer than under the canopy and much larger differences occurred in summer. Such contrasts will substantially affect the 

behaviour and activity levels of invertebrates. 

Farrow (1999) found no clear differences in the diversity of canopy-living insects in small burnt patches and unburnt areas in 

ACT grasslands and suggested that recolonisation occurred within 6 months of fire. Nearly twice as many individual insects 

were present in summer in areas burnt 6 months prior to sampling than in unburnt areas, and more than three times as many in 

the following spring. Species that benefit from more open ground, like many ants, probably commonly proliferate. Maintenance 

or periodic refreshment of plant diversity by fire will benefit a wide range of herbivorous taxa with particular food plants, and 

maintenance of the forb components will benefit nectar and pollen feeders. Maintenance or enhancement of animal diversity at 

the primary consumer level will flow on to higher tropic levels, benefitting the diverse array of predators and parasitoids. 

Greenslade (1994) found litter removal by fire was likely to markedly reduce Collembola diversity for at least two years. Other 

detritivores dependant on plant litter would be expected to be similarly disadvantaged. 

Any beneficial effects on fire on plant diversity may or may not flow on to invertebrates, depending, in part on whether source 

populations of new and recolonising species exist (Tscharntke and Greiler 1995). Driscoll (1994) concluded that some species 

may be completely intolerant of fire, while others may require it for survival. Any species that spend part of their lifecycle 

underground are probably relatively resistant to fire, while those that do not are likely to be more highly vagile and have an 

ability to recolonise from unburnt areas. Sedentary, non-subterranean species are probably most vulnerable to extinction, but 

there are unlikely to many of them in fire adapted vegetation. As with vascular plants (Stuwe 1994) and bryophytes (Morgan 

2004), the indigenous invertebrate flora of grasslands that have been subject to periodical fire over many thousands of years must 

consist largely of fire-adapted species, whether or not they are fire survivors or recolonisers. 

Fire regime management in Victoria now incorporates the goal of providing the conditions necessary for the persistence of the 

indigenous biota (Mansergh et al. 2006b) but implementation of this idealfor the various biotic elements, including animals, is 

difficult and is bound to produce contradictory results because different valuable elements of the biota may be advantaged or 

harmed by the same fire cycle. The difficulty for managers is to evaluate and juggle the often contrary needs and tolerances of 

the various components in each grassland remnant. 

Nutrients and soil factors 

Lowland native grassland in Australia is found on relatively fertile, organic-rich, cracking clay soils on substrates of volcanic 

rocks (basalt and dolerite), fine-grained sedimentary rocks (including limestone) or recent alluvium (Kirkpatrick et al. 1995). In 

the past, temperate grasslands have erroneously been characterised as low in nutrients, based on analysis of soils. Mott and 

Groves (1994 p. 376) state that “Australian grassland soils are of very low fertility, particularly of nitrogen and phosphorus, by 

comparison with those of grasslands elesewhere”. Sharp (1997) lists low soil nutrient content as one important factor 

determining the distribution of native grassland in the ACT. Australian soils have in general been considered to be nutrient 

deficient (e.g. Roberts et al. 2006). However, like tropical rainforest, much of the system nutrients are held in the plants. 

Above- and below-ground biomass 

In temperate herbaceous communities 60-80% of vascular plant biomass is underground (see Reynolds 2006 p. 57). In grasslands 

up to 90% is below-ground (Wijesuriya and Hocking 1999) and it is usual for a high proportion of total plant biomass to be 

represented by roots and buried crowns. More of the energy captured in photosynthesis in grasslands is directed to below-ground 

parts than those above-ground and most of the nutrient circulation occurs below ground (Soriano et al. 1992). In Australian 

grasslands most of the biomass is contained in the dominant grasses (Groves 1965). T. triandra- and Poa-dominated 

communities are more productive (i.e. produce more biomass) than the Austrodanthonia and Austrostipa grasslands that 

predominate in drier areas (Lunt and Morgan 2002). The mean root: shoot ratios in North American grasslands varies from 13 to 

2, being higher in cooler climates (Tscharntke and Greiler 1995). Cooler and drier grasslands generally have ratios between 13 

and 6, while warmer, more humid grasslands have ratios between 6 and 2 (Soriano et al. 1992). A ratio of 2.6 was calculated for 

one Argentinean pampas grassland (Soriano et al. 1992). Rodríguez et al. (1995) measured below-ground biomass down to 10 

cm depth in five grazed grassland communities in Spain and found that 61-68% of the total biomass was below ground, of which 

49-68% was in crowns and the remainder below 1 cm depth. Groves (1965) found that the root:shoot ratio of a T. triandra 

grassland had very high seasonable variability but that root biomass was generally 2 to 4 times that of above-ground parts, i.e. 

66-80% of the biomass of the overwhelming dominant grass was below ground. In one Pampas grassland 65% of the 

underground biomass was located above 10 cm, 85% above 30 cm and 100% above 70 cm, a similar distribution to most 

temperate, subhumid grasslands (Soriano et al. 1992). 

Underground net primary productivity in one Pampas grassland was estimated to be about 5000 kg ha -1 yr -1 , just 17% less than 

productivity above ground, with a below-ground: above-ground productivity ratio of 0.9 (Soriano et al. 1992). 

Soil nutrient levels 

The soils of Australian native grasslands usually have low nutrient levels (Table 8), and a high organic matter content with 

organically bound N and P (Wijesuriya and Hocking 1999). However most of the nutrients in the system are locked up in the 

biomass of the grasses. In the Victorian basalt plains, soils derived from sediments have higher nutrient levels than those derived 

from basalt, and soils on stony rises have the highest fertility (Williams 2007). 

124

Table 8. Typical nutrient levels in Australian grassland soils. Source: McIntyre and Lavorel (2007). 

Vegetation N P (mg kg -1 ha -1 ) 

Natural grasslands low 1-3 (low) 

Native pasture medium 1-3 (low) 

Fertilised (improved) pasture high 20 (high) 

Sown pasture high 20 (high) 

Enriched grassland medium medium 

Moore (1973) suggested that nitrate levels in the surface soils rarely exceeded a few parts per million at any time of year. In 

grasslands in general a “small active fraction” of soil organic matter dominates both C respiration and N mineralisation 

(microbial conversion of N into the plant-available nitrate and ammonium forms) (Wedin 1999 p. 194). Roberts et al. (2006 p. 

148) stated that pasture productivity may be limited when P falls below 25 mg/kg (Colwell), but that “native pastures ... tend to 

be more tolerant of lower P levels than those dominated by introduced species”. 

Weed invasion and nutrient enrichment 

Nutrient enrichment by legumes, application of fertiliser, runoff, deposition of atmospheric pollution etc. is a major cause of 

alien grass invasion worldwide (Milton 2004) and experimental addition of nutrients often rapidly leads to weed invasion (Carr 

1993). Eutrophication, particular with N and P, is a major cause of plant diversity decline in terrestrial ecosystems (Hobbs and 

Heunneke 1992, Hautier et al. 2009). Grasses tend to be particularly favoured by nutrient inputs, and biodiversity losses are 

usually associated with their increasing productivity and dominance (Hobbs and Heunneke 1992). 

Australian soils have generally been characterised as nutrient impoverished, particularly in relation to phosphorus, but also in 

nitrogen, minor nutrients and organic matter (Leeper 1970). Because most Australian native plants are adapted to these low 

nutrient levels, nutrient enrichment favours the establishment of exotic weeds that are better adapted to high levels of fertility 

(Brereton and Backhouse 2003). Cale and Hobbs (1991) found a strong positive correlation of nutrient gradients across roadsides 

with exotic plant diversity and suggested that nutrient enrichment may increase their competitiveness. Cover of exotics increased 

from

Nitrogen dynamics of competing C 3 and C 4 grasses 

C 4 grasses are superior competitors in undistrurbed grassland because of they way they use and sequester N and extraneous 

addition of soil N benefits alien annual grasses and decreases species richness of native flora that evolved under conditions of 

low soil N (Milton 2004). 

Moore (1973) suggested that the success of T. triandra as a warm-season C 4 grass in southern latitudes is due to an ability to 

sequester N and other nutrients as they are mineralised. Short term increases in N mineralisation after fire in ungrazed T. 

triandra grassland therefore advantage the dominant grass. N appears to be the key nutrient determining the balance between C 4 

and C 3 grasses (Wedin 1999, Groves and Whalley 2002, Groves et al. 2003a). Soils are more commonly deficient in it than any 

other nutrient, and plants growing in N-deficient soils usually have a high root-shoot ratio (Salisburyand Ross 1992). Plant 

growth is believed to be primarily limited by the availability of N in the soil (Eschen et al. 2007). Unlike other nutrients, N has 

little interaction with inorganic soil minerals and its plant-availability is almost totally regulated by biotic processes (Wedin 

1999). The amount of plant-available soil N is determined by mineralisation from organic matter by soil microbes, 

immobilisation in microbial and plant biomass, deposition (external fertilisation) and external losses (denitrification and 

leaching) (Moretto and Distel 2002). Soil inorganic nitrogen content largely reflects the balance between mineralization of N and 

immobilisation of N by soil microbial biomass (Andrioli and Distel 2008). Most soil N is contained in the humus, locked in C-N 

bonds that are energetically very expensive for decomposer organisms to break (Wedin 1999). High lignin concentrations and 

high C:N ratios in plant litter commonly immobilise the N pool and result in low available N in the soil (Moretto and Distel 

2002). C 4 grasses use N more efficiently than C 3 grasses (Monson 1989, Bouchenak-Khelladi et al. 2009) and therefore have 

roots and leaves with higher C:N ratios and litter of a lower quality. For example the ratio in Schyzachyrium scoparium (Michx.) 

Nash (C 4 ) roots was 100 and leaf litter 110 (Andrioli and Distel 2008) compared to roots c. 39-59 and leaf litter c. 40-60 in four 

Nassella spp. assessed (Andrioli and Distel 2008). T. triandra, as a C 4 species, produces a relatively high amount of biomass, low 

in protein, compared to co-occurring native C 3 grasses (Moore 1993, Nie et al. 2009), and its litter has a high C:N ratio, so it can 

be conceived of as a climax species in grassland succession, where the climax is characterised by a stable species assemblage, 

with low levels of available soil N and relatively high biomass (Moore 1993). 

Higher quality plant litter decomposes more rapidly and results in net mineralisation of N, whereas lower quality litter 

decomposes more slowly and results in net immobilisation of N (Andrioli and Distel 2008). Andrioli and Distel (2008) found 

little difference between the litter quality of several Nassella spp. (not including N. neesiana) in semiarid (mean annual rainfall 

400 mm) Argentina and no differences in their influence on soil inorganic N content or potential N mineralisation. Variation 

measured within the Nassella spp. was low in comparison to C 4 grasses. Presumably N. neesiana has similar higher quality litter, 

that markedly differs in its biodegradibiltiy to the low quality T. triandra litter. 

Mineralisation of soil N in temperate Australian takes place during the summer, reaching a peak at the end of summer (Moore 

1993). The nitrate content of the top 10 cm of soil under T. triandra grassland at the end of summer has been found to not exceed 

5 ppm, in comparison to >36 ppm under Austrodanthonia C 3 grasses (Moore 1993). Soils under the cool season native grasses 

therefore had large pools of labile N when the winter growing season commenced, facilitating invasion by cool-season annuals, 

while intact T. triandra was resistant to invasion: “By growing when mineralisation processes are actually or potentially active, 

and utilising or otherwise limiting the accumulation of labile nitrogen in the soil surface, Themeda, seemingly, gives stability to 

the grassland community” (Moore 1993 p. 352). 

When the C:N ratio of grass litter is more than 30:1, the rate of decomposition of the litter is slow, microbial decomposers are N- 

limited, N is largely immobilised, and there is little or no release of nitrate and ammonium into soil solution (Wedin 1999, 

Groves and Whalley 2002, Moretto and Distel 2002). In such situations with a low rate of N mineralistation, species with a high 

N use efficiency, usually the C 4 grasses, have a competitive advantage, and are able to efficiently deplete the low soil nitrate 

pools. The C:N ratio of litter and roots of such grasses is generally much greater than 30:1, and they immobilise the N, and 

buffer against N pulses created by disturbance, so perpetuate their own dominance (Wedin 1999, Groves and Whalley 2002). T. 

triandra is a more efficient user of N than N. neesiana, and healthy T. triandra stands lock up system pools of N in high-C 

biomass. This array of N-related mechanisms explains the resistance to invasion by exotic species of T. triandra grasslands 

reported by Hocking (1998) and Wijesuriya and Hocking (1999). 

When disturbance results in the death of T. triandra, levels of available N are increased and this results in changes in the floristic 

composition of the grassland (Moore 1993). Disturbances including continuous grazing, fertiliser addition and cultivation, result 

in increased rates of N mineralisation and higher soil nitrate and ammonium levels (Wijesuriya and Hocking 1999, Groves and 

Whalley 2002). Soil disturbance involving digging and homogenisation of soil of T. triandra grassland thus results in a major 

increase in the above-ground biomass of dicot weeds and annual grasses. Wijesuriya and Hocking (1999) found that 

approximately 90% of a total of 60 kg ha -1 dry weight above ground in late spring, 70 days after such disturbance, consisted of 

exotic annual grasses, “thistles” and “flat weeds”. This reflected a simultaneous nearly ten-fold increase in the amount of plantavailable 

soil N and an approximate doubling of available soil P. Addition of c. 25 kg ha -1 of both N and P fertiliser, combined 

with digging and homogenisation of the soil, produced total above ground biomass over the same period of c. 150 kg ha -1 dry 

weight, of which approximately 90% again consisted of these exotic weeds (Wijesuriya and Hocking 1999). 

These authors demonstrated that addition of N and P to T. triandra grassland at rates of c. 25 kg of N or P ha -1 resulted in weed 

flushes and that cultivation resulted in rapid mineralisation of organic matter and a consequent ‘pulse’ of N. In experiments on 

basaltic clays at Derrimut, Victoria, after 14 days, the rate of mineralisation of N in soil dug out, homegenised to very small 

particle size and replaced in plots was 4.7 times that of undisturbed soil, while the rate for P was not significantly different to 

undisturbed soil. Total available soil N continued to increase in the disturbed soil for 70 days, at which time it was 

approximately 10 times that of undisturbed soil. Available P was also significantly higher in the dug soil after 70 days. These 

authors also compared plots in which there was no soil disturbance with those in which soil was dug out, homogenised and 

returned, and subsequently unfertilised, fertilised with N, P, both N and P, or treated with sucrose (as a C source). In early 

summer, 70 days after disturbance, >95% of plant biomass in disturbed plots consisted of exotic weeds, mainly annual 

Asteraceae and annual Poaceae, with the Poaceae accounting for about half the biomass of the dicotyledonous species. Undug 

plots carried >90% T. triandra. Species diversity was similar in plots dug and treated with N and P, or both N and P, but annual 

126

grasses produced sigificantly more biomass when fertilsed with P than when fertilsed with N or not fertilised, and annual dicots 

were significantly more numerically dominant in plots receiving both nutrients. Annual grasses had significantly higher numbers 

when the dug soil was treated with both nutrients than when treated with N alone, but not P alone. Total biomass was 

significantly greater in plots receiving both nutrients. Addition of sucrose resulted in increased microbial activity and rapid, near 

complete exhaustion of soil nitrates. Biomass of dicot weeds and annual Poaceae was significantly lower in the sugar treatments 

than in disturbed, unfertilised plots. 

Small scale spatial and temporal N mineralisation fluxes across the C:N degradation threshold “are probably common in even 

strongly N-limited grasslands and may play a role in maintaining grassland diversity” (Wedin 1999) as well as offerring limited 

opportunities for exotic plant invasion. However disturbances involving death of T. triandra produce major nutrient pulses in the 

soil, which strongly facilitate invasion of weeds. Elimination of T. triandra tips the competitive balance in favour of C 3 grasses, 

which produce litter with low C:N ratios, below the threshold that limits microbial breakdown, increase the mineralisation rate 

and perpetuate their own dominance. If N. neesiana happens to be one of those weeds, the cycling of N in the system is 

permanently altered. N. neesiana litter, produced largely in the summer after flowering, breaks down more rapidly than that of T. 

triandra so presumably gives rise to soil nutrient fluxes at a time of year more suitable to its own needs than that of T. triandra. 

Levels of plant-available nutrients will remain higher than in T. triandra swards, so conditions for the growth of other exotic 

weeds will be enhanced. Thus, once N. neesiana becomes established it too would appear to be able to perpetuate its own 

dominance. Grassland N cycling processes therefore explain both the original persistence and abundance of T. triandra and the 

permanent change from C 4 (T. triandra) to C 3 grass dominance (whether native or exotic) that have resulted from European 

grazing regimes and addition of chemical fertilisers (Groves and Whalley 2002, Groves et al. 2003). 

Nutrient enrichment, nutrient reduction and grassland restoration 

Species richness of native forbs declines with increasing P levels (McIntyre andLavorel 2007). Morgan (1998d) compared native 

and non-native species richness and cover in a T. triandra grassland and found a strong positive correlation between soil P 

(unstated method, unclear if this was Olsen ‘available P’) and the number and cover of non-native species, and a negative 

association for native species. There was more than twice the level of soil P at the edge of the grassland than in its centre (30-50 

m from the edge) and the edges were also enriched with ammonium and organic C. There were weak positive correlations 

between ammonium, soil pH, % organic C and exotic richness, and a weak negative correlation between soil pH and native 

richness, but no significant relationships for nitrate and sulfur. Non-native, perennial, high-biomass Poaceae were most 

dependent on high levels of soil nutrients and were suggested to be resource limited in undisturbed soils. 

Various endangered native plants are only known from areas where superphosphate has not been applied, but usually there are 

compounding disturbance factors such as soil cultivation that have probably played a role in reducing population sizes. 

Amphibromus pithogastrus is only known from unploughed sites to which superphosphate has not been applied (Ashton and 

Morcom 2004 p. 2). Cultivation destroys the deep-rooted native perennials, although “there are several instances of the reestablishment 

of native grassland after ploughing” (Kirkpatrick et al. 1995 p. 80). Weed invasion is enhanced more by nutrient 

addition than by cultivation alone (Kirkpatrick et al. 1995). 

There are potentially large nutrient additions to small vegetation remnants from farmland via windblown fertiliser, soil and plant 

material (Hobbs 1989) and ‘run-on’ of nutrient enriched water (Sharp 1997) that could significantly enhance weed invasion 

(Hobbs 1989). Proximity to urban development presumably has similar risks, particularly in regard to nutrient enrichment by 

atmospheric nutrient deposition. N enrichment is occurring globally in the atmosphere, water and soils. As much as 70% of the 

reactive nitrogen in the global system is the result of human activity and deposition rates in industrialised countries increased 

500% in the last 100 years (Hooper 2006) and are expected to double globally by 2050 (Mooney et al. 2006). The increases have 

been largest in agricultural lands, where native grasslands are mostly found (Aguiar 2005). The N comes from various sources: 

fossil fuel combustion, agricultural fertilisers, use of agricultural legumes, mobilisation of soil nitrogen by soil disturbance, and 

ammonia from livestock and human sewage being the most important. Much reactive N enters the soil from acid rain and 

particulate deposition and this is largely ammonium or ammonia (Heil et al. 1988, Hooper 2006). Wind-blown dust contains high 

amounts of N (Eldridge and Mensinga 2007) and increased dust deposition has likely been a feature of the landscape associated 

with increased agricultural development. Heil et al. (1988) measured bulk and throughfall ammonium deposition in a 

Netherlands grassland and found that a significant proportion is captured by the canopy and assimilated by the plants. Total 

deposition was estimated to be 5.9 kg ha -1 over the 3.5 month growing period in undisturbed grassland with a leaf area index of 

2.0-5.7 m 2 m -2 , and 2.5 kg ha -1 over the same period in mown grassland with a leaf area index of 1.8-3.2 m 2 m -2 . Assimilation in 

the canopy was calculated to be 4.7 kg ha -1 in the undisturbed grassland and 1.2 kg ha -1 in the mown grassland. Such increases in 

ammonium availability are more than sufficient to alter competitive relationships between plants and enable fast-growing species 

to outcompete slow-growing ones. Grasslands as a whole are not greatly N-limited and are expected to be in the intermediate 

range of affected ecosystems, but N addition has been demonstrated to have profound effects on grassland biodiversity, 

promoting dominance of a few species, generally fast-growing with high shoot-root ratios, and the suppression of many (see 

Tilman 1987, Aguiar 2005). 

Long term N fertiliser application is also associated with increased soil acidity (Aguiar 2005), which in turn is likely to have 

differential effects on grassland species, Austrodanthonia spp. for example having low tolerance to acidic conditions than 

Microlaena stipoides (Sharp 1997). Fertilisation of grasslands also decreases species diversity of native invertebrates across a 

wide range of taxa and simplifies the trophic web, although normally a few species benefit (Driscoll 1994). 

Increased atmospheric CO 2 improves N use efficiency of C 3 grasses (Milton 2004 see her ref.). A meta-analysis indicates that 

under CO 2 enrichment, C 3 species increased their biomass by 44% while C 4 species increased theirs by 33%, so elevated levels 

are expected to have a relatively high impact in mixed C 3 /C 4 grass systems, although experimental findings have so far been 

mixed (Aguiar 2005). Higher water use efficiency by plants with increased CO 2 levels should decrease transpiration and increase 

soil water content (Aguiar 2005). Increasing concentration of CO 2 is also likely to increase rates of microbial denitrification in 

the soil (Hooper 2006). 

127

Water enrichment may also favour exotic grasses over native grasses. Stafford (1991) found that Lolium perenne was highly 

competitive with T. triandra in cultivation when irrigated, but that cessation of mid-summer watering allowed T. triandra to 

dominate. Seasonal disturbances in the water regime may be important as well as changes to water tables. Grassland restoration 

may also require management of the water regime and water tables. 

Restoration of grasslands to a semi-natural state involves transition from fertilised to low-fertility states (Oomes 1990, 

Kirkpatrick et al. 1995, McIntyre and Lavorel 2007, Eschen et al. 2007). Weedy species remain dominant as long as nutrient 

availability remains high. Reduction in plant available N appears to be the key prerequisite (Eschen et al. 2007) but other 

nutrients may be important. Particular species may be limited by low P or K levels when available N is adequate for their needs. 

Techniques to achieve long-term reductions are poorly developed, and the process is generally prolonged (Kirkpatrick et al. 

1995, Eschen et al. 2007). Some existing management strategies may be successful because they achieve this objective. Oomes 

(1990) suggested that the first stage of such management should aim to reduce annual above-ground dry matter production to 4-6 

t ha -1 . Appropriately managed grazing can also remove nutrients if the grazing animals are harvested. Annual crops have been 

used to reduce nitrate leaching from farm land, and species with proven abilities to sequester N, such as Secale cereale L. could 

be grown and harvested in some highly degraded situations to reduce nutrient levels (Sheley and Rinella 2001). 

Oomes (1990) reported on the restoration of fertilised grassland withdrawn from agricultural use on sand and clay soils in the 

Netherlands to more species-rich grasslands by mowing twice annually over periods of 14 and 11 years respectively, and 

removing the harvested biomass. On the sand substrate, dry matter production fell from 10.2 t ha -1 to 6.5 t ha -1 after 4 y and after 

9 y was similar to that of comparable unfertilised grassland (4.1 t ha -1 ) at which time N and P yields in the vegetation and soil 

were still higher than unfertilsed grassland, but K yields were similar, indicating that K was then the limiting nutrient. On clay, 

dry matter yield decreased from 10.2 t ha -1 to 5.0 t ha -1 after 3 y, but increased again after 6 y. After 10 y low soil N 

concentration was probably limiting biomass production but low P may have been having a similar effect. 

One promising method of nutrient reduction involves applications of C via sugar (sucrose), sawdust and woodchips, to 

manipulate grassland species compostion. These C sources are believed to feed or provide substates for soil microbe populations 

that can temporarily ‘mop-up’ available soil N, and decrease rates of N mineralistation and nitrification. Eschen et al. (2007) 

found that C addition affected the concentration of nitrate, but not that of ammonium and that effects varied between different 

microbial components and at different sites. Sucrose stimulates microbial activity, probably mostly of bacteria, within hours 

while sawdust acts more slowly and wood chips more slowly still, probably largely on fungi. Little is known about the the 

dynamics of the soil microbial community components in relation to C addition (Eschen et al. 2007). The soil microbial 

population may increase, or if it’s biomass remains stable, its N content may increase or microbe consumer populations may 

increase. The method has been used effectively to reduce the competitive ability of invasive plants and above-ground biomass. 

Grass biomass is reduced more than that of legumes, the root:shoot ratio of grasses is significantly increased, annuals are more 

affected than perennials, and more bare ground is created (Eschen et al. 2007). The effects of a single application reduce over 

time, rapidly with sugar and more slowly with wood, and inorganic N pools may be replenished by decay of the mircobial 

biomass. Crushed brown coal, as used by McDougall (1989) as a mulch, similarly increases microbial activity and may have 

similar effectiveness. McDougall (1989) found that 3.5 kg m -2 of crushed coal applied over a mulch of T. triandra culms 

significantly improved establishment and later the flowering of the T. triandra, but did not measure any soil nutrient parameters 

or directly test its effects on weeds. 

Soil disturbance by animals 

Disturbance to the soil surface by animal grazing, burrowing and foraging creates the conditions required for the establishment 

of many plants, including invasive species (Hobbs and Heunneke 1992). Increases in the availability of safe sites for 

establishment is probably the most important effect (Hobbs and Heunneke 1992). Soil disturbance by introduced livestock and 

rabbits in Australian native grasslands is a very important contributor to the establishment and survival of weed populations, but 

may also benefit native species. Livestock trampling often destroys the cryptogam crust, favouring exotic plants (Kirkpatrick et 

al. 1995). But the digging and burrowing of animals (biopedturbation) appears also to be a critical factor in maintaining diversity 

of vascular plants in native grasslands, by creating favourable microsites for germination and seedling survival (Reynolds 2006, 

Kirkpatrick 2007). These disturbances generally modify soil structure and destroy the soil crust (Eldridge and Rath 2002). When 

vertebrate diggings are associated with resting sites, rather than foraging sites, they are likely to also have higher concentrations 

of dung and urine, which can improve the chances of plant establishment (Eldridge and Rath 2002). Biopedturbation alone may 

increase nutrient availability, as well as reducing competition from existing plants (Hobbs and Heunneke 1992). Pits made by 

burrowing animals increase water infiltration and water holding capacity of the soil, trap litter and seeds, and otherwise alter soil 

properties in ways that can enhance seed germination and seedling survival (Noble 1993, Eldridge and Mensinga 2007, James et 

al. 2009). Major effects of bioturbation can persist long after the animals that caused them have disappeared from the landscape 

(Villarreal et al. 2008). 

Vertebrate burrows and warrens generally result in nutrient enrichment of the soil, particularly with total and available N 

(Garkaklis et al. 2003, Villarreal et al. 2008). In semi-arid Australian rangelands biopedturbation results in long-term changes to 

structure and spatial patterning of surface soils and to alterations in microtopography (Noble 1993). Even the shallow hip holes 

constructed by Macropus spp. as resting places can significantly alter soil erodibility and water infiltration rates, and concentrate 

plant litter, dung and nutrients, notably N and S (Eldridge and Rath 2002). James et al. (2009) found that litter accumulation and 

seedling emergence at an Australian desert site was almost entirely restricted to vertebrate foraging pits. Pits acted as resource 

sinks and their effectiveness was possibly releated more to their capability of retaining trapped material than capturing it. A 

variety of small pits were possibly as effective as few, large pits. Vertebrate biopedturbation can also have opposite effects 

including increased runoff, reduced litter concentrations and physical compaction of the soil, depending on the particular soil, the 

nature of the disturbance and environmental factors (Eldridge and Rath 2002). In Tasmania, marsupial and aboriginal 

biopedturbation created “a widespread ... frequently renewed, regeneration niche” (Kirkpatrick 2007 p. 222) which is now absent 

in many areas. 

128

Fossorial vertebrates are or were once a feature of the indigenous fauna of temperate grasslands worldwide. Darwin (1845 p. 51) 

reported that “considerable tracts” of grassland of the Maldonada region in Uruguay were undermined by the extensive shallow 

burrowing of the root-feeding Tucotuco Ctenomys brasiliensis Blainville (Rodentia: Ctenomyidae), while the commonest 

mammal in grasslands between the Rio Negro and Bahía Blanca in Argentina was the burrowing Agouti or Mara (Dolichotis 

patagonum Zimmerman) (as Cavia patagonica) (Caviidae). Another burrowing rodent, the Viscacha (or Plains Vizcacha) 

Lagostomus maximus (Desmarest) (Chinchillidae), larger than the Agouti, lives in large groups in the pampas between the 

Uruguay and Rio Rivers, makes deep, multi-chambered burrows with large soil mounds, known as viscacheras, and forages at 

night on grasses and herbs. It is considered a”key ecosystem engineer” whose grazing and burrowing activities change the 

structure and composition of the vegetation over extensive areas (Villarreal et al. 2008 p. 701). This speices remained abundant 

at least until the early 1970s (Soriano et al. 1992). Rheas were abundant in the grasslands of Bahía Blanca, and live on “roots and 

grass” (Darwin 1845 p. 89). Burrowing mammals have been implicated in significant changes in microtopography (soil mounds) 

in Argentina (Noble 1993). The effects of L. maximus activities cascade through the ecosystem, and include altered plant 

biomass distribution, nutrient cycling (primarily by deposition of faeces and urine in the burrow), nutrient content of plants and 

fire regimes (Villarreal et al. 2008). 

Other small mammals in the regions where N. neesiana is found include a group of Necromys mice (Rodentia: Cricetidae) of 

open areas (D’Elía et al. 2008). N. lactens (Thomas) is found in high altitude grasslands (over 1500 m) in north-western 

Argentina and southern Bolivia including Catamarca, south Jujuy and Tucuman; N. obscurus onthe Atlantic coast of Uruguay 

and in areas around the La Plata River and N. lasiurus is very widely distributed including southern Buenos Aires Province 

(D’Elía et al. 2008). The oldest Necromys fossils are 3.5-4 million years old from southern Buenos Aires Province and the group 

probably radiated in the Pampean region during the late Pliocene (D’Elía et al. 2008). 

The biopedturbation activities of mammals now extinct or rare in temperate Australian grasslands probably played a critical role 

in the maintenance of forb diversity (Reynolds 2006). The mechanisms include alteration to the spatial distribution and cycling 

of soil nutrients and water infiltration (Garkaklis et al. 2003). Bettongs, potoroos, bandicoots, bilbies and rodents are amongst the 

most important digging and burrowing groups (Garkaklis et al. 2003). Bilbies are powerful burrowers, prefering soft soils such 

as sand dunes (Noble 1993). Reptiles, including goannas, may also be important, and the feral animal diggings, including those 

of rabbits, may function in a similar fashion to those of native vertabrates (James et al. 2009). 

Bettongia species probably mostly feed on underground fungi and may be opportunistically insectivorous and omnivorous 

(Seebeck and Rose 1989). They sometimes bury and store seed, “which are eaten later, often after germination” (Seebeck and 

Rose 1989 p. 721). The Rufous Bettong Aepyprymnus rufescens (Gray) is primarily rhizophagous buts eats fungi throughout the 

year (Seebeck and Rose 1989). Foraging activities of Bettongia penicllata Gray for hypogeous fungi commonly cover the ground 

with small diggings of a range of different sizes and ages. In plan they are elliptial with a spoil heap at one end and a steep wall 

to a depth of 10-15 cm at the other. These excavations accumulate litter, leading to concentrations of buried fungal hyphae which 

become water repellent lenses in the soil after gradual infill of the excavations. The increased water infiltration in the 

excavations is the probable cause of decreased available nitrate and sulfur found in the soil of old, simulated diggings, while 

decreased ammonium may be due to rapid nitrification (Garkaklis et al. 2003). 

The Burrowing Bettong Bettongia lesueur is considered to be the most fossorial of the potoroids (Strahan) and constructed 

warrens up to 30 m in diameter that could contain over 100 entrances or were simple structures with only one or two entrances, 

the former sometimes associated with rock caps and the latter sometimes on plains (Noble 1993, Noble et al. 2007). It is “a 

powerful burrower ... capable of penetrating the underlying rock” (Noble 1993 p. 60) and its activities probably significantly 

promoted landscape heterogeneity and plant diversity (Noble et al. 2007). B. lesueur “may have helped maitain vast areas as 

grassland by eliminating shrubs” (Noble et al. 2007 p. 335). 

Among the native mammals still present in some temperate Australian native grasslands, the biopedturbation activites of the 

Short-beaked Echidna may be the most significant. Much of the foraging activity of the Echidna requires digging to obtain 

invertebrates, and the animals shelter in temporary digs and prepared tunnels (Menkhorst 1995c). Nursery burrows are shallow 

and 1-1.5 m long (Menkhorst 1995c) while foraging disturbances range from nose-poke holes, through shallow scrapes to deep 

digs and extensive bulldozing (Eldridge and Mensinga 2007). In semi-arid woodlands with a grassy understorey this soil 

disturbance has been found to make a large contribution to landscape patchiness (Eldridge and Mensinga 2007). The foraging 

pits accumulated greater quantities of plant litter than undug areas, have moister and more porous soil, are cooler and have a 

different suite and greater abundance of soil micro-arthropods than the surface soil. Contrary to expectations, increased litter in 

pits did not result in increases in total or active C, total N and available P, although the N was probably largely immobilised in 

the litter. Ultimately, echidna pits probably influence plant germination and establishment (Eldridge and Mensinga 2007). 

The hip hole diggings of Macropus spp. are constructed so as to assist the animals to cool, and are generally located in the shade 

of trees or shrubs (Eldridge and Rath 2002), so may be of little significance in grasslands. 

Biopedturbation by invertebrates may be more important than that of vertebrates. Ants and earthworms are probably the most 

important groups. Ants are a prominent feature of temperate south-eastern Australian grasslands. Many species construct surface 

mounds of excavated material around their nest entrances, associated with tunnels that may descend well under the surface. 

When the effects of their activity is quantified over the long term they can be viewed as ecological engineers whose activities 

restructure the landscape, generate heterogeneity, and affect soil structure and porosity, the distribution of soil nutrients and 

regeneration of the flora (Richards 2009). The vertical tunnels of Underground Grass Caterpillars Oncopera fasciculatus 

(Walker) (Hepialidae) reach a depth of up to 23 cm at maturity, depending on the ease with which the soil can be dug, and the 

excavated soil is deposited around the tunnel entrance and on the larval feeding runways (Madge 1954). Many other insects have 

larvae which live beneath the soil surface and contribute to biopedturbation. Earthworm activity in temperate Australian 

grasslands can also be substantial. However there appear to be no published studies of the effects of bioturbation by invertebrates 

in temperate native grasslands. 

129

Native temperate grasslands of south-eastern Australia and their conservation status 

According to the Australian Native Vegetation Assessment 2001 (Cofinas and Creighton 2001), 60,214 km 2 of the pre-European 

area of 589,212 km 2 of tussock grasslands in Australia had been cleared or substantially modifed by grazing by c. 2000. The 

category includes grasslands dominated by Astreleba, Sorghum etc., mostly in Queensland (282, 547 km 2 ) and the tussock 

grasslands of the dry inland, the largest proportion by far of which was in the north of the continent. Of all the major vegetation 

groups, grasslands are among the most affected by clearing in Victoria, ACT and NSW (See Table 9). The Victorian Midlands, 

Victorian Volcanic Plains (Victoria and South Australia) and the South East Coastal Plain (Victoria) had less than 30% of their 

orginal native vegetation remaining (Cofinas and Creighton 2001). 

Table 9. Areal extent (km 2 ) of tussock grasslands in the ACT, NSW and Victoria pre-European settlement and c. 2000. Source: 

Cofinas and Creighton (2001). 

State/Territory Pre European c. 2000 % remaining 

Australian Capital Territory 207 91 44 

New South Wales 40,790 19,318 47 

Victoria 19,175 614 3 

Thus it is the native temperate grasslands along with neighbouring grassy woodlands that are among the most threatened 

ecosystems in Australia, and amongst the most inadequately represented in reserves (Lunt 1991, Eddy et al. 1998). Whole 

grassland ecosystems have therefore been afford threatened status in south-eastern Australia (Table 10). 

Table 10. Threatened grassland communities in south-eastern Australia. CE = critically endangered, E = endangered, U = 

unlisted, - = not present 

Community ACT NSW Vic References 

Natural Temperate Grassland E U Sharp and Shorthouse 1996, ACT Govt 1997 

Native Vegetation on Cracking Clay Soils of the 

Liverpool Plains (partly Western Slopes Grasslands) 

- E - Keith 2004 

Natural Temperate Grassland of the Victorian 

Volcanic Plain 

- - CE DEWHA 2008 

In 1971 there were no areas >4,000 ha in conservation reserves (Groves 1979). In the late 1970s the possibility of conserving 

moderately large areas seemed remote, because only small remnants were then known to exist (Groves 1979). By the early 1990s 

natural temperate grasslands had declined by 99.5% since European settlement (McDougall and Kirkpatrick 1994, Sattler and 

Creighton 2002). The 0.5% estimated to remain included remnants in “semi-natural condition” (Kirkpatrick et al. 1995 p. v). By 

the late 1990s, in all areas of south-eastern Australia, less than 1% remained of the grasslands and grassy woodlands that once 

existed (Lunt et al. 1998). Widespread extinctions occur when habitats are reduced by 90% and extinctions of vertebrates are 

continuing to occur in southern Australia in areas where native vegetation is reduced by 70-80% (Traill and Porter 2001). 

Existing remnants of southern Australian temperate grassland on public land are mostly small and isolated (Lunt and Morgan 

2002). Those on private land are generally larger, but are of lower quality, and all have a history of grazing (Kirkpatrick et al. 

1995). Sattler and Creighton (2002) identified the NSW South Western Slopes, Riverina and Victorian Volcanic Plain as 

bioregions with the highest priority for improvement of reservation status. These areas cover a high proportion of the land once 

occupied by native grasslands. 

The extreme fragmentation of the conserved elements (Eddy et al. 1998, Groves and Whalley 2002) means that they are subject 

to inundation or impoverishment by the surrounding ‘cultural steppe’, an immature ecosystem with a higher proportion of energy 

going into production than into maintenance, rapid accumulation of biomass, reduced litter production, increased excrement 

production and reduced species diversity with most species being r strategists (rapid intrinsic rates of population increase, little 

food specialisation, etc.) (Matthews1976). Nutrients, biomass, pollutants, exotic organisms, native pests and other ecological 

disrupters may be flushed in (Hobbs and Heunneke 1992), while native organisms or their propagules, water, etc. are extracted. 

Many environmental changes that may substantially affect the composition and functioning of small remnant grasslands have 

rarely received any consideration, including altered climatic regimes, light availability and noise profiles. 

Thus ‘fragmentation’ as an outcome of development needs to be contrasted to with the increased connectivity that also results. 

Utility infrastructure including power lines, fences, drains and traffic corridors have provided a new network of corridors that 

facilitate dispersal of a different set of organisms and allow the movement of a range of non-biological resources between 

remaining grassland remnants (Aguiar 2005). A series of small grasslands may nevertheless sustain more diversity than a large 

patch with an equivalent total area, because different species may dominate in small patches, there are more edges and 

transitional habitats (Hobbs and Heunneke 1992) and a greater liklihood of heterogeneity in disturbance regimes. 

Victoria 

In simplistic terms, the major plant formations of southern Victoria are determined by geology: heathlands on sands, Eucalyptus 

forests and woodlands on sedimentary rocks, and grasslands on clays of volcanic origin (Patton 1935). Victorian grasslands 

occur mainly on relatively young geological materials, such as Tertiary volcanics and Pleistocene and Holocene sediments 

130

(Rosengren 1999). Approximately one third of Victoria (c. 8 million ha) was probably occupied by grassy ecosystems, including 

grasslands and grassy woodlands, at the time of European occupation, of which less than 0.5% remains (Lunt 1991). All 

Victorian native grasslands are severely depleted (DNRE 1997) and all are now listed as threatened communities (DSE 2009a), 

that is Central Gipssland Plains Grassland Community, Northern Plains Grassland Community, Plains Grassland (South 

Gippsland) Community and Western (Basalt) Plains Grassland Community. 

“Plains Grassland” is one of 28 Broad Vegetation Types recognised in Victoria, occurring on fertile plains with 300-1000 mm 

annual rainfall, on heavy clay soils of basaltic origin (Western Volcanic Plains), outwash clays (Sale Plains) and alluvial silts and 

clays (Northern Plains, Wimmera and West Gippsland). The category does not include coastal grasslands of Austrostipa 

stipoides, dune grasses and sedges. Pre-1750 this formation is estimated to have covered 1,882,411 ha or c. 8.2% of the State, of 

which 8277 ha (0.44%) was considered to be extant in the early 1990s. Only 2504 ha (30%) was in conservation reserves, while 

4938 ha (60%) was on private land (Traill and Porter 2001). These estimates are similar to those of DNRE (1997) (Table 11). 

Ecological Vegetation Classes (EVCs) are a more accurate categorisation of actual vegetation types present in the State. The 

grassland EVC on the Western Volcanic Plains is Western Plains Grassland, in Gippsland it is Gippsland Plains Grassland and in 

the north is Northern Plains Grassland. In general, much higher proportions of these Plains Grasslands EVCs have been lost than 

of other grassland formations (Table 12). 

Table 11. Bioregional pre-European areal distribution of grassland complexes in Victoria and proportion remaining. Source: 

DNRE (1997). ‘Area remaining’ derived by calculation from the ‘% remaining’. 

Region Victorian Bioregion Pre-European 

area (ha) 

Area 

remaining (ha) 

% remaining 

Victorian Mallee Murray Mallee 51,512 52 0.1 

Wimmera 329,737 1319 0.4 

Victorian Volcanic Plain Victorian Volcanic Plain 826,402 1652 0.2 

Midlands Dundas Tablelands 79,694 80 0.1 

Goldfields 4,601 0 0.0 

Victorian Riverina Victorian Riverina 444,713 3113 0.7 

Coastal Plains Gippsland Plain 133,547 1870 1.4 

Total 1,870,206 8086 0.4 

All Victorian grasslands are presumed to have lost significant biodiversity since European occupation. Native vegetation and 

species disappeared locally and regionally in many areas before any detailed records were made (Mansergh et al. 2006a). Exotic 

plants have been reported to comprise 29% of the Southern Victoria grassland vascular flora (Kirkpatrick et al. 1995). 

Gippsland Grasslands 

The West Gippsland plains consist largely of uplifted marine and fluviatile sediments and swamp deposits at Koo-wee-rup (Hills 

1967). The largest areas of grasslands on the Gippsland Plains at the time of European settlement were between Stratford, 

Rosedale and Sale, north of the Latrobe River with some onthe floodplains of the Latrobe and Macalister Rivers, in coastal South 

Gippsland east of Yarram between Seaspray and Welshpool, and north of Westernport around Kooweerup Swamp (DNRE 1997, 

Lunt et al. 1998, Rosegren 1999). The Central Gippsland grasslands occur on leached sandy soils in areas with perched 

watertables and an abundance of swamps and small lakes (Rosegren 1999). In the Nambrok-Denison area, between the Latrobe 

and Thomson Rivers, grassland occupied areas with heavy loam topsoil over heavy clay on flat, alluvial plains 3.5-7 m above the 

floodplain of the Thomson, and were infrequently flooded (Kirkpatrick et al. 1995). Moore (1993) mapped the coastal plain from 

Sale to Lakes Entrance as Temperate Shortgrass (Austrodanthonia- Austostipa-Enneapogon). To the north, in primarily forested 

country he mapped a wide zone of Temperate Tallgrass (Themeda-Poa-Dichelachne) continuous up the east coast to north of 

Sydney,west across Victoria to the Hamilton area and inland to Rutherglen and the ACT. 

The historical extent and nature of Gippsland grasslands is very poorly known. Indeed, Gullan et al. (1985) did not recognise the 

presence of any natural grassland community in South and Central Gippsland, and Moore (1993) mapped the whole of West and 

South Gippsland except Wilsons Promontory (!) as an area without native grasslands. However many secondary grasslands 

derived from grassy woodlands currently exist in the region (Lunt et al. 1998) and several remnant native grasslands have 

recently been recognised. 

Kirkpatrick et al. (1995) briefly described two Gippsland grassland communities: 1. South Gippsland Themeda Grassland, a 

closed tussock formation, occasionally with significant amounts of Hemarthria uncinata R.Br. (Poaceae) and a variety of other 

grasses and rushes, found in the Welshpool-Seaspray areas; and 2. Central Gippsland Themeda Grassland, also a closed tussock 

formation, but with abundant Asteraceae, Liliaceae and Orchidaceae and scattered trees at some sites, found on alluvial plains 

between Traralgon and Johnsonville and derived originally from woodland or open forest. 

Substantially different grasslands have recently been identified in the Westernport and Mornington Peninsula regions. These 

have been described and delineated by inspections of the very restricted remnants that remain, and by analysis of early survey 

maps, plans, fragmentory historical records, aerial photographs and other sources (Cook and Yugovic 2003, Yugovic and 

Mitchell 2006, Sinclair 2007). They include c. 12 km 2 north of Tooradin (Cook and Yugovic 2003) and small areas (< 6 km 2 ) at 

Safety Beach (Sinclair 2007). Poa labillaredieri was the major grass, dominating the wetter areas, with T. triandra on drier sites 

(Yugovic and Mitchell 2006). The Westernport grasslands were occasionally flooded, but aboriginal burning probably prevented 

them being overrun by Melaleuca ericifolia Sm., which also may have been constrained by soil factors (Lunt et al. 1998, 

131

Yugovic and Mitchell 2006). The Safety Beach grasslands are on alluvial clays, were seasonally swampy, and are dominated by 

Poa poiformis (Labill.) Druce (very close to or identical with P. labillardieri var. (Volcanic plains)) and Notodanthonia 

semiannularis (Labill.) Zotov, along with T. triandra, and are also considered to be Gippsland Plains Grasslands (Sinclair 2007). 

Similar grasslands once occurred at Carrum (Sinclair 2007). Both the Westernport region and Safety Beach grasslands included 

Acacia melanoxylon R.Br. in the overstorey in some areas, and were transitional into grassy woodlands and Melaleuca ericifolia 

swamp scrub. 

132

Table 12. Areal extent and conservation and reservation status of grassland and grassy woodland Ecological Vegetation Classes in Victoria by bioregion. Source: Traill and Porter (2001), June 2000 

data from Department of Natural Resources and Environment. Conservation status: X = extinct, E = endangered, V = vulnerable, R = rare, D = depleted, LC = least concern; Reservation status: X = 

EVC not reaching reservation target, Y = reaches reservation target. Excludes swamp and wetland categories and tree or shrub dominated formations including a grassland component. 

Ecological Vegetation Class Bioregion % of Bioregion 

mapped 

Conservation 

Status 

Reservation 

Status 

Pre-1750 

area (ha) 

2000 Extent 

(ha) 

Proportion 

existing (%) 

Plains Grassland Central Victorian Uplands 100 E X 10490 219 2 

Gippsland Plain 75 E X 37327 291 1 

Glenelg Plain 100 E X 46214 944 2 

Goldfields 99 E X 1560 0 0 

Murray Mallee 4 7862 3 0 

Otway Plain 100 E X 2064 76 4 

Victorian Riverina 77 E X 168816 579 0 

Victorian Volcanic Plain 100 E X 220073 2291 1 

Wimmera 41 24612 3 0 

Plains Grassland/Stony Knoll Shrubland Victorian Volcanic Plain 100 761 0 0 

Plains Grassland/Plains Grassy Woodland Mosaic Victorian Volcanic Plain 100 456163 3326 1 

Riverina Plains Grassy Woodland/Plains Grassland Mosaic Central Victorian Uplands 100 2799 241 9 

Goldfields 99 1696 5 0 


Victorian Riverina 77 6808 16 0 

Victorian Volcanic Plain 100 8873 333 4 

Wimmera 41 3607 3 0 

Wimmera Plains Grassy Woodland/Plains Grassland Mosaic Goldfields 99 90 4 4 


Wimmera 41 6739 2 0 

Riverina Plains Grassy Woodland/Plains Grassland/Gilgai Plain Woodland Mosaic Goldfields 99 11307 24 0 

Victorian Riverina 77 415 0 0 

Black Box Chenopod Woodland/Plains Grassland Mosaic Wimmera 41 190 0 0 

Swamp Scrub/Plains Grassland Mosaic Gippsland Plain 75 22218 0 0 

Plains Grassland/Gilgai Plain Woodland/Wetland Mosaic Victorian Riverina 77 16180 87 1 

Creekline Tussock Grassland Victorian Volcanic Plain 100 X 2591 0 0 

Coastal Tussock Grassland Otway Plain 100 V X 249 176 71 

Otway Ranges 100 V Y 62 54 86 

Gippsland Plain 75 R X 744 522 74 

Victorian Volcanic Plain 100 V X 30 20 68 

Warrnambool Plain 100 R Y 198 160 81 

Coastal Headland Scrub/Headland Coastal Tussock Grassland Mosaic Otway Plain 100 253 193 76 

Mangrove Shrubland/Coastal Salmarsh/Berm Grassy Shrubland/Coastal Tussock Grassland Otway Plain 100 61 26 43 

Coastal Headland Scrub/Headland Coastal Tussock Grassland Complex Warrnambool Plain 100 1094 490 45 

Coastal Dune Grassland Gippsland Plain 75 R X 32 32 100 

Scree-slope Grassland/Woodland Greater Grampians 100 32 7 22 

Calcareous Swale Grassland Gippsland Plain 75 R Y 558 310 55 

Montane Grassland Highlands - Northern Fall 100 E X 2013 52 3 

Montane Grassy Woodland/Montane Grassland Mosaic Highlands - Northern Fall 100 1867 1 0 

Sub-alpine Grassland Highlands - Northern Fall 100 R Y 879 879 100 

Highlands - Southern Fall 100 R X 987 983 100 

Victorian Alps 100 LC Y 13925 13492 97 

Sub-alpine Wet Heathland/Sub-alpineGrasslandMosaic Highlands - Northern Fall 100 315 285 90 

Highlands - Southern Fall 100 696 696 100 

Victorian Alps 100 2643 2437 92 

133

Grasslands of the Gippsland Plains have substantially disappeared and are now largely restricted to small linear remnants (DNRE 

1997) (Tables 11 and 12). Those at Safety Beach are on small urban blocks destined for housing (Sinclair 2007). Threats include 

weed invasion, livestock grazing and roadworks (DNRE 1997). Exotic grasses are the most serious weeds at Safety Beach and 

include Anthoxanthum odoratum, Paspalum dilatatum, Pennisetum clandestinum, Holcus lanatus, Festuca arundinacea, 

Sporobolus africanus, Phalaris spp. and Stenotaphrum secundatum (Walter) Kuntze (Sinclair 2007). Fire is now believed to be 

less frequent and more intense than in pre-European times (DNRE 1997). 

Northern Plains Grasslands 

Much of the Murray Valley Riverine Plains were reputedly covered with grassland at the time of European settlement. The 

Victorian Riverina plains grasslands are ecologically similar to those on the floodplains of the Lachlan and Murrumbidgee 

Rivers in New South Wales. The Northern Plains Grasslands largely occupied gently sloping, Quaternary alluvial stream 

deposits, often very deep, built of sedimentary and igneous materials derived from the Victorian highlands, deposited on flat 

plains with residual volcanic hills (Hills 1967, Rosengren 1999). 

The soils are

Foreman (1997) investigated the effects of livestock grazing, cultivation and burning on a long-grazed species-rich remnant. 

Only the diminuitive annual species were not grazed and they increased in abundance under grazing by exploiting the increased 

gaps. Cultivation increased the abundance and species richness of exotic species, and the abundance of some native annuals 

(probably because they had a soil seed bank). Fire reduced the abundance of exotics, particularly annuals, presumably by 

destroying surface seeds. He also found that drought had a very significant effect on the flora, causing a substantial drop in 

above-ground species richness and abundance, and that native annuals were more abundant when winter rainfall was high. 

Irrigated agriculture is the major landuse in the Riverina and salinity is a widespread problem, while other threats include 

cultivation, irrigation, inappropriate grazing and fire regimes, weed invasion, feral animals and increased fertiliser use (DNRE 

1997). 

The Northern Plains Grasslands are listed as a threatened community under the Flora and Fauna Guarantee Act. (Department of 

Sustainability and Environment 2009a). Terrick Terrick National Park near Mitiamo, a former grazing property, is the largest 

and most important conserved remnant of this ecosystem in Victoria (Lunt et al. 1998). 

Victorian Basalt Plains Grasslands 

The grasslands of the Victorian basalt plains were a very large area of grassland at the time of European settlement (Tables 11 

and 12). The grassy ecosystems of the plains extended from Melbourne to Hamilton (Stuwe and Parsons 1977) or to “beyond the 

South Australian border, broken only in a few places by deeply dissected ranges” (McDougall 1987 p. 17), at altitudes from less 

than 100 m to over 600 m (Stuwe and Parsons 1977). The Volcanic Plains as a land unit occupies c. 22,000 km 2 of which about 

37% or 8,260 km 2 (or over 10,000 km 2 according to Williams 2007) was occupied by grassland complexes and most of the 

remainder by grassy woodland (28%) and herb-rich woodland (14%) complexes (DNRE 1997, Barlow and Ross 2001). 

Numerous other vegetation types occurred, including wetland, forest and swamp scrub, forming a complex mosaic, in which, 

judging by the maps of early surveyors, “treeless areas occurred as discrete ‘openings’ in the landscape”, which they identifed as 

“Plains” (Barlow and Ross 2001 p. 26). 

The majority of the Victorian Volcanic Plains terrain is composed of Newer Volcanics, a series of lava flows and tuff plains with 

extinct volcanic cones (commonly scoria cones but also lava domes) aged from 46 million years ago, peaking at about 2.4 

million years and ceasing about 7,000 years ago, i.e. Middle Pliocene to recent (Hills 1967, Dahlhaus et al. 2003), although 

radiocarbon evidence indicates that the most recent volcanic activity was >20,000 years ago (Rosegren 1999). The most recent 

lava flows are represented by stony rises of ropy lava (Hills 1967). Smaller volcanoes, such as Mt Cottrell, produced the majority 

of the geological material, mostly highly fluid, basalt lava, that spred in thin sheets 20-200 cm thick, with overlapping flows to 

60 m thick in some areas (Rosengren 1999). Eruptions were infrequent and there was little pyroclastic material and ash in most 

areas, so vegetation would have survived widely beyond the edges of the lava flows (Rosengren 1999). 

Despite common parent materials, the soils are highly variable, ranging from gradational clayey chocolate soils and kraznozems 

on the younger basalts, to coarsely structured duplex soils on the older rocks (Dahlhaus et al. 2003). Much of the lava plains was 

“originally rough and stony, resulting in an irregular topography with many small depressions” (McDougall 1987). In areas that 

have not been worked, there is still abundant surface rock (Hills 1967), and in areas with developed soils, basalt corestones 

(‘floaters’) are common at the surface. Much of the rock in developed areas has been cleared, and many kilometres of dry stone 

walls were built (Patton 1935). The rock is fine-grained, blue-grey in colour and not very hard (Patton 1935) and underlying rock 

is very close to the surface in many areas (Sutton 1916-1917). The development of watercourses is generally poor, with lava 

flows often resulting in blockages and substantial areas of internal drainage to ephemeral wetlands, swamps and lakes (Dahlhaus 

et al. 2003), many of which have been drained (McDougall 1987) Waterlogging is widespread (Dahlhaus et al. 2003), partly 

because the soils are so shallow (Patton 1935), but much of the area is well drained due to strong fracturing of the underlying 

rocks (McDougall 1987). 

The eastern areas are Miocene to recent lava plains, with two dominant soil types, formed in situ: grey cracking clays, Northcote 

classification Ug 5.2, and hard alkaline red soils Northcote classification Dr 2.13, ranging in surface texture from sandy loam to 

clay, but mostly clay loams or light clays, and in colour from black to grey, brown and reddish brown (Stuwe and Parsons 1977). 

Patton (1935) thought the darker soils characteristic of more low-lying areas, the colour attributable to leaching or organic matter 

from upslope, and considered that soils higher up the slope had lower clay content. The clays have a high water-holding capacity 

(45% according to Patton 1935) and high shrink:swell ratio so become waterlogged with poor aeration in winter and deeply 

cracked in summer (Stuwe and Parsons 1977). With a moisture content of 19%, the soil is “definitely sticky” (Patton 1945 p. 

185). Patton (1935) provided a generalised soil profile of a red-brown earth on the Keilor Plains and noted that the B horizon, 

extending from c. 40-85 cm depth, is not generally penetrated by roots and rests directly upon bedrock. It contains abundant 

calcium carbonate giving it a white colour. “Most of the richest aricultural and pastoral land in the Western District is composed 

of ... recent tuffs” (Hills 1967). For Australian soils, the clays are relatively nutrient rich, particularly in phosphorus (Barlow and 

Ross 2001). 

The Victorian Volcanic Plain has average annual rainfalls of 500-700 mm (Dahlhaus et al. 2003) and western Victorian 

grasslands between 450-850 mm (Jones 1999b) , with some areas in the east in rain shadow (between Melton and Werribee) 

receiving slightly less (Dahlhaus et al. 2003), the areas between Anakie and Melbourne with a range of c. 400-600 mm 

(McDougall 1987), and areas at higher altitude receiving up to 1000 mm, and is evenly distributed throughout the year (Stuwe 

and Parsons 1977) or with a late winter peak (Jones 1999b). Eight major droughts occurred in the eastern area between 1887 and 

1987 (McDougall 1987). Average monthly temperatures are 7-20ºC (Jones 1999b). February is hottest, with maxima sometimes 

exceeding 40ºC and annual frost days in the eastern area range from 16 to 53 (McDougall 1987). Evaporation is very high in the 

summer (December to February). The precipitation to evaporation ratio appears to have fallen over the last 100 years, so the area 

seems to be drying (Dahlhaus et al. 2003). Much of the western area is severely affected by rising water tables and dryland 

salinity, predicted to dramatically worsen (Dahlhaus et al. 2003). Mean daily minimum and maximum temperatures are 5-12ºC 

in July and 12-26ºC in January (Stuwe and Parsons 1977). 

Victorian western plains grasslands a result of both soil and climate factors, not aboriginal burning (Cook and Yugovic 2003). 

Treelessness is explained by seasonal aridity,cold temperature and heavier clay soils (Willis 1964 p. 398). Fires and ash falls 

135

esulting from vulcanism would have periodically impacted on the vegetation (Rosegren 1999). Some secondary grasslands are 

present, resulting from tree removal or suppression of tree regeneration by livestock, notably on stony rises (Kirkpatrick et al. 

1995). 

Grasslands in the Victorian Midlands are included in this category. Little is known about the original extent and composition of 

the Midlands grasslands. They were mostly in the Dundas tablelands, an undulating area west of the Grampians, 90% of which 

has now been cleared and which currently supports sheep and cattle grazing (DNRE 1997). The principal areas were in the 

Glenelg and Wannon catchments west of Hamilton. Smaller areas were present in the Goldfields and Central Victorian Uplands 

Bioregions. Remnants are are often degraded (DNRE 1997). Prior to European occupation the heavy clays in the Wannon region, 

near the western end of the Volcanic Plains, were often occupied by grasslands (Lunt et al. 1998). 

The flora of the Victorian volcanic plains as a whole is the most highly modified vegetation in the State (Willis 1964). The 

grasslands were the first areas of Victoria occupied by squatters in the 1830s and 1840s for sheep and cattle grazing (Willis 

1964, Turner 1968) and “conversion of tribal territory to sheep-runs was rapid and decisive” (Mulvaney 1964 p. 427). These 

areas “were especially valuable because they required little or no clearing” (Wadham and Wood 1950 p. 84). According to 

Marshall (1968 p. 165) “the native flora was virtually exterminated over hundreds of square miles”. Nevertheless there is still 

high diversity in the area as a whole. Carr (1999) provided a checklist of the Victorian Volcanic Plains indigenous flora, listing 

the conservation status of each taxon and the vegetation formations in which they occur. Treeless grasslands and grassy 

woodlands were not considered to be separate formations. Stony Rise Complex, in many ways a similar floral assemblage to the 

grasslands, was treated as a separate entity. Asteraceae and Poaceae were by far the richest families (Table 13). McDougall 

(1987) provided a list of the common vascular plant species in the basalt plains tussock grassland in the eastern section of the 

Plains (Melbourne region) and a list and brief description of 32 remnant sites on private and public land. 

Table 13. Number of plant taxa in plant families with the greatest number of taxa in Victorian Volcanic Plains grasslands and 

grassy woodlands. Derived from Carr (1999). Taxa doubtfully occuring in this vegetation formation are recorded as “Uncertain”. 

Liliaceae is the combination of Carr’s Anthericaceae, Asphodelaceae, Colchichaceae, Hypoxidaceae and Phormicaceae. 

Family 

No. of taxa 

Known Uncertain Total 

Asteraceae 61 9 70 

Poaceae 56 7 63 

Orchidaceae 18 20 38 

Liliaceae 23 1 24 

Fabaceae 19 1 20 

Chenopodiaceae 16 2 18 

Cyperaceae 13 1 14 

Juncaceae 10 1 11 

Two main types of grassland occur, dominated respectively by T. triandra in the drier eastern portion and by Poa labillardieri in 

the higher rainfall western portion, and grassy wetlands occurred in areas subject to seasonal inundation (Lunt et al. 1998, 

Barlow and Ross 2001). The daisies Rutidosis leptorhynchoides and Senecio macrocarpus have been identified as subdominant 

herbs on red-brown earth soils (Hills and Boekel 2003). T. triandra was the original main dominant over a high proportion of the 

area, with subdominant Austrodanthonia and Austrostipa spp., forming discrete tussocks with a range of herbs in the intertussock 

spaces (Lunt 1990a). In the eastern area Austrodanthonia spp. are occasionally locally dominant, and low lying areas are 

dominated by this genus or P. labillardieri and are “usually species rich”, while Themeda-dominated areas may be depauperate 

due to dense cover (McDougall 1987 p. 17).. The T. triandra tussocks are rarely taller than 30 cm (not including panicles) and 

have a basal diameter of 5-15 cm (Specht 1970). 

Kirkpatrick et al. (1995) considered that most of the basalt plains grassland was a T. trindra – Eryngium ovinum – Schoenus 

apogon association, with other characteristic species including Acaena echinata (Rosaceae), Leptorynchos squamatus (Labill.) 

Less. (Asteraceae) and Convolvulus spp., and at drier sites, Calocephalus citreus Less. (Asteraceae). Other areas were dominated 

by a T. triandra - Austrodanthonia setacea community, similar to, and possibly a degraded form of the former. Stony rises were 

occupied by a community dominated by T. triandra, Austrostipa semibarbata and Poa sieberiana, sometimes with scattered 

shrubs of Bursaria spinosa or small trees (Acacia melanoxylon R.Br. and Allocasuarina verticillata (Lam.) L.A.S. Johnson), 

possibly derived from woodlands. Another T. triandra community, occurring in the Merri Creek Valley and on the Keilor Plains, 

is characterised by sparse shrubs of Acacia paradoxa DC., and commonly Austrostipa mollis (R.Br.) S.W.L. Jacobs and J. 

Everett and Austrodanthonia carphoides Kirkpatrick et al. (1995). McDougall (1987) found that many early maps referred to 

occasional ‘Honeysuckle’ trees in the grasslands to the north-west and west of Melbourne and concluded that these were 

probably Banksia marginata Cav. (Proteaceae), a species which no longer existed in this formation in the region. Barlow and 

Ross (2001) expanded this interpretation and considered this formation to be a Banksia marginata Cav./Allocasuarina 

verticillata/Acacia implexa Benth. woodland, none of which survived into modern times (Barlow and Ross 2001). 

The grasslands formed a complex mosaic with other vegetation types, including wetlands and riparian areas, grassy woodlands, 

shrublands and herblands, which have been similarly degraded or destroyed (McDougall 1999, Barlow and Ross 2001). In the 

western part of the plains, Grassy Woodland dominates wherever rainfall exceeds c. 700 mm per annum and trees naturally 

occured on the most rocky areas througout the region (Kirkpatrick et al. 1995) and on scoria cones (Barlow and Ross 2001). 

136

Willis (1964) considered that the flora of the Volcanic Plains to have the fewest species of higher plants of all the major 

vegetation provinces in Victoria (543 spp.), similarly for mosses (c. 85 spp.), lichens and fungi, but probably not for terrestrial 

and aquatic algae. This “floristic deficiency” (Willis 1964 p. 397) was not, in his opinion, due to degradation. McDougall (1987) 

calculated that the Melbourne area T. triandra grasslands had a total of 183 indigenous vascular species. Of the 108 taxa detected 

by Stuwe and Parsons (1977) in 59 remnants across the plains, 39 were exotics of which 52% were annuals, whereas annuals 

comprised only 10% of the native species and 76% of the annuals were alien species, while of the 151 species at Evans St., 

Sunbury, grassland 101 were native and 50 exotic. 

The Keilor Plains (Little, Werribee and Maribyrnong River catchments, along with some creeks that enter the lower Yarra River) 

at the eastern end of the Volcanic Plains is a rainshadow area with grasslands containing a larger component of ‘dry country’ 

species, typically found in the Northern Plains (Sutton 1916-1917, Lunt et al. 1998). The Keilor Plains and the plains between 

Geelong and Cressy “are and always have been open, dry tussock grassland, without any arboreal growth or even tall shrubs” 

(Willis 1964 p. 398 – except the Banksia (see above)) and with floristic and numerical dominance of Poaceae and Asteraceae, 

which together comprise nearly one quarter of the vascular plant species (Willis 1964). At least 40 species of native grass and 

about 150 other native plants are found on the Keilor Plains (Scarlett et al. 1992). These grasslands are also characterised by the 

absence of perennial plants that are obligate seed regenerators (Lunt and Morgan 2002). They are notable for the large areas of 

‘bare’ ground between the plants (‘intertussock spaces’) (Willis 1964),75-80% of the areas at Derrimut treated by by annual or 

biennial burning or grazing and 85-97% at Laverton North (Henderson 1999), but this space disappears as the grassland ages and 

the biomass of the dominant grass increases, leading to loss of the intertussock species, through shading, increased seed 

predation, etc. (Lunt and Morgan 2002). Frequent destruction or natural sparseness of the grass canopy is needed to maintain 

floristic diversity (Lunt and Morgan 2002). 

Stuwe and Parsons (1977) sampled 59 remnants in 1976 with wide variation in climate, substrate and position in the landscape, 

and found remarkable uniformity in floristic composition, which was most affected by the current management regime. Railway 

sites, burnt annually, usually in late spring and often with slow, patchy fires had significantly greater mean species richness, with 

abundant intertussock space. Roadsides, once commonly grazed, but then “virtually unmanaged” (Stuwe and Parsons 1977 p. 

473) generally were depauperate, with dense, litter-rich T. triandra stands and no bare ground. Pastures grazed by sheep, cattle 

and horses contained open, species poor vegetation. Morgan (1998c) examined five grasslands on the volcanic plains and found 

that vascular plant richness varied more between sites than within them, i.e. sites were internally homogeneous. 

Most of the flora commences growing from dormant buds after soaking rains in late autumn or winter, blooms in spring (October 

and November), sheds seeds in summer (December and January) and dies back to dormant buds with very limited growth in the 

driest period of the year (December to April) (Willis 1964, Lunt and Morgan 2002). 

By the mid 1970s only remnants remained, the richest and least weed-infested being along railway lines, managed by annual 

burning, with roadsides, largely unmanaged or burnt annually or less frequently by local fire brigades, and native pastures having 

lower plant diversity, the former dense and the latter relatively open (Stuwe and Parsons1977, Morgan 1998c). By the 1990s 

most intact T. triandra grasslands on the Victorian volcanic plains were no longer grazed or were subjected to minimal vertebrate 

grazing (Morgan 1998c 1998d) but the largest remnants had been degraded by long-term livestock grazing, generally contained 

few rare species, had low native plant diversity and in some instances a weed dominated soil seed bank (Lunt 1990b 1990c). 

Over 95% of all native vegetation on the Victorian Volcanic Plain has been cleared (Sattler and Creighton 2002). According to 

Sattler and Creighton (2002) 78 Ecological Vegetation Classes (EVCs) have been mapped in the region, 15% of which are 

probably extinct and 78% threatened. Barlow and Ross stated that 48 EVC occurred on the Volcanic Plain, as determined by 

direct observation or modelling. According to Dahlhaus et al. (2003 citing Ross et al. 2002) 115 EVCs, including mosaics and 

complexes occur in the region. Plains Grassland and Grassy Woodlands originally occupied three quarters of the region and only 

about 1% remains, much of it degraded (Sattler and Creighton 2002). Nature conservation reserves occupy approximately 1.3% 

of the region and contain approximately 40% of the EVCs, and biodiversity conservation is heavily reliant on private land, road 

and rail reserves and other public land (Sattler and Creighton 2002). 

Barlow and Ross (2001) cautioned that the several attempts to determine the area of grasslands currently existing in the region 

were beset with methodological problems, including the difficulty of identification where vegetation was highly altered. Using 

DNRE (1997) data, they concluded that grassland complexes had been reduced to an area of 1671 ha by 1997, an estimated 0.2% 

of the pre-1750 extent. But a more complete estimate was suggested to be 5000-6000 ha, or 0.6-0.7%. In the early 1990s 

possibly

on basalt duplex soils had the highest probablity of destruction but also supported the largest number of grasslands. These are 

more readily cultivated than stony and uniform soils and are relatively high in nutrients. In terms of the road class of the nearest 

roads, grasslands on local roads had the highest probability of destruction, followed by patches on highways, possibly because 

local roads are more likely to be grazed, illegally cropped or sprayed with herbicides. In terms of land tenure, grasslands on 

Crown land were most likely to have been destroyed (mostly by weed invasion), those on private land were less likely to have 

been destroyed than those on roadsides, and no railway line grasslands were lost. However the number of private grasslands in 

1984 was severely underestimated, and it is likely that in reality large numbers were probably destroyed during the period. 

Country Fire Authority brigades also had a major impact on destruction: those brigades area where management of grasslands 

had changed from burning to managment using herbicides experienced the highest levels of destruction. 

McDougall (1987) argued that management requirements for remnants needed to be site specific, depending on the particular 

weed problems and specific conservation requirements, and that much remained to be learned. He acknowledged that the effects 

of fire on the inter-tussock herbs was then largely unknown, but recommended burning at 3-5 year intervals between early 

October and mid-December to reduce weeds and prevent overdominance by T. triandra, except at sites with spring-flowering 

rare or endangered species, for which late summer or autumn fires were appropriate. The main management issues for these 

grasslands currently include inadequate knowledge of threatened taxa, management of introduced grasses, prevention of new 

weed incursions and generalised ecosystem degradation (Groves and Whalley 2002). 

Loss of unreserved remnants continues through development for housing and agriculture. Degradation of ‘protected’ remnants 

also continues, including decline in quality, loss of species etc., as a result of invasive species, pollution at the local and global 

levels (e.g. N and CO 2 enrichment of the atmosphere), alteration of hydrological processes, other changed ecological processes, 

and inappropriate or inadequate management. 


South Australia once had immense areas of temperate native grassland and grassy woodland in what is now the agricultural zone 

(Davies 1997). The grassy vegetation in that State occurred mainly on plateaus and in broad valleys on lower slopes from the 

Orroroo and Peterborough areas in the southern and western Flinders Ranges through the Mount Lofty Ranges to Peterborough 

and Murray Bridge. Smaller areas occurred on the sub-coastal plain of the South East between Bordertown and the Victorian 

border, on the southern Eyre Peninsula, the Yorke Peninsula, and on basalt soils near Mount Gambier (Lunt et al. 1998). 

However their composition and previous distribution is unclear (Davies 1997). As in Victoria, recognition that grasslands existed 

in some areas has been slow in coming. Lange (1983) failed to acknowledge the existence of grasslands in the South East. The 

grasses there, he noted, “do not form savannah; usually they grow sparsely” (p. 99). In the mid 1990s half of the significant 

South Australian remnants were on private land, one third on roadsides and one sixth on Crown land (Kirkpatrick et al. 1995). 

Exotic plants have been reported to comprise 30% of the South Australian temperate grassland vascular flora (Kirkpatrick et al. 

1995). 

Moore (1993) mapped these grasslands as Temperate Shortgrass (Austrodanthonia-Austrostipa-Enneapogon) formations, 

extending from south of Adelaide through the Mount Lofty Ranges to north of Port Augusta, and in parts of the South East. 

Two extant temperate grassland communities are under threat or inadequately conserved: Lomandra effusa/Lomandra multiflora 

subsp. dura (open) tussock grassland (co-dominant with Austrodanthonia and Austrostipa spp.), known as Mat-rush Grassland 

(Scented Mat-rush and Stiff Mat-rush Lomandra dura according to Lunt et al. 1998), or Irongrass communities (Carter et al. 

1993) and Austrodanthonia/Themeda tussock grassland. The former occurs on similar soil types and under similar ecological 

conditions to other temperate grassland communities in Australia, so despite the dominants not being grasses, the formations are 

considered in ecophysiological terms to be temperate grasslands (Carter et al. 1993). It is most common on skeletal soils in the 

eastern Mount Lofty Ranges and the Mid North (Kirkpatrick et al. 1995). Both communities were mainly found in the Mid 

North, in the Clare-Port Pirie-Peterborough region but have largely been destroyed by cropping and grazing. Grass dominated 

communities may have occurred near Adelaide and more widely in the lower South East, but are extremely poorly conserved 

(Lunt et al. 1998, Davies 1997). 

Areas of T. triandra grassland are “suspected to have existed” in the lower South East, “primarily on basalt soils (e.g. around 

Mount Schanck and Mount Gambier)”(Foulkes and Heard 2003 p. 48). These areas were described by early settlers as lighted 

wooded with Acacia melanxoylon R.Br, and eucalypts. 

A number of other grassland community types are present in South Australia, including Austrostipa grasslands and Murray 

Lakes grasslands (Davies 1997). The Austrodanthonia caespitosa – Enchylaena tomentosa Marsh Margin Grassland occurred on 

the margins of the Murray lakes, and “only survives in small roadside fragments” (Kirkpatrick et al. 1995). A Themeda- 

Geranium retrorsum-Arthropodium strictum community occurs in the Mount Lofty Ranges in the Mid North (Kirkpatrick et al. 

1995). Mathison (2004) stated that South Australian native grasslands are not climax communities but transitional states and can 

radically change their nature when there is a change in the management regime. Stafford (1991) provided some descriptive 

information on secondary grassland (transitional to woodland) developed in the East Torrens district after cessation of cropping 

and grazing. This probably constituted ‘enriched grassland’ in the sense of McIntyre and Lavorel (2007). 

Approximately half of the 20 most abundant vascular plant species in South Australian grasslands are exotic and annual grasses 

including Avena spp. Bromus spp. and Vulpia spp. are amongst the most invasive (Lenz et al. 2003) 

Tasmania 

As in other areas of southern Australia, the natural grasslands of Tasmania were rapidly occupied by European settlers (Benson 

and Redpath 1997), however they are better reserved than mainland grasslands (Kirkpatrick et al. 1995). Tasmanian non-alpine 

grasslands occur mostly on valley bottoms, on heavy clays or alluvial soils and lower slopes on shallow rocky soils, mostly in 

the drier Midlands” (Lunt et al. 1998). Moore (1993) mapped a narrow band of Temperate Shortgrass (Austrodanthonia- 

138

Austrostipa-Enneapogon) from around New Norfolk to south of Launceston. Frequent frosts appear to be the main cause of 

treelessness (Lunt et al. 1998). Secondary grasslands resulting from tree clearance are present in near-coastal areas where 

frequent burning prevent encroachment by Melaleuca ericifolia (Lunt et al. 1998). Poa labillardieri grasslands on Cape Barren 

and Flinders Islands were probably formed by repeated burning of forest (Kirkpatrick et al. 1995). Austrodanthonia, Austrostipa, 

Poa, Microlaena stipoides and T. triandra are the dominant grasses, the latter probably the major dominant at the time of 

European settlement (Lunt et al. 1998). Almost all of Tasmania’s remnant are grazed by livestock (Kirkpatrick 2007). Exotic 

plants have been reported to comprise 23% of the grassland vascular flora (Kirkpatrick et al. 1995). 

Several plant communities comprise the Tasmanian grasslands (Kirkpatrick et al. 1995): 

1. Austrodanthonia – Astroloma humifusum East Coast Tasmanian Grassland, a derived formation dominated by sparse A. 

caespitosa (Gaudich) H.P. Linder, A. pilosa (R.Br.) H.P. Linder and A. racemosa (R.Br.) H.P. Linder with other grasses, and 

intertussock spaces occupied by herbs and the epacrid A. humifusum (Cav.) R.Br. 

2. Poa labillardieri – Lomandra longifolia – Acaena novae-zelandiae Tasmanian Valley Grassland, mainly in the Midlands, 

dominated by P. labillardieri and L. longifolia Labill. (Xanthorrhoeaceae), with various components including Juncus spp. and 

A. novae-zelandiae Kirk (Rosaceae). 

3. P. labillardieri – Microlaena stipoides – Solenogyne dominii Tasmanian Valley Grassland, characterised by sparse Poa 

tussocks interspersed with grasses and herbs including the Flat Daisy, S. dominii L.G. Adams. 

4. P. labillardieri – Juncus – Hypericum japonicum Tasmanian Flood Plain Grassland, with large Poa tussocks, found on alluvial 

flats. 

5. Poa rodwayi – Astroloma humifusum Tasmanian Rock Outcrop Grassland, found on shallow sandy clay loams around dolerite 

outcrops. 

6. Themeda – Hibbertia hirsuta – Astroloma humifusum Tasmanian Grassland, widespread in the Midlands and on the east coast, 

much of it probably derived from grassy woodland. 

7. Themeda – Stipa stuposa – Chrysocephalum apiculatum Tasmanian Grassland, on fertile, well-drained sites in the Midlands. 

8. Themeda – Solenogyne gunnii – Microlaena stipoides Tasmanian Grassland, also in the Midlands and on the east coast, many 

remnants probably derived from woodlands. 

New South Wales and Australian Capital Territory 

For convenience, discussion of the NSW and ACT native grasslands is subdivided into three geographical regions. This ignores 

the occurrence of Temperate Montane Grasslands in outliers around Braidwood, Goulburn, Bathurst and the Albury area, and the 

whole Austrostipa aristiglumis-dominated Western Slopes Grasslands on the North West and Central West Slopes and Plains 

(Keith 2004). The previous extent of ‘natural’ grasslands in New South Wales remains controversial. There are large areas of 

derived grassland. Based on a survey of 126 selected landholders Garden et al. (2000) estimated that nearly 1.4 m ha (40%) of 

Central, Southern and Monaro Tablelands pastures (excluding the ACT) contained significant amounts of native perennial 

grasses, representing 48%, 31% and 40% respectively of the area of these regions. It was assumed that “cleared and undisturbed 

areas were essentially dominated by native perennial grasses” and that timbered areas and areas disturbed by cultivation or 

hebicides to sow pastures or crops were not (Garden et al. 2000 p. 1085). The figures probably significantly underestimate the 

true extent of native grass dominance, since many disturbed area have been recolonised and some sowing of exotic pastures has 

occurred amongst native grasses (Garden et al. 2000). Groves et al. (1973) considered that except for frost hollow areas the 

Themeda/Poa grasslands of south-eastern NSW were derived from woodland, and maintained by sheep grazing and regular 

burning. 

Compared to other States, large areas of native grassland in New South Wales exist in Travelling Stock Routes and Reserves, 

linear reserves up to 400 m wide, that are leased by Rural Land Protection Boards for grazing (Kirkpatrick et al. 1995). 

Southern Tablelands (NSW and ACT) 

The Murrumbateman (South Eastern Highlands 6) bioregion covers the Australian Capital Territory and surrounding areas in 

NSW and includes the Monaro Tablelands. Grassy woodlands are the major native vegetation type in the region but grasslands 

occurred throughout, especially on clay substrates in lower rainfall areas, in frost hollows and land with poor drainage (Lunt et 

al. 1998). The soils are typically fertile clays and are derived from basalt, limestone or other sedimentary rocks (Keith 2004). 

The grasslands in the region are included in Benson’s (2004) and Keith’s (2004) Temperate Montane Grasslands, which include 

the grasslands of the ACT, Southern Tablelands and Monaro Plains. According to Keith (2004) they occur at 550-1500 m 

altitudes in areas with 500-750 mm annual rainfall and are closely related to Tableland Clay Grassy Woodlands, widespread on 

both the Northern and Southern Tablelands of NSW. On clay soils they are dominated by T. triandra and Poa spp., on lighter 

soils and drier upper slopes by Austrostipa, Austrodanthonia and Bothriochloa and along drainage lines by Poa labillardieri. The 

inter-tussock spaces are occupied by erect, scrambling and rosette herbs, and geophytes including orchids and lilies (Keith 2004). 

Benson (2004) and to a lesser extent Keith (2004) include alpine grasslands dominated by Poa spp. in this class, although Keith 

acknowledges a gradation into Poa costiniana Vickery alpine herbfield/grassland. Temperate Themeda/ Austrodanthonia/ 

Austrostipa grassland was one of the dominant ecosytems in the region (Lunt et al. 1998, Sattler and Creighton 2002), with pre- 

European extent in the ACT of about 14,000 ha, or 20,000 ha if sparsely treed areas (

(Chan 1980). The region also contains large areas of native pasture and secondary grasslands (Sharp and Shorthouse 1996, ACT 

Government 2005). 

Despite the relatively high altitude of these tablelands grasslands, their vascular plant composition is more similar to the 

grasslands of the Victorian Volcanic Plains than to alpine grasslands (Lunt et al. 1998). T. triandra, Austrodanthonia auriculata, 

A. caespitosa, A. carphoides, A. laevis, Austrostipa bigeniculata, A. scabra, Bothrichloa macra, Poa labillardieri and P. 

sieberiana are generally the dominant native grasses (Sharp 1997). B. macra was dominant in some areas in the late 1970s 

(Groves 1979). Areas of the Monaro tablelands not dominated by Poa sieberiana were probably dominated by T. triandra, which 

was largely replaced under bovid grazing by less palatable Austrostipa spp. from the mid 1840s (Kirkpatrick et al. 1995). Similar 

to other areas, pasture improvement intensified dramatically from the 1940s, resulting in widespread losses of native grasslands 

or major modification of their components and structure, including massive incursions of weeds (Keith 2004). Exotic plants have 

been reported to comprise 29% of the Monaro grassland vascular flora (Kirkpatrick et al. 1995). Unlike the Victorian Volcanic 

Plains, rocks are not a common feature (Melbourne 1993, Kukolic 1994). 

Kirkpatrick et al. (1995) briefly described the following lowland grassland communities in the region: 1. Poa labillardieri – 

Austrostipa – Bothrichloa macra Monaro Basalt Grassland, a widespread, dense-tussock formation on clays in the ACT and 

NSW; 2. Poa sieberiana – Carex appressa – Juncus Monaro Grassland, probably best thought of a highland community; 3. 

Austrostipa scabra – Enneapogon nigricans Monaro Grassland, a mid-dense tussock grassland found mainly on upper slopes and 

ridges; 4. Themeda triandra – Poa sieberiana – Bulbine bulbosa Monaro Grassland, found on valley floors or slopes with a 

southerly aspect; 5. Themeda triandra – Eryngium ovinum – Carex inversa Canberra Grassland, largely confined to the ACT in 

valleys on limestone; 

Sharp (1997) provided an analysis of the floristic associations present in the ACT (Table 14) and previous floristic 

classifications. The Austrodanthonia association occurred on sites subject to historical levels of moderate to high disturbance on 

well drained clay soils with low nutrient levels, and had the largest areas of bare ground. A wet Themeda association occurred in 

areas with moderate to high historical disturbance regimes, on poorly drained (seasonally wet) sites with higher soil P and acidity 

levels. A dry Themeda assocation occurred on well drained sites with low levels of disturbance, and had high litter cover of c. 

40% and low soil P (7.9 ppm). About 70% of vascular plant species present were forbs, with Asteraceae an important component 

(Sharp and Shorthouse 1996). No community had consistently higher exotic richness, but the wet Themeda association had the 

highest exotic cover in spring, and the dry Themeda association the lowest (Table 14). 

Table 14. Floristic characteristics of natural grassland floristic associations in the ACT from a survey of 39 sites (Sharp 1997). 

Association Mean spp. richness (per 10 m 2 ) Mean spp. richness (sites) % exotic cover 

Native Exotic Total Native Exotic Total (spring) 

Austrodanthonia 24.7 12.4 37.1 33.2 21.7 54.9 32.8 

Wet Themeda 25.3 13.7 39.0 36.3 23.2 59.5 35.5 

Dry Themeda 21.8 8.6 30.4 41.8 23.6 65.4 11.1 

Remnant grassland on the Southern Tablelands is highly fragmented and adequate reservation was considered unlikely to be 

achievable (Sattler and Creighton 2002). Remaining remnants are in various states of degradation and exist in a matrix of exotic 

pasture (Keith 2004). Approximately 5% of the original 20,000 ha of grasslands in the ACT were in more of less natural 

condition (florisitically and structurally intact, with low weed cover) in 1996, with an additional 550 ha of low quality grassland 

(Sharp and Shorthouse 1996, Sharp 1997). Sharp and Shorthouse (1996) provided a map of the pre-European distribution and 

remaining remnants and estimated that c. 70% of the grasslands identified in the late 1970s had subsequently been destroyed by 

urban development, conversion to pasture or invasion by exotic plants. Less than 3% of the pre-European area remained in 

reasonable condition by 2005, reduced from 11% of the region pre-1750 to about 1% in 2000 (ACT Government 2005). Most 

had been destroyed by intensive agriculture and urban growth, and existing remnants were threatened by inappropriate 

management. Remaining sites were mostly

Woodlands and New England Grassy Woodlands (Benson 2004), by tree clearing and livestock grazing (Trémont 1994). 

Frequent aboriginal burning is believed to have opened up the woodlands and maintained grassy vegetation (McIntyre and 

Lavorel 1994b). The map of Moore (1993) included most of the Northern Tablelands as aTemperate Shortgrass 

(Austrodanthonia-Austrostipa-Enneapogon) community. The grasslands occur on soils derived from basalts, granite or 

sedimentary rocks, at altitudes from

pyramidata (Benth.) P.G. Wilson and M. sedifolia (F. Muell.) P.G. Wilson that were originally dominant over most of the area, 

and resulted in their replacement by low-growing, spiny, unpalatable Sclerolaena spp., or by “degraded grassland vegetation ... 

of annual, and to a lesser extent perennial, grasses and herbs”. Kirkpatrick et al. (1995) thought that the original vegetation 

before livestock grazing might best be characterised as chenopod shrubland with a grassy ground layer, and suggested that 

drought, fire or rabbit grazing may have destroyed the original shrubland, rather than livestock grazing which merely prevented 

regeneration of woody species. According to Moore et al. (1973 p. 236) no stands of the original vegetation then existed, but 

observations suggested that the trees had been up to 9 m high, the shrub layer was well-developed but discontinuous, that 

Atriplex predominated on grey and brown clay soils and the Acacia was dominant on red-brown earths. The other main shrubs 

were the chenopods Rhagodia spinescens R.Br., Enchylaena tomentosa R.Br. and M. aphylla. The grassy ground layer varied in 

composition according to soil texture. Chloris truncata and Austrostipa “variabilis” (= A. scabra/A.nodosa) were then common 

on light-textured soils. Austrodanthonia caespitosa and C. truncata were the main species on clays, although Austrostipa 

artistiglumis was probably once more common or dominant. Leigh and Mulham (1965) considered the latter species to be found 

on most soil types and to often occur in dense, localised stands. Other common and widespread native grasses included Chloris 

ramosus B.K. Simon, Eragrostis spp., Eriochloa pseudoarcrotricha (Stapf ex Thell.) J.M. Black, Panicum spp., Sporobolus 

caroli Mez, (Leigh and Mulham 1965). Other common native species included Bulbine bulbosa (R.Br.) Haw., Lomandra effusa 

(Lindl.) Ewart, Hypoxis glabella R.Br., other Atriplex spp., Chenopodium spp., Disphyma crassifolium (L.) L.Bolus, Swainsona 

spp., Sida spp., Haloragis spp., Plantago varia R.Br., Asperula conferta Hook. f., Wahlehbergia spp., herbaceous Goodenia spp. 

and a large suite of small daisies (Leigh and Mulham 1965). 

Leigh and Mulham (1965) provided an illustrated compendium of the important pastoral plants of the Riverina grasslands. 

Grassland fauna 

The fauna of temperate Australian native grasslands is very poorly known. Much of the vertebrate fauna, particularly the 

mammals, was eliminated during the early years of European occupation, and the invertebrate fauna has received limited 

attention, apart from a few agricultural pests and iconic native species, some exploratory inventory studies in Victoria and the 

ACT, and several detailed studies in the ACT. Australian grasslands have been said to have relatively few specialist animal 

species and “a modest but distinctive array of animals that feed on on the foliage or the seeds of grass” (Keith 2004 p. 104). 

Grasslands are characteristed by high rates of herbivory compared to many other terrestrial ecosystems, with consumption 

efficiencies generally c. 25% compared to c. 5% in forests (Tscharntke and Greiler 1995). High root: shoot ratios in temperate 

grasslands means there is a large subterranean plant biomass to support soil fauna, and the subterranean standing crop consumed 

by insects is 2-10 times higher than the above-ground crop (Tscharntke and Greiler 1995). Thus an underground life stage is 

typical in grassland insect genera (McQuillan 1999). 

The seral stage of T. triandra grassland affects its suitability for animals. Biomass of the grass determines the floral composition 

of the grassland, presence of food plants, amount of shade and the structure of habitat. Watson (1995 cited by Lunt and Morgan 

2002) “found substantial seed predation beneath a dense grass cover, but little in burnt open areas”. 

In this review, the vertebrates are first examined, with a particular emphasis on the extinct mammalian fauna, then invertebrates, 

with a particular emphasis on insects, including an examination of several rare and threatened taxa, and a detailed review of 

what is known about the fauna of grasses. 

Vertebrates 

Many of the mammals and birds that forage in grasslands require structural habitat features for shelter, nesting, etc. that 

grasslands do not provide (Keith 2004), this being the case for example with insectivorous bats. But today, and in geologically 

recent times, the often complex vegetation mosaics around or within south-eastern Australian grasslands have meant that 

vertebrates dependent on woodland and shrubland for shelter etc. have been widely able to use grasslands to meet some of their 

requirements. 

Vertebrate species can be detrimentally or beneficially impacted by the floristic and structural vegetation changes caused by 

weeds, including alterations to food supply and foraging potential, nest sites, cover, predator protection, etc. (Brown et al. 

1991). Conversely, the activities of vertebrates may hinder or facilitate weed populations by bioturbation, defecation and 

urination, feeding and other activities. 

Mammals 

Grassland mammal assemblages are characterised by the lack of arboreal species and of browsers, and restricted diversity of 

grazers (Webb 1977). Australian grasslands were once occupied by a marsupial megafauna, which disappeared, like similar 

faunas in North and South America, during the Miocene and the end of the Pleistocene (Webb 1978). The major extinction phase 

in Australia occurred in the late Pleistocene, possibly mostly c. 26 kbp and continuing to c. 18-15 kbp, and has been associated 

with general continental drying (Kershaw et al. 2000). The marsupial megafauna occurred in 250-750 mm rainfall zones, and one 

of three groups was specifically adapted to southern grasslands (Kershaw et al. 2000). Aboriginal Australians may have played 

little part in their disappearance, since fossil evidence suggests coexistence for at least 20,000 years, although recent redating 

appears to have markedly reduced this overlap period (Kershaw et al. 2000) and consensus now favors hunting as the major 

cause of extinctions (Flannery 1994, Johnson 2009). Subsequent to the extinction of the megafauna, another set of medium sized 

animals largely disappeared from south-eastern Australian grasslands, mainly during early historical times (Table 15). 

According to Lunt et al. (1998), 8 of 26 mammal species (excluding bats), mostly rodents and small to medium sized marsupials, 

once found in the grassy plains of south-eastern Australia are now no longer present. In the South East of South Australia the 

mammals that have disappeared since European occupation have been those that “lived mainly in woodlands and grasslands” 

(Aitken 1983 p. 127). Numerous species have been lost in Victoria including Potoroidae (Bettongs Bettongia spp., Rufous 

Bettong Aepyprymnus rufescens Gray), Macropodidae (e.g. Eastern Hare-wallaby Lagorchestes leporides (Gould), Bridled 

Nailtail Wallaby Onychogalea fraenata (Gould), Toolache Wallaby Macropus greyi Waterhouse, Tasmanian Pademelon 

142

Thylogale billardierii), Peramelidae (bandicoots) and Muridae (Rabbit Eared Tree Rat Conilurus albipes (Lichtenstein)) 

(Wakefield1964b, Menkhorst 1995a). Many grassland mammals have also become exinct in New South Wales, including L. 

leporides, Tasmanian Bettong Bettongia gaimardi Desmarest, Western Barred Bandicoot Perameles bougainville (Quoy and 

Gaimard), C. albipes and Plains Mouse Pseudomys australis Gray (Muridae) (Keith 2004). 

Bats are relatively common crespusclar and nocturnal foragers over native grasslands. The insectivorous Verpertilionidae are the 

dominant group in south-eastern Australia but their persistence is dependent on the availability of roosting and maternity sites in 

tree hollows, to which they must return on a daily basis, and the ongoing loss of old hollow-bearing trees is a threat to their 

continued existence (Mansergh et al. 2006b). 

Table 15. Grassland mammals that have disappeared from south-eastern Australian grasslands since European settlement. E = 

globally extinct, E(MSE) = extinct in mainland south-eastern Australia, En(Vo) = endangered in Victoria, only present in 

Victoria. ‘Mainland south-eastern Australia’ excludes Queensland and the Northern Rivers district of NSW. N.B. Grasslands 

were not a habitat of Leporillus apicalis, Dasyurus viverrinus and Phascogale calura according to Menkhorst (1995a). 

Species Common Name Family Current Status References 

Dasyurus viverrinus (Shaw) Eastern Quoll Dasyuridae E(MSE) Menkhorst 1995a, Lunt et al. 1998 

Phascogale calura Gould Red-tailed Phascogale Dasyuridae E(MSE) Menkhorst 1995a, Lunt et al. 1998 

Chareropus ecaudatus (Ogilby) Pig-footed Bandicoot Peramelidae E Menkhorst 1995a, Lunt et al. 1998 

Perameles bougainville (Quoy Western Barred Bandicoot Peramelidae E(MSE) Menkhorst 1995a, Lunt et al. 1998 

and Gaimard) 

Perameles gunnii Gray Eastern Barred Bandicoot Peramelidae En(Vo) Menkhorst 1995a, Backhouse and 

Crossthwaite 2003 

Aepyprymnus rufescens (Gray) Rufous Bettong Potoroidae E(MSE) Menkhorst 1995a, Lunt et al. 1998 

Bettongia gaimardi (Desmarest) Tasmanian Bettong Potoroidae E(MSE) Menkhorst 1995a, Lunt et al. 1998 

Bettongia lesueur (Quoy and Burrowing Bettong Potoroidae E(MSE) Noble 1993 

Gaimard) 

Bettongia penicillata Gray Brush-tailed Bettong Potoroidae E(MSE) Menkhorst 1995a, Lunt et al. 1998 

Lagorchestes leporides (Gould) Eastern Hare Wallaby Macropodidae E Wakefield1964b, Menkhorst 1995a 

Macropus greyi Waterhouse Toolache Wallaby Macropodidae E Menkhorst 1995a 

Onychogalea fraenata (Gould) Bridled Nailtail Wallaby Macropodidae E(MSE) Menkhorst 1995a, Lunt et al. 1998 

Thylogale billardierii 

Tasmanian Pademelon Macropodidae E(MSE) Williams 1995 

(Desmarest) 

Conilurus albipes (Lichtenstein) White-footed Rabbit-rat Muridae E Menkhorst 1995a, Lunt et al. 1998 

Leporillus apicalis (Gould) Lesser Stick-nest Rat Muridae E Menkhorst 1995a, Lunt et al. 1998 

Pseudomys sp. rat Muridae E Lunt et al. 1998 

Pseudomys australis Gray Plains Rat Muridae E(MSE) Menkhorst 1995a, Lunt et al. 1998, 

Mansergh andSeebeck 2003 

Information on the orginal habitat preferences and populations densities of the species that have dramatically declined are 

difficult to obtain (Noble et al. 2007). Opinions differ on whether some were grassland inhabitants. Seebeck and Mansergh (2003 

p. 2) considered the extinct Lagorchestes leporides to be an arid zone species, “limited to the central and southern sections of the 

Murray-Darling basin”, and the former range of the extinct C. albipes to be unclear. Hope (1994) argued that 30-12 kybp cave 

sediment fossils from south-western Victoria were of desert animals. But Wakefield (1964b) reported subfossil remains of L. 

leporides at Mt Hamilton and of C. albipes from Mt Eccles, Byaduk Caves and Mt Hamilton, while Seebeck and Mansergh 

(2003) mentioned historical records of C. albipes from the Port Phillip region and the Portland area. L. leporides was known in 

the South East of South Australia where it was last recorded near Naracoorte in 1870, and C. albipes was reported from the 

South East only before 1843 (Aitken 1983). Frith (1973 p. 76) considered both Lagorchestes and Onychogalea to have been 

“abundant in more open woodlands and savannahs”. Brown (1987) considered that three species of bandicoots extinct in Victoria 

were restricted to the semi-arid north-west of the State. Wakefield (1963a p. 328) stated that there is “evidence that C. albipes 

survived [in Victoria] until well after European settlement”. Williams (1995) stated that Thylogale billardierii was “apparently 

common” in coastal forest and scrub in southern Victoria before 1900, but recorded a specimen from Werribee in 1881. 

Pseuodmys australis, extinct in Victoria and NSW but still present in central Australia, is considered to have once inhabitated the 

Western Plains Grasslands of Victoria, “where it constructed large, shallow, complex burrow systems”, but there is no reliable 

(i.e. specimen based) evidence that it occurred in the State at the time of European occupation (Mansergh and Seebeck 2003). 

Wakefield (1964b) reported subfossil remains of the Bettongia gaimardi from Mt Hamilton and the Burrowing Bettong B. 

lesueur (Quoy and Gaimard) from Mt Hamilton and Bushfield (8 km north of Warrnambool). The former was probably present 

in the South East of South Australia during the period of early settlement (Aitken 1983) and probably became extinct on mailand 

Australia about 1900, but is still extant in Tasmania (Mansergh and Seebeck 2003). The latter, at the time of European 

settlement, was found across all regions of native grassland in western and northern Victoria, the NSW Riverina and South 

Australia, but not the NSW Southern Tablelands (Aitken 1983, Seebeck and Rose 1989, Noble 1993, Noble et al. 2007), and had 

the widest distribution of all native mammals (Noble et al. 2007), but became extinct on the mainland. Population densities of 

14-35 km-2 have been estimated in arid and semiarid areas (Noble et al. 2007). The Brushtailed Bettong B. penicillata was once 

common across southern Australia (Garkaklis et al. 2003), although Seebeck and Rose (1989) indicated a former range inland of 

current temperate grasslands, except in South Australia. The Rufous Bettong Aepyprymnus rufescens prefers “open grassy 

woodland and forest”, but in northern NSW the vegetation consisted of “only tall native grasses” and individuals have been 

observed feeding in pasture (Seebeck and Rose 1989 p. 726). It’s decline on the northern tablelands of NSW was partly the result 

143

of destruction as vermin and partly from extinction of its tussock grassland habitat (Seebeck and Rose 1989). Bettongs were 

considered to be agricultural pest species during early European settlement (Seebeck and Rose 1989, Noble et al. 2007). 

Wakefield (1964a p. 277) listed other species found as subfossils in caves near Byaduk and Mt Eccles that are not found in the 

Western Plains Grasslands today but were “Presumably ... present on the basalts of south-western Victoria at the time of 

European occupation ...”. These included the extinct Pseudomys auritus which was present “in the recent past ... eastward, 

apparently in abundance, to central Victoria” (Wakefield 1964a p. 278) and was “once widespread in western Victoria” 

(Wakefield 1963b p. 44). 

Perhaps at a more distant time the Western Basalt Plains were also inhabited by the Tasmanian Devil Sarcophilus harrisii 

(Boitard) and the Thylacine Thylacinus cyanocephalus (Harris). Wakefield (1963a) reported sub-fossil remains of both these 

large predators, extinct on mainland Australia, from lava caves at Mount Hamilton. Remains of of S. harrisii appeared to be 

“quite modern” with one mandible having “pieces of dried tissue adhering to the bone”, while the few remains of T. 

cyanocephalus appeared to be “not ... very old” (Wakefield 1963a p. 324). Remains of the latter species are known also in 

Victoria from Gisborne, while remains of S. harrisii in Victoria have been radiocarbon dated to 550 ±200 years B.P. (Wakefield 

1963a). Dixon (1989 p. 557) noted that T. cyanocephalus was well known from the main area of natural open grasslands in 

Tasmania, the midland plains region, and that most captures of the animal occured in “farmed flat areas”, where it was 

considered a serious predator of lambs. Loss of ‘top predators’ has serious cascade effects in food webs, and probably 

contributed greatly to the collapse of grassland ecosystems. 

Bandicoots are primarily nocturnal and crepuscular, rat to rabbit sized marsupials, which, along with bilbies (Thylacomyiidae) 

have sufferred more negative impact from European occupation than any other marsupial group except Potoroidae (Brown 1987) 

and the thylacine. The extinct Pig-footed Bandicoot Chaeropus ecaudatus (Peramelidae) was a grassland species which required 

“lush growing tips of grasses and herbs”, so livestock grazing is probably the major factor in its extinction (Menkhorst 1995a) 

along with habitat destruction by rabbits and predation by introduced mammals (Seebeck and Mansergh 2003). Seebeck and 

Mansergh (2003), on the other hand, considered it to be an arid zone species, restricted in the historical period in Victoria to the 

Mallee. Sheep grazing is also probably the major factor in disappearance of the Bridled Nailtail Wallaby (Menkhorst 1995a). 

Losses of such a large proportion of the native herbivores likely had a profound effect on plant diversity (Bloomfield and 

McPhee 2006). 

Studies of more arid and northern regions of/Australia, and other areas of the temperate pastoral zone show similar losses of 

medium sized (35 g to 55 kg) ground dwelling marsupials and murid rodents (Burbidge and McKenzie 1989, Short and Smith 

1994, Short and Turner 1994, Noble et al. 2007). Various factors in combination are blamed including hunting, increased 

predation from foxes and cats, rabbit and livestock grazing, habitat clearance and fragmentation, land use change and altered fire 

regimes (Recher and Lim 1990, Short and Smith 1994). Rat kangaroos (Potoroidae) never recovered from hunting under a 

bounty system on the Northern Tablelands of NSW (Johnson and Jarman 1975), although they were present “in plague 

proportions” in the 1880s and 1890s (Cameron 1975 p. 21). Altered fire regimes are possibly the most important: in areas where 

it has been possible to study the losses, in Central Australia, they are temporally correlated with the departure of aboriginal 

people and the cessation of regular aboriginal burning (Short and Turner 1994, Flannery 1994). The absence of regular burning 

resulted in decreased rates of nutrient recycling and the absence of new lush vegetation, creating a nutritional crisis for the more 

specialised feeders (Flannery 1994). A general increase in biomass of vegetation coupled with decreased patchiness meant that a 

fire, when it did arrive, burnt much more widely and intensely than previously, creating large areas with no food, nor shelter 

from predators, and later a uniform, even-aged vegetation (Flannery 1994). However a test of the mosaic burning decline 

hypothesis (Short and Turner 1994) found that mosaic scale had no significant effect on survival of three species in the critical 

weight range, and that declines on mainland Australia and survival on off-shore islands was better explained by the presence or 

absence of exotic predators (foxes and cats) and herbivores (rabbits and livestock). More generally, Burbidge and McKenzie 

(1989) found that the diverse effects of diversion of environmental production to human uses effectively resulted in a general 

trend to environmental aridity, that differentially affected moderately sized mammals with limited mobility, relatively high daily 

metabolic requirements and more specialised feeding strategies. 

Victorian grasslands now support little native mammalian diversity. Victorian mammals adapted to grasslands and grassy 

woodlands are one of Menkhorst’s (1995b) five groups of mammals requiring specific managment to secure their future. They 

include the Fat-tailed Dunnart Sminthopsis crassicaudata (Gould), Eastern Barred Bandicoot Perameles gunnii and Common 

Wombat Vombatus ursinus. The latter occurred throughout the volcanic plains in the 1800s but is now extinct in western 

Victorian grasslands. Another widespread species not threatened on a State or national basis, the Eastern Grey Kangaroo 

Macropus giganteus was once common in the Victorian basalt plains (Sutton 1916-1917, Coutts 1982) but is now rare in the 

region (Lunt et al. 1998) although some large populations exist (e.g. at Woodlands Historic Park). In the South East of South 

Australia it was restricted to four colonies by 1983, one in woodland with associated grassland (Aitken 1983). The Western Grey 

Kangaroo M. fuliginosus (Desmarest) was still common and included grassland in its habitat (Aitken 1983). This species prefers 

areas with heathy understorey but forages in grassland (Bennett 1995). 

In the South East of South Australia S. crassicaudata was considered common in grassland, including pasture, while V. ursinus 

was reduced to remnants, including populations in coastal grassland and sedgeland (Aitken 1983). 

Perameles gunnii, incorrectly said to be endemic to the Victorian Volcanic Plain (DNRE 1997), was formerly widespread and 

abundant across the Plains from near Melbourne north to Beaufort and west to Coleraine in Victoria and into the South East of 

South Australia, but is now critically endangered on mainland Australia (Aitken 1983, Brown 1987, Brown et al. 1991, 

Backhouse and Crossthwaite 2003). The mainland and Tasmanian populations are distinct, undescribed subspecies (Backhouse 

and Crossthwaite 2003). P. gunnii is “specifically adpated to grassland and savannah woodland” (Brown 1987). P. gunnii digs 

small conical burrows in the soil when foraging for its invertebrate prey. Earthworms are important in the diet in wetter months. 

Cockroaches (Blattidae), earwigs (Forficulidae), beetles, both larval and adult especially larval Scarabaeidae, Lepidoptera larvae 

and Romulea rosea bulbs are other common dietary items (Brown 1987). 

144

Its original habitat on the mainland “included” Austrodanthonia, Austrostipa, Poa labillardieri and Themeda triandra grasslands 

and it requires dense ground cover for shelter, and open areas with relatively short grass in which to forage (Brown et al. 1991 p. 

150). Replacement of native grasses by exotic pasture grasses and weeds has reduced protective cover, but other factors 

including habitat destruction, predation by introduced vertebrates and exotic disease appear to be of much greater importance 

than weeds in its decline (Brown et al. 1991, Backhouse and Crossthwaite 2003). One important factor is probably soil 

compaction by bovid livestock (Brown 1987). It was last recorded in South Australia nearr Mt Gambier in 1890 (Aitken1983). 

Several colonies possibly still existed in the Victorian Western District in the late 1940s but it probably survived only in the 

Hamilton and Penshurst districts in the 1960s (Brown 1987). From c. 1971 the last remaining wild population on mainland 

Australia continued to exist in a highly modified environment “almost totallydominated by exotic plant species” at Hamilton 

rubbish tip, being able to escape predation by hiding in old car bodies and thickets of gorse Ulex europaeus L., and foraging in 

surrounding paddocks (Backhouse and Crossthwaite 2003 p. 3). 

Hadden (2002) surveyed 24 remnant grassland sites on the Victorian Western Volcanic Plains and Northern Plains by pitfall 

trapping and systematic and opportunistic searching from January 1995 to February 1996. She evaluated habitat by measuring 

cover of cool- and warm-season perennial grasses, native herbs, exotic grasses, exotic herbs, bare ground, dry litter, total floristic 

composition (native/exotic), sheep grazing pressure and invertebrate richness. Three mammal species were captured in c. 7500 

trap nights. Sminthopsis crassicaudata (Gould) was found at all sites in the Northern Plains and 7 of 12 sites in the Western 

Plains. One inidividual of the Common Dunnart S. murina (Waterhouse) was captured at a Western Plains site, the first Victorian 

grassland record. The introduced House Mouse Mus musculus L. was found at 5 of the Western Plains sites and 8 Northern 

Plains sites. Rabbits O. cuniculus and European Hare Lepus capensis were almost ubiquitous in Northern Plains sites but were 

observed at approximately half the Western Plains sites. No mammals were recorded at five of the Western Plains sites. S. 

crassicaudata was abundant at some floristically rich sites and also in degraded remnants and exotic pastures. It was relatively 

more abundant at sites that were lightly grazed, at sites with more open vegetation and floristically rich sites. It is an active 

hunter, mainly of invertebrates, and may require open areas for foraging, as well as tussock cover as a protection from predators. 

It was found living in rock piles and stone walls, and utilised wolf spider burrows on the Northern Plains. This species has 

disappeared from grasslands near Melbourne where it was once common. The almost ubiquitous M. musculus was relatively 

more abundant on heavily grazed sites, was absent from lightly grazed sites on the Western Plains, was more abundant at densely 

vegetated sites on the Western Plains and lightly vegetated sites on the Northern Plains, and was more common on floristically 

poor sites. Two other species, the introduced Brown Rat Rattus norvegicus (Berkenhout) and native Swamp Rat R. lutreolus 

(Gray) have been recorded in Victorian grasslands in recent historical times. 

Mus musculus is a threat to some grassland plants. Mouse predation of tubers of Diuris fragrantissima is believed to have been 

responsible for mortality of perhaps 70% of plants at its single extant site during the mid-1980s (Webster et al. 2004). 

The Short-beaked Echidna, Tachyglossus aculeatus (Shaw), occurs widely in grasslands (Menkhorst 1995c) but appears to be 

absent from most of the smaller remnants, particularly in areas of closer settlement. 

The highly depauperate mammalian faunas or remnant grasslands means that biodiversity components directly dependent upon 

mammalian activities are also threatened or lost. Obvious dependents include dung beetles (Scarabaeidae: Scarabaeinae and most 

Aphodiinae) that require a dung resource, and a range of ectoparasites (fleas, etc.) and endoparasites (Nematoda, etc.) that 

require their host for survival. The complex consequences of the loss of native grazing species and mammalian plant-predators 

have only recently begun to be explored and very little is known about the ramifications. However absence of soil disturbance by 

mammals has probably had a strong negative impact on regeneration of many native forbs (Reynolds 2005) and the dispersal 

opportunities for plant seeds must have been radically altered. 

Birds 

Few Australian birds are restricted to grasslands and by necessity they are ground-nesting species. In New South Wales, 

populations of many grassland species that forage and nest on the ground have disappeared or shrunk to very low levels (Keith 

2004). In Victoria the most significant regions for threatened bird species include the Northern Plains and the Wimmera Plains, 

and few threatened species occur in the Western Volcanic Plains or the Gippsland Plains (Robinson 1991). Threatened birds in 

Victoria are more likely to be woodland or grassland species that are seed eaters or vertebrate predators, and to nest in tree 

hollows or on the ground (Robinson 1991). A number of species once common in native grasslands have declined markedly or 

disappeared from grasslands as the area of habitat has shrunk, however only two species that are more or less restricted to 

grasslands have endangered status (Table 16). 

Table 16. Endangered bird species of south-eastern Australian temperate native grasslands. U = unlisted,V= vulnerable, E = 

endangered, CE = critically endangered, T = threatened, X = extinct. 

Species Common Name ACT NSW SA Vic References 

Ardeotis australis(J.E. Gray) Australian Bustard X CE Sharp and Shorthouse 1996, Venn and 

Menkhorst 2003, Department of Sustainability 

and Environment 2009a 

Pedionomus torquatus (Gould) Plains Wanderer X V/T Baker-Gabb 2003, Department of Sustainability 


Emus Dromais novaehollandiae (Latham) were once common in the Victorian basalt plains (Sutton 1916-1917, Coutts 1982) 

and the grasslands of NSW (Keith 2004). Marshall (1968 p.76) reported observations by the Rev. James Backhouse in 1837 that 

emus in the Melbourne area were then “fast retiring before the white population and their flocks and herds”. They are now rarely 

reported on the volcanic plains (Lunt et al. 1998) and have disappeared or markedly declined in NSW grasslands (Keith 2004), 

145

their extermination probably having followed very similar lines to those reported by Marshall (1968): shot for food, oil and as 

vermin, but road deaths have probably also caused substantial populations losses. 

Of the threatened birds mentioned by Robinson (1991) the following may be the most restricted to grassland habitats: Australian 

Kestrel Falco cenchroides Vigors and Horsfield, Little Button Quail Turnix velox (Gould), Red-chested Button-quail T. 

pyrrhothorax (Gould), Singing Bushlark Mirafra javanica Horsfield, Rufous Songlark Cincloramphus mathewsi Iredale, Brown 

Songlark C. cruralis (Vigors and Horsefield), and possibly Southern Whiteface Aphelocephala leucopsis (Gould) (Blakers et al. 

1984). Notable subjects of conservation concern are the Plains Wanderer Pedionomus torquatus Gould, Australian Bustard 

Ardeotis australis (J.E. Gray), Bush Thick-knee or Stone Curlew Burhinus grallarius (Latham) and Painted Button-quail Turnix 

varia (Latham) (Lunt et al. 1998), although the Painted Button-quail reportedly lives only in heath, woodland and eucalypt forest 

(Blakers et al. 1984). Loyn and French (1991) discussed the diets of Stubble Quail, a relatively common species also found in 

grasslands, and Plains Wanderer in relation to exotic plant invasions. 

The Bush Thick-knee is mainly a woodland bird (Blakers et al. 1984), usually living in Victoria on farmland remnants of lightlytimbered, 

lowland grassy woodlands, most commonly treed with Eucalyptus microcarpa (Johnson and Baker-Gabb 1994). The 

bird is classified as Vulnerable in Victoria, where it has sufferred widespread regional extinction in the south due to habitat 

clearing and alteration. It is a long-lived, sedentary, nocturnal, cursorial ground-nesting bird, requiring low, sparse groundcover 

(10-70% bare ground, grasses 67%

Table 17. Endangered reptile species in south-eastern Australian native temperate grasslands. U = unlisted,V= vulnerable, E = 

endangered, T = threatened. 

Species Common Name ACT NSW SA Vic References 

Delma impar Fischer Striped Legless Lizard V V/E E V Coulson 1990, Kukolic 1994, Sharp and 

Shorthouse 1996, Eddy et al. 1998,Webster 

et al. 2003, Keith 2004, O’Shea 2005, 

Department of Sustainability and 

Environment 2009a 

Tympanocryptis pinguicolla 

(Mitchell) 

Eulamprus tympanum 

marnieae (Lvnnberg and 

Andersson) 

Aprasia parapulchella 

Kluge 

Grassland Earless Dragon E E E Sharp and Shorthouse 1996, Eddy et al. 

1998, Brereton and Backhouse 2003, Keith 

2004, ACT Government 2005, Department 

of Sustainability and Environment 2009a 

Corangamite Water Skink - - - T Department of Sustainability and 

Environment 2007 

Pink-tailed Worm Lizard, 

Pink-tailed Legless Lizard 

E or 

V 

Suta flagellum Eastern Whip Snake E or 

V 

Varanus rosenbergii Rosenbergs Monitor E or 

V 

E T Sharp and Shorthouse 1996,Eddy et al. 1998, 

Keith 2004, Department of Sustainability 


E or 

V 

E or 

V 

T 

Eddy et al. 1998 

Eddy et al. 1998, Department of 

Sustainability and Environment 2009a 

Delma impar, Striped Legless Lizard (Pygopodidae) 

Delma impar Fischer is one of 38 species in the Pygopodidae, a family of mostly surface active, fossorial or semi-fossorial, 

arthropod-eating lizards, endemic to Australia except for two species in Papua New Guinea, and most closely related to 

diplodacytline Geckonidae (O’Shea 2005). It is the most southerly occurring of the 17 Delma spp., is considered to be a diurnal, 

semi-fossorial species, active from September to late March or April (O’Shea 2005). It is slow growing, and could live for up to 

20 years (O’Shea 2005). According to Hadden (1995) D. impar had been recorded at 125 sites from the South East of South 

Australia, through eastern and northern Victoria to southern NSW and the ACT, of which possibly 45 were then considered still 

capable of supporting it. Additionally O’Shea (2005) mapped recent records from the South East of South Australia. The lizard’s 

rarity was considered to be due to its specificity to lowland grassland habitat and the widespread loss and degradation of such 

habitats. Western (Basalt) Plains Grassland was the major vegetation type occupied in Victoria (Webster et al. 2003). Population 

densities were reportedly highest in dense, relatively undisturbed native grassland, although capture rates were relatively high in 

an area dominated by Nassella trichotoma and total grass cover appeared to be the best predictor of population density (Coulson 

1990). Most of the extant sites (Hadden 1995) were native perennial tussock grasslands, either T. triandra in Victoria or T. 

triandra and Austrostipa bigeniculata in the ACT (Kukolic 1994), but a few were dominated by the exotic perennial Phalaris 

aquatica, while sites then dominated by Nassella spp. were found to no longer harbour the species (Hadden 1995). Kukolic 

(1995) recorded pitfall trap captures in T. triandra grasslands with mixed native and exotic grasses and in areas more dominated 

by exotic grasses at Yarramundi Reach, ACT. Hadden (1995) found that tussock cover >50% was usual at the majority of extant 

sites and tussocks were considered important for shelter and as basking sites. D. impar utilises deep cracks in clay soils for 

shelter, breeding and refuge, and surface rocks may be similarly used, but are not a necessary habitat component (Hadden 1995). 

Fire was considered an unknown risk, while livestock grazing was considered acceptable since it occurred on nearly half the 

known extant sites (Hadden 1995). Kukolic (1994) recommended that fire not be used to manage the habitat, mentioning finds of 

dead individuals at Derrimut, Victoria, after control burns. Webster et al. (2003) noted that fires in spring represented a clear risk 

to Victorian populations because soil cracks were seasonally unavailable. Weed invasion was listed as one of numerous threats to 

its habitat (Webster et al. 2003). O’Shea (2005) found that the population at Iramoo Grassland, Victoria, was present in areas 

dominated by N. neesiana and N. trichotoma as well as T. triandra areas, and persisted well in 0.5 ha areas that were deliberately 

burnt. She argued that grassland structure (presence of perennial tussocks) rather than floristic criteria determined suitable 

habitat. 

Spiders, larval Lepidoptera, Black Field Crickets Teleogryllus commodus (Walker) (Orthoptera: Gryllidae) and cockroaches 

(Blattodea) are the most frequently recorded prey items (O’Shea 2005). Coulson (1990) reported that the majority of faecal 

pellets obtained from individuals at Derrimut Grassland, Vic., from mid January to early February contained unidentified larval 

Noctuidae (Lepidoptera) and T. triandra fragments which were presumed to have been larval gut contents. Faecal pellets studies 

by Wainer (1992) indicated prey 10-30 mm in length including locusts (Acrididae), an earwig, an ant, adult moths, as well as 

items already mentioned. Invertebrates common in the habitat that were not eaten included millipedes, slaters, slugs, bugs 

(Hemiptera) and beetles (Coleoptera). He also found unidentified grass seeds in the scats. Kutt et al. (1998) reported that nocutid 

larvae consistuted a high proportion of the diet from November to January and crickets were predominant from December to 

February, corresponding with the abundance of these taxa. The lizard appears to have a flexible foraging strategy with seasonal 

food preferences, and forages in the daytime. Brumation takes place in the soil from late March or April to September (O’Shea 

2005). 

Victoria’s largest documented and best studied population (an estimated 600 individuals) is in the Iramoo Grassland Reserve, 

Cairnlea. O’Shea (1997) provided preliminary details of alternative detection and trapping techniques for the lizard. O’Shea 

(2005) found that the population at Iramoo was present in areas dominated by either T. triandra or N. neesiana and N. 

trichotoma, however the exotic habitat was contiguous with the native habitat. Small scale (0.5 ha) summer and autumn fires 

interspersed in larger areas of unburnt habitat appeared to have little effect on population size and structure or habitat quality. 

147

Roofing tiles were found to be regularly used as artificial shelters and determined to be a useful survey and population 

monitoring tool. She also developed an objective pictorial method of identifying individual animals based on head scale patterns. 

D. impar is listed as Vulnerable at the national level (Hadden 1995, Webster et al. 2003), Vulnerable in the ACT (Sharp 1997) 

and NSW and Endangered in South Australia and Victoria (O’Shea 2005). 

Tympanocryptis pinguicolla (Mitchell) Grassland Earless Dragon (Agamidae) 

Tympanocryptis Peters is an endemic Australian genus of small, terrestrial dragons distributed through most of mainland 

Australia (Cogger 1983). T. pinguicolla was originally described as T. lineata pinguicolla Mitchell in 1948 and raised to species 

status in 1999 with a change of common name from Eastern (ACT Government 2005) or Southern Lined Earless Dragon 

(Brereton and Backhouse 2003). The vernacular name refers to the absence of an external ear opening, and of a functional 

tympanum (ACT Government 2005) which is “hidden below the skin of the head” (Cogger 1983 p. 250). It was once abundant in 

the ACT and was recorded from near Cooma in the Southern Tablelands and Bathurst in NSW (ACT Government 2005). In 

Victoria, anecdotal evidence indicates it was once not uncommon in basalt plains grasslands north and west of Melbourne. All 

Victorian records with habitat data were from areas of rocky, open T. triandra grassland, including sites on the Jackson Creek at 

Holden Flora Reserve near Diggers Rest, the upper reaches of the Merri Creek, north of Donnybrook, and the Little River Gorge 

west of Werribee. All sites where it was observed between 1988 and 1990 were T. triandra grasslands with Red-leg Grass 

Bothrichloa macra and Silky Blue-grass Dichanthium sericeum (R.Br.) A. Camus in ungrazed or lightly grazed paddocks 

(Brereton and Backhouse 2003). In the ACT it is known from seven sites in natural temperate grassland and appears to prefer 

sites with little disturbance and grasslands that are short and open (ACT Government 2005). T pinguicolla is a diurnal, wholly 

terrestrial, insectivorous(Brereton and Backhouse 2003) and arachnivorous (ACT Government 2005)species which shelters in 

small burrows (Brereton and Backhouse 2003) including those of invertebrates, under rocks and within Austrostipa tussocks 

(ACT Government 2005). It is an oviparous species and females are believed in the main to breed only once, and to die after 1 

year (ACT Government 2005). Loss and modification of habitat is probably largely responsible for its great decline (Brereton 

and Backhouse 2003). 

Other species 

The Pink Worm Lizard Aprasia parapulchella Kluge was known only from Coppins Crossing on the Molongo River, ACT and 

from near Tarcutta and Bathurst, NSW, the type specimens recorded under weathered granite rocks on a grazed, grassy riverside 

(Cogger 1983) and was listed as nationally endangered (Sharp and Shorthouse 1996). A recovery plan for the species in the ACT 

was published in 1995 (Sharp and Shorthouse 1996). The Corangamite Water Skink Eulamprus tympanum marnieae (Lvnnberg 

and Andersson) is endemic to the Victorian Volcanic Plain (DNRE 1997, Department of Sustainability and Environment 2007). 

Its inhabits rocky areas on the margins of wetlands in a limited area of the Victorian volcanic plains so exists on the margins of 

grasslands. “Weed invasion” is one threat to this species (Department of Sustainability and Environment 2007). The Pygmy 

Bluetongue, Tiliqua ?adelaidensis (Peters) is another grasslands species that may be threatened. Cogger (1983 p. 388) recorded 

that its distribution was “not known” with most recorded specimens from the Adelaide region. It was listed as ‘Indeterminate’ in 

the IUCN Red List of Threatened Vertebrates (IUCN 1988) with not enough information available to determine whether it was 

enadangered or rare. The Pygmy Bluetongue reportedly “depends on spider holes for shelter in the grasslands around Burra in 

South Australia’s Mid North” (Lunt et al. 1998). 

Invertebrates 

Plant invasions have the potential to modify interactions amongst species at all trophic levels. Changes to the composition, 

structure and functioning of communities caused by alien plant invasions have a major impact on native insects (Samways 2005). 

These effects include alteration of species richness and the composition of the fauna. The mechanisms include displacement of 

indigenous food plants, alterations to solar insolation by shading, and to shelter characteristics of the vegetation (Samways 

2005). The impacts are generally complex, with individualistic responses by the range of taxa. 

On a world basis, ungulate mammals are the dominant herbivores in tropical grasslands whereas in temperate grassland this role 

is taken by insects (Tscharntke and Greiler 1995). Insects consume a greater proportion of the plant biomass compared to their 

own biomass than mammals (Tscharntke and Greiler 1995). Several features of grasses play a role in determining their insect 

fauna: lack of secondary thickening (woody tissue), simple architecture, protected buds, almost complete lack of secondary 

compounds with herbivore deterrence properties, but an abundance and great variety of silica bodies (phytoliths) in the epidermis 

that increases their resistance to invertebrate herbivory (Stebbins 1986, Tscharntke and Greiler 1995,Witt and McConnachie 

2004). 

Invertebrates are the major component of biodiversity in temperate native grasslands in Australia (Yen 1999, Gibson and New 

2007). New South Wales grasslands “have abundant herbivorous insects” (Keith 2004 p. 104). But surveys of grassland 

invertebrates in Australia had barely begun by the mid-1990s when Driscoll (1994) stated that the invertebrates of south eastern 

Australian grasslands had “never been surveyed’. Keith (2004 p. 104) exaggerated the exent of knowledge when he stated that 

the “invertebrate fauna of the grasslands before their modification and use as pastures is poorly documented”. Wapshere’s (1993 

p. 344) statement that “nothing” was known about the nematode, mite and insect faunas of Austrostipa spp., despite their 

economic importance in native pastures, while not strictly correct (see the section above on the curculionid predators of Nassella 

species in Australia), nicely reflects the general paucity of information. Gibson and New (2007) considered remnant lowland 

native grasslands to be “among the least investigated” ecosystems entomologically in south-eastern Australia. Nevertheless it is 

has been “presumed widely” that loss of native plants in these systems has been accompanied by similar loss of invertebrate 

diversity (New 2000 p. 154). 

Invertebrate diversity appears to be particularly suitable for study and assessment of remnants because most of the vertebrate 

diversity has been lost from native grasslands, and because invertebrates are highly diverse, abundant and easy to collect (Yen 

1995). It has been suggested that the suites of macroinvertebrates present are most appropriate for characterising grasslands for 

conservation, and can act as flagship taxa, and that microinvertebrates may be superior for monitoring ecological functioning 

(Yen 1995). However invertebrates hardly offer direct approaches to biodiversity assessment. Inventory studies are a primary 

148

equirement, but unlike plants and vertebrates, complete inventories are idealistic (Yen 1999). Basic studies are required to 

characterise the invertebrate communities present and identify threatened species, and long term monitoring of permanent sites is 

necessary to fully determine the biodiversity that exists and the ways that it fluctuates. Yet, studies of natural grassland 

invertebrate communities in Australia have until recently been totally lacking (Yen 1995) and the conservation significance of 

existing invertebrate populations is difficult to assess because of major anthropogenic alteration of the habitat and lack of 

baseline survey data (Yen 1995 1999). 

Invertebrates have functional roles in most ecological processes including decomposition, nutrient cycling, soil aeration, seed 

dispersal, herbivory and pollination, and in food chains as prey, predators and parasites (Yen 1995 1999, Ens 2002a, Samways 

2005, Stephens 2006).They therefore offer possibilities for exploring ecosystem structure and function, given an understanding 

of their biology. 

Identification of the effects of biodiversity threats and management practices on invertebrates is difficult (Yen 1995 1999) and 

knowledge of the how insect diversity and conservation is affected by alien plant invasion is “very limited” (Samways 2005 

p.115). Assessment of insect populations and communities is complicated by their high spatial (

abiotic influences affecting lower trophic levels”. However there has been very little comparable research into spider 

communities in any habitats in Australia (Churchill 1997), they interact with invasive plants only indirectly (Mgobozi et al. 

2008) and their specialisation by prey type also appears to be relatively low, limiting their usefulness. 

Correlations between the richness of taxa may be coincidental, a response to common environmental factors (e.g. climate) or to 

different factors that are spatially covariant, or the result of biological interactions between taxa (e.g. Hymenoptera parasitoids 

and their hosts) (Wolters et al. 2006). Unfortunately many studies have used groups for which there is no a priori reason to 

believe have good indicator value (Wolters et al. 2006). 

Driscoll (1994) recommended beetles, ants and grasshoppers as suitable indicator groups for south-eastern Australian grasslands, 

based on the relatively large size of the constituent species, their functional significance, variablity in dispersal abilities and, 

possibly most importantly, the availability of taxonomic knowledge and expertise. But what studies of these groups might 

‘indicate’ is unclear. New (2000) suggested that ants might be good indicators of habitat complexity, but lack of primary 

phytophagy in the group suggests they would show little response to simple floristic change. Gibson and New (2007) suveyed 

beetles and ants in a Victorian grassland, but found that nearly all were widespread regionally, and provided no correlative 

measures for environmental change. Farrow (1999) argued that the use of phytophagous Chrysomelidae and Curculionidae – the 

most diverse Coleoptera families he detected in ACT grasslands – as indicators was compromised by poor understanding of their 

life histories and larval and adult host plants. Grasshoppers were considered also to be not particularly useful because of their 

highly clumped distributions due to particular oviposition requirements, their need for solar insolation and inter-annual 

dependence on rainfall (Farrow 1999). 

New (2000) assessed previous studies of ant diversity in temperate Australian grasslands and evaluated the suitability of ants as 

indicators of habitat conditions. Surveys that detected most of the surface active resident species commonly detected 20-30 

morphospecies. He considered ants to be poor indicators in Victorian grasslands because the species present are relatively habitat 

tolerant and the habitat is relatively heterogeneous at the fine scale. Ant diversity of grasslands dominated by native or exotic 

grasses were similar, and species composition of the grasses was probably less influential than structural diversity in the 

vegetation and overall habitat complexity. Miller and New (1997) evaluated pitfall trapping of ants as an indicator of grassland 

disturbance as manifested by the invasion of the exotic grass Holcus lanatus L., in Austrodanthonia grasslands at Mount Piper, 

Victoria. They detected 36 morphospecies by fortnightly pitfall trapping over seven months and concluded that the rankings of 

generic diversity at each site “usually coincided” with ranking of sites on a scale from “most natural” to “most degraded”. The 

grassland ant fauna appeared to be easier to define in terms of sampling effort than the much more diverse fauna of woodlands, 

but the grasslands sampled nevertheless indicated a mosaic of ant assemblages that differred between sites with apparent 

vegetational homogeneity. Native and exotic dominated sites had a similar number of species, while rare species and many more 

common species were not habitat specific. Only the subordinate Camponotus group appeared to be a good indicator of ‘natural’ 

sites (New 2000). Hinkley and New (1997) found that short term sampling using pitfall traps was inadeqate to assess ant species 

diversity and that sampling across seasons and extensive sampling in summer were required to enable reasonably complete 

inventories. Single trapping events (of c. 14 days) collected no more than half the species found by repeated trapping over 

fortnightly intervals for 7 months. 

Despite Evans’ (1966 p. 9) comment that the Australian cicadelloid (Hemiptera: Cicadellidae, Eurymelidae and Membracidae) 

fauna of “grasses, and of annuals and perennials generally, is very sparse”, Farrow (1999) found that the diversity and 

distribution 32 species of Cicadellidae obtained in sampling at 11 grasslands in the ACT, provided a good approximation of the 

total number of canopy insect species at individual sites. He recommended the family as a relatively homogeneous grass-feeding 

group suitable for use as an indicator. However Yen at al. (1994a) found only five Cicadellidae spp. in Victorian basalt plains 

grasslands. It is not clear whether this contrast in regional diversity is real or a sampling artefact (Farrow 1999). 

Ease of sampling, taxonomic accesibility, ecological (trophic and microhabitat) diversity and functional importance are generally 

agreed to be the most important criteria for indicator groups (Melbourne 1983, New 1984, Churchill 1997). Melbourne (1993) 

found that the diversity of genera of Formicidae and to a questionable degree Carabidae (Coleoptera) provided sufficient 

taxonomic resolution to assess differences in the species diversity of these groups in ACT grasslands, and that plant community 

was a reasonably good predictor of diversity of these taxa. However it is likely that no single invertebrate group could be be used 

as an adequate estimater for all the others (Melbourne 1993). Farrow (1999) for example found that ACT grassland sites ranked 

for biodiversity using the Collembola data of Greenslade (1994) did not correspond with his own rankings based on canopydwelling 

insects. A recent meta-analysis of over 200 studies found that no vertebrate, invertebrate or vascular plant taxon has 

been found to be a good predictor for the richness of other taxa; indeed supposed indicator taxa have often proved to be 

disappointingly inadeqate for the task, although invertebrates at the

e assigned the same biodiversity score as those that are rare and narrowly distributed (New 1984, Driscoll 1994) and links with 

other biological studies where species names are used can rarely be made except by later reference to preserved voucher 

specimens. Imprecise taxonomic information severely limits connections with the whole range of existing biological and 

ecological information about particular taxa (Churchill 1997). 

Thus it is generally difficult to determine whether a particular grassland area contains a good representation of the invertebrate 

fauna. The criterion of rarity is also difficult to assess given the general low level of invertebrate knowledge. It requires 

examination over temporal and spatial scales largely beyond the reach of current resources. Endemism may be a better criterion 

(Driscoll 1993), but again often requires more knowledge than is available (Melbourne 1993), except for a very limited range of 

organisms, or at the biogeographical scale. 

Landscape criteria and the biological attributes of species 

Landscape criteria are intended to enable the integration of habitat patch attributes, dispersal characteristics and dispersal 

opportunities of the organism with population biology, conservation genetics and the features of the surrounding landscape to 

arrive at guidelines for conservation (Driscoll 1994). For example, small populations may frequently become extinct, but that is 

not a problem if recolonisation from surrounding habitat is inevitable. However knowledge of the dynamics of invertebrate 

metapopulations is very difficult to investigate in fragmented grassland remnants (Yen 1999). Consideration of these 

complications led Driscoll (1994) to conclude that until such time as a wide range of invertebrates with variable dispersal 

abilities and life strategies could be assessed, all native grassland remnants were potentially important for invertebrate 

conservation, and conservation approaches should attempt to ensure maximal connectivity between the remnants. 

Consideration of the biological significance of grassland invertebrate faunas must include identification of taxa restricted to 

grasslands and determination of the presence of threatened species (Yen 1995). The most important criteria in determination of 

species at risk are low vagility (Hill and Michaelis 1988, Farrow 1999), low reproductive rate, long development period and 

degree of endemicity (Hill and Michaelis 1988). These criteria are clearly apparent for grassland invertebrate taxa known to be 

endangered (see below). 

Knowledge of the distribution of invertebrate species is often too limited to determine whether a species is restricted to 

grasslands. Ants are amongst the better known groups and ants of native grasslands have been found to be largely a subset of 

those of neaby woodlands (New 2000), again paralleling the vascular plants. Most grassland plants are also found in other 

vegetation formations, particularly grassy woodlands, but if invertebrate foodplants or larval hosts are restricted to grasslands 

then breeding populations of the invertebrate must also be restricted to grasslands. Some species may be restricted to grasslands 

by multiple factors, e.g. microclimate and food plant distribution. The endangered Ptunurra Xenica Oriexenica ptunarra L.E. 

Couchman (Lepidoptera: Satyrinae), a Tasmanian endemic, has Poa-feeding larvae and occurs in “open plains and poorly 

drained areas bordering mountain lakes and swamps”, in open grassy woodlands and tussock grasslands, mainly in the Midlands 

(Braby 2000 pp. 495-496). Two other Xenica species O. orichora (Meyrick) and O. latialis Waterhouse and Lyell are mainly 

restricted to alpine grasslands in south-eastern Australia (Braby 2000). The thermal requirements of these species (including 

open sunny areas for the adults to bask) are probably important in determining the habitat occupied. Similarly, grasshopper 

diversity is impacted when trees reduce insolation in grasslands (Samways 2005). 

Low vagility occurs when species are wingless or have reduced aptery, e.g. female grass anthelids, Pterolocera spp., 

(Lepidoptera: Anthelidae) are virtually wingless (Common 1990); the primitive, diverse, endemic Australian tribe Amycterini 

(Coleoptera: Curculionidae) with some species in grasslands, are wholly flightless (Zimmerman 1993), and morabine 

grasshoppers are very sluggish and sedentary. Poor dispersal abilities are particularly important in highly fragmented habitats 

where remnant habitat patches are small and vulnerable to severe disturbances. Farrow (1999) noted that three of four 

endangered ACT insects were flightless, but found that only seven species, about 2% of species sampled in canopies of ACT 

grasslands, had flightless adults: one cicadellid, four Orthoptera and two micro-Hymenoptera spp. A higher proportion, of course 

could be expected to occur in the ground- and soil-dwelling faunas. 

Low reproductive rates probably occur in some large weevils and morabine grasshoppers. Species with long development 

periods probably include the Golden Sun Moth Synemon plana Walker (Edwards 1994). Rhopaea sp. (Coleoptera: 

Melolonthinae) may have a 2 or 3 year life cycle, making them more vulnerable to local extinction (Allsopp 2003). Local 

endemicity is exhibited for example by sun moths (Castniidae), anthelids, Amycterinae and Morabinae. These factors are 

considered in more detail below in relation to various grassland invertebrates identified as threatened or at risk in south-eastern 

Australia. 

Management effects 

Various effects of grassland management on invertebrate species have been studied. Intensification of use generally results in 

loss of specialist species and increases in the proportion and dominance of common generalist species and exotic species 

(Tscharntke and Greiler 1995). Driscoll (1994) provided a brief review of grazing, pasture improvement, chemical use and fire. 

Kirkpatrick et al. (1995 p. 87) claimed that herbicides “may badly affect native invertebrates”, but there appears to be little 

published evidence. Farrow (1999 2006) evaluated the effects of grazing, fire and mowing on sites he sampled in the ACT, but 

the small number of sites sampled provided largely inconclusive results. “There was no consistent evidence that burning had any 

long-term impact on diversity” and “limited evidence from one site ... that regular mowing did not appear to limit biodiversity” 

(Farrow 2006 p. 2). Natural variation due to drought had a more profound effect in reducing diversity than any of these 

management measures (Farrow 2006). 

Maintenance of microhabitat and habitat variability and plant diversity are basic requirements for invertebrate conservation. 

Manipulation of patch dynamics, by promoting variation in plant diversity, plant age and successional stage can enhance species 

richness (Tscharntke and Greiler 1995) but the increased diversity may comprise widespread, abundant species with little habitat 

specificity. Farrow (1999 2006) found that the small, isolated urban grasslands with relatively uniform vegetation in the ACT 

had markedly fewer species of canopy-living insects than the larger, better-connected, more vegetatively diverse, peri-urban 

grasslands, but Farrow (2006 p. 11) concluded that the diversity was “not related to measurable or easily observable 

environmental variables including vegetation diversity”. 

151

Insect diversity in diverse central European meadows is approximately 1500 species (Tscharntke and Greiler 1995). Comparable 

data is not available for Australia. Yen (1999) found what appeared to be lower invertebrate diversity at ordinal (Order) or higher 

levels in the fauna of Derrimut Grassland compared to other, non-grassland regions of Victoria, but cautioned that a range of 

methodological factors might be the cause. Farrow (1999) found 328 morphospecies by sweep net sampling of 11 native 

grasslands in the ACT, considered to be a good representation of the canopy fauna, but excluding Lepidoptera, almost all Diptera 

and some minor orders. Similarly, Farrow (2006) found an estimated 383 species at 15 grasslands sites. Yen, Horne, Kay and 

Kobelt (1994a) found a total of 210 spp. in Victorian Basalt Plains grasslands remnants, but surveyed only three orders as a 

whole (Coleoptera, Hemiptera and Orthoptera) and just a small part (Formicidae) of another (Hymenoptera). The rest of the 

Hymenoptera along with the unsurveyed orders Lepidoptera and Diptera are all highly speciose, with micro-Hymenoptera in 

particular being highly diverse (Farrow 1999 2006). 

Increased insect diversity and biomass is generally correlated with increased floristic diversity, structural complexity (including 

vertical stratification) of the vegetation and plant biomass (Tscharntke and Greiler 1995, Ens 2002a). However if increased 

biomass results in reduced floristic diversity, invertebrate diversity is likely to decline (e.g. Hadden and Westbrooke 1999, 

McQuillan 1999, Ens 2002a) and increased biomass in temperate south-eastern Australian grasslands is generally associated with 

overdominance by a limited number of grasses which suppress smaller inter-tussock species. 

Impact of invasive grasses 

Data on the impact of invasive grasses on insects is very limited. Chown and Block (1997 cited by Samways 2005) found that 

the beetle Hydromedion sparsutum was smaller in areas of South Georgia Island where alien grasses were dominant. New (2000) 

found that ant diversity in grasslands dominated by native or exotic grasses was similar and species composition of the grasses 

was probably less influential than structural diversity in the vegetation. However Gibson and New (2007) pointed out that 

disturbance effects were confounded. Hagiwara et al. (2009) demonstrated the potential benefits to the endangered butterfly 

Lycaeides argyrognmon praeterinsularis of removing Eragrostis curvula, which increased flowering and seed production of the 

butterfly’s host plant. Melbourne (1993) and Melbourne et al. (1997) investigated the variation in numbers of a range of 

invertebrate taxa in grasslands of the ACT dominated by exotic or native species. Miller and New (1997) compared the ant 

faunas of areas dominated by an exotic grass and by native grasses. 

Grassland insects 

Pasture pests have generally been the main focus of invertebrate research in Australian grasslands (e.g. Gregg 1997). The great 

majority are native species subject to occasional outbreaks and are valuable components of biodiversity in natural areas. Losses 

from insect attack in grasslands on a global basis are estimated to amount to 9-32% of plant biomass (Tscharntke and Greiler 

1995). Insect herbivory can have significant impacts on rare and endangered plants. Archer (1984) recorded the depradations of 

unidentified grasshoppers on colonies of Thesium australe, a plant once widespread in temperate native grasslands, attributed to 

severe depletion of other vegetation by mammalian grazers (Archer 1984 1987). Important agricultural grassland pest taxa in 

Australia include grasshoppers and locusts (Acrididae), Teleogryllus commodus (Walker) (Orthoptera: Gryllidae), Therioaphis 

trifolii (Monell) and Acyrthosiphon pisum (Harris) (Hemiptera: Aphididae), weevils (Coleoptera: Curculionidae), larvae of 

Elateridae, Tenebrionidae and Scarabaeidae (particularly Melolonthinae, Aphodius tasmaniae (Hope) (Aphodiinae) and 

Adoryphorus couloni (Burmeister) (Dynastinae)(Coleoptera), underground grass grubs (Lepidoptera: Hepialidae), larvae of 

Noctuidae and Pyralidae (particularly Hednota spp.) (Lepidoptera) and ant seed predators (Hymenoptera: Formicidae) (Gregg 

1997). Mites including Balaustium and Penthaleus spp. are important cereal pests (Anon. 2008b). 

Many adult Melolonthinae species feed on trees, particularly Eucalyptus spp., or use them for mating, while others do not feed in 

the adult stage (Roberts et al. 1982). Larvae live in the soil, eating roots, probably often of grasses, and other organic matter. In 

pastures on the New England tablelands (NSW), smaller species in the tree-feeding group were significantly more abundant in 

areas with low tree densities than larger species (Roberts et al. 1982). Highest densities of the non-tree-feeding group, which 

included the largest species (e.g. Rhopaea spp.), occurred in areas with 0-10% tree cover. Ridsdill-Smith (1975) noted that 

several of the Northern Tablelands species were known to eat living grass roots and found that Sericesthis nigrolineata 

(Boisduval) preferred grass roots to dead organic matter. Hardy (1976b) found that Scitala sericans Erichson predominantly 

inhabits grasslands and dry sclerophyll forests, the native grasslands inhabited being dominated byPoa spp. and also the exotic 

Agrostis capillaris L. in some areas. Larvae of Antitrogus Burmeister (Melolonthini) “feed on the roots of grasses and other 

similar plants” (Allsopp 2003 p. 159). 

Various Dynastinae are also common in pastures and lawns in south-eastern Australia, where many feed on grass roots. 

Cyclocephala signaticollis Burmeister, introduced to Australia from South America has larvae that damage pasture and turf in 

Australia (Carne 1956) and is now common in the ACT (Robin Bedding CSIRO pers. comm. via M. Malipatil). This species 

inhabits the core range of N. neesiana in Argentina and likely damages it. According to A. Martinez (reported in Carne 1956 p. 

220) it is found in “the provinces of Buenos Aires, the eastern part of Córdoba, southern Santa Fé, in Entre Rios and the northeast 

of the Pampa territory [and] Uruguay ... the roots of native grasses are the natural food .. while they also attack ... wheat, 

maize ... and barley”. The well-studied pasture pest Adoryphorus couloni was the most abundant beetle collected in pitfall traps 

at Craigieburn grassland by Gibson and New (2007). 

Dramatic fluctuations of native insect populations in pastures are commonplace (e.g. Davidson 1982), as they are in natural 

grasslands (Yen 1999). Population irruptions of phytophagous species may play a role in the patch dynamics of the lowland 

grasslands. For example, the underground feeding damage to Poa snow grass by the Alpine Grassgrub Oncopera alpina Tindale 

in the Australian Alps, described as “extensive patch death” by McDougall and Walsh (2007) is “part of an important ecological 

cycle opening up overgrown grass swards to invasion by the numerous flowering herbs for which the Kosciuszko area is 

famous” (Edwards 2002 p. 61). Similarly Green and Osborne (1994) reported that the Poa snow grass feeding larvae of the 

casemoth ‘Plutorectis’ caespitosae Oke (Psychidae, Lomera caespitosae in Common 1990) cause severe patch damage to large 

areas in the subalpine zone, especially below the treeline, when in large numbers, but recovery after winter is rapid. 

152

Nematodes are mostly minute animals that can be extremely abundant in soils and on plants, particularly on the roots. Numerous 

species are associated with grasses in Australia and several have been recorded from native Stipeae. These are discussed in an 

Appendix to this literature review. 

Exotic invertebrates are often widespread and common in native grasslands, particularly in urban fringe areas (personal 

observations). Collembola is one group with a substantial proportion of exotics, however many higher taxa including various 

Coleoptera families and Formicidae are largely or almost entirely dominated by native species (New 2000). Common exotics 

present in grasslands include molluscs (slugs and snails), red legged earth mites Halotydeus destructor (Tucker), insects such as 

the aforementioned aphids, the African black beetle Heteronychus arator (Scarabeidae: Dynastinae), the weevils Graphognathus 

leucoloma (Boheman) and Sitona discoideus (Gyllenhal) (Gregg 1997, Hill et al. 1997),the European honeybee Apis mellifera, 

and possibly earthworms (Annelida: Haplotaxida). Morgan (1995b) found that H. destructor grazed seedlings of the endangered 

Rudidosis leptorrhynchoides but did not significantly affect their survival, and that grazing by unidentified slugs occurred mostly 

on very young seedlings and caused little damage. Exotic Chriothrips spp. (Thysanoptera: Thripidae) are found on grasses in 

south-eastern Australia (Mound and Palmer 1972). The grassland daisy Senecio macrocarupus is subject to attack by red-legged 

earthmites Halotydeus destructor (Tucker) and aphids (Hills and Boekel 2003). H. destructor and blue oat mites Penthaleus 

major (Dugès) attack a wide range of native grassland forbs including species of Craspedia, Podolepis and Senecio (Robinson 

2005). The Portuguese Black Millipede Ommatoiulus moreletti (Lucas) (Julidae) has spread into Victorian basalt plains 

grasslands, and although there is no evidence of competition with native millipedes, it may influence litter decay rates (Yen 

1995). Many of the invasive insect species probably have significant effects on grassland plants and the ecological functioning of 

the system, but there appear to be no studies that have directly addressed particular impacts. The agricultural literature, although 

restricted to few species, contains some valuable data about invertebrate impact on ecosystem processes, particularly effects on 

plant biomass production (Gregg 1997) and soils. 

A range of exotic molluscs are considered pests of pastures (Smith and Kershaw 1979, Kershaw 1991) and have been suggested 

to be important predators of sensitive native plants such as orchids (Sydes 1994, Daniell 1994). Exotic slugs “appear to be highly 

invasive of native grasslands” and were found to be equally abundant in grasslands dominated by exotic and native grasses in the 

ACT (Melbourne et al. 1997 p. 366). Daniell (1994) recorded no or few species of native molluscs at four Melbourne grasslands, 

and observed that exotic slugs can occur at very high densities in native grasslands. Holland et al. (2007) found only exotic 

molluscs, three snail and five slug species, in extensive surveys in the Victorian volcanic plains. A few mollusc species have 

been identified as threats to particular native taxa, e.g. unidentified slugs and snails are considered a threat to Diuris 

fragrantissima (Webster et al. 2004), the Black-keeled Slug Milax gagates (Draparnaud) defoliates Rutidosis leptorrhynchoides 

(Daniell 1994), slug predation has caused the loss of a Pterostylis nutans R.Br. (Orchidaceae) colony (Sydes 1994), while 

unidentified slugs and possibly Common Garden Snails Helix aspersa (Müller) caused mortality of Thesium australe planted at 

Lake Omeo (Scarlett et al. 2003). Lenz et al. (2003 p. 30) applied molluscide in a low-diversity, weedy native grassland with a 

high population of the white snail Cernuella virgata da Costa and found that snail exclusion had no effect on vascular plant 

species composition over 9 months. The dietary preferences of introduced herbivorous molluscs in regard to native plants do not 

appear to have been investigated (Lenz et al. 2003) except for M. gagates which was found by Holland et al. (2007) to prefer 

some native plants over others, with the two native grasses tested having relatively low palatability. 

Grass-feeding insects 

Various insect groups that avoid plants with toxic metabolites are restricted to Poaceae including some Acrididae (Orthoptera) 

and Auchenorrhyncha (Hemiptera). Theoretical considerations suggest that the general absence of toxic principles should result 

in a lower ratio of endophagous to ectophagous feeders in grasses than in dicotyledons, and that appears to be the case, with 

external feeders being much more common than gall-makers, borers and miners. Ratios of c.30:10 and 30:1 have been recorded 

in six grass species (Tscharntke and Greiler 1995). Dominance of ectophages is also apparent in Sporobolus spp. (Witt and 

McConnachie 2004). Veldtman and McGeoch (2003) found five grass species with insect galls, caused by five gall-forming 

insect species, in a broad survey in South Africa. Poaceae had one of the highest number of galled species compared with other 

plant families. However, herbaceous plants generally have fewer gall-forming species than woody plants, possibly in part 

because lignified material is more long-lasting, and is thus ‘safer’ for gall formation (Veldtman and McGeoch 2003). Most of the 

chewing ectophages on grasses are oligophagous, whereas amongst the main ‘sucking’ groups (Hemiptera) Delphacidae are 

monophagous while Cicadellidae has a higher proportion of oligophages (Tscharntke and Greiler 1995). 

As with vertebrates, the nutritional quality of grass foliage, particularly silicate and lignin content, are important determinants of 

palatability for invertebrates. Juvenile foliage has a reduced content of these structural polymers but has better chemical defenses 

against herbivores. C 3 species have higher protein contents than C 4 species, so may be preferred by herbivores. Large abundant 

grasses generally have larger faunas (5-12x) than rare, small species. Two variables, shoot length and life-cycle dichotomy 

(annual or perennial) explain a high proportion of variance in the species richness of a particular grass. Unpredictably in the 

spatial and temporal distribution of annuals, rather than any difference in the number of possible niches they offer appears to 

explain this (Tscharntke and Greiler 1995). 

Data on the grass food plants of insects in Australia, as elsewhere in the world, is very fragmentory. The literature appears to be 

devoid of studies of whole faunas associated with Australian native grass species. According to Wapshere (1993) no arthropods 

had been recorded from Nassella trichotoma in Australia, despite its long presence in the country, and nothing was known of the 

invertebrate faunas of Austrostipa species (although this is no longer correct for these grasses). At attempt has been made to 

draw together some of the scattered information sources on grass invertbrates in an Appendix to this Literature Review. Taxa of 

Gondwanan origin such as Austrodanthonia may harbour a larger range of endemic coevolved invertebrates than species such as 

T. triandra which have colonised Australia more recently from the north (E.D. Edwards cited by Driscoll 1994). 

Lawton and Schroder (1977 p.137) compiled data on the insects associated with species of British plants but excluded Poaceae 

“because the insect data appeared to be particularly unreliable”, the entomological literature frequently recording “grass”, rather 

than particular grass species as food plants. However they found that the monocots studied had the fewest insect species 

associated with them compared with shrub, perennial herb, ‘weeds and annuals’, and aquatic dicot herb groups investigated. 

153

Nevertheless, based largely on Northern Hemisphere knowledge, the basic insect phytophage assemblage, for the smallest 

grasses, consists of one species each of Eurytomidae (Hymenoptera), Cecidomyiidae (Diptera) and Pseudococcidae (Hemiptera) 

(Tscharntke and Greiler 1995). The large, cosmopolitan Common Reed, Phragmites australis (Cav.) Trin. ex Steud., is the most 

speciose host known, attacked by c. 100 insect species (Tscharntke and Greiler 1995) and has a total of over 160 associated 

arthropods, approximately half of which are endophages (Witt and McConnachie 2004). Its diverse flora may be largely a result 

of a long evolutionary history, it being a “Palaeogenic relict” belonging to the”most ancient of modern grasses” the Arundineae 

(Tsvelev 1984 p. 59) and its broad geographical distribution. The phytophage complement is dependent on many factors, but 

larger faunas are associated with wider geographical range, large size, predictable occurrence and perenniality (Lawton and 

Schroder 1977). More than 24 arthropod species have been identified as potential biological control agents for the intertidal grass 

Spartina alternifolia Lois. in the USA, while Calamagrostis epigejos (L.) has 10 endophagous arthropod species (Witt and 

McConnachie 2004). 

Isoptera (Termites) 

Drepanotermes grass-harvesting termites are a conspicuous feature of arid Australia but are not found in the grasslands of 

temperate south-eastern Australia (Watson 1982), where most grasses appear to escape termite attack. 

Coleoptera 

Unlike on dicotyledonous groups, beetles (Coleoptera) are generally relatively depauperate on grasses. Only 8 spp. are known 

from Phragmites australis and only 2 of 24 apparent specialists on Spartina alternifolia are beetles (Witt and McConnachie 

2004). Coleoptera are “relatively abundant” on smaller grasses including Sporobolus spp. and Nassella trichotoma, but are 

mostly pollen feeders (Witt and McConnachie 2004). 

Thysanoptera 

Thrips (Thysanoptera) are common inhabitants of grasses. Chirothrips species breed only in the flowers of grasses and can 

reduce seed production and limit grass regeneration (Mound and Palmer 1972). Six species are recorded in Australia, all 

introductions. Native thrips of grass flowers are species of Odontothripella. Other species associated with Poeaceae flowers 

include Caliothrips striatopterus (Kobus), common in grass flowers in Queensland, Haplothrips froggatti Hood common in 

subtropical Australia, Desmothrips reedi Mound and D. tenicornis (Bagnall). Many other species are associated with grasses 

including Phibalothripis longiceps (Karny), leaf tissue-feeding species related to Anaphothrips, usually found in leaf funnels, 

Podothrips species predatory on coccoids at the base of grasses in the tropics, and species found in the bases of grass tussocks 

that feed on fungi (Mound and Palmer 1972). Desmothrips reedi (Aelothripidae)lives at the base of grasses, Moundothrips and 

Phibalothrips (Thripidae) live on grasses and Odontothripella are found on grass flowers (Mound and Heming 1991) including 

O. compta Pitkin, O. reedi Pitkin and O. unidentata Pitkin (Pitkin 1972). Introduced Limothrips is common on grasses in cooler 

areas of southern Australia often living on leaves, the introduced Aptinothrips rufus is abundant on grasses and Haplothrips 

(Phlaeothripidae) live in grass flowers (Mound and Heming 1991), five species (H. anceps Hood, H. angustus Hood, H. gowdeyi 

(Franklin), H. froggatti Hood and H. pallescens (Hood)) apparently breeding there, one species living in the base of tussocks, 

and other species including Antillothrips cingulatus (Hood), Apterygothrips australis Pitkin and Podothrips xanthopus Hood 

living on or amongst grasses, the latter probably predatory on small arthropods such as coccids, mites or other thrips (Pitkin 

1973). 

Diptera 

True flies are relatively more common on Poaceae than on dicotyledons, notably gall midges Cecidomyiidae, leafminer flies 

Agromyzidae and grass flies Chloropidae. On a world basis, many stem-boring and stem-galling Diptera on grasses have narrow 

host ranges (Witt and McConnachie 2004). 

Many cecidomyiid species appear to be monophagous at species level (Witt and McConnachie 2004). The Australian 

cecidomyiid fauna is extremely poorly known (Harris 1979, Gagné 2007, R. Adair and R. Gagné pers. comms.) but several 

species belonging to three genera, Contarinia, Lasioptera and Geromyia have been recorded from grasses, largely in northern 

Australia (Harris 1979). Larval Cecidomyiidae feed in the inflorescence, inside the culms, or under leaf sheaths often near the 

base of the plant. The Contarinia spp. inhabit inflorescences and seed heads, and include C. brevipalpis Harris which attacks 

Eragrostis brownii (Nees) Kunth, C. dichanthii Harris which attacks Dichanthium spp. including D. sericeum (R.Br.) A. Camus, 

and undescribed species on Heteropogon contortus (L.) P. Beauv. ex Roem.and Schult. and Themeda triandra (Harris 1979). 

There are “probably hundreds” of undescribed Australian species (Harris 1979 p. 168). Lasioptera species are found on 

inflorescences of grasses including Panicum, Setaria, Bothriochloa and Heteropogon, while the single Geromyia species is 

“probably restricted to Setaria” (Harris 1979 p. 164). No Cecidomyiidae species are known from Austrostipa or Austrodanthonia 

spp. It is highly likely that exploration of the temperate Australian fauna will reveal many species that inhabit native grasslands 

and attack grasses (Adair and Gagné pers. comms.). The biology of a New Zealand species that feeds on developing seeds of 

New Zealand ‘snow-tussocks’ Chionochloa spp. has been described by Kolesik et al. (2007). The larvae do not form galls but 

overwinter in the floret after feeding on developing seeds. The species appears to be the main seed predator driving mast seeding 

of these grasses, destroying up to 60% of florets (Kolesik et al. 2007, Kelly et al. 2008). 

Agromyzidae has relatively high host plant specificity with most species specific to a single plant genus or family, Asteraceae 

hosting the greatest number of species, and polyphagy (across plant families) being rare (Spencer 1977). Larvae may be leaf 

miners, internal feeders throughout the plant or selective feeders in roots, stems or flowers (Ferrar 1987, Spencer 1989). Adults 

feed on plant sap via punctures in the tissue made by the female (Ferrar 1987). No Australian Agromyzidae are known from 

Stipeae (Spencer 1977), but members of the family are recorded from other grasses in Australia. Spencer (1977) listed 

Pseudonapomyza ?spinosa Spencer from Urochloa subquadripara (Trin.) R.D. Webster at Darwin, NT (as Agromyza sp. on 

Brachiaria miliformis (Presl.) Chase in Kleinschmidt 1965), P. spinosa from the introduced Eleusine indica (L.) Gaertn. and 

Agromyza sp. from Oplismenus compositus (L.) Beauv., while Kleinschmidt (1965) also listed Cerodontha australis Malloch 

from Poa annua L. and Pseudonapomyza spicata (Malloch), a “common grass-mining species”, probably of exotic origin, from 

Eleusine indica. Elsewhere in the world P. spinosa attacks wheat and barley and “other wild grasses certainly serve as hosts” 

(Spencer 1977). U. subquadripara is widespread in NSW, but not in areas of temperate grassland, the common weed E. indica is 

154

found in all mainland states, O. compositus is not known from Victoria or NSW. Other Pseudonampomyza spp. with angulate 

third antennal segments have host plants restricted to Poaceae, but no known Australian hosts: P. probata Spencer, P. rara 

Spencer, P. salubris Spencer and P. pudica Spencer. Agromyza mellita Spencer and A. venusta, both known from northern 

Queensland have male genitalia with “the characteristic form of grass-feeders” (Spencer 1977 p. 122). 

Some Anthomyiidae (Diptera) also attack grasses including the wheat bulb fly Delia coarctata (Fallén), D. extreminata 

(Malloch) in Bromus in the USA, and Phorbia spp. mining stems and shoots (Ferrar 1987). Others are known to breed on 

Epichloe fungi on grasses (Ferrar 1987). 

Larvae of numerous Chloropidae species live in young shoots and stems of grasses and may feed largely on bacteria (Colless and 

McAlpine 1991) or be saprophagous rather than phytophagous (Ferrar 1987), however little is known of the biology of 

Australian species (Sabrosky 1989). Ferrar (1987) provided summary information on some pests of cereals and pasture grasses 

elsewhere in the world, plus a tabulation of the genera known from Poaceae. Most Chloropinae attack Poaceae or Cyperaceae, as 

do some Oscinellinae (Ferrar 1987). Some Oscinella species appear to be monophagous on particular grass species, while others 

have multiple known grass hosts (Ferrar 1987). Thirty-two Chloropidae spp., mostly endophages, have been collected from 

Phragmites, over 20 of which are monophagous (Witt and McConnachie 2004). Diplotaxa similis Spencer has been commonly 

found in the florets of Chionochloa spp. in New Zealand and has been viewed as the major seed predator, although they may 

actually often destroy the flower before seed formation, and there has been considerable confusion about whether the damage is 

caused by the chlropopid or Cecidomyiidae (Kelly et al. 1992). The wingless D. moorei (Salmon) Spencer is a predator of 

Festuca novae-zelandiae author (Kelly et al. 1992). 

Some Ephydridae, mostly Hydrellini, attack grasses, and some are cereal pests, but most breed in aquatic situations or wet areas 

and feed on decaying matter, algae etc. (Ferrar 1987). Hydrellia spp. are recorded from Lolium, Poa, Pennisetum, Holcus, 

Panicum, Echinochloa and other grasses outside Australia (Ferrar 1987) 

Hemiptera 

Hemiptera also appear to have good representation on grasses. Specialist insects on Spartina alternifolia are mainly Hemiptera 

(Witt and McConnachie 2004). Among the Heteroptera, the Leptocorisini (Alydidae: Leptocorisinae) is closely associated with 

grasses and includes a number of rice pests, while the Stenodernini (Miridae) and Blissinae (Lygaeidae) are also grass 

inhabitants (Ahmad 1965). Lygaeids are mostly seed eaters but the Blissinae (Chinch Bugs) feed on grass sap (Slater 1991). 

Subfamily Cyminae of Lygaeidae contains “small, brown bugs that live in seed heads and resemble seeds of various sedges and 

rushes” (Slater 1991 p. 502) and possibly includes some grass feeders. The pachygronthine Stenophyella macreta Horváth “is 

often common in seed heads of grasses” (Slater 1991 p. 502) and “appears to feed on a number of grasses even when they are 

completely dry” (Slater 1976 p. 135). Nabidae are predacious but oviposit in grass stems (Gross and Cassis 1991). Many grassfeeding 

Homoptera have a narrow host range, but mealybugs Pseudococcidae and soft scales Coccidae are generally 

polyphagous (Witt and McConnachie 2004). 

Farrow (1999) found that Cicadellidae (leafhoppers) were very numerous in grasslands in the Australian Capital Territory and 

this family along with other planthoppers (Delphacidae, Eurymelidae, Flatidae, Membracidae, Ricanidae), froghoppers 

(Cercopidae), cicadas (Cicadidae) and allies, collectively classified in the suborder Auchenorryhncha, has been identified as a 

highly appropriate higher taxon for evaluation of the conservation significance of grasslands and monitoring environmental and 

habitat change, at least in Europe (Biedermann et al. 2005). They are diverse, with many rare species, wholy herbivorous, their 

numerical abundance supports higher trophic levels, and perhaps most importantly, rhey have high and rapid sensitivity to a 

range of disturbances (Biedermann et al. 2005). The European grassland fauna however is much better described than that of 

Australia, baseline studies of whole faunas have been undertaken and some effects of grassland management on the faunas have 

been determined. 

Poaceae probably supports a higher diversity of Australian cicadas (Cicadidae) than any other plant family, but few species 

inhabit the grasslands of the south-east (Moulds 1990). Nymphs of the grass-feeding species feed on grass roots, and the adults 

of some species oviposit in grass stems or leaves (Moulds 1990). Evans (1966) considered the cicadelloid fauna of Australian 

grasses and herbs to be depauperate and Australian grasslands to carry “a very small leafhopper population and ... even fewer 

cercopids” (p. 20) in comparison with other parts of the world. However Day and Fletcher (1994 p. 1119) noted that there had 

been “little systematic collecting” of Cicadelloidea in Australia and that distribution data was “very inadequate”. According to 

Evans (1966 p. 21) both Cicadelloidea and Cercopoidea were “not usually abundant on grasses” in Australia, and most grassfrequenting 

species were introduced, but he acknowledged that records of food plants were very scanty. However Evans (1966 p. 

133) thought it “probable” that all species of Hecalini (Cicadellidae) are grass-feeders. Hecaleus arcuatus (Motschulsky) known 

from Queensland and outside Australia is recorded from T. triandra and Heteropogon (Day and Fletcher 1994). Day and Fletcher 

(1994) listed numerous species with known grass hosts in Australia but their designation of “host” is not rigorous; e.g. Evans 

(1966) stated that Mircrolopa minuta Evans was collected on grasses, which became in Day and Fletcher (1994 p. 1123): 

“Known hosts: Poaceae (Evans 1966: 87)”. Their host information “frequently reflects the plant species on which specimens 

were colleced and do not necessarily reflect the true hosts” (Day and Fletcher 1994 p. 1121). 

Hymenoptera 

Many ‘seed chalcids’ Eurytomidae (Hymenoptera) are host specific and different species develop in different positions within 

the culm (Witt and McConnachie 2004). They can cause major reductions in forage yield and seed weight, including a reduction 

of 60% in Heterostipa comata (Trin. and Rupr.) Barkworth, with consequent reduction in germinability (Witt and McConnachie 

2004). A Tetramesa sp. in Africa that bores in the stems of Sporobolus spp. was found to infest 33% of culms at one site. 

Infested culms were significantly shorter than uninfested and 60% had deformed infloresences (Witt and McConnachie 2004). 

The tribe Cephini of the sawfly family Cephidae consists of grass-mining specialists, a number of which are cereal pests, but is 

restricted to the Nearctic and Holarctic (Ivie 2001). 

155

Lepidoptera 

Amongst the Lepidoptera that attack Poaceae, the Noctuidae are generally polyphagous (Witt and McConnachie 2004). A high 

proportion of Satyrinae (Nymphalidae) and Hesperiidae are grass feeders (Braby 2000). A study of 13 satyrines and 3 hesperines 

showed that they all stored the grass secondary metabolite flavonoid tricin, while one species stored glycosylflavones, with the 

flavone stored in the wings and body and constituting 1-2% of their dry weight.(Harborne and Williams 1986). One important 

role of flavonoids in grasses is probably as grazing deterrents (Harborne and Williams 1986) and these butterflies have therefore 

to some extent overcome those defences. A species of Gelechiidae has been found rarely attacking the florets of Chionochloa 

spp. in New Zealand (Kelly et al. 1992). 

Australian Grassland Invertebrate Faunas 

A few studies of the faunas of particular natural grassland associations have been undertaken in Australia including Sharp 

(1997), Melbourne (1993), Farrow (1999 2006), New (2000), Gibson and New (2007) and the studies of Yen and colleagues in 

Victoria. The study of Farrow (1999) was a pioneering quantification of the biodiversity of a major part of the insect fauna of 

ACT grasslands. 

Victoria 

Terrestrial invertebrates of western Victorian Basalt Plains Grasslands have been surveyed and discussed by Yen (1999) and 

assessed by Yen, Horne, Kay and Kobelt (1994) and Yen et al. (1995), and in the western region of Melbourne by Yen, Kobelt, 

Lillywhite and Van Praagh (1994). These were the first baseline invertebrate surveys of native temperate grasslands in southeastern 

Australia (Farrow 1999), however these are long unpublished reports containing little analysis. Previous studies covered 

only a very limited number of individual taxa (e.g. Key 1978, Horne 1992, McDougall 1989), or were a part of dietary studies of 

endangered vertebrates including Delma impar (Coulson 1990, Wainer 1992) and Perameles gunnii (Yen 1995) or related to 

agricultural pest species (e.g. Schroder 1983). Approximately 50 taxa were identified from pitfall trap specimens taken at 

Derrimut Grassland Reserve by Kathy Ebert (Coulson 1990) but only ants and some spiders were identified, and only to genus. 

Initial pitfall trap sampling suggested that basalt plains grasslands have lower invertebrate divesity than other habitats in Victoria 

(Yen 1999), paralleling Willis’s (1964) view of the vascular plant and bryophyte diversity. 

Hadden (1997, 1998) investigated invertebrates in both the Northern and Western Plains, but did not provide specific 

identifications. Hadden and Westbrooke (1999) detected 160 arthropod morphospecies using pitfall traps, sweep net and hand 

searching in a grazed T. triandra pasture near Ballarat: 26 Formicidae, 90 Coleoptera and 44 Araneae. The Coleoptera and 

Araneae were identified to family level and the Formicidae to genus. Various features of the fauna were identified including the 

low number of aerial web-building spiders, the dominance of Iridiomyrmex spp. ants and the small size of those Coleoptera spp. 

that were abundant. Gibson and New (2007) detected 24 morphospecies of Formicidae and 27 Coleoptera species by pitfall 

trapping in spring and summer at Craigieburn grassland, Victoria. Both major taxa appeared to be representative of regional 

diversity and included no rare or apparently grassland-specific species. 

Collecting techniques used in the surveys of Yen et al. were pitfall trapping, weeping, suction sampling, canopy fogging and 

direct searching. Yen, Horne, Kay and Kobelt (1994) reported on 1992-3 seasonal surveys of 12 remant grassland sites 

representative of the range of site types, management practices and conservation ranking of all sites previously listed McDougall 

and Kirkpatrick (1994): 5 roadside reserves (4 burnt annually), 3 railway reserves (2 burnt annually), 2 privately owned pastures, 

1 conservation reserve and 1 cemetry; and an additional 5 sites were surveyed once: 2 rail, 1 roadside, 2 private, along with 6 

areas adjacent to seasonally sampled sites. Yen, Kobelt, Lillywhite and Van Praagh (1994) surveyed a variety of vegetation types 

during 1991-93 that included 2 Poa grasslands (Point Cook and Werribee) and 5 T. triandra grasslands (Truganina, Derrimut, St 

Albans rail reserve, Manor rail reserve, Evan St rail reserve). Yen et al. (1995) surveyed 29 additional grassland sites during 

1994, completing their coverage of all the remnant grasslands listed by McDougall and Kirkpatrick (1994) that were still in 

existence. 5 of these sites were in the Melbourne region and the remainder between Ballarat and Hamilton. 

Yen, Kobelt, Lillywhite and Van Praagh (1994) quantified the numbers of individuals collected by order, and by genus or 

species (where possible) for Coleoptera (beetles), Formicidae (ants) and Araneae (spiders). Yen, Horne, Kay and Kobelt (1994) 

and Yen et al. (1995) duplicated this analysis but also included species or genus level data for Orthoptera (grasshoppers and 

crickets) and Hemiptera (true bugs) and reported on spiders only at family level. 

A reference collection of grassland invertebrates has been established at the Museum of Victoria under the supervision of Peter 

Lillywhite, and identification of included taxa is proceeding gradually (Yen pers. comm. 2006, Kobelt pers. comm. 2007). 

Miller and New (1997) determined the ant fauna of (?derived) Austrodanthonia grasslands at Mount Piper, Victoria. The fauna 

was a far less diverse subset of that found in nearbye woodland and only one species was restricted to grassland. They found that 

sites invaded by Holcus lanatus supported an average of c. 16 and a total of 28 morphospecies while the more natural grasslands 

supported an average of c. 21 and a total of 33 morphospecies. However matched invaded and uninvaded sites”only a few tens of 

metres apart, did not differ significantly from each other in diversity” (p. 378) and “many species” showed “little apparent 

discrimination in relation to dominant grass species” (p. 381). Four morphospecies were trapped only at H. lanatus sites, while 

seven were trapped only at the Austrodanthonia sites. “Overall disturbance may be more significant to ants than simple 

replacement of native by exotic grass species” (Miller and New 1997 p. 381). 

Few unique features have been noted about Victorian grassland invertebrate assemblages. One such example is the presence of 

ants that make use of rocks as a habitat. These ants are absent from ACT grasslands (Melbourne 1993). 

Australian Capital Territory 

Little is known about the distribution and ecological requirements of most invertebrate species in ACT grasslands (Sharp 1997 p. 

4), although the fauna is much better studied that most other grassland areas, partly due to the presence of CSIRO Entomology 

and its predecessors in the national capital since 1928. Driscoll (1994) integrated knowledge then available to develop an 

invertebrate conservation and research strategy for the ACT. Numerous localised studies of selected grassland groups or species 

have been undertaken, including Edwards (1994) on the Golden Sun Moth, Rowell and Russell (1995) on the grasshopper 

156

Keyacris scurra, Melbourne (1993) on ants and carabid beetles, Greenslade (1994) on Collembola, Melbourne et al. (1997) on 

native crickets (Gryllidae) and exotic slugs (Arionidae, Limacidae and Milacidae), Sharp (1997) on all orders and their 

relationship to grassland composition, structure and functioning and Farrow (1999, 2006) on canopy-living insects and their 

relationships to management factors, season of sampling and vegetation type. 

Melbourne (1993) sampled three types of native grassland, dominated respectively by T. triandra, Austrostipa (typically A. 

bigeniculata) and Austrodanthonia and two types of exotic grassland dominated by Phalaris aquatica (improved pasture) and 

Avena fatua (an enriched grassland) maily using pitfall traps. A total of 37 ant species in 18 genera and 24 carabid species in 22 

genera were detected. The pitfall trap catch size of ants was significantly larger in plots where vegetation was experimentally 

cleared and plots where litter was removed. Some species were more commonly trapped in cleared areas, some in uncleared, 

with one species showing no response to vegetation density. Slug (Limacidae and Milacidae) catches were approximately equal 

across treatments. However analysis suggested that structural differences in the grassland vegetation effected the efficiency of 

the traps, and that the trap catches in different vegetation types did not reflect actual abundance. Further analysis indicated that 

ant catches in the relatively open Austrodanthonia grasslands were approximately twice those in the more dense T. triandra and 

Austrostipa associations, and were by far the lowest in Phalaris pasture. Grassland type also affected carabid abundance (almost 

all being Notiobia edwardsii), with T. triandra having the lowest numbers and Austrostipa the highest. Catches of crickets were 

also lowest in T. triandra, while slugs were most abundant in the two most highly modified grassland types and next in T. 

triandra. The number of ant species trapped also varied significantly between grassland types, Austrostipa and Austrodanthonia 

being highest, but the number of carabid species was not signifcantly different between types. Other sampling methods yielded 

only very low numbers of individuals. 

Melbourne et al. (1997) further reported on the crickets and slugs, the real abundances of which were determined to be properly 

reflected in the pitfall trap catches. Slug numbers increased with increasing density of the grasslands, possibly because slugs 

moved less on the drier substrates associated with more open habitat. The cricket Bobilla victoriae Otte and Alexander was 

several times more abundant in Phalaris aquatica grassland than the other grassland types, while T. commodus was also more 

common in P. aquatica. 

Greenslade (1994) sampled springtails (Collembola) at 29 grassland and grassy woodland sites in the ACT and compared 

numbers and abundances of native, exotic and rare species. One new species of in the Tomoceridae was found at an ungrazed T. 

triandra site. Results showed a high degree of congruence with assessments of biodiversity based on vascular plants but not with 

ants and only partially with carabid beetles. Sites with high disturbance and weed invasion consistently had low Collembola 

diversity. T. triandra and Austrostipa sites had distinct faunas. Abundance of exotic species correlated with the amount of bare 

ground, as did the abundance of the rare native species Australotomurus sp., probably corresponding with their high temperature 

threshold for activity. 

Sharp (1997) analysed invertebates collected from soil samples at grassland sites. Species representing 22 orders were found, 

dominated numerically by Acarina (mites), Collembola (springtails) and Coleoptera (beetles). Abundance and order richness of 

soil invertebrates were highest at sites dominated byT. triandra (rather than Austrodanthonia), at sites with darker wet-soil 

colour and at sites managed by mowing, and lowest in grazed sites, and was not significantly related to floristic association. 

Farrow (1999) sampled canopy-living insects by sweep netting in 11 of the most important ACT grasslands in January and 

November 1998 and February 1999. He found representatives of 8 orders and 48 families and ‘super groupings’, and 

approximately 328 morphospecies including approximately 150 micro-Hymenoptera but excluding Lepidoptera, all Diptera 

except Tephritidae and some minor orders. Apart from micro-Hymenoptera, four families each were represented by >10 

morphospecies: Chrysomelidae and Curculionidae (Coleoptera), Cicadellidae (Hemiptera) and Acrididae (Orthoptera), and 

Hemiptera were most speciose with 73 spp., followed by Coleoptera with 58 spp. and Orthoptera with 18 spp. Cicadellidae spp. 

were by far the most abundant family comprising 77% of the total individuals in summer1998 and 37% in summer 1999. Next in 

abundance was Acrididae, followed by micro-Hymenoptera, Lathrididae (Coleoptera), Tephritidae (Diptera) and Alydidae 

(Hemiptera). 34-40% of species were detected at only one location, and further 16-21% at two locations, confirming that most 

species are rare. There was no evidence of a total biodiveristy difference between habitats dominated by forbs and grasses. Insect 

predators were relatively uncommon. Spiders outnumbered insects in summer 1999. Farrow (2006) resurveyed the same sites 

and four additional ones in 1999, 2000 and 2001. He found representatives of 57 families and ‘super groupings’ and an estimated 

383 species, with similar family representation to the previous study. 

Conservation of grassland invertebrates 

A 1984 survey of Australian entomologists identifed Austrostipa, Austrodanthonia and T. triandra grasslands among the broad 

habitat types that were poorly conserved from the invertebrate viewpoint (Hill and Michaelis 1988). Indentified taxa of 

conservation significance with many species associated with these grasses included Synemon (Lepidoptera: Castniidae), 

Pterolocera (Lepidoptera: Anthelidae), Hesperiidae (Lepidoptera) and Acridoidea (Orthoptera) (Hill and Michaelis 1988). 

Small reserves can be sufficient for the conservation of grassland invertebrates (Key 1978, Hill and Michaelis 1988). In alpine 

grasslands, management regimes are a threat to dayfling Geometridae (Lepidoptera) (Hill and Michaelis 1988, McQuillan 1999). 

Trends apparent from sampling of mown vs. unmown grasslands in the ACT indicate that mown areas have fewer individual 

insects and that insects may aggregrate in unmown areas (Farrow 1999). 

Five relatively well known invertebrate taxa of conservation significance found in grasslands are discussed in more detail below: 

the xanthorhoinine geometrid moths (Lepidoptera), the sun moths, Synemon spp. (Lepidoptera: Castniidae), the morabine 

grasshoppers (Orthoptera) , the cricket Coorabooraama canberrae Rentz (Orthoptera) and the Perunga grasshopper Perunga 

ochracea (Sjöstedt). Two grassland invertebrate species are officially declared threatened species in the ACT, Synemon plana 

(endangered) and P. ochracea (vulnerable) (ACT Government 2005). Other species with notable conservation significance in the 

ACT include Lewis’s laxabilla Laxabilla smaragdina Sjöstedt (Orthoptera) and Whisker’s Springtail Tomocereridae new genus 

undescribed sp. (Yen 1995, Anonymous 1997). 

157

Other grassland insects of conservation significance include Tropiderus childreni Gray (Phasmatodea: Bacteriidae), SA, Vic, 

NSW, savannah, mallee, grass, reportedly threatened by urbanisation in the Adelaide Hills (Hill and Michaelis 1988); Anisynta 

albovenata Waterhouse (Lepidoptera: Hesperiidae), SA, WA, NSW, associated with Austrostipa, reportedly threatened by 

roadworks, clearing and overgrazing (Hill and Michaelis 1988); Oriexenica kershawi kanunda Tindale (Lepidoptera: Satyriinae) 

of SA and Vic , associated with open grassland, reportedly threatened by fire at Canunda National Park (Hill and Michaelis 

1988); Trapezites lutea (Tepper) (Lepidoptera: Hesperiidae) of SA and Vic found on Lomandra dura threatened by fire and 

drought in SA (Hill and Michaelis 1988). 

Xathorhoini, Larentiinae, Geometridae 

McQuillan (1999) reported on studies on the Xanthorhoini, a tribe of small, colourful, largely diurnal Geometridae (Lepidoptera) 

with high diversity and endemicity in Australian grasslands. Members of the subfamily Larentiinae are “numerous of the 

tablelands and mountains of sout-eastern Australia and Tasmania” (Common 1990 p. 375). The tribe is represented in Australia 

by about 20 spp. of Xanthorhoe Hűbner, mostly found in the south (Common 1990, but only 13 spp. listed by Nielsen et al. 

1996), many species of Chrysolarentia Butler, three species of Acodia Rosenstock, one of Austrocidaria Dugdale, eight of 

Epyaxa Meyrick and two of Visiana Swinhoe (Nielsen et al. 1996). Adults of many Chrysolarentia spp. are “common in summer 

at the higher altitudes on the tablelands and mountains” of southern NSW, Victoria and Tasmania (Common 1990, as Euphyia 

Hűbner). Larval Xanthorhoini are sluggish and nocturnal and most species appear to be stenophagous on a small number of lowgrowing 

annual native herbs, including species of Hydrocotyle, Geranium and Acaena (McQuillan 1999). McFarland (1988) 

recorded details of the habitats and diets of X. actinipha (Lower) and X. vicissata (Guenée) which both consumed Medicago 

polymorpha L. in captivity. No Australian Geometridae are known to consume grasses or any other monocots (McFarland 1988). 

The studies of McQuillan (1999), in Tasmanian grasslands mainly dominated by Poa labillardieri or other Poa spp., Themeda 

triandra, exotic grasses, Gymnoschoenus sphaerocephalus (R.Br.) (Cyperaceae) or other Cyperaceae and Restionaceae, found 

that species richness increased with altitude, rainfall and greater sedge-content, and declined in areas with greater grass 

dominance, livestock grazing and weediness. Site species richness was lowest in highly modified grasslands and reached a 

maximum of 15 in unmodified grasslands and sedgelands, and 25 spp. were detected overall. Many species survived in small 

remnant grassland patches. Some species required grass tussocks for shelter in the adult stage and used tussocks as pheromonal 

‘calling’ and mating sites. 

Synemon spp., Sun Moths (Lepidoptera: Castniidae) 

The Australian Castniidae or sun-moths consist of 20 (Douglas 2003b), 22 (Douglas and Marriott 2003) or 24 (Douglas 2000, 

Edwards 1996 1997a) named species and 22 (Douglas 2003b) or possibly 21 or more unnamed species of the endemic genus 

Synemon Doubleday (Edwards 1997a, Douglas and Marriott 2003). Douglas (2003a) provided a brief bibliography for the 

family. Adults normally fly only in bright sunshine (Edwards 1997a, Douglas 2003b), hence the common name, and resemble 

butterflies, having broad wings, clubbed antennae and usually bright colours (Douglas 2000). Females have very long, 

retractable ovipositors and deposit eggs underground between foodplant and soil or on the base of the tillers of the foodplant 

(Common and Edwards 1981, Edwards 1994, Dunn 1996, Douglas 2000). There is pronounced sexual dimorphism of adults, e.g. 

the the sexes of S. plana Walker were described as separate species in 1854 and 1874 and their identity was not recognised until 

1926 (Edwards 1994). 

The larvae of different species feed inside the rhizomes of Lomandra spp. (Xanthorrhoaeaceae), Ecdeiocolea sp. 

(Ecdeiocoleaceae), in the tillers and then against the rhizomes of Lepidosperma viscidum R.Br. (Cyperaceae) or in tunnels 

entirely undergound, on the roots of grasses (Poaceae) or other sedges (Cyperaceae) (Common and Edwards 1981, Edwards 

1994, Dunn 1996, Douglas 2000, Douglas and Marriott 2003). Douglas (1999) also mentions unnamed Juncaceae as larval food 

plants. Most species are found in areas with sandy or light textured soils (Edwards 1994). Final instar larvae pupate in their 

feeding gallery and the pupa works its way upwards to protrude from the ground or the foodplant to enable adult emergence 

(Edwards 1994, Douglas 2000). The whole life cycle occupies 1-3 years (Common and Edwards 1981, Edwards 1994, Douglas 

2000, Endersby and Koehler 2006) so the adults present in any one year may represent only a small proportion of the population 

(Edwards 1994). All species are localised in their occurrence and the adult flight period is brief (Dunn 1996), “as short as a 

fortnight” (Edwards 1997b). Adults are usually present in only small areas (hundreds of square metres) (Dear 1996). 

Determination of what plant is actually eaten is a vexed question for some species. Edwards (1994) stated that digging the roots 

to find pupae “is the surest method … Finding protruding pupal shells within and between tussocks is the next most reliable”, 

female oviposition probing is unreliable as a foodplant indicator, although local distribution of adults, particularly females, is 

indicative. None of these methods establish actual feeding relationships. Nevertheless the appear to be the basis for the food 

plant relationships recorded, until recently with little criticism, in the literature. Determination of the actual food of the larvae is 

difficult and requires a clear protocol to indicate the various levels of uncertainty inolved (Edwards 1997a) 

The literature records species of Austrodanthonia as known or probable foodplants for five species found in Victoria including A. 

setacea (R.Br.) H.P. Linder and A. laevis (Vickery) H.P. Linder (Marriott 2004), while A. carphoides and A. auriculata are also 

probable food plants (Dear 1996) as well as Austrostipa sp. (Marriott 2004). Food plants of Queensland Synemon spp. include 

the grasses Chrysopogon sp. and Thellungia advena (Edwards 1997a). 

Recently Braby and Dunford (2006) identified N. neesiana and red-leg grass Bothrichloa macra as probable larval food plants 

for the Golden Sun Moth S. plana on the basis of the location and distribution of empty pupal cases (protruding from or beside 

tussocks) in ACT grasslands. Despite Edwards’ reliance on this method himself (e.g. Edwards 1994) he disputes this attribution 

(pers. comm. October 2006) believing it more likely that the moth has bred on undetected Austrodanthonia or in plants no longer 

visible above ground at the site, the juvenile stages being prolonged. More recently Gilmore et al. (2008) reported female 

oviposition on N. neesiana and pupal cases amongst a dense N. neesiana sward at Greenvale, Victoria. Gilmore et al. (2008) also 

reported oviposition on Austrostipa spp. and Microlaena stipoides. Edwards (1994) suggested that S. plana larvae may consume 

a mixture of roots, rhizomes and culm bases. 

S. plana is one of the few invertebrate taxa that have been used as flagship taxa for grassland invertebrate conservation in 

Australia (New 2000) and is ranked as critically endangered in Australia (Endersby and Koehler 2006, Gilmore et al. 2008). It 

158

was once widespread in south- eastern Australia but in 1993 there were only three known populations recorded since 1950, one 

in the ACT and two in Victoria (Edwards 1994). Edwards (1994) investigated eight additional likely sites in Canberra and one in 

NSW and found populations at all except one in central Canberra. Recent surveys have found it at 43 sites in NSW and 12 in the 

ACT (Endersby and Koehler 2006). The South Australian distribution is based on a single individual from Bordertown (Edwards 

1994) and it has not been found recently in that State (O’Dwyer 2004). Endersby and Koehler (2006 citing C. O’Dwyer pers. 

comm.) stated that: “Prior to 2003, the species had been reported from just six areas in Victoria – Broadford, Tallarook, 

Flowerdale, Dunkeld, Hamilton and near Nhill-Salisbury”. However that is a highly misleading statement that should have been 

qualified, perhaps as ‘probably extant populations’. Edwards’ (1994 pp. 32-33) map and list of historical records showed many 

more localities with much wider distribution. The map provided by O’Dwyer(1999) apparently leaves out historical records 

noted by Edwards (1994) including Ararat, Bendigo, Castlemaine, Gisbourne, Maryborough, Monbulk, Nagambie and Woodend. 

The map provided by O’Dwyer (2004) is more complete, showing c. 24 distribution localities prior to 1970 and c. 7 sites after 

that time. Gilmore et al. (2008) stated that there were only four records in the Melbourne area prior to 2003, near Keilor, 

Broadmeadows and Laverton. Recent records in Victoria include grasslands in the Deer Park, Craigieburn and Epping areas 

(Endersby and Koehler 2006, Gilmore et al. 2008), at Greenvale and Woodlands Historic Park (personal observations, Gilmore 

et al. 2008) and Moyhu (personal observations). 

Edwards (1994) considered S. plana to be restricted to Austrodanthonia patches in native grasslands, mostly on low hills or rises, 

sometimes in grazed and mown areas. He stated (p.34): “The foodplant is known to be Danthonia”. Driscoll (1994) stated that 

the moth requires Austrodanthonia carphoides and A. auriculata grassland for survival. According to O’Dwyer (1999) and 

O’Dwyer and Attiwill (1999) S. plana inhabits native grasslands and grassy woodlands with >40% Austrodanthonia cover with 

some sites having up to 75% cover, on soils ranging from sands, through loams to clays, with available P below 14μg/g. 

O’Dwyer (2004) states that its habitat is native grasslands dominated by Austrodanthonia spp. particularly including also A. 

eriantha and A. setacea. She noted that oviposition by S. plana has not been observed. At Craigieburn and Epping, Victoria, it 

has recently been found in habitat previously considered atypical, dominated by T. triandra (Endersby and Koehler 2006). 

The flight period of S. plana is variable from place to place and year to year (Driscoll 1994), possibly late October peaking in 

mid November and early December and through to at least the end of January in ACT (Edwards 1994), mainly 11 am – 2 pm 

(Edwards 1994). According to Driscoll (1994) the likely duration of the larval stage is 21-22 months. O’Dwyer (2004 p. 3) 

suggests “about 11 months”. Edwards (1994) mentioned the absence of parasite records, and recorded some bird predators of 

adults. O’Dwyer (2004) mentioned Willie Wagtail Rhipidura leucophrys and robber fly (Asilidae) predation of adults. 

Much of the literature on S. plana repeats unproved suppositions or speculation as facts, usually including Austrodanthonia spp. 

being food plants and native grassland habitat specificity, and fails even to provide full lists of known sites of occurrence. A 

large number of research projects on the species have failed to establish its most basic life history, including hosts plants and the 

duration of the larval stage. The most recent information (e.g. Braby and Dunford 2006, Endersby and Koehler 2006, Gilmore et 

al. 2008) indicate it is a widespread and common species that thrives in degraded pastures, including areas dominated by N. 

neesiana. The development of knowledge of this species is instructive. It has never been, as claimed by New (2000) a species 

“appraised reasonably fully” and its use as a flagship taxon has been of little value even for its own preservation, since so much 

of its basic life cycle remains poorly understood. 

S. selene has been considered a particularly significant taxon because it exists in five distinct parthenogenetic forms (Douglas 

2000); however it has not been listed under Victoria’s Flora and Fauna Guarantee Act (Department of Sustainability and 

Environment 2009a) (Table 18). 

The Cryptic Sun Moth S. theresa Doubleday is morphologically similar to other Synemon species in the grass-feeding group and 

it has been suggested that it too feeds on Austrodanthonia (Douglas 2003b). No existing populations are known but locality data 

on old specimen labels indicate it was an inhabitant of open grassy woodland with Austrodanthonia and Austrostipa in the 

understorey (Douglas 2003b). It is listed as a threatened species under the Victorian Flora and Fauna Guarantee Act 

(Department of Sustainability and Environment 2009a). 

The Orange Sun Moth S. nais Klug occurs in the Mallee in Victoria and in South Australia and Western Australia. In the Mallee 

it is known from “a floristically diverse combination of open grassy areas interspersed with stands of trees and shrubs”, the 

grassy areas dominated by Austrodanthonia setacea and Austrostipa spp. (Douglas 2003b p. 7). Oviposition behaviour suggests 

that A. setacea and an unidentified Austrostipa sp. may be larval hosts (Douglas 2003b). It is listed as a threatened species under 

the Victorian Flora and Fauna Guarantee Act (Department of Sustainability and Environment 2009a). 

Only a single population of the Striated Sun Moth Synemon sp. aff. collecta is known in Victoria, from near Shelley. Its habitat 

may have originally been open grassy woodland and grassland dominated by Austrodanthonia spp. (Douglas 2003b). 

Edwards (1996) noted that the “grass feeding species in particular have sufferred drastic reductions in distribution from the use 

of grasslands for agriculture”; urban development threatens populations of some species and invasion by introduced weeds has 

been identified as a threat to S. plana: “Without grazing or mowing the low-growing natives can become shaded and eventually 

choked out by taller exotic plants” (Edwards 1993 p. 17, Edwards 1994). In western Victoria this species survives in sheepgrazed 

Austrodanthonia pastures in which fertilisers and pesticides are not used (Dear 1996). According to Douglas and Marriott 

(2003 p. 90): “Any disturbance, through ploughing and other types of cultivation and/or excessive invasion of exotic grasses and 

forbs” leads to the disappearance of Synemon spp. Exotic perennial and annual grasses, not including Nassella spp., are listed as 

threats by O’Dwyer (2004). Cultivation in Two Wells area of South Australia was blamed for the extinction of S. selene in South 

Australia (Douglas 2000). These opinions should be viewed with some scepticism since they are based almost solely on 

observational correlations. S. plana continues to exist in areas of the ACT subject to light grazing or mowing, and the 

underground larval stage probably confers resistance to fire (Driscoll 1994). Douglas (2000) suggested that moderate disturbance 

from grazing, slashing or possibly fire appears to be necessary for S. selene to flourish (Douglas 2000). Douglas and Marriott 

(2003) advocated weed control and regular mowing, burning or grazing to remove accumulated dead grass “that provides cover 

for predators and reduces the extent of acceptable sites for oviposition”. 

159

Table 18. Conservation status of south eastern Australian Synemon taxa. Status: CE = critically endangered, E = endangered, S = 

secure, V = vulnerable, X = extinct; Other symbols: – = not present. References: 1. Marriott 2003; 2. Douglas and Marriott 2003; 

3. Endersby and Koehler 2006; 4. O’Dwyer 1999; 5. Edwards 1994, 6. Sharp 1997; 7. O’Dwyer 2004; 8. Douglas 2003b; 9. 

Dept. of Sustainability and Environment 2009a. 

Species and form 

Status 

Aust 

Status 

ACT 

Status 

NSW 

Status SA 

Status 

Vic 

Habitat 

selene Klug – – E E/CE native perennial grassland y 1 2 

selene Klug pale morph – – E native perennial grassland y 1 

selene Klug dark morph – – E native perennial grassland y 1 

selene Klug Nhill morph – – CE native perennial grassland y 1 

selene Klug narrow-winged morph – – CE native perennial grassland y 1 

selene Klug Terrick Terrick morph – – E native perennial grassland y 1 

plana Walker CE E E X? T/E grassland y 1 3 4 5 6 7 

9 

nais Klug – – possibly 

widespread 

collecta Swinhoe (=? near collecta 

and sp. aff. collecta) 

T/E 

Belah/Callitris wodland and 

grassland; Walpeup, 

Sealake 

– prob S – CE open grassy woodland and 

grassland?; 1 site Vic: 

Shelley nr Corryong, many 

New England 

Grassland 

sp.? 

Refs. 

y 1 3 8 9 

y? 1 3 8 

theresa Doubleday – – E T/X grassy woodland?; 

n? 1 2 3 8 9 

Castlemaine, no extant 

populations know 

jcaria R. Felder – ? ? T/V woodland, mallee 

n 1 8 

heathland; Kiata, Big Desert 

parthenoides R. Felder S heathland n 1 

discalis Strand – – ? CE mallee, heathland. Big n 1 8 

Desert, Hattah 

undescribed sp. – V? – – 1 site Kosciusko National 

Park 

? 3 

Driscoll (1994) listed elucidation of the lifecycle and food plants as a research priority for S. plana. This remains to be achieved. 

Recent data (Braby and Dunford 2006) suggest it could be polyphagous on suitable Poaceae, or may be a narrow oligophage that 

has evolved new preferences. In grazed situations, eggs, larvae and perching adults of S. plana may selectively survive on or 

near tussocks of N. neesiana because they are taller and bulkier than Austrodanthonia tussocks, so are less liable to trampling 

damage, and because they are not grazed by livestock in the reproductive phase, which corresponds in time with flight and 

oviposition periods of the moth. Possibly increased survival rates of all life stages on N. neesiana under grazing and declining 

abundance of Austrodanthonia spp. may have combined to create selection pressure for host shifting. Any genetic change could 

persist more readily S. plana, with its very localised populations with restricted gene flow, than in more mobile insects. The 

several forms of S. selene suggest that S. plana may have adequate genetic variation on which selection could act. A similar host 

shift from native to exotic plant, demonstrated to be genetically based, occurred in the butterfly Euphydryas editha in pastures in 

Nevada USA. Larval survival was much greater on the alien plant (Cox 2004). Many other Lepidoptera species have adapted to 

use exotic plants (e.g. Shapiro 2002). 

In summary, the literature on S. plana contains many incorrect statements, assumptions taken to be facts etc. S. plana appears to 

be much more widespread, with less specific habitat requirements than previously assumed (Endersby and Koehler 2006). Larval 

feeding has never been observed and foodplant information is based on surmise. 

Morabine grasshoppers 

Morabinae (Orthoptera: Eumastacidae), known as matchstick grasshoppers (Rentz 1996), are a primitive, endemic Australian 

subfamily of about 250-260 species and 40 genera, of extraordinary biological interest as subjects for studies of genetic drift, 

cytology, chromosomal polymorphism, karyotype evolution and parapatric speciation (Key 1973 1976 1978 1981, Rentz 1996). 

Key (1976) provided a bibliography of this literature. They are wingless, mostly very small, slender, fragile, inconspicuous 

grasshoppers resembling matchsticks, with subtle colouration, mainly active at night (Rentz 1996). External morphology within 

the subfamily shows very little variation except for male cerci (Key 1981). Morabinae had probably evolved by the Cretaceous, 

preceding the appearance of Poaceae and a few species are known to consume ferns, assumably a primitive character, although 

diversification in a number of groups is associated with widespread grass genera (Key 1976). Genitalia dissection of males or 

examination of the karyotype is the best means of identification (Key 1981, Rentz 1996). 39 chromosomal fusions due to 

translocations and 22 dissociations due to chromosome breakage have been identified in the subfamily, making most species 

chromosomally unique (Hartl and Clark 1989 presumably from White 1968 1968). Congeneric species and races of morabines 

frequently exhibit parapatric distributions with no obvious mechanisms of pre-mating isolation. Detailed studies have 

demonstrated the extreme difficulty of determining whether these populations consist of races or species and have resulted in 

reappraisal and clarification of these taxonomic concepts (Key 1981).The species and races appear to be able to mate freely but 

produce no viable offspring, because of chromosomal incompatibility, so effectively “each population prevents the other from 

invading its territory by breeding with it” (Key 1981 p. 432).; or the offspring have (often little) reduced fertility; or interbreed 

occurs despite radical chromosome differences (Key 1981). 

Many morabines live on trees or shrubs but some on grasses and forbs, and many are monophagous (Rentz 1996) and dependent 

on specific features of plants for shelter (Key 1978). Like other groups that exhibit parapatric speciation they have low vagility, 

160

eing sluggish and wingless, and exist, or once existed, in numerous small sub-populations (Key 1978). Conservation of many of 

the remnant populations as possible is desirable because of their large genetic and chromosomal variation. 

Three species of Morabinae are known from areas of grassland in south-eastern Australia. Vandiemenella viatica (Erichson), the 

first morabine grasshopper described, occurs from south-eastern Tasmania and King Island through south and west Gippsland 

and western Victoria to Kangaroo Island, and has 19 and 17 chromosome races, the former present in Victoria and Tasmania 

(Key 1976 1981). Vandiemenella species are dicotyledon feeders (Key 1976). The genus has been the subject of the most 

comprehensive studies of parapatry in the Morabinae. Twelve species or races have been distinguished based on karyotype, male 

cercus and genitalia characters, and size relationships (Key 1981 p. 433). Natural hybrids between the chromosome race of V. 

viatica with an estimated fertility depression of only c. 10%, occur in the contact zone in South Australia. Hybridisation 

experiments with the various Vandiemenella races and species have found no evidence that offspring males are completely 

sterile (Key 1981). 

Keyacris Rehn contains c. 10 species that occur on forbs and low shrubs in south-eastern Australia through southern South 

Australia to south-east coastal Western Australia (Key 1976). Keyacris scurra (Rehn), Key’s matchstick, is found on the 

tablelands of southern NSW and the ACT (Key 1981, Rowell and Crawford 1995) in grassland and grassy woodland (Driscoll 

1994), and was considered by Key (1978) to be the second most threatened morabine. New (2000) listed it as one of few flagship 

invertebrate taxa for Australian temperate grassland. It was previously found over a large area of south-eastern NSW, the ACT 

and Victoria but its range has severely contracted due to destruction of habitat, including grazing and other disturbances (Rowell 

and Crawford 1995). It is a small species about 25 mm long with form and colouring that makes it difficult to distinguish from 

dry grass and litter (Driscoll 1994). It is “dependent for its survival” upon stands of T. triandra and associated forbs, is 

eliminated, “along with its habitat”, under intensive sheep grazing, and is now largely restricted to country cemetry reserves 

(Key 1978 p. 7), fenced areas along railway lines, and some paddocks only ever grazed by cattle or horses (Key 1981). Rowell 

and Crawford (1995) surveyed 700 ha of potentially suitable habitat in the northern ACT in 1994 and found populations at 7 sites 

with widely varying management histories, with a total area of 25 ha, and along the Queanbeyan to Williamsdale rail easement 

in adjacent areas of NSW. Colonies occupied areas from 0.1 ha to 20 ha. They also reported on an unconfirmed site in Namadgi 

National Park in the southern ACT. Three populations were new records, and two previously known populations were found to 

be extinct. Its presence was correlated with that of 14 uncommon to endangered plants, also threatened by grazing and with 

Delma impar. Sharp and Shorthouse (1996) mentioned a then current study to determine the effects of vegetation composition 

and past management on populations. Farrow (1999) recorded its presence at one ACT grassland. 

K. scurra is a winter grasshopper with a single generation per year, hatching in February and reaching maturity by May (males) 

or in spring. It has very low fecundity: a maximum of 21 eggs was produced by a female in a laboratory colony. The eggs are 

deposited shallowly in surface soil (Rowell and Crawford 1995). 

T. triandra is used for shelter and perennial Asteraceae “usually a species of Helichrysum” are eaten (Key 1981 p. 432). 

“Helichrysum” is an old portmanteau name, subsequently split into numerous genera. Driscoll (1994) stated that the preferred 

food plants are Chrysocephalum semipapposum and C. apiculatum, the former also being used for shelter. The actual diet may be 

much more diverse, since individuals ate a wide range of native and exotic herbs and shrubs when offerred no other food (Rowell 

andCrawford 1995, citing Blackith and Blackith 1966). Rowell and Crawford (1995) distinguished three inhabited plant 

assemblages: Chrysocephalum semipapposum woodland and wet and dry T. triandra grasslands sometimes with derived origins. 

Dense T. triandra was not suitable habitat. Other native daisies in the tribe Inulae were suspected to be the food plant in areas 

where Chrysocephalum was absent. 

K. scurra has two races, with 15 and 17 chromosomes respectively, which are believed to have once been in contact over a 

distance ofabout 200 km, but by the late 1950s were much depleted, although a 16 chromosome hybid individual, with an 

estimated reduction in fertility of 10% was detected in a relict of the contact zone. An artificial field population consisting of 

males of one race and females of the other survived for at least four generations (Key 1981). Rowell and Crawford (1995) 

provided additional details of populations, colouration, behaviour and habitat of K. scurra and discussed threatening factors. 

These include exotic grass invasion (namely Nassella trichotoma, Anthoxanthum odoratum, Eragrostis curvula and Holcus 

lanatus), increased cover of T. triandra, grazing and changed fire regimes. 

Achurimima Key contains 17 species found in the herbaceous stratum or on low shrubs, very irregularly in the southern two 

thirds of inland Australia (Key 1976). Achurimima sp. P42, appears to be the most threatened morabine, recorded from only 6 

localities in south-eastern NSW “where small populations survive on lightly grazed native pastures including the composite 

Helichrysum semipapposum” (Key 1978 p. 18) (now Chrysocephalum semipapposum). Driscoll (1994) noted that efforts by the 

NSW National Parks and Wildlife Service to reserve an area where this species was present along with K. scurra were frustrated 

by an uncooperative landholder. 

Replacement of forb-rich T. triandra grassland by Austrostipa-Austrodanthonia pasture due to sheep grazing on the southern 

tablelands of NSW produced “a complete change in the grasshopper fauna”, including the loss of two species which became 

restricted to cemetries (Key 1978). 

Canberra raspy cricket Coorabooraama canberrae Rentz 

C. canberrae (Orthoptera: Gryllacrididae) is a very large cricket, with a body about 40 mm long and antennae of twice this 

length, recorded only from urban areas in Canberra. The genus is monospecific (Rentz 1996). No populations were known in 

1994 (Driscoll 1994). It probably lives under bark and litter and is found in suburban gardens (Driscoll 1994). Rentz (1996) 

considered it likely to be a grassland species. Virtually nothing is known about its biology. Farrow (1999) collected an 

unidentifed first instar gryllacridid at the Majura grassland in the ACT and noted that adults are known to occur there. Rentz 

(1996) provided a colour photograph of a male. Gryllacridids are predatory and often have highly specific food preferences. 

They all stridulate and are nocturnal, spending the day in burrows or leaf shelters they construct using silk from the mouthparts 

(Rentz 1996). 

161

Perunga grasshopper Perunga ochracea (Sjöstedt) 

The Perunga grasshopper Perunga ochracea (Sjöstedt) (Acrididae: Catantopinae) is a medium sized, flightless (Farrow 1999), 

almost wingless grasshopper with a narrow distribution in the ACT and neighbouring areas of NSW (Rentz et al. 2003). Farrow 

(1999) found the species at 6 of 11 grasslands surveyed by sweep netting in the ACT, but only in spring, and considered it to be 

unusual amongst the grassland Orthoptera in being winter-spring active. It occurs in T. triandra (Farrow 1999), Austrostipa and 

Austrodanthonia grasslands, feeds on forbs, overwinters as a nymph and is present as adults during spring and summer (Rentz et 

al. 2003). In the ACT it has disappeared from areas where it was once common, possibly as a result of “encroachment of dense 

cover of introduced grasses” (Rentz et al. 2003). P. ochracea is officially declared vulnerable in the ACT (ACT Government 

2005). 

Lewis’s Laxabilla, Laxabilla smaragdina Sjöstedt 

Lewis’s Laxabilla, Laxabilla smaragdina Sjöstedt (Acrididae: Oxyinae) is a small grasshopper with wingless females and fully 

winged or brachypterous males, found in grasslands and “open savannah” from southern NSW to Mackay, Queensland (Rentz et 

al. 2003). Farrow (1999) noted that it had not been recorded in the ACT for 20 years. 

Impact of N. neesiana on invertebrates 

Weed invasion can eliminate native host plants and may enhance the spread of exotic invertebrates (Yen 1995). 

Ens (2002a) conducted the only study to date of the effects of N. neesiana on invertebrates. She studied two endangered 

ecological communities in New South Wales: the edge of remnant Cumberland Plain Woodland (grassy woodland) at St Clair 

and much altered Sydney Coastal River-flat Forest (a coastal swamp forest) at Mt Annan. Pitfall trap and vacuum sampling were 

undertaken to enable comparison of areas dominated by N. neesiana and relatively devoid of native ground cover species, and 

native areas relatively free of N. neesiana. Sites had similar distrubance history, geology, topography and proximity to water. 

Point quadrats were assessed to quantify basal cover of N. neesiana, other exotic plants, native plants, bare ground, Eucalyptus 

litter, grass litter and sticks. Tree canopy cover was estimated using charts. Vegetation community structure was assessed by a 

point-height method with 4 height classes. Temperature and light were also measured above and below foliage, as well as 

distance to the nearest tree. Ens (op. cit.) reported significant quantitative impact, with a negative effect of N. neesiana on 

Formicidae and 3 Formicidae spp., reportedly “by altering the ground cover composition”, and on mean abundance of 

Thysanoptera and Cicadidae moults, but a beneficial effect (“significant habitat”) on Blattodea and two unidentified Coleoptera 

spp. Abundance of Collembola, Hemiptera, Gastropoda, Lepidoptera larvae and Araneae was significantly reduced in invaded 

areas. These results were attributed to the altered habitat structure and “change in plant architecture” i.e. the scale, complexity 

and heterogeneity of plants in the invaded community, and “indirect effects on the trophic heirarchy”. Ens (2005) summarised 

her results as reduced ant abundance and alteration of “the entire invertebrate community composition”. 

However the higher proportion of bare ground in the native vegetation explained the effects on Formicidae at one site and an 

increased cover of Eucalyptus bark at the other, neither necessarily related to N. neesiana effects. The effects on one ant species 

was best explained by the higher weed richness in the native vegetation. Multiple regression analysis failed to reveal a sensible 

cause for decreased Thysanoptera or inreased Blattodea. The abundance of cicada moults was explained by Eucalyptus bark 

cover, suggesting the native plots were closer to trees, the roots of which some Cicadidae nymphs feed upon, however the 

reduction in bark cover was attributed as an effect of N. neesiana (Ens 2002a p. 67) and there was no correlation with the 

variable ‘distance to nearest tree’. Some cicadas are grass feeders, so this may be a host plant influence. Identification of the 

species would have helped resolve this question. A number of correlations between environmental variables or higher taxa and 

other taxa with significantly different abundance in the N. neesiana areas do not make much biological sense e.g. Hemiptera 

were more abundant in greater litter depths of the native areas (?protection from predation), gastropods were more abundant in 

areas with more sticks (?protected from predation), Araneae with abundance of larvae (which they don’t consume) but not larvae 

with abundance of Araneae, but these perhaps await fuller explanation. No trophic cascades or indirect effects on the trophic 

heirarchy are clear. No attempt was made to distinguish exotic and native invertebrates, pest or beneficial species, or widespread 

versus rare taxa and no trophic links to N. neesiana were identified. 

The main effects were attributed to “changes in habitat parameters, cascade effects to higher trophic levels, changes to 

invertebrate community structure … decreases [in] ground temperature and ground incident light ... [and a] thick layer of foliage 

10-20 cm above ground when a thick monoculture” (Ens 2005). These however were correlations and may not represent true 

causative relationships. Dense growth of T. triandra appears likely to produce a very similar set of effects and alter community 

structure in a similar way. 

Grassland restoration 

Restoration of degraded, weed-invaded grasslands is difficult. Corbin et al. (2004) advocated an integrated approach utilising all 

available tools, including traditional weed management techniques, fire, grazing, and reduction of soil N availability, along with 

measures that increase the abundance of native seeds and seedlings. Reintroduction of the disturbance regimes to which the 

systems are adapted is usually the first step in restoration (MacDougall and Turkington 2007). But the original functioning of the 

system may not be properly understood, the disturbances may function in different ways because the degraded systems differ 

from the original ones and there may be substantial risks, especially related to small species populations, and costs (MacDougall 

and Turkington 2007). “The more degraded the site, the less likely the recovery by native species other than those already 

present. Supplemental measures targeting dispersal and the survival of juveniles are needed in conjunction with treatments that 

reduce the disturbance-sensitive competitive dominants” (MacDougall andTurkington 2007 p. 270). 

Preservation of native grassland remnants has been both politically and ecologically challenging. Simultaneous limited 

understanding of ecology. A critical turning point where most of the saveable remnants have been saved or their value 

appreciated and are being managed conservatively to preserve or enhance their biodiveristy. Future challenge of widespread 

162

estoration and, if Australia follows the European path, of extensifying agricultural land back to native grassland to achieve 

biodiversity and ecosystem services goals. 

Biomass reduction techniques – fire, grazing, cutting and raking, manual removal – shown to have similar effects in removal of 

exotic grasses (MacDougall and Turkington 2007). 

Nutrient reduction technqiues There is good evidence that the ability of perennial C 4 (such as T. triandra) grasses to outcompete 

C 3 grasses (including Nassella spp.) is enhanced under low soil N conditions (Badgery et al. 2002). 

Reintroduction of species 

MacDougall and Turkington (2007) compared the effects of annual cutting and raking, fire and hand weeding of the dominant 

invasive exotics Poa pratensis and Dactylis glomerata, in the restoration of invaded Garry Oak (Quercus garryana) savannah in 

British Columbia, unburnt for decades. All treatments significantly increased light levels at the ground surface and the amount of 

bare ground, reduced the growth and reproduction of the invasive grasses, and increased native plant cover and flowering. After 

four years of summer treatments the cover of P. pratensis and D. glomerata was reduced on average to

Determining cause and effect in relation to N.neesiana invasion is complicated by this range of compounded factors. Changes in 

community composition and biodiversity coinciding with invasion by an alien grass may be the result of the anthropogenic 

disturbances, rather than the invader. This situation is exemplified by the natural grasslands and subsequent ‘invaded’ pastures of 

Australian and North America (see Mack 1989), where it is not possible in the historical context to separate the effects of 

introduced grazing animals from that of the invasive plants (Woods 1997). Distinguishing between direct effects of disturbance 

on native plants and the competitive effects of invasive species has generally been difficult, as has determination of the 

interaction between disturbance and the invader (McIntyre 1993). Morgan (1998c p. 153) stressed that “conservation of many 

perennial native plants of species-rich grasslands would appear to be critically dependent on the conditions that maintain the 

standing [native] flora (i.e. low levels of canopy shading by the dominant grasses); that maintain flowering ... and that maintain 

opportunities for occasional successful recruitment, i.e. large canopy gaps”. The extent to which N. neesiana invasions prejudice 

these objectives are a critical issue in restricting its biodiversity impact. 

N. neesiana responds positively to anthropogenic disturbance. It appears to invade when there is nutrient enrichment with N and 

P, particularly the former, a condition that occurs when the major nutrient pool of the grassland system, contained in the roots 

and crowns of the dominant grasses, is mobilised by the death of these grasses. Exogenous nutrient inputs of diverse kinds, 

including water enrichment, may also be driving invasions. Although the dominant native Poaceae generally survive fire, 

burning appears to promote exotic annual weeds, and possibly releases smaller labile nutrient pools which N. neesiana may 

utilise in initial colonisation. Burning, grazing and herbicide application can all create bare ground which N. neesiana seedlings 

require to establish. Best practice mangement of T. triandra grasslands in Victoria generally includes regular burning to reduce 

T. triandra cover. Areas that are not burned rapidly loose their vascular plant diversity through loss of forbs, which generally 

have very small short-lived soil seed banks. N. neesiana invades T. triandra grasslands that are not burnt regularly or otherwise 

managed to reduce their biomass. T. triandra becomes senescent and dies under such conditions, resulting in liberation of the 

major nutrient stores in the crowns and roots. Other types of native grassland in south-eastern Australia are usually managed by 

low intensity livestock grazing to reduce grass biomass and maintain plant diversity. Periodical harvest of the livestock removes 

nutrients from the system and is probably a significant factor in protecting them from invasion. 

Native vascular plant diversity, overwhelmingly comprised of species that grow in the inter-tussocks spaces, also responds 

positively to disturbance. Regular biomass reduction is required to maintain open areas in T. triandra grasslands. The 

germination requirements of inter-tussock forbs are poorly understood and current poor recruitment levels are probably linked to 

the loss of vertebrate biodiversity. It is not clear whether the disturbances that facilititate invasion of N. neesiana and other 

weeds are of the same magnitude and type that are required to maintain existing native plant diversity,and it is not clear whether 

N. neesiana invasion is a cause of plant biodiversity loss, or a consequence of it. 

N. neesiana seed appears to be widely and commonly dispersed by human agencies. Propagule pressure is a pre-requisite for any 

plant invasion. Seed dispersal processes and rates are difficult to significantly reduce in the highly developed landscapes in 

which most remnant grasslands exist. All remnants are located in the cultural steppe or urban contexts and most have high 

perimeter : area ratios, allowing propagule pressure to be more or less relentless while the grass remains uncontrolled on 

roadsides, in pastures and in neglected areas. Protection of native grassland therefore requires greater emphasis on post-dispersal 

processes and minimisation of significant disturbance. 

Occupation of a site by an exotic perennial grass can effectively be permanent if there is no managment intervention. Heavily 

invaded, high nutrient grasslands constitute a metastable state that is very difficult to shift. Extensification of native pasture and 

more highly developed grassland tracts, and the restoration of native grasslands from degraded enriched grasslands require a 

similar set of techniques to reduce their nutrient levels and establish a large range of native plants. These techniques are poorly 

developed in Australia and have limited effectiveness. 

Given the trajectories of these factors, what is the prognosis for the impact of N. neesiana on biodiversity? Previous studies of 

invasive grass impacts on biodiversity in Australia have correlated presence of the grass with reduced native vascular plant 

diversity (e.g. McArdle et al. 2004) and alterations in invertebrate communities (Ens 2002a), but have generally failed to 

demonstrate cause and effect relatioships: ‘matched’ invaded and uninvaded areas are selected, compared, found to be similar, 

and the grass is ultimately assumed to be the cause of the impacts detected. Prediction of impact requires an understanding of the 

causes and mechanisms of invasion, many of which are undoubtedly disturbance-related, and require historical ecological 

understanding of the area invaded. The prognosis is continued invasion, as management failures accumulate across the grassland 

realm, to the climatic limit. New biiotic constraints on N. neesiana will develop as predators, parasites and competing plants 

evolve. There is some evidence that N.neesiana is used by native organisms and has begun to accumulate predators. General 

considerations suggest that whatever native invertebrates utilise Austrostipa spp. may have some preadaptations that enable them 

to exploit N. neesiana or to host shift on to it, because of the close generic relationship of the hosts. 

N. neesiana is expected to support a small fauna of generalist phytophagous insects and may according to biotic resistance theory 

be favoured over a native grasses by some of the polyphagous generalist invertebrate herbivores. Native herbivores mammals 

should favour N. neesiana over native grasses. Where it displaces native grasses, specialist grass feeders with a narrow host 

range can be expected to be displaced. What this fauna might be is largely unknown. Detritovores dependent on grass litter may 

be little affected unless there are major differences in the nutrient content or indigestible components. N. neesiana has a different 

seasonal growth pattern which may restrict its utilisation by grass-feeding insects that have lifecycles synchronised with the 

native grasses. Little specificity is expected for insects with larvae that feed on grass roots, and a number of these probably 

exploit N. neesiana. Post dispersal seed feeders may be little affected or advantaged due to the similarilty of N. neesiana and 

Austrostipa seeds and seeding patterns. Some specialist Austrostipa feeders may be able to shift hosts to N. neesiana given the 

close relationship between the genera. The balance of evolved and adaptive changes might be expected to reduce the dominance 

of N. neesiana, but it too will evolve. Some evolved changes may happen rapidly and reduce the significance of N. neesiana as a 

weed, but predicting outcomes is far beyond our capabilities. 

There are many similarities in the biology and ecology of N. neesiana and dominant native caespitose grasses it displaces, and 

some differences. Both N. neesiana and T. triandra resprout post-fire and retain dead leaf material that promotes fire frequency. 

Both have large, awned seeds, probably adapted for dispersal over longer distances mainly by animals, but which usually fall 

164

close to the plant are able to bury themselves in the ground. Both are drought adapted. T. triandra has a C 4 photosynthetic 

pathway, while the subdominant native grasses and N. neesiana are C 3 species. N. neesiana can form dense closed swards, with 

high tussock densities, as does T. triandra. 

One factor that may contribute to the superior competitive abilities of N. neesiana is its early-mid spring growth peak, which 

coincides better with periods of high soil moisture than the late-spring-early summer T. triandra. In years when rainfall is 

limiting to native grassland growth, and that may be most years, established N. neesiana presumably depletes soil moisture pools 

that would otherwise be available for use by native forbs and by the later growing T. triandra. A similar process has been 

demonstated in competition between seedlings of Bromus tectorum and native perennial grasses in the USA (Evans and Young 

1972). Earlier growth may also enable preemption of any soil nutrient pools that form during autumn and winter. These two 

impacts on the physical environment reinforce the benefits for N. neesiana: lower success of the dominant native grass means 

more resources for N. neesiana in the next growing season. Cool season native grasses may thereby be advantaged. 

N. neesiana reportedly excludes all other species (Kirkpatrick et al. 1995), but any long-lived grass may be able to exclude other 

plants from the areas it occupies, i.e. effectively hold its ground under metastable management regimes, unless its competitors 

have large advantages. Distel et al. (2008) suggested that dominance of unpalatable grasses under livestock grazing is a stable 

vegetation state in central Argentine grasslands, caused by continual grazing pressure against palatable species, low seed banks 

of palatable species and low availability of safe germination sites for the palatable species. There, successful establishment of 

native perennial grasses requires adequate soil moisture in autumn and winter, and the replacement of dominant unpalatable 

casespitose species by palatable species requires their destruction, e.g. by disc ploughing, otherwise they are “impervious to 

invasion” (Distel et al. 2008). 

Native grassland at any density and cover of the dominant native grass is resistant to N. neesiana invasion because N. neesiana 

seed germination requires more sunlight than is present in dense swards and seedling survival requires a soil nutrient pool not 

available unless existing vegetation is killed. Similarly, Barger et al. (2003) found that native Trachypogon plumosus Nees 

grassland in Brazil, in the absence of soil disturbance and external fertiliser addition, was resistant to invasion of Melinis 

minutiflora. 

Reducing the impact of N. neesiana on biodiversity in grasslands can be achieved by: 

1. Maintaining cover of the dominant grass Themeda triandra which is able to resistant invasion (Lunt and Morgan 2002, 

Hocking 2005b). 

2. Eliminating disturbance that kills native grassland plants, especially the dominant grasses. (e.g. better off not to spray weeds) 

including vehicle traffic 

3. Creation of larger buffer zones around native grassland, whether or not weed invaded and managing the weeds within that 

zone so as to minimise propagule pressure on the grassland. 

3. Burning T. triandra grasslands in late spring-early summer (after most forbs have flowered and fruited and before the main 

growing period of T. triandra) so that bare ground is not created at a time when it can best be occupied by N. neesiana seedlings. 

Hypotheses 

1. Soil disturbance that kills dominant native grasses enables N. neesiana invasion. 

2. N. neesiana reduces angiosperm diversity 

3. N. neesiana supports a greater abundance and diversity of polyphagous native invertebrate phytophages than Themeda 

triandra and other dominant native grasses. 

N. neesiana may passively occupy voids created by disturbance. 

165

Literature Review Appendix 

Appendix L1. Grass-feeding invertebrates of south-eastern Australian temperate grasslands 

This is a partial compendium of published records of invertebrates occurring or thought likely to occur in south-eastern Australian lowland grasslands that are known to feed on grasses or particular 

grass species found in these grasslands, plus a limited number of extra-Australian records of invertebrates known to feed on such grasses. Such taxa , except for Nematoda are listed in Table A2.1. 

Nematodes of grasses are recorded in Table A2.2 and are discussed in more detail below. When an invertebrate taxon (other than nematodes) with recorded grass host is known not to be present in 

south-eastern Australia the invertebrate is not included. Taxa where the recorded hosts are not known from temperate south-eastern Australia are also excluded (e.g. sugarcane). In many instances these 

distributions could not be determined, so many taxa are probably listed that may not occur in temperate grasslands. The vast literature has been incompletely surveyed and there are certainly major 

gaps. The tabulation provides the name of the invertebrate taxon, the life stage which feeds on grass, the known host plants, the host tissue eaten, the reference source, and miscellaneous notes for 

various entries. An entry is included if both the plant and the invertebrate are recorded from south-eastern Australia, unless otherwise stated (in the ‘Notes’ column). Records for Nassella spp. are 

highlighted in bold. * = introduced sp. 

Table A2.1. Literature records of invertebrates and their grass hosts recorded in south-eastern Australia, excluding Nematoda. 

Species 

Life Family Order Host Host tissue References Notes 

stage 

Halotydeus destructor (Tucker) all Penthaleidae Acarina grasses Gregg 1997 seedlings most vulnerable 

Penthaleus major (Dugès) all Penthaleidae Acarina grasses Gregg 1997 readily damages grasses, seedling most vulnerable 

Nala lividipes (Dufour) 

nymphs, 

adults 

Labiduridae Dermaptera *maize, *sorghum, *winter cereals seed, roots, 

stubble 

Allsopp & Hitchcock 

1987 

“usually feeds on decaying stubble but also eats 

newly-sown and germinating seed and the roots of 

crops” (Allsopp & Hitchcock 1987 p. 86) 

grasshoppers all Orthoptera alpine grasses foliage Carr & Turner 1959 “extremely abundant in some years and in severaly 

infested areas they shorten the sward 

considerably“(Carr & Turner 1959 p. 19) 

all Acrididae Orthoptera grasses foliage Rentz et al. 2003 most species “can be raised on grass”: see Rentz et 

al. (2003) for details of individual acridid spp. 

Teleogryllus commodus (Walker) 

Gryllotalpa spp. 

*Chirothrips ah Girault 

*Chirothrips atricorpus Girault 

*Chirothrips frontalis Williams 

*Chirothrips manicatus Haliday 

Table A2.1 (continued) 

nymphs, 

adults 

nymphs, 

adults 

larva, 

adult 

larva, 

adult 

larva, 

adult 

larva, 

adult 

Gryllidae Orthoptera grasses and forbs, mainly damages 

grass foliage 

seeds, 

leaves, 

crowns 

Browning 1954, 

Heath 1968, Allsopp 

& Hitchcock 1987, 

Panetta et al. 1993, 

Gregg 1997 

“pasture seed loss – the most serious effect” 

(Allsopp & Hitchcock 1987 p. 81). Feeds on the 

crowns of grasses (Heath 1968, New Zealand). 

“The main source of food ... is green grass” 

(Browning 1954). Seedling predator. 

*Nassella neesiana seeds Slay 2001 hollows out fallen seed in New Zealand (Slay 

Gryllotalpidae Orthoptera *cereals, *sugarcane, lawns roots Allsopp & Hitchcock 

1987 

Thripidae Thysanoptera grasses, Austrostipa aristiglumis flowers Mound & Palmer 

1972 

Thripidae Thysanoptera *Sorghum, *Melinis repens flowers Mound & Palmer 

1972 

Thripidae Thysanoptera grasses flowers Mound & Palmer 

1972 

Thripidae Thysanoptera grass flowers Mound & Palmer 

1972 

2001) 

breed only in the flowers of grasses 




166

Species 

*Chirothrips mexicanus Crawford 

Life 

stage 

larva, 

adult 

Family Order Host Host tissue References Notes 

Thripidae Thysanoptera grasses, *Avena barbata, 

Leptochloa fusca 

flowers 

Mound & Palmer 

1972 


Odontothripiella 

Odontothripiella aloba Pitkin 

Odontothripiella buloba Pitkin 

Odontothripiella compta Pitkin 

Odontothripiella reedi Pitkin 

Odontothripiella unidentata Pitkin 

Caliothrips striatopterus (Kobus) 

Haplothrips 

Haplothrips anceps Hood 

Haplothrips angustus Hood 

Haplothrips gowdeyi (Franklin) 

Haplothrips froggatti Hood 

larva, 

adult 

larva, 

adult 

larva, 

adult 

larva, 

adult 

larva, 

adult 

larva, 

adult 

larva, 

adult 

larva, 

adult 

larva, 

adult 

larva, 

adult 

larva, 

adult 

larva, 

adult 

Thripidae Thysanoptera grasses flowers Pitkin 1972, Mound 

& Palmer 1972, 

Mound & Heming 

1991 

Thripidae Thysanoptera grasses? Pitkin 1972 

Thripidae Thysanoptera grasses Pitkin 1972 

Thripidae Thysanoptera grasses, Austrostipa aristiglumis. flowers Pitkin 1972 “lives in grass flowers” 

A. stuposa, Themeda triandra 

Thripidae Thysanoptera grasses, Themeda triandra. Pitkin 1972 

Thripidae Thysanoptera grasses, reeds, *barley grass, 

Themeda triandra, *Hordeum 

leporinum, *Hordeum vulgare 

flowers Pitkin 1972 “feeds in the flowers of grasses” 

Thripidae Thysanoptera grasses flowers Mound & Palmer “common in grass flowers in Queensland” 

1972 

Phlaeothripidae Thysanoptera grasses flowers Mound & Heming “particularly on grasses” (Mound & Heming 1991) 

1991 

Phlaeothripidae Thysanoptera grasses, Themeda triandra flowers Pitkin 1973 “grass-flower living ... apparently confined to 

Phlaeothripidae Thysanoptera grasses, *Pennisetum 

clandestinum, Paspalidium sp., 

*Secale cereale, “Danthonia 

linkii”, Themeda triandra, 

*Hordeum leporinum 

Queensland” 

flowers Pitkin 1973 “grass-flower living ... apparently widespread” 

(Pitkin 1973) 

Phlaeothripidae Thysanoptera grasses, Melinis repens flowers Pitkin 1973 “apparently breeds on grasses” 

Phlaeothripidae Thysanoptera grasses, Leptochloa digitata, 

Leptochloa fusca, Sorghum sp., 

Paspalidium sp., *Triticum 

aestivum, *Zea mays 

flowers 

Mound & Palmer 

1972, Pitkin 1973 

“common in flowers of Gramineae” (Mound & 

Palmer 1972); “grass-flower living ... apparently 

widespread” (Pitkin 1973) 

Haplothrips pallescens (Hood) Phlaeothripidae Thysanoptera grasses, Themeda triandra, Melinis 

Pitkin 1973 

found on grasses, possibly predatory 

repens 

Antillothrips cingulatus (Hood) Phlaeothripidae Thysanoptera grasses, *Oryza sativa. Pitkin 1973 found on grasses, possibly predatory 

Podothrips xanthopus Hood Phlaeothripidae Thysanoptera grasses, Themeda triandra Pitkin 1973 found on grasses, probably predatory 

Apterygothrips australis Pitkin Phlaeothripidae Thysanoptera grasses, “Danthonia linkii”, 

“Stipa” sp., Paspalidium sp., 

Sporobolus virginicus 

Pitkin 1973 

“apparently widespread in southern Australia”, 

possibly predatory 

Desmothrips reedi Mound 

Desmothrips tenuicornis (Bagnall) 

*Limothrips 

Phibalothrips spp. 

larva, 

adult 

larva, 

adult 

larva, 

adult 

larva, 

adult 

Aeolothripidae Thysanoptera grasses flowers Mound & Palmer 

1972 

Aeolothripidae Thysanoptera grasses flowers Mound & Palmer 

1972 

Thripidae Thysanoptera grasses leaves Mound & Heming 

1991 

Thripidae Thysanoptera grasses Mound & Heming 

1991 

“associated with grass flowers” 

“associated with grass flowers” 

“in cooler southern areas ... common on grasses ... 

often on leaves” 

167


Species 


stage 

Stenchaetothrips biformis (Bagnall) larva, Thripidae Thysanoptera *Oryza sativa Mound & Heming “found on grasses in moist habitats in Queensland” 

adult 

1991 

Aptinothrips rufus (Haliday) 

larva, Thripidae Thysanoptera Mound & Heming “common on grasses” 

adult 

1991 

Phibalothrips longiceps (Karny) larva, Thripidae Thysanoptera grasses leaves Mound & Palmer “feed on leaf tissue” 

adult 

1972 

Species related to Anapothrips Uzel Thripidae Thysanoptera grasses leaves Mound & Palmer “feed on leaf tissue” 

1972 

Leptocorisa acuta Thunberg, L. 

Alydidae Hemiptera grass - Gross 1991b “common on grasses” 

oratorius (Fabricius) 

Pemphiginae Aphididae Hemiptera grasses roots Gair et al. 1983, 

Carver 1991 

*Metopolophium dirhodum all Aphididae Hemiptera *cereals, grasses Carver 1991 

*Rhopalosiphon insertum (Walk.) all Aphididae Hemiptera *oat,*wheat, *barley, *rye; grasses roots Gair et al. 1983, anholocyclic on grass roots (Carver 1991) 

Carver 1991 

*Rhopalosiphum maidis (Fitch) all Aphididae Hemiptera *oat,*wheat, *barley, *rye, *maize, 

*cereals, grasses 

leaves Gair et al. 1983, 

Carver 1991 

*Rhopalosiphum padi (L.) all Aphididae Hemiptera *oat,*wheat, *barley, *rye, 

*cereals, grasses 

leaves Gair et al. 1983, 

Carver 1991 

*Rhopalosiphum rufiabdominalis 

Aphididae Hemiptera grasses, *rice roots Carver 1991 anholocyclic on grass roots 

(Sasaki) 

*Sitobion spp. Aphididae Hemiptera *cereals, grasses Carver 1991 

Arawa pulchra Knight - Cicadellidae Hemiptera grass Day & Fletcher 1994 

*(?) Arawa taedius (Kirkaldy) - Cicadellidae Hemiptera collected on grasses Evans 1966 probably synonymous with A. pulchra (Day & 

Fletcher 1994) 

Balclutha incisa (Matsumura) - Cicadellidae Hemiptera Poaceae Day & Fletcher 1994 plus hosts in several other families 

Balclutha punctata (Fabricius) - Cicadellidae Hemiptera mainly Poaceae Day & Fletcher 1994 

Balclutha rosea (Scott) - Cicadellidae Hemiptera mainly Poaceae Day & Fletcher 1994 

Chiasmus varicolor (Kirkaldy) - Cicadellidae Hemiptera collected on grass Evans 1966 

Deltocephalinae - Cicadellidae Hemiptera grasses Fletcher & Semeraro 

2001 

Deltocephalinae: Macrostelini - Cicadellidae Hemiptera grasses Evans 1966 “many species feed on grasses” 

Exitianus spp. - Cicadellidae Hemiptera grasses Evans 1966 

Hecalinae: Hecalini - Cicadellidae Hemiptera grasses Evans 1966 “probably that all species are grass-feeders” 

Day & Fletcher 1994 no Hecalini spp. known from NSW, Vic, SA, Tas 

*Nephotettix apicalis Motschulsky - Cicadellidae Hemiptera collected on grasses Evans 1966 not listed as an Australian sp. by Day & Fletcher 

1994 

Mircrolopa minuta Evans - Cicadellidae Hemiptera collected on grasses Evans 1966 

Nesoclutha pallida (Evans) - Cicadellidae Hemiptera collected on grasses Evans 1966 

- *cereals, *maize, Chloris, 

Day & Fletcher 1994 vector of viruses that attack these plants 

*Paspalum 

Pardorydium menalus Kirkaldy - Cicadellidae Hemiptera collected on grasses Evans 1966 

Recilia hospes (Kirkaldy) - Cicadellidae Hemiptera *Cynodon dactylon, Digitaria 

Day & Fletcher 1994 other hosts: sedges, low shrubs 

henryi 

* Stirellus fatigandus (Kirkaldy) - Cicadellidae Hemiptera collected on grass Evans 1966 

Togacephala vetus (Knight) - Cicadellidae Hemiptera grass Day & Fletcher 1994 plus other monocots and dicots 

168


Species 


stage 

Cicadetta waterhousei (Distant) nymph Cicadidae Hemiptera grass roots Moulds 1990 habitat: “grasses of several species; usually long 

and partly or completely browned ... In South 

Australia ... often found on on dried-out seedbearing 

...Avena spp.” 

Diemeniana euronotiana (Kirkaldy) nymph Cicadidae Hemiptera grasses roots Moulds 1990 habitat: “at low altitudes on grass ...” 

Diemeniana frenchi (Distant) nymph Cicadidae Hemiptera grasses roots Moulds 1990 habitat: “Grass, usually growing along river and 

creek flats” 

Oliarius nymph Cixiidae Hemiptera grass roots Gross 1983 

Eriococcus Coccidae Hemiptera “Danthonia” Boucek 1988 New Zealand record; ?both plant and insect in SE 

Aust. 

Symonicoccus Coccidae Hemiptera grasses Williams1991 ?both plant and insect in SE Aust. 

Phaenacantha australiae Kirkaldy Colbathristidae Hemiptera ?grasses, *sugar cane Gross 1991a “occurs in large numbers on grasses in coastal Qld 

and may be a pest of sugar cane” ; ?both plant and 

insect in SE Aust. 

Cyminae? - Lygaeidae Hemiptera ?grasses seeds? Slater 1991 as Kosmioplex varicolor Kirkaldy 

Blissinae? - Lygaeidae Hemiptera grasses sap Slater 1991 subfamily occurs “most commonly on grasses” 

Opistholepis spp. - Lygaeidae Hemiptera grasses Slater 1976 ? in SE Australia 

Stenophlegyas spp. - Lygaeidae Hemiptera grasses Slater 1976 ? in SE Australia 

Stenophyella macreta Horváth - Lygaeidae Hemiptera grasses Slater 1976 1991 “often common in seed heads” (Slater 1991); 

“appears to feed on a number of grasses even when 

they are completely dry” (Slater 1976 p. 135) 

Anzac bipunctatus (Fabricius) - Membracidae Hemiptera collected on grasses Evans 1966 

Euciodes suturalis Pascoe larva Anthribidae Coleoptera *Arrhenatherum elatius, *Bromus 

sp., *Dactylis glomarata, *Festuca 

stems Penman 1978, May 

1994, Kuschel 1972, 

New Zealand host records, no published Australian 

data (Zimmerman 1994a) 

arundinacea,*Holcus lanatus 

Holloway 1982 

Hispellinus 

larva 

adult 

Chrysomelidae Coleoptera Themeda leaves Jolivet & 

Hawkeswood 1995 

Hispellinus australicus 

larva Chrysomelidae Coleoptera grasses leaves Matthews &Reid larva is a leafminer 

(Motschulsky) 

2002 

Oulemma spp. larva Chrysomelidae Coleoptera native and introduced grasses leaves Matthews &Reid 

2002 

Rupilia spp. ? larva Chrysomelidae Coleoptera grasses prob leaves Matthews &Reid 

2002 

hosts unclear. Eulalia in Central Australia 

Steriphus caudata (Pascoe) larva Curculionidae Coleoptera *cereals and pastures seeds, 

seedling 

stems, tiller 

bases 

*Graphognathus leucoloma 

(Boheman) 

larva, 

adult 


1987 

Curculionidae Coleoptera wild grasses, *maize Allsopp & Hitchcock 

1987 

in cereals “larvae eat out the swelling seeds .. bore 

into the underground part of the stem ... tillers may 

be affected without the plant dying” (Allsopp & 

Hitchcock 1987 p. 60). Wide host range outside 

Poaceae. 

“Adult weevils chew the foliage of most plants but 

rarely cause economic damage” (Allsopp & 

Hitchcock 1987 p. 51); wide host range outside 

Poaceae 

*Listronotus bonariensis (Kuschel) larva Curculionidae Coleoptera *Zea mays, *Lolium stem-miner May 1994 New Zealand 

Cubicorhynchus spp. 

larva Curculionidae Coleoptera native and introduced grasses roots, stems Zimmerman 1993 

adult 

Cubicorhynchus calcaratus Macleay larva Curculionidae Coleoptera ?Austrostipa Zimmerman 1993 found in a clump of “Stipa” 

adult 

Cubicorhynchus crenicollis 

Waterhouse 

larva Curculionidae Coleoptera unidentifed grass May 1994 

169


Species 


stage 

Cubicorhynchus sordidus Ferguson Curculionidae Coleoptera *Nassella trichotoma and “some roots, stems Zimmerman 1993 

other grasses” 

Cubicorhynchus taurus Blackburn larva Curculionidae Coleoptera Austrostipa nitida, A. nodosa, 

Zimmerman 1993, possibly restricted to semiarid grasslands 

Enneapogon nigricans, Eragrostis 

eriopoda, unidentified grasses 

May 1994 

*Floresianus sordidus Hustache Curculionidae Coleoptera Poaceae May 1994 ?insect found in SE Aust. 

*Linogeraeus urbanus (Boheman) larva 

adult 

Curculionidae Coleoptera *Paspalum distichum roots, 

stolons 

May 1994 

as Paspalum paspaloides (sic); ?insect found in SE 

Aust. 

Maleuterpes spinipes Blackburn larva Curculionidae Coleoptera grass roots May 1994 main hosts Rutaceae; ?insect found in SE Aust. 

Phalidura abnormis (Macleay) larva 

adult 

Curculionidae Coleoptera *Nassella trichotoma Zimmerman 1993, 

May 1994 

Phalidura elongata (Macleay) larva Curculionidae Coleoptera *Nassella trichotoma and “other 

grasses” 

underground 

parts 

Zimmerman 1993, 

May 1994 

Sclerorinus sp. Curculionidae Coleoptera “Stipa sp.” May 1994 

*Sphenophorus brunnipennis 

(Germar) 

Curculionidae Coleoptera *Pennisetum clandestinum and 

*“bent grass” 

basal 

nodes, 

crowns, 

stolons 

May 1994 

?insect found in SE Aust. 

- larva Elateridae Coleoptera grasses etc. Gregg 1997 damage seedlings 

Agrypnus and Hapatesus spp., larva Elateridae Coleoptera *wheat, *arley, *oats, *sugar cane, 


wireworms 

*maize, *sweet corn, *sorghum 

1987, Calder 1996 

“Grasslands and pasture are the natural habitat ... 

sporadic ... pests of newly-sown cereals” (Allsopp 

& Hitchcock 1987 p. 44) 

Arachnodima bribbarensis (Calder) 

& A. xenikon (Calder) 

larva Elateridae Coleoptera *Triticum aestivum seeds, 

seedlings 

Calder1996 

larvae “attack germinating wheat seeds and 

seedlings” (Calder 1996 p. 139) 

Arachnodima ourapilla (Calder) larva Elateridae Coleoptera *Triticum aestivum Calder1996 “attacks wheat crops” (Calder 1996 p. 139) 

Arachnodima opaca Candèze larva Elateridae Coleoptera **Hordeum vulgare. Calder1996 “attacks barley crops” (Calder 1996 p. 139) 

Dicranolaius bellulus (Guérin- 

Méneville)/D. cinctus 

(Redtenbacher) 

adult Melyridae Coleoptera aquatic grasses, *rice flowers, 

seeds 

Hely 1958, Booth et 

al. 1990 

normal foods of adults are flowers of sedges, 

rushes, and aquatic grasses “of the millet type” 

(Hely 1958 p. 30) 

Mordellistena spp. larva Mordellidae Coleoptera grasses stems Booth et al. 1990 no Australian grass host records located 

Melolonthinae - various larva Scarabaeidae Coleoptera pasture grasses roots Allsopp & Hitchcock 

1987 

Melolonthinae: Scitalini - various larva Scarabaeidae Coleoptera probably grasses roots Britton 1987 

Adoryphorus couloni (Burmeister) larva Scarabaeidae Coleoptera grasses roots Blackburn 1983, Hill 

et al. 1993, 

McQuillan & Webb 

1994Gregg 1997 

“Pasture species that form a shallow-rooted turf 

(for example Yorkshire fog, barley grass and 

perennial ryegrass) are more susceptible to attack” 

(Blackburn 1983 p. 3); “reduce the productivity 

and persistence of perennial pastures” (McQuillan 

& Webb 1994 p. 49) 

Antitrogus spp. larva Scarabeidae Coleoptera grasses and other plants roots Allsopp 2003 

Aphodius tasmaniae Hope larva Scarabaeidae Coleoptera grasses foliage Allsopp & Hitchcock 

1987, Panetta et al. 

1993, Gregg 1997 

Chlorochiton sp. larva Scarabaeidae Coleoptera *Nassella trichotoma roots Lowe 1954 New Zealand sp., large patches killed in NZ 

*Cyclocephala signaticollis 

Burmeister 

larva Scarabaeidae Coleoptera pasture and lawn grasses roots Carne 1956 most abundant in Cynodon dactylon and Paspalum 

dilatatum swards (Carne 1956) 

170


Species 

*Heteronychus arator (Fabricius), 

Life 

stage 

larva 

adult 

Family Order Host Host tissue References Notes 

Scarabaeidae Coleoptera mainly damages grasses, larvae on 

matt-forming grasses incl. *kikuyu, 

*paspalum; *maize, *sweet corn, 

lawns, pastures 

roots, 

rhizomes, 

shoots, 

felled 

heads 


1987, Panetta et 

al.1993, Gregg 1997 

Saulostomus villosus 

Waterhouse 

Scitala sericans Erichson larva Scarabaeidae Coleoptera damages pastures roots Hardy 1976b, 


1987 

larvae on roots, adults on shoots or chew holes in 

stems of young plants 

larva Scarabaeidae Coleoptera damages pastures and turf roots Hardy 1976a the soil-dwelling larvae “feed on plant material 

including roots” (Hardy 1976a p. 282) 

Sericesthis consanguinea 

(Blackburn) 

larva Scarabaeidae Coleoptera *wheat, *oats roots Allsopp & Hitchcock 

1987 

Sericesthis geminata Boisduval larva Scarabaeidae Coleoptera grasses, including *Lolium perenne roots Carne & Chinnick 

1957, Wensler 1971, 

Allsopp & Hitchcok 

1987, Panetta et 

al.1993, Gregg 1997 

Sericesthis harti Sharp larva Scarabaeidae Coleoptera *cereals roots Allsopp & Hitchcock 

1987 

Sericesthis nigrolineata (Boisd.) larva Scarabaeidae Coleoptera grasses, incl. *Lolium perenne roots Ridsdill-Smith 1975, 

Hardy 1976b, 


1987, Britton 1987, 

Gregg 1997 

- larva Tenebrionidae Coleoptera grasses etc. Gregg 1997 damage seedlings 

Helea, Pterohelaeus, Gonocephalum, larvae, Tenebrionidae Coleoptera *cereals, incl. *Triticum aestivum seeds, 

Isopteron and Saragus spp. 

adults 

seedlings 

- larva Agromyzidae Diptera grasses, *rice leaves, 

stems 


1987, Matthews & 

Bouchard 2008 

Wapshere 1990, 

Colless & McAlpine 

1991, Hely 1958 

“after entering the third instar ... the larvae confine 

their feeding to the roots of grasses” (Carne & 

Chinnick 1957 p. 608); “prefer grasses over broadleaf 

plants” (Allsopp & Hitchcock 1987 p. 20); 

“feed upon young roots” (Britton 1987 p.687). 

“larvae preferentially select and feed on living 

roots” (Ridsdill-Smith 1975 p. 75); “feed upon 

young roots of grasses” (Britton 1987 p.687) 

larvae damage seed and seedlings before 

emergence, adults usually eat the seedling at 

ground level; Pterohelaeus are “major root and 

seedling pests”, Isopteron “recorded feeding on 

geminating wheat” (Matthews & Bouchard 2008) 

Europe (Wapshere 1990). Rice leaf miner (Hely 

1958). Leaf and stem miners, gall makers 

Asphondylini sp. larva Cecidomyiidae Diptera Themeda triandra spikelets McDougall 1989 undescribed sp., prob a new genus (Robin Adair, 

pers. comm. 2006) 

larva Cecidomyiidae Diptera Austrostipa spp. spikelets Yen 1999 “may ... be the same species” as on T. triandra 

(Yen 1999) 

- larva Cecidomyiidae Diptera Andropogon Boucek 1988 p. 561 gall former 

- larva Cecidomyiidae Diptera grasses stems Boucek 1988, Europe (Wapshere 1990) 

bloodworms, ?Chironomus tepperi 

Skuse 

Wapshere 1990 

larva Chironomidae Diptera *rice ?roots Hely 1958, Colless & 

McAlpine 1991 

- larva Chloropidae Diptera grasses shoots and 

stems 

Hydrellia spp. larva Ephydridae Diptera *rice, *barley, *irrigated cereals leaves, 

stems 

Colless & McAlpine 

1991 

Mathis 1989 

“failure of seedling development” (Hely 1958 p. 

31), mainly physical disturbance to roots (Colless 

& McAlpine); may similarly effect other grasses in 

submerged situations 

inside young growth; may be largely bacteriafeeders 

171


Species 


stage 

Atherigona spp., shoot flies larva Muscidae Diptera grasses, *cereals shoots Pont 1989 “primary pests” 

- larva microlepidoptera Lepidoptera grasses Wapshere 1990 Europe 

Anthela basigera (Walker) larva Anthelidae Leidoptera grasses foligage Marriott 2008 

Anthela denticulata (Newman) larva Anthelidae Leidoptera grasses,*cereals; *Triticum 

aestivum 

foliage French 1911, 

Common 1990, 

Edwards & Fairey 

Anthela euryphrica Turner larva Anthelidae Leidoptera grasses; causes periodic damage to 

crops 

1996 , Marriott 2008 

foliage Common 1990; 


1996 

Anthela ferruginosa Walker larva Anthelidae Lepidoptera grasses, *cereals foliage Common 1990 

Anthela ocellata (Walker) larva Anthelidae Lepidoptera grasses, *cereals foliage French 1911, 

Common 1990, 

Marriott 2008 

various introduced grasses, 

*Ehrharta erecta 

foliage Coupar & Coupar 

1992 

Anthela oressarcha Turner larva Anthelidae Leidoptera grass foliage Common 1990 

Anthela ostra Swinhoe larva Anthelidae Leidoptera causes periodic damage to crops foliage Edwards & Fairey 

1996 

Pterolocera spp. larva Anthelidae Lepidoptera grasses; periodic damage to native 

pastures 

“frequently found feeding on grasses in gardens” 

(Common 1990) 

? grass feeder; ?insect found in SE Aust. 

foliage Common 1990, 


1996, Marriott 2008 

larvae “accept Poa spp. and many introduced 

grasses” (Marriott 2008); also “clover and low 

growing shrubs” (Common 1990) 

Synemon plana Walker larva Castniidae Lepidoptera Austrodanthonia laevis roots Common 1990 See general discussion on Synemon spp. 

- larva Elachistidae Lepidoptera grasses leaf and 

stem 

miners 

Cosmiotes synethes (Meyrick) larva Elachistidae Lepidoptera grasses, *Bromus catharticus, 

*Triticum aestivum 


Nielsen & Common 

1991 

leaf miner Common 1990, 

Nielsen & Common 

1991 

Fraus simulans Walker larva Hepialidae Lepidoptera grasses Allsopp & Hitchcock 

1987 

Oncopera spp. larva Hepialidae Lepidoptera grasses roots, 

stems, 

leaves 

Chadwick 1966, 


1987, Common 

1990, Gregg 1997, 

Edwards 2002 

Europe, Australia 

“larva produces an elongate blotch mine” 

(Common 1990) 

feeds “mostly on grasses” 

Oncopera alboguttata Tindale larva Hepialidae Lepidoptera *Nassella trichotoma foliage Campbell 1998 killed small areas 

Oncopera alpina Tindale larva Hepialidae Lepidoptera Poa “australis”, Poa spp. 

snowgrasses 

Oncopera fasciculatus (Walker) larva Hepialidae Lepidoptera grasses and other pasture plants, 

annual grasses, *Phalaris, 

*Dactylis glomerata 

tussock 

bases and 

roots? 

foliage 

Carr & Turner 1959, 



1987, Common 

1990, 

Madge 1954, Erlich 

1980, Allsopp & 

Hitchcock 1987 

“prefer grasses, especially reyegrass” (Gregg 

1997); “Young larvae ... feed on the roots and 

leaves of young grasses” (Allsopp & Hitchcock 

1987 p. 74). Larve “feed on the bases of snow 

grass tussocks” (Edwards 2002 p. 61). Stem and 

blade-base most damaging (Chadwick 1966) 

“extensive patch death” in Australian Alps 

(McDougall &Walsh 2007); “damage to the aerial 

parts of the tussocks is slight” (Carr & Turner 1959 

p. 20). 

“eat new green pasture as it appears ... larger 

numbers ... eat both new green feed and old dry 

pasture”, mature Phalaris tussocks little damaged, 

D. glomerata tussocks killed (Madge 1954 p. 193) 

172


Species 


stage 

Oncopera intricata Walker larva Hepialidae Lepidoptera grasses foliage French 1909, 


1987 

Oncopera rufobrunnea Tindale larva Hepialidae Lepidoptera *Nassella trichotoma, grasses foliage, 

stems 

Campbell 1998, 

Gregg 1997, 

Common 1990, 


1987, Hill et al. 1993 

Oncopera tindalei Common larva Hepialidae Lepidoptera grasses foliage Allsopp & Hitchcock 

1987 

Oxycanus antipoda (Herrich- 

Schäffer) 

larva Hepialidae Lepidoptera grasses foliage Erlich 1980, 

Common 1990 

Oxycanus fuscomaculatus Walker larva Hepialidae Lepidoptera grasses foliage Allsopp & Hitchcock 

1987 

Dispar compacta (Butler) larva Hesperiidae Lepidoptera Poa tenera, Poa sp. foliage Braby 2000 

Herimosa albovenata (Waterh.) larva Hesperiidae Lepidoptera Austrostipa scabra, A. 

foliage Braby 2000 

semibarbata, ?Austrodanthonia 

setacea 

Ocybadistes flavovitta (Latreille) larva Hesperiidae Lepidoptera Poaceae foliage Braby 2000 

Ocybadistes walkeri Heron larva Hesperiidae Lepidoptera *Cynodon dactylon, *Ehrharta 

erecta, *Lolium sp, *Pennisetum 

clandestinum 


Taractrocera papyria (Boisduval) larva Hesperiidae Lepidoptera Austrodanthonia sp., Austrostipa 

scabra, *Cynodon dactylon, 

Microlaena stipoides, Poa sp., 

*Ehrharta spp., *Paspalum 

dilatatum, *Pennisestum 

clandestinum 


“the most destructive of all grass-eating grubs 

known to myself” French (1909 p. 103) 

killed small areas of N. trichotoma (Campbell 

1998); “widely distributed in native and sown 

pastures in Tasmania” (Common 1990) 

“Grasses are the favoured food” (Allsopp & 

Hitchcock 1987 p. 76) 

- larva Noctuidae Lepidoptera Themeda triandra leaves Wainer 1992 larval gut contents in Delma impar faecal pellets, 

Derrimut, Vic. 

Agrotis infusa (Boisduval) larva Noctuidae Lepidoptera grasses, *Hordeum vulgare, 


Common 1990, 

Gregg 1997 

cut through stems;”omnivorous ... all kinds of ... 

crops, fodder, and grass” (Froggatt 1910) 

*Hordeum vulgare, *Triticum 

aestivum 

Coupar & Coupar 

1992 

Agrotis munda Walker, larva Noctuidae Lepidoptera grasses, *cereals, *Hordeum 

Common 1990, cut through stems (Gregg 1997) 

vulgare, *Zea mays 

Gregg 1997 

Agrotis ypsilon (Hufnagel) larva Noctuidae Lepidoptera *Zea mays. Common 1990, cut through stems 

Gregg 1997 

Bathytricha truncata (Walker) larva Noctuidae Lepidoptera grasses incl. *Oryza sativa, *Zea 

mays, *Paspalum dilatatum, 


stems Common 1990 usually grass stem borers 

Dasygaster padockina (Le Guillou) larva Noctuidae Lepidoptera native and introduced grasses, 

Common 1990 

*cereals, *Triticum aestivum 

Helicoverpa armigera (Hűbner) larva Noctuidae Lepidoptera *Sorghum, *Zea mays L. heads Common 1990 

Helicoverpa punctigera (Walleng.) larva Noctuidae Lepidoptera *Sorghum, *Zea mays L. heads Common 1990 

Heliocheilus cramboides Guenée larva Noctuidae Lepidoptera Sorghum intrans seedhead Matthews 1999 

Heliocheilus eodora Meyrick larva Noctuidae Lepidoptera Eulalia aurea seedhead Matthews 1999 

Heliocheilus pallida Butler larva Noctuidae Lepidoptera Dichanthium tenuiculum seedhead Matthews 1999 

173


Species 


stage 

Metopiora sanguinata (Luc.) larva Noctuidae Lepidoptera grasses Common 1990 

Mocis frugalis (Fabricius) larva Noctuideae Lepidoptera grasses, *Avena sativa Common 1990 

Mythimna convecta (Walker) larva Noctuidae Lepidoptera Astrebla pectinata leaves McDonald 1991 

grasses, *cereals, *graminaceous seed heads, Common 1990, eat or sever flower heads (Goodyer 1983) 

forage crops, *Avena, *Hordeum 

(barley), *Oryza sativa, *Setaria, 

*Triticum aestivum, *Zea mays 

foliage, 

stems 

Goodyer 1983, 

Gregg 1997 

Mythimna separata (Walker) larva Noctuidae Lepidoptera grasses, *cereals, *Saccharum 

officinarum, pastures and crops 

Common 1990, 

Goodyer 1983 

Neocleptria punctifera (Walker) larva Noctuidae Lepidoptera *Triticum aestivum, *Zea mays Common 1990 highly polyphagous, many families 

Numichtis nigerrima (Guen.) larva Noctuidae Lepidoptera *Zea mays Common 1990 highly polyphagous 

Persectania dyscrita Common larva Noctuidae Lepidoptera pastures and crops, *Triticum 

aestivum 

Goodyer 1983, 

Common 1990 

Persectania ewingii (Westwood) larva Noctuidae Lepidoptera pasture grasses, primarily grasses; 

*cereals and *graminaceous forage 

crops, *Avena, *Hordeum (barley), 


seed heads, 

foliage, 

stems 

Schroder 1983, 

Goodyer 1983, 

Common 1990, 

Gregg 1997 

Proteuxoa spp. larva Noctuidae Lepidoptera native and introduced grasses foliage Common 1990 

Proteuxoa sanguinipuncta (Guen.) larva Noctuidae Lepidoptera native and introduced grasses foliage Common 1980 

Spodoptera exempta (Walker) larva Noctuidae Lepidoptera grasses, grass pastures, *cereals, 

Goodyer 1983, 

Urochloa mutica, *Sorghum, 

Common 1990 

*Paspalum dilatatum, 

*Pennisetum clandestinum 

Spodoptera exigua (Hubner) larva Noctuidae Lepidoptera pastures and crops, *Oryza sativa, 

Goodyer 1983, 

*Sorghum, *Zea mays 

Common 1990 

eat or sever flower heads (Goodyer 1983) 

U. mutica is present in NC NSW 

found in NSW (Goodyer 1983) 

Spodoptera litura (Fabricius), larva Noctuidae Lepidoptera Urochloa mutica Common 1990 Brachiaria the old name? Plant on NC NSW 

Spodoptera mauritia (Boisduval) larva Noctuidae Lepidoptera grasses, *cereals, pastures, lawns, 

*Cynodon dactylon, *Pennisetum 

clandestinum, *Sorghum 

leaves Goodyer 1983, 

Common 1990, 

Gregg 1997 

found north from N NSW (Common 1990) 

Argynnina hobartia (Westwood) larva Nymphaliidae Lepidoptera *Lolium perenne foliage Braby 2000 “probably ... native grasses such as Poa 

labillardieri and Austrodanthonia” (Braby 2000 p. 

491) 

Geitoneura acantha (Donovan) larva Nymphalidae Lepidoptera native grasses, Themeda triandra, 

Poa spp., Microlaena stipoides 

Geitoneura klugii (Guérin- 

Méneville) 

larva Nymphalidae Lepidoptera native grasses, Microlaena 

stipoides, Austrostipa flavescens, 

Joycea pallida, Poa labillardieri, 

P. morrisii, P.tenera, T. triandra, 

*Brachypodium distachyon, 

*Ehrharta calycina, *Vulpia sp. 

Heteronympha merope (Fabricius) larva Nymphalidae Lepidoptera native and introduced grasses, 

*Lolium perenne, *Cynodon 

dactylon, Microlaena stipoides, 

Poa poiformis, P. tenera, Themeda 

triandra, *Brachpodium 

distachyon, *Bromus catharticus, 

*Ehrharta erecta 

foliage 

foliage 

foliage 


1992 


1992, Braby 2000 


1992, Braby 2000 

174


Species 


stage 

Heteronympha penelope Waterh. larva Nymphalidae Lepidoptera Austrodanthonia penicillata, 

Austrodanthonia pilosa, Poa sp., 

leaves Braby 2000 butterfly not a grassland sp. Sources: Jarvis 1908, 

Tindale 1952a, Common & Waterhouse 1981 

Poa snowgrass, Themeda triandra 

Hypocysta metirius Butler larva Nymphalidae Lepidoptera Alexfloydia repens, *Cynodon leaves Braby 2000 source for C. dactylon E.O. Edwards 1948 

dactylon, 

Oriexenica ptunarra L.E. Couchman larva Nymphaliidae Lepidoptera Poa gunnii, P. labillardieri, P. foliage Braby 2000 

rodwayi 

Philobota productella (Walker), 

P. chionoptera Meyrick, 

P. diaereta Turner, P. physaula 

Meyrick 

larva Oecophoridae Lepidoptera grasses foliage, 

litter 


1987, Hill et al. 

1993, Common 1990 

plus “other herbaceous plants ... at times damage 

pastures” (Common 1990); “Foliage is cut and 

often left uneaten on the ground” (Allsopp & 

Hitchcock 1987 p. 80) 

‘Lomera’ caespitosae Oke 

(“Plutorectis” caespitosae) 

larva Psychidae Lepidoptera Poa Snow Grass, “Poa australis” leaves, leaf 

lamina just 

above the 

sheath 

Carr & Turner 1959, 


Common 1990, 

Green & Osborne 

1994, Edwards 2002 

“feeds exclusively on snow grass”, “heavily 

infested ... tussocks are completely cut through at 

the base and do not recover” (Carr & Turner 1959 

p. 19). “Extensive patch death” in Australian Alps 

(McDougall &Walsh 2007). But damage confused 

with Oncopera alpina, which does most of the 

damage. 

‘Lomera’ boisduvalii (Westwood) larva Psychidae Lepidoptera probably grasses Common 1990 

‘Lomera’ spp. larva Psychidae Lepidoptera probably grasses Common 1990 most other spp. probably feed on grasses. 

Achyra affinitalis (Led.) larva Pyralidae Lepidoptera *Sorghum, *Zea mays Common 1990 highly polyphagou – many families 

Callamatropha leptogramella larva Pyralidae Lepidoptera grasses, cereals, *Paspalum Common 1990 “minor pasture pest” 

(Meyrick) 

Conogethes punctiferalis (Guen.) larva Pyralidae Lepidoptera *Sorghum, *Zea mays heads Common 1990 occurs S to N. NSW 

Cnaphalocrocis medinalis (Guen.) larva Pyralidae Lepidoptera *Oryza sativa 

Culladia cuneiferellus (Walker) larva Pyralidae Lepidoptera grasses, *cereals, *Cynodon 

dactylon 

lawns 

Hednota spp. larva Pyralidae Lepidoptera grasses, *cereals incl. wheat, 

barley and rye, *Hordeum spp., 

*Bromus spp., *Vulpia spp. 

Hednota bivitella (Don.), 

H. crypsichroa Lower, 

H. longipalpella (Meyrick), 

H. panteucha (Meyrick), 

H. pedionoma (Meyrick), 

H. pleniferellus (Walker) 

leaves Common 1990 grass both exotic and native, sometimes damages 

foliage 


1987, Panetta et al. 

1993, Gregg 1997 

larva Pyralidae Lepidoptera grasses Allsopp & Hitchcock 

1987, Common 1990 

“minor pests of native grasses” (Gregg 1997); 

damage evident from “chopped off shoots” 

(Allsopp & Hitchcock 1987 p. 78) 

larvae in silk galleries in crowns or surface soil 

Herpetogramma licarsisalis (Walk.) larva Pyralidae Lepidoptera grasses, *cereals, *Paspalum, leaves Common 1990 lawn and pasture pest, S to N NSW 

*Pennisetum clandestinum 

Mampava rhodoneura (Turn.) larva Pyralidae Lepidoptera *Cenchrus ciliaris seed Common 1990 insect in C Qld (Common 1990 p. 348) ; ?both 

plant and insect in SE Aust. 

Marasmia venilialis (Walker) larva Pyralidae Lepidoptera grass Common 1990 larva in rolled leaf shelter 

Sclerobia trialis (Walker) larva Pyralidae Lepidoptera *Cynodon dactylon leaves Common 1990 C. dactylon is both exotic and native; “minor 

pasture pest ...also damages lawns” 

Eretmocera dioctis (Meyr.) larva Scythrididae Lepidoptera *Chloris gayana, Enteropogon 

Common 1990 insect at least in N NSW (Common 1990 p. 266) 

acicularis 

- larva Tineidae Lepidoptera grasses Common 1990 larvae of some spp. “tunnel in the soil and feed on 

grass” (Common 1990 p. 183); however true 

phytophagy is “very rare” and generally facultative 

(Robinson & Nielsen 1993 p. 25) 

175


Species 


stage 

Nemapogon granella (L.) larva Tineidae Lepidoptera *Triticum aestivum, *Hordeum seed French 1900 

vulgare 

Crocidosema plegejana Zell. larva Tortricidae Lepidoptera *Triticum aestivum. Common 1990 highly polyphagous; “sometimes damage ears of 

wheat” in NSW and SA 

Aprostocetus asperulus (Graham) larva Eulophidae Hymenoptera *Hyparrhenia hirta Boucek 1988 extralimital host record (Madiera), insect known 

from Brisbane, may be a parasite/ inquiline of 

another insect 

Eurytoma spp. larva Eurytomidae Hymenoptera grasses stems Boucek 1988 normally develop with & consume Tetramesa spp. 

Tetramesa spp. larva Eurytomidae Hymenoptera grasses seeds, 

internodes 

Boucek 1988, 


Naumann 1991 

3 Australian spp. associated with grasses 

(Naumann 1991); Wapshere: Europe, USA; pests 

of cereals elsewhere in the world 

- adult Formicidae Hymenoptera grasses seed Gregg 1997 small, light seed readily taken, larger seeds are not 

Pheidole spp. adult Formicidae Hymenoptera pasture grasses, *cereals, 

seed Allsopp & Hitchcock 

*sorghum 

1987 

“occasional pest sof newly-sown grass ... pastures 

and cereal crops, especially sorghum ... stands of 

annual grasses can also be severely reduced” 

(Allsopp & Hitchcock 1987 p. 91) 

Ormocerinae larva Pteromalidae Hymenoptera Poaceae Naumann 1991 “associated with galls especially on ... Poaceae” 

Systasis larva Pteromalidae Hymenoptera grasses seeds Naumann 1991 “associated with grass seeds” 

Systasis graminis (Cameron) larva Pteromalidae Hymenoptera Panicum sp. seeds Boucek 1988 

176

Nematodes of grasses and grasslands 

Nematodes (Phylum Nematoda) constitute a vast, poorly explored taxon and are “the most abundant multicellular animals on 

earth” (Baldwin et al. 2004 p. 84). As many as 500,000 species have been predicted to exist, of which only 10,000 were known 

in 1951 (Meglitsch 1972). Most are entirely free-living in soil, water, etc. and a comparitively small proportion are plant 

parasites (Baldwin et al. 2004). Many of the parasitic groups have free-living stages and there is a full suite of intermediate 

lifestyles from fully sedentary ecto- or endoparasitism to migratory endoparasitism (Siddiqi 1986, Baldwin et al. 2004). Many of 

the plant parasitic nematodes are ecto- or endo-parastites of grasses. Secretion products from plant parasistic species promote 

plant pathologies including nurse cells, root-knots and cysts specific to particular taxa, while some groups transmit viruses to 

their hosts (Baldwin et al. 2004). The order Tylenchida, commonly called plant nematodes, is the largest and most economically 

important group of plant parasites, and its members damage all plant organs including seeds, flowers and especially roots 

(Siddiqi 1986). However the orders Dorylaimida and Triplonchida also include plant parasites (Baldwin et al. 2004). 

Johnston (1938) recorded numerous species from unidentified grasses in Australia, along with species that attack sugar cane and 

cereals, but no records from native grasses. McLeod et al. (1994) recorded 42 taxa from unspecified grass or grasses in Australia 

and listed the nematode species found on a wide range of other grasses in Australia. 

In the Stipeae, Anguina sp. is recorded from Austrostipa drummondii (Steud.) Jacobs & Everett and A. nitida (Summerh. & C.E. 

Hubb.) Jacobs & Everett in Victoria, A. trichophylla (Benth.) Jacobs & Everett in Western Australia and from “Stipa” sp. in 

NSW, Neodolichodorus adelaidensis and Xiphinema monhystereum are recorded from “Stipa” sp. in South Australia, and 

Pratylenchus neglectus is recorded from Austrostipa trichophylla (Benth.) Jacobs & Everett in Queensland (McLeod et al. 

1994). 

No species were known from Themeda triandra in south-eastern Australia, but 19 species have been found associated with the 

plant in Queensland (McLeod et al. 1994). Radopholus intermedius Colbran was described from material taken from around the 

roots of Allocasuarina torulosa (Aiton) L. Johnson and T. triandra, and R. laevis Colbran has also been found in this situation 

(Colbran 1971). The genus Radopholus Thorne contains 22 root endoparasitic species all but one of which are indigenous to 

Australia (Siddiqi 1986). 

Table A2.2. Literature records of the hosts of grass-inhabiting Nematoda recorded in south-eastern Australia. Grasses included 

are inhabitants or probably occur in temperate south eastern Australian grasslands. South-eastern Australia excludes Queensland. 

Nematodes recorded from unspecified grass or grasses are not included. The list is indicative only, since a large proportion of the 

distribution records are not from grasslands. All data from McLeod et al. (1994). Nematode family assignment from Siddiqi 

(1986). 

Grass Nematode Family Distribution 

*Agrostis capillaris Belonolaimus sp. Dolichodoridae NSW 

Helicotylenchus dihystera Tylenchidae NSW 

*Agrostis stoloniferea Paratrichodorus lobatus ? NSW 

Paratylenchus nanus Paratylenchidae NSW 

*Avena fatua Paratylenchus sp. Paratylenchidae Vic 

*Avena sativa Ditylenchus dipsaci Anguinidae SA 

Heterodera avenae Heteroderidae NSW, SA, Vic 

Meloidogyne javanica Meloidogynidae NSW 

Paratylenchus neglectus Paratylenchidae SA, Vic 

Pratylenchus thornei Pratylenchidae SA 

Pratylenchus sp. Pratylenchidae Vic 

Scutellonema brachyurum Hoplolaimidae NSW 

Austrodanthonia caespitosa Paralongidorus sacchari ? SA 

Xiphinema monohysterum Neotylenchidae SA 

Austrodanthonia setacea Paratylenchus sp. Paratylenchidae SA 

Austrostipa nitida Anguina sp. Anguinidae Vic 

*Bromus catharticus Ditylenchus dipsaci Anguinidae NSW 

*Bromus sp. Meloidogyne sp. Meloidogynidae SA 

*Chloris gayana Meloidogyne incognita Meloidogynidae NSW 

*Cynodon dactylon Belonolaimus lolii Dolichodoridae NSW 

Criconema mutabilie Criconematidae SA 

Ditylenchus intermedius Anguinidae NSW 

Helicotylenchus dihystera Tylenchidae NSW 

Hemicriconemoides minor Criconematidae Vic 

Hemicycliophora labiata Hemicycliophoridae NSW, SA 

Hemicycliophora saueri Hemicycliophoridae Vic 

Hemicyliophora truncata Hemicycliophoridae NSW 

Heterodera graminis Heteroderidae NSW 

Heterodera sp. Heteroderidae NSW 

Macroposthonis sp. Criocnematidae Vic 

Morulaimus whitei Dolichodoridae Vic 

Paralongidorus eucalypti ? Vic 

Paratrichodorus lobatus ? NSW, SA 

Paratrichodorus minor ? NSW 

Rotylenchus brevicaudatus Hoplolaimidae NSW 

Scutellonema brachyurum Hoplolaimidae NSW, SA 

177

Tylenchorhynchus annulatus Dolichodoridae Vic 

Tylenchorhynchus clarus Dolichodoridae SA 

Xiphinema americanum Neotylenchidae SA 

Xiphinema monohysterum Neotylenchidae Vic 

*Digitaria sanguinalis Scutellonema clariceps Hoplolaimidae NSW 

*Eragrostis cilianensis Paratrichodorus lobatus ? SA 

*Holcus lanatus Anguina sp. Anguinidae SA, Vic 

Pratylenchus penetrans Pratylenchidae Vic 

*Hordeum leporinum Heterodera avenae Heteroderidae NSW, Vic 

Pratylenchus neglectus Pratylenchidae Vic 

*Hordeum marinum Pratylenchus neglectus Pratylenchidae SA 

Lachnagrostis filiformis Anguina sp. Anguinidae NSW 

*Lolium perenne Heteodera avenae Heteroderidae NSW, Vic 

Pratylenchus crenatus Pratylenchidae NSW 

Pratylenchus penetrans Pratylenchidae Vic 

Subanguina radicola Anguinidae Tas 

Xiphinema italiae Neotylenchidae NSW 

Xiphinema radicicola Neotylenchidae NSW 

*Lolium rigidum Anguina funesta Anguinidae SA 

Heterodera avenae Heteroderidae SA, Vic 

Merlinus brevidens Dolichodoridae SA 

Pratylenchus neglectus Pratylenchidae SA 

Tylenchorhynchus clarus Dolichodoridae SA 

*Lolium sp. Longidorus elongatus ? SA 

Pratylenchus neglectus Pratylenchidae SA 

Pratylenchus thornei Pratylenchidae SA 

Microlaena stipoides Anguina microlaenae Anguinidae NSW, Vic 

*Paspalum dilatatum Heterodera graminis Heteroderidae NSW 

*Pennisetum clandestinum Helicotylenchus dihytera Tylenchidae NSW 

Paratrichodorus lobatus ? NSW, SA 

Pratylenchoides leiocauda Pratylenchidae NSW 

Radopholus similis Pratylenchidae NSW 

Xiphinema americanum Neotylenchidae NSW 

Xiphinema ensicliferum Neotylenchidae NSW 

Xiphinema insigne Neotylenchidae NSW 

*Poa annua Belonolaimus lolii Dolichodoridae NSW 

Subanguina radicicola Anguinidae Tas 

Poa labillardieri Scutellonema sp. Hoplolaimidae Vic 

Poa sieberiana Hemicriconemoides minor Criconematidae Tas 

Hemicylciophora truncata Hemicycliophoridae Tas 

Poa sp. Criconema pasticum Criconematidae Tas 

Pratylenchus coffeae Pratylenchidae NSW 

*Polypogon monspeliensis Anguina sp. Anguinidae NSW, SA 

“Stipa” sp. Anguina sp. Anguinidae NSW 

Neodolichodorus adelaidensis Dolichodoridae SA 

Xiphinema monhystereum Neotylenchidae SA 

178

REFERENCES 

Aceñolaza, F.G. (2004) Paleobiogeografía de la Región Mesopotámica, In F.G. Aceñolaza (Ed.), Temas de la Biodiversidad del Litoral Fluvial 

Argentino, Instituto Superior De Correlación Geologica ( INSUGEO ) Miscelánea 12, 31 - 38 

ACT (Australian Capital Territory) Government (1997) Natural Temperate Grassland: an Endangered Ecological Community. Action Plan 1. 

Environment ACT, Canberra. 

ACT (Australian Capital Territory) Government (2005) A Vision Splendid of the Grassy Plains Extended. ACT Lowland Native Grassland 

Conservation Strategy. Action Plan 28, Environmnet ACT, Arts Heritage and Environment, Canberra. 

ACT (Australian Capital Territory) Weeds Working Group (2002) ACT Chilean Needle Grass Management Plan. Final Draft. ACT Government, 

Canberra, 11 pp. 

Adair, R.J. (1995) The threat of environmental weeds to biodiversity in Australia: a search for solutions. In R.A. Bradstock, T.D. Auld, D.A. 

Keith, R.T. Kingsford, D. Lunney and D.P. Sivertsen (Eds.), Conserving Biodiversity: Threats and Solutions. Surrey Beatty and Sons, Sydney, 

pp. 184-201. 

Adair, R.J. and Groves, R.H. (1998) Impact of Environmental Weeds on Biodiversity: A Review and Development of a Methodology. Biodiversity 

Group. Environment Australia, Canberra, iv + 51 pp. 

Aguiar, M.R. (2005) Biodiversity in grasslands. Current changes and future scenarios. In S.G. Reynolds and J. Frame (Eds.), Grasslands: 

Developments Opportunities Perspectives, United Nations Food and Agriculture Organisation and Science Publishers Inc., pp. 261-280. 

Ahmad, I. (1965) The Leptocorisinae (Heteroptera: Alydidae) of the world. Bulletin of the British Museum (Natural History) Entomology Supplement 

5, 1-156. 

Aldous, D.E., Isaacs, S. and Mebalds, M.I. (1998) Endophytes in the genus Neotyphodium are not found in Australian native grasses. Australasian 

Plant Pathology 28, 183-186. 

Allsopp, P.G. (2003) Synopsis of Antitrogus Burmeister (Coleoptera: Scarabaeidae: Melolonthini). Australian Journal of Entomology 42, 159- 

178. 

Allsopp, P.G. and Hitchcock, B.E. (1987) Soil Insects Pests in Australia: Control Alternatives to Persistent Organochlorine Insecticides. 

Australian Agricultural Council, Plant Health Committee, Canberra. 

Amiotti, N.M., Bravo, O., Zalba, P. and Peinemann, N. (2007) Effect of landscape position on the acidification of loess-derived soils under Pinus 

radiata. Austral Ecology 32(5), 534-540. 

Amme, D. (2003) Nassella notes. Grasslands(California Native Grass Association)13(4), 3 pp. 

Andersen, A.M. (1963) Germination of seed of Texas needlegrass, Stipa leucotricha. Proceedings of the Assocation of Official Seed Analysts of 

North America 53, 74-79. 

Anderson, F.E. (2002a) Progress Report on Investigation of Fungal Pathogens of Serrated Tussock (Nassella trichotoma) and Chilean Needle 

Grass (N. neesiana) July 2002-November 2002. CSIRO Entomology, CERZOS-UNS, Bahía Blanca, Argentina, 8 pp. 

Anderson, F.E. (2002b) Report on Field Trip to Collect Rust on Serrated Tussock (Nassella trichotoma) and Chilean Needle Grass (N. neesiana) 

November-December 2002. CERZOS-UNS, Bahía Blanca, Argentina, 9 pp. 

Anderson, F.E., Diaz, M.L. and McLaren, D.A. (2006) Current status of research on potential biological control agents for Nassella neesiana and 

Nassella trichotomoa (Poaceae) in Australia. In C. Preston, J.H. Watts and N.D. Crossman (Eds.), 15th Australian Weeds Conference Papers and 

Proceedings. Weed Management Society of South Australia Inc., Adelaide, pp. 591-594. 

Anderson, F.E., Flemmer, A.C., Hansen, P.V., McLaren, D.A. and Barton, J. (2008) An update on progress towards biological control of Nassella 

neesiana in Australia and New Zealand. In R.D. van Klinken, V.A. Osten, F.D. Panetta and J.C. Scanlan (Eds.), Proceedings of the 16th 

Australian Weeds Conference, Queensland Weeds Society, Brisbane, pp. 202-204. 

Anderson, F., Pettit, W., Briese, D.T. and McLaren, D.A. (2002) Biological control of serrated tussock and Chilean needlegrass. Plant Protection 

Quarterly 17(3), 104-110. 

Anderson, F., Pettit, W., Morin, L., Briese, D. and McLaren, D. (2004) Pathogens for the biological control of weedy stipoid grasses in Australia: 

completion of investigations in Argentina. In J. M. Cullen, D.T. Briese, D.J. Kriticos, W.M. Lonsdale, L. Morin and J.K. Scott (Eds.), 

Proceedings of the XI International Symposium on Biological Control of Weeds, CSIRO, Canberra, pp. 69-74. 

Anderson, H. (1963) Leather industry. The Australian Encyclopaedia Volume V, Grolier Society of Australia, Sydney, pp. 268-270. 

Andrioli, R.J. and Distel, R.A. (2008) Litter quality of C3 perennial grasses and soil inorganic nitrogen in a semiarid rangeland of central 

Argentina. Journal of Arid Environments 72, 1684-1689. 

Anonymous (1968) Edgar Tatlow. New Zealand Trotting Calendar, 13 March. Sourced from Addington Raceway Timetable, Racing History - 

People. http://www.addington.co.nz/Timeline.asp?Case=1andID=1968 [Accessed 18 July 2007] 

Anonymous (1997) Disappearing insects of native grasslands. ACT Parks and Conservation and CSIRO Entomology, Canberra, ACT. [Poster] 

Anonymous (2005) Victoria declares twelve WONS. Under Control – Pest Plant and Animal Management News No. 32, December, p. 1. 

Anonymous (2008a) Chilean needle grass alert. SPIFFA Newsletter (Southern Peninsula Indigenous Flora and Fauna Association), 18(5), 1. 

Anonymous (2008b) Balaustium mites. Blue oat mites. PestFacts South-Eastern (Grains Research and Development Corporation and CESAR 

University of Melbourne) No. 7, 1 August, pp. 2-3. 

Archer, W.R. (1984) Thesium australe R.Brown (Santalaceae) – field notes and observations. Victorian Naturalist 101(2), 81-85. 

Archer, W.R. (1987) Additional field notes and observations of Thesium australe R. Brown (Santalaceae). Victorian Naturalist 104(2), 46-49. 

ARMCANZ (Agriculture and Resource Management Council of Australia and New Zealand), ANZECC (Australian and New Zealand 

Environment and Conservation Council) and Forestry Ministers (2001). Weeds of National Significance Chilean Needle Grass (Nassella 

neesiana) Strategic Plan. National Weeds Strategy Executive Committee, Launceston. 

Arriaga, M.O. and Jacobs, S.W.L. (2006) An anatomo-ecological experiment in Austrostipa aristiglumis, a lowland stipoid species. Telopea 

11(2), 161-170. 

Arrillaga-Maffei, B.R. (1966) Anomalous stomata of Stipa neesiana (Gramineae). Canadian Journal of Botany 44(6), 845-846 + 1 pl. 

179

Ashton, A. and Morcom, L. (2004) Plump Swamp Wallaby-grass Amphibromus pithogastrus. Flora and Fuana Guarantee Action Statement No. 

109, Department of Sustainability and Environment, East Melbourne, 3 pp. 

Aitken, P.F. (1983) Mammals. In M.J.Tyler, C.R.Twidale, J.K.Ling and J.W. Holmes (Eds.), Natural History of the South East. Royal Society of 

South Australia Inc., pp. 127-133. 

Auckland Regional Council (2002) Auckland Regional Pest Management Strategy 2002-2007. Auckland Regional Council, Auckland, New 

Zealand. 

Audas, J.W. (1950) The Australian Bushland. W. A. Hamer Pty. Ltd., North Melbourne. 

Australian Weeds Commitee (2007) Noxious weed list. Weeds Australia web site. http://weeds.org.au/ [Accessed 22 June 2007] 

Australian Wool Burean (1963) Wool. The Australian Encyclopaedia Volume IX, Grolier Society of Australia, Sydney, pp. 352-359. 

Backhouse, G. and Crossthwaite, J. (2003) Eastern Barred Bandicoot Perameles gunii. Flora and Fauna Guarantee Action Statement No. 4, 

Department of Sustainability and Environment, East Melbourne, 8 pp. 

Badgery,W.B., Kemp, D.R., Michalk, D.L. and King, W.M. (2002) Competition between native grasses and serrated tussock (Nassella 

trichotoma) at low fertility – initial results. Plant Protection Quarterly 17(3), 95-98. 

Baeza, C.M., Marticorena, C., Stuessy, T., Ruiz, E. and Negritto, M. (2007) Poaceae en el archipelago de Juan Fernandez (Robinson Crusoe). 

Gayana Botánica 64(2), 125-174. 

Baker-Gabb, D. (1993) Managing native grasslands to maintain biodiversity and conserve the plains-wanderer. RAOU Conservation Statement 

No. 8, Supplement to Wingspan 10, June, 8 pp. 

Baker-Gabb, D. (2003) Plains Wanderer Pedionomus torquatus. Flora and Fauna Guarantee Action Statement No. 66, Department of 

Sustainability and Environment, East Melbourne, 6 pp. 

Baldwin, I.T., Kessler, A. and Halitschke, R. (2002) Volatile signaling in plant-plant-herbivore interactions: what is real? Current Opinion in 

Plant Biology 5, 1-4. 

Baldwin, J.G., Nadler, S.A. and Adams, B.J. (2004) Evolution of plant parasitism among nematodes. Annual Review of Phytopathology 42, 83- 

105. 

Barbosa, E.G., Pivello, V.R. and Meirelles, S.T. (2008) Allelopathic evidence in Brachiaria decumbens and its potential to invade the Brazilian 

cerrados. Brazilian Archives of Biology and Technology 51(4), 825-832. 

Barger, N.N., D’Antonio, C.M.,Ghneim, T. and Cuevas, E. (2003) Constraints to colonization and growth of the African grass, Melinis 

minutiflora, in a Venezuelan savanna. Plant Ecology 167, 31-43. 

Barker, G.M. (2008) Mollusc herbivory influenced by endophytic clavicipitaceaous fungal infection in grasses. Annals of Applied Biology 153, 

381-393. 

Barker, R.D. and Vestjens, W.J.M. (1989) The Food of Australian Birds 1. Non-Passerines. CSIRO, Melbourne. 

Barker, R.D. and Vestjens, W.J.M. (1990) The Food of Australian Birds 2. Passerines. CSIRO, Melbourne. 

Barkworth, M.E. (1990) Nassella (Gramineae, Stipeae): revised interpretation and nomenclatural changes. Taxon 39, 597-614. 

Barkworth, M.E. (1993) North American Stipeae (Gramineae): taxonomic changes and other comments. Phytologia 74(1), 1-25. 

Barkworth, M.E. (2006) Stipeae. Web site of the Intermountain Herbarium, Utah State University, Logan, Utah, USA. 

http://herbarium.usu.edu./stipeae/ (Accessed 24 September 2006, 16 July 2007). 

Barkworth, M.E., Capels, K.M., Arderton, L., Long, S. and Piep, M.B. (Eds.) (2007) Manual of Grasses for North America. Flora North America 

North of Mexico Volumes 24 and 25. Intermountain Herbarium, Utah State University http://herbarium.usu.edu/grassmanual [Accessed 17 August 

2007] 

Barkworth, M.E. and Everett, J. (1986) Evolution in the Stipeae: identification and relationships of its monophyletic taxa. In T.S. Soderstrom, 

K.W. Hilu, C.S. Campbell and M.E. Barkworth (Eds.), Grass Systematics and Evolution, Smithsonian Institution Press, Washington, USA, pp. 

251-264. 

Barkworth, M. E. and Torres, M.A. (2001) Distribution and diagnostic characters of Nassella (Poaceae: Stipeae). Taxon 50, 439-468. 

Barlow, B.A. (1981) The Australian flora: its origin and evolution. In A.S. George (Ed.), Flora of Australia Volume 1. Introduction. 1st Ed., 

Australian Government Publishing Service, Canberra, pp. 25-75. 

Barlow, T.J. and Ross, J.R. (2001) Vegetation of the Victorian volcanic plain. Proceedings of the Royal Society of Victoria 113, 25-28. 

Barrett, S.C. (1983) Crop mimicry in weeds. Economic Botany 37, 255-288. 

Bartley, M.J., Parsons, R.F. and Scarlett, N.H. (1990) Stipa neesiana (Chilean Needle Grass) threat to grassland reserves and endangered plants in 

the Melbourne region. Letter from Department of Botany, LaTrobe University, to John Taylor, Acting Regional Manager, Melbourne Region, 

Department of Conservation and Environment, 7 May, 2 pp. 

Barwick, P. (1999) Weeds and roadsides. In A.C. Bishop, M. Boersma and C.D. Barnes (Eds.), 12th Australian Weeds Conference Papers and 

Proceedings, Tasmanian Weed Society, Devenport, pp. 678-680. 

Beames, L., Hocking, C. and Wlodarcyzk, P. (2005) Best practice management of Chilean needle grass (Nassella neesiana) in conservation 

reserves – the seed bank story. Plant Protection Quarterly 20(1), 2-8. 

Beauglehole, A.C. (1983) The Distribution and Conservation of Vascular Plants in the Melbourne Area, Victoria. Western Victorian Field 

Naturalists Clubs Association, Portland. 

Beauglehole, A.C. (1987) The Distribution and Conservation of Vascular Plants in the Wimmera Area, Victoria. A.C. and H.M. Beauglehole, 

Portland. 

Bedggood,W. and Moerkerk, M. (compilers) (2002) Integrated weed management results from a workshop on Chilean needle grass management 

held at Victoria University of Technology, St Albans, Victoria, 2 October 2002. Cooperative Research Centre for Weed Management Systems, 

Adelaide, 12 pp. 

Bennett, A.F. (1995) Wetern Grey Kangaroo. In P.W. Menkhorst (Ed.), Mammals of Victoria. Distribution, Ecology and Conservation. Oxford 

University Press, Melbourne, pp. 136-138. 

Benson, D. and von Richter, L. (2009) Ecology of Cumberland Plain Woodland. 

http://www.rbgsyd.nsw.gov.au/science/current_research/Ecology_of_Cumberland_Plain_Woodland [Accessed 10 June 2009] 

Benson, J.S. (2004) The native grasslands of the Monaro region: Southern Tablelands of NSW. Cunninghamia 3(3), 609-650. 

180

Benson, D. and McDougall, L. (2005) Ecology of Sydney plant species Part 10. Monocotyledon families Lemnaceae to Zosteraceae. 

Cunninghamia 9(1), 16-212. 

Benson, J.S. and Redpath, P.A. (1997) The nature of pre-European native vegetation in south-eastern Australia: a critique of Ryan, D.G., Ryan, 

J.R. and Starr, B.J. (1995) The Australian Landscape – Observation of Explorers and Early Settlers. Cunninghamia 5(2), 285-328. 

Berg, R.Y. (1975) Myrmecochorous plants in Australia and their dispersal by ants. Australian Journal of Botany 23, 475-508. 

Berry, S. and Mulvaney, M. (1995) An Environmental Weed Survey of the Australian Capital Territory. A Report Prepared for the Conservation 

Council of the South-east Region and Canberra. August. 

Biedermann, R., Achtziger, R., Nickel, H. and Stewart, A.J.A. (2005) Conservation of grassland leafhoppers: a brief review. Journal of Insect 

Conservation 9, 229-243. 

Blackburn, M. (1983) The redheaded pasture cockchafer. Agnote 2042/83, Department of Agriculture Victoria, 3 pp. 

Blakers, M., Davies, S.J.J.F. and Reilly, P.N. (1984) The Atlas of Australian Birds. Melbourne University Press, Carlton. 

Bloomfield, T. and McPhee, S. (2006) Rabbits, foxes and feral pigs- how do they impact on weeds? Plant Protection Quarterly 21(4), 148-151. 

Boeckmann, B., Bairoch, A., Apweiler, R., Blatter, M.C., Estreicher, A., Gasteiger, E., Martin, M.J., Michoud, K., O’Donovan, C., Phan, I., Pibout, S. 

and Schneider, M. (2003) The SWISS-PROT protein knowledgebase and its supplement TrEMBL in 2003. Nucleic Acid Research 31, 365-370. 

Booth, R.G., Cox, M.L. and Madge, R.B. (1990) IEE Guides to Insects of Importance to Man 3. Coleoptera. International Institute of Entomology, 

Wallingford, UK. 

Bond, W.J., Silander, J.A., Ranaivonasy, J. and Ratsirarson, J. (2008) The antiquity of Madagascar’s grasslands and the rise of C 4 grassy biomes. 

Journal of Biogeography 35, 1743-1758. 

Boucek, Z. (1988) Australasian Chalcidoidea (Hymnoptera). A Biosystematic Revision of Genera of Fourteen Families, with a Reclassification of 

Species. CAB International, Wallingford, UK. 

Bouchenak-Khelladi, Y., Verboom, G.A., Hodkinson, T.R., Salamin, N., Francois, O., Chonghaile, G.N. and Savolainen, V. (2009) The origins and 

diversification of C 4 grasses and savanna-adapted ungulates. Global Change Biology 15, 2397-2417. 

Bourdôt, G.W. (1988) Chilean needle grass management for eradication. Guidleines for farmers. New Zealand Ministry of Agriculture and 

Fisheries Aglink FPP884, 1st Ed., 2 pp. 

Bourdôt, G.W. (1989) Letter to Dr. R.F. Parsons, Botany Department, LaTrobe University (addressed to University of Melbourne) from 

Ministryof Agriculture and Fisheries New Zealand, re Stipa neesiana, 13 November, 1 p. 

Bourdôt, G.W. and Hurrell, G.A. (1987a) Chilean needle grass (Stipa neesiana). 1. Control with herbicides in pasture. Proceedings of the 40th 

New Zealand Pest and Weed Control Conference, pp. 204-207. 

Bourdôt, G.W. and Hurrell, G.A. (1987b) Chilean needle grass (Stipa neesiana) 2. Herbicidal control in lucerne. Proceedings of the 40th New 

Zealand Pest and Weed Control Conference, pp. 208-211. 

Bourdôt, G.W. and Hurrell, G.A. (1989a) Cover of Stipa neesiana Trin. and Rupr. (Chilean needle grass) on agricultural and pastoral land near 

Lake Grassmere, Marlborough. New Zealand Journal of Botany 27, 415-420. 

Bourdôt, G.W. and Hurrell, G.A. (1989b) Ingress of Stipa neesiana Trin. and Rupr. into swards of Lolium perenne L., Dactylis glomerata L. and 

Phalaris aquatica L., on a dry, low-fertility soil in Marlborough as affected by fertiliser and 2,2-DPA. New Zealand Journal of Agricultural 

Research 32, 317-326. 

Bourdôt, G.W. and Hurrell, G.A. (1992) Aspects of the ecology of Stipa neesiana Trin. and Rupr. seeds. New Zealand Journal of Agricultural 

Research 35, 101-108. 

Bourdôt, G.W. and Ryde, D.H. (1986) Chilean needle grass Stipa neesiana. Significance, identification, control. Ministry of Agriculture and 

Fisheries aglink FPP 675, 1st Ed., Information Services MAF, Wellington, 2 pp. 

Bowden, W.M. and Senn, H.A. (1962) Chromosome numbers in 28 grass genera from South America. Canadian Journal of Botany 40, 1115- 

1124 

Braby, M.F. (2000) Butterflies of Australia. Their Identification, Biology and Distribution. CSIRO Publishing, Collingwood, Victoria. 

Braby, M.F. and Dunford, M. (2006) Field observation on the ecology of the golden sun moth, Synemon plana Walker (Lepidoptera: Castniidae). 

Australian Entomologist 33(2), 103-110. 

Bramwells, H. (2003) Brittle Greenhood Pterostylis truncata. Flora and Fauna Guarantee Action Statement No. 63, Department of 

Sustainabilityand Environment, East Melbourne, 8 pp. 

Brereton, R. and Backhouse, G. (2003) Southern Lined Earless Dragon Tympanocryptis lineata lineata. Flora and Fauna Guarantee Action 

Statement No. 63, Department of Sustainabilityand Environment, East Melbourne, 6 pp. 

Briese, D.T. and Evans, H.C. (1998) Biological control of serrated tussock (Nassella trichotoma): is it worth pursuing? Plant Protection Quarterly 

13(2), 94-97. 

Briese, D.T., McLaren, D.A., Pettit, W.J., Zapater, M., Anderson, F., Delhey, R. and Distel, R. (2000) New biological control initiatives against weeds 

of South American origin in Australia: Nassella tussock grasses and blue heliotrope. Proceedings of the X International Symposium on Biological 

Control of Weeds 4-14 July 1999, Montana State University, Bozeman, Montana, USA, N.R. Spencer (Ed.), pp. 215-223. 

Britt, D. (2001) An investigation into the taxonomy, natural enemies and control methods for Chilean needle grass in Australia. B.Appl.Sci. (Hons.) 

thesis, Department of Applied Biology/Biotechnology, RMIT University. 

Britt, D.L., Lawrie, A.C. and McLaren, D.A. (2002) Genetic variation in Nassella neesiana (Chilean needlegrass) in Australia. Plant Protection 

Quarterly 17(3), 125. 

Britton, E.B. (1987) A revision of the Australian chafers (Coleoptera: Scarabaeidae: Melolonthinae) Vol. 5. Tribes Scitalini and Comophorini. 

Invertebrate Taxonomy 1, 685-799. 

Brown, P.R. (1987) Draft Management Plan for the conservation of the Eastern Barred Bandicoot, Perameles gunnii, in Victoria. Arthur Rylah 

Institute for Environmental Research Technical Report Series No. 41, 70 pp. 

Brown, P.R., Wallis, R.L., Simmons, D. and Adams, R. (1991) Weeds and wildlife. Plant Protection Quarterly 6(3), 150-153. 

Brown, W.V. (1952) The relation of soil moisture to cleistogamy in Stipa leucotricha. Botanical Gazette 113, 438-444. 

Browning, T.O. (1954) Observations on the ecology of the Australian field cricket, Gryllus commodus Walker, in the field. Australian Journal of 

Zoology 2, 205-222. 

Bruce, N. (2001) ACT Survey of Chilean Needle Grass (Nassella neesiana). A Report Prepared for Environment ACT, February, 55 pp. 

181

Burbidge, A.A. and McKenzie, N. L. (1989) Patterns in the modern decline of Western Australia’s vertebrate fauna: causes and conservation 

implications. Biological Conservation 50, 143-198. 

Burbidge, N.T. and Gray, M. (1970) Flora of the Australian Capital Territory. Australian National University Press, Canberra. 

Burkart, A. (1969) Gramineas. La familia botánica de los pastos. Flora Ilustrada de Entre Ríos (Argentina) Iia. 

Burke, M.J. and Grime, J.P. (1996) An experimental study of plant community invasibility. Ecology 77, 776-90. 

Byers, J.E., Reichard, S., Randall, J.M., Parker, I.M., Smith, C.S., Lonsdale, W.M., Atkinson, I.A.E., Seastedt, T.R., Williamson, M., Chornesky, E. and 

Hayes, D. (2002) Directing research to reduce the impacts of nonindigenous species. Conservation Biology 16, 630-640. 

Cabrera, A.L. (1949) Las comunidades vegetales de los alrededores de La Plata. Lilloa 20, 269-347. [Abstract by GMR, Herbage Abstracts 21(4), 206] 

Cale, P. and Hobbs, R. (1991) Condition of roadside vegetation in relation to nutrient status. In D.A. Saunders and R. J. Hobbs (Eds.), Nature 

Conservation 2: the Role of Corridors. Surrey Beatty and Sons, Sydney, pp. 353-362. 

Caley, P., Groves, R.H. and Barker, R. (2008) Estimating invasion success of introduced plants. Diversity and Distributions 14, 196-203. 

Calder, A.A. (1996) Click Beetles. Genera of the Australian Elateridae (Coleoptera). Monographs on Invertebrate Taxonomy Vol. 2, CSIRO 

Publishing, Collingwood. 

Caldwell, M.M., Richards, J.H., Johnson, D.A., Nowak, R.S. and Dzuree, R.S.(1981) Coping with herbivory: photosynthetic capacity and 

resource allocation in two semiarid Agropyron bunchgrasses. Oecologia 50, 14-24. 

Callaway, R.M.and Maron, J.L. (2006) What have exotic plant invasions taught us over the past 20 years? Trends in Ecology and Evolution 21(7), 369- 

374. 

Cameron, A.W. (1975) Changes in the wild life community of the Waterloo Valley between 1866 and 1975. In Proceedings of a Workshop on 

Agriculture, Forestry and Wildlife: Conflict or Coexistence? held at the University of New England, 17-19 August 1975. University of New England, 

Armidale, pp. 19-25. 

Campbell, D.A. (1975) Influence of pest parrot populations on agriculture in an isolated area. In Proceedings of a Workshop on Agriculture, 

Forestry and Wildlife: Conflict or Coexistence? held at the University of New England, 17-19 August 1975. University of New England, Armidale, pp. 

133-135. 

Campbell, M.H. (1977) Serrated tussock. Part 1. Life history, identification. New South Wales Department of Agriculture Division of Plant Industry 

Bulletin P476. [Reference not seen] 

Campbell, M.H. (1997) Effect of low rates of flupropanate on selective removal of serrated tussock (Nassella trichotoma (Nees) Arech.) seedlings from 

a young improved pasture. Plant Protection Quarterly 12(4), 175-6. 

Campbell, M.H. (1998) Biological and ecological impact of serrated tussock (Nassella trichotoma (Nees) Arech.) on pastures in Australia. Plant 

Protection Quarterly 13(2), 80-86. 

Carne, P.B. (1956) Cyclocephala signaticollis Burmeister, an introduced pasture scarab. Proceedings of the Linnean Society of New South Wales 81, 

217-221. 

Carne, P.B. and Chinnick, L.J. (1957) The pruinose scarab (Sericesthis pruinosa Dalman) and its control in turf. Australian Journal of Agricultural 

Research 8, 604-616. 

Caro, J.A. (1966) Las especies de Stipa (Gramineae) de la region central Argentina. Kurtziana 3, 7-119. 

Carolin, R.C. and Tindale, M.D. (1994) Flora of the Sydney Region. 4th Ed., Reed, Chatsworth, NSW. 

Carlini, A.A., Zurita, A.E., Gasparini, G. and Noriegas, Y.J.Y. (2004) Los mamíferos del Pleistoceno de la Mesopotamia Argentina y su relación 

con los del Centro Norte de la Argentina, Paraguay y Sur de Bolivia, y los del Sur de Brasil y oeste de Uruguay: Paleobiogeografía y 

Paleoambientes. In F.G. Aceñolaza (Ed.), Temas de la Biodiversidad del Litoral Fluvial Argentino, Instituto Superior De Correlación Geologica ( 

INSUGEO ) Miscelánea 12, 83-91. 

Carr, G.W. (1993) Exotic flora of Victoria and its impact on indigenous biota. In D.B. Foreman and N.G. Walsh (Eds.), Flora of Victoria Volume 

1. Introduction, Inkata Press, Melbourne, Ch.11, pp. 256-297. 

Carr, G.W. (1999) Checklist of the vascular flora of the Victorian Volcanic Plain, June 1997. In R.N. Jones (Ed.), The Great Plains Crash. 

Proceedings of a Conference on the Grasslands and Grassy Woodlands of Victoria. Victoria Institute of Technology, October 2nd and 3rd, 1992, 

Indigenous Flora and Fauna Association and Victorian National Parks Association, pp. 91-119. 

Carr, G.W., Robin, J.M. and Robinson, R.W. (1986) Environmental weed invasion of natural ecosystems: Australia’s greatest conservation 

problem. In R.H. Groves and J.J. Burdon (Eds.), Ecology of Biological Invasions: An Australian Perspective. Australian Academy of Science, 

Canberra, p. 150. 

Carr, G.W. and Yugovic, J.V. (1989) Weeds of Native Vegetation in Victoria. Unpublished report to Department of Conservation, Forests and 

Lands, Victoria, Melbourne. [reference not seen] 

Carr, G.W., Yugovic, J.V. and Robinson, K.E. (1992) Environmental Weed Invasion in Victoria. Conservation and Management Implications. 

Department of Conservation and Environment and Ecological Horticulture Pty Ltd, East Melbourne and Clifton Hill. 

Carr, S.G.M. and Turner, J.S. (1959) The ecology of the Bogong High Plains. I. The environmental factors and the grassland communities. 

Australian Journal of Botany 7, 12-33. 

Carter, O., Murphy, A.M. and Cheal, D. (2003) Natural temperate grassland. Australian Government Department of the Environment and Water 

Resources. http://www.environment.gov.au/biodiversity/publications/grasslands/ 

Carver, M. (1991) Superfamily Apidoidea. In CSIRO, The Insects of Australia. A Textbook for Students and Research Workers. 2nd Ed., 

Melbourne University Press, Carlton, pp. 453-457. 

Castillo-Ruiz, A. and Lundrigan, B. (2007) Lama pacos. Animal Diversity Web, University of Michigan, Accessed 20 June 2007. 

http://animaldiversity.ummz.umich.edu/site/accounts/information/Lama_pacos.html 

Chalmers, A.C. (2006) Trait differences between exotic and native species within a long-grazed temperate grassland on the Northern Tablelands 

of New South Wales. In C. Preston, J.H. Watts and N.D. Crossman (Eds.), 15th Australian Weeds Conference Papers and Proceedings. Weed 

Management Society of South Australia Inc., Adelaide, pp. 235-238. 

Chadwick, C.E. (1966) Investigations on Plutorectis caespitosae Oke (Lep., Psychidae) and Oncopera alpina Tindale (Lep., Hepialidae) in the 

Australian Alps. Journal of the Entomological Society of Australia (NSW) 3, 5-29. 

Champion, P.D. (1995) Assessment of Plant Pests Proposed for Inclusion in Auckland Regional Council’s Pest Management Strategies to be 

Prepared under the Biosecurity Act 1993. National Institute for Water and Atmosphere, Consultancy Report No. ARC315, prepared for Auckland 

Regional Council Environment, January. 

182

Chan, C.W. (1980) Natural grasslands in Canberra: their distribution, phenology and effects of mowing. MSc thesis, Australian National 

University, Canberra. 

Chejara, V.K., Nadolny, C., Kristiansen, P., Whalley, R.B.D. and Sindel, B.M. (2006) Impacts of Hypparhenia hirta (L.) Stapf (Coolatai grass) 

on native vegetation in a travelling stock route in northern New South Wales. In C. Preston, J.H. Watts and N.D. Crossman (Eds.), 15th 

Australian Weeds Conference Papers and Proceedings. Weed Management Society of South Australia Inc., Adelaide, pp. 207-210. 

Chown, S.L. and Block, W. (1997) Comparative nutritional ecology of grass-feeding in a sub-Antarctic beetle: the impact of introduced species 

on Hydromedion sparsutum from South Georgia. Oecologia 111, 216-224. [Reference not seen - obtained] 

Christin, P.-A., Salamin,N., Kellogg, E.A., Vicentini, A. and Besnard, G. (2009) Integrating phylogeny into studies of C 4 variation in the grasses. 

Plant Physiology 149, 82-87. 

Churchill, D.M. and de Corona, A. (1972) The Distribution of Victorian Plants. Royal Botanic Gardens and National Herbarium and Monash 

University Botany Department, Clayton. 

Churchill, T.B. (1997) Spiders as ecological indicators: an overview for Australia. Memoirs of the Museum of Victoria 56(2), 331-337. 

Clark, J. (1941) Notes on the Argentine ant and other exotic ants introduced into Australia. Memoirs of the Museum of Victoria 12, 59-70. 

Clark, L.G. and Fisher, J.B. (1986) Vegetative morphology of grasses: shoots and roots. In T.S. Soderstrom, K.W. Hilu, C.S. Campbell and M.E. 

Barkworth (Eds.), Grass Systematics and Evolution, Smithsonian Institution Press, Washington, USA, pp. 37-45. 

Clark, L.G., Judziewicz, E., Laegaard, S., Londono, X., Morrone, O., Zuloaga, F., Davidse, G., Peterson, P.M. and R. Soreng (1999-2008) 

Catalogue of the Vascular Plants of Ecuador. Missouri Botanical Garden, the Herbarium QCA at Pontificia Universidad Católica del Ecuador 

and Herbario Nacional del Museo Ecuatoriano de Ciencias Naturales, Department of Systematic Botany at Aarhus University. 

http://www.mobot.org/mobot/Research/ecuador/welcome.shtml [Accessed 15 July 2008] 

Clayton, W.D., Harman, K.T. and Williamson, H. (2002 onwards). World Grass Species: Synonymy. http://www.kew.org/data/grasses-syn.html. 

[Accessed 17 July 2007]. 

Cluff, D. (2001) Solving the dryland salinity problems of the Murray Darling Basin with historical research, observation and logic. Proceedings 

of the Australian Agronomy Conference 2001. Australian Society of Agronomy, http://www.regional.org.au/au/asa/2001/5/b/cluff.htm [Accessed 

1 July 2009] 

Coates, F. (2003a) Shelford Leek-orchid Prasophyllum fosteri. Flora and Fauna Guarantee Action Statement No. 165, Department of 


Coates, F. (2003b) Fragrant Leek-orchid Prasophyllum sp. aff. suaveolens (Western Basalt Plains). Flora and Fauna Guarantee Action Statement 

No. 165, Department of Sustainability and Environment, East Melbourne, 4 pp. 

Coates, F. (2003c) Basalt Sun-orchid Thelymitra gregaria. Flora and Fauna Guarantee Action Statement No. 138, Department of Sustainability 

and Environment, East Melbourne, 4 pp. 

Cofinas, M. and Creighton, C. (compilers) (2001) Australian Native Vegetation Assessment 2001. National Land and Water Resources Audit, 

Land and Water Australia, Braddon, ACT. http://audit.ea.gov.au/ANRA/vegetation/docs/Native_vegetation/nat_veg. Accessed 7February 2007. 

Colbran, R.C. (1971) Studies of plant and soil nematodes. 15. Eleven new species of Radopholus Thorne and a new species of Radopholoides de 

Guiran (Nematoda: Tylenchoidea) from Australia. Queensland Journal of Agricultural and Animal Sciences 27(1970), 437-460. 

Cogger, H.G. (1983) Reptiles and Amphibians of Australia. Revised Edition, A.H. and A.W. Reed Pty Ltd, Frenchs Forest NSW. 

Cole, M.H. (1975) Notes on bird damage in agriculture: the problem from a farmer’s point of view. In Proceedings of a Workshop on Agriculture, 

Forestry and Wildlife: Conflict or Coexistence? held at the University of New England, 17-19 August 1975. University of New England, Armidale, 

pp.136-139. 

Colless, D.H. and McAlpine, D.K. (1991) Diptera. In CSIRO, The Insects of Australia. A Textbook for Students and Research Workers. 2nd Ed., 


Common, I.F.B. (1990) Moths of Australia. Melbourne University Press, Carlton. 

Common, I.F.B. and Edwards, E.D. (1981) The life history and early stages of Synemon magnifica Strand (Lepidoptera: Castniidae). Journal of 

the Australian Entomological Society 20, 295-302. 

Connor, D.J. (1966) Vegetation studies in north-west Victoria II. The Horsham area. Proceedings of the Royal Society of Victoria 79, 637-647. 

Connor, H.E. (1986) Reproductive biology in the grasses. In T.S. Soderstrom, K.W. Hilu, C.S. Campbell and M.E. Barkworth (Eds.), Grass 

Systematics and Evolution, Smithsonian Institution Press, Washington, USA, pp. 251-264. 

Connor, H.E., Edgar, E. and Bourdôt, G.W. (1993) Ecology and distribution of naturalised species of Stipa in New Zealand. New Zealand 

Journal of Agricultural Research 36, 301-307. 

Conole, L. (1994) The Red-rumped Parrot Psephotus haematonotus as a dispersal agent of the environmental weed Serrated Tussock Nassella 

trichotoma. Indigenotes 7(1), 6. 

Cook, D. and Yugovic, J. (2003) Clyde-Tooradin grassland rediscovered. Victorian Naturalist 120, 140-146. 

Cook, G.D. and Dias, L. (2006) It was no accident: deliberate plant introductions by Australian government agencies during the 20th century. 


Cook, T. (1999) Chilean needle grass: is it much different to other perennial grass weeds? 10th New South Wales Biennial Noxious Weeds 

Conference - Ballina 1999, pp. 91-97. 

Cooke, D.A. (1998) Bulbil watsonia is a variety of Watsonia meriana (L.) Miller (Iridaceae). Journal of the Adelaide Botanical Gardens 18(1), 5- 

7. 

Corbin, J.D., D’Antonio, C.M. and Bainbridge, S.J. (2004) Tipping the balance in the restoration of native plants. Experimental approaches to 

changing the exotic:native ratio in California grassland. In M.S. Gordon and S.M. Bartol (Eds.), Experimental Approaches to Conservation 

Biology, University of California Press, Berkley, USA, pp. 154-179. 

Cosse, M., Gonzáles, S. and Ginenez-Dixon, M. (2009) Feeding ecology of Ozotoceros bezoarticus: conservation implications in Uruguay. 

Iberingia. Série Zoologia 99, 158-164. 

Coulson, G. (1990) Conservation biology of the Striped Legless Lizard (Delma impar) an initial investigation. Arthur Rylah Institute for 

Environmental Research Technical Report Series No. 106, 40 pp. 

Coupar, P. and Coupar, M. (1992) Flying Colours. Common Caterpillars, Butterflies and Moths of South-eastern Australia. New South Wales 

University Press, Kensington. 

183

Coupland, R.T. (1992) Overview of South American grasslands. In R.T. Coupland (Ed.), Ecosystems of the World 8A. Natural Grasslands. 

Introduction and Western Hemisphere. Elsevier Sceince Publishers B.V., Amsterdam, Netherlands, pp. 363-366. 

Coutts, P.J.F. (1982) Prehistory. In J.S. Duncan (Ed.) Atlas of Victoria. Victorian Government Printing Office, Melbourne, pp. 67-73. 

Coutts-Smith, A. and Downey, P. (2006) Impact of weeds on threatened biodiversity in New South Wales. Cooperative Research Centre for 

Australian Weed Management Technical Series No. 11. 

Cox, G.W. (2004) Alien Species and Evolution. The Evolutionary Ecology of Exotic Plants, Animals, Microbes and Interacting Native Species. 

Island Press, Washington, USA. 

Craigie, V. (1993) Laverton North Grassland Reserve Draft Management Plan. Melbourne Region, Department of Conservation and Natural 

Resources. 

Cribb, A.B. and Cribb, J.W. (1974) Wild Food in Australia. Collins, Sydney. 

Crosby, A.W. (1986) Ecological Imperialism. The Biological Expansion of Europe, 900-1900. Cambrdige University Press, Cambridge. 

Culley, T.M. and Klooster, M.R. (2007) The cleistogamous breeding system: a review of its frequency, evolution, and ecology in angiosperms. 

The Botanical Review 73(1), 1-30. 

Cumber, R.A. (1959) The insect complex of sown pastures in the North Island IV. The Coleoptera as revealed by summer sweep-sampling. New 

Zealand Journal of Agricultural Research 2, 763-772. 

Dahlhaus, P., Cox, J., MacEwan, R. and Codd, P.(2003) Victorian Volcanic Plains Scoping Study. CSIRO Land and Water, Final Report for 

Corangamite Catchment Management Authority, 84 pp. 

Dale, A.B. (1975) Birds and grain crops in northern New South Wales. In Proceedings of a Workshop on Agriculture, Forestry and Wildlife: 

Conflict or Coexistence? held at the University of New England, 17-19 August 1975. University of New England, Armidale, pp.140-142. 

Dalton, S. (2000) Chilean needle grass the latest weed threat. The Weekly Times, 2 February, p. 29. 

Daniell, A. (1994) The impact of terrestrial molluscs on native vegetation in south-eastern Australia. Victorian Naturalist 111, 218-222. 

D’Antonio, C.M., Dudley, T.L. and Mack, M. (1999). Disturbance and biological invasions: direct effects and feedbacks. In L.R. Walter (Ed.), 

Ecosystems of the World 16. Ecosystems of Disturbed Ground, Elsevier, Amsterdam, pp. 413-52. 

D’Antonio, C.M. and Vitousek, P.M. (1992) Biological invasions by exotic grasses, the grass/fire cycle and global change. Annual Review of 

Ecology and Systematics 23, 63-87. 

Dare, R. and Hocking, C. (1997) Control of Chilean needle-grass using selected herbicides and replacement with kangaroo grass. Paper presented 

to the Research Advisory Group of the Victorian Grassy Ecosystem Group Seminar 11 October. 

Darwin, C. (1845?) The Voyage of the ‘Beagle’. 2nd Ed., Heron Books. 

Davidse, G. (1986) Fruit dispersal in the Poaceae. In T.S. Soderstrom, K.W. Hilu, C.S. Campbell and M.E. Barkworth (Eds.), Grass Systematics 

and Evolution, Smithsonian Institution Press, Washington, USA, pp. 143-155. 

Davidson, R.L. (1982) An ecological view of pasture pests on the Northern Tablelands of NSW. In K.E. Lee (Ed.), Proceedings of the 3rd 

Australasian Conference on Grassland Invertebrate Ecology, Adelaide 30 November – 4 December, pp. 67-74. 

Davies, R. J.-P. (1997). Weed Management in Temperate Native Grasslands and Box Grassy Woodlands in South Australia. Botanic Gardens of 

Adelaide/Environment Australia, Athelstone, South Australia. 

Davis, M.A., Grime, J.P. and Thompson, K. (2000) Fluctuating resources in plant communities: a general theory of invasibility. Journal of 

Ecology 88, 528-534. 

Day, M.F. and Fletcher, M.J. (1994) An annotated catalogue of the Australian Cicadelloidea (Hemiptera: Auchenorryncha). Invertebrate 

Tasxonomy 8, 1117-1288. 

Dear, C. (1996) Distribution of Synemon plana: a new encounter. Victorian Entomologist 26(2), 26-28. 

D’Elía, G., Pardinas, U.F.J., Jayat, J.P. and Salazar-Bravo, J. (2008) Systematics of Necromys (Rodentia, Cricetidae, Sigmodontinae): species 

limits and groups, with comments on historical biogeography. Journal of Mammalogy 89(3), 778-790. 

Department of Sustainability and Environment (2007) Corangamite Water Skink (Eulamprus tympanum marnieae). Department of Sustainability 


Department of Sustainability and Environment (DSE) (2009a) Flora and Fauna Guarantee Act threatened list – July 2009. 

http://www.dse.vic.gov.au/CA256F310024B628/0/4AC71110D293ED14CA25760800132DD7/$File/20090804+FFG+threatened+list.pdf 

Department of Sustainability and Environment (DSE) (2009b) Flora and Fauna Guarantee Act processes list – July 2009. 

http://www.dse.vic.gov.au/CA256F310024B628/0/4A38EF8FD5BAE502CA257601000134BF/$File/Updated+FFG+processes+list+July+2009.p 

df 

De Santis, L. and Loiácono de Silva, M.S. (1981) Insectos que destruyen el espartillo Stipa brachychaeta y sus parasitoides. Journadas 

Fitosantarias Argentinas IV. Resumenes (Universidad Nacional de Cordoba, Facultad de Cincias Agropecuraris, 19, 20 and 21 August 1981), 

115-116. 

De Santis, L. and Loiácono de Silva, M.S. (1983) Insectos que destruyen el espartillo (Stipa brachychaeta) y sus parasitoides. Contributions of 

the American Entomological Institute 20, 402-427. 

De Triquell, A.A. (1986) Grass gametophytes: their orign, structure and relation with the sporophyte. In T.S. Soderstrom, K.W. Hilu, C.S. 

Campbell and M.E. Barkworth (Eds.), Grass Systematics and Evolution, Smithsonian Institution Press, Washington, USA, pp. 11-20. 

De Wet, J.M.J. (1986) Hybridization and polyploidy in the Poaceae. In T.S. Soderstrom, K.W. Hilu, C.S.Campbell and M.E. Barkworth (Eds.), 

Grass Systematics and Evolution, Smithsonian Institution Press, Washington, USA, pp. 188-194. 

DEWHA (Department of Environment, Heritage, Water and the Arts) (2008) Commonweatlh listing device on Natural Temperate Grassland of 

the Victorian Volcanic Plain. http://www.environment.gov.au/cgibin/sprat/public.publicshowcommuninty.pl?id=42andstatus=Critically%20Endangered 

Distel, R.A., Pietragalla, J., Rodríguez Iglesias, R.M., Didoné, N.G. and Andrioli, R.J. (2008) Restoration of palatable grasses: a study case in 

degraded rangelands of central Argentina. Journal of Arid Environments doi:10.1016/j.jaridenv.2008.04.007 

Dixon, J.M. (1989) Thylacinidae. In D.M. Walton and B.J. Richardson (Eds.), Fauna of Australia Volume 1B. Mammalia. Australian Government 

Publishing Service, Canberra, pp. 549-559. 

DNRE (Department of Natural Resources and Environment) (1997) Victoria’s Biodiversity- Directions in Management. Department of Natural 

Resources and Environment, Melbourne, 149 pp. 

184

DNRE (Department of Natural Resources and Environment) (2000) Code of Practice for Providing Chemical Information to Clients. Chemical 

Standards Branch, DNRE, Melbourne. 

Dodd, A.J., Williams, N.S.G. and McDonnell, M.J. (2007) Seed bank spatial dynamics and plant invasions in Victoria’s western basalt plains 

grasslands. Proceedings Third Victorian Weed Conference, ’Earth Wind Fire Water and Weeds’, Bendigo, Victoria, 3-4 October 2007, Weed 

Society of Victoria, p. 121. 

Dorrough, J., Ash, J. and McIntyre, S. (2004) Plant responses to livestock grazing in an Australian temperate grassland. Ecography 27, 798-810. 

Douglas, F. (2000) Sun moths in the back paddock. In T. Barlow and R. Thorburn (Eds.), Balancing Conservation and Production in Grassy 

Landscapes. Proceedings of the Bushcare Grassy Landscapes Conference, Clare, South Australia, 19-21 August 1999. Environment Australia, 

Canberra, pp. 56-57. 

Douglas, F. (2003a) Sun-moths, Lepidoptera: Castniidae: a reference resource. Victorian Entomologist 33(6), 96-97. 

Douglas, F. (2003b) Five threatened Victorian Sun-moths (Synemon species). Flora and Fauna Guarantee Action Statement No. 146, Department 

of Sustainability and Environment, East Melbourne, 11 pp. 

Douglas, F. and Marriott, P. (2003) Some notes on the Sun-moths, Lepidoptera: Castniidae. Victorian Entomologist 33(6), 90, 95. 

Downey, P.O. (2008) Determination and management of alien plant impacts on biodiversity: examples from New South Wales, Australia. In B. 

Tokarska-Guzik, J.H. Brock, G. Brundu, L. Child, C.C. Daehler and P. Pyšek (Eds.), Plant Invasions: Human Perception, Ecological Impacts 

and Management, pp. 369-385. 

Downey, P.O. and Cherry, H. (2005) Delivering strategic conservation outcomes through the integrated management of bitou bush, a Weed of 

National Significance. In Proceedings Second Victorian Weed Conference, 17-19 August 2005, Bendigo, Weed Society of Victoria, Frankston, 

pp. 41-45. 

Downey, P.O. and Coutts-Smith, A.J. (2006) Biodiversity threatened by weeds in New South Wales. In C. Preston, J.H. Watts and N.D. 

Crossman (Eds.), 15th Australian Weeds Conference Papers and Proceedings. Weed Management Society of South Australia Inc., Adelaide, pp. 

803-806. 

DPIW (Department of Primary Industries and Water), Tasmania (2007) Chilean Needle Grass – Statutory Weed Management Plan. 

http://www.dpiw.tas.gov.au/intertext.nsf/WebPages/TPRY-5D82GC?open, 19 June. [Accessed 27 June 2007] 

Driscoll, D.A. (1994) Invertebrates of lowland native grasslands in the Australian Capital Territory: conservation and research strategies for a 

recovery plan. ACT Parks and Conservation Service Technical Report 9, v + 41 pp. 

Dukes, J.S. (2002) Species composition and diversity affect grassland susceptibility and response to invasion. Ecological Applications 12(2), 602- 

617. 

Duncan, M.R. (1993) Chilean needle grass. New South Wales Agriculture and Fisheries Agnote 2/082, 1st Ed., 4 pp. 

Dunin, F. (1999) Water use by Kangaroo Grass Themeda triandra communities. Native Grass South Australia 1(3), 49, 59. 

Dunin, F.X. and Reyenga, W. (1978) Evaporation from a Themeda grassland I. Controls imposed on the process in a sub-humid environment. 

Journal of Applied Ecology 15, 317-325. 

Dunn, K.L. (1996) Territorial and oviposition behaviour in the sun moth Synemon laeta Walk. (Lepidoptera: Castniidae). Victorian Entomologist 

36(2), 36-39. 

Dunstan, P.K. and Johnson, C.R. (2006) Linking richness, community variability, and invasion resistance with patch size. Ecology 87(11), 2842- 

2850. 

Durrell, G. (1964) The Whispering Land. Penguin Books, Harmondsworth, UK. 

Dyksterhuis, E.J. (1945) Axillary cleistogenes in Stipa leucotricha and their role in nature. Ecology 26(2), 195-199. 

Earl, G., Barlow, T. and Morrees, A. (2003) Twelve threatened Swainson-peas and Darling-peas (Swainsona species). Flora and Fauna 

Guarantee Action Statement No. 126, Department of Sustainability and Environment, East Melbourne, 19 pp. 

Eddy, D., Mallinson, D., Rehwinkel, R. and Sharp, S. (1998) Grassland Flora. A Field Guide for the Southern Tablelands (NSW and ACT). 

World Wide Fund for Nature Australia, Australian National Botanic Gardens, NSW National Parks and Wildlife Service, NSW Dept. of Land and 

Water Conservation, Natural Heritage Trust, Snowy Mountains Hydro-Electric Authority and Environment ACT, Canberra. 

Edgar, E. and Connor, H.E. (2000) Flora of New Zealand Volume V. Gramineae. Manaaki Whenua Press, Lincoln, New Zealand. 

Edgar, E., O’Brien,M.-A. and Connor, H.E. (1991) Checklist of pooid grasses naturalised in New Zealand. 1. Tribes Nardeae, Stipeae, 

Hainardieae, Meliceae, and Aveneae. New Zealand Journal of Botany 29, 101-116. 

Edwards, E.D. (1993) Golden sun moth. Australian Natural History 24(6), 16-17. 

Edwards, E.D. (1994) Survey of lowland grassland sites in A.C.T. for the Golden Sun Moth, Synemon plana. Report to the Wildlife Research 

Unit, ACT Parks and Conservation Service, 42 pp. 

Edwards, E.D. (1996) Castniidae. In E.S. Nielsen, E.D. Edwards and T.V. Rangsi (Eds.), Checklist of the Lepidoptera of Australia. CSIRO, 

Canberra, p. 138. 

Edwards, E.D. (1997a) Moths in the sun. ANIC News (Australian National Insect Collection, CSRIO Entomology) No. 10, pp.7-8. 

Edwards, E.D. (1997b) Castniidae and the Republic. ANIC News (Australian National Insect Collection, CSRIO Entomology) No. 11, pp.8-9. 

Edwards, E.D. (2002) Lepidoptera (Moths and Butterflies). In K. Green (Ed.), Biodiversity in the Snowy Mountains, Australian Institute of Alpine 

Studies, Jindabyne, NSW, pp. 58-72. 

Edwards, E.D. and Fairey, K.D. (1996) Anthelidae. In E.S. Nielsen, E.D. Edwards and T.V. Rangsi (Eds.), Checklist of the Lepidoptera of 

Australia. CSIRO, Canberra, pp. 258-260. 

Elias, S.A. (1994) Quaternary Insects and their Environments. Smithsonian Institution Press, Washington. 

Eldridge, D.J. and Rath, D. (2002) Hip holes: kangaroo (Macropus spp.) resting sites modify the physical and chemical environment of woodland 

soils. Austral Ecology 27(5), 527-536. 

Eldridge, D.J. and Mensinga, A. (2007) Foraging pits of the short-beaked echidna (Tachyglossus aculeatus) as small-scale patches in a semi-arid 

Australian box woodland. Soil Biology and Biochemistry doi:10.1016/j.soilbio.2006.11.016 [obtain final ref] 

Ellis, B.A. (1975) Macropods and their diet. In Proceedings of a Workshop on Agriculture, Forestry and Wildlife: Conflict or Coexistence? held at 

the University of New England, 17-19 August 1975. University of New England, Armidale, pp. 73-79. 

Endersby, I. and Koehler, S. (2006) Golden Sun Moth Synemon plana: discovery of new populations around Melbourne. Victorian Naturalist 

123(6), 362-365. 

185

Ens, E-J. (2002a) Chilean Needle Grass (Nassella neesiana) on the Cumberland Plain: exposé of distribution and impacts on invertebrates. B.Sc. 

Honours Thesis, School of Geography, University of New South Wales, 102 pp. 

Ens, E-J. (2002b) Chilean Needle Grass (Nassella neesiana) on the Cumberland Plain: exposé of distribution and impacts on invertebrates. 

Honours Thesis, School of Geography, University of New South Wales, Thesis abstract, Bulletin of the Ecological Society of Australia 32(4), 33. 

Ens, E.-J. (2005) Nassella neesiana. Global Invasive Species Database, Invasive Species Specialist Group, International Union for the 

Conservation of Nature, http://www.issg.org/database/species/ecology.asp?si=458andfr=1andsts= Accessed 23 October 2006. 

Entwisle, T. (1994) Orchidaceae. In N.G. Walsh and T.J. Entwisle (Eds.), Flora of Victoria Volume 2. Ferns and Allied Plants, Conifers and 

Monocotyledons, Inkata Press, Melbourne, pp. 740-901. 

Entwisle, T., Albrecht, D. and Walsh, N. (1994) Ground flora – diversity and associations in Victoria. Victorian Naturalist 111(3), 80-86. 

Erakovic, L. (2005) Improved control of Chilean needle grass spread through novel slasher design. M.Eng. thesis, School of Aerospace, 

Mechanical and Manufacturing Engineering, RMIT University, Melbourne. 

Erakovic, L., Subic, A., Abu-Hijleh, B. and McLaren, D. (2003) Simulation models of Chilean needle-grass spread during slashing. Under 

Control - Pest Plant and Animal Management News No. 23, 8-9. 

Erlich, S. (1980) The underground grass caterpillar. Department of Agriculture Victoria Agnote 1026/80, 2 pp. 

Esau, K. (1977) Anatomy of Seed Plants. 2nd Ed., John Wiley and Sons, New York. 

Eschen, R., Mortimer, S.R., Lawson, C.S., Edwards, A.R., Brook, A.J., Igual, J.M., Hedlund, K. and Schaffner, U. (2007) Carbon addition alters 

vegetation composition on ex-arable fields. Journal of Applied Ecology 44, 95-104. 

Esler, A.E., Liefting, L.W. and Champion, P.D. (1993) Biological Success and Weediness of the Noxious Plants of New Zealand. Ministry of 

Agriculture and Fisheries, New Zealand, April. 

Evans, H.C. (1991) Biological control of tropical grassy weeds. In F.W.G. Baker and P.J. Terry (Eds.), Tropical Grassy Weeds, CAB 

International, Wallingford, UK, pp. 52-72. 

Evans, J.W. (1966) The leafhoppers and froghoppers of Australia and New Zealand. Australian Museum Memoir 12, 1-347. 

Evans, R.A. and Young, J.A. (1972) Competition within the grass community. In V.B. Younger and C.M. McKell (Eds.), The Biology and 

Utilization of Grasses. Academic Press, New York, pp. 230-246. 

Everson, C.S., Everson, T.M. and Tainton, N.M.(1988) Effects of intensity and height of shading on the tiller initiation of six grass species from the 

Haighland sourveld of Natal. South African Journal of Botany 54, 315-318. (reference not seen) 

Farrow, R.A. (1999) Diversity and abundance of canopy-living insects in natural grassland remnants of high conservation value in the ACT: 

effects of site characteristics and management. Tilembeya Consultants, Queanbeyan, Report to Environment ACT, December, 30 pp. 

Farrow, R.A. (2006) Diversity and abundance of canopy-living insects in natural grassland remnants of high conservation value in the ACT: II 

interannual variations 1999-2002. Tilembeya Consultants, Queanbeyan, Report to Environment ACT, April, 15 pp. 

Feldman, S.R., Bisaro, V.,Biani, N.B. and Prado, D.E. (2008) Soil salinity determines the relative abundance of C 3 /C 4 species in Argentinean 

grasslands. Global Ecology and Biogeography 17, 708-714. 

Ferdinands, K., Setterfiled, S., Douglas, M. and Barratt, J. (2006) Africanising the tropical woodlands: canopy loss and tree death following 

gamba grass Andropogon gayanus invasion. In C. Preston, J.H. Watts and N.D. Crossman (Eds.), 15th Australian Weeds Conference Papers and 

Proceedings. Weed Management Society of South Australia Inc., Adelaide, p. 296. 

Fernández, O.A., Gil, M.E. and Distel, R.A. (2009) The challenge of rangeland degradation in a temperate semiarid region of Argentina: the 

Caldenal. Land Degradation and Development 20, 431-440. 

Ferrar, P. (1987) A Guide to the Breeding Habits and Immature Stages of Diptera Cyclorrhapha. E.J. Brill/Scandinavian Science Press, Leiden. 

FFG (Flora and Fauna Guarantee) SAC (Scientific Advisory Committee) (1996) Final recommendation on a nomination for listing. The invasion 

of native vegetation by environmental weeds. Nomination No. 360, Item No. T9610, 12 March, 4 pp. 

Flannery, T. F. (1994) The Future Eaters. An Ecological History of the Australasian Lands and People. Reed Books, Chatswood, NSW. 

Fletcher, M.J. and Semeraro, L. (2001) The genus Scaphoideus Uhler in Australia (Hemiptera: Cicadellidae: Deltocephalinae). Australian Journal 

of Entomology 40, 1-8. 

Font, J., Viñas, X. and Izquierdo, C. (2001). Nasella neesiana (Trin. and Rupr.) Barkworth and N. mucronata (Kunth) R.W. Pohl in Barkworth 

dues gramínies al•lòctones noves a Catalunya. Butlletí de la Institucio Catalana D’Història Natural 69: 67- 69. Barcelona. 

Foreman, P. (1997) The effects of grazing, cultivation and burning on a long-grazed species-rich remant grassland on Victoria’s Northern 

(Riverine) Plain. Paper presented to the Research Advisory Group of the Victorian Grassy Ecosystem Group Seminar 11 October. 

Foulkes, J.N. and Heard, L.M.B. (Eds.) (2003) A Biological Survey of the South East South Australia 1991 and 1997. Department of Environment 

and Hertiage, South Australia, Adelaide. 

Fox, M.D. and Fox, B.J. (1986) The susceptibility of natural communities to invasion. In R.H. Groves and J.J. Burdon (Eds.), Ecology of 

Biological Invasions: An Australian Perspective. Australian Academy of Science, Canberra, pp. 57-66. 

Frederick, J. (1992) The North Central Chilean Needle Grass Project 1998-2002. Department of Primary Industries, Victoria, 29 pp. 

French, C. (1900) A Handbook of the Destructive Insects of Victoria with Notes on the Methods of Prevention and Extirpation. Part III. 

Department of Agriculture, Melbourne. 

French, C. (1909) A Handbook of the Destructive Insects of Victoria with Notes on the Methods of Prevention and Extirpation. Part IV. 


French, C. (1911) A Handbook of the Destructive Insects of Victoria with Notes on the Methods of Prevention and Extirpation. Part V. 


Frith, H.J. (1973) The herbivorous wild animals. In R.M. Moore (Ed.) Australian Grasslands, Corrected edition, Australian National University 

Press, Canberra, pp. 74-83. 

Froggatt, W.W. (1910) Insects which damage saltbush. Agricultural Gazette of New South Wales 21, 465-470. 

Froyd, C.A. and Willis, K.J. (2008) Emerging issues in biodiversity and conservation management: The need for a palaeoecological perspective. 

Quaternary Science Reviews 27, 1723-1732. 

Fryer, R. (2006) Appendix 1. Seed germination records. In L. Sweedman and D. Merritt (Eds.) Australian Seeds. A Guide to their Collection, 

Identification and Biology. CSIRO Publishing, Collingwood, pp. 199-219. 

186

Fuhrer, B. (1993) A Field Companion to Australian Fungi. Field Naturalists Club of Victoria, South Yarra. 

Gagné, R.J. (2007) Species numbers of Cecidomyiidae (Diptera) by zoogeographical region. Procedings of the Entomological Society of 

Washington 109(2), 499. 

Gair, R., Jenkins, J.E.E. and Lester, E. (1983) Cereal Pests and Diseases. 3rd Ed., Farming Press Limited, Ipswich, UK. 

Gallego, L. and Distel, R.A. (2004) Phytolith assemblages in grasses native to Central Argentina. Annals of Botany 94, 865-874. 

Gallinal Heber, J.P., Campal, E.F., Bergalli, I., Aragone, L. and Rosengurtt, B. (1946) Descripción y análysis de los campos de Juan Jackson. 

Estudios sobre praderas naturales del Uruguay 5a. Contribución, pp. 95-213. [Abstract by GMR, Herbage Abstracts 17, 252-253] 

Gammage, B. (2009) Galahs. Australian Historical Studies 40, 275-293. 

Garden, D.L., Dowling, P.M., Eddy, D.A.and Nicol, H.I. (2000) A survey of farms on the Central, Southern and Monaro Tablelands of New 

South Wales: management practices, farmer knowledge of native grasses, and extent of native grass areas. Australian Journal of Experimental 

Agriculture 40, 1081-1088. 

Gardener, M.R. (1998) The biology of Nassella neesiana (Trin. and Rupr.) Barkworth (Chilean needle grass) in pastures on the Northern 

Tablelands of New South Wales: weed or pasture? PhD thesis, University of New England, Armidale. 

Gardener, M.R. and Sindel, B.M. (1998) The biology of Nassella and Achnatherum species naturalized in Australia and the implications for 

management on conservation lands. Plant Protection Quarterly 13(2), 76-79. 

Gardener, M.R., Sindel, B.M. and Whalley, R.D.B.(1999) Chilean needle grass (Nassella neesiana) –what we know and where to next? In A.C. 

Bishop, M. Boersma and C.D. Barnes (Eds.), 12th Australian Weeds Conference Papers and Proceedings, Tasmanian Weed Society, Devenport, 

pp. 8-12. 

Gardener, M.R., Sindel, B.M.,Whalley, R.D.B. and Earl, J.M.(2005) Can Nassella neesiana, Chilean Needle Grass, be incorporated into a grazing 

mangement system in Australia? In Proceedings of the Tussock Terminators Research Forum, Lake Hume Resort, Albury, NSW, 2-3 November 

2005, Cooperative Research Centre for Australian Weed Management, National Serrated Tussock Management Group, National Chilean Needle 

Grass Taskforce, Natural Heritage Trust, Department of Primary Industries Victoria, New South Wales Department of Primary Industries, pp. 2-7. 

Gardener, M.R., Whalley, R.D.B. and Sindel, B.M. (1996a) The failure of management technology for reproductively efficient grassy weeds: the 

Chilean needle grass example. In R.C.H. Shepherd (Ed.), Proceedings Eleventh Australian Weeds Conference, Melbourne 30 September - 3 

October 1996, Weed Science Society of Victoria, Frankston, pp. 243-246. 

Gardener, M.R., Whalley, R.D.B. and Sindel, B.M. (1996b) The usefulness and ecology of Chilean needle grass (Nassella neesiana) as a pasture 

in Argentina. Report to Cooperative Research Centre for Weed Management Systems, the Meat Research Corporation, Weed Society of NSW 

and the Matheson Landcare Group, October-November, 13 pp. 

Gardener, M.R., Whalley, R.D.B. and Sindel, B.M. (2003a) Ecology of Nassella neesiana, Chilean needle grass, in pastures on the Northern 

Tablelands of New South Wales. I. Seed production and dispersal. Australian Journal of Agricultural Research 54, 613-619. 

Gardener, M.R., Whalley, R.D.B. and Sindel, B.M. (2003b) Ecology of Nassella neesiana, Chilean needle grass, in pastures on the Northern 

Tablelands of New South Wales. II. Seed bank dynamics, seed germination, and seedling recruitment. Australian Journal of Agricultural 

Research 54, 621-626. 

Garkaklis, M.J., Bradley, J.S. and Wooler, R.D. (2003) The relationship between animal foraging and nutrient patchiness in south-west Australian 

woodland soils. Australian Journal of Soil Research 41, 665-673. 

Gassó, N., Sol, D., Pino, J., Dana, E.D., Lloret, F., Sanz-Elorza, M., Sobrino, E. and Vilà, M. (2009) Exploring species attributes and site 

characteristics to assess plant invasions in Spain. Diversity and Distributions 15, 50-58. 

Gaur, S., McLaren, D. and Butler, K. (2005) Spraytopping as a management tool to reduce seed production in Chilean needle grass infestations. 

In Proceedings Second Victorian Weed Conference, 17-19 August 2005, Bendigo, Weed Society of Victoria, Frankston, pp. 121-125. 

Germishuizen, G. and Meyer, N.L. (Eds.) (2003) Plants of southern Africa: an annotated checklist. Strelitzia 14, National Botanical Institute, 

Pretoria, South Africa, 1231 pp. 

Gibbs Russell, G.E., Reid, C., Van Rooy, J. and Smook, L. (1985) List of species of Southern African plants. Edition 2. Recent literature and 

synonyms. Part 1. Cryptogams, Gymnosperms, Monocotyledons. Memoirs of the Botanical Survey of South Africa No. 51. 

Gibson, L.A. and New, T.R. (2007) Characterising insect diversity on Australia’s remnant native grasslans: ants (Hymenoptera: Formicidae)and 

beetles (Coleoptera) at Craigieburn Grasslands Reserve,Victoria. Journal of Insect Conservation 11, 409-413. 

Giles, B.E. (1994) Genetic biodiversity: analysing the data. Trends in Ecology and Evolution 9(9), 317-319. 

Gill, G.S. and Davidson, R.M. (2000) Weed interference. In B.M. Sindel (Ed.), Australian Weed Management Systems, R.G. and F.J. Richardson, 

Melbourne, pp. 61-80. 

Gilmore, D., Koehler, S., O’Dwyer, C. and Moore, W. (2008) Golden Sun Moth Synemon plana (Lepidoptera: Castniidae): results of a broad 

survey of populations around Melbourne. Victorian Naturalist 125(2), 39-46. 

Glanznig, A. and Kessal, O. (2004) Invasive Plants of National Importance and their Legal Status by State and Territory. WWF Australia, 

Sydney, June, 28 pp. 

Goodyer, G.J. (1983) Armyworm caterpillars. Department of Agriculture New South Wales AGFACT AE.15, 6 pp. 

Gott, B. (1983) Murnong - Microseris scapigera: a study of a staple food of Victorian Aborigines. Australian Aboriginal Studies 1983(2), 2-18. 

Gott, B. (1999) Koorie use and management of the Plains. In R.N. Jones (Ed.), The Great Plains Crash. Proceedings of a Conference on the 

Grasslands and Grassy Woodlands of Victoria. Victorian Institute of Technology October 2nd and 3rd, 1992, Indigenous Flora and Fauna 

Association and Victorian National Parks Association, pp. 41-45. 

GPWG (Grass Phylogeny Working Group) (2001) Phylogeny and subfamilial classification of the grasses (Poaceae). Annals of the Missouri 

Botanical Gardens 88(3), 373-457. 

Gray, A.J., Raybould, A.F. and Brown, S.L. (1997)The environmental impacts of Spartina anglica: past, present and predicted. In K. Patten (Ed.), 

Second International Spartina Conference Proceeedings, Washington State University, pp. 39-45. 

Grech, C. (2004) Does Chilean needle grass have any feed value? Chilean needle grass grazing management – interim results. Under Control – 

Pest Plant and Animal Management News 28, 9. 

Grech, C. (2005a) Getting the taste for Chilean needle grass – results of grazing trials. Under Control – Pest Plant and Animal Management News 

29, 8. 

Grech, C. (2005b) Spread the good seed – NOT the bad. Under Control – Pest Plant and Animal Management News 29, 8. 

187

Grech, C. (2007a) Grazing management for the long term utilisation and control of Chilean needle grass (Nassella neesiana). PhD thesis, School 

of Rural Science and Agriculture, University of New England, Armidale, NSW. 

Grech, C. (2007b) Response of Chilean needle grass (Nassella neesiana) and phalaris (Phalaris aquatica) seedlings and mature plants to changes 

in soil phosphorus, nitrogen and soil pH. Proceedings Third Victorian Weed Conference, ’Earth Wind Fire Water and Weeds’, Bendigo, Victoria, 

3-4 October 2007, Weed Society of Victoria, p. 77. 

Grech, C., Bennett, H., Butler, K. and McLaren, D. (2009a) Effects of flypropanate on non-target species – glasshouse. Proceedings Fourth 

Victorian Weed Conference ‘Plants Behaving Badly in Agriculture and the Environment’, Geelong, Victoria, 7-8 October 2009, Weed Society of 

Victoria, p. 57. 

Grech, C., Gaur, S. and McLaren, D. (2009c) Integrating herbicide wipers for the the management of Chilean needle grass (Nassella neesiana). 

Proceedings Fourth Victorian Weed Conference ‘Plants Behaving Badly in Agriculture and the Environment’, Geelong, Victoria, 7-8 October 

2009, Weed Society of Victoria, pp. 102-103. 

Grech, C., McLaren, D. and Butler, K. (2009b) Effect of low rtes of flupropanate for control of Chilean needle grass seedlings. Proceedings 

Fourth Victorian Weed Conference ‘Plants Behaving Badly in Agriculture and the Environment’, Geelong, Victoria, 7-8 October 2009, Weed 

Society of Victoria, p. 104. 

Grech, C., McLaren, D., Chapman, D. and Sindel, B.M. (2004) Effect of clipping and fertiliser addition on the feed value of Chilean needle grass 

(Nassella neesiana) during reproductive growth stages. In B.M. Sindel and S.B. Johnson (Eds.), 14th Australian Weeds Conference Papers and 

Proceedings. Weed Society of New South Wales, Sydney, p. 110. 

Grech, C., McLaren, D., Chapman, D. and Sindel, B.M. (2005) Chilean needle grass (Nassella neesiana) – integrated grazing for success. In 

Proceedings of the Tussock Terminators Research Forum, Lake Hume Resort, Albury, NSW, 2-3 November 2005, Cooperative Research Centre 

for Australian Weed Management, National Serrated Tussock Management Group, National Chilean Needle Grass Taskforce, Natural Heritage 

Trust, Department of Primary Industries Victoria, New South Wales Department of Primary Industries, pp. 11-14. 

Greene, H.C. and Cummins, G.B. (1958) A synopsis of the Uredinales which parasitize grasses of the genera Stipa and Nassella. Mycologia 50, 

6-36. 

Green, K. and Osborne, W. (1994) Wildlife of the Australian Snow-Country. Reed Books, Chatswood, NSW. 

Greenslade, P. (1994) A study of the use of indicator groups in assessing the conservation value of native grassland sites in the ACT and NSW. 

Project Report to the Australian Capital Territory Parks and Conservation Service, May, 27 pp. 

Gregg, P. (1997) Insect and mite pests of pastures. In: J.V.Lovett and J.M. Scott (Eds.), Pasture Production and Management, Inkata Press, 

Melbourne, pp. 231-253. 

Grice, A.C. (1993) Ecology of Stipa and Aristida in native pastures of western New South Wales. In E.S. Delfosse (Ed.), Pests of Pastures. Weed, 

Invertebrate and Disease Pests of Australian Sheep Pastures. CSIRO Information Services, Melbourne, pp. 158-162. 

Grice, A.C. (2004a) Weeds and the monitoring of biodiversity in Australian rangelands. Austral Ecology 19, 51-58. 

Grice, A.C. (2004b) Perennial grass weeds in Australia: impacts, conflicts of interest and management issues. Plant Protection Quarterly 19(2), 

42-47. 

Grice, A.C. (2006) The impacts of invasive plant species on the biodiversity of Australian rangelands. The Rangeland Journal 28, 27-35. 

Grice, A.C. and Ainsworth, N. (2003) Sleeper weeds – a useful concept? Plant Protection Quarterly 18(1), 35-39. 

Grice, A.C., Field, A.R. and McFadyen, R.E.C. (2004) Quantifying the effects of weeds on biodiversity: beyond Blind Freddy’s test. In B.M. 

Sindel and S.B. Johnson (Eds.), 14th Australian Weeds Conference Papers and Proceedings. Weed Society of New South Wales, Sydney, pp. 

464-468. 

Griffen, G.F., Stafford Smith, D.M., Morton, S.R., Allan, G.E., Masters, K.A. and Preece, N. (1989) Status and implications of the invasion of 

tamarisk (Tamarix aphylla) on the Finke River, Northern Territory, Australia. Journal of Environmental Management 29, 297-315. 

Gross, G.F. (1983) Myriapods, insects and allied forms. In M.J.Tyler, C.R.Twidale, J.K.Ling and J.W. Holmes (Eds.), Natural History of the 

South East. Royal Society of South Australia Inc., pp. 205-213. 

Gross, G.F. (1991a) Superfamily Lygaeoidea. In CSIRO, The Insects of Australia. A Textbook for Students and Research Workers. 2nd Ed., 

Melbourne University Press, Carlton, p. 499. 

Gross, G.F. (1991b) Superfamily Coreoidea. In CSIRO, The Insects of Australia. A Textbook for Students and Research Workers. 2nd Ed., 


Gross, G.F. and Cassis, G. (1991) Infraorder Cimicomorpha. In CSIRO, The Insects of Australia. A Textbook for Students and Research Workers. 

2nd Ed., Melbourne University Press, Carlton, pp. 488-491. 

Groves, C.P. (1989) Bovidae. In D.M. Walton and B.J. Richardson (Eds.), Fauna of Australia Volume 1B. Mammalia. Australian Government 

Publishing Service, Canberra, pp. 1061-1066 

Groves, R.H. (1965) Growth of Themeda australis tussock grassland at St. Albans, Victoria. Australian Journal of Botany 13, 291-302. 

Groves, R.H. (1975) Growth and development of five populations of Themeda australis in response to temperature. Australian Journal of Botany 

23, 951-963. 

Groves, R.H. (1979) The status and future of Australian grasslands. New Zealand Journal of Ecology 2, 76-81. 

Groves, R.H. (1999) Sleeper weeds. In A.C. Bishop, M. Boersma and C.D. Barnes (Eds.), 12th Australian Weeds Conference Papers and 

Proceedings, Tasmanian Weed Society, Devenport, pp. 632-636. 

Groves, R.H. (2002) The impacts of alien plants in Australia. In D. Pimental (Ed.), Biological Invasion. Economic and Environmental Costs of 

Alien Plant, Animal and Microbe Species. CRC Press, Boca Raton, pp. 11-23. 

Groves, R.H. (2004) CAWS Oration. Weed management at the border, at the garden fence and in the bush. In B.M. Sindel and S.B. Johnson 

(Eds.), 14th Australian Weeds Conference Papers and Proceedings. Weed Society of New South Wales, Sydney, pp. 1-9. 

Groves, R.H., Austin, M.P. and Kaye, P.E. (2003a) Competition between Australian native and introduced grasses along a nutrient gradient. 

Austral Ecology 28, 491-498. 

Groves, R.H. (Convenor), Hosking, J.R., Batianoff, G.N., Cooke, D.A., Cowie, I.D., Johnson, R.W., Keighery, G.J., Lepschi, B.J., Mitchell, A.A., 

Moerkerk, M., Randall, R.P., Rozenfelds, A.C., Walsh, N.G. and Waterhouse, B.M. (2003b) Weed Categories for Natural and Agricultural 

Ecosystem Management. Bureau of Rural Sciences, Canberra, ACT. 

Groves, R.H., Keraitis, K. and Moore, C.W.E. (1973) Relative growth of Themeda australis and Poa labillardieri in pots in response to 

phosphorus and nitrogen. Australian Journal of Botany 21, 1-11. 

188

Groves, R.H. and Whalley, R.D.B. (2002) Grass and grassland ecology in Australia. In: K. Mallett and A.E. Orchard (Eds.), Flora of Australia 

Volume 43. Poaceae 1. Introduction and Atlas. Australian Biological Resources Survey/CSIRO, Melbourne, pp. 157-182. 

Groves, R.H. and Willis, A.J. (1999) Environmental weeds and loss of native plant biodiversity: some Australian examples. Australian Journal of 

Environmental Management 6, 164-171. [Reference not seen - obtained] 

Gullen, P.K., Forbes, S.J., Earl, G.E., Barley, R.H. and Walsh, N.G. (1985) Vegetation of South and Central Gippsland. Muelleria 6, 97-146. 

Gurr, G.M., Aston, S.C., Wheatley, W.W., Woodward, R.A., Hunt, E.R. and Morrison, C.G.(1996) Management of pasture weeds on 

thoroughbred horse stud farms in eastern Australia. In R.C.H. Shepherd (Ed.), Proceedings Eleventh Australian Weeds Conference, Melbourne 

30 September - 3 October 1996, Weed Science Society of Victoria, Frankston, pp. 301-304. 

Haase, P., Pugnaire,F.I and Incoll, L.D. (1995) Seed production and dispersal in the semi-arid tussock grass Stipa tenacissima L. during masting. 

Journal of Arid Environments 31, 55-65. 

Hadden, S. (1995) Distribution, status and habitat requirements of the Striped Leglesss Lizard Delma impar (Fischer). Flora and Fauna Branch, 

Department of Conservation and Natural Resources, Final Report to the Australian Nature Conservation Agency, April, xx pp. 

Hadden, S. (1997) The composition and abundance of the Coleoptera, Araneida and Formicidae fauna of two long-grazed native grassland 

remnants of Victoria and the effects of cessation of grazing. Paper presented to the Research Advisory Group of the Victorian Grassy Ecosystem 

Group Seminar 11 October. 

Hadden , S.A. (1998) Composition and ecology of the flora and fauna of remnant native grasslands of the Western Basalt Plains and Northern 

Plains of Victoria - implications for management on private property. PhD thesis, School of Science, University of Ballarat, Victoria. 

Hadden, S.A. (2002) The mammal fauna of remnant native grasslands of the Western Basalt Plain and Northern Plains of Victoria. Victorian 

Naturalist 119(1), 14-20. 

Hadden, S.A. and Westbrooke, M.E. (1999) A comparison of the Coleoptera, Araneae and Formicidae fauna in a grazed native grassland remnant 

of Victoria. In: W. Ponder and D. Lunney (Eds.), The Other 99%. The Conservation and Biodiversity of Invertebrates. Transactions of the Royal 

Zoological Society of New South Wales, Mosman, pp. 101-106. 

Hagiwara, Y, Yamashita, M. and Sawada, H. (2009) Invasion status of Eragrostis curvula in the Lycaeides argyrognomon praeterinsularis 

habitat along the Ave River, and the effects of its removal. Japanese Journal of Conservation Ecology 14, 25-35. 

Hamilton, M.A., Murray, B.R., Cadotte, M.W., Hose, G.C., Baker, A.C., Harris, C.J. and Licari, D. (2005) Life-history correlates of plant 

invasiveness at regional and continental scales. Ecology Letters 8, 1066-1074. 

Hansford, M. (2006) Lobed needle grass: management strategies for a Priority Agricultural Sleeper Weed, National Environmental Alert List 

Weed and State Prohibited Weed in Victoria. In C. Preston, J.H. Watts and N.D. Crossman (Eds.), 15th Australian Weeds Conference Papers and 

Proceedings. Weed Management Society of South Australia Inc., Adelaide, pp. 645-648. 

Harborne, J.B. and Williams, C.A. (1986) Flavonoid patterns of grasses. In T.S. Soderstrom, K.W. Hilu, C.S. Campbell and M.E. Barkworth 

(Eds.), Grass Systematics and Evolution, Smithsonian Institution Press, Washington, USA, pp. 107-113. 

Hardy, R.J. (1976a) Observations on the pasture beetle Saulostomus villosus Waterhouse (Scarabaeidae: Rutelinae). Journal of the Australian 

Entomological Society 15, 433-440. 

Hardy, R.J. (1976b) The biology and behaviour of the pasture beetle Scitala sericans Erichson (Scarabaeidae: Melolonthinae). Journal of the 

Australian Entomological Society 15, 281-284. 

Harradine, A.R. (1991) The impact of pampas grasses as weeds in southern Australia. Plant Protection Quarterly 6(3), 111-115. 

Harris, I. (1990) Letter to M.J. Bartley, R.F. Parsons and N.H. Scarlett, Department of Botany, LaTrobe University re Stipa neesiana from 

Melbourne Regional Office, Department of Conservation, Forests and Lands. 27 August, 1 p. 

Harris, K.M. (1979) Descriptions and host ranges of the sorghum midge, Contarinia sorghicola (Coquillet) (Diptera: Cecidomyiidae), and of 

eleven new species of Contarinia reared from Gramineae and Cyperaceae in Australia. Bulletin of Entomological Research 69, 161-182. 

Hartl, D.L. and Clark, A.G. (1989) Principles of Population Genetics. 2nd Ed., Sinauer Associates Inc., Sunderland, Massachusetts. 

Hartley, M.J. (1994) Dalapon resistant individuals in a susceptible population of Chilean needle grass. Proceedings of the 47th New Zealand 

Plant Protection Conference, pp. 75-79. 

Hattersley, P.W. (1986) Variations in photsynthetic pathway. In T.S. Soderstrom, K.W. Hilu, C.S. Campbell and M.E. Barkworth (Eds.), Grass 


Hayes, K.R. and Barry, S.C. (2008) Are there any consistent predictors of invasion success? Biological Invasions 10, 483-506. 

Hayman, D.L. (1960) The distribution and cytology of the chromosome races of Themeda australis in southern Australia. Australian Journal of 

Botany 8(1), 58-68. 

Hayward, I.M. and Druce, G.C. (1919) The Adventive Flora of Tweedside.T. Buncle and Co., Arbroath, Scotland. 

Heath, E. (1968) New Zealand Farm Pests In Colour. A Guide to Identification. A.H. and A.W. Reed, Wellington, New Zealand. 

Heil, G.W., Werger, M.J.A., de Mol, W., van Dam, D. and Heijne, B. (1988) Capture of atmospheric ammonium by grassland canopies. Science 

239(4841), 764-765. 

Henderson, L. (2001) Alien Weeds and Invasive Plants. Plant Protection Research Institute Handbook No. 12, Agricultural Research Council, 

South Africa. 

Henderson, M. (1999) How do urban grasslands respond to fire and slashing? Effects on gaps and weediness. In V. Craigie and C. Hocking 

(Eds.), Down to Grass Roots. Proceedings of a Confence on Management of Grassy Ecosystems, 9 and 10 July 1998, Victoria University, St 

Albans, Vic., pp. 38-43. 

Herbold, B. and Moyle, P.B. (1986) Introduced species and vacant niches. American Naturalist 128(5), 751-760. 

Heyligers, P.C. (1986) Impact of introduced grasses on foredunes in south-eastern Australia. In R.H. Groves and J.J. Burdon (Eds.), Ecology of 

Biological Invasions: An Australian Perspective. Australian Academy of Science, Canberra, p. 152. 

Hill, L. and Michaelis, F.B. (1988) Conservation of insects and related wildlife. Report on the Australian National Parks and Wildlife Service 

survey of insect conservation needs. Australian National Parks and Wildlife Service Occasional Paper No. 13, 40 pp. 

Hill, R.D., Quigley, P.E. and Ward, G.N. (1993) Weeds, diseases and arthropod pests of sheep pasture in southwestern Victoria. In E.S. Delfosse 

(Ed.), Pests of Pastures. Weed, Invertebrate and Disease Pests of Australian Sheep Pastures. CSIRO Information Services, Melbourne, pp. 47- 

51. 

189

Hills, A. and Boekel, R. (1996) Large-fruit Groundsel Senecio macrocarpus. Flora and Fauna Guarantee Action Statement No. 68, Department 

of Natural Resources and Environment, East Melbourne, 10 pp. 

Hills, A. and Boekel, R. (2003) Large-fruit Groundsel Senecio macrocarpus. Flora and Fauna Guarantee Action Statement No. 68, Department 

of Sustainability and Environment, East Melbourne, 6 pp. 

Hills, E.S. (1967) The Physiography of Victoria. An Introduction to Geomorphology. Whitcombe and Tombs Pty Ltd, Melbourne. 

Hinkley, S. and New, T.R. (1997) Pitfall trapping for surveying ant assemblages: lessons from a study at Mount Piper, Victoria. Memoirs of the 

Museum of Victoria 56(2), 369-376. 

Hitchcock, A.S. and Chase, A. (1971) Manual of the Grasses of the United States. 2nd (revised) Ed., Dover Publications Inc., New York. 

Hobbs, R.J. (1989) The nature and effects of disturbance relative to invasions. In J.A.Drake, H.A. Mooney, F. di Castri, R.H. Groves, F.J. Kruger, 

M. Rejmanek and M. Williamson (Eds.), Biological Invasions: a Global Perspective, Scientific Committee on Problems of the Environment 

(SCOPE) 37, John Wiley and Sons, Chichester, UK, pp. 389-405. 

Hobbs, R.J. (1991) Disturbance as a precursor to weed invasion in native vegetation. Plant Protection Quarterly 6(3), 99-104. 

Hobbs, R.J. and Huenneke, L.F. (1992). Disturbance, diversity, and invasion: implications for conservation. Conservation Biology 6, 324-337. 

Hocking, C. (1998) Land management of Nassella areas - implications for conservation. Plant Protection Quarterly 13(2), 87-91. 

Hocking, C. (2002) Impacts and control of exotic stipoid grasses in Australia- what have we learnt since 1998? Plant Protection Quarterly 17(3), 

83-86. 

Hocking, C. (2005a) People as the forgotten ecological element of lowland grassland ecology – new perspectives on values and management. 

Plant Protection Quarterly 20(1), 9-11. 

Hocking, C. (2005b) Chilean needle grass – potential impacts and control. In Proceedings of the Tussock Terminators Research Forum, Lake 

Hume Resort, Albury, NSW, 2-3 November 2005, Cooperative Research Centre for Australian Weed Management, National Serrated Tussock 

Management Group, National Chilean Needle Grass Taskforce, Natural Heritage Trust, Department of Primary Industries Victoria, New South 

Wales Department of Primary Industries, pp. 27-31. 

Hocking, C. (2007) Grassy weeds as fiery competitors or phantom companions? – lessons learnt from conservation and land management. 

Proceedings Third Victorian Weed Conference, ’Earth Wind Fire Water and Weeds’, Bendigo, Victoria, 3-4 October 2007, Weed Society of 

Victoria, pp. 55-57. 

Hocking, C. (2009) Removal of Chilean needle grass, Nassella neesiana, from roadsides across Australia using wick wiping. Proceedings Fourth 

Victorian Weed Conference ‘Plants Behaving Badly in Agriculture and the Environment’, Geelong, Victoria, 7-8 October 2009, Weed Society of 

Victoria, pp. 63-66. 

Hocking, C. and Mason, B. (2001) Ecology of repair and management of native grasses. Stipa Native Grass Association Second National 

Conference: Our Valuable Native Grasslands, Better Pastures Naturally, Dookie College, University of Melbourne, 27-28 November. 

Hogan, T.W. (1940) The Argentine ant in Victoria. A report on survey and control measures follwing its discovery in Melbourne. Journal of 

Agriculture, Victoria 38, 486-493. 

Honaine, M.F., Osterrieth, M.L. and Zucol, A.F. (2009) Plant communities and soil phytolith assemblages relationship in native grasslands from 

southeastern Buenos Aires province, Argentina. Catena 76, 89-96;lum. 

Honaine, M.F., Zucol, A.F. and Osterrieth, M.L. (2006) Phytolith assemblages and systematic associations in grassland species of the southeastern 

Pampean Plains, Argentina. Annals of Botany 98, 1155-1165. 

Holland, K.D., McDonnell, M.J. and Williams, N.G. (2007) Abundance, species richness and feeding preferences of introduced molluscs in 

native grasslands of Victoria, Australia. Austral Ecology 32, 626-634. 

Holloway, B.A. (1982) Fauna of New Zealand Number 3. Anthribidae (Insecta: Coleoptera). Science Information Division, DSIR, Wellington, 

New Zealand, 264 pp. 

Hooper, R. (2006) Something in the air. New Scientist 189(2535), 40-43. 

Hope, G.S. (1994) Quaternary vegetation. In R.S. Hill (Ed.), History of the Australian Vegetation: Cretaceous to Recent, Cambridge University 

Press, Cambridge, UK, pp. 368-389. 

Hopper, K.R. (2001) Research needs concerning non-target impacts of biological control introductions. In E. Wajnberg, J.K. Scott and P.C. 

Quimby (Eds.), Evaluating Indirect Ecological Effects of Biological Control, CABI Publishing, Wallingford, UK, pp. 39-56. 

Howden, A.T. (1986) Notes on the biology of adult and immature Amycterinae (Coleoptera, Curculionidae). Proceedings of the Linnaean Society 

of New South Wales 109(2), 91-105. 

Hubbard, C.E. (1968) Grasses. A Guide to their Structure, Identification, Use and Distribution in the British Isles. Revised Ed., Penguin Books, 

Harmondsworth, Middlesex. 

Huenneke, L.F., Hamburg, S.P., Koide, R., Mooney, H.A. and Vitousek, P.M. (1990) Effects of soil resources on plant invasion and community 

structure in Californian serpentine grassland. Ecology 71(2), 478-491. 

Hume, I.D., Jarman, P.J., Renfree, M.B. and Temple-Smith,P.D. (1989) Macropodidae. In D.M. Walton and B.J. Richardson (Eds.), Fauna of 

Australia Volume 1B. Mammalia. Australian Government Publishing Service, Canberra, pp. 679-715. 

Humphries, R.K. and Webster, A. (1992) Button Wrinklewort Rutidosis leptorrhynchoides. Flora and Fauna Guarantee Action Statement No. 

28, Department of Conservation and Environment, East Melbourne, 6 pp. 

Humphries, R.K. and Webster, A. (2003) Button Wrinklewort Rutidosis leptorrhynchoides. Flora and Fauna Guarantee Action Statement No. 

28, Department of Sustainabilityand Environment, East Melbourne, 6 pp. 

Hunt, P. (1996) Action urged on Chilean needle grass. The Weekly Times, 18 December, p. 7. 

Hunziker, J.H. and Stebbins, G.L. (1986) Chromosomal evolution in the Gramineae. In T.S. Soderstrom, K.W. Hilu, C.S. Campbell and M.E. 


Hurrell, G.A. and Bourdôt, G.W. (1988) A comparison of granular herbicides for Chilean needle grass (Stipa neesiana) control. Proceedings of 

the 41st New Zealand Weed and Pest Control Conference, pp. 235-237. 

Hurrell, G.A., Bourdôt, G.W. and Hunter, R.M. (1994) Effect of three soil fumigants on the viability of buried seeds of Chilean needle grass 

(Stipa neesiana). Proceedings of the 47th New Zealand Plant Protection Conference, pp. 80-83. 

Iaconis, L. (2003) Chilean needle-grass - why is it a Weed of National Significance? Under Control - Pest Plant and Animal Management News 

No. 23, pp. 6-7. 

190

Iaconis, L. (2004) Chilean needle grass research update. Under Control - Pest Plant and Animal Management News No. 28, p. 8. 

Iaconis, L. (2006a) Review of Progress Against the National Chilean Needle Grass Strategic Plan (2001-2006). Victorian Department of Primary 

Industries/National Chilean Needle Grass Taskforce, August, 36 pp. 

Iaconis, L. (2006b) Minutes of the meeting of the National Chilean Needle Grass Taskforce, Wodonga, Victoria, 26 October. 

Ingeme, Y. (2003) Basalt Greenhood Pterostylis basaltica. Flora and Fauna Guarantee Action Statement No. 137, Department of Sustainability 


Iriarte, J. (2006) Vegetation and climate change since 14,810 14 C yr B.P. in southeastern Uruguay and implications for the rise of early formative 

societies. Quaternary Research 65, 20-32. 

Irrigation Research and Extension Committee (1975) A statement on the effects of wildfowl on agriculture in the irrigation regions of the 

Riverina. In Proceedings of a Workshop on Agriculture, Forestry and Wildlife: Conflict or Coexistence? held at the University of New England, 17-19 

August 1975. University of New England, Armidale, pp.163-167. 

ITIS (Integrated Taxonomic Information System) (2006) Nassella neesiana. Department of Agriculture USA online database. http://www.it 

is.usda.org 

IUCN (International Union for Conservation of Nature and Natural Resources) (1988) 1988 IUCN Red List of Threatened Animals. IUCN 

Conservation Monitoring Centre, Gland, Switzerland and Cambridge, UK. 

Ivie, M.A. (2001) On the geographic origin of the wheat stem sawfly (Hymenoptera: Cephidae): a new hypothesis of introduction from 

northeastern Asia. American Entomologist 47, 84-97. 

Jacobs, S.W.L. and Everett, J. (1993) Nassella. In G.J. Harden (Ed.), Flora of New South Wales. Volume 4. (1st Ed.), New South Wales 

University Press, Kensington, pp. 638-639. 

Jacobs, S.W.L. and Everett, J. (1996) Austrostipa, a new genus, and new names for Australasian species formerly included in Stipa (Gramineae). 

Telopea 6(4), 579-595. 

Jacobs, S.W.L., Everett, J., Barkworth, M.E. and Hsiao, C. (2000) Relationships within the stipoid grasses (Gramineae). In S.W.L. Jacobs,and J. 

Everett (Eds.) Grasses. Systematics and Evolution. International Symposium on Grass Systematics and Evolution (3rd: 1998). CSIRO Publishing, 

Collingwood, Vic., pp. 75-82. 

Jacobs, S.W.L., Everett, J., Connor, H.E. and Edgar, E. (1989) Stipoid grasses in New Zealand. New Zealand Journal of Botany 27, 569-582. 

Jacobs, S.W.L., Everett, J. and Torres, M.A. (1998) Nassella tenuissima (Gramineae) recorded from Australia, a potential new weed related to 

Serrated Tussock. Telopea 8, 41-46. 

Jallow, M.F.A., Dugassa-Gobena, D. and Vidal, S. (2008) Influence of an endophytic fungus on host plant selection by a polyphagous moth via 

volatile spectrum changes. Arthropod-Plant Interactions 2, 53-62. 

James, A.I., Eldridge, D.J. and Hill, B.M. (2009) Foraging animals create fertile patches in an Australian desert shrubland. Ecography 32, 723- 

732. 

Jeanes, J.A. (1999a) Solanaceae. In N.G. Walsh and T.J. Entwisle (Eds.) (1994) Flora of Victoria Volume 4. Dicotyledons. Cornaceae to 

Asteraceae. Melbourne, Inkata Press, pp. 332-365. 

Jeanes, J.A. (1999b) Asteraceae. In N.G. Walsh and T.J. Entwisle (Eds.) (1994) Flora of Victoria Volume 4. Dicotyledons. Cornaceae to 

Asteraceae. Melbourne, Inkata Press, pp. 652-984. 

Jessop, J., Dashorst, G.R.M. and James, F.M. (2006) Grasses of South Australia. An Illustrated Guide to the Native and Naturalised Species. 

Wakefield Press, Kent Town, South Australia. Victorian Naturalist 111(4), 131-134. 

Johnson, B.L. (1972) Polyploidy as a factor in the evolution and distribution of grasses. In V.B. Youngner and C.M. McKell (Eds.), The Biology 

and Utilization of Grasses. Academic Press, New York, pp. 18-35. 

Johnson, C.N. (2009) Ecological consequences of Late Quaternary extinctions of megafauna. Proceedings of the Royal Society B 276, 2509-2519. 

Johnson, G. and Baker-Gabb, D. (1994) The Bush Thick-knee in Northern Victoria (Part 1): Conservation and Management. Arthur Rylah 

Institute for Environmental Research Technical Report Series No. 129. 

Johnson, K.A. and Jarman, P.J. (1975) Records of wildlife as pests in the Armidale district, 1812-1975. In Proceedings of a Workshop on 

Agriculture, Forestry and Wildlife: Conflict or Coexistence? held at the University of New England, 17-19 August 1975. University of New England, 

Armidale, pp. 26-32. 

Johnston, T.H. (1938) A census of the free-living and plant-parasitic nematodes recorded as occurring in Australia. Transactions of the Royal 

Society of South Australia 62(1), 149-167. 

Jones, M.M. (1991) Marine organisms transported in ballast water. A review of the Australian scientific position. Department of Primary Industries and 

Energy, Bureau of Rural Resources Bulletin No. 11, 48 pp. 

Jones, R.N.(1999a) The biogeography of the grasses and lowland grasslands of south-eastern Australia. In R.N. Jones (Ed.), The Great Plains 

Crash. Proceedings of a Conference on the Grasslands and Grassy Woodlands of Victoria. Victorian Institute of Technology, October 2nd and 

3rd, 1992, Indigenous Flora and Fauna Association and Victorian National Parks Association, pp. 11-18. 

Jones, R.N.(1999b) Natural and human influences on the distribution and extent of Victorian lowland grasslands. In R.N. Jones (Ed.), The Great 

Plains Crash. Proceedings of a Conference on the Grasslands and Grassy Woodlands of Victoria. Victorian Institute of Technology, October 2nd 

and 3rd, 1992, Indigenous Flora and Fauna Association and Victorian National Parks Association, pp. 19-39. 

Julien, M.H. and Griffiths, M.W. (Eds.) (1998) Biological Control of Weeds. A World Catalogueof Agents and their Target Weeds. 4th Ed., CABI 

Publishing, Wallingford, UK. 

Keane, P.J. (1994) Soil microflora and its importance in the restoration and management of vegetation. Victorian Naturalist 111(4), 131-134. 

Keane, R.M. and Crawley, M.J. (2002) Exotic plant invasions and the enemy release hypothesis. Trends in Ecology and Evolution 17(4), 164- 

170. 

Keith, D. (2004) Ocean Shores to Desert Dunes. The Native Vegetation of New South Wales and the ACT. Department of Environment and 

Conservation (NSW), Hurstville. 

Kellerman, T.S., Coetzer, J.A.W. and Naude, T.W. (1988) Plant Poisonings and Mycotoxicoses of Livestock in Southern Africa. Oxford 

University Press, New York. [Reference not seen]. 

Kelly, D., McKone, M.J., Batchelor, K.J. and Spence, J.R. (1992) Mast seeding of Chionochloa (Poaceae) and pre-dispersal seed predation by a a 

specialist fly (Diplotaxa, Diptera: Chloropidae). New Zealand Journal of Botany 30, 125-133. 

191

Kelly, D. and Skipworth, J.P. (1984) Tradescantia fluminensis in a Manawatu (New Zealand) forest: I. Growth and effects on regeneration. New 

Zealand Journal of Botany 22, 393-397. 

Kelly, D., Turnbull, M.H., Pharis, R.P. and Sarfati, M.S. (2008) Mast seeding, predator satiation, and temperature cues in Chionochloa (Poaceae). 

Population Ecology 50, 343-355. 

Kershaw, A.P., Martin, H.A. and McEwen Mason, J.R.C. (1994) The Neogene: a period of transition. In R.S. Hill (Ed.), History of the Australian 

Vegetation: Cretaceous to Recent, Cambridge University Press, Cambridge, UK, pp. 299-327. 

Kershaw, P., Quilty, P.G., David, B., Van Huet, S. and McMinn, A. (2000) Palaeobiogeography of the Quaternary of Australasia. In J.A. Wright, 

G.C. Young, J.A. Talent and J.R. Laurie (Eds.), Palaeobiogeography of Australasian Faunas and Floras, Memoir 23 of the Association of 

Australian Palaeontologists, Canberra, pp. 471-515. 

Kershaw, R.C. (1991) Snail and Slug Pests of Tasmania. Queen Victoria Museum and Art Gallery, Launceston. 

Key, K.H.L. (1973) The primary types of the Australian Eumastacidae (Orthoptera: Eumastacoidea). Australian Journal of Zoology 

Supplementary Series No. 19. 

Key, K.H.L. (1976) A generic and suprgeneric classification of the Morabinae (Orthoptera: Eumastacidae), with description of the type species 

and a bibliography of the subfamily. Australian Journal of Zoology Supplementary Series No. 37. 

Key, K.H.L. (1978) The conservation status of Australia’s insect fauna. Australian National Parks and Wildlife Service Occasional Paper No. 1 

1978, Australian Government Publishing Service, Canberra, 24 pp. 

Key, K.H.L. (1981) Species, parapatry, and the morabine grasshoppers. Systematic Zoology 30(4), 425-458. 

Khidir, H.H., Eudy, D.M., Porras-Alfaro, A., Herrera, J., Natvig, D.O. and Sinsabaugh, R.L. (2009) A general suite of fungal endophytes 

dominate the roots of two dominant grasses in a semiarid grassland. Journal of Arid Environments doi:10.1016/j.jaridenv.2009.07.014 

King, S.A. and Buckney, R.T. (2001) Exotic plants in the soil-stored seed bank of urban bushland.. Australian Journal of Botany 49, 717-720. 

King, S.A. and Buckney, R.T. (2002) Invasion of exotic plants in nutrient-enriched urban bushland. Austral Ecology 27, 573-583. 

Kirkpatrick, J. (1995) The status of lowland temperate grasslands in south-eastern Australia. In S. Sharp and R. Rehwinkel (Eds.), Management of 

Relict Lowland Grasslands. Proceedings of a Workshop and Public Seminar September 24 and 25, 1993. ACT Parks and Conservation Service 

Conservation Series No. 8, pp. 75-79. 

Kirkpatrick, J.B. (2007) Collateral benefit: unconscious conservation of threatened plant species. Australian Journal of Botany 55, 221-224. 

Kirkpatrick, J., McDougall, K. and Hyde, M. (1995) Australia’s Most Threatened Ecosystem. The Southeastern Lowland Native Grasslands. 

SurreyBeatty and Sons and the World Wide Fund for Nature Australia, Chipping Norton, NSW. 

Kleinschmidt, R.P. (1965) Some leaf-mining Agromyzidae (Diptera) from Queensland. Journal of the Entomological Society of Queensland 4, 1- 

2. 

Klink, C.A. (1996) Germination and seedling establishment of two natives and one invading African grass species in the Brazilian cerrado. 

Journal of Tropical Ecology 12, 139-147. 

Knapp, A.K. and Seastedt, T.R. (1986) Detritus accumulation limits productivity in tall-grass prairies. BioScience 36, 662-668. 

Kolesik, P., Sarfati, M.S., Brockerhoff, E.G. and Kelly, D. (2007) Description of Eucalyptodiplosis chionochloae sp. nov., a cecidomyiid feeding on 

inflorescences of Chionochloa (Poaceae) in New Zealand. New Zealand Journal of Zoology 34, 107-115. 

Krebs, C.J. (1985) Ecology. The Experimental Analysis of Distribution and Abundance. 3rd Ed., Harper and Row Publishers, New York. 

Kreyling, J., Beierkuhnlein, C., Ellis, L. and Jentsch, A. (2008) Invasibility of grassland and heath communities exposed to extreme weather events – 

additive effects of diversity resistance and fluctuating physical environments. Oikos 117, 1542-1554. 

Kukolic, K. (1994) Survey for the Striped Legless Lizard Delma impar during 1993 at the National Museum of Australia site, Yarramundi Reach, 

ACT. Report to the National Capital Planning Authority. Wildlife Research Unit, ACT Parks and Conservation Service, Tuggeranong, ACT, 

December, v + 28 pp. 

Kuschel, G. (1972) The foreign Cuculionoidea established in New Zealand (Insecta: Coleoptera). New Zealand Journal of Science 15, 273-289. 

Kuschel, G. (1990) Beetles in a suburban environment: a New Zealand case study. The identity and status of Coleoptera in the natural and 

modified habitats of Lynfield, Auckland (1974-1989). New Zealand Department of Scientific and Industrial Research Plant Protection Report 

No. 3, Auckland, 119 pp. 

Kutt, A.S., Coulson, G. and Wainer, J. (1998) Diet of the Striped Legless Lizard Delma impar (Squamata: Pygopodidae) in a western (basalt) 

plains grassland, Victoria. Australian Zoologist 30(4), 412-418. 

Lange, R.T. (1983) Native vegetation. In M.J.Tyler, C.R.Twidale, J.K.Ling and J.W. Holmes (Eds.), Natural History of the South East. Royal 

Society of South Australia Inc., pp. 95-113. 

Lawton, J.H. and Schroder, D. (1977) Effects of plant type, size of geographical range and taxonomic isolation on number of insects species 

associated with British plants. Nature 265, 137-140. 

Lee, C.E. (2002) Evolutionary genetics of invasive species. Trends in Ecology and Evolution 17(8), 386-391. 

Leeper, G.W. (1970) Soils. In G.W. Leeper (Ed.) The Australian Environment. 4th Ed., CSIRO and Melbourne University Press, Carlton, pp. 21- 

31. 

Leigh, E.G. (2007) Neutral theory: a historical perspective. Journal of Evolutionary Biology 20, 2075-2091. 

Leigh, J.H. and Briggs, J.D. (1992) Threatened Australian Plants: Overview and Case Studies. Australian National Parks and Wildlife Service, 

Canberra. 

Leigh, J.H. and Mulham, W.E. (1965) Pastoral Plants of the Riverine Plain. The Jacaranda Press, Brisbane. 

LeJeune, K.D. and Seastedt, T.R. (2001) Centaurea species: the forb that won the west. Conservation Biology 15(6), 1568-1574. [Not used?] 

Lemerle, D. and Murphy, C.E. (2000) Cultural management methods. In B.M. Sindel (Ed.), Australian Weed Management Systems, R.G. and F.J. 

Richardson, Melbourne, pp. 123-138. 

Lenz, T.I., Moyle-Croft, J.L. and Facelli, J.M. (2003) Direct and indirect effects of exotic annual grasses on species composition of a South 

Australian grassland. Austral Ecology 28,23-32. 

Leslie, A.J. and Spotila, J.R. (2001) Alien plant threatens Nile crocodile (Crocodylus niloticus) breeding in Lake St. Lucia, South Africa. 

Biological Conservation 98, 347-355. 

Levin, D.A. (2006) Ancient dispersals, propagule pressure, and species selection in flowering plants. Systematic Botany 31(3), 443-448. 

192

Levine, J.M., Adler, P.B. and Yelenik, S.G. (2004) A meta-analysis of biotic resistance to exotic plant invasions. Ecology Letters 7, 975-989. 

Levine, J.M., Vilà, M., D’Antonio, C.M., Dukes, J.S.,Grigulis, K. and Lavorel,S. (2003) Mechanisms underlying the impacts of exotic plant 

invasions. Proceedings of the Royal Society of London Series B 270, 775-781. 

Liebert, A. (1996) Chilean Needle Grass: Stipa neesiana. Assessment Report to the North Central CALP Region. Department of Conservation 

and Natural Resources North Central Catchment and Land Protection Region, unpublished report, February, 39 pp. 

Lock, J.M. and Millburn, T.R. (1970) The seed biology of Themeda triandra Forsk. in relation to fire. In E. Duffey and A.S. Watt (Eds.), The 

Scientific Management of Animal and Plant Communities for Conservation. Blackwell Scientific Publications, Oxford, UK, pp. 337-349. 

Lockwood, J.L., Hoopes, M.F. and Marchetti, M.P. (2007) Invasion Ecology. Blackwell Publishing, Malden, Massachusetts, USA. 

Lockwood, J.L., Cassey, P. and Blackburn, T.M. (2009) The more you introduce the more you get: the role of colonization pressure and 

propagule pressure in invasion ecology. Diversity and Distributions 15, 904-910. 

Long, J.L. (2003) Introduced Mammals of the World. Their History, Distribution and Influence. CABI Publishing and CSIRO Publishing, 

Collingwood, Victoria, and Wallingford, UK 

Longhi-Wagner, H.M. and Zanin, A. (1998) Padrões de distribuição geográfica das espécies de Stipa L. (Poaceae-Stipeae) ocorrentes no Brasil. 

Revista Brasiliera de Botânica 21(2), 167-175. 

Lonsdale, W.M. (1994) Inviting trouble: introduced pasture species in northern Australia. Australian Journal of Ecology 19, 345-354. 

Lonsdale, W.M. (1999) Global patterns of plant invasions and the concept of invasibility. Ecology 80(5), 1522-1536. 

Lorentz, P.G. (1876) Vegetationsverhältnisse der agentinischen Republik. Buenos Aires. 

Low, T. (1989) Bush Tucker. Australia’s Wild Food Harvest. Angus and Robertson, Pymble, NSW. 

Low, T. (2002) The New Nature. Winners and Losers in Wild Australia. Viking/Penguin Group, Camberwell, Victoria. 

Lowe, A.D. (1954) A note on the death of nassella tussock Nassella trichotoma (Nees.) Hack., due to the depradations of Chlorochiton sp. New 

Zealand Entomologist 1(4), 15-16. 

Loyn, R.H. and French, K. (1991) Birds and environmental weeds in south-eastern Australia. Plant Protection Quarterly 6(3), 137-149. 

Lundgren, J.G. and Rosentrater, K.A. (2007) The strength of seeds and their destruction by granivorous insects. Arthropod-Plant Interactions 1, 

93-99. 

Lunt, I.D. (1990a) A floristic survey of the Derrimut Grassland Reserve, Melbourne,Victoria. Proceedings of the Royal Society of Victoria 102(1), 

41-52. 

Lunt, I.D. (1990b) The soil seed bank of a long-grazed Themeda triandra grassland in Victoria. Proceedings of the Royal Society of Victoria 

102(1), 53-57. 

Lunt, I.D. (1990c) Impact of an autumn fire on a long-grazed Themeda trianda (Kangaroo Grass) grassland: impications for management of 

invaded, remnant vegetations. Victorian Naturalist 107(2), 45-51. 

Lunt, I.D. (1990d) Species area curves and growth-form spectra for some herb-rich woodlands in western Victoria, Australia. Australian Journal 

of Ecology 15,155-161. 

Lunt, I.D. (1991) Management of remnant lowland grasslands and grassy woodlands for nature conservation: a review. Victorian Naturalist 

108(3), 56-66. 

Lunt, I.D. (1994) Variation in flower production of nine grassland species with time since fire, and implications for grassland management and 

restoration. Pacific Conservation Biology 1, 359-366. Surrey Beatty and Sons, Sydney. 

Lunt, I.D. (1995a) Seed longevity of six native forbs in a closed Themeda triandra grassland. Australian Journal of Botany 43, 439-449. 

Lunt, I.D. (1995b) European management of remnant grassy forests and woodlands in south-eastern Australia- past, present and future? Victorian 

Naturalist 112(6), 239-249. 

Lunt, I.D. (1995c) A flexible approach to grassland management based on plant life-forms and growth-forms. In S. Sharp and R. Rehwinkel 

(Eds.), Management of Relict Lowland Grasslands. Proceedings of a Workshop and Public Seminar, September 24 and 25 1993. ACT Parks and 

Conservation Service Conservation Series No. 8, pp. 112-121. 

Lunt, I.D. (1996) A transient soil seed bank for the yam-daisy Microseris scapigera. Victorian Naturalist 113, 16-19. 

Lunt, I.D. (1997) Effects of long-term vegetation management on remnant grassy forests and anthropogenic native grasslands in south-eastern 

Australia. Biological Conservation 81, 287-297. 

Lunt, I., Barlow, T. and Ross, J. (1998) Plains Wandering. Exploring the Grassy Plains of South-eastern Australia. Victorian National Parks 

Association and Trust for Nature (Victoria), Melbourne. 

Lunt, I.D. and Morgan, J.W. (1999a) Effect of fire frequency on plant competition at the Laverton North Grassland Reserve, Victoria. Victorian 

Naturalist 116(3), 84-89. 

Lunt, I.D. and Morgan, J.W. (1999b) Vegetation changes after 10 years of grazing exclusion and intermittent burning in a Themeda triandra 

(Poaceae) grassland reserve in south-eastern Australia. Australian Journal of Botany 47, 537-552. 

Lunt, I. and Morgan, J. (1999c) Lessons from grassland management of Laverton North and Derrimut Grassland Reserves: what can we learn for 

future grassland management? In V. Craigie and C. Hocking (Eds.), Down to Grass Roots. Proceedings of a Conference on Management of 

Grassy Ecosystems, 9 and 10 July 1998, Victoria University, St Albans, Vic., pp. 69-74. 

Lunt, I.D. and Morgan, J.W. (2000) Can competition from Themeda triandra inhibit invasion by the perennial exotic grass Nassella neesiana in 

native grasslands? Plant Protection Quarterly 15(3), 92-94. 

Lunt, I.D. and Morgan, J.W. (2002) The role of fire regimes in temperate lowland grasslands of southeastern Australia. In R.A. Bradstock, J.E. 

Williams and M.A. Gill (Eds.), Flammable Australia. The Fire Regimes and Biodiversity of a Continent. Cambridge University Press, Cambridge, 

UK, pp. 177-196. 

MacDougall, A.S. and Turkington, R. (2007) Does the type of disturbance matter when restoring disturbance-dependent grasslands? Restoration 

Ecology15(2), 263-272. 

Mack, M.C. and D’Antonio, C.M. (1998) Impacts of biological invasions on disturbance regimes. Trends in Ecology and Evolution 13(5), 195- 

198. 

Mack, R.N. (1989) Temperate grasslands vulnerable to to plant invasions: characteristics and consequences. In J.A. Drake, H.A. Mooney, F. di 

Castri, R.H. Groves, F.J. Kruger, M. Rejmanek and M. Williamson (Eds.), Biological Invasions: a Global Perspective, Scientific Committee on 

Problems of the Environment (SCOPE) 37, John Wiley and Sons, Chichester, UK, pp. 155-179. 

193

Mack, R.N. and Lonsdale, W.M. (2001) Humans as global plant dispersers: getting more than we bargained for. Bioscience 51(2), 95-102. 

Macphail, M.K., Alley, N.F., Truswell, E.M. and Sluiter, I.R.K. (1994) Early Tertiary vegetation: evidence from spores and pollen. In R.S. Hill 

(Ed.), History of the Australian Vegetation: Cretaceous to Recent, Cambridge University Press, Cambridge, UK, pp. 189-261. 

Madge, P.E. (1954) A field study on the biology of the underground grass caterpillar,Oncopera fasciculata (Walker) (Lepidoptera: Hepialidae), in 

South Australia. Australian Journal of Zoology 2, 193-204. 

Madin, R.W. (1993) Weed, invertebrate and disease pests of Australian sheep pastures- an overview. In E.S. Delfosse (Ed.), Pests of Pastures. 

Weed, Invertebrate and Disease Pests of Australian Sheep Pastures. CSIRO Information Services, Melbourne, pp. 3-20. 

Maguire, A. (2005) Mexican feather grass. New South Wales Department of Primary Industries Agfact P7.6.60, 4 pp. 

Maher, P.N. and Baker-Gabb, D.J. (1993) Surveys and conservation of the Plains-wanderer in northern Victoria. Arthur Rylah Institute for 

Environmental Research Technical Report Series No. 131. 

Mallett, K. and Orchard, A.E. (Eds.) (2002) Flora of Australia Volume 43. Poaceae 1. Introduction and Atlas. Australian Biological Resources 

Survey/CSIRO, Melbourne. 

Mansergh, I., Anderson, H. and Amos, H. (2006a) Victoria’s living natural capital - decline and replenishment 1800-2050. (Part 1). Victorian 

Naturalist 123(1), 4-28. 

Mansergh, I., Anderson, H. and Amos, H. (2006b) Victoria’s living natural capital - decline and replenishment 1800-2050. Part 2. The new 

millenium: replenishment. Victorian Naturalist 123(5), 288-313. 

Mansergh, I. and Seebeck, J. (2003) Extinct mammals 2. Flora and Fauna Guarantee Action Statement No. 14, Department of Sustainability and 

Environment, East Melbourne, 5 pp. 

Marco, C.M. (1950) Descripción parcial del Departmento Gualeguay (Provincia de Entre Rios) desde el punto de vista de su producción 

forragera. Darwiniana 9, 188-224. (Abstract: NGS, Forage Abstracts 21(3), 113-114, August 1951). 

Mares, M.A. and Rosenzweig, M.L. (1978) Granivory in North and South American desert: rodents, birds and ants. Ecology 59(2), 235-241. 

Marriott, N. and Marriot, J. (1998) Grassland Plants of South-eastern Australia. Bloomings Books, Melbourne. 

Marriott, P. (2004) Victorian Sun-moths, Lepidoptera: Castniidae. Victorian Entomologist 34(3), 32-33. 

Marriott, P. (2008) Moths of Victoria. Part 1. Silk Moths and Allies Bombycoidea. Entomological Society of Victoria, Melbourne. 

Marshall, A.J.(1968) Of the plume trade, and emu oil. In A.J. Marshall (Ed.), The Great Extermination. A Guide to Anglo-Australian Cupidity, 

Wickedness and Waste. Panther Books, London, pp. 55-89. 

Martin, H.A. (1994) Australian Tertiary phtyogeography: evidence from palynology. In R.S. Hill (Ed.), History of the Australian Vegetation: 

Cretaceous to Recent, Cambridge University Press, Cambridge, UK, pp. 104-142. 

Martín Osorio,V.E.,Wildpret de la Torre, W. and Reyes Betancort, J.A. (2000) Nassella neesiana (Trin. and Rupr.) Barkworth, una especie 

invasora en los espacios naturales protegidos de Canarias. Anuario del Instituto de Estudios Canarios 44, 35-46. 

Martinovský, J.O. (1980) Stipa L. In T.G. Tutin, V.H. Heywood, N.A. Burges, D.M. Moore, D.H. Valentine, S.M. Walters, D.A. Webb (Eds.), 

Flora Europaea Volume 5 Alismataceae to Orchidaceae (Monocotyledones). Cambridge University Pres, Cambridge, pp. 247-252. 

Mason, B.(1998) Replacement of Chilean needle grass (Nassella neesiana Trin. and Rupr.) by the native kangaroo grass (Themeda triandra 

Forssk.): managing native grasslands in Central Victoria using the ‘spray and hay’ method. MSc thesis, Victoria University, St. Albans. 

Mason, B. (2005) Can acetic acid by used to control Nassella spp. – modifications to the ‘spray and hay’ method. Plant Protection Quarterly 

20(3), 82-84. 

Mason, B.D. (2004) Directions for best management of Kangaroo Grass (Themeda triandra Forssk.) re-establishment in southeastern Australian 

native grassland remnants. PhD thesis, Victoria University, Melbourne. 

Mason, B. and Hocking, C. (2002) The potential for repair of exotic stipoid grass infested sites with kangaroo grass (Themeda triandra Forrsk.) 

with special reference to remnant native grasslands. Plant Protection Quarterly 17(3), 98-101. 

Mathison, M.J. (2004) Uses and impacts of perennial grasses in South Australia: an environmental perspective. Plant Protection Quarterly 19(2), 

50-51. 

Matthews, E.G. (1976) Insect Ecology. University of Queensland Press, St Lucia. 

Matthews, E.G. and Bouchard, P. (2008) Tenebrionid Beetles of Australia. Descriptions of Tribes, Keys to Genera, Catalogue of Species. 

Australian Biological Resources Study, Canberra. 

Matthews, E.G. and Reid, C.A.M. (2002) A Guide to the Genera of Beetles of South Australia Part 8. Polyphaga: Chrysomeloidea: 

Chrysomelidae. South Australian Museum, Adelaide. 

Matthews, J. (2006) The Chilean needle-grass invasion south-west Victoria. Under Control – Pest Plant and Animal Management News 33, 7. 

Matthews, M. (1999) Heliothine Moths of Australia. A Guide to Pest Bollworms and Related Noctuid Groups. Monographs on Australian 

Lepidoptera Volume 7. CSIRO Publishing, Collingwood. 

Mathis, W.N. (1989) Family Ephydridae. In N.L. Evenhuis (Ed.), Catalog of the Diptera of the Australasian and Oceanian Regions. Bishop 

Museum Press and E.J. Brill, Honolulu and Leiden, pp. 639-649. 

May, B.M. (1993) Larvae of Curculionoidea (Insecta: Coleoptera): a Systematic Overview. Fauna of New Zealand No. 28, Manaaki Whenua 

Press, Lincoln, New Zealand. 

May, B.M. (1994) An introduction to the immature stages of Australian Curculionoidea. In E.C. Zimmerman, Australian Weevils (Coleoptera: 

Curculionoidea) Volume II. Brentidae, Eurhynchidae, Apionidae and a Chapter on Immature Stages by Brenda May. CSIRO and Entomological 

Society of America, East Melbourne and Lanham, Maryland, pp. 365-703. 

Mayr, E. (1982) The Growth of Biological Thought. Diversity, Evolution and Inheritance. Harvard University Press, Cambridge, Massachustts, 

USA. 

McArdle, S.L., Nadolny, C. and Sindel, B.M. (2004) Invasion of native vegetation by Coolatai Grass Hyparrhenia hirta: impacts on native 

vegetation and management implications. Pacific Conservation Biology 10(1), 49-56. 

McBarron, E.J. (1976) Medical and Veterinary Aspects of Plant Poisons in New South Wales. Department of Agriculture New South Wales. 

McDonald, G. (1991) Development and survival of Mythimna convecta (Walker) and Persectania ewingii (Westwood) (Lepidoptera: Noctuidae) 

on cereal and grass hosts. Journal of the Australian Entomological Society 30, 295-302. 

194

McDougall, K.L. (1987) Sites of Botanical Significance in the Western Region of Melbourne. Department of Geography, University of 

Melbourne, report for Department of Conservation, Forests and Lands, Melbourne, 122 pp. 

McDougall, K.L. (1989) The re-establishment of Themeda triandra (Kangaroo Grass): implications for the restoration of grassland. Arthur Rylah 

Institute for Environmental Research Technical Report Series No. 89, 53 pp. 

McDougall, K.L. (1999) Grassland, grassland über alles. In R.N. Jones (Ed.), The Great Plains Crash. Proceedings of a Conference on the 

Grasslands and Grassy Woodlands of Victoria. Victorian Institute of Technology October 2nd and 3rd, 1992, Indigenous Flora and Fauna 

Association and Victorian National Parks Association, pp. 47-52. 

McDougall, K.L. and Kirkpatrick, J.B. (Eds.) (1994) Conservation of Lowland Native Grasslands in South-eastern Australia. World Wide Fund 

for Nature, Sydney, 187 pp. 

McDougall, K.L. and Morgan, J.W. (2005) Establishment of native grassland vegetation at Organ Pipes National Park near Melbourne, Victoria: 

vegetation changes from 1989 to 2003. Ecological Management and Restoration 6(1), 34-42. 

McDougall, K.L. and Walsh, N.G. (2007) Treeless vegetation of the Australian Alps. Cunninghamia 10(1), 1-57. 

McGregor, B. (2003) Eco-grazing – the use of diversified grazing ecosystems as part of integrated weed management. Weed Society of Victoria 

Proceedings First Biennial Conference,20-21 August 2003, pp. 33-36. 

McFarland, N. (1988) Portraits of South Australian Geometrid Moths. Allen Press Inc., Lawrence, Kansas, USA. 

McGowran, B., Archer, M., Bock, P., Darragh, T.A., Godthelp, H., Hageman, S., Hand, S.J., Hill, R., Li, Q., Maxwell, P.A., McNamara, K.J., 

Macphail, M, Mildenhall, D., Partridge, A.D., Richardson, J., Shafik, S., Truswell, E.M. and Warne, M. (2000) Australasian palaeobiogeography: 

the Palaeogene and Neogene record. In J.A. Wright, G.C. Young, J.A. Talent and J.R. Laurie (Eds.), Palaeobiogeography of Australasian Faunas 

and Floras, Memoir 23 of the Association of Australian Palaeontologists, Canberra, pp. 405-470. 

McIlroy, W., Brake, J., Strom, H.G., Lakeland, W.J. and O’Carroll, J.D. (1938). Report of Committee Appointed to Investigate Erosion in 

Victoria. Government Printer, Melbourne. 

McIntyre, M., Morcom, L., Hadden, S. and Butler, M. (2004) Small Milkwort Comesperma polygaloides. Flora and Fanua Guarantee Action 

Statement No. 96, Department of Sustainability and Environment, East Melbourne, 6 pp. 

McIntyre, S. (1993) The nature of exotic invasions in herbaceous vegetation of high and low species richness: implications for conservation and 

management. In J.T. Swarbrick, C.W.L. Henderson, R.J. Jettner, L. Streit and S.R. Walker (Eds.), 10th Australian Weeds Conference and 14th 

Asia Pacific Weed Society Conference, Brisbane, 6-10 September, pp. 466-470. 

McIntyre, S. (1995) Disturbance: does it have a role in the maintenance of rare grassland plants? In S. Sharp and R. Rehwinkel (Eds.), 

Management of Relict Lowland Grasslands. Proceedings of a Workshop and Public Seminar, September 24 and 25 1993. ACT Parks and 


McIntyre, S. and Lavorel, S. (1994a) How environmental and disturbance factors influence species composition in temperate Australian 

grasslands. Journal of Vegetation Science 5, 373-384. 

McIntyre, S. and Lavorel, S. (1994b) Predicting richness of native, rare, and exotic plants in response to habitat and disturbance variables across a 

variegated landscape. Conservation Biology 8, 521-531. 

McIntyre, S. and Lavorel, S. (2007) A conceptual model of land use effects on the structure and function of herbaceous vegetation. Agriculture, 

Ecosystems and Environment 119, 11-21. 

McIntyre, S., Lavorel, S. and Trémont, R.M. (1995) Plant life-history attributes: their relationship to disturbance response in herbaceous 

vegetation. Journal of Ecology 83, 31-44. 

McLaren, D.A., Coventry, R. and Thorp, J. (2002a) Stipoid grasses as Weeds of National Significance. Plant Protection Quarterly 17(3), 119- 

124. 

McLaren, D., Pritchard, G. and Gaur, S. (2005) Herbicide label changes for serrated tussock and needle grasses (Nassella spp.). Under Control – 

Pest Plant and Animal Management News 29, 14-15. 

McLaren, D.A., Morfe, T.A. and Weiss, J. (2002b) Distribution, economic impact and attitudes towards Chilean needlegrass (Nassella neesiana 

(Trin. and Rupr.) Barkworth) in Australia. In H. Spafford Jacob, J. Dodd and J.H. Moore (Eds.), 13th Australian Weeds Conference Papers and 

Proceedings, 8-13 September 2002, Plant Protection Society of Western Australia Inc., Perth, pp. 749-752. 

McLaren, D.A., Stajsic, V. and Gardener, M.R. (1998) The distribution and impact of South/North American stipoid grasses (Poaceae: Stipeae) in 

Australia. Plant Protection Quarterly 13(2), 62-70. 

McLaren, D.A., Stajsic, V. and Iaconis, L. (2004) The distribution, impacts and identification of exotic stipoid grasses in Australia. Plant 

Protection Quarterly 19(2), 59-66. 

McLaren, D.A., Weiss, J.E.R. and Faithfull, I. (2004) Weeds of significance to the grazing industries of Victoria. In T. Grice, Weeds of 

Significance to the Grazing Industries of Australia. Final Report to Meat and Livestock Australia, Project Number COMP.045, CRC for 

Australian Weed Management and Meat and Livestock Australia Ltd, North Sydney, pp. 48-69. 

McLaren, D.A., Whattam, M., Blood, K., Stajsic, V. and Hore, R. (1999) Mexican feather grass (Nassella tenuissima) a potential disaster for 

Australia. In A.C. Bishop, M. Boersma and C.D. Barnes (Eds.), 12th Australian Weeds Conference Papers and Proceedings, Tasmanian Weed 

Society, Devenport, pp. 658-662. 

McLeod, R., Reay, F. and Smyth, J. (1994) Plant Nematodes of Australia Listed by Plant and by Genus. New South Wales Agriculture and Rural 

Industries Research and Development Corporation. 

McQuillan, P.B. (1999) The effect of changes in Tasmanian grasslands on the geometrid moth trip Xanthorhoini (Geometridae: Larentiinae). In: 

W. Ponder and D. Lunney (Eds.), The Other 99%. The Conservation and Biodiversity of Invertebrates. Transactions of the Royal Zoological 

Society of New South Wales, Mosman, pp. 121-128. 

McQuillan, P.B. and Webb, W.R. (1994) Selective soil organic matter consumption by larvae of Adoryphorus couloni (Burmeister) (Coleoptera: 

Scarabaeidae). Journal of the Australian Entomological Society 33, 49-50. 

McTaggart, A. (1936) A survey of the pastures of Australia. Embodying ecological information and discussions explanatory of the accompanying 

pasture map of the Commonwealth. Council for Scientific and Industrial Research Bulletin No. 99. 

McWhirter, L., Lowien, J., Grech, C., McLaren, D.A. and Sindel, B.M. (2006) Regional best practice management of Chilean needle grass 

(Nassella neesiana) – New South Wales preliminary trial results. In B. Hackney, K. Bailes, J. Piltz and H. Burns (Eds.), Proceedings of the 21st 

Annual Conference of the Grassland Society of NSW Inc., Wagga Wagga, July, pp. 139-140. 

Meagher, D. and Fuhrer, B. (2003) A Field Guide to the Mosses and Allied Plants of Southern Australia. Australian Biological Resources Study 

and Field Naturalists Club of Victoria, Canberra and Blackburn. 

195

Meglitsch, P.A. (1972) Invertebrate Zoology. 2nd ed., Oxford University Press, London. 

Melbourne, B.A.(1993) The assessment of insect biodiversity. A case study: ants and carabid beetles in grasslands of the ACT. BSc (Hons.) 

thesis, Division of Botany and Zoology, Australian National University, Canberra. 

Melbourne, B.A., Gullan, P.J. and Su, Y.N. (1997) Interpreting data from pitfall-trap surveys: crickets and slugs in exotic and native grasslands of 

the Australian Capital Territory. Memoirs of the Museum of Victoria 56(2), 361-367. 

Melbourne, B.A., Cornell, H.V., Davies, K.F., Dugaw, C.J., Elmendorf, S., Freestone, A.L., Hall, R.J., Harrison, S., Hastings, A., Holland, M., 

Hoyloak, M., Lambrinos, J., Moore, K. and Yokomizo, H. (2007) Invasion in a heterogeneous world: resistance, coexistence or hostile takeover? 

Ecology Letters 10, 77-94. 

Memmott, J., Fowler, S.V., Paynter, Q., Sheppard, A.W. and Syrett, P. (2000) The invertebrate fauna on broom, Cytisus scoparius, in two native 

and two exotic habitats. Acta Oecologica 21(3), 213-222. 

Menkhorst, P.W. (Ed.) (1995a) Mammals of Victoria. Distribution, Ecology and Conservation. Oxford University Press, Melbourne. 

Menkhorst, P.W. (1995b) The Victorian mammal fauna. In P.W. Menkhorst (Ed.), Mammals of Victoria. Distribution, Ecology and Conservation. 

Oxford University Press, Melbourne, pp. 5-15. 

Menkhorst, P.W. (1995c) Short-beaked Echidna. In P.W. Menkhorst (Ed.), Mammals of Victoria. Distribution, Ecology and Conservation. 

Oxford University Press, Melbourne, pp. 38-39. 

Menkhorst, P.W. (1995d) European rabbit. In P.W. Menkhorst (Ed.), Mammals of Victoria. Distribution, Ecology and Conservation. Oxford 


Mgobozi, M.P., Somers, M.J. and Dippenaar-Schoeman, A.S. (2008) Spider responses to alien plant invasion: the effect of short- and long-term 

Chromolaena odorata invasion and management. Journal of Applied Ecology 45, 1189-1197. 

Michalk, D., Kemp, D.R., Campbell, M.H. and McLaren, D.A. (1999) Control of serrated tussock – problems in developing IWM systems. In 

A.C. Bishop, M. Boersma and C.D. Barnes (Eds.), 12th Australian Weeds Conference Papers and Proceedings, Tasmanian Weed Society, 

Devenport, pp. 20-24. 

Michelmore, M. (2003) The Serrated Tussock Managers’ Factpack. NSW Agriculture, iv + 74 pp. 

Miller, L.J. and New, T.R. (1997) Mount Piper grasslands: pitfall trapping of ants and interpretation of habitat variability. Memoirs of the 

Museum of Victoria 56(2), 377-381. 

Milton, S.J. (2004) Grasses as invasive alien plants in South Africa. South African Journal of Science 100, 69-75. 

Minkey, D. and Spafford Jacob, H. (2004) The significance and variability of predation of weed seeds in an agricultural landscape in Western 

Australia. In J. M. Cullen, D.T. Briese, D.J. Kriticos, W.M. Lonsdale, L. Morin and J.K. Scott (Eds.), Proceedings of the XI International 

Symposium on Biological Control of Weeds, CSIRO, Canberra, p. 64. [Abstract] 

Mitchell, T.L.(1848) Journal of an expedition into the interior of tropical Australia, in search of a route from Sydney to the Gulf of Carpentaria. 

Longman, Brown, Green and Longmans, London. 

Moerkerk, M. (2005a) Machinery hygiene - what is on our vehicles? In Proceedings Second Victorian Weed Conference, 17-19 August 2005, 

Bendigo, Weed Society of Victoria, Frankston, pp. 99-101. 

Moerkerk, M. (2005b) Machinery hygiene - what is on our vehicles? Proceedings of the Tussock Terminators Research Forum at Lake Hume 

Resort Albury NSW 2nd-3rd November 2005. Cooperative Research Centre for Australian Weed Management, National Serrated Tussock 

Management Group, National Chilean Needle Grass Taskforce, Natural Heritage Trust, Department of Primary Industries Victoria, New South 

Wales Department of Primary Industries, pp. 50-52. 

Moerkerk, M. (2006a) Risk of weed movement through vehicles, plant and equipment: results from a Victorian study. In C. Preston, J.H. Watts 

and N.D. Crossman (Eds.), 15th Australian Weeds Conference Papers and Proceedings. Weed Management Society of South Australia Inc., 

Adelaide, pp. 458-461. 

Moerkerk, M. (2006b) Are you imorting weeds onto your property? In B. Hackney, K. Bailes, J. Pilz and H. Burns (Eds.), Proceedings of the 21st 

Annual Conference of the Grassland Society of New South Wales Inc., pp. 55-57. 

Monson, R.K. (1989) On the evolutionary pathways resulting in C4 photosynthesis and crassulacean acid metabolism (CAM). Advances in 

Ecological Research 19, 57-110. 

Moon, C.D., Guillaumin, J-J., Ravel, C., Craven, K.D. and Schardl, C.L.(2007) New Neotyphodium endophyte species from the grass tribes 

Stipeae and Meliceae. Mycologia 99(6), 895-905. 

Mooney, H.A. and Drake, J.A. (1989) Biological invasions: a SCOPE program overview. In J.A.Drake, H.A. Mooney, F. di Castri, R.H. Groves, 

F.J. Kruger, M. Rejmanek and M. Williamson (Eds.), Biological Invasions: a Global Perspective, Scientific Committee on Problems of the 

Environment (SCOPE) 37, John Wiley and Sons, Chichester, UK, pp. 491-508. 

Mooney, H.A., Zavaleta, E.S. and Hobbs, R.J. (2006) Invasive alien species – are we up to the challenge? In C. Preston, J.H. Watts and N.D. 

Crossman (Eds.), 15th Australian Weeds Conference Papers and Proceedings. Weed Management Society of South Australia Inc., Adelaide, pp. 

1-5. 

Moore, D.M. (1982) Flora Europaea Check-list and Chromosome Index. Cambridge University Press, Cambridge. 

Moore, R.M. (1973) South-eastern temperate woodlands and grasslands. In R.M. Moore (Ed.) Australian Grasslands, Corrected edition, 

Australian National University Press, Canberra, pp. 169-190. 

Moore, R.M. (1993) Grasslands of Australia. In R.T. Coupland (Ed.), Ecosystems of the World 8B. Natural Grasslands Eastern Hemisphere and 

Résumé, Elsevier Science Publishers B.V., Amsterdam, Netherlands, pp. 315-360. 

Moore, R.M., Condon, R.W.and Leigh, J.H. (1973) Semi-arid woodlands. In R.M. Moore (Ed.) Australian Grasslands, Corrected edition, 

Australian National University Press, Canberra, pp. 228-245. 

Moraldo, B. (1986) Il genere Stipa L. (Gramineae) in Italia. Webbia 40(2), 203-278. 

Morcom, L. (2004) Curly Sedge, Carex tasmanica. Flora and Fauna Guarantee Action Statement No. 88, Department of Natural Resources and 


Moretto, A.S. and Distel, R.A. (1998) Requirement of vegetation gaps for seedling establishment of two unapalatable grasses in native grassland 

of central Argentina. Australian Journal of Ecology 23, 419-423. 

Moretto, A.S. and Distel, R.A. (2002) Soil nitrogen availability under grasses of different palatability in a temperate semi-arid rangeland of 

central Argentina. Austral Ecology 27, 509-514. 

196

Morfe, T.A., McLaren, D.A. and Weiss, J. (2003) Treatment of Chilean needle-grass (Nassella neesiana) in Victoria: a benefit-cost analysis. 

Proceedings II. Nineteenth Asian-Pacific Weed Science Society Conference, 17-21 March, Manilla, Phillipines, Weed Science Society of the 

Philippines, pp.875-881. 

Morgan, J.W. (1994) The ecology of grasses and grasslands in lowland Victoria. Victorian Naturalist 111(3), 87-92. 

Morgan, J.W. (1995a) Ecological studies of the endangered Rutidosis leptorrhynchoides. I. Seed production, soil seed bank dynamics, population 

density and their effects on recruitment. Australian Journal of Botany 43, 1-11. 

Morgan, J.W. (1995b) Ecological studies of the endangered Rutidosis leptorrhynchoides. II. Seedling emergence and survival in a native 

grassland. Australian Journal of Botany 43, 13-24. 

Morgan, J.W. (1996) Secondary juvenile period and community recovery following late-spring burning of a kangaroo grass Themeda triandra 

grassland. Victorian Naturalist 113(2), 47-57. 

Morgan, J. (1997) Monitoring small-scale spatial and temporal dynamics of plants in basalt plains grasslands. Paper presented to the Research 

Advisory Group of the Victorian Grassy Ecosystem Reference Group Seminar, 11 October. 

Morgan, J.W. (1998b) Importance of canopy gaps for recruitment of some forbs in Themeda triandra-dominated grasslands in south-eastern 

Australia. Australian Journal of Botany 46, 609-627. 

Morgan, J.W. (1998c) Composition and seasonal flux of the soil seed bank of species-rich Themeda triandra grasslands in relation to burning 

history. Journal of Vegetation Science 9, 145-156. 

Morgan, J.W. (1998d) Patterns of invasion of an urban remnant of a species-rich grassland in southeastern Australia by non-native plant species. 

Journal of Vegetation Science 9, 181-190. 

Morgan, J.W. (1998e) Small-scale plant dynamics in temperate Themeda triandra grasslands of southeastern Australia. Journal of Vegetation 

Science 9, 347-360. 

Morgan, J.W. (2001) Seedling recruitment patterns over 4 years in an Australian perennial grassland community with different fire histories. 

Journal of Ecology 89, 908-919. 

Morgan, J.W. (2004) Bryophyte composition in a native grassland community subjected to different long-term fire regimes. Cunninghamia 8(4), 

485-489. 

Morgan, J.W. and Lunt, I.D. (1999) Effects of time-since-fire on the tussock dynamics of a dominant grass (Themeda triandra) in a temperate 

Australian grassland. Biological Conservation 88, 379-386. 

Morgan, J.W. and Rollason, T.S. (1995) Base-line monitoring of a significant grassland remnant at Evans Street, Sunbury, Victoria. Victorian 

Naturalist 112(4), 148-159. 

Morton, S.R. (1985) Granivory in arid regions: comparison of Australia with North and South America. Ecology 66(6), 1859-1866. 

Mott, J.J. and Groves, R. H. (1994) Natural and derived grasslands. In R.H. Groves (Ed.) Australian Grasslands. 2nd Ed., Cambridge University 

Press, Cambridge, UK, Ch.13, pp. 369-392. 

Moulds, M.S. (1990) Australian Cicadas. New South Wales University Press, Kensington. 

Mound, L.A. and Palmer, J.M. (1972) Grass-flower infesting thrips of the genus Chirothrips Haliday in Australia. Journal of the Australian 

Entomological Society 11, 332-339. 

Mound, L.A. and Heming, B.S. (1991) Thysanoptera. In CSIRO, The Insects of Australia. A Textbook for Students and Research Workers. 2nd 

Ed., Melbourne University Press, Carlton, pp. 510-515. 

Mueller Dombois, D. (1973) A non-adapted vegetation interferes with water removal in a tropical rain forest area in Hawaii. Tropical Ecology 24, 

1-18. 

Mueller-Dombois, D. and Ellenberg, H. (1974) Aims and Methods of Vegetation Ecology. John Wiley and Sons, New York. 

Muir, A. (1994) Western (Basalt) Plains Grassland Community. Flora and Fauna Guarantee Action Statement No. 53, Department of 

Conservation and Natural Resources, Victoria, 12 pp. 

Muir, A. (2003) Small Psoralea Psoralea parva. Flora and Fanua Guarantee Action Statement No. 31, Department of Sustainability and 


Mulham, W.E. and Moore, C.W.E. (1970) The grass seed problem. The Pastoral Review 80, 105 and 107, 25 February. 

Mulvaney, D.J. (1964) Prehistory of the basalt plains. Proceedings of the Royal Society of Victoria (New Series) 77(2), 427-432. 

Murbach, L. (1900) Note on the mechanics of the seed-burying awns of Stipa avenacea. Botanical Gazette 30, 113-117. 

Murphy, B.P. and Bowman, D.M.J.S. (2007) The interdependence of fire, grass, kangaroos and Australian aborigines: a case study from central 

Arnhem Land, northern Australia. Journal of Biogeography 34, 237-250. 

Muyt, A. (2001) Bush Invaders of South-east Australia. A Guide to the Identification and Control of Environmental Weeds found in South-east 

Australia. R.G. and F.J. Richardson, Meredith, Victoria. 

Muyt, A. (2005) Yarramundi Reach: Vegetation Survey Methods, Management Issues and Flora List. Report to the National Capital Authority, 

March, 20 pp. 

Nair, P. (2003) Farmers’ weed plea. The Regional [newspaper, Sunbury area, Victoria], 30 November. 

Nannfeldt, J.A. (1949) In ‘News and Notes’, Herbage Abstracts 19(1), 72. 

Nathan, R., Sapir, N., Trakhtenbrot, A., Katul, G.G., Bohrer, G., Otte, M., Avissar, R., Soons, M.B., Horn, H.S., Wikelski, M. and Levin, S.A. 

(2005) Long-distance biological transport processes through the air: can nature’s complexity be unfolded in silico? Diversity and Distributions 

11, 131-137. 

National Herbarium of New South Wales (2009) New South Wales Flora Online. Nassella neesiana. Royal Botanic Gardens and Domain Trust, 

Sydney. http://plantnet.rbgsyd.nsw.gov.au/cgi-bin/NSWfl.pl?page=nswflandlvl=spandname=Nassella~neesiana [Accessed 10 June 2009]. 

Naumann, I.D. (1991) Hymenoptera. In CSIRO, The Insects of Australia. A Textbook for Students and Research Workers. 2nd Ed., Melbourne 

University Press, Carlton, pp. 916-1000. 

Nelis, L.C. (2006) Plantas nativas y endemicas de la Ilsa Robinson Crusoe. http://fm2.fieldmuseum.org.plantguides/guide_pdfs/222-01.pdf 

[accessed 14 December 2006] 

New, T.R. (1984) Insect Conservation – An Australian Perspective. Dr W. Junk Publishers, Dordrecht, Netherlands. 

New, T.R. (1994) Exotic Insects in Australia. Gleneagles Publishing, Glen Osmond, South Australia. 

197

New, T.R. (2000) How useful are ant assemblages for monitoring habitat disturbances on grasslands in southeastern Australia. Journal of Insect 


Newsome, A.E. and Noble, I.R. (1986) Ecological and physiological characters of invading species. In R.H. Groves and J.J. Burdon (Eds.), 

Ecology of Biological Invasions: An Australian Perspective. Australian Academy of Science, Canberra, pp. 1-20. 

Nie, Z.N., Zollinger, R.P. and Jacobs, J.L. (2009) Performance of 7 Australian native grasses from the temperate zone under a range of cutting 

and fertiliser regimes. Crop and Pasture Science 60, 943-953. 

Nielsen, E.S. and Common, I.F.B. (1991) Lepidoptera. In CSIRO, The Insects of Australia. A Textbook for Students and Research Workers. 2nd 

Ed., Melbourne University Press, Carlton, pp. 817-915. 

Noble, J.C. (1993) Relict surface-soil features in semi-arid mulga (Acacia aneura) woodlands. Rangeland Journal 15, 48-70. 

Noble, J.C., Müller, W.J., Detling, J.K. and Pfitzner, G.H. (2007) Landscape ecology of the burrowing bettong: warren distribution and patch 

dynamics in semiarid eastern Australia. Austral Ecology 32, 326-337. 

Noriega, J.I., Carlini, A.A. and Tonni, E.P. (2004) Vertebrados del Pleistoceno tardío de la cuenca del Arroyo Ensenada (Departamento 

Diamante, Provincia de Entre Ríos). In F.G. Aceñolaza (Ed.), Temas de la Biodiversidad del Litoral Fluvial Argentino, Instituto Superior De 

Correlación Geologica ( INSUGEO ) Miscelánea 12, 77-82. 

Norton, M.R., Leliévre, F., Fukai, S. and Volaire, F. (2008) Measurement of summer dormancy in temperate perennial pasture grasses. Australian 

Journal of Agricultural Research 59, 498-509. 

NSW Scientific Committee (2003) Invasion of native plant communities by exotic perennial grasses –Key Threatening Process declaration. NSW 

Scientific Committee, Hurstville. 

Obst, C. and How, T. (2004) Field survey results for Chilean needlegrass (Nassella neesiana (Trin. and Rupr.) Barkworth) and Texas needlegrass 

(Nassella leucotricha (Trin. and Rupr.) R.W. Pohl) in the Mt. Lofty Ranges, Fleurieu Peninsula and greater Adelaide regions of South Australia. 

Plant Protection Quarterly 19(2), 67-68. 

Oomes, M.J.M. (1990) Changes in dry matter and nutrient yields during the restoration of species-rich grasslands. Journal of Vegetation Science 

1, 333-338. 

O’Donoghue, J. (2008) Petal power. New Scientist 200(2680), 36-39. 

O’Dwyer, C. (1999) The habitat of the Golden Sun Moth Synemon plana (Lepidoptera; Castniidae). In: W. Ponder and D. Lunney (Eds.), The 

Other 99%. The Conservation and Biodiversity of Invertebrates. Transactions of the Royal Zoological Society of New South Wales, Mosman, pp. 

322-324. 

O’Dwyer, C. (2004) Golden Sun Moth Synemon plana. Flora and Fauna Guarantee Action Statement No. 106, Department of Sustainability and 


Oram, R.N. (1990) Register of Australian Herbage Plant Cultivars. 3rd Ed., CSIRO, East Melbourne. 

O’Shea, M. (1997) Alternative detection techniques for Delma impar. Paper presented to the Research Advisory Group of the Victorian Grassy 

Ecosystem Group Seminar 11 October. 

O’Shea, M.B. (2005) Methods for assessment and techniques for management of Striped Legless Lizard Delma impar populations in southeastern 

Australia. PhD thesis, Sustainability Group, Victoria University, St Albans, Victoria. 

Overbeck, G.E. and., Műller, S.C., Pillar, V.D. and Pfadenhauer, J. (2006) Floristic composition, environmental variation and species distribution 

patterns in burned grassland in southern Brazil. Brazilian Journal of Biology 66(4), 1073-1090. 

Overbeck, G.E. and., Műller, S.C., Fidelis, A., Pfadenhauer, J., Pillar, V.D., Blanco, C.C., Boldrini, I.I., Both, R. and Forneck, E.D. (2007) 

Brazil’s neglected biome: the South Brazilian Campos. Perspectives in Plant Ecology, Evolution and Systematics 9, 101-116. 

Overbeck, G.E. and Pfadenhauer, J. (2007) Adaptive strategies in burned subtropical grassland in southern Brazil. Flora 202, 27-49. 

Panetta, F.D. and Lane, D.W. (1996) Managing weeds in Australia’s public lands and forests. In R.C.H. Shepherd (Ed.), Proceedings Eleventh 

Australian Weeds Conference, Melbourne 30 September - 3 October 1996, Weed Science Society of Victoria, Frankston, pp. 391-397. 

Panetta, F.D., Ridsdill-Smith, T.J., Barbetti, M.J. and Jones, R.A.C. (1993) The ecology of weeds, invertebrate pests and diseases of Australian 

sheep pastures. In E.S. Delfosse (Ed.), Pests of Pastures. Weed, Invertebrate and Disease Pests of Australian Sheep Pastures. CSIRO Information 

Services, Melbourne, pp. 87-114. 

Parker, J.D. and Hay, M.E. (2005) Biotic resistance to plant invasions? Native herbivores prefer non-native plants. Ecology Letters 8, 959-967. 

Parker, J.D., Burkepile, D.E. and Hay, M.E. (2006a) Opposing effects of native and exotic herbivores on plant invasions. Science 311, 1459- 

1461. 

Parker, J.D., Burkepile, D.E. and Hay, M.E. (2006b) Response to comment on “Opposing effects of native and exotic herbivores on plant 

invasions”. Science 313, 298b. 

Parr, J., Sullivan, L. and Quirk, R. (2009). Sugarcane phytoliths: encapsulation and sequestration of a long-lived carbon fraction. Sugar Tech 

11(1), 17-21. 

Parsons, W.T. (1973) Noxious Weeds of Victoria. Inkata Press, Melbourne. 

Parsons, W.T. and Cuthbertson, E.G. (1992) Noxious Weeds of Australia. Inkata Press, Melbourne. 

Patton, R.T. (1930) The factors controlling the distribution of trees in Victoria. Proceedings of the Royal Society of Victoria 42 (NS)(2), 154-210. 

Patton, R.T. (1935) Ecological studies in Victoria. Part IV. – Basalt Plains Association. Proceedings of the Royal Society of Victoria 48 (NS)(2), 

172-191. 

Peart, M.H. (1979) Experiments on the biological significance of the morphology of seed-dispersal units in grasses. Journal of Ecology 67, 843- 

863. 

Peart, M.H. (1984) The effects of morphology, orientation and position of grass diaspores on seedling survival. Journal of Ecology 72, 437-453. 

Pedigree Online (2007) Pedigree Online. All Breeds Database. http://www.allbreedpedigree.com/raider3 [Accessed 18 July 2007] 

Pedrano, J., Isacch, J.P.andBó, M.S.(2008) Habitat relationships of diurnal raptors at local and landscape scales in southern temperate grasslands 

of Argentina. Emu 108, 301-310. 

Peñailillo, B. (1996) Anatherostipa, un nuevo genero de Poaceae (Stipeae). Gayana Botánica 53, 277-284. 

Penman, D.R. (1978) Biology of Euciodes suturalis (Coleoptera: Anthribidae) infesting cocksfoot in Canterbury. New Zealand Entomologist 

6(4), 421-425. 

198

Perelman, S.B., León, R.J.C. and Oesterheld, M. (2001) Cross-scale vegetation patterns of Flooding Pampa grasslands. Journal of Ecology 89, 

562-577. 

Peter, J.M. (2000) Birds and boxthorn. Victorian Naturalist 117(2), 63-66. 

Peterson, R.A. (1962) Factors affecting resistance to heavy grazing in needle-and-thread grass. Journal of Range Management 15, 183-189. 

Petit, R.J. (2004) Biological invasions at the gene level. Diversity and Distributions 10, 159-165. 

Petroski, R.J., Powell, R.G. and Clay, K. (1992) Alkaloids of Stipa robusta (sleepygrass) infected with an Acremonium endophyte. Natural 

Toxins 1(2), 84-88. 

Phillips, A. and Hocking, C. (1996) Results of trials for replacing serrated tussock with weed-free kangaroo grass in degraded native Western 

Plains Grassland. In R.C.H. Shepherd (Ed.), Proceedings Eleventh Australian Weeds Conference, Melbourne 30 September - 3 October 1996, 

Weed Science Society of Victoria, Frankston, p. 524. 

Piperno, D.R. and Sues, H.-D. (2005). Dinosaurs dined on grass. Science 310, 1126-1128. 

Pitkin, B.R. (1972) A revision of the Australian genus Odontothripiella Bagnall, with descriptions of fourteen new species (Thysanoptera: 

Thripidae). Journal of the Australian Entomological Society 11, 265-289. 

Pitkin, B.R. (1973) A revision of the Australian Haplothripini, with descriptions of three new species (Thysanoptera: Phaleothripidae). Journal of 

the Australian Entomological Society 12, 315-339. 

Platt, S., Adair, R., White, M. and Sinclair, S. (2005) Regional priority-setting for weed management on public land in Victoria. In Proceedings 

Second Victorian Weed Conference, 17-19 August 2005, Bendigo, Weed Society of Victoria, Frankston, pp. 89-98. 

Pohl, R.W. (1986) Man and the grasses: a history. In T.S. Soderstrom, K.W. Hilu, C.S. Campbell and M.E. Barkworth (Eds.), Grass Systematics 

and Evolution, Smithsonian Institution Press, Washington, USA, pp. 355-358. 

Pont, A.C. (1989) Family Muscidae. In N.L. Evenhuis (Ed.), Catalog of the Diptera of the Australasian and Oceanian Regions. Bishop Museum 

Press and E.J. Brill, Honolulu and Leiden, pp. 675-699. 

Powell, J.M. and Duncan, J.S. (1982) History. In J.S. Duncan (Ed.) Atlas of Victoria. Victorian Government Printing Office, Melbourne, pp. 74- 

81. 

Prasad, V., Strömberg, C.A.E., Alimohammadian, H. and Sahni, A. (2005) Dinosaur coprolites and the early evolution of grasses and grazers. 

Science 310, 1177-1180. 

Preston, C. (2000) Herbicide mode of action and herbicide resistance. In B.M. Sindel (Ed.), Australian Weed Management Systems, R.G. and F.J. 

Richardson, Melbourne, pp. 209-225. 

Pretty, G.L., Paton, R.C. and Weathersbee, R.D.J. (1983) Tribal man. In M.J.Tyler, C.R.Twidale, J.K.Ling and J.W. Holmes (Eds.), Natural 

History of the South East. Royal Society of South Australia Inc., pp. 115-126. 

Prieur-Richard, A.-H. and Lavorel, S. (2000) Invasions: the perspective of diverse plant communities. Austral Ecology 25, 1-7. 

Pritchard, A. and Ingeme, Y. (2003) Gorae Leek-orchid Prasophyllum diversiflorum. Flora and Fauna Guarantee Action Statement No. 148, 

Department of Sustainabilityand Environment, East Melbourne, 4 pp. 

Pritchard, G.H. (2002) Evaluation of herbicides for the control of mature Chilean needle grass. Unpublished report, Keith Turnbull Research 

Institute, Department of Primary Industries, Frankston, Victoria, 4 pp. 

Puckey, H. and Albrecht, D. (2004) Buffel grass (Cenchrus ciliaris L.): presenting the arid Northern Territory experience to our south Australian 

neighbours. Plant Protection Quarterly 19(2), 69-72. 

Puhar, P. (1996) The effects of herbicides on in vitro seed germination of native Kangaroo Grass (Themeda triandra) and introduced weed 

species Serrated Tussock (Nassella trichotoma) and Chilean Needle Grass (Stipa neesiana). B.App.Sci. (Hons.) thesis, Department of 

Environmental Management, Victoria University of Technology, St Albans, Victoria. 

Puhar, P. and Hocking, C. (1996) The effects of herbicides on in vitro seed germination of kangaroo grass (Themeda triandra) and introduced 

weed species, serrated tussock (Nassella trichotoma) and Chilean needle grass (Stipa neesiana). In R.C.H. Shepherd (Ed.), Proceedings Eleventh 

Australian Weeds Conference, Melbourne, 30 September - 3 October 1996, Weed Science Society of Victoria, Frankston, p. 533. 

Pulliam, H.R. and Brand, M.R. (1975) The production and utilization of seeds in plains grassland of southeastern Arizona. Ecology 56, 1158-116. 

Quattrocchi, U. (2006) CRC World Dictionary of Grasses: Common Names, Scientific Names, Eponyms, Synonyms and Etymology. CRC Press, 

Taylor and Francis, Boca Raton, London, New York. 

Ramasamy, S. (2008) Investigations into group J herbicide resistance in Nassella trichotoma and Sporobolus fertilis and biological control of S. 

fertilis using the pathogen Nigrospora oryzae. PhD thesis, School of Applied Science, RMIT University. 

Ramirez, J.R. (1951) Notas sobre algunas Gramíneas y Leguminosas nuevas o interesantes de la flora Paraguaya I. Darwiniana 9 (3-4), 618-621. 

Randall, R.P. (2002) A Global Compendium of Weeds. R.G. and F.J. Richardson, Melbourne. 

Read, J.L. (1994) The diet of three species of firetail finches in temperate South Australia. Emu 94, 1-8. 

Ratcliffe, F.N. (1934) Soil Drift in the Arid Pastoral Areas of South Australia. Council for Scientific and Industrial Research Pamphlet No. 64, 

Melbourne, 96 pp. 

Recher, H.F. and Lim, L. (1990) A review of current ideas of the extinction, conservation and management of Australia’s terrestrial vertebrate 

fauna. Proceedings of the Ecological Society of Australia 16, 287-301. 

Reddy, G.V.P. and Guerrero, A. (2004) Interactions of insect pheromones and plant semiochemicals. Trends in Plant Science 9(5), 253-261. 

Rejmánek, M. (1989) Invasibility of plant communities. In: J.A.Drake, H.A. Mooney, F. di Castri, R.H. Groves, F.J. Kruger, M. Rejmanek and M. 

Williamson (Eds.), Biological Invasions: a Global Perspective, Scientific Committee on Problems of the Environment (SCOPE) 37, John Wiley 

and Sons, Chichester, UK, pp. 3-13. 

Rejmánek, M. (1995) What makes a species invasive? In: P. Pyšek, K. Prach, M. Rejmánek and M.Wade (Eds.), Plant Invasions – General 

Aspects and Special Problems, SPB Academic Publishing, Amsterdam, Netherlands, pp. 3-13. 

Rejmánek, M. (1996) A theory of seed plant invasiveness: the first sketch. Biological Conservation 78, 171-181. 

Rejmánek, M. (2000) Invasive plants: approaches and predictions. Austral Ecology 25, 497-506. 

Rejmánek, M. and Richardson, D.M. (1996) What attributes make some plant species more invasive? Ecology 77(6), 1655-1661. 

Ren, A.Z., Gao, Y.B., Wang, Y.H., Zhang, X. and Wei, Y.K. (2008) Diversity of clavicipitaceous fungal endophytes in Achnatherum sibiricum in 

natural grasslands of China. Symbiosis 46(1), 21-31. 

199

Rentz, D.C.F. (1996) Grasshopper Country. The Abundant Orthopteroid Insects of Australia. University of New South Wales Press, Sydney. 

Rentz, D.C.F., Lewis, R.C., Su, Y.N. and Upton, M.S. (2003) A Guide to Australian Grasshoppers and Locusts. Kota Kinabalu, Malaysia, 

Natural History Publications (Borneo). 

Reyna, J.V. and Barkworth, M.E. (1994) El genero Nassella (Poaceae: Stipeae) en Mexico. Acta Botánica Mexicana 26, 63-75. 

Reynolds, D. M. (2006) The ecological importance of soil disturbance on the recruitment, growth and reproduction of temperate lowland 

grassland forbs and implications for restoration. B Sc (Honours) Thesis. Sustainability Group, Faculty of Science, Health and Technology, 

Victoria University, St Albans. 

Reynolds, O.L., Keeping, M.G. and Meyer, J.H. (2009) Silicon-augmented resistance of plants to herbivorous insects: a review. Annals of Applied 

Biology 155, 171-186. 

Ricciardi, A. and Ward, J.M. (2006) Comment on “Opposing effects of native and exotic herbivores on plant invasions”. Science 313, 298a. 

Richards, P.J. (2009) Aphaenogaster ants as bioturbators: impacts on soil and slope processes. Earth-Science Reviews 96, 92-106. 

Richardson, D.M. and van Wilgen, B.W. (2004) Invasive alien plants in South Africa: how well do we understand the ecological impacts? South 

African Journal of Science 100, 45-52. 

Ridenour, W.M. and Callaway, R.M. (2001) The relative importance of allelopathy in interference: the effects of an invasive weed on a native 

bunchgrass. Oecologia 126(3), 444-450. 

Ridsdill Smith, T.J. (1975) Selection of living grass roots in the soil by larvae of Sericesthis nigrolineata (Coleoptera: Scarabaiedae). Entomolgia 

Experimentalis et Applicata 18, 75-86. 

Roberts, R.J., Campbell, A.J., Porter, M.R. and Sawtell, N.L. (1982) The distribution and abundance of pasture scarabs in relation to Eucalyptus 

trees. In K.E. Lee (Ed.), Proceedings of the 3rd Australasian Conference on Grassland Invertebrate Ecology, Adelaide 30 November – 4 

December, pp. 207-213. 

Roberts, S.E., Andersson, K., Murphy, N., McMaster, C. and Tate, S. (2006) Distribution of phosphorus levels across the Healthy Soils for 

Healthy Landscape and Benchmarking Soil Chemistry projects areas for cropping and pasture systems. In B. Hackney, K. Bailes, J. Pilz and H. 

Burns (Eds.), Proceedings of the 21st Annual Conference of the Grassland Society of New South Wales Inc., pp. 148-149. 

Robertson, D. (1985) Interrelationships between kangaroos, fire and vegetation dynamics at Gellibrand Hill Park, Victoria. PhD thesis, 

Universithy of Melbourne. 

Robinson, D. (1991) Threatened birds in Victoria: their distributions, ecology and future. Victorian Naturalist 108(3), 67-77. 

Robinson, G.S. and Nielsen, E.S. (1993) Tineid Genera of Australia (Lepidoptera). Monographs on Australian Lepidoptera Volume 2. CSIRO 

Publishing, Melbourne. 

Robinson, R. (2003) Natural recruitment of native forbs in the grassy ecosytems of south-eastern Australia. MSc thesis, Victoria University of 

Technology, St. Albans. 

Robinson, R. (2005) Direct seeding in weed infested remnant grassland situations using native forbs. Plant Protection Quarterly 20(3), 84-89. 

Rogers, A.L., Anderson, G.W., Biddiscombe, E.F., Arkell, P., Glencross, R., Nicholas, D.A. and Paterson, J.G. (1979) Perennial Pasture Grasses 

in South Western Australia. 1. Preliminary Evaluation of Species. Department of Agriculture Western Australia Technical Bulletin No. 45, 15 pp. 

Rolls, E. (1984) They All Ran Wild. The Animals and Plants that Plague Australia. Revised Ed., Angus and Robertson, Sydney. 

Rosengren, N. (1999) Geology and geomorphology of Victoria’s grassland regions. In R.N. Jones (Ed.), The Great Plains Crash. Proceedings of 

a Conference on the Grasslands and Grassy Woodlands of Victoria. Victorian Institute of Technology, October 2nd and 3rd, 1992, Indigenous 

Flora and Fauna Association and Victorian National Parks Association, pp. 7-10. 

Rosengurtt, B., Arrillaga de Maffei, B.R. and Izaguirre de Artucio, P. (1970) Gramineas Uruguayas. Universidad de la República, Departamento 

de Pulbicaciones, Montevideo. 

Ross, J.H. and Walsh, N.G. (2003) A Census of the Vascular Plants of Victoria. 7th Ed., Royal Botanic Gardens, South Yarra, Victoria.. 

Rossiter, N.A., Setterfield, S.A., Douglas, M.M. and Hutley, L.B. (2003) Testing the grass-fire cycle: alien grasses invasion in the tropical 

savannas of northern Australia. Diversity and Distributions 9,169-176. 

Rossiter, N., Setterfield, S., Douglas, M., Hutley, L. and Cook, G. (2006) The impact of exotic grass invasions on nitrogen cycling: a mini review. 

In C. Preston, J.H. Watts and N.D. Crossman (Eds.), 15th Australian Weeds Conference Papers and Proceedings. Weed Management Society of 

South Australia Inc., Adelaide, pp. 815-818. 

Rout, M.E. and Chrzanowski, T.H. (2009) The invasive Sorghum halepense harbors endophtyic N 2 -fixing bacteria and alters soil 

biogeochemistry. Plant and Soil 315, 163-172. 

Rowell, A. and Crawford, I. (1995) Survey of the morabine grasshopper Keyacris scurra (Rehn) in the A.C.T. CSIRO Report to the Wildlife 

Research Unit, ACT Parks and Conservation Service, June, 63 pp. 

Rudgers, J.A., Afkhami, M.E., Rua, M.A., Davitt, A.J., Hammer, S. and Huguet, V.M. (2009) A fungus among us: broad patterns of endophyte 

distribution in the grasses. Ecology 90(6), 1531-1539. 

Ruprecht, E., Donath, T.W., Otte, A. and Lutz Eckstein, R. (2008) Chemical effects of a dominant grass on seed germination of four familial pairs 

of dry grassland species. Seed Science Research 18, 239-248. 

Saba, S.L. and Toyos, A. (2003) Seed removal by birds, rodents and ants in the Austral portion of the Monte Desert, Argentina. Journal of Arid 

Environments 53, 115-124. 

Sabrosky, C.W. (1989) Family Chloropidae. In N.L. Evenhuis (Ed.), Catalog of the Diptera of the Australasian and Oceanian Regions. Bishop 


Salisbury, F.B. and Ross, C.W. (1992) Plant Physiology. 4th Ed., Wadsworth Publishing Company, Belmont, California. 

Samways, M.J. (2005) Insect Diversity Conservation. Cambridge University Press, Cambridge, UK. 

Sands, D. and Scott, S. (1997) Richmond birdwing. Nature Australia 25(7), 24-29. 

Sanz-Elorza, M., Dana E.D. and Sobrino, E. (2005) Aproximación al listado de plantas vasculares alóctonas invasoras reales y potenciales en las 

islas Canarias. Lazaroa 26, 55-66. 

Sattler, P. and Creighton, C. (Eds.) (2002) Australian Terrestrial Biodiversity Assessment 2002. National Land and Water Resources Audit, Land 

and Water Australia, Braddon, ACT. http://audit.ea.gov.au/anra/vegetation/docs/biodiversity/bio_assess 

200

Saunders, D.A. (1999) Biodiversity does matter. In T. Barlow and R. Thorburn (Eds.), Balancing Conservation and Production in Grassy 

Landscapes. Proceedings of the Bushcare Grassy Landscapes Conference, Clare, South Australia, 19-21 August 1999. Environment Australia, 

Canberra, pp. 14-18. 

Scarlett, N.H. (1994) Soil crusts, germination and weeds – issues to consider. Victorian Naturalist 111(4), 125-130. 

Scarlett, N., Brawell, M. and Earl, G. (2003) Austral Toad Flax Thesium australe. Flora and Fauna Guarantee Action Statement No. 56, 


Scarlett, N.H. and Parsons, R.F. (1993) Rare or threatened plants in Victoria. In D.B. Foreman and N.G. Walsh (Eds.), Flora of Victoria Volume 

1. Introduction, Inkata Press, Melbourne, Ch.10, pp. 227-255. 

Scarlett, N.H., Wallbrink, S.J. and McDougall, K. (1992) Field Guide to Victoria’s Native Grasslands. Native Plants of Victorian Lowland 

Plains. National Trust of Australia (Victoria), Victoria Press, South Melbourne. 

Schimper, A.F.W. (1903) Plant Geography upon a Physiological Basis. Translated by W.R.Fisher, revised and edited by P. Groom and I.B. 

Balfour, Clarendon Press, Oxford, UK. 

Schmidt, C.D., Hickman, K.R., Channell, R., Harmoney, K. and Stark, W. (2008) Competitive abilities of native grasses and non-native 

(Bothriochloa spp.) grasses. Plant Ecology 197, 69-80. 

Scholz, H. and Krigas, N. (2004) Nassella neesiana (Trin. and Rupr.) Barkworth, pp. 78-79 in W.Greuter and T. Raus (Eds.) Med-checklist 

Notulae, 22. Willdenowia 34, 71-80. 

Schomburgk, R. (1875) The Flora of South Australia. Adelaide. [Reference not seen] 

Schroder, P. (1983) Insect pests of pasture in the Hamilton district. Agnote 2295/83, Department of Agriculture Victoria, 3 pp. 

Seabloom, E.W., Harpole, W.S., Reichman, O.J. and Tilman, D. (2003) Invasion, competitive dominance, and resource use by exotic and native 

California grassland species. Proceedings of the National Academy of Sciences (USA) 100(23), 13384-13389. 

Seebeck, J. and Mansergh, I. (2003) Extinct mammals 1. Flora and Fauna Guarantee Action Statement No. 13, Department of Sustainability and 


Seebeck, J.H. and Rose, R.W. (1989) Potoroidae. In D.M. Walton and B.J. Richardson (Eds.), Fauna of Australia Volume 1B. Mammalia. 


Sendulsky, T., Filgueiras, T.S. and Burman, A.G. (1986) Fruits, embryos, and seedlings. In T.S. Soderstrom, K.W. Hilu, C.S. Campbell and M.E. 

Barkworth (Eds.), Grass Systematics and Evolution, Smithsonian Institution Press, Washington, USA, pp.31-36. 

SGAP (Society for Growing Australian Plants) Maroondah, Inc. (1991) Flora of Melbourne. A Guide to the Indigenous Plants of the Greater 

Melbourne Area. 1st Ed., SGAP Maroondah, Inc., Ringwood, Victoria. 

Shapiro, A.M. (2002) The Californian urban butterfly fauna is dependent on alien plants. Diversity and Distributions 8, 31-40. 

Sharp, S. (1997) Diversity, patterns and processes of vegetation and invertebrate orders in natural temperate grassland in the Australian Capital 

Territory. M.App.Sci. thesis, University of Canberra. 

?Sharp, S. (2002) Chilean needle grass survey 2002. Summary of results of consultant surveys in ACT. Unpublished report, Environment ACT, 4 

pp. 

Sharp, S. (2006) Assessment of vegetation condition of grassy ecosystems in the Australian Capital Territory. Ecological Management and 

Restoration 7, S63-S65. 

Sharp, S.B. and Shorthouse, D.J. (1996) Native grasslands and threatened species in the Australian Capital Territory: steps towards protection and 

recovery. In S. Stephens and S. Maxwell (Eds.), Back from the Brink: Refining the Threatened Species Recovery Process, Surrey Beatty and 

Sons, pp. 138-149. 

Shea, K. and Chesson, P. (2002) Community ecology theory as a framework for biological invasions. Trends in Ecology and Evolution 17(4), 

170-176. 

Sheehan, M. (2008) Chilean Needle Grass has shot ... in Balranald! Posting to cngnetwork@lists.weeds.org.au, 23 September. 

Sheley, R.L. and Rinella, M.J. (2001) Incorporating biological control into ecologically based weed management. In E. Wajnberg, J.K. Scott and 

P.C. Quimby (Eds.), Evaluating Indirect Effects of Biological Control. CAB International, Wallingford,UK, pp. 211-228. 

Shepherd, R.C.H., Richardson, R.G. and Richardson, F.J. (2001) Plants of Importance to Australia. A Checklist. R.G. and F.J.Richardson, 

Meredith, Victoria. 

Shigesada, N. and Kawasaki, K. (1997) Biological Invasions: Theory and Practice. Oxford University Press, Oxford, UK. 

Short, J. and Smith, A. (1994) Mammal decline and recovery in Australia. Journal of Mammalogy 75(2), 288-297. 

Short, J. and Turner, B. (1994) A test of the vegetation mosaic hypothesis:a hypothesis to explain the decline and extinction of Australian 

mammals. Conservation Biology 8(2), 439-449. 

Siddiqi, M.R. (1986) Tylenchida. Parasites of Plants and Insects. Commonwealth Institute of Parasitology, Commonwealth Agricultural Bureau, 

Farnham Royal, Slough, UK. 

Siebert, B.D. and Newman, D.M.R. (1989) Camelidae. In D.M. Walton and B.J. Richardson (Eds.), Fauna of Australia Volume 1B. Mammalia. 


Sinclair, R. (2002) Ecophysiology of grasses. In: K. Mallett and A.E. Orchard, A.E., Flora of Australia Volume 43. Poaceae 1. Introduction and 

Atlas. Australian Biological Resources Survey/CSIRO, Melbourne, pp. 133-156. 

Sinclair, S. (2007) Native grassland at Safety Beach, Morninton Peninsula, Victoria. Victorian Naturalist 124(3), 132-149. 

Singh, D.K., Bird, P.R. and Saul, G.R. (2003) Maximising the use of soil water by herbaceous species in the high rainfall zone of southern 

Australia: a review. Australian Journal of Agricultural Research 54, 677-691. 

Slater, J.A. (1976) The biology, distribution and taxonomy of some Lygaeidae of southwest Australia (Hemiptera: Heteroptera). Journal of the 

Australian Entomological Society 15, 129-151. 

Slater, J.A. (1991) Lygaeidae. In CSIRO, The Insects of Australia. A Textbook for Students and Research Workers. 2nd Ed., Melbourne 

University Press, Carlton, pp. 499-503. 

Slay, M. (2001) Annual Report of Activities of the Hawke’s Bay Chilean Needle Grass Group. Homelea AgTechnology and Advisory Service and 

AgResearch Poukawa, Hastings and Havelock North, 59 pp. 

201

Slay, M.W.A. (2002a) Review: Distribution biology and control of Chilean needle grass. Report prepard for the Hawke’s Bay Chilen Needle 

Grass Group, Homelea AgTechnology and Advisory Sevices and AgResearch Poukawa, Hastings and Havelock North, New Zealand, March, 50 

pp. 

Slay, M.W.A. (2002b) Chilean needle grass control in Hawke’s Bay. Chemical topping. Report prepared for the Hawke’s Bay Chilean Needle 

Grass Group, Homelea AgTechnology and Advisory Sevices and AgResearch Poukawa, Hastings and Havelock North, New Zealand, July, 30 pp. 

Slay, M. (2002c) Chilean Needle Grass. A Guide to Identification and Management in Hawke’s Bay. Hawke’s Bay Regional Council, Napier, 

New Zealand, 44 pp. 

Smith, B.J. and Kershaw, R.C. (1979) Field Guide to the Non-marine Molluscs of South Eastern Australia. Autralian National University Press, 

Canberra. 

Smith, Z.F., James, E.A., McDonnell, M.J. and McLean, C.B. (2009) Planting conditions improve translocation success of the endangered 

terrestrial orchid Diuris fragrantissima (Orchidaceae). Australian Journal of Botany 57, 200-209. 

Snell, K., Grech, C. and Davies, J. (2007) National Best Practice Management Manual Chilean Needle Grass. Defeating the Weed Menace, 

Victorian Department of Primary Industries and National Chilean Needle Grass Taskforce, Melbourne. 

Soriano, A., León, R.J.C., Sala, O.E., Lavado, R.S., Deregibus,V.A., Cauhépé, M.A., Scaglia, O.A., Velázquez, C.A. and Lemcoff, J.H. (1992) 

Río de la Plata grasslands. In R.T. Coupland (Ed.), Ecosystems of the World 8A. Natural Grasslands. Introduction and Western Hemisphere. 

Elsevier Sceince Publishers B.V., Amsterdam, Netherlands, pp. 367-407. 

Soto, L.K. (1980) Descripcion de las praderas naturales de Magallanes Continental. 1. Area cubierta y composicion botanica. Agricultura tecnica 

44, 185-193. [Reference not seen] 

Specht, R.L. (1970) Vegetation. In G.W. Leeper (Ed.), The Australian Environment, 4th Ed., CSIRO and Melbourne University Press, Carlton, 

Vic., pp. 44-67. 

Specht, R.L. (2000) Savannah woodland vegetation in the South-East District of South Australia: the influence of evaporative aerodynamics on 

the foliar structure of the understorey invaded by introduced annuals. Australian Journal of Ecology 25(6), 588-599. 

Spencer, K.A. (1977) A Revision of the Australian Agromyzidae (Diptera). Western Australian Museum Special Publication No. 8, 255 pp., 

Perth. 

Spencer, K.A. (1989) Family Agromyzidae. In N.L. Evenhuis (Ed.), Catalog of the Diptera of the Australasian and Oceanian Regions. Bishop 


Stace, C. (1997) New Flora of the British Isles. 2nd Ed., Cambridge University Press, Cambridge, UK. 

Stafford, J.L. (1991) Techniques for the establishment of kangaroo grass in South Australian conservation reserves. Plant Protection Quarterly 

6(3), 120-122. 

Stanton, R. (2006) Weed seed dispersal by sheep and cattle. Plant Protection Quarterly 21(4), 145-147. 

Stebbins, C.L. (1972) The evolution of the grass family. In V.B. Youngner and C.M. McKell (Eds.), The Biology and Utilization of Grasses. 

Academic Press, New York, pp. 1-17. 

Stebbins, C.L. (1986) Grass systematics and evolution: past, present and future. In T.S. Soderstrom, K.W. Hilu, C.S. Campbell and M.E. 


Stephens, C. (2006) Claire Stephens. Research topic and project outline. Impact of weed invasion on insect biodiversity and associated 

community structure and processes. School of Earth and Environmental Sciences, University of Adelaide. 

http://www.ees.adelaide.edu.au/people/enviro/PhD/csteve01.html [Accessed 29 September 2006]. 

Stewart, J. (1996) Observations after fire in a degraded grassland. Victorian Naturalist 113, 102-105. 

Stohlgren, T.J. (2007) Measuring Plant Diversity. Lessons from the Field. Oxford University Press, Oxford, UK. 

Stoner, J.R., Adams, R. and Simmons, D. (2004) Management implications of increased fuel loads following exotic grass invasion. Ecological 

Management and Restoration 5(1), 68-69. 

Storrie, A. (2006) The weed control paradox or how and when do I learn to love my weeds? Major pasture weeds in northern NSW. In B. 

Hackney, K. Bailes, J. Pilz and H. Burns (Eds.), Proceedings of the 21st Annual Conference of the Grassland Society of New South Wales Inc., 

pp. 28-32. 

Storrie, A.M. and Gardener, M.R. (1998) Chilean needle grass. New South Wales Agriculture Agnote DPI 194. 

Storrie, A.M. and Lowien, J.C. (2003) Chilean needle grass. New South Wales Agriculture Agnote DPI 194, 3rd Ed., 5 pp. 

Stott, P. (2007) Comparisons of digestive function between the European hare (Lepus europaeus) and the European rabbit (Oryctolagis 

cuniculus): mastication, gut passage and digestibility. Mammalian Biology 73, 276-286. 

Strong, D.R. and Pemberton, R.W. (2002) Food webs, risks of alien enemies and reform of biological control. In E. Wajnberg, J.K. Scott and P.C. 

Quimby (Eds.), Evaluating Indirect Ecological Effects of Biological Control, CABI Publishing, Wallingford, UK, pp. 57-79. 

Stuwe, J. (1986) An assessment of the conservation status of native grasslands on the Western Plains, Victoria, and sites of botanical significance. 

Arthur Rylah Institute for Environmental Research Technical Report Series No. 48. 

Stuwe, J. (1994) The role of fire in ground flora ecology. Victorian Naturalist 111(3), 93-95. 

Stuwe, J. and Parsons, R.F. (1977) Themeda australis grasslands on the Basalt Plains, Victoria: floristics and management effects. Australian 

Journal of Ecology 2, 467-476. 

Sutton, C.S. (1916-1917) A sketch of the Keilor Plains flora. Victorian Naturalist 33, 112-123, 128-143. 

Swarbrick, J.T. (1991) Towards a rating scheme for environmental weeds. Plant Protection Quarterly 6(4), 185. 

Swarbrick, J.T. and Skarratt, D.B. (1994) The Bushweed 2 Database of EnvironmentalWeeds in Australia. 2nd Ed., Department of Plant 

Production, University of Queensland Gatton College, Gatton. 

Sweedman, L. and Merritt, D. (Eds.) Australian Seeds. A Guide to their Collection, Identification and Biology. CSIRO Publishing, Collingwood. 

Sydes, M. (1994) Orchids: indicators of management success? Victorian Naturalist 111, 213-217. 

Symstad, A.J. (2000) A test of the effects of functional group richness and composition on grassland invasibility. Ecology 81(1), 99-109. 

Systematic Entomology Laboratory USDA (2001) Eurytomidae generic synopsis. 

http://www.sel.barc.usda.go/hym/chalcids/eurytomid/gen_syn.html [Accessed 27 July 2007] 

202

Thellung, A. (1912) La flore adventice de Montpellier. Mémoires de la Société Nationale des Sciences Naturelles et Mathématiques de 

Cherbourg 38, 57-728. 

Thomasson, J.R. (1986) Fossil grasses: 1820-1986 and beyond. In T.S. Soderstrom, K.W. Hilu, C.S. Campbell and M.E. Barkworth (Eds.), Grass 


Thorp, J.R. and Lynch, R. (2000) The Determination of Weeds of National Significance. National Weeds Strategy Executive Committee, 

Launceston. 

Tomlin, C.D.S. (Ed.) (1997) The Pesticide Manual. 11th Ed., British Crop Protection Council, Farnham, Surrey, UK. 

Torres, M.A. (1993) Revision del genero Stipa (Poaceae) en la Provincia de Buenos Aires. Provincia de Buenos Aires, Comisión de 

Investigaciones Científicas, Monografia 12, 62 pp., La Plata. 

Torres, M.A. (1997) Nassella (Gramineae) del noroeste de la Argentina. Stipa (Gramineae) del noroeste de la Argentina. Nicoraella (Gramineae) 

un nuevo género para América del Sur., Ministerio de la Producción y el Empleo, Provincia de Buenos Aires, Comisión de Investigaciones 

Científicas, Monografia 13, La Plata. 

Traill, B. and Porter, C. (2001) Nature Conservation Review Victoria 2001. Victorian National Parks Association Inc., East Melbourne. 

Trémont, R.M. (1994) Life-history attributes of plants in grazed and ungrazed grasslands on the Northern Tablelands of New South Wales 


Trémont, R.M. and McIntyre, S. (1994) Natural grassy vegetation and native forbs in temperate Australia: structure, dynamics and life histories. 


Trengrove, M. (1997) Letter to David McLaren, Keith Turnbull Research Institute, from Geelong Indigenous Nursery, re Nassella trichotoma and 

N. neesiana, 4 December, 1 p. 

Tscharntke, T. and Greiler, H.-J.(1995) Insect communities, grasses and grasslands. Annual Review of Entomology 40, 535-558. 

Tsvelev, N.N. (1977) On the origin and evolution of feathergrasses (Stipeae, Poaceae). In Problemi Ekologii, Geobotaniki, Botanichheskoi 

Geografii i Floristiki, Academiya SSSR, Leningrad, pp. 139-150.[Translated by K. Gonzalez and M.E. Barkworth, 2007, 

http://utc.usu.edu/Stipeae/Tsvelev.htm, accessed 16 July 2007] 

Tsvelev, N.N. (1984) Grasses of the Soviet Union. Translated by B.R. Sharma. A.A. Balkema, Rotterdam. 

Turner, J.S.(1968) The decline of the plants. In A.J. Marshall (Ed.) The Great Extermination. A Guide to Anglo-Australian Cupidity, Wickedness 

and Waste. Panther Books, London, pp. 154-177. 

Tumino, M. (2004) Spiny Rice-flower, Pimelea spinescens subsp. spinescens. Flora and Fanua Guarantee Action Statement No. 132, 


Twigg, L.E., Lowe, T.J., Taylor, C.M., Calver, M.C., Martin, G.R., Stevenson, C. and How, R. (2009) The potential of seed-eating birds to spread 

viable seeds of weeds and other undesirable plants. Austral Ecology 34, 805-820. 

United Nations (1992) Convention on Biodiversity. Rio de Janeiro, 5 June 1992 http://www.cbd.int/doc/legal/cbd-un-en.pdf 

USDA ARS (United State Department of Agriculture, Agricultural Research Service) (2006) Germplasm Resources Information Network (GRIN) 

[Online database]. National Germplasm Resources Laboratory, Beltsville, Maryland. http://www.ars-grin.gov/cgi-bin/npgs/html/taxon.pl?402089 

(accessed 14 December 2006). 

USDA FEIS (United State Department of Agriculture, Fire Effects Information System) (2006) Nassella viridula [Online database] 

http://www.fs.fed.us/database/feis/plants/graminoid/nasvir/all.html (accessed 14 December 2006). 

USDA NRCS (Natural Resources Conservation Service) (2006) Plants Profile. http://plants.usda.gov (accessed 14 December 2006). 

USFDA (United States Food and Drug Administration) (2006) FDA Poisonous Plant Database. [Online database]. Centre for Food Safety and 

Applied Nutrition, Office of Plant and Dairy Foods. http://www.cfsan.fda.gov/~djw/pltx.cgi?QUERY=Nassella+neesiana 

Uys, R., Bond, W.J. and Everson, T.M. (2004) The effects of different fire regimes on plant diversity in southern African grasslands. Biological 


van Andel, J. and van den Bergh, J.P. (1987) Disturbance of grasslands. Outline of the theme. In J. van Andel, J.P. Bakker and R.W. Snaydon (Eds.), 

Disturbance in Grasslands. Causes, Effects and Processes. Dr. W. Junk Publishers, Dordrecht, pp. 3-13. 

Vánky, K. and Shivas, R.G. (2008) Fungi of Australia. The Smut Fungi. Australian Biological Resources Study, Canberra. 

Vázquez, F.M. and Barkworth, M.E. (2004) Resurrection and emendation of Macrochloa (Gramineae: Stipeae). Botanical Journal of the Linnean 

Society 144, 483-495. 

Vázquez, F.M. and Devesa, J.A. (1996) Revisiòn del género Stipa L. y Nassella Desv. (Poaceae) en la Península Ibérica e Islas Baleares. Acta 

Botanica Malacitana 21, 125-189. 

Valentine, L.E. (2006) Habitat avoidance of an introduced weed by native lizards. Austral Ecology 31, 732-735. 

Veldtman, R. and McGeoch, M.A. (2003) Gall-forming insect species richness along a non-scleromorphic vegetation rainfall gradient in South 

Africa: the importance of plant community composition. Austral Ecology 28, 1-13. 

Venn, D. and Menkhorst, P. (2003) Australian Bustard Ardeotis australis. Flora and Fanua Guarantee Action Statement No. 179, Department of 


Verbeek, B. (2006)) New and existing weed threats in southern New South Wales. In B. Hackney, K. Bailes, J. Piltz and H. Burns (Eds.), 

Proceedings of the 21st Annual Conference of the Grassland Society of NSW Inc., Wagga Wagga, July, pp. 25-27. 

Verloove, F. (2005) A synopsis of Jarava Ruiz and Pav. and Nassella E. Desv. (Stipa L. s.l.) (Poaceae: Stipeae) in southwestern Europe. 

Candollea 60(1), 97-117. 

Versfeld, D.B., Le Maitre, D.C. and Chapman, R.A. (1998) Alien invading plants and water resources in South Africa: a preliminary assessment. 

Water Research Commission Report No. TT99/98, Pretoria. 

Versfeld, D.B. and van Wilgen, B.W. (1986) Impact of woody aliens on ecosytem properties. In I.A.W. Macdonald, F.J. Kruger and A.A. Ferrar 

(Eds.), The Ecology and Management of Biological Invasions in Southern Africa. Oxford University Press, Cape Town, pp. 239-246. 

Vickery, J.W., Jacobs, S.W.L. and Everett, J. (1986) Taxonomic studies in Stipa (Poaceae) in Australia. Telopea 3(1), 1-132. 

Vidler, S.J. (2004) Using your cute and furries: the role of threatened species in weed awareness. In B.M. Sindel and S.B. Johnson (Eds.), 14th 

Australian Weeds Conference Papers and Proceedings. Weed Society of New South Wales, Sydney, pp. 652-658. 

Villarreal, D., Clark, K.L., Branch, L.C., Hierro, J.L. and Machicote, M. (2008) Alteration of ecosystem structure by a burrowing herbivore, the 

Plains Vizcacha (Lagostomus maximus). Journal of Mammalogy 89(3), 700-711. 

203

Vines, G. (2006) Aliens in sheep’s clothing. New Scientist 190(2546), 54-55. 

Virtue, J.G., Crossman, N.D. and Richards, N. (2004) Future directions for managing perennial grass weeds in South Australia. Plant Protection 

Quarterly 19(2), 84-88. 

Virtue, J.G. and Melland, R.L. (2003) The Environmental Weed Risk of Revegetaion and Forestry Plants. Department of Water, Land and 

Biodiversity Conservation, South Australia, 169 pp. 

Vitou, J., Scott, J.K. and Sheppard, A.W. (2004)The patterns of harvester ant removal of wild radish seeds inthe native range: the importance of 

generalist seed predators to weed management. In J. M. Cullen, D.T. Briese, D.J. Kriticos, W.M. Lonsdale, L. Morin and J.K. Scott (Eds.), 

Proceedings of the XI International Symposium on Biological Control of Weeds, CSIRO, Canberra, pp. 188-192. 

Vitousek, P.M., D’Antonio, C.M., Loope, L.L., Rejmánek, M. and Westbrooks, R. (1997) Introduced species: a significant component of humancaused 

global change. New Zealand Journal of Ecology 21(1), 1-16. 

Vitousek, P.M. and Walker, L.R. (1989) Biological invasion by Myrica faya in Hawai'i: plant demography, nitrogen fixation, ecosystem effects. 

Ecological Monographs 59(3), 247-265. 

Vitousek, P.M.,Walker, L.R. ,Whiteaker, L.D., Mueller-Dombois, D. and Matson, P.A. (1987) Biological invasion by Myrica faya alters 

ecosystem development in Hawaii. Science 238(4828), 802-4.: 

Von der Lippe, M. and Kowarik, I. (2007a) Long distance dispersal of plants by vehicles as a driver of plant invasions. Conservation Biology 

21(4), 986-996. 

Von der Lippe, M. and Kowarik, I. (2007b) Do cities export biodiversity? Traffic as dispersal vector across urban-rural gradients. Diversity and 

Distributions 14(1), 18-25. 

Wadham, S.M. and Wood, G.L. (1950) Land Utilization in Australia. 2nd Ed., Melbourne University Press, Carlton. 

Wainer, J.W. (1992) Diet of the Striped Legless Lizard (Delma impar) at the Derrimut Grassland Reserve, to the west of Melbourne. Unpublished 

report. 6 pp. 

Wajnberg, E.,Scott, J.K. and Quimby, P.C. (Eds.) (2001) Evaluating Indirect Ecological Effects of Biological Control. CABI Publishing, 

Wallingford, UK. 

Wakefield, N.A. (1963a) Sub-fossils from Mount Hamilton, Victoria. Victorian Naturalist 79(11), 323-330. 

Wakefield, N.A. (1963b) Mammal sub-fossils from near Portland, Victoria. Victorian Naturalist 80(2), 39-45. 

Wakefield, N.A. (1964a) Mammal sub-fossils from basalt caves in south-western Victoria. Victorian Naturalist 80(9), 274-278. 

Wakefield, N.A. (1964b) Recent mammalian sub-fossils of the basalt plains of Victoria. Proceedings of the Royal Society of Victoria 77(2), 419- 

425. 

Walsh, N.G. (1994) Poaceae. In N.G. Walsh and T.J. Entwisle (Eds.) (1994) Flora of Victoria Volume 2. Ferns and Allied Plants, Conifers and 

Monocotyledons. Melbourne, Inkata Press, pp. 356-627. 

Walsh, N.G. (1998) Identification of South American tussock weeds. Plant Protection Quarterly 13(2), 59-62. 

Walsh, N.G. (1999) Senecio. In N.G. Walsh and T.J. Entwisle (Eds.) (1994) Flora of Victoria Volume 4. Dicotyledons. Cornaceae to Asteraceae. 

Melbourne, Inkata Press, pp. 941-965. 

Walsh, N.G. and Stajsic, V. (2007) A Census of the Vascular Plants of Victoria. 8th Ed., National Herbarium of Victoria, South Yarra. 

Walter, H.S. and Levin, G.A. (2008) Feral sheep on Socorro Island: facilitators of alien plant colonization and ecosystem decay. Diversity and 

Distributions 14, 422-431. 

Wang, B. and Qiu, Y.-L. (2006) Phylogenetic distribution and evolution of mycorrhizas in land plants. Mycorrhiza 16, 299-363. 

Wang, Q., Jin, S. and Ruan, X. (2009) Ecological explanations for successful invasion of exotic plants. Frontiers of Biology in China 4, 271-281. 

Wapshere, A.J. (1990) Biological control of grass weeds in Australia: an appraisal. Plant Protection Quarterly 5(2), 62-75. 

Wapshere, A.J. (1993) Aspects of biological control of grass weeds in pastures. In E.S. Delfosse (Ed.), Pests of Pastures. Weed, Invertebrate and 

Disease Pests of Australian Sheep Pastures. CSIRO Information Services, Melbourne, pp. 342-346. 

Waterhouse, D.F. (1998) Foreword. In M.H. Julien and M.W. Griffiths, M.W. (Eds.) Biological Control of Weeds. A World Catalogueof Agents 

and their Target Weeds. 4th Ed., CABI Publishing, Wallingford, UK, p. vi. 

Watson, J.A.L. (1982) Distribution, biology and speciation in the Australian harvester termites, Drepanotermes (Isoptera: Termitinae). In W.R. 

Barker and P.J.M. Greenslade, Evolution of the Flora and Fauna of Arid Australia, Peacock Publications in association with Australian 

Systematic Botany Society and ANZAAS South Australian Division Inc., Frewville, South Australia, pp. 263-265. 

Watson. L. and Dallwitz, M.J. (2005) The grass genera of the world: descriptions, illustrations, identification, and information retrieval; including 

synonyms, morphology, anatomy, physiology, phytochemistry, cytology, classification, pathogens, world and local distribution, and references. 

Version: 28 November 2005. http://delta-intkey.com/grass [Accessed 16 July 2007] 

Weaver, V. and Adams, R. (1996) Horses as vectors in the dispersal of weeds into native vegetation. In R.C.H. Shepherd (Ed.), Proceedings 

Eleventh Australian Weeds Conference, Melbourne 30 September - 3 October 1996, Weed Science Society of Victoria, Frankston, pp. 383-387. 

Webb, S.D. (1977) A history of savanna vertebrates in the New World. Part I: North America. Annual Review of Ecology and Systematics 8, 355- 

380. 

Webb, S.D. (1978) A history of savanna vertebrates in the New World. Part II: South America and the Great Interchange. Annual Review of 

Ecology and Systematics 9, 393-426. 

Weber, E. (2003) Invasive Plant Species of the World. A Reference Guide to Environmental Weeds. CABI Publishing, Wallingford, UK. 

Webster, A., Backhouse, G., Murphy, A.H. and Knight, C. (2004) Sunshine Diuris Diuris fragrantissima. Flora and Fanua Guarantee Action 

Statement No. 50, Department of Sustainability and Environment, East Melbourne, revised ed., 8 pp. 

Webster, A., Fallu, R. and Preece, K. (2003) Striped Legless Lizard Delma impar. Flora and Fanua Guarantee Action Statement No. 17, 


Wedin, D.A. (1999) Nitrogen availability, plant-soil feedbacks and grassland stability. In D. Eldridge and D. Freudenberger (Eds.), VIth 

International Rangeland Congress Proceedings Vol. 1, International Rangeland Congress Inc., Townsville, pp. 193-197. 

Wei, Y.K., Gao, Y.B., Zhang, X., Su, D., Wang, Y.-H., Xu, H., Lin, F., Ren, A. –Z., Chen, L. and Nie, L. Y. (2007) Distribution and diversity of 

Epichloë/Neotyphodium fungal endophytes from different populations of Achnatherum sibiricum (Poaceae) in the Inner Mongolian Steppe, 

China. Fungal Diversity 24, 329-345. 

204

Weiss, J. and McLaren, D. (2002) Victoria’s pest plant prioritisation process. In H. Spafford Jacob, J. Dodd and J.H. Moore (Eds.), 13th 

Australian Weeds Conference Papers and Proceedings, 8-13 September 2002, Plant Protection Society of Western Australia Inc., Perth, pp. 509- 

512. 

Weiss, J.,McLaren, D. and Itami, B. (1999) Utilising a decision support system to prioritise Victoria’s noxious weeds. In A.C. Bishop, M. 

Boersma and C.D. Barnes (Eds.), 12th Australian Weeds Conference Papers and Proceedings, Tasmanian Weed Society, Devenport, p. 586. 

Weiss, P.W. and Noble, I.R. (1984) Status of coastal dune communities invaded by Chrysanthemoides monilifera. Australian Journal of Ecology 

9, 93-98. 

Wells, M.J., Balsinhas, A.A., Joffe, H., Engelbrecht, V.M., Harding, G. and Stirton, C.H. (1986) A Catalogue of Problem Plants in Southern 

Africa incorporating the National Weed List of Southern Africa. Memoirs of the Botanical Survey of South Africa No. 53. 

Wells, M.J. and Stirton, C.H. (1982) Weed problems of South African pastures. In W. Holzner and N. Numata (Eds.), Biology and Ecology of 

Weeds. Dr W. Junk Publishers, The Hague, Netherlands, pp. 429-434. 

Wensler, R.J. (1971) Locomotor and feeding activity of larvae of the scarabaeid Sericesthis geminata (Coleoptera). Entomologia Experimentalis 

et Applicata 14, 270-282. 

Wheeler, D.J.B., Jacobs, S.W.L. and Norton, B.E. (1990) Grasses of New South Wales. 2nd Ed., University of New England, Armidale. 

Whelan, C.J., Wenny, D.G. and Marquis, R.J. (2008) Ecosystem services provided by birds. Annals of the New York Academy of Sciences 1134, 

25-60. 

Whitney, K.D. and Gabler, C.A. (2008) Rapid evolution in introduced species, ‘invasive traits’ and recipient communities: challenges for 

predicting invasive potential. Diversity and Distributions 14, 569-580. 

Whittet, J. N. (compiler) (1969) The Farmers’ Handbook Series. Pastures. 2nd Ed., Department of Agriculture, New South Wales. 

Wigney, R. (Coordinating Ed.) (1994) Plants of the Merri Merri. A Home Gardener’s Guide to Using Indigenous Plants in Northern Suburbs of 

Melbourne. Revised Ed., Merri Creek Management Committee, Northcote, Victoria. 

Wijesuriya, W.S. (1999) Investigation of the relationships between biomass reduction, soil disturbance, soil nutrients and weed invasion in basalt 

plains native grassland remnants in Victora, Australia. PhD thesis, School of Life Sciences and Technology, Faculty of Engineering and Science, 

Victoria University, St Albans. 

Wijesuriya, S. and Hocking, C. (1999) Why do weeds grow when you dig up native grasslands? The effects of physical disturbance on available 

nutrients, mineralisation and weed invasion in grassland soils. In V. Craigie and C. Hocking (Eds.), Down to Grass Roots. Proceedings of a 

Conference on Management of Grassy Ecosystems, 9 and 10 July 1998, Victoria University, St Albans, Vic., pp. 31-37. 

Wilkie, L., Cassis, G. and Gray, M.(1999) Quality control in invertebrate biodiversity data compilations. In: W. Ponder and D. Lunney (Eds.), 

The Other 99%. The Conservation and Biodiversity of Invertebrates. Transactions of the Royal Zoological Society of New South Wales, Mosman, 

pp. 147-153. 

Willatt, S.T. and Pullar, D.M. (1983) Change in soil physical properties under grazed pastures. Australian Journal of Soil Research 22, 343-348. 

Williams, D.J. (1991) Superfamily Coccoidea. In CSIRO, The Insects of Australia. A Textbook for Students and Research Workers. 2nd Ed., 


Williams, J.A. and West, C.J. (2000) Environmental weeds in Australia and New Zealand: issues and approaches to management. Austral 

Ecology 25, 425-444. 

Williams, L.M. (1995) Tasmanian Pademelon. In P.W. Menkhorst (Ed.), Mammals of Victoria. Distribution, Ecology and Conservation. Oxford 


Williams, N.S.G. (2007) Environmental, landscape and social predictors of native grassland loss in western Victoria, Australia. Biological 


Williams, N.S.G., Morgan, J.W., McCarthy, M.A. and McDonnell, M.J. (2006) Local extinction of grassland plants: the landscape matrix is more 

important than patch attributes. Ecology 87(12), 3000-3006. 

Williamson, M.H. and Fitter, A. (1996) The characteristics of successful invaders. Biological Conservation 78,163-170. 

Willis, J.H. (1964) Vegetation of the basalt plains in Western Victoria. Proceedings of the Royal Society of Victoria (New Series) 77(2), 397-418. 

Willis, J.H. (1970) A Handbook to Plants in Victoria. Volume I. Ferns, Conifers and Monocotyledons. 2nd Ed., Melbourne University Press, 

Carlton. 

Wink, M. and van Wyk, B.-E. (2008) Mind-altering and Poisonous Plants of the World. Briza Publications, Pretoria, South Africa. 

Witt, A.B.R. and McConnachie, A.J. (2004) The potential for classical biological control of invasive grass species with special reference to 

invasive Sporobolus spp. (Poaceae) in Australia. In J.M.Cullen, D.T. Briese, D.J. Kriticos, W.M. Lonsdale, L. Morin and J.K. Scott (Eds.), 

Proceedings of the XI International Symposium on Biological Control of Weeds, Canberra, Australia, 27 April-2 May 2003, pp. 198-202. 

Wolters, V., Bengtsson, J. and Zaitsev, A.S. (2006) Relationship among the species richness of different taxa. Ecology 87(8), 1886-1895. 

Woods, K.D. (1997) Community response to plant invasion. In J.O. Luken and J.W. Thieret (Eds.), Assessment and Management of Plant 

Invasions. Springer-Verlag, New York, pp. 57-68. 

Yen, A.L. (1995) Some issues involved in the conservation of lowland native grassland invertebrates. In S. Sharp and R. Rehwinkel (Eds.), 

Management of Relict Lowland Grasslands. Proceedings of a Workshop and Public Seminar, September 24 and 25 1993. ACT Parks and 


Yen, A.L. (1999) Grassland invertebrates of the Western Victorian Basalt Plains: plant crunchers or forgotten lunches? In R.N. Jones (Ed.), The 

Great Plains Crash. Proceedings of a Conference on the Grasslands and Grassy Woodlands of Victoria. Victorian Institute of Technology 

October 2nd and 3rd, 1992, Indigenous Flora and Fauna Association and Victorian National Parks Association, pp. 57-68. 

Yen, A.L., Horne, P.A., Kay, R. and Kobelt, A.J. (1994a) The use of terrestrial invertebrates to rank sites of the remnant western Victorian Basalt 

Plains Grasslands. Report to the Endangered Species Unit, Australian Nature Conservation Agency, June, 155 pp. 

Yen, A.L., Kobelt, A.J., Lillywhite, P.K. and Van Praagh, B. (1994b) The terrestrial invertebrates of the western region of Melbourne. Report for 

the Department of Planning and Development with assistance from the Western Suburbs Action Program 1989/90, Department of Planning and 

Urban Growth, July, 116 pp. 

Yen, A.L., Horne, P.A. and Kobelt, A.J. (1995) Invertebrates of the Victorian Basalt Plains Grasslands. Report to the Australian Heritage 

Commission for National Estate Program Grant 1993/94 Project No. 122 from the Museum of Victoria, July, 88 pp. 

Yeoh, H. and Watson, L. (1986) Taxonomic patterns in protein amino acid profiles of grass leaves and caryopses. In T.S. Soderstrom, K.W. Hilu, 

C.S. Campbell and M.E. Barkworth (Eds.), Grass Systematics and Evolution, Smithsonian Institution Press, Washington, USA, pp.88-96. 

205

Yugovic, J. and Mitchell, S. (2006) Ecological review of the Koo-Wee-Rup Swamp and associated grasslands. Victorian Naturalist 123(5), 323- 

334. 

Zallar, S. (1981) Report on Plant Introduction, Screening, Observation and Evaluation for the Period from 1962 to 1979. Soil Conservation 

Authority of Victoria, Melbourne, II + 72 pp. 

Zanin, A. (2008) Nassella. In Z.E.Rúgolo de Agrasar, E. Nicora, A..M. Molina, J.M. Sallés and A. Zanin, Flora del Paraguay 38. Gramineae VII. 

Conservatoire et Jardin botaniques de la Ville de Genève, Switzerland. 

Zanin, A., Mujica Salles, J. and Longhi Wagner, H.M. (1992) Flora ilustrada do Rio Grande do Sur/22 Gramineae Tribo Stipeae. Beletim do 

Instituto de Biociencias 51, 1-174. [Reference not seen] 

Zimmerman, E.C. (1993) Australian Weevils (Coleoptera: Curculionoidea) Volume III. Nanophyidae, Rhynchophoridae, Erirhinidae, 

Curculionidae: Amycterinae, Literature Consulted. CSIRO and Entomological Society of America, East Melbourne and Lanham, Maryland. 

Zipcode Zoo (2006) Nassella nessiana. http://zipcodezoo.com (Accessed 14 December 2006). 

Zola, N. and Gott, B. (1992) Koorie Plants Koorie People. Traditional Aboriginal Food, Fibre and Healing Plants of Victoria. Koorie Heritage 

Trust, Melbourne. 

Zucol, A.F. (1996) Microfitolitos de las Poaceae Argentinas: I. Microfitolitos foliares de algunas especies del genero Stipa (Stipeae: 

Arundinoideae), de la Provincia de Entre Ríos. Darwiniana 34(1-4), 151-172. 

Zuloaga, F.O, Nicora, E.G., Rúgolo de Agrasar, Z.E, Morrone, O., Pensiero, J. Y Cialdella, A.M. (1994) Catálogo de la familia Poaceae en la 

República Argentina. Missouri Botanical Garden Monographs in Systematic Botany 47. 

206

Literature review: Impact of Chilean needle grass ... - Weeds Australia

Create successful ePaper yourself

Delete template?

Save as template?