DAMIAN CHMURA
Biology and ecology of an invasion
of Impatiens parviflora DC in
natural and semi-natural habitats
Bielsko-Biała
2014
Redaktor Naczelny:
prof. dr hab. Kazimierz Nikodem
Redaktor Działu:
prof. ATH dr hab. inż. Stanisław Rabiej
Recenzenci:
prof. dr hab. Krystyna Falińska
prof. dr hab. Eugenija Kupčinskienė
Sekretarz Redakcji:
mgr Grzegorz Zamorowski
Rozprawy Naukowe nr 50
Publikacja będzie dostępna po wyczerpaniu nakładu w wersji internetowej :
Śląska Biblioteka Cyfrowa www.sbc.org.pl
Wydawnictwo Naukowe Akademii Techniczno-Humanistycznej
w Bielsku-Białej
43-309 Bielsko-Biała, ul. Willowa 2
tel. (33) 827 92 68
e-mail: wydawnictwa@ath.bielsko.pl
ISBN 978-83-63713-68-3
ISSN 1643-983-X
Table of contents
Acknowledgements .................................................................................................................... 5
Introduction ................................................................................................................................ 6
1. The object of the study: Impatiens parviflora .................................................................. 13
1.1. Characteristics of the species ............................................................................................ 13
1.1.1. Taxonomy............................................................................................................... 13
1.1.2. Morphology ............................................................................................................ 13
1.1.3. Biology ................................................................................................................... 14
1.1.4. Habitats................................................................................................................... 17
1.1.5. Biotic interactions .................................................................................................. 18
1.2. History of invasion ............................................................................................................ 23
1.2.1. History of spread in alien range ............................................................................. 23
1.2.2. Impact on native biodiversity and the nature conservancy viewpoint ................... 26
1.2.3. Methods of control ................................................................................................. 27
1.2.4. Economic importance ............................................................................................. 29
2. Methods ................................................................................................................................ 30
2.1. Study areas ........................................................................................................................ 30
2.2. Habitat research ................................................................................................................. 32
2.3. Population studies ............................................................................................................. 34
2.4. Phytosociological research ................................................................................................ 40
2.5. Studies on permanent plots and long-term research.......................................................... 41
2.6. Mycorrhizal research and biotic studies ............................................................................ 46
2.7. Data processing ................................................................................................................. 48
3. Results and discussion .......................................................................................................... 49
3.1. Biotopic requirements of species ...................................................................................... 49
3.1.1. Diversity of substratums and soil conditions ......................................................... 49
3.1.2. Phytoinidication of the patches of the communities with a contribution of small
balsam............................................................................................................................... 56
3.2. Life history traits ............................................................................................................... 58
3.2.1. Diversity of seeds and capacity of germination ..................................................... 58
3.2.2. Morphological variation of individuals .................................................................. 63
3.2.3. Seasonal dynamics of Impatiens parviflora ........................................................... 70
3.2.4. Life history modifications that are dependent on habitat heterogeneity ................ 76
3.3. Contribution of Impatiens parviflora to plant communities and biotic relations .............. 95
3.3.1. The occurrence in plant communities .................................................................... 95
3.3.2. Dynamics of abundance of populations in various plant communities ................ 100
3.3.3. Role of plant functional groups and biotic diversity of plants in Impatiens
parviflora invasion ......................................................................................................... 118
3.3.4. Ecological conditions of the occurrence of mycorrhiza ....................................... 135
3.3.5. Interactions with coexisting plant species ............................................................ 140
4. Synthesis............................................................................................................................. 170
4.1. Traits of invasive species ................................................................................................ 170
4.2. Causes of biological invasions ........................................................................................ 174
4.3. Model of Impatiens parviflora invasion.......................................................................... 179
5. Conclusions ........................................................................................................................ 187
6. References .......................................................................................................................... 189
Appendix 1 ............................................................................................................................. 209
Appendix 2 ............................................................................................................................. 210
Appendix 3. ............................................................................................................................ 213
Streszczenie (Polish summary) .............................................................................................. 215
Acknowledgements
First of all, I would like to express my gratitude to Prof. Krystyna Falińska for our
discussion on the earlier concept of the work. I am especially grateful to my colleagues: Dr.
Edyta Sierka, Dr. Ewa Gucwa-Przepióra, Dr. hab. Anna Orczewska, Dr. hab. Andrzej Urbisz
and MSc Kamil Najberek who helped me at various stages of the research. I am indebted to
my bosses: the late Prof. Halina Piękoś-Mirkowa, Prof. Zygmunt Denisiuk, Prof. Jan
Żarnowiec and Prof. Henryk Klama who offered me support and encouraged me in my work.
Thanks are also due to other “impatientologists”: Prof. Eugenija Kupcinskiene, Dr.
Peter Csontos, Dr. Wojciech Adamowski, Dr. Emanuela Komosińska and Dr. Tomasz
Wyszomirski for their insights and many fruitful comments. I also thank Dr. Wojciech Bąba
and Dr. hab. Gabriela Woźniak for their many valuable suggestions over the years.
I owe a lot to the late Prof. Krzysztof Rostański who introduced me to the world of
botany and to Prof. dr. hab. Barbara Tokarska-Guzik who introduced me to the world of
invasive alien plants. I will keep in memory the late Prof. Janusz Bogdan Faliński for
infecting me with an enthusiasm for doing research.
I would also like to thank to my co-workers at my department: Dr. Anna Salachna for
help in edition of some figures and Dr. Marek Krywult for drawing my attention to
ecophysiological aspects. Michele Simmons improved the language of this manuscript.
I give my warm feelings to my wife Olga as well as to my parents for their support and
constant presence. My son, Piotruś, deserves thanks for keeping me smiling from the day that
he was born.
The major part of the studies was conducted owing to grants: no 3 P04 G093 25 in the
years 2003-2006 and no. N N304 092434 in the years 2008-2011 and to a bilateral project
between the Polish and Hungarian Academy of Sciences in the years 2010-2013.
5
Introduction
In recent decades since the publication of Charles Elton’s book “Invasion by animals
and plants” in 1958, we have been witnesses to the birth of a new branch of ecology –
invasion science (Richardson, Pyšek 2008). The phenomenon of the introduction, spread and
naturalisation of species of alien origin, called invasive alien species (IAS) in a new area is
considered to be the second, after habitat loss, threat to native biodiversity (Wiliamson 1996;
Weber 2003). Thus, besides botanists, zoologists and ecologists, invasions are receiving
attention from naturalists, conservationists and practitioners in nature protection.
Ecologists and invasion biologists seek to answer the question of which species become
invasive or potentially invasive (Williamson, Brown 1986; Weber 1997; Rejmánek et al.
2005) and what traits promote their invasiveness? (Baker 1965; Noble 1989; Rejmánek 1995,
1996; Jackowiak 1999). What are factors that control the vulnerability of habitats, plant
communities and ecosystems to invasion, i.e., the so-called invasibility? (Tilman 1997;
Lonsdale 1999; Davis et al. 2000). And finally, what is the impact of IAS on the native flora
and fauna (Parker et al. 1999; Wilgen et al. 2001; Yurkonis, Meiners 2004)? The most
important question, however, is how can we prevent and control IAS?
The body of literature related to biological invasions is enormous and therefore the
terminology that is used to discuss biological invasions is very complex and often
inconsistent. Authors use many terms to define invasive species and the processes of invasion.
Thus, it is often stressed that this multiplicity of concepts and words can lead to
misunderstandings. In addition to the term “alien”, the species are referred to as “introduced”,
“acclimatized”, “adventive”, “invasive”, “non-indigenous”, “non-native”, “allochthonous” or
“exotic”. Moreover, there are also such terms as “invader”, “weed”(plant) and “pest”(animal),
which reflect a more anthropocentric point of view than an ecological or biogeographical
approach (Rejmánek 1995).
The following definitions were used for the purpose of the present study. All of them are
definitions that have been agreed upon by the Convention on Biological Diversity (CBD) and
then implemented in the "European Strategy on Invasive Alien Species" that was adopted by
the Standing Committee of the Bern Convention. They are also used in the database “Alien
species in Poland” at www.iop.krakow.pl/ias.
Alien species – a species, subspecies or lower taxon, introduced outside its natural past or
present distribution; includes any part, gametes, seeds, eggs or propagules of such species that
might survive and subsequently reproduce.
6
Invasive alien species (IAS) – an alien species whose introduction and/or spread threatens
biological diversity.
Alien plants – plant taxa whose presence in a given area is due to the intentional or accidental
introduction as a result of human activity. Practically, they are a subset of alien species that
are plants, i.e. plant IAS.
Naturalized plants – alien plants that form sustainable populations year on year without the
influence of humans.
Invasive plants – naturalized plants that produce a large number of offspring that may persist
even at large distances from the parent plant. They can (not necessarily) threaten native
biodiversity and/or human economy and/or human health.
Weeds – plants (not necessarily of an alien origin) that grow on sites where they are not
wanted and that usually have detectable negative economic or environmental effects.
Introduction – the movement by human agency, indirect or direct, of an alien species outside
of its natural range (past or present). This movement can either be within a country or
between countries or areas beyond national jurisdictions.
Exotic range and introduced range – can justifiably be defined as the same. The exotic
range is an area into which the species has spread due to human activity and from which it
would otherwise be restricted from due to geographical barriers.
Native range – A region where a species naturally occurs without direct or indirect human
interference. It is the area in which a species evolved or migrated without human help.
Invasion – The multi-stage process whereby an alien organism negotiates a series of potential
barriers in the naturalization-invasion continuum. It begins with an introduction by humans,
followed by naturalization and further spread. Invasion in this way is the same as range
expansion but only in cases in which it concern alien species and human interference
(Richardson et al. 2011) or chorological expansion sensu Jackowiak (1999). Originally, one
of two variants of “ecological explosion” by Elton (1958), i.e. the introduction and spread in
a new area corresponds to the present definition of invasion.
Taking into account the biogeographical-historical classifications of plants, the situation is
even more complex which is shown in the dictionary of the synanthropization of vegetation
by Sudnik-Wójcikowska and Kożniewska (1988). Many systems for the classification of
synanthropic plants were developed in Europe, especially in the central part. The majority of
them are adaptations of the Thellung system (Kowarik, Pyšek 2012). The terms that are used
in Europe can be incomprehensible to people in other regions of the world. According to
contemporary systems of classification, invasive alien plant species may be a subset of
7
neophytes (=kenophytes in Poland) (Tokarska-Guzik 2005b; Tokarska-Guzik et al. 2012), i.e.
alien species that were introduced after 1500. Those that are able to enter into (semi)natural
plant communities are agriophytes, i.e. hemi- or holoagriophytes, respectively. Species that
colonize man-made habitats and enter into ruderal and segetal plant communities are called
epecophytes. Plants that are escapees from a cultivated area but that are not naturalized
correspond to the term ergasiophygophytes, whereas casual weeds, i.e. species that have
been introduced but that are not naturalized are ephemerophytes. Some authors have stressed
that along the naturalization-invasion continuum plant species can be classified into different
categories (Faliński 1998a).
There is no doubt that some species such as goldenrods Solidago gigantea and S. candensis as
well as knotweeds like Reynoutria (Fallopia) japonica, R. sachalinensis and hybrid R x
bohemica are true invasive taxa. In addition to their rapid range expansion, they have
a negative impact on native (resident) plants in recipient communities. Their presence leads to
a decrease in the species richness of phytocoenoses (e.g., Gerber et al. 2008; de Groot et al.
2011). Customarily or for practical reasons many alien plant species, especially those that
were introduced after 1500, are treated as invasive taxa, although according to the CBD
definition, not every species meets these criteria.
Small balsam Impatiens parviflora DC, which is native to East Asia, is among many
species that have traditionally been regarded as invasive alien plant species. This plant, which
was introduced into Europe in the 1830s, has been the subject of various types of research
before. In 1956 the monographic paper about this species was published in the Journal of
Ecology within the series Flora of the British Isles (Coombe 1956). Later, especially in the
late 1950s and 1960s, small balsam was a model plant for ecophysiological studies (Causton
et al. 1978; Evans, Coombe 1959; Evans, Hughes 1961; Hughes 1959, 1965abc, 1966;
Hughes, Evans 1961, 1962ab, 1963, 1964; Elias, Causton 1975; Young 1981ab; Whitelam,
Johnson 1982). In 1970s a set of papers that were chiefly devoted to the germination and seed
ecology of small balsam were published (Jouret 1974ab, 1976, 1977ab). Finally, in 1984
a monograph about the history of its spread in Europe and a description of some of the
phytogeographic, phytosociological and autecological aspects of the invasion Impatiens
parviflora was published (Trepl 1984). Invasions by Impatiens parviflora in different
geographical regions or countries were well-documented but unfortunately, some of them
were not published in the form of papers despite presenting interesting and novel data about
the ecology of the species, e.g., in Poland (Adamska-Wachowiak 1983; Kujawa-Pawlaczyk
1990; Budziszewska 2006; Dańko 2009; Jończyk 2007; Kozłowska 2013; Kraszewski 2007;
8
Michalik 2005; Piskorz 2004; Komosińska 2008; Lipińska 2011), Slovakia (Maćkova 2012,
Tóthová 2011), the Czech Republic (Urban 2009), Hungary (Csiszar 2004), Belgium (Perez
2006; Vervoort 2011), Austria (Lemmerer 2010), Germany (Hertel 1994; Schaddach 2008).
The problem of invasions was observed by governments, organisations, NGOs etc. Small
balsam was listed and described in some online databases, e.g., CABI (Tanner 2008);
NOBANIS (2013); European and Mediterranean Plant Protection Organisation (EPPO)
(EPPO…2011). The majority of the papers that are concerned with or that mention small
balsam present floristic and phytosociological studies.. Purely autecological papers or works
in terms of community ecology research that focus on I. parviflora as the main object of study
are rarer. One of the first examples that showed its distribution and presence in plant
communities in a large geographical region, Pomerania, was the work by Ćwikliński (1978).
Case studies from the Primeval Białowieża Forest were conducted by Adamska-Wachowiak
(1983), Kujawa-Pawlaczyk (1990,1991) and Adamowski and Keczyński (1998). The process
of the invasion of small balsam in this large forest complex was described in those works.
Synecological studies on permanent study plots in various forest communities in a chosen
forest nature reserve (Warsaw) were performed by Obidziński and Symonides (2000). As
a result of research in the Wielkopolski National Park, a series of autoecological and
biocenotical papers by Klimko and Piskorz 2003; Klimko et al. 2009; Piskorz 2005; Piskorz
and Klimko 2001, 2002, 2006, 2007; and Piskorz and Urbańska 2007 were published. Both an
autoecological and genetic study were conducted by Komosińska et al. (2006). Several
phytosociological surveys have been done in the Silesian Upland and the Jurassic Upland,
which are situated in southern Poland, (Chmura, Orczewska 2004; Chmura, Urbisz 2005;
Chmura, Sierka 2007). Autoecological studies (Chmura et al. 2007) and studies on permanent
study plots (Chmura and Sierka 2006ab; Sierka et al. 2009) have also been done. In addition,
an interdisciplinary autoecological-mycorrhizal study was performed by Chmura and GucwaPrzepióra (2012).
The enormous body of literature on this plant is very often ambiguous on this topic and
often presents conflicting data. This reflects the differences in the specificity of the regions
that are studied, the history of invasions and the methods that are applied rather than the
general patterns of the invasiveness of small balsam. In spite of this some, review publications
on Impatiens parviflora, including one monograph (Coombe 1956; Trepl 1984; Eliáš 1999)
and more recently a monographic paper (Csiszar, Bartha 2008) that were published did not
explore or fully explain all aspects of the ecology of this plant. Hopefully, this work will
contribute to the knowledge about the ecology of small balsam. The main goal of this
9
publication is to summarise and discuss the mechanisms of invasion by Impatiens parviflora
on the basis of the author’s own research and other authors’ earlier studies.
Main aims and hypotheses
I. parviflora can be regarded as a “true” invader species because it was introduced
unintentionally. Since its establishment, the processes of naturalisation and its further spread
have been spontaneous. Small balsam has been successfully established in seminatural and
natural habitats, i.e., forest margins and forests. This plant very frequently found in protected
areas. It occurs in 22 of the 23 national parks and in the majority of landscape parks in Poland
(Najberek, Solarz 2011). Moreover, observed lag of phase during its invasion history makes
the species more interesting. The majority of the research presented in this work was done in
the field in well-preserved or only slightly disturbed forest communities. Most of the study
plots are situated in nature reserves and more rarely in managed forests. The knowledge about
species behaviour in natural and semi-natural communities for the purpose of the present
study is the most important and does matter a great deal from a nature conservation
viewpoint. The observations that are presented and the results of the fieldwork in a spatial and
temporal design are related to other, interesting studies that were conducted in greenhouses,
laboratories and in field experiments.
The main objective of the work is to estimate to what extent the success of an invasion of
I. parviflora is dependent on the invasiveness of the species, i.e., the set of biological
(determined by genetic factors) and/or on ecological characters (determined chiefly by
environmental factors) and on the environment in the sense of the intrinsic factors of the
habitat that make it vulnerable to an invasion, i.e., invasibility. Both the phenomena and the
terms by which they are described are difficult to separate because they partially overlap.
Based on the literature that is available, the research of the author and others and the different
patterns in invasions of I. parviflora that have been observed, the following detailed questions
were posed and particular hypotheses that were related to them were formulated:
Are there any relationships between the edaphic conditions and the abundance of
I. parviflora?
H0: The abundance of I. parviflora increases with an increase in the content of nitrogen,
phosphorus and potassium.
H1: The abundance of I. parviflora does not depend on the edaphic conditions but on the
disturbances of habitats.
10
If the morphological plasticity of I. parviflora is dependent on the soil parameters, then
how are variations in selected plant traits affected by the type of forest community?
H0: The morphometric traits, phenology and germination ability vary significantly among
forest communities under the tree canopy in forest interiors.
H1: The morphometric traits, phenology and germination ability differ significantly only from
individuals growing on forest paths/forest margins.
Was the synergic effect of light availability and the high content of nutrients studied in
lab conditions on the morphological plasticity of I. parviflora that is known from
literature (Peace, Grubb 1982) observed in field conditions? Do the canopy openings and
the nutrients of decaying dead wood enhance plant fecundity? Does a natural disturbance
favor some phenotypes and lead to persistence in the forest interior?
H0: The size of individuals is biggest on sites near decaying dead logs where the soil is
affected by the decomposition of dead wood and under canopy openings.
H1: Only light availability has an impact on the size of individuals of I. parviflora.
How does the invasiveness of the species change among various microhabitats? Which
microhabitats are occupied by individuals that are characterised by a large size and
a higher number of flowers and fruits?
H0: Disturbed habitats (forest paths and margins) are characterised by a massive presence of
large and fecund individuals of I. parviflora. Other microhabitats do not play an important
role in the spread and persistence of the species.
H1: In addition to forest paths/forest margins and the soil under the tree canopy, there are
other microhabitats (associated with dead wood complex) that are frequently colonised by
I. parviflora where it is able to grow and finally set seeds.
What is the temporal variation of the occurrence of I. parviflora? Under which conditions
does the abundance of the species change? Which natural and anthropogenic factors
determine an increase in abundance? As an annual does small balsam establish the same
places using old and new gaps in the forest layer? To what extent is this process
determined by its earlier presence or is a random?
H0: Small balsam is invasive species; thus it increases in abundance over time.
H1: The dynamics of small balsam are random or depend on many factors that are hard to
predict.
What is the role of arbuscular mycorrhiza (AM) in I. parviflora? What abiotic factors
have an impact on it? How does AM influence the size and fecundity of the species?
11
H0: The abiotic factors that occur in particular plant communities are manifested in the
presence/absence of AM in I. parviflora. The size of plants and their fecundity are first of
all correlated with the frequency of mycorrhiza followed by other arbuscular mycorrhiza
indices.
H1: Abiotic factors mainly have an impact on the values of the AM colonisation indices.
Forest plant communities do not differ in the presence of AM+ plants. The size of plants
and the fecundity of I. parviflora depend on various traits of mycorrhiza colonisation.
What is the role of functional plant groups in relations with I. parviflora? Which plant
functional group is displaced or affects the presence/abundance of small balsam?
H0: There are some plant functional groups whose representatives have a higher cover in
uninvaded sites when compared to sites that have been invaded by I. parviflora. The
species of some plant functional groups are able to displace I. parviflora.
H1: The contribution of particular plant functional groups does not matter in the
absence/presence of I. parviflora.
Does I. parviflora colonise “empty sites” sensu Harper (Tilman 1997) in forest
communities? Is the naturalisation of I. parviflora an effect of the displacement of native
species due to its competitive ability? Does the cover of native and resident species have
an influence on the abundance of I. parviflora? If yes, how does this impact change
among various forest types?
H0: There are some species that are affected by I. parviflora and they can be hindered or
displaced by the species. Then, the species meets criteria of invasiveness according to
definition by Convention on Biological Diversity.
H1: No, I. parviflora prefers sites where there is bare soil or a lower species richness and cover
of native plants. In species rich communities, it is displaced by them.
12
1. The object of the study: Impatiens parviflora
1.1. Characteristics of the species
1.1.1. Taxonomy
I. parviflora DC, was described by De Candolle in 1824 (Coombe 1956). The synonym
name is Impatiens nevskii Pobed. It belongs to the family Balsaminaceae. According to
various sources, the genus Impatiens is estimated to contain from about 850 (Csiszar, Bartha
2008) to more than 1000 species (Janssens et al. 2006). As a result of molecular phylogenetic
studies, Balsaminaceae was reclassified as a family in the Ericales (an order of 26 families)
sitting as a sister group to all of the other Ericales in the Balsaminoid Ericales. The
Balsaminoid
Ericales
consists
of
the
families
Balsaminaceae,
Marcgraviaceae,
Pellicieraceae and Tetrameristaceae. Together this group comprises approximately 1130
species (Tanner 2008). Based on molecular evidence, it can be concluded that Impatiens
parviflora evolved ca. one million years ago – in the mid Pleistocene and its most nearly
related congener species is I. balfourii. A probable ancestor of small balsam originated from
southwestern China (Janssens et al. 2009). An older taxonomical classification indicated that
I. parviflora belonged to the Brachycentron section and Micropetalae series inside the
Cauliimpatiens subgenus (Jørgensen 1927 after Csiszar, Bartha 2008). Nowadays,
I. parviflora is a representative of one of the 15 clades within the genus Impatiens together
with, among others, I. glandulifera and I noli-tangere (Janssens et al. 2006). Below is the
taxonomic classification of small balsam (Tanner 2008).
Domain: Eukaryota
Kingdom: Plantae
Phylum: Spermatophyta
Subphylum: Angiospermae
Class: Dicotyledonae
Order: Balsaminales
Family: Balsaminaceae
Genus: Impatiens
Species: Impatiens parviflora
1.1.2. Morphology
Small balsam is a glabrous annual herb that is usually 20-60 cm tall, but which can also
be as short as 4 cm (Chmura 2008a). Rarely it reaches 150 cm (Coombe 1956) and even taller
– 152 cm (Chmura 2008a), 165 cm (Chmura npbl) and ca. 170 cm (Adamowski personal
com). The tallest recorded plant was 205 cm (Dańko 2009). The root system is shallow with
a short-lived primary root that is supplemented by growing stronger lateral and adventitious
13
roots from the lower node (Hegi 1965; Coombe 1956; Sebald et al. 1998; Csiszar, Bartha
2008). The adventitious roots sometimes develop on the aboveground part of plants. The plant
is single stemmed or branched from the lower nodes with third-order branches in welldeveloped plants that grow under favourable conditions in terms of habitat fertility and
a medium density of population (Coombe 1956; Chmura npbl). The leaves are simple,
alternate, ovate, elliptic or wide-lanceolate and are usually 5-12 cm long and 2.5-5 cm wide
and sharply serrated at the edges with 13-35 teeth on each side (Tanner 2008). Petioles carry
stalked glands that may serve as extrafloral nectaries. The length of the petioles gradually
decreases with the height of the stem. In the lowest part, they can be as long as four cm but on
the upper part they are only about 0.5 cm. Zygomorphic flowers are located in inflorescences,
which are axillary racemes. An inflorescence bears from (1-) 4-10 to (-15) flowers. Racemes
are as long as the upper leaves or longer. The flowers are 10-15 mm long including their spur
and upright. The corolla is pale yellow with red spots on the inside but white flowers with
yellow patterns have also been described. There are five sepals, the two frontal ones are
reduced, the two lateral ones are rudimentary green and about three mm long (Csiszar, Bartha
2008). The back sepal forms a straight spur that is 5-10 mm in length and has the same colour
as the petals. The lateral and back petals of corolla are in pairs; the frontal petal is large (up to
10 mm long). Five stamens stand alternately with the petals inside the flower; the filaments
grow freely and the anthers stand together. Five carpels form the pistil and the ovary is
superior. The fruit is a capsule that is linear to club-shaped, 15-25 mm long, glabrous and
green. The brown coloured seeds are 3-5 mm long with one to five fine longitudinal striations
per capsule (Csiszar, Bartha 2008; Tanner 2008). Records for the I. parviflora seed mass vary
from 6.91 to 9 g per thousand seeds according to the database of Royal Botanic Gardens, Kew
(RBG 2008) but according to Moravcová et al. (2010), the mass of 25 propagules (g), which
are probably seeds, is 0.1422±0.0135. The mean length of a propagule (seed) is 4.01±0.34
mm and the width is estimated at 1.97±0.29 mm.
1.1.3. Biology
Variablity and molecular biology
It is commonly assumed that there is little genetic variation in the invasive populations
as compared to the native ones. Two forms are known in Europe – varietas albiflora with
white flowers, yellow spots inside and the forma albescens that has white flowers with orange
spots in the throat of the flower (Jørgensen 1927). Some initial reports about the genetic
variability of the species in Europe that were performed in Poland (Komosińska et al. 2006)
14
and in Lithuania (Janulionienė et al. 2011; Kupcinskiene et al. 2013b) shed some light on this
problem. In the latter study, it was revealed that those genetic distances among populations
are correlated with geographical distances. In the former, smaller study conducted in Poland,
no genetic differences between the two populations were detected; however, a common
gardens experiment suggests little genetic differences. Kupcinskiene et al. (2013b)
demonstrated that multiple introductions occurred in Lithuania, and therefore it can be
assumed that a similar phenomenon took place in other parts of Europe. The chromosome
numbers that were recorded are 2n=20, 2n=24 and 2n=26. No hybrids are known in Europe
(Coombe 1956); however, it can coexist with I. noli-tangere, which is native to Europe and
more frequently with another alien species, I. glandulifera, in an adventive range.
Phenology and physiology
I. parviflora plants in Europe usually germinate in late March or April. The time from
germination to flowering is eight to nine weeks with the seeds ripening three to four weeks
later (Coombe 1956). Flowering usually begins in May or June and lasts until late September
or mid-October with the oldest recorded plants being seven months old. The phenology of the
plants differs depending on the types of habitats in terms of aspect, light availability and
disturbance (Chmura 2008ab; Piskorz, Klimko 2002; Piskorz 2005). The species is
autoperiodic, i.e., it can grow from seed to fruit at any time of the year in a greenhouse under
suitable conditions of seed dormancy and germination (Coombe 1956). A series of field and
laboratory experiments (Evans, Hughes 1961; Hughes 1959, 1965abc, 1966; Hughes, Evans
1961, 1962ab, 1963, 1964) revealed some patterns of the plant’s physiological response to
environmental factors with regard to its seasonal cycle and daily balance. It was confirmed
that the minimum period of dormancy after ripening is 90 days in wet conditions and 5° C.
During growth, the hypocotyl increases with a decreasing light intensity and a higher
temperature in red light. A low temperature and blue light as well as a greater vapour pressure
deficit and a higher nutrient status are typical for shorter hypocotyls. Under natural conditions
in the field, such a situation enhances the development of adventitious roots in cases in which
the hypocotyls are too long and too weak to support the growth of plant. The distribution of
the total dry weight and biomass allocation of small balsam is affected by many factors
(Hughes 1965; Eliaś 1992). The unit leaf rate is proportional to daylength at a high light
intensity. The quantum efficiency for apparent photosynthesis is almost identical in blue and
red light and it appears to be optimum at 15° C. The unit leaf area decreases with increasing
plant size. A high nutrient status reduces the specific leaf area, whereas blue light and an
15
increasing plant size reduce both the specific leaf area and leaf weight ratio. A low
temperature reduces the specific leaf area but does not influence the leaf weight ratio, which
matters in early spring and limits the growth of plants in field conditions. The highest rate of
increase in leaf area takes place in spring and early summer at 10% of the daylight factor
(Hughes 1965). Whitelam and Johnsson (1982) observed a response of nitrate reductase
activity to simulated canopy radiation of a low photon influence, which characterises the light
conditions in shade. Small balsam showed a marked and rapid response to changes in the
photochrome photoequlibrium with low-fluence rate sources. The authors explain the ability
of Impatiens parviflora to display strong responses to added red light by its ability to tolerate
much reduced fluence rates of PAR (photosynthetically active radiation), which is often found
in natural habitats. Thus, this clarifies why small balsam can grow both in very deep shade
and more open habitats. Ugoletti et al. (2011) studied 18 ecophysiological traits in three alien
Impatiens species including I. parviflora. These were, among others, traits associated with
plant growth and allometry, reproductive capacity and leaf physiology. In the beginning the
plants were reared in pods in a greenhouse and later were moved outside to an open area. The
maximum growth in these plants was observed at the end of August and amounted to an
average ± SE of 1.48 ± 0.25cm per a day. The leaf weight ratio (LWR) was 0.299±0.047
(mean ± SE), leaf area (LAR) was 0.0065± 0.0004 (mean ± SE), stem weight ratio (SWR)
was ca. 0.6 of g stem g-1 plant and specific leaf area (SLA) was 0.025 m2g-1 leaf. The mean
percentage of water in fresh mass, which reaches as much as 92%, is one of the highest
amongst herbaceous plants (Drobnik 2007). Small balsam is listed as an ozone-sensitive plant
and is often the subject of studies that are focused on the impact of ozone on the condition of
plants (Godzik 1996; Bergmann et al. 1999; Skelly et al. 1999; Manning et al. 2002;
Manning, Godzik 2004).
Reproductive biology
Small balsam is an annual plant, i.e., a therophyte. Propagation is entirely by seed.
Capsules can contain from one to five seeds. The flowers are protandrous with a male phase
of two to four hours and a female phase of one to two days (Tanner 2008). Flowers can be
cleistogamous although chasmogamous ones are much more common. Flowers are visited
mainly by insects from the Syrphidae family of which 19 species were found on I. parviflora
(Schmitz 1998). All flowers usually set seeds independent of the intensity of insect visitation.
The plant is self-compatible, geitonogamous and allogamous pollination results in no
differences in a set of seeds. The most fecund individuals can produce 1000-2000 per plant
16
(Trepl 1984); however, Coombe (1956) estimated a maximum amount of 10,000 seeds. In an
alder forest a single plant had 90 seeds on average while a single plant had only 10-30 seeds
in hornbeam-oak forests (Trepl 1984). According to Moravcová et al. (2010), the average
number of propagules is 279 per single plant and 2689 per 1m2 (in the maximum population
density found in a locality). Many authors have conducted germination experiments and have
obtained different results because of various storage conditions. Seeds require low
temperatures to break dormancy but temperatures below -20° C kill all seeds (Coombe 1956).
Kinzel (1927) found that at room temperature none of seeds germinate but with seeds stored
at -5° C, the germination rate was 100%. A surprising result was obtained by Coombe (1956)
who after three days of an experiment (+5C, wet) noted 100% germination. According to
Jouret (1974ab) temperatures between 0° C and 5° C are ideal for germination. The shortest
stratification period that results in germination is 13 days. The duration of the stratification
has an influence on the increase in the germination rate. Seeds remain viable for less than
three years when they are stored dry at room temperature but they still germinated after four
years when kept in wet conditions (Coombe 1956). Burying seeds delays the time of
germination slightly (Schaddach 2008). Opinions about seed banks of I. parviflora are
contradictory. Csontos (1986) in a removal experiment suggested the existence of seed banks.
Two years after seeds were put into the soil 14% of the previous population regenerated.
During the time there was no seed output from neighbourhood. According to Perglová et al.
(2009), only alien I. capensis and native I. noli-tangere form short-term persistent seed banks.
The native balsam was known from the existence of seed banks from many earlier studies
(e.g., Pirożnikow 1983; Falińska 1990).
1.1.4. Habitats
I. parviflora is a species that grows on a wide range of mineral soils with a pH range in
the rhizosphere from 4.5 to 7.6 in UK (Coombe 1956). The species has been found in heavily
acidic sites (pH= 4.12) (Csiszar, Bartha 2008) in Hungary and even those with a little less
(pH<4.1) in Poland (Chmura et al. 2007). Most soils are brown soils or rendzinas and the
plant avoids waterlogged conditions (Coombe 1956). The species has also been encountered
in alkaline soil in calcareous grassland (Chmura 2008a). Small balsam showed a positive
response measured as an increase of the percent of cover to the compactness of soil
(Godefroid, Koedam 2004a). It is a temperate species that prefers shade and half-shade and is
mostly found in 5-40% relative daylight (Tanner 2008). According to Borhidi (1999), while
17
small balsam is plant of shade or half-shade, it can also survive on open sites. Survival in
open habitats is influenced by the moisture of the stratum (Coombe 1956).
It mainly penetrates forests that are under a strong human influence, such as managed forests
and timber plantations, as well as natural old-growth forest types. It is found more often in
moist to wet forests from floodplains, alder forests mad beech forests rather than in mixed or
coniferous forests. In addition, the species occurs in the ruderal vegetation in settlements,
mainly on roadsides and in gardens. Human disturbance in forest habitats generally promotes
the spread of I. parviflora (Trepl 1984; Eliáš 1999; Csiszar, Bartha 2008); however, fire has
a negative impact. Comparison of the flora from burnt and unburnt sites indicates that the
species avoids former burnt sites (Maringer et al. 2012).
Taking into account the number of types of habitats small balsam is to be found in a wide
range of biotopes. According to the European Nature Information System (EUNIS)
classification I. parviflora grows in 68% of 33 main types of habitats (Chytrý et al. 2008),
whereas in more detailed classification which corresponds to phytosociological alliances or
groups of alliances amounting in a total of 68 types (Sádlo et al. 2007) I. parviflora was
recorded in 45 types (Pyšek et al. 2012). On the basis of comparison with other neophytes and
invasive species it turned out that small balsam is one of the species with the widest
ecological amplitude.
1.1.5. Biotic interactions
Associated plant species
In Trepl’s opinion (1984) I. parviflora occurs frequently in seven phytosociological
classes and 20 alliances in Central Europe; however, the range of vegetation units where it is
present is wider. These are mostly deciduous forests (Querco-Fagetea class), i.e., oakhornbeam forests (Carpinion betuli) consisting in Quercus spp, Carpinus betulus, Tilia
cordata, Acer pseudoplatanus, Acer platanoides, beech woods (Fagion sylvaticae), floodplain
forests (Alnenion glutinosae-incanae, Ulmenion minoris) and alder and willow carrs (Alnetea
glutinosae class) with Alnus glutinosa. It also occurs in mixed and coniferous forests
(Vaccinio-Piceetea class) and plantations under Pinus sylvestris, Picea abies, Abies alba,
Larix decidua etc. The species can also be found in acidophilus oak forests (Quercetea
robori-petraeae) and can grow in scrub communities (Rhamno-Prunetea). It usually occupies
disturbed sites in forests, e.g., forest paths and forest edges. In forest interiors it can grow on
sites that are not favourable for other herbaceous plants because of low light levels, heavy
competition by tree roots or thick litter layers. In nitrophilous forest edges and eutrophicated
18
forests, it is associated with Geranium robertianum, Geum urbanum, Chaerophyllum
temulum, Alliaria petiolata, Chamaenerion angustifolium, which are mainly representatives
of the Epilobietea angustifolii class (Trepl 1984; Schmitz 1998; Kowarik 2003; Tanner 2008).
There are reports about massive or quite abundant occurrences of small balsam in other types
of vegetation, e.g., chasmophytic vegetation, i.e., vegetation of rocks – the Asplenietea
Trichomanis class (Anioł-Kwiatkowska, Świerkosz 1992; Świerkosz et al. 2011) and postbog meadows on habitats of fens and transitional mires (the Scheuchzerio-Caricetea fuscae
class) (Podlaska 2010). Small balsam occurs very rarely in natural patches of fens and bogs
and grows only in drainage and disturbed parts of these types of vegetation (Sadowska 2011).
It was reported that small balsam can occur in arable fields including crop fields (wheat, rye,
sugar beets, oilseed rape and maize), but the plants were chiefly found in the shaded edges of
fields along the shrub and tree zones (Dajdok, Wuczyński 2008).
The anthropogenic, ruderals, habitats that are occupied by the species include parks,
cemeteries, garden allotments, wastelands, cottage yards, refuse tips and railway tracks and
embankments (Tokarska-Guzik 2005a, Ciosek, Bzdon 2003; Faliński 1966; Klimko, Bozio
2003; Sowa, Warcholińska 1980; Szmajda 1974; Załuski 1974; Celka, Żywica 1994;
Warcholińska 2005; Rostański 2006). Small balsam is a characteristic species for the
Alliarion alliance (Matuszkiewicz 2011) and sometimes the plant association Impatientum
parviflorae, which occurs in ruderal habitats, is distinguished (Brzeg, Wojterska 2001; Wika
et al. 2002).
Associated fungi species
No mycorrhiza were known until the beginning of the 1950s (Coombe 1956) and this
information was copied and cited in many further publications (e.g., Tanner 2008). Probably
Truszkowska (1953) was the first who provided information about arbuscular mycorrhiza in
I. parviflora among other species in floodplain forests. Others who mentioned mycorrhizal
fungi in small balsam were Peace and Grubb (1982) who were later cited by (Harley, Harley
1987; Wang, Qiu 2006). In 2009 species gained the definite status of a mycorrhizal species
(Štajerová et al. 2009).
Five phytopathogenous fungi were found on I. parviflora: (Shaerotheca balsaminae, order
Erysiphales) Aschyta impatiensis, Phyllosticta impatientis from Sphaeropsidales and two rust
fungi (Puccinia argentea, P. komarovii) (Schmitz 1998; Csiszar, Bartha 2008). The rust
Puccinia komarovii is one the greatest enemies of small balsam, which can cause mortality
that can even reach 100% (Eliáš 1995; Bacigálová et al. 1999). This species was brought from
19
Central Asia, which is the native region of I. parviflora. Its westward spread has been
observed since 1921 when it was first found in the Ukraine (Kiev), in Germany in 1935,
Switzerland in 1938, Slovakia in 1942 and still continues ever westward. The first record of
this fungus in Poland was in 1934 (Trepl 1984; Majewski 1979). Piskorz and Klimko (2006)
studied the impact of this rust on the population of small balsam in detail and demonstrated
that infected populations differ from healthy ones in the seasonal dynamics of changes in
abundance; high mortality is observed as early as May and the first part of June; the infection
and its intensity stimulates the growth of the stem and the hypocotyls and fresh weight of
infected plants is about 30% less than the weight of uninfected plants but the weight of the
generative organs decreases significantly by more than 50% in some cases. The rust infection
also affects reproduction efforts, which is expressed by the ratio of the weight of the fruits (or
only seeds) to the total biomass, which shows a distinct downward trend. Other pathogenic
fungi that can be encountered on the seeds of small balsam are Rhizopus stolonifer, Rhizopus
oryzae and Absidia glauca; however, they occur more frequently on native I. noli-tangere
(Budziszewska 2006).
Pathogenous arthropods and flower visitors
Schmitz (1998) mentioned 13 phytophagous species that consume small balsam.
Among them there are 12 insect taxa. Eight are polyphagous taxa: Sminthuridae sp of
Colembola order, Anthropophora alni, Philaneus spumarius, Aphis fabae cirsiiacanthoidis,
Aphis nasturtii of Homoptera. Lygus spp (e.g., Lygus pabulinus) from Heteroptera and two
species of Lepidoptera: Xanthorhoe birivata and Pergesa elpenor. There are also two
monophagous species of the 2nd order – aphid Impatientinum asiaticum and from the Diptera
order – Phytoliriomyza melampyga. The former species is native to Central Asia and followed
its host plant but was introduced later; probably more than 100 years later (Schmitz 1998;
Eliašova 2011). It can also be found on Impatiens glandulifera; however, it has not been
noted on Impatiens noli-tangere, which is native to Europe yet. It cannot be excluded that the
distribution will continue to the native congener (Ripka, Csiszar 2008). The leaf-mining fly
P. melampyga (=Liriomyza impatiensis – previous synonym) was already described by Vogel
(1943) after Csiszar, Bartha (2008) and Coombe (1956). Aphids are the dominant group. As
Csiszar and Bartha (2008) reported, ca. 94% of all individuals found belonged to this group of
insects. Aphids were present on the 74% of all of the investigated plants.
Hoverflies (Syrphidae family) are the most important group from among flower visitors. In
total 19 species and their larvae were recorded on small balsam. They play the role of
20
pollinating species because hoverflies collect nectar and pollen from the flowers (Csiszar,
Bartha 2008). Perhaps one of the reasons that hoverflies are attracted to the flowers of
I. parviflora is their colour, which is especially favoured by this group of insects (Lunau,
Maier 1995; Arnold et al. 2009). Schmitz (1998) lists 14 species as among those that are true
pollinators, e.g., Melanostoma mellinum, Scaeva pyrastri, Sphengia clumpe. Rich
aphidophagous fauna is associated with the aphids that are present on I. parviflora. Four
species of hoverflies are both pollinators and aphidophagous in the larva that feed on
Impatientum asiaticum, e.g., Bacca elongate. The remaining three hoverflies species are
exclusively aphidophagous insects. They are usually attracted by dense colonies of
Impatientinum asiaticum. Other representatives of this group are species from Heteroptera
(three species), Coccinellidae (four species), Neuropteroidea (five species), Sphecidae (two
species) and two other parasitoid aphid species. When compared to I. noli-tangere and
I. glandulifera, the most aphidophagous species are on small balsam, while in turn, the lowest
number of phytophagous species was found on them but they were more abundant. Small
balsam does not have well-developed extrafloral nectarines like I. glandulifera and I. nolitangere, and therefore the species richness of extrafloral visitors is the lowest (Csiszar, Bartha
2008). One of the reasons for the relatively small number of pollinators is its low nectar
secretion compared to the two congeners mentioned above (Vervoort et al. 2011). Najberek et
al. (in preparation) confirmed that among arthropods aphids are the most important group
followed by representatives of Diptera, Hymenoptera, Coleoptera. The representatives of the
Hymenoptera order, i.e., Bombus spp., which fly from flower to flower, were observed in both
Poland and Hungary (Csontos, Chmura unpublished). Representatives of Apidae (Bombus
spp, Apis melilifera) can also be found on two other congeners – I. glandulifera and
I. noli-tangere (Vervoort et al. 2011). Small balsam plants are also visited by ants
(Formicidae) and spiders. The first group is attracted by the aphids, which provide honey-dew
and the latter group are predators (representatives of four families – Linyphiidae, Aegelnidae,
Thomisidae and Clubionidae) (Csiszar, Bartha 2008). The species composition of
arachnofauna does not differ from the fauna of spiders that visit other herbaceous plants in the
neighbourhood. Two species of Acari, Anystidae sp and Trombidiidae, are parasites of aphids
(Schmitz 1998).
Other invertebrates
Schmitz (1998) listed one snail species, the grove snail (Cepaea sp) of Helicidae family,
which feeds on the leaves of small balsam. It is native to Western and Central Europe but is
21
invasive in North America. Piskorz and Urbańska (2007) observed Columella edentula on
leaves in an oak-hornbeam forest. It was revealed that I. parviflora was used as food for the
snail and as a protection from predators and direct sunlight. The leaves of small balsam under
which the snails occurred provided stable humidity as well. What is important is that
a comparison of seasonal variations and abundances with other species that were present in
the herb layer showed that I. parviflora was most frequently colonised by C. edentula.
Vertebrates
It is rarely grazed by herbivorous mammals. Only roe deer, Capreolus capreolus, was
reported to feed on the shoots of this plant; animals other than the deer avoid this species
when compared to I. noli-tangere (Schmitz 1998). Rabbits, Oryctolagus cuniculus and other
forest mammals such as rodents do not feed on this plant (Coombe 1956).
Allelopathy
Several studies have showed that small balsam is an allelopathic plant (Vrchotová et al.
2009; Csiszar, Bartha 2008; Csiszar et al. 2012). In these studies other methods were used but
they revealed that under laboratory conditions I. parviflora had intermediate inhibitory effects
after I. glandulifera. Vrchotová et al. (2009) tested the effect of water, methanol and
dichloromethane extracts from the leaves of three Impatiens (I. noli-tangere, I. parviflora,
I. glandulifera) on the germination of the seeds of two model plants – Leucosinapis alba and
Brassica napus. All of the tested extracts had inhibitory effects on the seeds of all of the
plants studied (except for the dichloromethane extracts).
The extracts from I. parviflora inhibited both the percentage of germination and the lengths of
the radicle and hypocotyl in the germinated seeds of the plants studied. Csiszar and Bartha
(2008) proved that mustard seeds (Sinapis alba) that are treated with a 5 g/100 ml
concentration extract differed considerably from the control (only 86.66% of the seeds
germinated). In another laboratory study, Csiszar et al. (2012) used the juglone index, which
is based on comparing the effects of treatment with 1 mM juglone and a substance that is
extracted from a plant species with an unknown allelopathic potential. The juglone index is
the quotient that is created by the germination rate, shoot length and root length of white
mustard (Sinapis alba L.) that has been treated with juglone and a substance from
a potentially allelopathic plant. According to that study, small balsam at both lower and
higher concentrations, i.e., 1- and 5 g plant material/100 ml distilled water revealed an
inhibitory effect. The index scored 1.05 and 1.17, respectively. Among 14 herbaceous and 20
22
woody species, the juglone index for a lower concentration of I. parviflora was second after
Amorpha fruticosa (1.11) and the seventh one (Phytolacca esculenta – the highest 5.49) for
a higher concentration. I. parviflora has a repellent and toxic effect on some insects (Pavela et
al. 2009). During an experiment with three Impatiens species (including I. noli-tangere and
I. glandulifera as well) and their influence on Myzus persicae (aphid, Homoptera) was
demonstrated. M. persicae (green peach aphid) is an important pest for many plants. After 54
h of exposure, the most active extract was Impatiens parviflora with a 99.7 and 90.0 %
mortality at concentrations of 0.5 and 0.1 %, respectively and with high percentage of
repellency (90-100%) at different times (from five to 48 h). The extract of I. parviflora
contained tryptophan (2.13 mg/g); 2-methoxy-1,4-naphthoquinone (0.02 mg/g), total flavone
(7.64 mg/g) and total derivatives of caffeic acid (15.60 mg/g). The authors believe that the
plant can be used as an insecticide.
1.2. History of invasion
1.2.1. History of spread in alien range
Native range
The native range of I. parviflora encompasses the mountains of central Asia. According
to Trepl (1984), there are parts of the range that consist of scattered areas in which the species
is interspersed with areas without it in countries such as Turkmenistan, Afghanistan,
Turkmenistan and Mongolia (Trepl 1984). The USDA-ARS (2008) database indicates part of
central Asia, including Kazakhstan, Kyrgystan and Uzbekistan, also Xinjiang, China as its
native range. Other parts of Asia, i.e., the Russian Far East are treated as a naturalised range.
I. parviflora occurs along rivers and streams, in shady and humid localities in parts of the
former USSR (Komarov 1934-1964 after Obidziński, Symonides 2000). In areas with steppe
or semi-desert vegetation, the species can only occur in more humid forest patches, e.g., in
floodplains or on northern slopes (Tanner 2008).
Spread in an alien range
Apart from Europe the alien range also includes East Asia, Western and Northeastern
America (Adamowski 2008). The history of its introduction into Europe was described in
detail by Trepl (1984), who gave the approximate dates of its introduction for all of the
countries where small balsam is now present. The putative first record of the introduction of
I. parviflora is often estimated to be around 1830 or shortly before that date (Trepl 1984;
23
Galera, Sudnik-Wójcikowska 2010) or in 1831 (Coombe 1956; Trepl 1984). Its seeds were
then introduced into the botanical garden in Geneva (Switzerland) on purpose for cultivation.
The first naturalised stand (in the wild) in the neighbourhood of the above-mentioned
botanical garden was reported in 1831 (Coombe 1956); however, according to De Candolle
(after Höck 1900 after Trepl 1984), the first date of a naturalised record was 1837. In turn,
Probst (1949) after (Trepl 1984) indicated Solothurn, which is also in Switzerland, as the first
naturalised locality. The online atlas of British & Irish Flora indicates 1823 as the date of
introduction (BRC 2008). This is probably an error. Galera and Sudnik-Wójcikowska (2010)
analysed the migration history of the species whose introduction was associated with
botanical gardens at the beginning of the early stages of invasion. Based on their studies, it
was noted that I. parviflora had 46 historical records from 30 European botanical gardens as
an escapee. The next known naturalised stand dates back to 1842 in Dresden five years after
its introduction into the town’s botanical garden in 1837 (Trepl 1984). It appeared in two
places in Poland around 1850 – botanical garden in Kwidzyn (German: Marienwerder) and
Dłużyna (Marianaue) near Gryfino (Klinggraeff 1880 after Trepl 1984) and in the vicinity of
Kraków (Berdau 1859). It was observed for the first time in Prague, the Czech Republic in
1872, the same year as it was observed in France. The other dates of its introduction are as
follows: Belgium (1868); Austria (1870), the Netherlands (1885), Sweden (1870) and the
Ukraine (1868) (Hegi 1965; Trepl 1984; Hultén, Fries 1986). It is not clear when the species
arrived into the territory of Great Britain as two dates are given – 1848 and 1851 (Coombe
1956; Williamson 1996). The remaining part of the invasive range covers most of central
Europe, France and the UK, with scattered localities in Scandinavia, the Baltic states of
Latvia (1904) and Lithuania (1934) (Gudžinskas 1998; Hylander 1971; Hultén, Fries 1986;
Verloove 2006) and in North America (Barkworth 1973). In North America the species is
naturalised in northeast Canada, i.e., Quebec, New Brunswick, Nova Scotia and Prince
Edward Island. At present it is observed in USA (EPPO 2011; Tanner 2008) but until recently
it was not reported (Tabak, Wettberg 2008). As was mentioned earlier, the first habitats were
botanical gardens and their close vicinity. In the next stages of invasion, the habitats that were
invaded were predominantly gardens, parks and other sites in settlements (Trepl 1984; Tanner
2008). By the second half of the 19th century the species had been naturalised mostly in the
deciduous forests of the northern and central part of Europe. Small balsam invaded forest
interiors and their edges in forests; the former were usually disturbed sites at the beginning of
invasion. Later in the 1900s, it became more and more able to penetrate into less disturbed
and more natural forest habitats (Trepl 1984).
24
Dynamics and means of spread
Since its first introduction, the pace of the spread of I. parviflora has become much
faster. For instance, its massive expansion in Poland started in the 1960s and now the species
is very common on the territory of the country (Tokarska-Guzik 2005a). Small balsam is the
third most common neophyte that is naturalised in the semi-natural vegetation in Germany
(Kühn et al. 2004). The rate of increase of its range in Europe is unknown except for a few
countries, e.g., the maximum rate of spread in the UK was calculated as 24 km per year in
1915 (Williamson 1996). Many authors stress that at the beginning of invasion, I. parviflora
occupied only human-made or disturbed habitats such as botanical gardens, parks, cemeteries
and roadsides (Trepl 1984; Csiszar, Bartha 2008; Obidziński, Symonides 2000). It appears
that the species underwent the process of a lag-phase, which resulted in its faster spread and
its entering into natural communities in the second half of the 20th century (Trepl 1984;
Obidziński, Symonides 2000). During this time, the species changed its status from an
epecophyte to (holo)agriophyte according to classification of synanthropic plants. It is not
certain precisely when this happened. On the one hand, an extensive phytosociological survey
in Westfalia (Wittig 1977 after Csiszar, Bartha 2008) showed that presence of I. parviflora
was recorded in four of 800 relevés. On the other hand, a year later Ćwikliński (1978)
presented the contribution of small balsam in natural forest communities including, among
others, a floodplain forest, Circaeo-Alnetum; an oak hornbeam forest, Stellario-Carpinetum
and a beechwood forest, Melico-Fagetum, Fago-Quercetum.
Autochorous dispersal mechanisms are not enough for the quick spread because seeds can
only reach distances of up to 3.4 m (Trepl 1984). Kamieński (1884) after Tokarska-Guzik
(2005a) claimed that the species was introduced accidentally by travelers who had arrived
from Western Europe mainly by the sea. The transport of floating seeds by water is possible
but probably of limited importance, although transport in river sediments with fast-moving
waters in winter floods may contribute to long-distance dispersal (Tanner 2008). The majority
of researchers agree that humans aided in the transport of the seeds of I. parviflora. Long
distance forest management and the transport of timber played the most significant role. The
occurrence of I. parviflora in various timber-yards in local parks and forests can be attributed
to this pathway of migration. The frequent occurrence of I. parviflora along railways and
tracks where trains that transported timber also supports this theory (Trepl 1984).
Single reports indicated that seeds might have been introduced along with buckwheat for
peasants (Coombe 1956) or with soil and in the roots of garden plants, with flower seeds and
25
with compost as well. Epizoochorous dispersal in the fur of mammals and in the dirt on their
feet was also reported as an important mode of long-distance dispersal (Trepl 1984). Graae
(2002), in her experiment with domestic dog in a forest, found that despite the smooth surface
of small balsam’s seeds, dispersal on the backs of animals is possible. It was proved that roe
deer and wild boar can transport the seeds of the plant (Heinken, Raudnitschka 2002).
Another possibility for spread is the creation of microhabitats that are suitable for the
germination of seeds. For instance, Piskorz, Klimko (2001) reported that soil that is rooted by
wild boars (Sus scrofa) is also intensively colonised by I. parviflora, especially in strongly
shaded sites that were previously covered with a thick layer of leaf litter. However, these are
short-term changes that are closely related to repeated disturbances by animals. As Obidziński
and Głogowski (2005) and Obidziński, Kiełtyk (2006) proved, zoopression, which is best
exemplified by the red fox (Vulpes vulpes) and the Eurasian badger (Meles meles), may
promote the spread of some species including therophytes. I. parviflora did not occur in the
regions and forests that they studied – the Rogów District in central Poland and the center of
Bialowieza Primeval Forest, respectively. However, it can be assumed that these animals
could also facilitate the spread of small balsam.
Birds also transport seeds, which is manifested by the occurrence of I. parviflora in hollows
or in crevices in the bark of living trees. Trepl (1984) reported that the dirt on the vehicles that
a reused by foresters may contain up to 22 seeds per one litre of soil that is trapped in the
tyres and other parts of a vehicle. Although some sources claim that the entire European
population is derived from a single introduction in 1830 (Coombe 1956) such a fast spread
contradicts this theory. According to Trepl (1984), individuals were brought into gardens out
of botanical curiosity and were also sown into near-natural vegetation with the aim of
“enriching” the natural flora (Trepl 1984, Tanner 2008). This method of propagation both in
case of Impatiens parviflora and other plants is no longer used.
1.2.2. Impact on native biodiversity and the nature conservancy viewpoint
Opinions about the possible impact of I. parviflora on native and resident herbaceous
flora are diverse. Hegi (1965) stated that the species in near natural forests can displace its
native congener, I. noli-tangere. Later Sukopp (1962) disagreed because of the different
biotopic requirements of the species. Trepl’s (1984) observations showed that I. noli-tangere
is substituted by small balsam only at suboptimal, bit drier habitats and in an optimal position;
in moist habitats it retains its area and dominance. According to Faliński (1968; 1998a,b) and
Kujawa-Pawlaczyk (1991), small balsam represents a substitutive relationship under
26
conditions of high light-availability and I. noli-tangere could possibly be replaced but this
was a rather theoretical assumption. To date, none of the previous studies have indicated the
negative impact of I. parviflora that resulted in the complete displacement of native species
(Schmitz 1998; Obidziński, Symonides 2000; Kowarik 2003; Tanner 2008; Hejda 2012).
When I. parviflora begins to penetrate into the herb layer sometimes, it can lead to the
dominance of this species. This depends on the composition of the previous species and the
availability of empty niches. The cover of bare ground and the low species diversity of
resident plants seem to facilitate of the invasion of small balsam. Kujawa-Pawlaczyk (1991)
believes that I. parviflora is an indicator of the degeneration of forest communities rather than
a factor that causes it. The same problem was analysed by MacDougall and Turkington
(2005) but in the case of alien invasive plants in general. They posed the question of whether
alien plants benefit from disturbances or are the cause of disturbances in an ecosystem. Some
studies (e.g., Łysik 2008) demonstrated an increase in I. parviflora cover and simultaneously
a decrease of the cover of native species but any possible competitive ability of small balsam
was not proven. Such changes that result in species poverty and disturbances or loss of
ecosystem function can be a consequence of increasing human pressure rather than the
influence of invasive alien species.
By analysing insects that feed on I. parviflora and aphidophagous insects that are associated
with Impatientinum asiaticum, Schmitz (1998) drew the conclusion that the species supports
biodiversity by enhancing rich fauna. Stary and Laska (1999) already formulated a similar
opinion but only in relation to syrphid flies, especially in cultivated landscapes because small
balsam plays a seasonal role as a food resource for them. Kujawa-Pawlaczyk (1991) and
Faliński (1998b) claim that in deep forests with low light availability, I. parviflora is
suppletive component of the herbaceous layer and this leads to an increase in the local
biodiversity in natural plant communities. To date, no effect of the species and its distribution
and abundance on the soil fauna and microorganisms has been observed (Csiszar, Bartha
2008).
1.2.3. Methods of control
In 1942 during World War 2, Impatiens parviflora was referred to as a "Bolshevik
Mongolian invader” by the Nazis. The Reich Central Office of Vegetation Mapping, which
was headed by the well-known phytosociologist, Reinhold Tüxen, demanded a “war of
extermination” against small balsam in the areas that were supposed to belong to the
homeland of Germany, i.e., the occupied areas of Poland (Barbour 1996; Simberloff 2003). In
27
spite of such declarations, no attempts to remove this species were implemented. Only a few
experiments to control I. parviflora (Csontos 1986; Adamowski, Keczyński 1998, 1999) have
been published in literature. Both mentioned experiments that consisted of the removal of
individuals by hand. In the previously mentioned study, 86% efficiency was noted and the
abundance of the population decreased considerably. The author attributed the survival of
I. parviflora in the locality that was studied to the seed bank of the species. The possible input
of seeds from and adjacent area was excluded due to method that was applied. In the latter
study, the authors described a successful attempt to stop an invasion of I. parviflora, although
they emphasised that the treatment and monitoring should be repeated every year. Coombe
(1956) indicated that cutting and pulling up the plants in their flowering phase before seed-set
might be an effective control method due to the fact that the seeds germinate early in the
spring. These kinds of methods are time and labour-consuming and thus they are only
applicable at the initial stage of an invasion and in small areas. The control of populations of
I. parviflora across a larger area or an entire country is not possible. Control activities can
yield results only on a local scale and are desirable primarily in protected areas (nature
reserves, national parks etc). A relatively efficient way to reduce the spread of small balsam
and to reduce the size of the existing population in these areas is the elimination of ruderal
habitats. Existing populations should be eliminated mechanically by pulling, digging or
mowing. If the treatment takes place at the time of fruiting, plants that are removed should be
burnt on the spot, thus reducing the risk of the further spread of the seeds (Solarz et al. 2005).
Csiszar and Bartha (2008) reported that in some afforested sites that were overgrown by
I. parviflora using herbicides successfully hindered germination, although herbicides such as
glyphosate (commonly called roundup) cannot be used everywhere and e.g., in river valleys
its usage is limited and it cannot be used at all times. Godefroid et al. (2007) showed that
a recovery of I. parviflora was noted during tillage treatments. It turned out that from among
several mechanical methods and pesticides, disc ploughing and glyphosate promoted the
development of this plant. It should be noted that I. parviflora was not target species for
application of methods of control in that study.
Biological agents could be another possible method of control. As was mentioned earlier, 13
phytophagous insects, at least two snails and five phytopathogenic fungi have been found on
I. parviflora. Among them, Puccinia komarovii, which could kill whole population seems to
be the most effective.
28
1.2.4. Economic importance
Small balsam, I. parviflora, is an alternative host for the black bean aphid, Aphis fabae,
(Schmitz 1998), which is a major pest of sugar beet, bean and celery crops where a massive
occurrence of aphids can cause the stunting of the plants. It is also a host for the cucumber
mosaic virus CMV (Polak 1967; Brcak, 1979), which can attack many crop plants, apart from
cucumbers, such as tomatoes, carrots, celery, lettuce, spinach and beets. It is well-known that
CMV has a very wide host range (Zitter, Murphy 2009), so the possible role of I. parviflora in
the spread of CMV is probably minimal.
Some reports have indicated that I. parviflora is possibly an edible plant. Griebel (1948)
wrote that the leaves of small balsam contain a lot of vitamin C and when eaten raw could be
source of this substance. This cannot be said about the shoots, which, if eaten raw can cause
nausea. In addition, they contain a lot of oxalates and thus they are not recommended for
people who are prone to kidney stones or arthritis, although the seeds have a pleasant nutty
flavor and can even be eaten raw (Łuczaj 2002). Düll and Kutzelnigg (1988) after Tanner
(2008) reported that the dried stems of the plants have been a source of food for people during
times of famine. Like many other plants, small balsam can be used as a medicinal plant.
Hydro-alcoholic extracts of the fruit and herbs made from the fruits are the most useful.
Therapeutic treatments using extracts of Impatiens spp, including I. parviflora, are broad. The
extract can be used as an anti-inflammatory, a diuretic, an antispasmodic, and an antipsoriasis treatment. It inhibits autoimmunity immune and can be used against lupus, against
atopic dermatitis, as an anti-acne treatment, an anti-androgenic treatment, a hypoglycemic
treatment, an anti-atherosclerotic treatment, an anti-bacterial treatment, an anti-fungal
treatment (systemic), an anti-allergic treatment. It works also as a mild laxative, protective
liver, kidney and heart treatment. Preparations of balsam also prevent the hypertrophy of the
prostate, kidney and urinary disorders of micturition (Różański 2009).
29
2. Methods
2.1. Study areas
The studies were conducted in several regions located in southern Poland – the SilesianKraków Upland (Silesian Upland, the Jurassic Upland) (Fig. 1) and the Cieszyn Foothills
region and on single sites in the Republic of Hungary.
The Silesian Upland is a physical-geographical region that covers an area of nearly
4000 km2, which is located in the southern part of Poland 50°15′N, 19°0′E. The region is
characterised by a differentiated relief and geological structure with its central part being built
of Carboniferous formations. It has also been heavily affected by the coal mining industry. As
a result of human interference, the plant and soil cover has been substantially altered
(Kozyreva et al. 2004). The exploitation of mineral resources began at the beginning of the
Middle Ages and intensified in second half of the 18th century due to changes in economic
activity, technical-scientific progress and urbanisation. The average air temperature reaches
about 8° C and the annual precipitation is estimated to be ca. 700-800 mm. The prevailing
winds are from the western sector (SW, W, NW). Many of the depressions and land
deformations in woodland areas are the result of previous coal mining activity. This has led to
the humidification or the desiccation of large areas. Colliery waste heaps, ground pits and
subsidence reservoirs are characteristic features of the Silesian landscape. According to
phytosociological studies by Cabała (1990), there are 17 forest associations and 26 syntaxa of
lower ranks. The most dominant forest communities are pinewoods, Leucobryo-Pinetum
W.Mat. (1962) 1973, Molinio-Pinetum W.Mat. et J.Mat. 1973, Calamagrostio villosaePinetum Staszk. 1958, and alder-ash carrs, Fraxino-Alnetum W.Mat. 1952. Oak-hornbeam
forests, Tilio-Carpinetum Tracz. 1962, along with beech woods, Dentario-glandulosaeFagetum W.Mat. 1964 ex Guzikowa et Kornaś 1969 are less common. Floodplain forests,
Salici albo-fragilis R.Tx. 1955, Populetum albae Br.-Bl. 1931, are the most weakly
developed. One of the most dominant forest communities is Querco roboris-Pinetum (W.Mat.
1981) J.Mat. 1988 due to massive plantation of Scots pine, Pinus sylvestris, in the habitats of
deciduous forests. Therefore, the origin of this type of vegetation is anthropogenic. The
majority of primary forests were cleared and converted into arable lands or wastelands and urbanised industrial areas. Some of these were afforested mainly by coniferous species and
occasionally by deciduous species (Nyrek 1975). There are probably some remnants of
ancient forests resulting from habitat continuity and quality but they have been exploited by
foresters (Cabała 1990). Thus, the forests that exist in the area are rather recent forests.
30
The Jurassic Upland (Kraków-Częstochowa Upland) is located in Southern Poland and
borders on the Silesian Upland. This region covers an area of about 2615 km2 and is mostly
built from Jurassic dolomites. The characteristic elements of the landscape are limestone
rocks and numerous caves. The mean elevation of the area is about 350 m a.s.l. The soils of
this area are rather poor; 60% are podzolic soils and brown soils occur more rarely. These
most frequently occur in the Olkusz Upland. The river network is weakly developed due to
the high permeability to precipitation of Jurassic lime. The biggest rivers are: the Przemsza,
the Rudawa, the Prądnik, the Dłubnia and the Pilica. All of them supply the Wisła with water.
The mean annual temperature is ca. 7.5°C and is lower for the whole country. The mean
annual precipitation amounts to ca. 700 mm but it increases in a southward direction. The
vegetation period lasts from 201 to 211 days and is also about two weeks shorter than in the
adjacent areas except for a part of the upland in the east (Urbisz 2004).
Fig. 1. Study area – The Silesian-Kraków Upland. Numbers indicate the nature reserves
where the permanent study plots were established. For a description of the nature reserves see
Table 4
31
Impatiens parviflora is quite a frequent species in both areas. It is more common in the
western-southern part of the Silesian Upland while the species grows more frequently in the
southern part of the region in the Jurassic Upland (Fig. 2).
Fig. 2. Distribution of Impatiens parviflora in the study area (Chmura, Urbisz 2005). A –
Silesian Upland, B – Jurrasic Upland
2.2. Habitat research
Soil samples were taken on the sites where the permanent plots were established and
on randomly selected sites where the phytosociological relevés were performed.
Four soil sub-samples were collected from the topsoil from 0-20 cm depths and then mixed
into one composite sample. Each sub-sample was taken from the rhizosphere of Impatiens
parviflora. After air-drying and sieving over 2 mm, the samples were analysed for pH,
measured potentiometrically in H2O and in 1N KCl. Total organic C (%) was measured
according to the Tiurin method. Loss on ignition was tested in a muffle furnace (%) and total
N content - NT (%) was determined using the Kjeldahl method. Available Mg was detected
using FAAS (Flame Atomic Absorption Spectrometry), available phosphorus P in an
ammonium lactate extraction was done using the colorimetry method, sodium Na and
potassium K were detected using flame emission spectroscopy and Ca by spectrophotometry
in 1 N ammonium acetate (mg/kg) (Lityński et al. 1976; Ostrowska et al. 1991). The content
32
of CaCO3 was analysed using the Scheibler method. Granulometric composition was
measured using the aerometric method and the sieve method. Fractions of mineral grains in
sizes of between <0.0002 to 0.02 mm were classified as floatable parts. The results of the
analyses are presented on figures in the form of ecodiagrams (Zarzycki 1976; Borysiak 1984;
Szwed 1986) that show the percentage of the occurrence of Impatiens parviflora as
distinguished by aspect, soil properties and litter depth and other soil variables. Soil reaction,
i.e., pH in water and KCL solution are given after Piękoś-Mirkowa et al. (1996). In order to
characterise the pH of soil, a modified scale according to Fotyma et al. (1987) after PiękośMirkowa et al. (1996) (Tab. 1) was used. The scales of the available nutrient content of
phosphorus, potassium, magnesium, organic carbon and total nitrogen were adopted after
Piękoś-Mirkowa et al. (1996) and the literature cited there. The following classes were used:
0-10%, 10.1-20.0%, 20.1-30.0%, 30.1-40.0 and more than 40.0% in the diagrams that show
the content of loss on ignition. The content of calcium was divided into six classes [mg/100g]:
0.0-200.0, 200.1-400.0, 400.1-600.0, 600.1-800.0, 800.1-1000.0, >1000.0. In addition, six
classes were applied in the case of sodium [mg/100g]: 0.0-10.0, 10.1-20.0, 20.1-30.0,
30.1-40.0, 40.1-50.0, >50.0.
The 7-degree scale of the percentage of the floatable fraction and the division of soils into
groups according to granulometric composition were employed according to Kuźnicki et al.
(1979). The 5-degree scale of litter depth was used after Obmiński (1977).
Tab. 1. Scale of soil reaction
Reaction
Very strongly acid
Strongly acid
Acid
Slightly acid
Neutral
Alkaline
pH
in H2O
<4.1
4.1- 5.0
5.1- 6.0
6.1- 6.7
6.8 -7.4
>7.4
in KCL
<3.5
3.5 - 4.5
4.6 - 5.5
5.6 - 6.5
6.6 - 7.2
>7.2
In the diagrams showing the C:N ratio, the scale of humus was adopted after
Duchaufour (1970) after Piękoś-Mirkowa et al. (1996). According to these authors, the C:N
ratio estimated at ca. 10 demonstrates humus of a calcimorphic type. A forest of a mull type is
indicated by C:N < 20 and it most frequently ranges from 12 to 15. The range for moder is 1525, while for mor it varies from 30 to 40. Thus, the intervals were established as follows: C:N:
<10.0, 10.1-15.0, 15.1-20.0, 20.1-25.0, >25.0.
33
The scale of the content of the elements (P, K, Mg) that was available to plants was adopted
after Piękoś-Mirkowa et al. (1996) (Tab. 2). A Lutron LX-105 light-meter was used to
characterise the light conditions under the tree canopy layer. Five records were taken within
the plot. The light quantity was expressed as a percentage of the light conditions as measured
in an open area nearby at the same time. The canopy-scope was used as an alternative method
of canopy openness measurement (Brown et al. 2000; Salachna, Chmura 2013).
Tab. 2. The scale of contents of nutrients
Content
Low
Middle
High
Very high
P2O5
K2O
MgO
<3
3.- 9
9.- 18.
>18
<7
7. - 14.
14. - 28.
>28
<10
10. - 20.
20. - 40.
<40
The Spearman rank correlation test was used to test relationships between the coverabundance of I. parviflora and selected soil parameters.
In addition to the direct field measurements, the Ellenberg indicator values system, EIVs,
were used. The indicator values for light (L), continentality (K), temperature (T), moisture
(F), reaction (R) and nitrogen (N) were taken from Ellenberg et al. (1992). The mean
arithmetic EIVs were computed on the basis of the presence/absence of the species in herb
layer. They were assigned as: mL, mR, mK, mT, mF and mN, respectively. The species that
were assigned x, i.e., considered to be indifferent were omitted from the calculations.
Impatiens parviflora was also excluded from the calculations. In order to assess whether the
invaded sites could be studied using phytoindication methods, the Spearman rank correlation
coefficients between environmental variables (soil parameters) and the mean indicator values
were calculated.
2.3. Population studies
In order to study variations in the seed mass of Impatiens parviflora, seed samples
were collected in two countries, Poland and Hungary, in 2010 in natural and semi-natural
forest communities (Csontos et al. 2012; Chmura et al. 2013) and they were collected in
Poland once again in 2011 and 2012. They were collected at the full-ripe stage. The seeds
were kept in paper bags under the same room temperature conditions until the date of the
measurements. During storage, the weight of the seeds was monitored until it was stable at
34
which time five 50-seed lots were formed from each sample and measured using an analytical
balance with 0.0001 g accuracy. The viability of seeds was tested using the so-called
“apparent viability” method (Zelenchuk 1961; Csontos et al. 2012, Chmura et al. 2013). The
mass of the 50-seed lots between sites was analysed using the analysis of variance ANOVA
test and the LSD Fisher test for multiple comparisons.
Different regimes were applied to study the conditions of germination. In total there were 60
samples of seeds (100 seeds in each). Seeds were kept air-dry in paper bags. During
stratification they were wet and kept in the dark. The effect of stratification temperature,
storage temperature and time of stratification as well as habitat were analysed. Seeds were
germinated in Petri dishes at 5° C in the dark.
Tab. 3. Basic parameters of the study sites where the seed samples of Impatiens parviflora
were collected. H1-4 samples were collected in Hungary; P1-7 samples were collected in
Poland (Csontos et al. 2012; Chmura et al. 2013 and author’s own studies)
Code
H1
H2
H3
H4
P1
P2
P3
P4
P5
P6
P7
Locality
Vadálló rocks,Visegrádi Mts,
Hungary, karst forest
Rumi forest at the Rába River,
Hungary
Kis-Sváb Hill, Budai Mts, thickets
Near the village of Rábahídvég;
along an earth-covered road in a
Floodplain forest (mainly formed by
willow, poplar and ash) on the bank
of the Rába River
Katowice, Poland, managed forest
with P. sylvestris
“Bukowica” nature reserve near
Wygiełzów, Dentario glandulosae
Fagetum
Soblówka near Ujsoły, Beskid
Żywiecki Mts., Poland
“Skała Kmity” nature reserve,
Zabierzów near Kraków, Poland
Jasieniczanka, Bielsko-Biała
Village of Cisownica near Ustroń,
Tilio-Carpinetum
Katowice, along a forest path, mixed
deciduous forest along the Mleczna
River
Latitude N
Longitude E
Date
47°44'22"
18°54'50"
24.08.2010
47°06'41"
16°50'45"
02.09.2010
47°30'12"
19°00'47"
22.07.2010
47°03'54"
16°44'59"
31.08.2010
50°12'8.31"
18°57'26.13" 04.08.2011
50°4'45.81"
19°24'59.48" 18.09.2011
49°26'13.6"
19°08'33.7"
20.09.2011
50°06'13.24"
19° 49'8.06"
18.09.2011
49°52'53.34"
18°57'16.45" 22.09.2011
49°42'57.43"
18°45'52.26" 23.09.2012
50°12'8.3"
35
18°57’26"
23.09.2012
Differences in the germination percentage due to the types of habitats (oak-hornbeam forest,
Tilio-Carpinetum; alder forest, Fraxino-Alnetum and forest path (seeds were gathered in the
vicinity of the two forest communities mentioned), stratification temperature (3.5° C vs -2.5°
C), storage temperature (8.5° C vs 20° C, only for 3.5° C and 2.5° C of stratification
temperature and all habitats combined), time of stratification (for all of the remaining habitats
and treatments combined) were analysed using G – tests.
Morphological variation
In order to characterize the morphological variation of Impatiens parviflora among
various forest types 415 individuals in total were selected for morphometric studies. The
individuals were selected from forest interiors under the tree canopy in deep or moderate
shading conditions. Sites were situated in patches of the four forest communities that are
found most frequently in the Jurassic Upland – Dentario glanduloase-Fagetum, TilioCarpinetum, Fraxino-Alnetum and Querco-roboris-Pinetum. These four plant communities
are quite common in the nature reserves in this region. As was pointed out by Szary and
Michalik (1998) using the example of the Dolina Racławki nature reserve, the topographic
diversity of the vegetation that covers the hills shows a similar pattern. At the bottom of the
hills, usually in the river valleys, there are habitats of the Ulmenion alliance communities,
while patches of oak-hornbeam forests (Tilio-Carpinetum) are situated in the higher areas and
finally close to top and on the top of the hills, there are phytocoenoses of beechwoods
(Dentario glandulosae-Fagetum, Luzulo-pilosae-Fagetum and on the southern slopes CariciFagetum or mixed coniferous forests. Ten randomly chosen individuals were subjected to
measurements on the site. The following variables were measured: the height of plant, the
width and length of leaves and the number of flowers and fruits per plant. The height of plant
was defined as the distance from the ground level up to the top of a shoot. The length of a leaf
was determined for the longest leaf of a plant. The width of a leaf was determined as the
widest area of the longest leaf. The number of flowers included all of the flowers that were
counted per a plant as was the number of fruits.
Variations in the above-mentioned plant traits are shown on histograms with a fit to the
normal distribution. The Spearman rank correlation test was used to determine whether the
soil parameters had an influence on the plant traits that were chosen or whether there was
a correlation between the morphological plasticity and environmental variables. The KruskalWallis test and Conover test for pair-wise comparisons were used to determine the
significance of any differences in particular plant traits between the distinguished forest
36
communities. Principal Component Analysis (PCA) was used to reduce the dimensionality
and to select the most explanatory variables based on the value of the eigenvalues.
Phenology
Based the Piskorz and Klimko (2002), the following phenophases in I. parviflora were
distinguished: 1 the two-cotyledon phase, 2 the one-cotyledon phase, 3 the vegetative phase
(stems without flower buds), 4 the preflowering phase (flower buds present), 5 the flowering
phase (open flowers present), 6 the prefruiting phase (with unripe fruits), 7 the fruiting phase
(with ripe fruits), 8 the subsenile phase (with declined leaves), 9 the senile phase (stems
without leaves). The frequency of the particular phenophases between the four forest
communities, Dentario glanduloase Fagetum, Tilio-Carpinetum, Fraxino-Alnetum, Querco
roboris-Pinetum and at forest margins of the above-mentioned forest types over the
vegetation season from 16.04.2007 to 30.09.2007 in a 12-time series was observed. From 57
to 80 marked individuals in each type of habitat were studied in terms of the number of
cotyledons, flower buds and flowers and unripe and ripe fruits.
The participation of the phenophases in particular forest types was analysed using
contingency tables that were designed to find any significant differences between them. The
frequencies between the time series and the distinguished forest type were checked using the
G-test for each phenophase. Only those time series for which at least one individual
representing a given phenophase in at least one forest type were taken into account.
Spatial and temporal variation in life history
The morphological variation of I. parviflora among microhabitats was analysed in one
forest community, Dentario glandulosae Fagetum, in the area of the “Bukowica” nature
reserve. Microhabitats within the forest environment were chosen based on the study by
Klama (2002) with the author’s modifications. Klama (2002) distinguished 30 types of
microhabitats, which he referred as “substrates” or “terrain microforms”. Of the list of
microhabitats, only those were chosen that were available in the study area (in at least five
replicates) and that were occupied by I. parviflora. The situation of the chosen microhabitats
is shown on Figure 3. Only one type represented an anthropogenic one – a forest path that was
created because of forest management. The others are natural forest types of habitats that can
be encountered in deciduous forests and that are associated with natural disturbances such as
wind tree falls or forest self-thinning processes.
37
The list of microhabitats includes: Forest interior – soil under tree canopy (1); Canopy
opening – soil under the canopy opening (2); Area near a log – area near a lying decaying
dead log (3); Log under canopy – bark on a decaying dead log of Fagus sylvatica from 1st to
7th of 8 classes of dead wood decomposition according to the classification by Holeksa (2001)
(4); Log under the canopy opening – bark on a decaying dead log under the canopy opening
(5); Root plate – soil in the root plate of dead logs (6); Tree fall disturbance – holes in the
ground after a tree has fallen (7); Hollow – tree hollow in a living tree (8); Path – soil on a
forest path (9), Root collar – base of a living tree (10); Stump – decaying top of stump (11)
(Fig 3.). For sites situated on soil, i.e., forest interiors, random 1 m2 squares of the canopy
openings and forest paths were established. For other types of microhabitats, the selected area
depended on the particular objects.
Only ten individuals were randomly chosen on each site for morphometric studies because the
populations of I. parviflora were small on some sites.
Fig 3. The scheme that shows the types of microhabitats that can occur in a forest.
Explanations: 1: canopy; 2: canopy opening; 3: area near a log; 4: log under the canopy; 5:
log under the canopy opening; 6: root plate; 7: tree fall disturbance; 8: hollow; 9: path; 10:
root collar; 11: stump
38
The following parameters were considered – the height of the stem; the number of flower
buds; the number of flowers; the number of fruits and the presence of cotyledons [%]. The
measurements were conducted in mid-June 2006.
Because of the high degree of variations in the habitat conditions within the selected
microhabitats, the light conditions were measured using a Lutron LX-105 light meter. In
addition to light availability, selected edaphic conditions were also analysed, i.e., C:N, the
content of Ca, CaCO3, loss on ignition, Mg, Na, P and K. In order to estimate the edaphic
conditions, soil samples or samples of the substrate (decaying bark with initial humus;
remnants of soil at the root plate of fallen trees in the case of a dead wood complex), were
collected if they were available. The substrate material was sieved and underwent the
procedures that were described in Chapter 2.2 Habitat Research. In addition to the
environmental variables for each site, the cover of all of the vascular plants were estimated
visually using 1,2,5,10…100% intervals. The entire area of the distinguished types of
microhabitats was taken into account with the exception of forest interiors, canopy openings
and forest paths. A similar procedure was applied by Zielonka and Piątek (2004) for the rotten
logs of spruce, Picea abies.
At least 10 randomly and marked plants were observed from 16.04.2007 to 22.09.2007 for the
temporal study of variations in the life-history traits among the distinguished microhabitats.
Seedling survival (%) was counted and seedling height, the number of leaves, the number of
flower buds, the number of flowers as well as the number of fruits was assessed during the
vegetation season in a six-time series.
In order to examine any differences between microhabitats in the spatial study of variations in
the life history traits, the Kruskal-Wallis test was used followed by a post-hoc Conover test.
G-statistics were computed to test any differences in the percentage of cotyledons between
microhabitats. PCA was used to analyse which traits were the most explanatory. Redundancy
Analysis (RDA) was used to assess the impact of environmental variables that were measured
and the number of accompanying species on the traits of I. parviflora. The Monte Carlo test
with 499 permutations was used to calculate the first type of error – p.
The Kruskal-Wallis and Conover tests were also used to assess the significance of any
differences in the mean variables of the life history traits between microhabitats over the time
in a six-time series.
39
Influence of the presence of dead wood and light on the behaviour of the species.
In order to estimate the impact of light availability (canopy openings) and the trophy
of a forest habitat due to the enrichment of soils by nutrients from dead wood decomposition,
152 individuals were measured in four types of habitats: C – canopy (control habitat: soil
under a canopy of trees, CW – canopy and dead wood (neighbourhood of decaying dead logs
under a tree canopy), O – openings (soil under canopy openings) OW – openings and dead
wood present (neighbourhood of decaying dead logs under canopy openings). In order to
avoid pseudoreplication sensu Hurlbert (1984) and a high variation in the measurements of
the distinguished habitats, the plant harvests were conducted at the same time but from many
sites and were repeated in two succesive years (2007-2008) beginning in 2006. The following
plant traits were analysed: plant height (cm), the size of the longest leaf area, LA, (leaf area
cm2); the biomass of a leaf, LB, [g] and the ratio of the leaf area to the dry biomass of a leaf,
SLA, (specific leaf area). The leaves were dried for 48 hours at a temperature of 60o C.
Two-way ANOVA was used to test the impact of light availability in the canopy openings
and the presence of dead wood on the size of plants and leaf traits. The Student test with the
Bonferroni-Holm correction was adopted for multiple comparisons (Holm 1979).
2.4. Phytosociological research
A modified Braun-Blanquet method with 0,1,2,5,10,20…100 % intervals was applied
for the estimation of species abundance in the phytosociological records (10 m x 10 m). Using
the random-stratified method (Dobosz 2004; Godefroid et al. 2005; Chmura 2013, 2014), 485
relevés were taken in 52 forest complexes (Fig. 1) in the Silesian Upland. The relevés were
distributed according to a pre-established framework on the basis of forest stands, pedological
(Strzemski and Witek 1978), topographical and geological (Kondracki 1998) and a potential
vegetation map (Matuszkiewicz 2008) for the stratified sampling. In order to avoid the risk of
omitting smaller topographic units, i.e., parts of the region that differed in relief, soils and
land use, the types of forests sites for the relevés were distributed within the entire area of the
Silesian Upland. The relevés were only taken in forest interiors. Crop plants sites and those
sites that were places of a disturbance such as the vicinity of forest paths, roads and cutting
areas were excluded from the study. All relevés were done in the patches where invasive alien
plant species can be encountered in this area (Chmura 2004). The invasive alien species were
supposed to be in the phase of naturalization in the patches of vegetation, i.e., the presence of
seedling or juvenile individuals in herb layer in the case of woody plants.
40
The collected material was stored in a table using JUICE 6.3 software (Tichý 2002).
The phytosociological affiliations of species were adopted after Matuszkiewicz (2008),
the
names
of
plants
after
Mirek
et
al.
(2002)
(also
available
online:
http://info.botany.pl/czek/check.htm).
Based on the obtained cover data in order to estimate the degree of the naturalisation
of an invasive species, the percentage cover of neophytes in relevés was calculated according
to the following formula (Chmura, Sierka 2006b):
DN
C
C
NI
100%
ALL
where: DN = the degree of neophytisation (sensu Olaczek 1974, see also Łaska 2001),
CNI – the sum of the cover of all neophytes in the relevé, CALL – the sum of the cover of all of
the species in the relevé.
In order to estimate the role of a particular neophyte in the neophytisation of
a community, its percentage of relative cover PN was computed according to the following
formula:
PN
CN
100%
C
NI
where: PN = participation in the neophytisation, CN – the cover of neophyte in relevé,
CNI – the sum of cover of all of the neophytes in the relevé.
In order to test which syntaxonomical groups of species I. parviflora is accompanied
by, the Spearman rank correlation was used to analyse the relationships between the cover of
I. parviflora and the total cover of representatives of particular phytosociological classes –
Querco-Fagetea, Vacccinio-Piceetea, Alnetea glutinosae and Quercetea robori-petraeae.
2.5. Studies on permanent plots and long-term research
In the years 2004-2012 fourteen nature reserves in total were subjected to
phytosociological and population studies (Tab. 4). A total of 68 permanent study plots were
laid out including 38 in the Silesian Upland and 30 ones in the Jurassic Upland in 2004. Ten
permanent study plots were established in six nature reserves of the second region for longterm research (LTR) between 2005 and 2012. The 3-year research was conducted on one
permanent study plot in the Ochojec nature reserve during the period of 2005-2007 (Sierka et
al. 2009).
41
The size of study plots covered an area 100m2. They were laid out in various forest
communities including riparian forests, oak-hornbeam forests, beech woods and mixed
coniferous forests. Each study plot was divided into 100 subplots of 1 m2 (Chmura, Sierka
2006a). The species composition of field layer was noted on each subplot. The cover of plants
was estimated visually using the following scale: 0,1,2,5,10,20…100 %. The total cover of all
of the plants that were present could exceed 100%.
The biodiversity indices were calculated for all of the subplots:
Shannon-Wiener index:
H' = -
p
ln p j
j
where: pj is the contribution of a species (percent cover) of the j-th species.
Shannon’s evenness index (= Pielou’s evenness index=J’ H/H max):
E = H/Hmax
where H is the value of the Shannon-Wiener index and H max is the H value when all of the
species in a sample have an equal contribution. The greater the value of the index, the more
equal the abundances of the species that were present in a sample.
Simpson’s diversity index:
S
D =1/
( p
i 1
2
i
)
The value of this index ranged between 0 and 1; the greater the value, the greater the
sample diversity. In this case, the index represented the probability that two individuals that
had been randomly selected from a sample would belong to different species.
S was also determined for each study plot as a measure of species richness.
The species from the study plots were classified into several plant functional types according
to the concept of Semenova and Van der Maarel (2000). Among others, species were
classified by their life form according to Raunkiaer, given after Zarzycki et al. (2002) and by
their dispersal mode following the classification of Kornaś (1972) (i.e., autochore, barochore,
endozoochore, epizoochore, myrmecochore and hydrochore). Affinities of a species to
a particular plant functional type were taken from several sources – Kornaś (1972), Dzwonko
and Loster (2001), Frank and Klotz (1990) and Jacquemyn et al. (2001). Ecological strategies
according to Grime (1979) were adopted after several sources – Frank and Klotz (1990) and
Dzwonko and Loster (2001). Seven strategy types were used – three fundamental ones, i.e.,
competitor (C), stress tolerater (S) and ruderal (R) and four intermediate strategies that are
42
indicated by the corresponding letters (CR, CS, SR, CSR). Further intermediate types were
pooled as suggested by Hermy et al. (1999).
Tab. 4. The list of the nature reserves situated in the Silesian-Krakow Upland (Jurassic
Upland - KC and Silesian Upland - SU) that were studied. The number of study plots are
given for studies in the years 2005-2012, *-2005-2007
Name of reserve
1. Bukowica
2. Dolina Eliaszówki
3. Dolina Kluczwody
4. Dolina Racławki
5. Lipowiec
6. Skała Kmity
7. Wąwóz Bolechowski
8. Hubert
9. Dolina Żabnika
10. Łężczok
11. Ochojec
12. Segiet
13. Las Murckowski
Number of
Area
permanent study
Latitude
Longitude Region
ha
plots for longterm research
KC
2
22.7 50˚04’43"N 19˚23’55"E
KC
2
109.57 50˚10’18"N 19˚38’02"E
KC
1
35.22 50˚09’54"N 19˚49’10"E
KC
1
473.9 50˚09’49"N 19˚41’33"E
KC
2
12.44 50˚04’42"N 19˚26’38"E
KC
2
19.36 50˚06’06"N 19˚48’38"E
KC
22.44 50˚09’30"N 19˚46’55"E
SU
13.47 50°32'42"N 18°26'49"E
SU
42.33 50°13'03"N 19°23'53"E
SU
396.21 50°08'01"N 18°16'27"E
SU
1*
25.73 50°12'24"N 19°00'07'' E
SU
24.99 50°24'18"N 18°50'53"E
SU
100.67 50°11'32"N 19°3'41"E
In order to study the effect of scale on the relations (Stohlgren et al. 2006) between
I. parviflora and native species, differences in plant functional types were compared at several
levels – the floristic level, a high scale and a low scale. The floristic level comprised the
number of species belonging to the aforementioned groups, which were divided into two
groups – those accompanying I. parviflora (species that occurred in subplots that were
occupied by the species) and species that were present in the invaded subplots. The high
scale, i.e., level of 100m2 – total frequencies of species in the subplots that were either
occupied or non-occupied by I. parviflora was taken into account. For the third level, the low
scale of 1 m2, the total cover of plant functional types were computed for subplots of 1 m2.
TriGraph software (Legg 2004a) was used to compare the Grime strategies at the floristic
level and CSR Bubble Plot was done. The Mann-Whitney test was used to examine any
differences between the two groups of species at the distinguished scale levels.
The Spearman rank correlation test was used to test the hypothesis that the number of
species and the cover of resident species affect the cover of I. parviflora in relation to all 68
plots combined and for the plots in the Jurassic Upland and the Silesian Upland separately for
2005. The correlations were repeated in 2006 for the plots of the Kraków-Częstochowa
43
Upland. In addition, the density (the number of individuals of I. parviflora) was correlated
with species richness and the cover of native plants in this region in 2005 and 2006.
The Mann-Whitney test (Wilcoxon sum rank test) was used to examine any
differences in the number and cover of native species in the presence and absence of small
balsam. This procedure was done separately for the study plots from the Jurassic Upland and
the Silesian Upland.
For ten selected study plots (LTR plots), each of which is situated in the Jurassic
Upland, the studies were continued. Sampling of herb layer vegetation was carried out in the
peak of vegetation season (mid-June till mid-August). These study plots differed in the
invasion level by I. parviflora at the beginning of the long-term research. For two first years,
2005 and 2006, within a five-metre belt of the border of the LTR plot, each individual of
small balsam was removed in order to isolate the population from the plants in the vicinity.
Any other new plants that appeared in the remaining years were regarded as individuals that
had developed due to the self-dispersal of the population within the LTR plots.
The LTR plots can be divided into the three groups – 1) the initial invasion group –
frequency not higher than ten. This group contained two study plots nos. 3 and 8 – the patch
of Dentario glandulosae-Fagetum in the “Dolina Racławki” nature reserve and the patch of
oak-hornbeam forest, Tilio-Carpinetum” with Aconitum moldavicum, in the “Skała Kmity”
nature reserve, respectively. The next group 2) intermediate and advanced invasion group –
contained five study plots with a frequency of 37-66 occupied subplots per study plot. These
were study plots nos. 2, 10, 5, 9, 1 – two patches of Tilio-Carpinetum in the “Dolina
Eliaszówki” nature reserve and one from the “Dolina Kluczwody” nature reserve and one site
of beechwood, Dentario glandulosae-Fagetum, from the “Bukowica” nature reserve. The
next one was the riparian forest, Fraxino-Alnetum, in “Skała Kmity”. The last group 3)
saturation invasion group encompasses three study plots, with a frequency at least 90 invaded
subplots, nos. 4,6,7 – two patches of Tilio-Carpinetum from the “Lipowiec” nature reserve
and the phytocoenosis of Dentario glandulosae-Fagetum in “Bukowica”. In addition to
recording the cover of each of the vascular plants in the subplots within the study plots, the
spatial structure of the Impatiens parviflora population was determined following Chessel
(1977) after Falińska (2004).
Additionally, in the LTR plots in 2005 and 2006 the cover of bare ground and ground
that was exclusively covered by I. parviflora, the cover of litter and the total cover of gaps
(other empty sites in the herb layer, including stones, mobile wood rests, trees – all substrates
that were not occupied by a species) were estimated. The density of I. parviflora, i.e., the
44
number of shoots (individuals), was assessed in each subplot of the study plots. A 1m x 1m
wooden frame was used to collect all of the data within the subplots for measurements in the
field. In the remaining years only the cover of all species were noted and measuring tape was
used instead of a wooden frame in order to minimise the effect of trampling on the ground
flora and the dispersal of seeds by I. parviflora.
The Spearman rank correlation coefficients were calculated in order to test the relations
between a disturbance in the herb layer (bare ground, ground covered by litter) and the
density and cover of I. parviflora.
The contingency tables (G-test) were used to assess any differences in the frequency of
I. parviflora between the years. Differences in the total cover over the years was checked
using the Kruskal-Wallis test followed by a post-hoc Conover test, whereas the Friedman rank
test followed by the Conover test were used between the years in the same occupied subplots.
The phenomenon of autocorrelation may occur in the afore-mentioned studies,
especially when the variables that are computed based on vegetation data from the subplots
were subjected to correlation tests. It is noteworthy that similar correlation tests, in which the
data from subplots within transects, i.e., spatially autocorrelated sites, were performed by
Obidziński and Symonides (2000). This phenomenon of autocorrelation accompanies almost
all natural processes. It is the spatial relationship of the values of variables that results from
the spatial continuity of the environment (Kapusta 2004 after Franiel 2012). This problem will
always be present in cases in which the study plots are divided into smaller subplots. There is
a high probability that adjacent study plots are similar in terms of habitat properties and
species composition due to their close proximity. Thus, subplots cannot be regarded as
independent samples as observations, but on the other hand, it was revealed that the study
plots are rather heterogenic, which was manifested by the high beta-diversity (among the
subplots within the study plots) (Chmura, Sierka 2006a). The correlation coefficients, which
were computed based on the variables that were taken from subplots, showed a similar trend
as the correlations between the variables from the independent samples (study plots), e.g.,
a negative significant correlation between the beta-diversity of the native flora and the
frequency of I. parviflora in the study plots. Moreover, to reduce any possible autocorrelation,
additional data sets, i.e., 1000 randomly 1 m2 subplots with different abundances of
I. parviflora in each of the two regions that were studied were analysed in terms of the cover
of all of the species that were present and were also subjected to correlation analyses.
45
2.6. Mycorrhizal research and biotic studies
Studies on arbuscular mycorrhiza, AM
The methods in mycorrhizal studies were described in detail in the paper by Chmura
and Gucwa-Przepióra (2012). A total of 900 one-cm long root pieces were taken from the
topsoil of 30 sites in the Jurassic Upland and subjected to laboratory analyses according to the
modified method by Phillips and Hayman (1970). The following parameters of arbuscular
mycorrhisation (AM) were recorded: mycorrhizal frequency (F%) – the ratio between root
fragments that had been colonised by AMF mycelium and the total number of root fragments
that were analysed; relative mycorrhizal root length (M%) – an estimate of the amount of root
cortex that was mycorrhizal relative to the whole root system; the intensity of the colonisation
within individual mycorrhizal roots (m%); relative arbuscular richness (A%) – arbuscule
richness in the whole root system and (a%) – arbuscule richness in root fragments where
arbuscules were present (Trouvelot et al. 1986). The parameters of AM colonisation were
described for a sample of the roots of ten individuals.
In addition to the mycorrhizal colonisation parameters, the number of various structures –
arbuscules, vesicles and coils was also counted in each sample.
The results of AM colonisation were related to the soil variables on each site where root
samples were collected and were related to plant height, width and the length of the longest
leaf, the number of flowers and the number of fruits per plant of I. parviflora individuals
(Chmura, Gucwa-Przepióra 2012).
An alternative statistical approach was used in this work. In the paper by Chmura and
Gucwa-Przepióra (2012), the association between the morphometric traits of individuals and
AM colonisation was studied using correlation tests between the mean values of the plant
traits based on the measurements of ten individuals and AM colonisation indices. Such
a procedure leads to the loss of some information about any variations in the plants in
a sample. Therefore, in this work all of the individuals that were measured were counted in
the correlation matrix data; however, the values of the AM colonisation parameters were
pooled (repeated). Only the Spearman rank correlation tests were used because of repeated
values. The Bonferroni correction for multiple tests was not applied assuming the argument
given by Moran (2003) in order not to omit possible important results.
46
Research on interactions with coexisting species in the forest floor and other substrata
For the purpose of this research, floristic and vegetation data from both spatial studies
on the permanent plots from the Silesian Upland and the Jurassic Upland were analysed as
well as data from the long-term research. The species composition of each study plot was
compared using Euclidean distance between the years. The median of the biodiversity indices
including the number of species, the values of Shannon-Wiener and Evenness were compared
using the Friedman tests and the Conover test for pair-wise comparisons, i.e., differences
between succeeding years.
The classification of indicator species into two groups of sites –invaded and uninvaded
– at 1 m2 was performed using the indicator value, i.e., the IndVal method (Dufręne, Legendre
1997) as modified by Cáceres and Legendre (2009) and finally improved by Cáceres et al.
(2010). The statistical significance of this relationship is tested using a permutation test. This
classification was performed separately or specific distinguished vegetation units. Only those
indicator species for both groups that had an IndVal higher than 0.6 of the range (0.0-1.0)
were taken into account. Habitat overlap and the coexistence and exclusion of Impatiens
parviflora and accompanying species across plant associations were tested by calculating the
Spearman rank correlation coefficients between the cover of I. parviflora and the cover of the
most common coexisting species. The correlations were performed separately for all of the
study plots combined from the nature reserves of the Silesia Upland and for the study plots of
the Jurassic Upland.
Any changes between the abundance of I. parviflora and the most abundant species in the
herb layer in the particular study plots in the Jurassic Upland were analysed by comparing the
values of the Spearman rank correlation coefficients between the covers of the species in
respective years.
The following microhabitats were included in the analyses for research on relations between
I. parviflora and accompanying species in the microhabitats that were associated with dead
wood – the area near a log; hollow; log under a canopy; log under a canopy opening; root
plate; stump and tree fall disturbance. The association between the density of I. parviflora and
the number of native species and their total density was tested using the Spearman rank
correlation test. To measure the density of other accompanying species, the number of shoots,
leave rosettes or clumps that were due to the life form of a species were used. The density was
calculated as the total density of plants per unit of area (1 m2). The density was correlated
with cover (see paragraph Spatial and temporal variation in life history in Chapter 2.3
Population Studies). Rarefaction gave similar results, e.g., (Chmura 2008c).
47
2.7. Data processing
The choice of tests that were used for statistical analyses was made due to the
character of data. Non-parametrical tests were performed for the ordinal and categorical data.
In each case percent of cover was treated as ordinal data because it was derived from a visual
estimation. Categorical data were analysed using contingency tables, mainly the G-test and
less frequently chi-square test. The former is less conservative than the latter. The G-test was
used with or without Williams’ correction depending on the size of the samples and the
presence of zeros. When the data were of an interval scale, then the distribution of normality
was checked using the Shapiro-Wilk, Kolmogorov or D’Agostino-Pearson tests. The
homogeneity of variance was checked using the Levene test (Sokal, Rohlf 1995). For
convenience and clarity, in some cases the descriptive statistics were presented as means ± SD
(instead of medians) or simply as arithmetic means as in the case of the analysis of life traits
in time. Statistical analyses were conducted chiefly using the R language and environment
(several versions), (R Development Core 2012) and with some exceptions: Pop Tools (Hood
2011), PAST (Hammer et al. 2001) and contingency tables software (Legg 2004b). The
majority of the ordinal analyses, PCA, RDA and DCA, were conducted using CANOCO 4.5
software (ter Braak, Šmilauer 2002). An alpha level of significance at p<0.05 was accepted
throughout the entire work.
48
3. Results and discussion
3.1. Biotopic requirements of species
3.1.1. Diversity of substratums and soil conditions
Ecological amplitude
The species tends to be more frequent on “cold” slopes, i.e. northern-facing (N, NE,
NW) than on “warm” slopes, i.e. southern-facing (S, SW, SE) (Fig. 4). The former contribute
38.7% to all slopes, the latter – 25.8%. The majority of the stands of small balsam in forests
are sites where the litter depth varies between 1-4 cm – almost 51% followed by a litter depth
of between 4 and 6 cm (27%) (Fig. 4).
Fig. 4. Aspect of the sites and the litter depth on sites with Impatiens parviflora
Fig. 5. Variations of soil pH on the sites with Impatiens parviflora
49
Almost 70% of all of the stands that were studied are sites with definitely acid soils
(very strongly, strongly and acid); only 30% are slightly acid, neutral and alkaline habitats
both in terms of pH in the water and in pH in KCl (Fig. 5).
Fig. 6. Variations in the organic carbon and C/N on the sites with Impatiens parviflora
Impatiens parviflora prefers sites with an intermediate concentration of organic carbon
(Fig. 6). The majority (i.e. almost 40%) of the sites represent a moder type of humus (Fig. 6).
Fig. 7. Variations in the amount of available phosphorus and magnesium on the sites with
Impatiens parviflora
50
Fig. 8. Variations in the amounts of available potassium and total nitrogen on the sites with
Impatiens parviflora
Fig. 9. Variations in the amounts of calcium and sodium on the sites with Impatiens
parviflora
51
As far as the trophy, which is characterized by the concentrations of available phosphorus,
magnesium, potassium and total nitrogen, is concerned, sites with the presence of small
balsam are rather poor (Fig 7-8). More than half of the sites are located in the lowest classes
of particular scales. See Table 2 for a comparison. The content of calcium indicates that soils
that are occupied by I. parviflora are rather poor or intermediate rich in Ca2+ ions in this area
(Fig. 9). Soils also are very poor in relation to their content of sodium (Fig. 9).
The loss on ignition is rather low and almost 90% are contributed by the two lowest classes in
the scale (Fig. 10). Variations in the participation of floatable parts in the granulometric
composition are interesting.
Fig. 10. Variations in the loss on ignition and the percentage of floatable parts in the
granulometric composition of soils on the sites with Impatiens parviflora
52
There are two peaks – participation of less than 5.0% in a soil sample is the most
frequently represented (ca. 43%); another quite abundant contribution (ca. 35%) is for a class
that varied from 35.1-50% of floatable parts in the soil (Fig. 10).
Since the work by Brothers and Spingarn (1992), many studies have revealed that some
species prefer southern-facing and that others prefer northern-facing slopes. Godefroid et al.
(2003) showed that species tend to be more frequent on eastern slopes. Coombe (1956)
claimed that small balsam is more abundant on the northern-facing slopes but only when there
is less shade. Indeed, in the initial stages of an invasion, small balsam was found on northern
slopes (Csontos 1986). In their ordination analysis, Ling and Asmore (1999) found that small
balsam is generally associated with steeper slopes but they did not specify the aspect.
Previous studies conducted only in the Jurassic Upland (Chmura 2006) or those performed on
a larger area that included a smaller number of sites in the analysis demonstrated similar
results (Chmura et al. 2007) and showed a preference for northern slopes. In a recent work
that included an analysis of indicator species, Godefroid et al. (2006) proved that small
balsam is confined to northern slopes to a significant degree. As regards litter thickness,
Coombe (1956) wrote that litter decomposition and nitrification is active but mentioned
nothing about the thickness of litter although it can be concluded that sites with litter are
preferred by the species.
Węglarski (1991) showed that the roots of I. parviflora can grow down to a depth of 12
cm and therefore litter should not pose a barrier for species establishment.
The frequency of I. parviflora on sites that differed in soil parameters were analyzed by
Węglarski (1991) who used his own ecological indicators values to study the ecological
amplitudes of species that occur in the Wielkopolski National Park. According to his research,
small balsam is an indicator of weakly acid and acid soils that are rich in phosphorus and
potassium and very rich in humus. As for the percentage of floatable parts, soils were
classified as sandy clays and clays. The present results confirm that small balsam is confined
to soils with lower pH; however, it seems that there are two optima. The majority of stands
are characterized by lower contents of phosphorus and potassium, which is in contrast to
Węglarski’s (1991) study. Indeed, soils that were occupied by the species can be classified as
loose sands (less than 5.0% of floatable parts) and the second most optimum are medium
clays (35%-50%). Thus, the present study partially confirms this.
Vervoort et al. (2011) analyzed the habitat overlap of small balsam and congener I. nolitangere based on 13 sites with deciduous and coniferous forests in Belgium. What is
interesting is that there were no differences between the two species, but no comparison was
53
done with Impatiens sites. Sites with I. parviflora had higher pH than uninvaded sites (5.2 vs
3.0) and higher concentrations of magnesium and potassium (1285±446 cmolc kg 1554±108) vs non-occupied sites (98±20, 112±10), respectively. Those results are
contradictory to those presented in the study but this should be treated with caution because of
the different methods that were applied. Both ecodiagrams with a priori assigned classes of
particular variables and a comparison of invaded and uninvaded sites in a limited area can
give different and biased results. An analysis of the environmental requirements of species
that takes into account its abundance seems to be more reliable.
Relationship between abiotic parameters and species abundance
Four soil parameters are significantly and positively correlated with the cover-
4
5
6
7
8
0 20
60
rs = 0.26, p<0.01
0
20
40
60
rs = 0.38, p<0.001
rs = 0.26, p<0.01
10
30
50
0 20
60
Content of Mg [mg/100g]
Cover of small balsam [%]
pH
60
0
Cover of small balsam [%]
0 20
60
rs = 0.2, p<0.05
0 20
Cover of small balsam [%]
Cover of small balsam [%]
abundance of I. parviflora (Fig. 11).
Content of K [mg/100g]
0
500
1000
1500
Content of Ca [mg/100g]
Fig 11. Results of the Spearman rank correlation (rs) between the cover-abundance of
Impatiens parviflora and the properties of soils
54
The strongest and medium correlation was with the concentration of potassium and the
remaining ones with magnesium and calcium have a medium and weak correlation with pH.
Despite the fact that the ecodiagrams showed that sites with the presence of I. parviflora were
typified by low contents of magnesium, potassium and calcium (Fig. 7-9), analyses of the
abundance of the species and the content of the elements in substratum produced opposite
results.
Dobravolskaite (2012) showed that a lower density of I. parviflora was observed on
soils that were poorer in nutrients and that sites with rich humus soils had a higher number of
plants. The correlation analysis is more in agreement with the above mentioned results by
Vervoort and Jacquemart (2012). In their study, the presence of species was typified by
a higher concentration of nutrients, and in this study the cover of small balsam also increases
with increasing values of some nutrients. Dyguś (2008), who investigated the effect of
fertilization on changes in vegetation and the concentrations of nutrients in the organs of
plants, stated that I. parviflora is a potassium-demanding plant. Despite the fact that his was
a different type of research, his finding corresponds to the present study, which showed
a positive correlation with potassium. Čuda et al. (2014) indicated that moisture was
negatively correlated with the cover of small balsam, whereas tree cover was positively
correlated. In the present work, a negative correlation (rs= -0.07) with moisture was revealed
but it was non-significant. No significant relationship with nitrogen content was detected in
the present study. Buriánek et al. (2013), when analyzing phytosociological and soil data in
the Czech Republic, did not find significant changes in Impatiens parviflora coverage due to
the nitrogen content in soil in either the humus and M01 (0-10 cm) and M12 (10-20 cm)
layers. However, with a decreasing value of C/N in humus and M01, M012: 0-10 cm and 1020 cm, the cover of small balsam significantly increased. Ratio of carbon to nitrogen is
considered as a good evidence of nitrogen saturation in ecosystem (Burianek et al. 2013, Aber
et al. 1989). The lowered value of C/N indirectly indicates demand of the species for nitrogen
and therefore I. parviflora can be considered as nitrophilous species. Moreover, it is known
that input of nitrogen into soil may cause increase of cover of small balsam as in the research
in oak pine by Turnau et al. (1992). The other important thing is that total nitrogen content
perhaps, which did not showed significant result in the present study, is not a good measure to
investigate plant responses to nitrogen in soil. The available forms of this element for plants
are ions NH4+ and NO3. The evidence for unfit of total nitrogen for such analyses was given
by Rahmonov et al. (2013) who studied soils from pine forest, river banks to colliery waste
tips overgrown by Reynoutria japonica no significant differences in total nitrogen content was
55
exhibited. Kupcinskiene et al. (2013a) showed that eutrophication of the environment,
especially in anthropogenic habitats, is an important factor in the spread of alien Impatiens
spp including I. parviflora in the Baltic regions. Moreover, the uptake of heavy metals also
does not hinder an invasion of this species.
As regards soil reaction, there are ambiguous data because Macková (2012) obtained opposite
results the number of individuals decreased at higher pH.
3.1.2. Phytoinidication of the patches of the communities with a contribution of small
balsam
The mean Ellenberg indicator values EIVs, which were calculated on the
presence/absence data, differ from those EIVs that were originally assigned to Impatiens
parviflora (Ellenberg et al. 1992) (Fig. 12). The differences concern, among others: L original value light = 4 vs calculated value mvL = 5.12; moisture F =5.7 vs mF = 5. The
remaining calculated values for other Ellenberg indicator indices are lower than the original
values.
7
6
5
Observed
4
3
x
Ellenberg
orig
2
1
0
L
T
K
F
R
N
Fig. 12. Original Ellenberg indicator values vs the obtained EIVs (Means±SD) for the Silesian
Upland
According to Ellenberg et al. (1992), Impatiens parviflora is indifferent as to its
tolerance to soil reaction. For the sites in the Silesian Upland, mR = 5.09 and it is also
characterized by the highest value of the variation coefficient cv =14.5%.
The EIVs were calculated for I. parviflora from five sites in the Czech Republic (Čuda et al.
2014). The mean mL was 4.52, which was an intermediate value between the original value of
L and mL for Silesian Upland, while the moisture mF in the Czech Republic was almost
equal, i.e. 5.72 vs 5.76 and soil reaction was higher mR 5.87 vs 5.09. The nutrient content
56
espressed by mN for the Czech Republic was also higher (6.48 vs 5.05). These results showed
that the soils in this area are poorer than those that are most prefered by the species. Chmura
and Urbisz (2005) demonstrated differences between two adjacent regions – the Silesian
Upland and the Jurrasic Upland in southern Poland using EIVs as well as between regions but
only for particular plant communities. These findings were confirmed by the differences in
the soils that were studied among these two mesoregiones and the Glubczyce Plateau
(Chmura et al. 2005). The EIVs were used to study a comparison of the responses of native
Impatiens noli-tangere and I. parviflora in the forests of Belgium (Godefroid, Koedam 2010).
The optimal EIVs for small balsam were as follows: mL = 3, mF = 4, mN = 3 and mR = 4.5.
The authors highligtened that their results were not in accordance with the original data as
I. parviflora turned out to be a more shady plant that prefered more acidic and nutrient poor
soil conditions. After recalibrating the indices for the British Isles, Hill et al. (1999) obtained
a relatively high value for nutrients (N=8).
It is interesting to what extent differences in EIVs indicate differences in the environmental
conditions, the responses of the ecotypes of I. parviflora or the accuracy of methods that are
applied.
The results of the calibration of the calculated EIVs with environmental measurements
showed that the correlations are positve and medium for the sites with I. parviflora in the
Jurrasic Upland but for the Silesian Upland only the soil reaction mR and ph_H20 are
significantly correlated. The remaining correlations turned out to be non-significant despite
the fact that a larger sample had been used in the analyses in the case of the Silesian Upland
(Tab. 5).
Tab. 5. Spearman rank correlation coefficients between the Ellenberg acidity reaction value
mR and the chosen environmental viariables
Variables
mR (SU)
mR(JU)
rs
p
rs
p
pH aqua
0.32
0.05
0.40
0.02793
pH KCl
0.28
ns
0.39
0.03397
Ca
0.30
ns
0.42
0.02189
Number of plots
38
30
SU – Silesian Upland, JU – Jurassic Upland
Many studies have demonstrated a high correlation of the mean Ellenberg indicator values
with the measured variables (Dzwonko 2001 and literature cited therein). Most of the
calibrating studies of EIVs using field measurements have been conducted for forest
57
communities (e.g., Diekman 1995; Dzwonko, Loster 2000; Dzwonko 2001; Balkovič et al.
2012). One of the best predictors is mR, which was proved for many types of ecosystems by
Ertsen et al. (1998); however, there are arguments that calibrations should take the effect of
vegetation type into account (Wamelink et al. 2002). On the other hand, it was shown that
even in strongly transformed and secondary habitats such as colliery waste tips (Woźniak,
Błońska 2009) mR can be a good predictor. In the present study, it was demonstrated that pH
of soil was better predicted in one region (Jurassic Upland) than in the other (Silesian
Upland). The main reason for this result is probably the difference in the quality of the forests,
which is associated with age, substratum and origin. Forests that are located in nature reserves
in the Jurassic Upland are characterized by a high degree of naturalness and diversity
(Babczyńska-Sendek et al. 2005). The location of the majority of them on calcareous hills has
meant that they have been preserved in an agricultural landscape that has been untouched by
farming practices. The forests in the Silesian Upland, in regard to the degree of
synanthropization (sensu Faliński 1986), resemble recent forests. The Ellenberg indicator
system has already been tested in respect to light, pH_H2O, total nitrogen, cation exchange
capacity (CEC) for ancient and recent forests (Dzwonko, Loster 2000; Dzwonko 2001). This
study revealed that EIVs are better predictors of soil parameters in ancient forests than in
recent forests. Wulf (2003) identified Impatiens parviflora as an indicator for ancient woods,
whereas Graae et al. (2004) showed that there is no clear preference for ancient or recent
forest in the case of small balsam. The answer to the main question of whether invaded forest
communities can be predicted by EIVs would be biased by the type of forest that is invaded.
Various studies proved that I. parviflora can occur in almost all types of forests, especially
deciduous ones and that their naturalness and species diversity does not matter. The presence
of small balsam as such probably does not affect the reliability of EIVs prediction.
3.2. Life history traits
3.2.1. Diversity of seeds and capacity of germination
Diversity of seeds
The highest mean mass was reported for the population from the Beskid Mts in Poland
and the lowest was recorded for the Visegrádi Mts in Hungary (Fig. 13). These populations
were distinct from other stands in both countries. The Hungarian population with the heaviest
seeds on average was located in a river valley, whereas the population with the lightest seeds
in Poland was also observed in the same type of habitat. There are populations in each
country that significantly differ, which was new data and contrary to the previous study
58
(Csontos et al. 2012). Differences in the mass of seeds in small balsam between Poland and
Hungary were already discussed in that work. The authors pointed out climatic factors: high
temperatures and drought as the main drivers of the lower seed mass in Hungary. On other
similar research that focused on Impatiens glandulifera (Chmura et al. 2013) and additionally
included Germany climatic conditions were also determined to be the main causes for the
variability of seed mass – differences in precipitation and temperature were mentioned. The
present study shows that differences can concern not climatic variables but environmental
factors and ecotypes as well. Seed mass for I. parviflora was reported from 5.69 (Moravcová
et al. 2010) and 6.862 (Dostál 2010) for the Czech Republic and from 6.91 to 9.0 g of
thousands of seeds according to the database of Royal Botanic Gardens Kew (RBG 2008).
Such a high variation may be a consequence of morphological plasticity that is associated
with the adaptation to local conditions. The two populations with the lightest seeds were from
an alder floodplain forest and a margin managed forest with Pinus sylvestris. There were no
significant differences between wet and more open vs drier and more closed habitats (Student
T-test, t=-0.7461, p=0.46). The only factor that explained the differences between the
populations was time (t=3.99, p=0.0003617). The populations from which seeds were
sampled in 2011 had heavier seeds (0.33±0.04 g) on average than those sampled the following
year (0.27±0.04 g). In 2011 a higher mean temperature in summer was noted compared to
2012.
a
ab
de bcd cde
0.1
0.2
0.3
ab ab bcd ab abc
0.0
mass of 50 seed lots
e
H1 H2 H3 H4 P1 P2 P3 P4 P5 P6 P7
code of population
Fig. 13. Comparison of air-dry seed weights among the selected populations of I. parviflora.
Means ± SE are presented. Different letters above the bars indicate non-significant differences
between the samples (ANOVA followed by LSD test). Abbreviations of code of population
see table 3
59
There were serious droughts like those in Hungary so the higher temperature and more light
availability positively influenced the ripeness of the fruits and the seeds inside them.
The number of seeds per capsule differed significantly among forest communities
(G=103.3, P<0.0001) (Fig. 14). Only one-seed capsules were noted in patches of beechwood
Dentario glandulosae-Fagetum. In the three other forest communities, this type of capsule
was the most frequent reaching 92.5, 87.5 and 90 for an oak-hornbeam forest, an alder forest
and a mixed coniferous forest, respectively (Fig. 14). Three and four seeds were only present
in capsules in an alder forest and in a population on a forest path. In the latter capsules with
two or more seeds were present in more than half of all of the fruits that were measured and
analyzed. Csiszar and Bartha (2008) wrote as many as five seeds can develop in capsules;
however, no fruits containing five seeds were observed during this study. The majority of
capsules had no more than three seeds per capsule although fruits with four seeds were also
present, which is in disagreement with the research by Perrins et al. (1993), who recorded
fewer than three seeds in a garden experiment conducted in the UK. However a similar
percentage of fruits containing from one to four seeds was observed in Wielkopolski National
100
Park by Piskorz (2005).
60
40
0
20
percentage
80
4 seeds
3 seeds
2 seeds
1 seed
DF
TC
FA
QP
FP
Fig. 14. Comparison of seed production (number of seeds per capsule) between forest
communities and the contact zone. DF – beechwood Dentario glandulosae Fagetum, TC –
oak hornbeam forest Tilio-Carpinetum, FA – alder forest Fraxino-Alnetum, QP – mixed
coniferous forest Querco roboris-Pinetum, FP – forest path
60
Plants growing in canopy gaps had three and four seeds but they were rare 6% and 1%,
respectively. In the present study the population in which three- and four-seeded plants were
the most abundant grew on a forest path where the light availability was higher and the cover
of other plants was lower. The lower number of seeds that was observed by Perrins et al.
(1993) may be attributed to a different morphological variation caused by climatic factors.
Jończyk (2007) studied variations in the dormancy of I. parviflora seeds of 11 European
populations. The British population was characterized by constantly differing times of
germination. It can be expected that some other traits would differ among remote sites.
Germination conditions
A comparison of the different treatments that were applied in the germination
experiment revealed that all of the treatments that were used had an influence on the
percentage of germination of I. parviflora (Tab 6). Detailed data about germination [%] are
given in Table 7. On average, 34.5% of seeds for all of the treatments from the forest path
habitat combined germinated, followed by the alder forest (25.9%) and the oak-hornbeam
forest (21.8%) (Tab. 7).
The mean percentage of germination for all seeds that had been stratified at -2.5°C amounted
to 50.7%, whereas the seeds that had been stratified in 3.5°C scored 12.2% on average. The
seeds stored at 8.5°C amounted to a mean germination of 37.6% and seeds stored at 20°C
germinated at 23.8%.
Tab. 6. Results of the comparison of treatments and types of habitat on the percentage of
germinated seeds of I. parviflora
G
P
df
Effect of habitat (only for stratification at 3.5°C and -2.5°C
86.2 <0.0001 8
combined)
Effect of stratification temperature (3.5°C vs -2.5°C on all habitats
60.96 <0.0001 4
combined)
Effect of storage temperature (8.5°C vs 20°C, for only 3.5°C and 73.70 <0.0001 4
2.5°C of stratification temperature and for FA and TC)
Effect of the time of stratification (for all remaining types of
516.84 <0.0001 4
habitat and treatments combined)
Time influenced the germination of Impatiens parviflora seeds and the mean percentage of
germinated seeds decreased as follows: 40.25% (14 weeks), 38.2% (18 weeks), 37.2% (16
weeks), 9.1% (11 weeks) and 6.7% (9 weeks).
61
The results that were obtained confirmed previous findings (Kinzel 1912; Coombe 1956,
Shaddach 2008; Perglowa et al. 2009) that storage of seeds at room temperature without cold
stratification causes no germination or that only a few seeds are able to germinate. The
present study shows that independent of the habitat from which plants derive a lack of
stratification has the same effect (Tab. 7). Kinzel (1912) and later (Coombe 1956), proved
that exposure of seeds to temperatures around -5°C enhances the germination percentage.
Temperatures of 0-5°C and wet conditions are almost always pivotal for starting germination
in this species, which was showed by Jouret (1976).
Tab.7. Percentage of germination of Impatiens parviflora in various types of habitat and
treatments
Weeks of stratification
9
11 14 16 18
FA
20° C
8.5° C
0
0
0
0
2
20° C
3.5° C
0
0
0
0
10
20° C
-2.5° C
0
25 70 67 87
8.5° C
3.5° C
5
10 48 63 52
8.5° C
-2.5° C
60
54 89 90 85
TC
20° C
8.5° C
0
0
0
0
0
20° C
3.5° C
0
0
10 18
5
20° C
-2.5° C
0
2
92 51 40
8.5° C
3.5° C
0
0
8
12
4
8.5° C
-2.5° C
0
2
86 45 39
FP
20° C
3.5° C
0
0
0
10 50
20° C
-2.5° C
15
16 80 90 84
FA – Fraxino-Alnetum, TC – Tilio-Carpinetum, FP – forest path
Habitat Temperature of storage
Stratification
This study demonstrated that temperatures below 0° C induced a more abundant and faster
germination. Sometimes, natural stratification is not enough (Godefroid et al. 2011), and
therefore a stable below-zero temperature is needed to induce germination. The same was
showed by Komosińska (pers comm., 2008) who demonstrated that seeds germinate earlier
when the temperature is lower. Coombe (1956) claimed that seeds younger than six months
old never managed to geminate. The present and other studies indicate that germination is
possible within a shorter period of time. (Komosińska 2008) demonstrated that 19-week-old
seeds are able to germinate. It is not certain what the effect of habitat on the ability to
germinate is. Komosińska et al. (2006) pointed out that seeds from a floodplain ash-elm forest
had a shorter mean time of germination than a population from a ruderal site, i.e. railroad
tracks. Indeed, in the present study, plants from the floodplain forest germinated earlier and
62
the percentage was higher when compared to the oak-hornbeam forest population. Seeds on
the forest path were taken from the neighborhood of both forest communities and mixed
them. The participation of seeds from the floodplain forest and the effect of light on plants
setting seeds might have an impact on the results. The seeds used in the experiment probably
differed in size and ripeness but as Trepl (1984) revealed in his experiments, the size of seeds
did not indicate any differences in germination. Thus, the differences among habitats are
caused by the environmental factors that are associated with the sites. Both Jończyk (2007)
and Komosińska (2008) believed that due to the maternal effect in the first generation in small
balsam, traits including germination features that are inherited from maternal plants that help
in their adaptations to new environments. In the present study germination was only
performed using seeds that had been directly harvested from habitats that were studied and
therefore a possible maternal effect could have occurred.
3.2.2. Morphological variation of individuals
The individuals that were studied had shoots of different heights ranging from less
than 10 cm to 125 cm. The shoot heights of the majority of individuals were between 10 and
25 cm. Distribution of this trait was right-skewed and small plants were dominant (Fig. 15A).
The distribution of the length of the longest leave was also somewhat right-skewed (Fig. 15B)
although there was no real deviation from the normal distribution. A similar pattern was
observed in the case of the width of the longest leaf, which was normally distributed
(D’Agostino test, p=0.11) (Fig 15C). Both the number of flowers and fruits were similar (Fig.
16AB). Among the individuals that were measured, ca. 20% had no flowers at all and 16.7%
did not develop fruits. The majority of plants had fewer than 20 flowers and 20 fruits per
plant, which amounted to ca. 85% and 90%, respectively.
Under the conditions of deep forest interiors, the height of plants is much smaller than in open
habitats such as forest edges and forest paths, which has been demonstrated in many studies
(Coombe 1956; Trepl 1984; Eliáš 1992, 1999; Chmura 2008a; Kujawa-Pawlaczyk 1991;
Klimko, Piskorz 2003; Piskorz 2005; Dobravolskaite 2012). Both the percentage of flowering
plants and the height of the generative specimens of I. parviflora in forest interiors were small
(Fig. 15A, 16A).
63
Fig. 15. Histogram of plant height (A), length of leaves (B) and width of leaves (C) with fit of
normal distribution (line) based on measurements of 415 individuals
Fig. 16. Histogram of the number of flowers (A) and the total number of fruits (B) based on
measurements of 415 individuals growing on mineral soil in forest interiors
64
Individuals in the blooming phase reached 10-20 cm most frequently, which was
mentioned earlier; however, some individuals can even grow up to ca. 140 cm. The frequency
distribution was an inverse J-shape. Eliáš (1992) found a J-shape of the frequency distribution
for a monospecific stand of small balsam on a clearing along a forest edge on which 60% of
the tallest plants (from the upper class of height) prevailed. This diversity might be
a consequence of a shift in phenology or differences in environmental factors (nutrients in the
substratum, biotic interactions and light conditions). Nevertheless, these results are evidence
of a high intrapopulation variation in plant traits in the species. Both the length and width of
the longest leaf showed a more normal distribution. These two parameters are associated with
leaf area, i.e. they indirectly reflect the photosynthetic ability of a plant. It can be assumed
that light conditions with the exception of gaps in the canopy are of a normal distribution and
therefore this is manifested in a variation in leaf parameters.
Coombe (1956) wrote that taller plants have more fruits and flowers. In this study nonflowering and non-fruiting plants were in the majority. On the other hand, single individuals
had 100-200 seeds per a plant.
The results of Dobravolskaite (2012) suggest that the growth of I. parviflora is
continuous over time and that it can be expected that the number of generative organs
corresponds to the size of plants. However, taking biomass into account like Eliáš (1992) did,
the main pattern could be different. It turned out that the tallest plants and those that were
dominant within a population of I. parviflora allocate more biomass in comparison with
codominant and suppressed individuals that are smaller in size. The approximate input of
biomass into flowers and fruits reached 9.4%, 7.4 and 6.2% for dominant, codominant and
suppressed specimens, respectively (Eliáš 1992).
The acidity of soils was positively correlated with the height of the shoots of Impatiens
parviflora, although it was a rather weak relationship (Tab. 8). The concentrations of calcium,
calcium carbonate, phosphorus and the cover of shrubs were positively correlated with the
height of plants.
Neither the length nor the width of leaves was significantly correlated with the parameters of
the soils and substratum The number of accompanying species was also not significantly
correlated with these parameters (Tab. 8). Only the content of phosphorus, the content of
floatable fraction and the cover of shrubs and herbaceous species on the forest floor were
positively significantly correlated with the leave traits. The content of total nitrogen and
sodium negatively correlated with the number of flowers, whereas total nitrogen positively
65
correlated with the number of fruits. Slope was negatively correlated in relation to the number
of fruits and flowers (Tab. 8).
Tab. 8. Spearman intercorrelation coefficients between the chosen plant traits and
environmental variables. Significance at p<0.05 (ns- non-significant)
pHaqua
pHKCl
C org [%]
C/N
Ca [mg/100g]
CaCO3
K [mg/100g]
LOI
Mg [mg/100g]
NT
Na [mg/100g]
P [mg/100g]
Floatable fraction [%]
litter depth
Slope
species richness
cover of herbs
cover of shrubs
cover of trees
stem height
0.15
0.14
ns
-0.17
0.11
0.16
ns
ns
ns
ns
ns
0.28
ns
ns
-0.19
-0.12
ns
0.31
ns
leaf length
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
0.12
0.13
ns
ns
ns
0.13
0.13
ns
leaf width
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
ns
0.13
ns
ns
ns
ns
0.13
0.14
ns
flowers
ns
ns
ns
ns
ns
ns
ns
ns
ns
-0.22
-0.12
ns
ns
ns
-0.15
ns
0.16
0.02
ns
fruits
0.28
0.30
0.16
ns
0.26
0.18
ns
0.15
ns
0.16
ns
0.23
-0.24
ns
-0.16
ns
-0.14
0.40
ns
The cover of herbs positively correlated with the number of flowers but negatively correlated
with the number of fruits. The number of fruits per plant revealed many other significant
correlations. Acidity, the concentration of organic carbon, calcium ions, calcium carbonate,
loss on ignition and phosphorus were all positively correlated although the percentage of
floatable fraction was negatively correlated (Tab. 8).
Slope (only for sites situated on hills) and species richness were negatively correlated with the
height of plants.
The remaining individuals from forest communities (oak-hornbeam forest, beechwoods and
mixed coniferous forest) did not differ significantly in the height of plants. Plants from mixed
coniferous forest sites were the smallest and the least productive in the development of
flowers, the number of flowers and the length and width of leaves (Fig. 17D). Individuals
from the oak-hornbeam forest and mixed coniferous forest had a similar number of fruits (Fig.
17E).
66
Fig. 17. Comparison of plant height (A) length of leaves (B), width of leaves (C), total
number of flowers (D) and total number of fruits (E) per plant in individuals growing on
mineral soil in forest interiors. The different letters above the box-whiskers indicate
significant differences at p<0.05 (Kruskal-Wallis test followed by Conover test). FA –
Fraxino-Alnetum, FP – forest path, DF – Dentario glandulosae-Fagetum, TC – TilioCarpinetum, QP – Querco roboris-Pinetum
67
Principal Components Analysis produced five (orthogonal not correlated) components that
explain 100% of the variation in the data. According to the results of PCA, the first
component accounted for 63% (Tab. 9). Loadings of the first component for the
morphometric variables that were used varied between 0.304 and 0.509. The highest value
was obtained for the width of leaves, thus this variable is responsible for the highest
percentage of variation in the analyzed data. Other important variables are the number of
fruits, the number of flowers and the height of plants, which all contributed to the loadings of
the following components 2, 3, 4 and 5, respectively.
Tab. 9. The variance and component loadings of particular variables based on measurements
of 415 individuals of Impatiens parviflora
Proportion of variance
Cumulative proportion
Flowers
Fruits
Leaf length
Leaf width
Height of stem
Comp.1 Comp.2 Comp.3 Comp.4 Comp.5
0.603
0.183
0.141
0.061
0.013
0.603
0.786
0.926
0.987
1.000
0.405
0.070
0.791 -0.453 -0.016
0.304
0.783 -0.421 -0.342 -0.018
0.716
0.509 -0.375 -0.274 -0.107
0.499 -0.411 -0.292 -0.103 -0.697
0.485
0.268
0.192
0.810 -0.010
The height of plants usually is associated with enrichment with nutrients. In this study
a higher content of available phosphorus, calcium and calcium carbonate influenced the size
of plants. Dobravolskaite (2012) revealed that individuals of I. parviflora that were growing
on humus-rich soil at the edge of a pine forest were the tallest.
The shortest plants were found in the soils that were poorest in terms of nutrients and low pH
on sites of the Picea abies community. Thus, the results in the present study are in accordance
with those that were obtained by Dobravolskaite (2012).
In a previous work by Chmura (2008a), individuals representing populations from different
forest and non-forest plant communities (calcareous grassland) and across plant communities
that occurred in various habitats (forest interiors, paths, margins, dead logs) were analyzed in
regard to the same morphological traits. A greater variation was observed in all of the
parameters. The tallest plants were found in a contact zone between an oak-hornbeam and
floodplain forest on bare (unvegetated) soil with thick litter. These individuals had many
branches of the 3rd order and adventive above-ground roots. The investigated population
increased in density but mean height decreased after three years of the study (Chmura npbl).
The shortest flowering plants grew on a very thin layer of soil (up to 2 cm) on a limestone
68
outcrop. It seemed that the thickness of the soil, temperature and full sun conditions were
limiting factor for the growth of plants. In contrast, Vervoort and Jacquemart (2011) observed
flowering individuals growing in river beds where the water table was 1-2 cm above the root
collar. This not only proves very high morphological plasticity but also a very high
morphological adaptation to extreme environmental conditions.
Most studies that investigate variations in the morphometric traits of I. parviflora focus on
differences among the types of habitats within the gradient of naturalness/disturbance, e.g.,
road-forest borders (Kujawa-Pawlaczyk 1991; Klimko, Piskorz 2003), forest fringe
communities (Uherčiková, Eliáš 1987; Eliáš 1992), two forest communities and forest edge
(Dobravolskaite 2012) or forest communities and their contact zone with another one (Piskorz
2005). These case studies mainly considered only the size of plants and seldom examined any
additional morphometric properties (Piskorz 2005; Dobravolskaite 2012) or plant architecture
and biomass allocation (Eliáš 1992). Dobravolskaite (2012) did not find any considerable
differences in plant height, plant height up to the first inflorescence or to the first lateral
branch in two forest communities – a spruce and a pine forest. Only individuals from the edge
of the pine forest had higher values of the parameters. In the present study the four forest
communities that were analyzed are much more diversified with respect to habitat properties
and structure and plant composition and therefore the differences in the selected plant traits
that were observed were significant (Fig. 17). The highest variation and the highest outliers in
the height of plants that were growing along a forest path can be explained by the border
effect. This phenomenon is known in Impatiens noli-tangere, which grows more abundantly
in the ecotone between a forest and meadow. The border effect can be manifested by larger
plants and higher fecundity (Falińska 2004). There are diverse biotopic conditions along
forest paths. Light conditions are poorer on sites that are closer to a dense forest than on more
open sites. In each case individuals from the Fraxino-Alnetum community had higher values
of the plant traits that were studied. Floodplain forests (Ulmenion alliance) are prone to
invasion due to the sufficient amount of light in the forest undergrowth as well as nutrientrich soil (Petrášová et al. 2013). Knowing the biotopic requirements of the species, it is not
surprising that small balsam achieved the highest degree of robustness in the patches of this
type of forest. In contrast to that forest, Querco roboris-Pinetum plants were the smallest in
a mixed coniferous forest. The research by Trepl (1984) clarified that I. parviflora prefers
broad-leaved forests than mixed and coniferous forests. Unfavorable habitat conditions hinder
penetration by this species. It can be presumed that the same conditions that influence the
frequency and abundance of the species also have an impact on plant traits. The history of
69
I. parviflora invasion into Poland was clarified in great detail in the Primeval Białowieża
Forest (Kujawa-Pawlaczyk 1991; Adamowski, Keczyński 1991; Faliński 1998b). It was
reported that I. parviflora only naturalized in deciduous forests, i.e. oak-hornbeam and
beechwood. The history of the establishment and spread of the species into the Jurassic
Upland is not known although it can be assumed that mixed coniferous forests were colonized
later than deciduous forests. The differences between oak-hornbeam forest Tilio-Carpinetum
and beechwood Dentario glanduloase-Fagetum have not been identified except for the
number of fruits, which might be associated with differences in the phenology of these
phytocoenoses.
The variations in the selected plant properties among the four forest communities that had
been distinguished are rather stable because the same variables were used by Chmura and
Gucwa-Przepióra (2012) even though individuals that were measured for other purposes had
similar results.
3.2.3. Seasonal dynamics of Impatiens parviflora
All of the phenophases that had been distinguished were observed in the forest habitats
that were analyzed, which is shown on Figures 18-22. They differ in the contribution of
almost all of the phenophases with the exception of the preflowering phase (stadium with
a presence of flowers buds) (Tab. 10).
Tab. 10. Results of the comparison of the participation of phenophases between the forest
habitats that had been distinguished (G-test, ns – non-significant)
Phenophase
Time
G
df
p
two-cotyledon phase
16.04-30.05
45.48
12
0.000009
one-cotyledon phase
30.04-30.06
55.38
12
0.0000002
vegetative phase
17.05-22.09
60.30
32
0.0018
preflowering phase
17.05-30.08
27.39
28
ns
flowering phase
17.05-30.08
65.74
28
0.00007
prefruiting phase
17.05-30.08
45.56
28
0.019
fruiting phase
17.05-30.09
83.84
36
0.00001
subsenile phase
17.07-30.09
33.11
20
0.03
senile phase
14.06-30.09
47.39
28
0.012
70
Plants with one or two cotyledons were present until the end of June. Plants with only one
cotyledon appeared in the second half of May. At the end of May, the contribution of twocotyledon individuals dropped to less than 20% (Fig. 18). The first fully developed vegetative
plants appeared at the end of May but mid-June was optimum for this phenophase. Plants
were in the blooming phase from the second half of June until the end of August which was
also optimum for the fruiting phase. In beechwood community, the two-cotyledon phase
lasted from mid-April until the end of May when single plants with two cotyledons were
present (Fig. 19). The highest decrease occurred between mid-May and the end of May when
the participation of this phenophase fell from 80% to around 10%. The last plants with single
cotyledons were recorded at the end of June. The first individuals that had flowers were found
in mid-July; however, the first plants with buds were noted a month earlier. Plants were in the
fruiting phase from second half of August until the end of month. The first signals of
senescence also appeared during that time frame.
Fig. 18. Participation of phenophases and the survival of individuals of Impatiens parviflora
in the Tilio-Carpineum forest community. Abbreviations: 1 – two-cotyledon phase, 2 – onecotyledon phase, 3 – vegetative phase (stems without flower buds), 4 – preflowering phase
(flower buds present), 5 – flowering phase (open flowers present), 6 – prefruiting phase (with
unripe fruits), 7 – fruiting phase (with ripe fruits), 8 – subsenile phase (with declined leaves),
9 – senile phase (stems without leaves)
71
Fig. 19. Participation of phenophases and the survival of individuals of Impatiens parviflora
in the Dentario glandulosae-Fagetum forest community. Abbreviations: 1 – two-cotyledon
phase, 2 – one-cotyledon phase, 3 – vegetative phase (stems without flower buds), 4 –
preflowering phase (flower buds present), 5 – flowering phase (open flowers present), 6 –
prefruiting phase (with unripe fruits), 7 – fruiting phase (with ripe fruits), 8 – subsenile phase
(with declined leaves), 9 – senile phase (stems without leaves)
Fig. 20. Participation of phenophases and the survival of individuals of Impatiens parviflora
in the Fraxino-Alnetum forest community. Abbreviations: 1 – two-cotyledon phase, 2 – onecotyledon phase, 3 – vegetative phase (stems without flower buds), 4 – preflowering phase
(flower buds present), 5 – flowering phase (open flowers present), 6 – prefruiting phase (with
unripe fruits), 7 – fruiting phase (with ripe fruits), 8 – subsenile phase (with declined leaves),
9 – senile phase (stems without leaves)
72
During the study 80% of plants died out from mid-April until the end of September (Fig. 19).
The first single senile plants appeared in mid-June, despite the fact that no subsenile
phenophase had been recorded prior to that date. Almost 30% of individuals survived until the
end of the research (Fig. 19). Only two cotyledons individuals were present at the beginning
of the research but half a month later one plant in the one-cotyledon phase was observed. The
optimum for this phenophase was at the end of May. The contribution of the vegetative phase
was similar to the plants in the Tilio-Carpinetum community. The participation of plant that
represented the preflowering and flowering phases differed in the optimum periods from the
other plant communities. These plants were flowering over a period of two months from June
30 until August 30. The optimum fruiting phase occurred in August and September. The first
subsenile plants were observed in mid-July and the highest percentage occurred at the end of
August. The survival percentage amounted to almost 36%. The first plants of one cotyledon
phenophase after the two-cotyledon phase appeared at the end of April in the mixed
coniferous forests (Fig. 21). The optimum time for this phenophase was observed at the end
of May. In mid-June ca. 50% of plants represented the vegetative phenophase. The
preflowering phenophase lasted quite a long time from mid-June until the end of August.
Fig. 21. Participation of phenophases and the survival of individuals of Impatiens parviflora
in the Querco roboris-Pinetum forest community. Abbreviations: 1 – two-cotyledon phase, 2
– one-cotyledon phase, 3 – vegetative phase (stems without flower buds), 4 – preflowering
phase (flower buds present), 5 – flowering phase (open flowers present), 6 – prefruiting phase
(with unripe fruits), 7 – fruiting phase (with ripe fruits), 8 – subsenile phase (with declined
leaves), 9 – senile phase (stems without leaves)
73
Fig. 22. Participation of phenophases and the survival of individuals of Impatiens parviflora
in the forest edge community. Abbreviations: 1 – two-cotyledon phase, 2 – one-cotyledon
phase, 3 – vegetative phase (stems without flower buds), 4 – preflowering phase (flower buds
present), 5 – flowering phase (open flowers present), 6 – prefruiting phase (with unripe
fruits), 7 – fruiting phase (with ripe fruits), 8 – subsenile phase (with declined leaves), 9 –
senile phase (stems without leaves)
Individuals in the fruiting phenophase existed until the end of the study and its optimum
period was the end of August. An early process of senescence was observed beginning in the
second half of June. Twenty percent of the labeled plants survived until the end of
observation.
The individuals of I. parviflora that represented the first two phenophases (one- and twocotyledon phase) in the forest edge community were present until the end of June (Fig. 22).
The vegetative phenophase lasted for quite long time from mid-May until the third week of
September. The preflowering phase also lasted for a long time but it ended at the end of
August, and the optimum period for this phenophase was at the end of July. The flowering
and prefruiting phenophases lasted for the same period of time. The fruiting individuals were
present from the second half of May until the end of the research. The period of the subsenile
and senile phenophases occurred simultaneously but they differed in the optimum time period
(Fig. 22). The percentage of plants that survived dropped to 34%.
Despite the fact that only 12 time series were taken into account, the results of the
observations showed significant differences (Tab. 10). This variation is manifested by
different optimum periods rather than the time that particular phenophases begin and end. The
biggest differences are in the generative phenophases (preflowering, flowering, prefruiting
74
and fruiting phases). The plants on the forest fringe sites entered into these phases faster than
those in the remaining forest communities.
It is known that there are significant differences in the phenology of I. parviflora among
different geographical regions. For instance, seeds germinate not sooner than in May in
Finland (Erkamo 1952 after Piskorz and Klimko 2002), but in Great Britain seeds are already
germinating in March (Coombe 1956). Similar phenological observations, which were the
inspiration for present study, were conducted by Piskorz and Klimko (2002) in Wielkopolski
National Park. The only differences are related to the number of visits that were made to
record the observations and types of forest habitats. Their study was conducted in an oakhornbeam forest Galio silvatici-Carpinetum typicum that resembles Tilio-Carpinetum
ecologically. Both plant associations are biogeographical variants of the same forest
community (Matuszkiewicz 2011). Thus, they can be compared in regard to the contribution
of particular phenophases of I. parviflora although the authors stressed that their phenological
study was conducted in a thinned forest and they also mentioned that the population that was
studied was infected by Puccinia komarovii – a rust and a natural enemy of small balsam.
Moreover, in the present study observations in five habitats were performed in 12 time series
while the Piskorz and Klimko (2002) study was conducted every 3-7 days in one community.
Nevertheless, some similarities and differences can be observed. First of all, phenophases
were present and the time of the beginning of the flowering phase was quite fast. In the
present study, it occurred at the end of June but when taking into account the longer breaks
between measurements, it can be presumed that the start of flowering phenophase may have
occurred earlier. The time of flowering was quite long and lasted until the end of September
while the fruiting phase was also quite long and lasted about two months. The phenophases in
the beechwood forest were shifted in comparison with the oak-hornbeam forest probably due
to its denser tree canopy cover. For this reason, i.e. light availability, most of the differences
are between the forest interiors of the investigated forest communities vs the forest path. The
first fruiting individuals were found in the second half of May and the earliest senile
individuals were recorded at the end of May. Pilkova (2013) compared the participation of
phenophases between a forest interior and a forest glade. She noted that the time of particular
phenophases was prolonged by a month in an area that had been cut. From the viewpoint of
the survival of a population, the most important factor in phenology is the time of special
developmental phases. Trepl (1984) believed that such a long period of flowering and fruiting
is one of key elements in the invasion success of the species. Several populations that differ in
phenology including an extended period of flowering and fruiting have an increased chance of
75
the persistence of the species when they coexistence in the gradient of a forest interior – forest
fringe or forest interior canopy gaps in a specific area.
3.2.4. Life history modifications that are dependent on habitat heterogeneity
Spatial patterns in life history traits
According to the Principal Components Analysis, the first component (axis 1) explains
nearly 42% of the variation, whereas the second axis explains ca. 22%. The number of
flowers and the height of plants are correlated with the first axis and the number of fruits and
the number of flower buds per plant correlate with the second axis. The presence of
cotyledons only plays a minor role. PCA showed that some individuals that represented
various microhabitats differ from other plants, but the majority of individuals that were
analyzed are situated close together on the diagram (Fig. 23).
Fig. 23. Arrangement of individuals of I. parviflora along the two axes of the Principal
Components Analysis on the basis of the selected morphometric traits and the chosen types of
microhabitats
76
The variation within particular microhabitats is quite large when compared to the other types.
Plants that are growing in logs under canopy openings; forest paths; canopy openings and
stumps are highly variable. The plant traits that differentiated individuals of I. parviflora the
most were the height of a plant and the number of flower buds (Tab. 11). The number of
flowers correlated with the size of a plant (Fig. 23).
Tab. 11. The component loadings of particular plant traits
Proportion of Variance
Cumulative Proportion
number of flowers
number of fruits
number of flower buds
cotyledons
stem height
Comp.1 Comp.2
0.419
0.217
0.419
0.636
-0.509
0.124
-0.523
0.262
-0.244
-0.744
0.090
-0.591
-0.113
-0.632
Comp.3
0.189
0.825
-0.097
-0.311
0.505
-0.799
0.027
Comp.4 Comp.5
0.128
0.047
0.953
1.000
-0.257
0.806
-0.565 -0.492
-0.092 -0.352
0.060
0.026
-0.138
0.753
The plants differ in mean values among the microhabitats that had been distinguished,
which is shown in Table 12. The tallest individuals were found under canopy openings, on
treefall disturbances and forest paths, whereas the smallest plants in terms of the height of
stem grew on dead logs under a canopy of trees (Tab. 12). The highest number of flower buds
was reported for plants that were growing in a canopy opening, on treefall disturbances and in
areas near dead logs. Plants that were growing in forest interiors, dead logs and stumps had
a lower number of flower buds on average.
Tab. 12. Comparison of plant traits (Mean±SD) among the selected types of habitats. Different
letter near values in a column indicate significant differences in the medians (Kruskal-Wallis
test followed by Conover test)
n
Habitat
65
50
60
20
50
50
50
30
50
50
40
Forest interior
Canopy opening
Area near log
Log under canopy
Log under canopy opening
Root plate
Treefall disturbance
Hollow
Path
Root collar
Stump
Height of stem
Number of
Number
flower buds of flowers
13.9±10.0
32.7±15.5
24.9±13.0
10.9±4.8
23.9±26.1
14.5±7.0
30.7±12.9
20.6±12.2
32.9±17.4
18.8±8.4
28.8±21.9
0.8±1.3
3.2±4.3
2.5±2.0
0.8±1.2
0.1±0.2
1.3±1.3
2.9±2.8
1.0±1.7
1.8±2.4
1.1±1.3
0.9±1.3
fg
a
bc
g
ef
fg
a
cde
ab
de
cd
77
c
a
a
c
d
bc
a
bc
b
bc
c
0.3±0.5
1.8±3.2
0.6±1.6
0.1±0.2
1.5±2.8
0.5±0.6
1.5±1.1
0.5±0.6
1.6±1.5
0.2±0.4
3.2±6.2
efg
ab
ef
g
cd
de
a
de
a
fg
bc
Number of
fruits
0.6±1.6
2.7±3.9
0.2±0.6
0.1±0.2
5.8±10.6
0.1±0.1
2.3±2.4
0.8±1.4
5.1±5.1
0.1±0.5
1.21.8
ef
bc
fg
fg
b
g
b
de
a
fg
cd
Presence
of
cotyledons
%
16.9
12.0
1.7
0.0
0.0
4.0
4.0
33.3
0.0
10.0
0.0
The pattern for the number of flower per plant is not congruent with previous results.
Generally, the number of flowers was lower in all types of microhabitats. The highest was
observed in stumps but with a high degree of variation (Tab. 12). Other microhabitats with the
highest number of flowers in I. parviflora were: canopy openings, forest paths and dead logs
under canopy openings. The plants with a large number of flowers per plant were the least
abundant in dead logs under canopies, in forest interiors and on root collars (Tab. 12).
The plants with the highest number of fruits per plant occurred on the forest paths followed by
on treefall disturbances, dead logs under canopy openings, whereas plants with the lowest
number of fruits were recorded in such microhabitats as dead logs under canopy openings, on
root collars and areas near dead logs (Tab. 12).
The habitats also differed in relation to the presence of cotyledons in plants (χ 2= 58,
p<0.0001). The highest percentage was observed in individuals that were growing in hollows
followed by those that were growing in forest interiors, under canopy openings and on root
collars (Tab. 12).
The tallest plants were associated with well-lit sites such as paths, treefall disturbances and
canopy openings. It is already known that I. parviflora grows better and in large masses in
such places (Coombe 1956; Trepl 1984; Eliaś 1999). A recent study by Nowińska (2010)
showed that small balsam, even when it is equally frequent under canopies and gaps in
forests, scores a higher cover under gaps. It is not surprising that individuals that occur under
canopies and on soil were relatively small. The plants that were growing on substratums other
than soil, i.e. bark or decaying wood, were strange and atypical. Some of them such as those
that were growing on stumps and on dead logs but with good light conditions were taller than
plants in forest interiors. The latter still had at least one cotyledon. Only plants that were
growing in hollows in trees had cotyledons more frequently. A comparison of plants from
various microhabitats suggests that the specimens also differed in phenology. In the literature
there are some reports of the colonization of dead wood by small balsam (Piskorz, Klimko
2001; Chmura 2008c; Nowińska et al. 2009; Staniaszek-Kik, Żarnowiec 2013). Previous
studies analyzed the density and height of plants on various microsites in the vicinity of
uprooted hornbeam Carpinus betulus and spruce Picea abies trees.
In the nature reserves of the Wielkopolski National Park, the mean height of plants
that were found on dead logs was between ca. 20-55 cm and the tallest individuals reached ca.
90 cm (Piskorz and Klimko 2001). Thus, they were markedly taller than the specimens of
I. parviflora that were observed on fallen beech trees in the present study. Piskorz and Klimko
(2001) also gave the heights of other plants that occurred in the remaining structures near
78
treefalls. The plants in that study were of similar sizes as the individuals in the present study
(between 20 cm-55 cm). It is possible that plants that were growing on fallen trees were
measured close to root plate of uprooted trees in the Wielkopolski National Park. The authors
wrote about sections of fallen trees from both the side of the trunk or root collar and at
a farther distance from the root plate and these areas are sometimes covered by remnants of
soil. When the amount of soil is quite large, it forms suitable conditions for plants to develop.
A Redundancy Analysis RDA (Fig. 24) of plant traits of Impatiens parviflora from the
microhabitats that were studied and the environmental variables showed that light availability
followed by ratio of nitrogen and carbon are the most important explanatory factors. Other
important significant variables are the content of magnesium, sodium, calcium and
phosphorus (Tab. 13). The light influences the height of plants and the number of flowers per
plant and indirectly and more weakly the number of fruits. The morphometric variables such
as the presence of cotyledons and the number of flower buds do not seem to be strongly
affected by abiotic variables.
Fig. 24. Relations between plant traits and selected environmental variables based on RDA
79
Tab. 13. Summary of the Monte Carlo test of RDA based on
I. parviflora and selected environmental variables
Variable
P
Light
0.001
C/N
0.010
Ca
ns
CaCO3
0.037
LOI
ns
Mg
0.006
Na
0.008
P
0.042
Total cover of native species
ns
K
ns
the morphometric traits of
F
52.96
5.92
2.69
4.27
1.16
8.75
7.21
3.37
0.39
0.70
This analysis also confirmed the well-recognized fact that light conditions (Coombe 1956;
Trepl 1984) are the most important factor for the growth of small balsam. It has an impact on
all of the parameters except the presence of cotyledons (Fig. 24). What is interesting is that
the content of nutrients is correlated with the cover of native plants, which suggests that in
these microhabitats nutrients mainly influence native species and small balsam is not as
dependent and has low biotopic requirements.
Temporal changes in habitat occupancy and association with plant traits
Taking into account changes over the course of time and variations among the
microhabitats, statistical analyses revealed many significant differences. Seedlings of
I. parviflora that were growing in various microhabitats differed in survival during the time of
the study (G=297.8, df=40, p<0.0001). From mid-April until mid-May, 100% of the seedlings
in dead logs under canopy openings, on root plates, treefall disturbances and on root collars
and stumps survived. During the last time series, i.e. in the second half of September, 70%
and 60% in canopy openings on soil and forest paths survived, respectively. No plants were
recorded in habitats such as dead logs under canopies, treefall disturbances and hollows (Fig.
25) in the last time series. The mean survival percentage of seedlings of I. parviflora was
ranked as follows: forest paths (88.4%), area near dead logs (88.0%), canopy openings
(86.0%), forest interiors (82.6%), root plates (71%), hollows (68.6%), root collars (60.8%),
stumps (65%), logs under canopy openings (53.1%), treefall disturbances (38.6%) and logs
under canopies (32.5%).
The mean number of cotyledons also differed among the habitats (Kruskal-Wallis, χ 2=
30.75, df = 10, p< = 0.001) and among the time series (Kruskal-Wallis χ 2= 529.47, df = 5, p<
80
= 0.001). In the second half of April, all of the plants had two cotyledons except for those on
forest paths where single individuals with one cotyledon were found but these did not survive
until the next time series (Fig. 26). The mean number of cotyledons was estimated at up to
two until the second half of May in the following microhabitats: forest interiors, canopy
openings, areas near logs, root plates, treefall disturbances, hollows and root collars. In the
remaining microhabitats there were single plants that had only one cotyledon, thus the mean
values were more than 1.5 cotyledons per plant. In forest interiors, canopy openings and
hollows, no individuals with at least one cotyledon were observed in the second half of June;
however, in the remaining microhabitats, plants with cotyledons were still observed.
Individuals with one cotyledon were noted along forest paths until the second half of July.
The seedling height varied greatly during the time of the study and among the habitats
(Tab. 14). Significant differences occurred in each of the time series among microhabitats
(Tab. 15).
In the second half of April seedlings of I. parviflora were tallest along forest
paths and in areas near dead logs and on average reached 5.9 and 4.3 cm, respectively. In midMay the tallest plants were found on dead logs that were growing under canopies – ca. 12 cm,
followed by plants that were growing near tree root collars, treefall disturbances, hollows and
stumps (Fig. 27). In mid-June the tallest plants of small balsam were observed in hollows (ca.
27 cm), on root collars and in canopy openings. Because some seedlings died, the mean
height of individuals on dead logs under canopies decreased to ca. 6 cm – the lowest value
that was recorded among the microhabitats that had been distinguished (Fig. 27). The tallest
plants on stumps were observed in July, August and September and reached 39-45 cm. From
mid-August until September a significant increase in the mean height of plants was noted for
forest interiors and canopy openings which finally reached ca. 34 cm (Fig. 27). There were no
significant differences in the mean height of plants among the microhabitats in September
(Tab. 15).
The number of leaves that were related time and the variations among
microhabitats showed a trend that was similar to that of the mean height of plants (Fig. 28).
Differences among microhabitats were noted for the second (the second half of May) and
third time series (the second half of June). The first leaves were only observed in May. The
mean number of leaves per plant, from the highest to the lowest, is ranked as follows: stumps
(11.3), hollows (10.6), root collars (8.7), areas near logs (7.6), forest paths (7.2), root plates
(6.1), forest interiors and canopy openings (5.8), logs under canopy openings (5.7), treefall
disturbances (5.0) and logs under canopies (4.9).
81
The first flower buds in individuals of I. parviflora only appeared in mid-June.
In July plants differed significantly in the mean number of flower buds among the
microhabitats (Tab. 14-15). The highest number was observed on plants that were growing on
stumps and the lowest on plants in treefall disturbances (Fig. 29) during this period. In August
plants in hollows had the highest mean number of flower buds. The mean number of flower
buds decreased in all of the habitats in September.
The mean number of flowers per plant differed significantly only in June when
the first flowers developed (Tab. 14-15). Generally, the number of flowers was not high in
any of the microhabitats during the time of the study. The highest was noted for plants that
were growing in stumps and hollows (Fig. 30).
Single fruits on plants were observed for the first time in June (Fig. 31). The
plants that were growing near root collars, stumps and on root plates had a significantly
higher number of fruits (Tab. 14-15, Fig. 31) in comparison with the other types of
microhabitats. Individuals of small balsam that were growing in treefall disturbances or on
dead logs under canopies developed fruits singly and only in July. These plants usually died
out faster than in the other types of microhabitats.
Similar plant history traits were analyzed in temporal and spatial research. The height of
adult plants was a factor for spatial variation, whereas for changes in time, the height of
seedlings was analyzed until their death. The number of leaves was considered to be
important as well as the mean number of cotyledons, which obviously varies from zero to
two, when examining the development of individuals. Only the presence of cotyledons was
important for the spatial research. The above-mentioned life history traits such as plant height,
the number of flowers and fruits are studied very often. The first one is a measure of the
gregariousness of plants and the next two indirectly indicate reproductive potential. The
changes in density that were observed were similar to those that were demonstrated by
Kujawa-Pawlaczyk (1991) with a peak in the abundance of plants in June and July; however,
due to method that was applied, individuals that emerged later were not included in the
observations.
The highest seedling mortality on dead logs or stumps without a layer of normally
developed soil is determined by unfavorable abiotic conditions. It was shown that the
mortality of seedlings for the native congener I. noli-tangere is caused by moisture and light
conditions and its neighbors (Falińska 2004). In the present study the lack of moisture
appeared to be the most crucial factor. The exceptions were plants that were growing in
treefall disturbances, which died out quite quickly. The highest density of plants was observed
82
on these sites, which could lead to higher intraspecific competition. All of the observations in
relation to phenology and differences in morphometric traits were done on labeled
individuals. Thus, only one cohort of plants was analyzed in the present study. The model of
seasonal dynamics of one or more cohorts was elaborated on by Symonides (1988a).
Germination and seedling emergence usually take place in a short time and the development
of particular stages of the life cycle run almost simultaneously. Obviously, differences among
individuals that are derived from various microhabitats are more considerable in disturbed and
nutrient-poor microcosms such as those in the present study.
The most important finding was that in almost all of the microcosms that were studied,
with the exception of plants that were growing on dead logs under canopies plants developed
fruits, i.e. on sites without a developed layer of soil and with poor light conditions. High
fecundity is a typical attribute for annual plants and is also the optimal model of actual
strategy. Seeding fruits is the only chance to survive and therefore it is obvious that plants
allocate biomass to the generative organs even under unfavorable conditions (Symonides
1988b).
Other important features that were not investigated in the present study are: selfpollination, which was quite well-recognized by Vervoort et al. (2011) and Piskorz (2005),
who studied the presence of cleistogamous flowers. Symonides (1988b) believes that
amphicarpy in annuals, i.e. the presence of both cleistogamous and chasmogamous flowers,
gives them an evolutionary advantage, especially in disturbed habitats.
83
Tab. 14. Effect of habitat and the time of measurement on the morphometric features of Impatiens parviflora seedlings (Kruskal-Wallis test)
Habitat
Time
Habitat:
series I
series II
series III
series IV
series V
series VI
height of seedlings
χ2
df P
27.09 10 0.002516
331.53 5 <0.0001
number of leaves
χ2
df P
20.0
10 0.0286
404.88 5 <0.0001
number of flower
buds
χ2
df P
21.21 10 0.01961
311.73 5 <0.0001
67.66
41.90
46.07
28.62
30.55
6.63
30.58
22.79
17.85
15.77
12.50
16.10
25.28
18.94
4.47
10
10
10
10
9
7
<0.0001
<0.0001
<0.0001
0.001434
0.0003532
ns
10
10
10
9
7
0.0006869
0.01155
ns
ns
ns
10
10
9
7
ns
0.004823
0.0257
ns
number of flowers
χ2
df P
28.32 10 0.001599
190.40 5 <0.0001
number of fruits
χ2
df P
37.05 10 <0.0001
268.89 5 <0.0001
37.00
13.87
14.04
5.85
22.51
30.39
8.78
10
10
9
7
<0.0001
ns
ns
ns
10
9
7
0.0127
0.0003751
ns
Abbreviations: series I: 16.04.07, series II: 18.05.07, series III: 14.06.07, series IV: 17.07.07, series V: 17.08.07, series VI: 22.09.2007
84
Tab. 15. Result of the post-hoc Conover test. Different letters in a column indicate significant differences at p<0.05. See Table 13 and Figures
nos. 27-31
number of
number of flower
number of
number of
height of seedlings
leaves
buds
flowers
fruits
series
1
2
3
4
5
2
3
4
5
3
4
5
Forest interior
de
g
cde abc de
cd
bc
abcd
abc
b
abcd
cde
Canopy opening
a defg a bc e
d
a
abcde
bc
b
de
de
Area near log
ab cdef bcd a abc bc
abc
ab
a
b
ab
abc
Log under canopy
e
a
e bc
a
c
cde
b
cde
Log under canopy opening de efg de bc de
ab
abc
de
c
b
bcde
cd
Root plate
e
fg abc a abc ab
a
abcd
ab
b
abc
ab
Treefall disturbance
bc abc bcd bc cde
a
c
e
abc
b
e
e
Hollow
cd abcd a
a abc ab
a
abcd
ab
a
abcd
cd
Forest Path
a bcde abc ab bcd
a
ab
abc
abc
a
abcd
ab
Root collar
cd ab
a
a
a
abc
ab
bcde
abc
b
bcde
a
Stump
e abcde ab a ab
ab
abc
a
a
b
a
a
Abbreviations: series 1: 16.04.07, series 2: 18.05.07, series 3: 14.06.07, series 4: 17.07.07, series 5: 17.08.07, series 6: 22.09.2007
85
Stump
Root collar
Forest Path
Hollow
Tree fall disturbance
Root plate
Log under canopy
opening
Log under canopy
Area near log
Canopy opening
Forest interior
Mean number of cotyledons
Stump
Root collar
Forest Path
Hollow
Tree fall disturbance
Root plate
Log under canopy opening
Log under canopy
Area near log
Canopy openning
Forest interior
Survival of seedlings [%]
100
90
80
70
60
50
40
30
20
10
0
18.05.07
14.06.07
17.07.07
17.08.07
22.09.07
Fig. 25. Percentage of survival of Impatiens parviflora seedlings over time in particular
microhabitats since 16.04.2007
2
1
16.04.07
18.05.07
0
14.06.07
17.07.07
Fig. 26. Comparison of the mean number of cotyledons in particular microhabitats over time
86
Stump
Root collar
Forest Path
Hollow
Tree fall disturbance
Root plate
Log under canopy opening
Log under canopy
Area near log
Canopy openning
Forest interior
Mean number of number of leaves
Stump
Root collar
Forest Path
Hollow
Tree fall disturbance
Root plate
Log under canopy opening
Log under canopy
Area near log
Canopy openning
Forest interior
Means of height of seedlings
50
45
40
35
30
25
20
15
10
5
0
16.04.07
18.05.07
14.06.07
17.07.07
17.08.07
22.09.07
Fig. 27. Effect of time and microhabitat on the height of Impatiens parviflora seedlings
20
18
16
18.05.07
14
14.06.07
12
17.07.07
10
17.08.07
8
22.09.07
6
4
2
0
Fig. 28. Effect of time and microhabitat on the number of leaves of Impatiens parviflora seedlings.
No leaves were recorded on 16.04.2007
Mean number of flower buds
7
14.06.07
6
17.07.07
5
17.08.07
4
22.09.07
3
2
1
Stump
Root collar
Forest Path
Hollow
Tree fall disturbance
Root plate
Log under canopy opening
Log under canopy
Area near log
Canopy openning
Forest interior
0
Fig. 29. Effect of time and microhabitat on the number of flower buds of Impatiens parviflora.
No flower buds were recorded on16.04.2007 and 18.05.2007
3
2
14.06.07
1
17.07.07
Stump
Root collar
Forest Path
Hollow
Tree fall disturbance
Log under canopy
Area near log
Canopy opening
Forest interior
0
Root plate
17.08.07
Log under canopy…
Mean number of flowers
4
22.09.07
Fig. 30. Effect of time and microhabitat on the number of flowers of Impatiens parviflora.
No flower buds were recorded on 16.04.2007 and 18.05.2007
87
10
8
6
Stump
Root collar
Forest Path
Hollow
Tree fall disturbance
17.08.07
Root plate
0
Log under canopy opening
17.07.07
Log under canopy
2
Area near log
14.06.07
Canopy openning
4
Forest interior
Mean number of fruits
12
22.09.07
Fig. 31. Effect of time and microhabitat on the number of fruits of Impatiens parviflora
Irrespective of the type of flowers, the majority of individuals from the microhabitats that were
studied survived until June and July and bore fruits. It turned out that those that had the highest
reproductive potential were plants that were growing where the soil layer was available (on root
collars and root plates where a remnant of soil is present. The tops of stumps seem to have
unsuitable conditions for I. parviflora. However, the structure of the wood enables them to hold
moisture, which is believed to significantly increase during wood decay (Bütler et al. 2007). Due to
numerous slits, fallen leaves and low amounts of decaying bark they can undergo decomposition as
well thereby beginning the process of the formation of humus can occur, which creates quite
favorable conditions for small balsam.
Impact of aspect, light and nutrients on the occurrence of species
Figure 32 shows the differences in the heights of plants of Impatiens parviflora on two
types of slopes and in various time series. There are significant differences between southernfacing and northern-facing slopes and the time of the measurements (one-month intervals).
Statistical analyses revealed significant differences in space and time (Tab. 16). A higher degree of
growth and the heights of the plants were generally detected on northern-facing slopes (Fig. 32).
It has been known for some time that small balsam grows more abundantly on northern-facing
slopes when compared to southern-facing slopes (Coombe 1956; Trepl 1984). Brothers and
Spingarn (1992) and Small & McCarthy (2003) proved that southern-facing slopes are
88
characterized by a higher species richness and a lower dominance and abundance of some species.
More favorable conditions, especially temperature and light conditions, create niches for more
species and therefore small balsam has a smaller ecological barrier, which is manifested by the
lower biocenotic resistance that it has to overcome on northern-facing slopes. Chmura (2008b)
demonstrated that northern-facing slopes are overgrown by herbaceous vegetation (including
I. parviflora and 11 other species) whose plants are taller on average. The present study shows in
detail how this aspect has an impact on the height of shoots in small balsam. Possible mechanism
of this phenomenon may include suppression by other plants. It was revealed that the density of
individuals of I. parviflora affects the size of plants even in a monoculture population (Coombe
1956).
b
a
a
a
d
bc
c
b
b
20
10
0
Height of plants [cm]
30
40
e
N-04 N-05 N-06 N-07 N-08 S-04 S-05 S-06 S-07 S-08
Code of group
Fig. 32. The comparison of height of Impatiens parviflora individuals due to time and aspect.
N- northern slopes, S- southern slopes, 04-08- successive months
Tab. 16. Influence of aspect and time on height of plant (Wilcoxon sum rank test and KruskalWallis test)
Czynnik
Time
Aspect
Time X aspect
Statistics
χ 2= 49.5
W = 1046
χ 2= 64.20
89
P
<0.0001
0.03101
<0.0001
An analysis of the presence/absence of small balsam in microhabitats that were associated
with living trees (sites close to root collars and hollows), snags (root collars, hollows and the tops
of broken trees) and dead log complexes (areas near dead logs, bark on logs, treefall disturbances
and root plates) demonstrated significant differences in the frequency of the species (G=56.59,
p<0.001).
The plants that were living in dead log complexes were present (60%) more than absent (40%) in
location in which such forms and types of substratum were available. Snags, i.e. both unbroken and
wind-broken dead trees, had more individuals of Impatiens parviflora close to their root collar than
100
living trees (Fig. 33).
60
40
0
20
percentage
80
absence
presence
living trees
snags
dead log complex
type of wood
Fig. 33. Association of Impatiens parviflora with dead logs complex and snags in comparison with
living standing trees
There is a negative medium but significant correlation between the distance from dead logs or
snags and the density of Impatiens parviflora individuals (Fig. 34).
90
80
rs = -0.51, p<0.0001
70
60
50
40
30
20
10
0
0
2
4
6
8
10
Fig. 34. The influence of distance from dead logs complex on density of Impatiens parviflora in
1m2 subplots (Spearman rank correlation test)
A study by Piskorz and Klimko (2001) revealed that Impatiens parviflora is an efficient colonizer
of microhabitats that are associated with fallen dead trees. They studied fallen logs, root plates and
holes in the ground. The results that were obtained are descriptive but without any statistical
analyses, although they seem to be congruent with the present study. For instance, they
demonstrated that with an increasing distance from the sites of tree falls the density of small
balsam showed a downward trend. In both that study and the present one such differences in height
can be attributed to the impact of light intensity. Some authors such as Hegi (1965), Zarzycki
(1984), Zarzycki et al. (2002) believe that I. parviflora is a sciophyte. Moreover, in laboratory
research, Hughes & Evans (1962) and Hughes (1965) revealed a decrease in density under full
sunlight although in another lab study, Whitelam and Johnson (1980) proved that small balsam can
behave like a shade-liking and open habitat species. Other authors who conducted research in the
field claimed that the species is rather a photophilous species (Kujawa-Pawlaczyk 1991; Piskorz,
Klimko 2001; Klimko, Piskorz 2003).
The analysis of the height of Impatiens parviflora plants in the four types of plots
confirmed the impact of light and the presence of dead wood on their height (Tab. 17). Only the
presence of dead wood does not significantly affect the size of plants. Post-hoc tests showed
significant differences among individuals from the control plots (ground layer under tree canopies)
and individuals from the other types of plots. There were no significant differences in the height of
plants that were growing in the vicinity of dead logs under canopies (CW) and the control (C).
Likewise, there were no differences between plants that were growing on plots with canopy
openings (O) and plants that were growing near dead logs under canopy openings (OW) (Fig. 35).
91
Moreover, plants that were growing near dead logs did not differ among the plots with or without
a tree canopy (CW vs. OW). The mean leaf area differed significantly between the control (C) and
canopies and dead wood plots (CW) vs canopy openings (O) and plots with dead wood under
canopy openings (OW) (Fig. 35), which was confirmed by the significance of the impact of light
availability (Tab. 18). The individuals of the second group were characterized by a larger leaf area.
The same situation concerns the biomass of leaves (Fig. 35, Tab. 18). When specific leaf area
(SLA) was considered, plants from the control plots had, on average, the lowest value of this
variable (Fig. 35, Tab. 19).
Tab. 17. Influence of dead wood and light availability on the height of plants (two-way ANOVA)
Czynnik
Df
1
1
1
146
Canopy openings
Dead wood
Canopy openings x dead wood
Residuals
F
45.7962
0.0775
14.9858
P
<0.0001
ns
0.000163
Table 20 presents the differences in the mean content of chemical elements that are responsible for
the trophy of soils. There was a significantly lower concentration of potassium in the control and
the plots with canopy opening as well as those with dead wood under canopy openings (Tab. 20).
A significantly higher content of phosphorus was observed in both types of dead wood plots. There
were no significant differences with respect to the concentration of magnesium due to high degree
of variation within the study plots (Tab. 20). The concentration of total nitrogen was similar in all
of the plots that had been distinguished (Tab. 20).
Tab. 18. The influence of presence of dead wood and canopy openings on leaf area (LA) and leaf
biomass (LB) (two-way ANOVA)
LA
Canopy openings
Dead wood
Canopy openings x dead wood
Residuals
df
F
P
1 78.712 <0.0001
1 0.4075
ns
1 0.0598
ns
72
92
LB
df
1
1
1
72
F
54.8198
3.7968
0.1772
P
<0.0001
0.05525(ns)
ns
a
ab
b
b
a
a
C
CW
O
OW
C
CW
O
OW
b
b
a
a
b
ab
ab
a
C
CW
O
OW
C
CW
O
OW
40
bc
30
10
20
LA
25
15
0
0 100
300
SLA
0.04
0.00
LB
0.08
500
0
5
Plant height
35
c
Fig. 35. Comparison of plant height (cm), leaf area LA (cm2), specific leaf area (SLA), leaf
biomass LB (g). between the types of plots. Abbreviations: C – canopy, CW – canopy and dead
wood, O – openings OW – openings and dead wood present
Tab. 19. The influence of the presence of dead wood and canopy openings on specific leaf area
(SLA). (Wilcoxon sum rank test and Kruskal-Wallis test)
Variable
Opening
Wood
Opening x Wood
statistics
W=542
W=485
χ 2= 8.48
93
P
0.06(ns)
0.025
0.036
Tab. 20. Comparison of chosen physical-chemical parameters of soils between the types of plots
that had been distinguished (Kruskal-Wallis test followed by Conover test). The same letter or no
letters in a column mean non-significance of differences at p<0.05
Canopy and dead wood
Control
Canopy openings
Openings and dead wood
K
34.4±48.8 a
9.9±5.4 b
9.6±2.6 b
14.0±10.5 b
P
20.7±37.0 a
5.2±6.4 b
7.5±5.0 b
11.7±16.3 a
Mg
30.2±59.4
10.8±11.5
13.6±12.0
15.0±19.8
Nt
0.3±0.1
0.5±0.4
0.3±0.2
0.3±0.1
The results of the analysis of height among the four habitats that had been distinguished are in
partial agreement with a greenhouse experiment by Elemans (2004), who proved that with a 66%
level of full light and the highest level of nutrients equivalent to 300 kg N ha–1 per year, plants had
the highest biomass that reached 10 g of the total biomass per plant. If it is assumed that the height
of individuals of Impatiens parviflora is correlated with its biomass then the results of this study
show a similar trend. Plants on plots with canopy openings were the tallest but not significantly
taller than those from plots with canopy openings with dead wood. Under natural conditions the
highest light intensities are at sites where the falls of trees have caused canopy gaps. As a result,
plants that were growing near dead logs under canopies were taller than those that were growing on
mineral soil under canopies. Thus, the trend is similar to the result of Elemans (2004) who showed
that the total biomass of plants was higher at a high dose of nutrients and was at 2% when
compared to low dose of nutrients with the same amount of light.
No significant interaction of light and the fertility of a habitat for SLA was observed, although the
highest value was recorded for sites under canopy openings (O) and near dead logs (CW, OW)
(Fig. 35). This is in contrast with Elemans (2004) who revealed the highest value of SLA for
a high-nutrient treatment and 2% of light. However, SLA on sites near dead logs was higher than
on sites on mineral soil under canopies. It is known that plants are capable of changing their
biomass allocation as well as their leaf morphology during growth, i.e. leaf weight, specific leaf
area, etc. At low light intensities plant usually allocate more biomass to the leaves and then SLA
increases, whereas when the amount of nutrients is low, biomass is allocated to the root system
(Poorter et al. 2011), although SLA can sometimes increase during nitrogen fertilization (Wang et
al. 2012).
The discrepancy between the findings obtained in this study and the works by Peace and Grubb
(1982) and Elemans (2004) result from the research design and growing conditions.
In the
previous work, the experiment was performed in controlled environment cabinets, whereas in the
latter one the conditions were less controlled, i.e. in greenhouses. Moreover, light and nutrients
94
were not controlled in the present study. Differences in light and nutrients are vague because they
are supposed to reflect the natural conditions in forest ecosystems. The most controversial aspect is
the presence of allegedly higher levels of nutrients that can be released from decaying wood.
However, higher values of potassium and phosphorus were found near dead logs in the present
study. Bütler et al. (2007), when studying the decomposition of Picea abies, reported higher values
of carbon and nitrogen under dead logs. In spite of the recognized knowledge that coarse wood
debris contributes to nutrient cycling, little is known about this phenomenon. One more noteworthy
point is that plants that grow near dead decaying logs are partially shaded and moisture conditions
are certainly better.
3.3. Contribution of Impatiens parviflora to plant communities and biotic
relations
3.3.1. The occurrence in plant communities
Phytocenological spectrum
Small balsam is encountered in many types of forest communities, which is shown in the
synoptic table (Appendix 1). Not all vegetation units can be classified into a particular plant
association within the Braun-Blanquet system. Despite some difficulties in the classification of
plant communities in which the species is present, it has been demonstrated that Impatiens
parviflora tends to be confined to deciduous or mixed deciduous forests. An analysis of species
that accompany small balsam in regard to their syntaxonomical affinity showed that the cover
abundance of I. parviflora is positively correlated with the cover of representatives of broad-leaved
deciduous forests (Tab. 21). The species that represent coniferous and mixed coniferous forests
(Vaccinio-Piceetea) negatively correlated with the cover of small balsam. There was also
a negative but weaker correlation with representatives of swamp riparian alder forests (Alnetea
glutinosae). No significant correlation was found with species of acidophilous oak forests (Tab.
21).
Tab. 21. Spearman rank correlation coefficients between the cover of representatives of particular
classes of forest vegetation and the cover of I. parviflora
class
Querco-Fagetea
Vacccinio-Piceetea
Alnetea glutinosae
Quercetea robori-petraeae
Rs
0.37
-0.28
-0.13
-0.07
95
P
<0.0001
<0.0001
0.0041
ns
During phytosociological research that was conducted in the Silesian Upland, 12 vegetation
units that had been distinguished in forest areas were recorded (Appendix 1). Among the 485
phytosociological relevés where neophyte species were present, small balsam occurred in 131.
Impatiens parviflora was the most frequent in the community named after its abundant
occurrence – Pinus sylvestris – Impatiens parviflora, where the median cover was estimated at
63% (Appendix 1). It is a secondary and ephemeral forest community that formed after the
removal of shrubs and digging of soils and it was encountered 11 times. Other forest communities
in which the species was found are given in a descending order in terms of their frequency:
Ficario-Ulmetum (75% of relevés), Fraxino-Alnetum (71%), beechwoods of the Fagenion alliance
(6.4%), oak-hornbeam forests Tilio-Carpinetum (60.6%), swamp alder forests Ribeso nigriAlnetum (60%), acidophilous beechwood Luzulo pilosae Fagetum (58.3%), acidophilous oak
forests Calamagrostio-Quercetum (41.9%), mixed coniferous forests Querco roboris-Pinetum
(41.4%), forest communities of Querco-Fagetea class (35.3%), Molinio-Pinetum (25%),
Calamagrostio villoase-Pinetum (25%) and Leucobryo-Pinetum (12.5%).
The median cover of I. parviflora was not very high. Its median cover only exceeded 10% in
deciduous forests: oak hornbeam forests, mixed coniferous forests and swamp alder forests. Its
cover on average amounted to 7% in disturbed anthropogenic communities of Querco-Fagetea
while in the remaining types of forest communities, the species cover of small balsam varied
between 2-3% (Appendix 1).
The results that were obtained show a phytosociological amplitude that is typical for small
balsam. Since Ćwikliński (1978) and (Trepl 1984), it has been known that the species occurs in
many types of plant associations, especially in broad-leaved deciduous forests. However, during
the collection of the standard phytosociological research, some sites that do not fully fit into the
known phytosociological units – syntaxon at the rank of plant association are omitted, which was
pointed out and criticized by Holeksa and Woźniak (2005). Nevertheless, Cabała (1990) while
studying the forest vegetation in the Silesian Upland and collecting around 2000 phytosociological
relevés revealed the presence of small balsam in communities of Querco roboris-Pinetum,
Calamagrostio-Quercetum petraeae, Ficario-Ulmetum campestris, Carici remotae-Fraxinetum,
Tilio-Carpinetum, Luzulo pilosae-Fagetum as well as in a community of Eu-Fagion, Dentario
glanduloase-Fagetum. The species was especially abundant and frequently represented in oakhornbeam forests. Small balsam is rare and not abundant in the remaining forest communities,
which could be the result of the selection of the sites for collecting phytosociological records,
a priori the special selection of relevés for synoptic tables, or it could reflect the state of the
invasion level of I. parviflora in the late 1980s.
96
Irrespective of classification of the forest communities within the Braun-Blanquet approach,
the analysis of the relationships between the cover abundance of I. parviflora and representatives
of particular phytosociological classes showed a distinct fidelity of the species to deciduous
vegetation. A median and weaker negative relationship with representatives of coniferous forests
and bog forests, respectively, is also relevant. It usually has a positive relationship with
representatives of acidophilous oak forests; however, in this case the relationship is slightly
negative and non-significant, which may result in fewer syngenetic species of Quercetea roboripetraeae, which could bias the statistical analyses.
The occurrence of species against the background of invasion level in plant communities
In addition to small balsam, 16 other invasive alien species were found (Appendix 1). The
most frequent were: Quercus rubra, Padus serotina, Solidago gigantea and S. canadensis. Among
the 13 vegetation units that had been invaded, some phytocoenoses were characterized by
a relatively large contribution of neophytes. Taking into account the ratio of cover of all of the
neophytes to the cover of all of the species that were present DN, the most invaded forest
communities were: Pinus sylvestris – Impatiens parviflora ca. 88%, beechwoods (communities of
the Fagenion alliance) and floodplain forests (Ficario-Ulmetum 70% and Fraxino-Alnetum 63%),
while those that had an intermediate degree of invasion were swamp alder forests Ribeso nigriAlnetum, oak-hornbeam forests and acidophilous beech forests Luzulo pilosae Fagetum (Fig. 36).
The forest community with the lowest abundance of neophytes was a pine forest LeucobryoPinetum 3.8%.
The participation of I. parviflora in the invasion of forest communities is rather high. The
mean PN is 10.6. The value of PN among the forest communities varied between 2.3% for MolinioPinetum and 36.9% for a Pinus sylvestris community (Fig 37). For comparison, the PN of Padus
serotina and Quercus rubra were only 0.18 and 0.23, respectively.
The number of alien species that can be found in forests depends on their source. Chmura (2004)
listed 40 species that occur in forest areas in this region, whereas Tokarska-Guzik, et al. (2007)
mentioned nine species among the 45 that they found in the area. According to her and
collaborators in the research on the Silesian Upland, I. parviflora prefers deciduous forests on more
fertile and humid soils, while the other most frequent species, Quercus rubra and Padus serotina,
were primarily recorded in pine, mixed and acidophilous deciduous forests. The lowest
contribution of small balsam to the neophytization of forest communities in Silesian Upland was
observed in coniferous and mixed coniferous forests (Fig. 37).
97
80
60
40
0
20
Mean value of DN [%]
CP CQ EF
FA FU
LF
LP MP
PI
QP RA TC
ZA
code of plant community/association
Fig. 36. Comparison of degree of neophytization DN (ratio of cover of alien to cover of all species)
among the vegetation units that had been distinguished in the Silesian Upland. Explanations: CP –
Calamagrostio villosae-Pinetum; CQ – Calamagrostio-Quercetum; EF – Fagenion; FA – FraxinoAlnetum; FU – Ficario-Ulmetum; LF – Luzulo pilosae-Fagetum; LP – Leucobryo-Pinetum; MP –
Molinio-Pinetum; PI – anthropogenic community with P. sylvestris; QP – Querco roborisPinetum; RA – Ribeso nigri-Alnetum; TC – Tilio-Carpinetum; ZA – anthropogenic community of
Querco-Fagetea class
The pattern for Quercus rubra, which sometimes forms monospecific associations, is the reverse
(Chmura (2013,2014). A negative correlation was found between the cover of Quercus rubra and
the cover of I. parviflora (Tokarska-Guzik et al. 2007). It was stressed that when compared to
other regions in the country, the Silesian Upland belongs to areas that are especially vulnerable to
invasions of alien plant species. Such a frequent occurrence in forests in the southern part of
Poland is the consequence of forest management in the past in both the case of Q. rubra and P.
serotina, (Tokarska-Guzik 2005a,b). Black cherry was recommended and cultivated as an
admixture and pioneer species in tree stands (Halarewicz 2011).
98
40
30
20
0
10
Mean value of PN [%]
CP CQ EF
FA FU
LF
LP MP
PI
QP RA TC
ZA
code of plant community/association
Fig. 37. Comparison of the participation in neophytization PN (ratio of the cover of a species to the
cover of all alien species) of I. parviflora among the vegetation units that had been distinguished in
the Silesian Upland. Explanations: CP – Calamagrostio villosae-Pinetum; CQ – CalamagrostioQuercetum; EF – Fagenion; FA – Fraxino-Alnetum; FU – Ficario-Ulmetum; LF – Luzulo pilosaeFagetum; LP – Leucobryo-Pinetum; MP – Molinio-Pinetum; PI – anthropogenic community with
P. sylvestris; QP – Querco roboris-Pinetum; RA – Ribeso nigri-Alnetum; TC – Tilio-Carpinetum;
ZA – anthropogenic community of Querco-Fagetea class
As a result, the more frequent occurrence of Q. rubra in habitats of coniferous or mixed forests is
a result of the choice of foresters to plant this tree in these types of soil rather than being an effect
of the habitat preferences of the species (Otręba and Ferchmin 2007). The two most frequent and
abundant tree species followed by small balsam were introduced intentionally. Unfortunately,
because they migrated from the cultivation sites and spontaneously spread, their massive presence
is chiefly the result of forest management. It is very often found that their common coexistence is
not the result of an invasion meltdown or an invasion complex (Richardson et al. 2011 and
literature cited therein) but the consequence of human activity.
99
3.3.2. Dynamics of abundance of populations in various plant communities
Year-to-year changes
On three of the ten study plots significant differences were found in the density of individuals
of Impatiens parviflora between two consecutive years. On study plots 2 and 9, an increase in the
mean density was observed in the following year but on study plot 6, there was a significant
decrease in mean density of plants (Tab. 22).
In the common subplots, significant changes were observed on five of the study plots – on study
plots 1, 2, 9 an increase of density was noted and on study plots 4, 6 the mean density decreased
significantly (Tab. 22). In some cases plants grew on different subplots (i.e. study plots 3 and 8)
when compared to the previous year, so there were no common subplots at all.
Tab. 22. Differences in mean density of Impatiens parviflora on study plots (U Mann-Whitney
test) and common subplots between two subsequent years (Wilcoxon test)
Study Plot
1
2
3
4
5
6
7
8
9
10
Year
2005
2006
2005
2006
2005
2006
2005
2006
2005
2006
2005
2006
2005
2006
2005
2006
2005
2006
2005
2006
Total
Mean±SD
19.58±12.83
19.36±14.27
5.35±3.31
10.15±9.87
1.5±0.53
1.83±1.60
5.24±3.62
4.52±3.52
6.89±6.80
7.2±6.16
29.28±11.18
20.78±12.63
17.36±10.92
23.99±20.06
1.56±0.88
9.33±11.31
4.72±5.22
6.2±4.98
4.05±3.03
5.09±4.91
P
ns
0.036
ns
ns
ns
<.0001
ns
ns
0.034
ns
Common subplots
Mean±SD
P
19.8±12.7
0.0009712
22.1±13.4
4.0±3.0
<0.0001
5.8±5.4
5.5±3.7
0.03585
4.5±3.5
10.0±7.4
ns
8.2±6.5
30.1±10.8
<0.0001
20.3±12.5
18.0±10.8
ns
24.8±19.9
4.8±5.5
ns
5.4±4.7
5.4±3.3
0.0199
13.4±10.7
Under favorable conditions small balsam can markedly increase in cover year by year.
Obidziński and Symonides (2000) noted an increase in both cover and density (number of
individuals) in six types of communities – Tilio-Carpinetum and Ficario-Ulmetum campestris and
their degraded variants. It is noteworthy that they chose patches of vegetation that had been
established perpendicular to the forest path, thus the distinct gradient of the disturbance was
apparent in this research. In the present study non-significant changes occurred on two plots of the
100
initial invasion group and one from the intermediate group. A significant decrease occurred on
only one study plot from the intermediate invasion group. Significant changes in the mean cover
were observed more frequently on the common subplots on which individuals had been present in
the previous year except for one study plot. It can be inferred that patches of I. parviflora generally
persisted where it had been before in this study, which may be the result of favorable microsites,
lower biocenotic resistance or short-distance dispersal.
Long term changes
During the eight years of the study from 2005 until 2012, the spatial distribution (Fig. 38–
47), frequency and mean cover (Fig. 48–51) of Impatiens parviflora changed considerably on
permanent study plots. The frequency, i.e. the number of subplots that were occupied by small
balsam within the study plots changed significantly in seven cases (Tab. 23). The Friedman rank
test revealed significant changes in the mean cover in paired subplots within all of the study plots
except for study plot 3. The Kruskal-Wallis test indicated the significance of the mean cover
changes in all of the study plots except for study plots 3, 5, 7 and 8 (Tab. 24).
On study plot 1, which was growing near a floodplain forest Fraxino-Alnetum and which
represented an intermediate degree of invasion, there was an increase in the frequency of
I. parviflora at the beginning of the study but it later decreased systematically from 75% down to
10%. Meanwhile, the mean cover of the invaded subplots decreased from 40% to ca. 5% (Fig.
48AB). The distribution of individuals also changed. At the beginning of the study, the spatial
structure was gradient but later on it became scattered (Fig. 38).
Study plot 2, which was situated in an oak-hornbeam forest, also underwent significant
changes. Two peaks in the increase in the frequency of small balsam occurred in 2007 and 2010
when the number of invaded subplots varied between 79 and 71. The mean frequency was ca. 54%
but in the last year of the study, there were only 18, which was the lowest value during the entire
period of the research (Fig. 48CD). The median cover was similar during the eight years and
amounted to 10%, although it decreased to ca. 5% in the invaded subplots in the last two years.
Study plot 3, which was from the initial invasion group and was situated in a beech wood Dentario
glanduloase-Fagetum, resembled Carici-Fagetum and was characterized by either random
dynamics in the frequency of small balsam, which ranged from five to ten, while its median cover
in invaded subplots decreased somewhat from 8% down to 5% (Fig. 48EF). The population was
characterized by a scattered type of distribution (Fig. 40).
Study plot 4 was highly invaded at the beginning of the study and therefore it represented
a saturation invasion group. This area was an oak-hornbeam forest in which there was a rapid
101
decrease in the frequency of I. parviflora from 90 down to 24 (Fig. 49A). The highest median
cover was observed in the years 2007–2011 but in the last year of the study, it dropped to ca. 5%
(Fig. 49B). The spatial distribution of individuals changed from areal type to areal-clumped (Fig.
41).
Study plot 5 was representative of an intermediate degree of invasion and was situated in an
oak-hornbeam forest. The frequency of I. parviflora there was highest in the first year and in last
year, reaching 46 and 47, respectively. The median cover in the invaded subplots was similar over
the years and varied between 10% and 15% (Fig. 49CD). The spatial distribution of the population
was gradient-clumped during the entire period of the study (Fig. 42).
On study plot 6 a considerable decrease in the frequency of small balsam was recorded from
98 to 20, and simultaneously the median cover in the invaded subplots dropped from 70% to 10%
(Fig. 49EF). This area represented highly invaded areas and was located in an acidophilous
beechwood Luzulo pilosae Fagetum. The spatial distribution changed profoundly. In the first years
of the study (2005-2006), it was areal-clumped, but in the years 2007-2008 it changed into arealtussocky. For two years (2009-2010) the spatial structure became gradient-tussocky and finally
clumped-tussocky (Fig. 43).
On study plot 7, which was also a highly invaded area and represented Dentario
glanduloase-Fagetum, no major changes in frequency were observed except for 2010 when the
number of subplots that were invaded by I. parviflora dropped from 95 to 79 (Fig. 50A). The
median cover of small balsam varied from 15% to 20% (Fig. 50B). The distribution of individuals
of small balsam was similar over the years and resembled a gradient-tussocky type of structure
(Fig. 44).
Study plot 8 was situated in Dentario glandulosae-Fagetum, which at the beginning of the
study was in the initial phase of an invasion and subsequently underwent significant changes in the
frequency of I. parviflora, which increased significantly from 9 to 23 (Fig. 50C). The median cover
in invaded subplots increased from 5% to 10% (Fig. 50D). The spatial structure of the population
was rather scattered in every year of the observation (Fig. 45).
In both study plots 9 and 10, which are situated in a beechwood Dentario glandulosaeFagetum and an oak-hornbeam forest Tilio-Carpinetum, there was an increase in the number of
subplots that were invaded by small balsam. In the former from 37 to 56 were detected and in the
latter, which occurred more rapidly, from 49 to 93 were detected (Fig. 50E, Fig. 51A). The spatial
structure of the population on study plot 9 was of an areal type but in the last year of the study it
resembled a gradient type (Fig. 46).
102
Tab. 23. Results of the G-test of the comparison of frequencies (number of invaded vs
uninvaded subplots by Impatiens parviflora) in the years 2005-2012
Number of study plot
1
2
3
4
5
6
7
8
9
10
Statistics
169.7
124.4
4.0
175.7
12.3
333.7
28.6
28.0
11.0
61.4
Probability
<0.000
<0.000
ns
<0.000
ns
<0.000
0.00017
0.00021
ns
<0.000
Tab. 24. Results of Friedman rank test (all subplots included) and Kruskal-Wallis test (only
invaded plots) on the basis of comparison of Impatiens parviflora mean cover over the years 20052012
No. of study plot
Friedman rank test
Kruskal-Wallis test
1
χ2=306.9; p <0.0001
χ2= 82.3; p < 0.0001
2
χ2=223.6; p <0.0001
χ2= 16.0; p = 0.02441
3
χ2= 6.0663; ns
χ2= 7.59; ns
4
χ2= 172.5; p <0.0001
χ2= 81.5; p<0.0001
5
χ2= 105.4; p <0.0001
χ2=6.0; ns
6
χ2= 418.0; p <0.0001
χ2=158.6; p< 0.0001
7
χ2= 32.21; p <0.0001
χ2= 1.8; ns
8
χ2=37.2; p <0.0001
χ2=6.0; ns
9
χ2=306.9; p <0.0001
χ2=27.0; p = 0.00032
10
χ2=105.4; p <0.0001
χ2= 22.1; p = 0.0023
103
Fig. 38. Dynamics of distribution of Impatiens parviflora on study plot no. 1. Explanations: 1–
0%; 2– >1-10%; 3– >10-20%; 4– >20-30%; 5– >30-40%; 6– >40-50%; 7– >50-60%; 8– >60-70%;
9– >70%; 10– >80% of plant cover
104
Fig. 39. Dynamics of distribution of Impatiens parviflora on study plot no. 2. Explanations: 1–
0%; 2– >1-10%; 3– >10-20%; 4– >20-30%; 5– >30-40%; 6– >40-50%; 7– >50-60%; 8– >60-70%;
9– >70%; 10– >80% of plant cover
105
Fig. 40. Dynamics of distribution of Impatiens parviflora on study plot no. 3. Explanations: 1–
0%; 2– >1-10%; 3– >10-20%; 4– >20-30%; 5– >30-40%; 6– >40-50%; 7– >50-60%; 8– >60-70%;
9– >70%; 10– >80% of plant cover
106
Fig. 41. Dynamics of distribution of Impatiens parviflora on study plot no. 4. Explanations: 1–
0%; 2– >1-10%; 3– >10-20%; 4– >20-30%; 5– >30-40%; 6– >40-50%; 7– >50-60%; 8– >60-70%;
9– >70%; 10– >80% of plant cover
107
Fig. 42. Dynamics of distribution of Impatiens parviflora on study plot no. 5. Explanations: 1–
0%; 2– >1-10%; 3– >10-20%; 4– >20-30%; 5– >30-40%; 6– >40-50%; 7– >50-60%; 8– >60-70%;
9– >70%; 10– >80% of plant cover
108
Fig. 43. Dynamics of distribution of Impatiens parviflora on study plot no. 6. Explanations: 1–
0%; 2– >1-10%; 3– >10-20%; 4– >20-30%; 5– >30-40%; 6– >40-50%; 7– >50-60%; 8– >60-70%;
9– >70%; 10– >80% of plant cover
109
Fig. 44. Dynamics of distribution of Impatiens parviflora on study plot no. 7. Explanations: 1–
0%; 2– >1-10%; 3– >10-20%; 4– >20-30%; 5– >30-40%; 6– >40-50%; 7– >50-60%; 8– >60-70%;
9– >70%; 10– >80% of plant cover
110
Fig. 45. Dynamics of distribution of Impatiens parviflora on study plot no. 8. Explanations: 1–
0%; 2– >1-10%; 3– >10-20%; 4– >20-30%; 5– >30-40%; 6– >40-50%; 7– >50-60%; 8– >60-70%;
9– >70%; 10– >80% of plant cover
111
Fig. 46. Dynamics of distribution of Impatiens parviflora on study plot no. 9. Explanations: 1–
0%; 2– >1-10%; 3– >10-20%; 4– >20-30%; 5– >30-40%; 6– >40-50%; 7– >50-60%; 8– >60-70%;
9– >70%; 10– >80% of plant cover
112
Fig. 47. Dynamics of distribution of Impatiens parviflora on study plot no. 10. Explanations: 1–
0%; 2– >1-10%; 3– >10-20%; 4– >20-30%; 5– >30-40%; 6– >40-50%; 7– >50-60%; 8– >60-70%;
9– >70%; 10– >80% of plant cover
113
Fig. 48. Dynamics of the frequency (left) and the median percent cover (right) only in subplots
that were occupied by Impatiens parviflora. Explanations AB – study plot 1, CD – study plot 2, EF
– study plot 3
114
Fig. 49. Dynamics of the frequency (left) and the median percent cover (right) only in subplots
that were occupied by Impatiens parviflora. Explanations AB – study plot 4, CD – study plot 5, EF
– study plot 6
115
Fig. 50. Dynamics of the frequency (left) and the median percent cover (right) only in subplots
that were occupied by Impatiens parviflora. Explanations AB – study plot 7, CD – study plot 8, EF
– study plot 9
On study plot 9 the median cover increased from 10 to 20 (Fig. 50F), whereas on study plot 10, it
varied over the years and ranged from 8 to 18% (Fig. 51B).
116
Fig. 51. Dynamics of the frequency (left) and the median percent cover (right) only in subplots
that were occupied by Impatiens parviflora. Explanations AB – study plot 10
The spatial distribution of the population of I. parviflora on study plot 10 was rather
scattered in 2005 but later on it started to be a gradient type (Fig. 47).
To the best of the author’s knowledge, no similar studies that were designed to identify the
dynamics of I. parviflora every year over a long period of time have been undertaken. Thus, it is
impossible to compare these results with other studies. Examples of research that showed changes
after some period can be found in the literature. For instance, Łysik (2008), who studied Dentario
glandulosae-Fagetum on 23 permanent study plots in Ojców National Park, revealed that in 2003
in comparison with 1993 small balsam appeared on eight plots and occupied ca. 12% of the entire
vegetation cover. Kiedrzyński et al. (2011) reported the appearance and increase in the cover of
small balsam after 40 years in a nature reserve in Central Poland. Both cases presented “before and
after studies” without investigating the dynamics of the species over the course of time. On the
other hand, some reports have indicated a decrease in the cover of the species year by year. For
instance, small balsam showed a downward trend and finally disappeared within five years during
the recovery of beech forest (Godefroid et al. 2003).
All of the possible scenarios of population dynamics occurred in the present study – a significant
increase, a significant decrease and no considerable changes. Two LTR plots that represented the
saturation of an invasion underwent the process of a decrease in population and only one study plot
from an intermediate degree of invasion showed a decrease in the frequency and median cover of
small balsam. A distinct increase in the frequency and finally an increase in the median cover of
the species were noted on three LTR plots – one study plot during the initial phase of an
I. parviflora invasion that showed a distinct increase and two that were invaded to an intermediate
degree showed a further increase in the invasion of the species. One of the three saturation invaded
plots turned out to be relatively stable. The frequency and median cover of small balsam persisted
117
at a similar level. On the other hand, one of the two initial invasion plots also underwent some
fluctuations in the abundance of small balsam, and finally did not show any trend in the dynamics
of population. Such patterns of the dynamics of populations may result from many intrinsic and
extrinsic factors. It is not possible that there is only one scenario, i.e. a gradual invasion from
introduction until post-invasion, e.g. dominance of the invader. As Hejda (2012) wrote I. parviflora
can form dense stands locally on small spatial scales, while Vervoort and Jacquemart (2012)
believe that I. parviflora can be limited locally by an unfavorable environment. As evidence they
pointed out some plots that were still uncolonized by the species. It can be added that, apart from
phenology and changes in abundance during the vegetation season, year-to-year and long-term
changes can also be very marked. Usually an abundant occurrence of the species in forest interiors
is temporary. The massive occurrences that are very often observed are usually in open habitats
within forest complexes.
3.3.3. Role of plant functional groups and biotic diversity of plants in Impatiens parviflora
invasion
The influence of species richness and species diversity on invasion success
When considering number of accompanying species in invaded and uninvaded subplots for
both regions on sites without small balsam, there are differences. Uninvaded subplots in Silesian
Upland were characterized by a higher species richness of native plants but the opposite situation
was observed in Jurassic Upland (Tab. 25).
The analysis of relationship between the density of I. parviflora individuals and the cover
of bare ground and the relation between the density of the species and the cover that was occupied
by litter gave different results (Tab. 25). A cover of litter positively correlated with both the density
and the cover of small balsam, whereas a cover of bare ground was negatively correlated with the
cover of I. parviflora.
The associations between the cover of native species and the cover of Impatiens parviflora
are different because of the regions. There was a negative relationship in the case of the nature
reserves of the Jurassic Upland and a weak positive in the case of the Silesian Upland (Fig. 52)
based on the data from the study plots and additional data sets. The relationship between the cover
of native species and the cover of small balsam from only 68 study plots exhibited a negative
correlation for the Jurassic Upland (rs=-0.19, p<0.0001) and a weak negative for the Silesian
Upland (rs=-0.03, p<0.0001).
118
Tab. 25. Comparison of subplots with the absence and presence of Impatiens parviflora in the
number of accompanying native species in the herb layer of two regions and the values of the
Spearman rank correlation between the cover of bare ground, the cover of litter and the density and
cover of Impatiens parviflora DC. (*p<0.05, **p<0.01, ***p<0.001)
Absence
Presence
Jurassic Upland
5.57±2.6***
4.79±2.53
Silesian Upland
1.26±1.25
2.15±1.33***
Variable (Jurassic Upland)
Cover of litter
Bare ground
Density
Percent cover
0.35***
0.20**
ns
-0.11*
Fig. 52. Comparison of the values of Spearman rank correlation coefficient between the cover of
Impatiens parviflora and the cover of native species for the Jurassic Upland (A) and the Silesian
Upland (B) based on the data from 68 permanent study plots and additional data sets.
These aforementioned results are in accordance with findings about the relationship
between the occurrence of I. parviflora and the presence of gaps in the ground flora layer
(Obidziński, Symonides 2000), but differ in relation to the role of bare ground. It has not been
confirmed that an increase in the cover of bare ground enhances an increase in the cover of the
species, which Obidziński and Symoindes (2000) explained as a favorable factor for an increase in
its density. However, an increase both of the cover and density on soil that is covered by litter
confirmed the observation that even a thick layer of litter does not hinder the germination and
growth of I. parviflora on the forest floor (Coombe 1956). The same result was obtained by
Obidziński and Symoindes (2000), although as was already mentioned before they used transects
in disturbed sites from forest paths towards forest interiors, on which the influence of the
disturbance that is associated with trampling is permanent and relatively constant. In the present
119
work the selected study plots differed only by their level of invasion and were laid out far from
forest margins such as paths, cut-areas, etc. Therefore, this indicates a more spontaneous,
stochastic process that was induced by species invasiveness rather than one that was caused by
extrinsic factors.
Another important feature is the biocenotic resistance of ecosystems. This character of an
ecosystem or phytocoenosis is dependent on species richness and species diversity (Kennedy et al.
2002, Van Holle et al. 2003). In a previous work (Chmura, Sierka 2006a), it was shown that
subplots that had been invaded by I. parviflora had higher biodiversity indices in relation to native
species than uninvaded ones. Thus, the detailed analyses in which the two regions were analyzed
separately gave different results (Fig. 52). For the Jurassic Upland there is an opposite pattern, the
invaded plots had lower species richness than uninvaded. For Silesian Upland there is the other
way round. This discrepancy could be explained by several factors. Firstly, forest in Jurassic
Upland grow on limestone substratum, secondly the quality of forests, even in nature reserves, is
better. They are less disturbed due to forest management. Thirdly, according to cadastral maps,
Jurassic forests are rather ancient forest than recent forests.
Interesting results about differences in species richness and cover of species were shown by
Vervoort et al. (2012). Both number of species and cover was lower on plots that had not been
colonized by I. parviflora and I. noli-tangere when compared to sites with I. parviflora, whereas
sites invaded only by I. parviflora had significantly a lower species richness than sites with I. nolitangere only and plots on which two species were present. In the same study, a negative
relationship between the coverage of small balsam and species richness was also shown and it was
quite high (R2 = -0.41), which indicates that I. parviflora preferred sites with lower species
richness when compared to its native congener. Plots with two Impatiens species did not differ
significantly in species richness and cover from plots with native balsam only.
A negative correlation between herbaceous species as well as with mosses was
demonstrated by Dobravolskaitė (2012) for pine and spruce forests and pine forest edges.
Negative relations between density, the cover of small balsam and the cover of native species were
revealed by Obidziński, Symonides (2000) and Csontos (1984). Godefroid and Koedam (2010)
demonstrated that species richness in the herb layer significantly decreased with an increasing
abundance of I. parviflora. The opposite pattern was found in an analysis of cover relations.
I. parviflora had its highest abundance in communities that had a dense herb layer. Golivets (2013)
found a very weak positive (r=0.12) but significant correlation between the abundance of small
balsam and other species at the scale of 1m2.
120
Fig. 53. Comparison of the association between the cover of I. parviflora and the cover of native
species between the years in the Jurassic Upland (data for 10 study plots)
121
In the work by Chmura and Sierka (2006a), the patches of the same forest communities
from two regions were pooled and subjected to statistical analysis. An analysis across the region
obtained biased patterns. As is shown in Figure 52, the correlation for the forests of the Jurassic
Upland is negative whereas for the Silesian Upland it is slightly positive. Such a divergence may
be a consequence of the impact of extrinsic factors that permit small balsam and other components
of ground flora to use disturbance sites on the forest floor together, thus there is a higher number of
native species in the case of the Silesian Upland. For instance, a significant and positive correlation
with the cover of herbaceous species was revealed for native but expansive presence of
Calamagrostis epigejos in the forests of the region (Sierka, Chmura 2005). As regards small
balsam, the only case in which there was also a positive correlation with the cover of the herb layer
was the Sonian Forest in Belgium. The forest complex was built mainly by Fagus sylvatica and
coniferous trees occupy ca. 8%. Perhaps a key reason is a disturbance of that area that is
manifested by the massive occurrence of I. parviflora, which was stressed by Godefroid and
Koedam (2010).
Obidziński and Symonides (2000) showed that the relationship, between density of I. parviflora
and cover of other herbaceous species, which is expressed as a higher value of the correlation
coefficient, was higher in the following year (1998) than in the previous one (1997). There was an
opposite result for the number of native species. The values of both correlations analyses were
similar and they are probably meaningless. In the present work when 2006 was compared with
2005, lower values for the correlation coefficients between density, the cover of small balsam and
number and cover of native species were obtained. This might have resulted from a decrease in the
biocenotic resistance of the LTR plots that were studied that were due to frequent trampling, which
enhanced the explosion of the fruits and the seed dispersal of the plants owing to their penetration
within the study plots. In the following years, the methods were modified in order to eliminate the
effect of disturbances that were associated with performing the research, which could have
contributed to the higher occurrence of small balsam within the LTR plots.
The effect of scale on relations with native species
Analysis of flora
In total 68 study plots were analyzed in terms of the accompanying flora and the frequency
and cover of accompanying species. These study plots were situated in seven forest communities
that represent oak-hornbeam forests Tilio-Carpinetum – TC (26 plots), swamp alder forests Ribeso
nigri-Alnetum – RA (2), mixed coniferous forests Querco roboris-Pinetum – QP (19), acidophilous
beech forests Luzulo pilosae-Fagetum – LF (7), thermophilous beech forests Carici-Fagetum – CF
122
(2), beechwood forests Dentario glanduloase-Fagetum – DF (9) and floodplain forests of AlnoUlmion alliance – AU (3).
Fig. 54. Abundance of Impatiens parviflora and the distribution of the study plots that represent
the various forest communities along the two first DCA axes in the Silesian-Krakow Upland.
Abbreviations of plant communities see text.
These plots are diversified in regard to the abundance of I. parviflora (Fig. 54A). DCA, which
was based on the sum of the covers of particular species on subplots within the study plots,
produced four axes with eigenvalues 0.556, 0.400, 0.229, 0.202 and the length of gradients 3.840,
3.625, 2.463, 3.003. Their cumulative explained variation is 19.4%. At the level of the study plots,
I. parviflora is not a significant variable in explaining species variation (Monte Carlo test CCA,
p=0.309) based on their frequency (number of subplots occupied), whereas based on the sum of
their covers, it is significant (p=0.015) although the inflation factor is too high because it was
correlated with the other species that were present. Moreover, it explained only 2.3% of the
variation in native species.
There were 191 species on the study plots but in 3628 subplots of 1m2 with the presence of
I. parviflora, there were 171 species while 173 species occurred in 3172 subplots without
I. parviflora. In total 152 species were common for both types of subplots.
The participation of Grime’s life strategies was very similar in both groups – plots with and
without I. parviflora (Fig. 55).
The C-S stress tolerant competitors dominated (23.7%-22.8%) followed by C-S-R
strategists (15.8-16.4%) and competitive ruderals (10.5%-11.5%). Representatives of two strategies
were not found – SR – stress tolerant ruderals and S/SR strategists.
123
The distribution of Raunkiaer’s life forms was as follows (Fig. 56): hemicryptophytes were
the
dominant
group
(44.8%-46.6%)
of
the
total
flora;
geophytes
(19.4%-19.0%),
megaphanerophytes (12.1%-12.3%) and nanophanerophytes (11.5%-11.7%). The remaining
groups played minor roles.
Anemochores were predominant among the accompanying species on subplots with
I. parviflora – 41.4% and on subplots without – 40.2% (Fig. 55). Endozochores were the second
most abundant group constituting 17.7% and 17.8% of flora, respectively.
Fig. 55. Grime’s C-S-R triangle for the accompanying flora of Impatiens parviflora on the
sites with a presence (invaded) and on sites with an absence of the species (uninvaded)
Myrmecochores contributed 12.8% and 11.8%, respectively. The least abundant were
hydrochores and autochores, which did not exceed 3.7% of the flora of the two groups of species
(Fig. 57).
The distribution of the species that represent particular Ellenberg indicator values was
similar in both groups of plants. With regard to light (L), the plants of the L6 and L7 values
predominate, i.e. plants that prefer semi-shade and well-lit places but that can also tolerate partial
shade. They both constitute 33.7% and 34.7% of the flora, respectively, for subplots both with and
without I. parviflora. Plants that prefer shade and semi-shade conditions were quite numerous (L =
3 and L = 4) (Fig. 58 L).
Taking into account temperature (T) species of the T5 value dominate in both groups of
plants reaching 46.3% and 46.9%, followed by species of the T6 value (24.5% and 26.9%) (Fig.
124
58T). Edaphic conditions, which were characterized by flora on sites with and without
I. parviflora, showed the same trend. In regard to moisture (F), representatives of mesophytic
habitats and fresh soils were the dominant groups on sites with or without small balsam (F= 5:
39.0% and 38.7%). Moreover, species that prefer a slightly moist substratum were numerous
(21.3% and 19.1%) (Fig 59F). The intermediate and slightly acid or weakly alkaline soils
indicators dominate in both groups of plants, i.e. values 6-7 contributed 19.5%, 20.2% and 29.6%
and 28.3%, respectively. The species that are indicators of calcium carbonate constituted 10.6%
and 10.4% of both floras (Fig. 59R). Among the introgen indicator species plants that occur in
80
intermediate-rich mineral nitrogen compounds were the most numerous (19.5% and 20%).
0
20
40
60
invaded
uninvaded
C
CH
G
H
HY
M
N
T
70
Fig. 56. Distribution of Raunkiaer’s life forms of the accompanying flora of Impatiens
parviflora on sites with a presence (invaded) and sites with an absence of the species (uninvaded).
Abbreviations: C – chamaephytes, CH – herb chamaephytes, H – hemicryptophytes, HY –
hydrophytes, G – geophytes, N – nanophanerophytes, M – megaphanerophytes, T – therophytes
0
10
20
30
40
50
60
invaded
uninvaded
AN
AU
B
D
EN
EP
HY
M
Fig. 57. Distribution of the types of the dispersal modes of the accompanying flora of
Impatiens parviflora on sites with a presence and on sites with an absence of the species. AN –
anemochores, AU – autochores, B – barochores, D – dyszoochores, EN – endozochores, EP –
epizochores, HY – hydrochores, M – myrmecochores
125
The plants that tend to occurr in nitrogen-rich soils (values 6 and 7) are also a major part of
both floras (Fig. 59N).
50
L
0
10
20
30
40
invaded
uninvaded
1
2
3
4
5
6
7
8
9
X
60
T
0
10
20
30
40
50
invaded
uninvaded
3
4
5
6
7
X
Fig. 58. Distribution of representatives of particular Ellenberg values for the climatic
conditions (L–light, T–temperature) of the accompanying flora of Impatiens parviflora on sites
with a presence and on sites with an absence of the species
126
70
F
0
10
20 30
40
50 60
invaded
uninvaded
1
3
4
5
6
7
8
9
X
50
R
0
10
20
30
40
invaded
uninvaded
2
3
4
5
6
7
8
9
X
35
N
0
5
10
15
20
25
30
invaded
uninvaded
1
2
3
4
5
6
7
8
9
X
Fig. 59. Distribution of representatives of particular Ellenberg values for the edaphic
conditions (F – humidity, R – soil reaction, N – nitrogen) of the accompanying flora of Impatiens
parviflora on sites with a presence and sites with an absence of the species
127
Analysis of differences in plant functional types of accompanying species on a large scale
There were no significant differences in the mean total number of occurrences between the
species that were present on all of the invaded subplots and all of the uninvaded subplots in terms
of life forms, dispersal mode or Grime’s strategies (data not shown). Moreover, the mean cover of
plant functional types did not differ significantly between all of the invaded and uninvaded
subplots combined (Tab. 26).
Tab. 26. Comparison of the mean cover of accompanying species on sites with (invaded) or
without Impatiens parviflora (uninvaded) within a 100 m2 area based on 68 permanent plots.
Invaded
Life form:
Chamaephytes
Herb chamephytes
Geophytes
Hemicryptophytes
Hydrophytes
Megaphanerophytes
Nanophanerophytes
Therophytes
Dispersal mode:
Anemochores
Autochores
Barochores
Dyszoochores
Endozoochores
Epizoochores
Hydrochores
Myrmecochores
Grime strategy:
Competitors
Stress tolerators
Ruderals
Competitive ruderals
Stress tolerant competitors
CSR strategists
Uninvaded
Statistics, probability
10.68±15.25
5.15±12.90
52.60±38.82
62.09±71.53
0.13±1.09
28.56±32.45
14.59±18.26
7.13±16.27
12.26±22.53
3.94±11.07
53.07±45.42
52.37±60.82
0.06±0.29
27.60±32.59
12.85±14.29
3.15±8.83
W = 2370, p = 0.7884
W = 2364, p = 0.7916
W = 2379.5, p = 0.7706
W = 2448.5, p = 0.5538
W = 2245.5, p = 0.3264
W = 2409.5, p = 0.6728
W = 2354, p = 0.8557
W = 2482.5, p = 0.3919
44.59±38.31
19.63±23.65
16.25±20.54
12.26±17.45
28.76±36.04
23.35±36.22
0.79±4.51
35.22±43.47
42.13±36.06
16.65±18.40
14.75±17.65
8.57±11.62
27.01±22.70
18.18±28.19
1.96±15.27
36.06±59.24
W = 2415.5, p = 0.6539
W = 2394.5, p = 0.7193
W = 2377, p = 0.7783
W = 2499, p = 0.4122
W = 2171, p = 0.5407
W = 2371, p = 0.7975
W = 2248.5, p = 0.5013
W = 2452.5, p = 0.5393
27.31±31.35
13.03±19.17
2.34±6.76
17.56±29.71
56.21±50.75
62.87±59.09
27.74±33.36
12.06±16.04
1.44±5.14
13.07±19.10
55.07±57.91
54.13±55.06
W = 2294, p = 0.9393
W = 2319, p = 0.9766
W = 2338.5, p = 0.8776
W = 2390.5, p = 0.7272
W = 2451.5, p = 0.5451
W = 2529.5, p = 0.3448
Analysis of differences in plant functional types of accompanying species on a small scale
The invaded subplots had a lower mean cover of chamaephytes (Fig. 60C) and geophytes
(Fig 60G), whereas therophytes had a higher mean cover (Fig 60T).
128
CH
NS
2.0
1.0
Cover [%]
3
2
0
0.0
1
Cover [%]
4
3.0
C p<0.0001
invaded
uninvaded
invaded
10 15 20
0
invaded
uninvaded
invaded
uninvaded
M
6
0
0.00
2
4
Cover [%]
0.08
0.04
NS
8 10
NS
0.12
HY
Cover [%]
NS
5
5 10
Cover [%]
20
H
0
Cover [%]
G p<0.001
uninvaded
invaded
uninvaded
invaded
uninvaded
Fig. 60. Comparison of the mean cover of Raunkiaer life forms representatives within
accompanying species among 1 m2 plots with and without Impatiens parviflora. C –
chamaephytes, CH – herb chamaephytes, H – hemicryptophytes, HY – hydrophytes, G –
geophytes, M – megaphanerophytes
129
invaded
uninvaded
1.5 2.0
p<0.001
0.0 0.5 1.0
1 2 3
Cover [%]
4 5 6
T
0
Cover [%]
N NS
invaded
uninvaded
Fig. 60. continued. Comparison of mean cover of Raunkiaer life forms representatives within
accompanying species between 1 m 2 plots with and without Impatiens parviflora. N –
nanophanerophytes, T – therophytes, NS – nonsignificant
Competitors had a lower cover on invaded plots than on uninvaded (Fig. 61C), whereas
ruderals had a much higher (almost twofold) cover on invaded subplots (Fig. 61R). The mean
cover of competitive ruderals was a little higher on invaded subplots than on uninvaded ones (Fig
61CR). In turn, stress tolerant competitors had a higher mean cover on uninvaded subplots (Fig
61CS).
On average, anemochores had a higher cover on uninvaded subplots (Fig. 62A) as did
endozochores (Fig. 62EN). Invaded plots had a significantly higher mean cover of dyszoochores
and epizoochores (Fig. 62D,EP). Myrmecochores reached a higher mean cover on subplots without
Impatiens parviflora.
The relationship between the abundance of small balsam and the cover of species
representing plant functional groups was also affected by scale (Tab. 27). When the frequencies of
a species were taken into account, only two groups significantly correlated, i.e. the cover of
chamaephytes (negative correlation) and the cover of therophytes and anemochores (positive
correlation). At the scale of 1m2 there were many significant relationships but they are very weak
(Tab. 27). A comparison of the correlation coefficients for anemochores led to contradictory
results. At the lower scale, the correlation was very weak and negative but at the large scale –
medium and positive.
130
p<0.001
S
4
3
uninvaded
invaded
P<0.001
uninvaded
CR p<0.05
4
2
0
0.0
0.2
0.4
Cover [%]
6
0.6
R
invaded
P<0.001
Cover [%]
NS
0
0
5
15
uninvaded
CSR
25
CS
uninvaded
20
invaded
5 10
Cover [%]
2
0
invaded
Cover [%]
NS
1
Cover [%]
8
4
0
Cover %
12
C
invaded
uninvaded
invaded
uninvaded
Fig. 61. Comparison of the mean cover of representatives of Grime strategies within the
accompanying species among 1 m2 plots with and without Impatiens parviflora. Explanations: C –
competitors, S – stress tolerators, R – ruderals, CR – competitive ruderals, CS – stress tolerant
competitors, CSR – CSR strategists, NS – nonsignificant
131
Cover [%]
10
0
invaded
uninvaded
invaded
D
p<0.001
2.0
0.0
1.0
Cover [%]
4
2
0
invaded
p<0.001
uninvaded
EP p<0.001
6
0
2
0 2 4 6 8
Cover [%]
12
8
EN
uninvaded
4
Cover [%]
invaded
Cover [%]
uninvaded
3.0
NS
6
B
NS
0 1 2 3 4 5 6
AU
20
P<0.05
5
Cover [%]
AN
invaded
uninvaded
invaded
uninvaded
Fig. 62. Comparison of the mean cover of dispersal mode representatives within
accompanying species among 1 m2 plots with and without Impatiens parviflora. Explanations: AN
– anemochores, AU – autochores, B – barochores, D – dyszoochores, EN – endozochores, EP –
epizochores, NS – nonsignificant
132
M
NS
invaded
uninvaded
0 2 4 6 8
0.15
Cover [%]
12
0.30
NS
0.00
Cover [%]
HY
invaded
uninvaded
Fig. 62. continued. Comparison of the mean cover of dispersal mode representatives within
accompanying species among 1 m2 plots with and without Impatiens parviflora. Explanations: HY
– hydrochores, M – myrmecochores, NS – nonsignificant
Some studies have shown that a relationship exists between the species cover of native and nonnative species and their covers: positive at the landscape scale (e.g., Stohlgren et al. 1999; Sax
2002), a vegetation scale lower that 100 m2 and negative at scales of < 1 m2 (neighborhood scales).
Thus, it can be expected that when comparing flora, the frequency and cover of native species
among invaded and uninvaded plots were also affected by the scale. There were no significant
differences in accompanying flora from all of the invaded and uninvaded subplots that were
pooled. The total flora from 68 100 m2 study plots was comparable to the flora at the landscape
level. The majority of studies have shown that at large areas > 1000 m2 and at vegetation scales as
well – 100 m2 (Stohlgren et al. 1999, 2003, 2006), there is a strong positive relationship between
native species richness and the cover, biomass and relative cover of alien plant species. At lower
scales – subplot (1m2), the differences in particular functional groups (life strategy, dispersal mode,
life form) show which sites are preferred by small balsam. Perennial plants (geophytes,
chamaephytes), competitors, stress-tolerant competitors, myrmecochores, anemochorous and
endozoochorous plants are more abundant on uninvaded plots, which may indicate
overcompetition by these species. Native annual plants (therophytes, ruderals) indicate
a disturbance on microsites where small balsam grows. It is more difficult to explain the higher
cover of dyszoochorous (tree seedlings) and epizoochorous plants on invaded plots. Small balsam
tolerates a high, but not too abundant, cover of tree seedlings due to its small aboveground biomass
and shallow root system. Such places are avoided by other herbaceous plants but are used by
I. parviflora.
The relationship between the cover and the total cover of the groups that had been distinguished is
modified by scale. A significant negative or positive correlation between the invasive species and
133
cover of particular functional groups at the scale of 100 m2 was demonstrated by Chmura (2013)
for red oak Quercus rubra. In the present study at the same scale, only two correlations turned out
to be significant, whereas at a lower scale the majority were significant but they were very weak
and unimportant. This proves that small balsam when compared to other species plays an
insignificant role in interspecific relations. If a species has an impact on coexisting plants, it could
be expected that not only on some species as such but in general plant traits or plant functional
groups are affected.
Tab. 27. Relationships (Spearman rank test) between the cover of particular functional groups and
the abundance of I. parviflora
Life form:
Chamaephytes
Herb chamaephytes
Geophytes
Hemicryptophytes
Hydrophytes
Megaphanerophytes
Nanophanerophytes
Therophytes
Dispersal mode:
Anemochores
Autochores
Barochores
Dyszoochores
Endozoochores
Epizoochores
Hydrochores
Myrmecochores
Grime strategy:
Competitors
Stress tolerators
Ruderals
Competitive ruderals
Stress tolerant competitors
CSR strategists
Scale of subplot [1m2]
rs
p
-0.0585
<.0001
0.0523
<.0001
-0.1148
<.0001
ns
0.0340
0.0050
-0.0293
0.0155
ns
0.2688
0.0864
<.0001
Scale of plot [100 m2]
rs
p
-0.3510 0.0033
ns
ns
ns
ns
ns
ns
0.3127 0.0094
-0.0334
-0.0524
-0.0259
0.0492
ns
ns
-0.0478
-0.1009
0.0059
<.0001
0.0330
<.0001
0.2442
ns
ns
ns
ns
ns
ns
ns
-0.0585
-0.0428
ns
ns
-0.0280
-0.0758
<.0001
0.0004
<.0001
<.0001
0.0210
<.0001
134
ns
ns
ns
ns
ns
ns
0.0448
3.3.4. Ecological conditions of the occurrence of mycorrhiza
The history of the evolution of the views on mycorrhiza in I. parviflora was described in detail
in a previous work (Chmura, Gucwa-Przepióra 2012). A detailed list of samples of the data on the
presence/absence of AM is presented in Table 28.
As a reminder, one can first say that the report indicates that there were no mycorrhiza in this
species and that recently, the species is definitely mycorrhizal (Štajerová et al., 2009). An
arbuscule structure was found in each AM+ sample (Fig. 63). Moreover, other structures such as
vesicles were observed in seven cases and a coil only once (Tab. 29). The mycorrhiza structures
that were found in I. parviflora are typical for the arbuscular type (Fig. 64). There are significant
difference in the frequency of AM+ plants of Impatiens parviflora among forest communities (G =
11.03, p = 0.011). AM+ plants were encountered most frequently in oak-hornbeam patches and
they were rarest in mixed coniferous forests (Fig. 63).
Tab. 28. The number of samples per site and the forest type in the study area. The number of
samples where AM+ plants were found are shown in parentheses (Chmura, Gucwa-Przepióra 2012,
supplemented)
Nature reserve
Bukowica
(50˚04’43’’N, 19˚23’55’’E).
Dolina Eliaszowki
(50˚10’18’’N, 19˚38’02’’E).
Dolina Kluczwody
(50˚09’54’’N, 19˚49’10’’E).
Dolina Racławki
(50˚09’49’’N, 19˚41’33’’E).
Lipowiec
(50˚04’42’’N, 19˚26’38’’E).
Skała Kmity
(50˚06’06’’N, 19˚48’38’’E).
Wąwóz Bolechowicki
(50˚09’30’’N, 19˚46’55’’E).
Total
FA
Forest type
DF
TC
3(1)
2(2)
3(3)
QP
2
2
3
1(1)
1
4
4
2
3(3)
2(2)
2(2)
1(1)
8
9
5
1(1)
1(1)
3
Total
1(1)
6
4
4
7
6
30
Some samples were characterized by maximum values of mycorrhiza colonization such as F% –
100% and (a%) – arbuscule richness in root fragments where arbuscules were present – 94 (Tab.
30). These samples originated from floodplain forests. The sites where AM+ plants were
encountered were characterized by a higher concentration of magnesium and higher values of soil
reaction (Chmura, Gucwa-Przepióra 2012, Tab. 31).
135
Tab. 29. Presence of mycorrhiza structure in particular samples
AMAM+
0
1
2
3
4
5
6
7
No sample
arbuscules vesicles coils
code of forest community
1
x
FA
2
x
FA
3
x
FA
4
x
DF
5
x
DF
6
x
x
x
FA
7
x
x
FA
8
x
TC
9
x
x
TC
10
x
TC
11
x
TC
12
x
TC
13
x
x
DF
14
x
x
TC
15
x
DF
16
x
x
DF
17
x
x
QP
18
x
TC
DF – beechwood forests Dentario glandulosae-Fagetum, TC – oak hornbeam forests TilioCarpinetum, FA – alder forests Fraxino-Alnetum, QP – mixed coniferous forests Querco roborisPinetum
DF
FA
QP
TC
Fig. 63. Frequency of the presence of arbuscular mycorrhiza in Impatiens parviflora in the
forest communities that had been distinguished. Explanations: DF – beechwood forests Dentario
glandulosae-Fagetum, FA – alder forests Fraxino-Alnetum, QP – mixed coniferous forests Querco
roboris-Pinetum, TC – oak hornbeam forests Tilio-Carpinetum
136
The AM+ from floodplain forests Fraxino-Alnetum were typified by higher values of all of the
mycorrhization indices in comparison with the three remaining forest communities. The lowest
values were observed in fertile beech forests (Tab. 32 after Chmura, Gucwa-Przepióra 2012).
Fig. 64. Presence of arbuscules in roots of Impatiens parviflora (photo by Gucwa-Przepióra)
Tab. 30. Descriptive statistics of arbuscular mycorrhization indices
F
M
m
Mean 57.51 28.03 39.93
SD
29.86 23.27 25.31
min
3.33 0.16
3.4
max
100 76.33 76.33
a
A
68.20 22.01
19.17 20.84
26.47 0.08
94.44 68.5
Tab. 31. Differences in the soil properties of samples from forests where mycorrhizal individuals
(AM+) and non-mycorrhizal individuals of I. parviflora were present (after Chmura, GucwaPrzepióra 2012, modified)
Soil parameter
pH
pHKCL
C
Loss on ignition (%)
NT
C:N
P (mg/100 g)
Mg (mg/100 g)
K (mg/100 g)
Ca (mg/100 g)
Na (mg/100 g)
CaCO3 (mg/100 g)
AM+
5.9±1.5
5.4±1.6
4.9±2.9
12.0±5.6
0.3±0.1
15.6±3.5
3.0±2.9
6.7±9.2
6.0±5.5
422.0±265.2
3.1±0.7
1.5±3.8
137
AM4.7±1.6
4.2±1.6
5.0±2.3
12.0±3.3
0.4±0.3
16.6±5.5
8.2±16.1
2.5±3.8
4.7±2.5
341.2±461.4
3.5±0.7
1.9±5.7
P
*
*
ns
ns
ns
ns
ns
*
ns
ns
ns
ns
Tab. 32. Comparison of mycorrhiza colonization parameters (Mean±SE) between studied forest
communities (after Chmura, Gucwa-Przepióra 2012, modified)
F
M
m
a
A
AU
66.8±4.5a 34.7±3.5a 40.1±3.4a 74.9±3.9a 30.0±3.1a
DF
13.3±2.9c 6.4±1.6c 16.4±4.1c 19.5±4.2c 3.9±1.0c
TC
35.7±5.0b 13.9±2.9b 21.1±3.4b 45.7±3.7b 9.6±2.2b
QP
21.9±5.8b 11.9±3.6b 17.1±4.8c 22.2±6.0c 8.9±2.9b
DF – beechwood forests Dentario glandulosae-Fagetum, TC – oak hornbeam forests TilioCarpinetum, FA – alder forests Fraxino-Alnetum, QP – mixed coniferous forests Querco roborisPinetum
The main difference between the present study and the work by Chmura and Gucwa-Przepióra
(2012) is related to the analysis of data that was obtained. A restricted analysis or choice of method
of analysis can cause the loss of some data and an underestimation of the results. Using the new
F% approach seems to be a more explanatory variable to explain the vegetative and generative
attributes of plants. It explains the number of flowers and fruits and some leaf morphology traits
better. The remaining AM indices were positively correlated with the height of plant, whereas in
the original study (Chmura, Gucwa-Przepióra 2012) index a% and m% did not correlate with any
of the morphometric attributes (Tab. 33).
Obviously, morphometric variables are correlated with each other. The taller the plant, the higher
the number of flowers and seeds (Coombe 1956). AM colonization indices were also correlated
with each other. However, it is very interesting to what extent the results show a statistical artifact
or causal associations. Can the presence of arbuscules or their quantity stimulate the development
of flowers or seeding seeds? Does AM influence biomass allocation? The answers to these
questions in the future in relation to invasive species can contribute to wider knowledge about
plant invasions. A series of experiments with the inoculation of AM fungi under controlled
conditions (laboratories, greenhouses) and the observation of life history traits or biomass
allocation may be useful and help to solve some of the problems.
This study and the one done by Chmura and Gucwa-Przepióra (2012) demonstrated that a greater
number and more fecund individuals of I. parviflora were characterized by higher values of the
AM colonization indices. It was therefore inferred that AM may have caused the gregariousness
and fecundity of the species. There is another possible explanation for this phenomenon. An
opposite pattern is that taller and more fecund plants are colonized by AM fungi more frequently.
138
Tab. 33. Comparison of the Pearson and Spearman rank correlation coefficients between the soil
variables and mean morphometric features of I. parviflora and mycorrhiza colonization (P<0.05),
(after Chmura, Gucwa-Przepióra 2012, modified). The results between pooled AM colonization
and morphometric features of individuals in the sample are shown in brackets.
(F%) – mycorrhizal frequency (M%) – relative mycorrhizal root length, (m%) – intensity of
colonization within individual mycorrhizal roots, (A%) – relative arbuscular richness, (a%) –
arbuscule richness in root fragments where arbuscules were present, ns – non-significant
Soil properties
pH(H2O)
pH(KCL)
C:N
Ca (mg/100g)
Mean height of plants (height of a
plant)
Mean number of flowers (number of
flowers)
Mean number of fruits (number of
fruits per a plant)
Mean length of leave (length of leave
per a plant)
Mean width of leaves (width of leave
per a plant)
Mycorrhizal parameters
F%
A%
0.56
ns
0.58
ns
-0.51
ns
ns
ns
a%
0.64
0.65
-0.56
0.51
M%
ns
ns
ns
ns
m%
ns
ns
ns
ns
0.62(0.28) 0.77(0.21) ns(0.32) 0.71(0.21) ns(0.15)
ns(0.15)
0.74(ns)
ns(ns)
0.69(ns)
ns(ns)
ns(0.14)
0.76(ns)
ns(ns)
0.71(ns)
ns(ns)
ns(0.14)
0.66(ns)
ns(ns)
0.62(ns)
ns(ns)
ns(0.15)
0.65(0.13) ns(0.21) 0.62(0.12)
ns(ns)
All congeners within the genus Impatiens in Europe are not equally mycorrhizal plants. Himalayan
balsam is considered to be sparsely colonized by the arbuscular mycorrhizal fungi AMF (Tanner,
Gange 2013). Moreover, its presence in a community may affect AMF and decrease their
percentage of colonization and as a consequence, this may lead to a decrease in the shoot biomass
of native plants. However, mycorrhiza in this plant was not investigated in that study. In a previous
work (Chmura, Gucwa-Przepióra 2012), the hypothesis was that when small balsam is colonized
by AM fungi, it can be a beneficent of AM presence. There is no information about its impact on
AMF in the soil. If Himalayan balsam affects microbial community, it is possible that role of
I. parviflora in a plant community is similar. If small balsam, under favorable conditions, is more
frequently colonized by AMF that other species in a community, this can indirectly have an impact
on the remaining plant species that are present. Future research should focus on AMF in coexisting
species on sites that have invaded and those that are uninvaded by I. parviflora in order to explain
the role of arbuscular mycorrhiza in the invasiveness of the species.
Other fungi-like pathogenic fungi and dark septate endophytes (DSE) were absent in the
samples that were investigated. DSE are sometimes reported to be present only in non-mycorrhizal
plants and a negative correlation between AM and DSE fungal colonization has been observed
(Chaundry et al. 2009; Muthukumar, Vediyappan 2010). Thus, it is possible that these fungi do not
139
colonize the roots of I. parviflora; however, further research should be conducted in order to solve
this problem. This is important because parasitic ones are also known among the various symbiotic
interactions of DSE with plants (Jumpponen 2001). The absence of pathogenic fungi and
facultative pathogenic fungi in roots might also enhance the invasion success of small balsam.
A reduced control by natural enemies in the introduced range compared with the native range,
which is known as the enemy release hypothesis, is believed to be one of the important causes of
alien plant species invasion (Colautti et al. 2004).
3.3.5. Interactions with coexisting plant species
Changes in species composition of plant communities invaded
A cluster analysis (CA) of the particular plots in the subsequent years based on the frequency
of all of the species that were present on the study plots revealed different patterns of changes on
the sites (Fig. 65). The vegetation data of native species was the most similar in the first two years
on study plots 1, 3,4, 6, 7, 8 and 10 (Fig. 65A,C,D,F,G,H,I respectively), whereas on the other
study plots – 2, 5 and 9, distant years occurred in the same clusters (Fig. 65B,E,J). It could have
been expected that the successive years would have be close in common clusters or in the closest
clusters. Assuming that time is the only factor that affects distance among the subsequent
vegetation seasons, then a cluster analysis of study plot 1 would be a good example of such
a phenomenon (Fig. 65A). There, the nearest years are grouped close to each other. All of the study
plots from the initial invasion group and saturation invasion group were found in the first group in
which the first two vegetation years were the most similar. The second group, which included
study plots 2, 5 and 9, belongs to the intermediate invasion group. This group underwent the most
significant changes in the population dynamics of small balsam, which could be reflected in the
frequency and abundance of accompanying species.
There were significant differences in the total mean cover of native plants among the
vegetation seasons on all of the study plots. Changes even occurred in two subsequent years on six
of the study plots (3, 5-9) (Fig. 68, 70-74).
140
A
50
Height
100
IV
I
VII
III
V
II
VIII
150
dist(dane)
hclust (*, "ward")
II
40
VI
250
0 50
V
IV
I
III
VIII
VII
dist(dane)
hclust (*, "ward")
I
II
VI
D
60
80 100
C
II
I
50
VII
V
VI
II
IV
III
I
0
IV
I
III
V
II
VIII
VI
100
150
dist(dane)
hclust (*, "ward")
VII
Height
VIII
40
60
80
dist(dane)
hclust (*, "ward")
20
0
VII
F
E
Height
IV
III
VI
V
VIII
IV
III
VIII
VI
V
VII
0
0
20
Height
150
Height
60
0 20
Height
100
350
B
Fig. 65. Cluster analysis (Euclidean distance, Ward method) of successive stages of
vegetation based on thedist(dane)
total cover of species on study plots 1–6. Explanations:
dist(dane)I – 2005…VIII –
2012
hclust (*, "ward")
hclust (*, "ward")
141
G
40
30
20
Height
V
VIII
VIII
VII
dist(dane)
hclust (*, "ward")
VI
III
VII
V
IV
0
III
VII
VI
IV
V
VIII
II
I
0
20
20
II
40
40
Height
60
60
80
dist(dane)
hclust (*, "ward")
80
VI
IV
III
I
VII
0
IV
V
VI
VIII
II
I
0
II
10
III
J
I
Height
I
100
50
Height
150
50
200
H
Fig. 65 continued. Cluster analysis (Euclidean distance, Ward method) of successive stages
of vegetation based on the total cover of species in study plots 7–10. Explanations: I – 2005…VIII
– 2012
dist(dane)
dist(dane)
hclust (*, "ward")
hclust (*, "ward")
These variations were accompanied by significant changes in the values of the species
richness of native taxa as well as the biodiversity indices – Shannon-Wiener and Simpson indexes.
Non-significant differences in the Shannon evenness index was noted on study plots 2, 3 and 8.
The quantitative variation in species richness and diversity does not always seem to be related to
the dynamics of the population of small balsam. For instance, on study plot 2-4, 5, 7-8 and 10, the
curve of the mean cover of the species over the years of the study is more or less constant and it
does not affect the biodiversity of native species irrespective of the value of the cover. On study
plot 6, an increase in the biodiversity indices were recorded simultaneously with a decrease in the
mean cover of Impatiens parviflora (Fig. 71). On the other hand, a decrease in the mean cover of
small balsam was accompanied by a decrease in the biodiversity indices although the total cover of
native plants increased (Fig. 66).
142
Fig. 66. Changes in the total mean cover of native plants (Friedman test) against the mean
cover of Impatiens parviflora (A) species richness of native plants (B) and the biodiversity indices:
H – Shannon-Wiener, E – Evenness, D – Simpson index (C) on study plot 1. *** – p<0.001, I –
2005…VIII – 2012
143
Fig. 67. Changes in the total mean cover of native plants (Friedman test) against the mean
cover of Impatiens parviflora (A) species richness of native plants (B) and the biodiversity indices:
H – Shannon-Wiener, E – Evenness, D – Simpson index (C) on study plot 2. *** – p<0.001, NS –
non-significant, I – 2005…VIII – 2012
144
Fig. 68. Changes in the total mean cover of native plants (Friedman test) against the mean
cover of Impatiens parviflora (A) species richness of native plants (B) and biodiversity indices:
H – Shannon-Wiener, E – Evenness, D – Simpson index (C) on study plot 3. Arrows indicate
significant differences between adjacent years. *** – p<0.001, NS – non-significant, I –
2005…VIII –2012
145
Fig. 69. Changes in the total mean cover of native plants (Friedman test) against the mean
cover of Impatiens parviflora (A) species richness of native plants (B) and biodiversity indices:
H – Shannon-Wiener, E – Evenness, D – Simpson index (C) on study plot 4. Arrows indicate
significant differences between adjacent years. ** – p<0.01, *** – p<0.001, I – 2005…VIII – 2012
146
Fig. 70. Changes in the total mean cover of native plants (Friedman test) against the mean
cover of Impatiens parviflora (A) species richness of native plants (B) and the biodiversity indices:
H – Shannon-Wiener, E – Evenness, D – Simpson index (C) on study plot 5. Arrows indicate
significant differences between adjacent years. *** – p<0.001, I – 2005…VIII – 2012
147
Fig. 71. Changes in the total mean cover of native plants (Friedman test) against the mean
cover of Impatiens parviflora (A) species richness of native plants (B) and the biodiversity indices:
H – Shannon-Wiener, E – Evenness, D – Simpson index (C) on study plot 6. Arrows indicate
significant differences between adjacent years. *** – p<0.001, I – 2005…VIII – 2012
148
Fig. 72. Changes in the total mean cover of native plants (Friedman test) against the mean
cover of Impatiens parviflora (A) species richness of native plants (B) and biodiversity indices:
H – Shannon-Wiener, E-Evenness, D-Simpson index (C) on study plot 7. Arrows indicate
significant differences between adjacent years are shown. *** – p<0.001, I – 2005…VIII – 2012
149
Fig. 73. Changes in the total mean cover of native plants (Friedman test) against the mean
cover of Impatiens parviflora (A) species richness of native plants (B) and the biodiversity indices:
H – Shannon-Wiener, E – Evenness, D – Simpson index (C) on study plot 8. Arrows indicate
significant differences between adjacent years. *** – p<0.001, NS –nonsignificant, I – 2005…VIII
– 2012
150
Fig. 74. Changes in the total mean cover of native plants (Friedman test) against the mean
cover of Impatiens parviflora (A) species richness of native plants (B) and the biodiversity indices:
H – Shannon-Wiener, E – Evenness, D – Simpson index (C) on study plot 9. Arrows indicate
significant differences between adjacent years. *** – p<0.001, I – 2005…VIII – 2012
151
Fig. 75. Changes in the total mean cover of native plants (Friedman test) against the mean
cover of Impatiens parviflora (A) species richness of native plants (B) and the biodiversity indices:
H – Shannon-Wiener, E – Evenness, D – Simpson index (C) on study plot 10. Arrows indicate
significant differences between adjacent years. *** – p<0.001, I – 2005…VIII – 2012
152
The changes in species composition and the cover of native species may have had many
different causes. The body of literature on this topic is enormous. There are many examples of
long-term studies on permanent plots including forest communities (Faliński 1993 after Falińska
2004; Durak 2009; Dierschke 2006; Łysik 2008, 2009; Kiedrzyński et al. 2011; Chmura et al.
2013) just to mention few. The high degree of variation concerning species cover over time very
often reflects changes that are due to the natural process of succession and regeneration after
disturbance, climatic conditions or biocenotic factors. These are mainly air temperature,
precipitation or a massive gradation of rodents or insects that are seed predators, respectively. In
relatively ecologically stable communities, fluctuations can occur in an herb layer that has been
built up by perennial species, for instance geophytes (Anemone nemorosa), chamaephytes
(Galeobdolon luteum) or hemicryptophytes (Milium effusum) (Towpasz, Tumidajowicz 1989).
Sometimes these phenomena occur within a short period of time, e.g., four years, in populations of
Oxalis acetosella, Anemone nemorosa and Galebdolon luteum (Pirożnikow 1991). Falińska (1991)
stated that that succession and the process of the turnover of species is a result of the demographic
processes within a population of plants. A similar approach was proposed by Hubbell (2001), who
used the term ecological drift to describe the sum of demographic stochasticities for each
population in a community. Moreover, he wanted to treat all of the components in a community at
the individual level rather than at the species level, which is typical for a neutral versus niche
concept of plant communities.
In the present study it was difficult to relate the changes in species composition or species
diversity to the dynamics of I. parviflora. If there is a casual relationship, then it appears that small
balsam is affected species rather than vice versa.
Relations with coexisting species in ground forest flora
A complete list of accompanying species is presented in Appendix 2 based on the number of
occurrences on the permanent study plots in the Silesian Upland and the Jurassic Upland.
A percentage of frequency was assigned to each species. The most frequent coexisting species are:
Oxalis acetosella, Galebdolon luteum, Galium odoratum, Asarum europaeum, Aegopodium
podagraria, Viola reichenbachiana and tree seedlings of Fagus sylvatica and Acer
pseudoplatanus.
Thirty-three species that are indicators of microsites that have been invaded by small balsam are
presented in Table 34. The highest number of indicator species was detected for floodplain forests
including Urtica dioica, a nitrophilous species on invaded sites, and Myosotis palustris, which
prefers more humid places than small balsam. Hedera helix occupies uninvaded places in
153
thermophilous beech woods, whereas Oxalis acetosella accompanies I. parviflora. In some cases,
the same species are indicators of either invaded or uninvaded sites depending on type of forest
community. One example is Maiantheum bifolium, which occupies invaded sites in acidophilous
beech and mixed coniferous forests and uninvaded sites in oak-hornbeam forests. In some
phytocoenoses of Tilio-Carpinetum, the grass Melica nutans more frequently occupies invaded
sites, whereas in other patches in they occur on uninvaded sites.
Tab. 34. Values of the species indicator IndVal for sites that were invaded or uninvaded by
I. parviflora in the forest communities that had been distinguished
AU Urtica dioica
Euonymus europaeus c
Festuca gigantea
Lamium maculatum
Oxalis acetosella
Ajuga reptans
Poa palustris
invaded
0.856
0.847
0.707
0.707
0.707
0.69
0.63
0.001
0.001
0.001
0.001
0.001
0.001
0.001
uninvaded
Myosotis palustris
Mentha longifolia
Petasites hybridus
Stachys sylvatica
Rubus idaeus
-
0.835
0.707
0.707
0.7
0.609
0.001
0.001
0.001
0.001
0.001
CF
CF
Oxalis acetosella
Fagus sylvatica c
0.827 0.001 Hedera helix
0.645 0.017 -
DF
Hieracium murorum
Circaea lutetiana
0.669 0.022 0.622 0.025 -
LF
Maianthemum bifolium
0.662 0.001 -
TC
Allaria officinalis
Melica nutans
Impatiens noli-tangere
0.877 0.001 Maianthemum bifolium
0.709 0.001 Aegopodium podagraria
0.668 0.001 Hedera helix
Melica nutans
Mercurialis perennis
Carex brizoides
Athyrium filix-femina
0.649
0.92
0.863
0.63
0.816
0.833
0.63
0.001
0.001
0.002
0.003
0.001
0.001
0.001
QP
Geranium robertianum
Galeobdolon luteum
Maianthemum bifolium
Pteridium aquilinum
Vaccinium myrtillus
0.823
0.664
0.836
0.641
0.637
0.725
0.627
0.656
0.615
0.001
0.003
0.002
0.001
0.001
0.001
0.001
0.003
0.002
Hedera helix
Fagus sylvatica c
Aegopodium podagraria
Convallaria majalis
0.675 0.008
RA Scirpus sylvaticus
0.643 0.001 Chaerophyllum hirsutum
0.673 0.02
Abbreviations: QP – Querco roboris-Pinetum; DF – Dentario glandulosae-Fagetum; TC – TilioCarpinetum; LF – Luzulo pilosae-Fagetum; CF – Carici-Fagetum, AU – communities of the AlnoUlmion alliance; RA – Ribeso nigri-Alnetum, c – in case of trees and shrubs indicate seedlings.
154
The same situation was revealed for beech Fagus sylvatica seedlings. Beech forest sites on which it
was present were invaded by small balsam but in mixed coniferous forests it was the other way
round.
The relationships between Impatiens parviflora and the most frequent species are presented
separately for each region (Tab. 35–36).
A negative correlation was found for, among others, Carex brizoides, Aeogopodium podagraria,
Athyrium felix-femina (Tab. 35) and Carex sylvatica, Convallaria majalis (Tab. 36).
It is difficult to state whether a negative correlation indicates a displacement or a shift in biotopic
requirements based on the values of the correlation coefficients.
Tab. 35. Values of the Spearman rank correlation coefficient between the percentage of the cover
of Impatiens parviflora and the cover of native species – a study of the Jurassic Upland
Species
rS
P
Acer platanoides c
0.36
<.0001
Acer pseudoplatanus c
-0.14
0.0047
Aegopodium podagraria
-0.24
<.0001
Ajuga reptans
0.36
0.0029
Allaria petiolata
0.75
<.0001
Anthriscus silvestris
0.98
<.0001
Athyrium filix femina
-0.20
0.0044
Carex brizoides
-0.36
<.0001
Chaerophyllum aromaticum 0.51
<.0001
Dryopteris filix mas
0.29
0.0323
Epipactis helleborine
0.42
0.0016
Euonymus europaeus c
0.45
0.0006
Fagus sylvatica c
0.39
<.0001
Ficaria verna
0.49
0.0106
Galium odoratum
-0.10
0.043
Hieracium murorum
-0.39
0.0121
Mercurialis perennis
-0.18
0.0457
Oxalis acetosella
-0.09
0.0219
Sanicula europaea
0.16
0.05
Stachys sylvatica
0.30
0.0081
Viola reichenbachiana
0.19
0.0003
c – in case of trees and shrubs indicate seedlings
N
130
393
515
65
60
13
202
190
74
54
55
53
450
26
407
40
124
712
139
75
341
The positive values of the correlation coefficient are mainly related to its coexistence with
seedlings or juvenile individuals of trees or shrubs, e.g., Acer platanoides, Fagus sylvatica and
Euonymus europaeus. Among herbaceous species, those that positively correlated are: Ajuga
reptans, Allaria officinalis, Chaerophyllum aromaticum, Dryopteris filix-mas, Stachys sylvatica,
Geum urbanum, Luzula pilosa and many others (Tab. 35–36).
155
More light can be shed by an analysis of changes in the correlation coefficients of particular study
plots among the years of the research (Tab. 37–44). The changes in values of the correlation
between I. parviflora and Carex brizoides can be an example of possible displacement by the latter
on study plot 1 (Tab. 37). The values of the correlation decreased with decreasing decrease in the
mean cover of small balsam (Fig. 76). Another negative relationship that was observed was the
correlation with Anemone nemorosa.
Tab. 36. Values of the Spearman rank correlation coefficient between the percentage of the cover
of Impatiens parviflora and the cover of native species – a study of the Silesian Upland
rs
P
N
Species
Asarum europaeum
0.56
0.020222
17
Carex sylvatica
-0.53
0.043937
15
Carpinus betulus c
-0.31
0.009054
72
Convallaria majalis
-0.34
0.000410
103
Fragaria vesca
0.28
0.039281
56
Geum urbanum
0.73
0.003008
14
Luzula pilosa
0.22
0.042894
89
Lysimachia vulgaris
0.42
0.012845
34
Oxalis acetosella
0.20
0.000703
292
Rubus hirtus
0.40
0.000358
75
Rubus plicatus
0.60
0.007203
19
Senecio nemorensis
0.72
0.001689
16
Stellaria holostea
0.49
0.015723
24
Vaccinium myrtillus
0.30
0.000001
264
c – in case of trees and shrubs indicate seedlings
Tab. 37. The relations between cover of I. parviflora and selected native species in particular
seasons: a study of plot 1. In the table Spearman rank correlation coefficient are given
Anemone nemorosa
Carex brizoides
1
2
ns
ns
-0.66*** -0.52***
3
ns
-0.33***
4
-0.21*
ns
5
ns
ns
6
ns
ns
7
ns
ns
Abbreviations: * – p<0.01, ** – p<0.01, *** – p<0.001, ns – non-significant
156
8
ns
ns
Fig. 76. Comparison of dynamics tendencies between I. parviflora and selected native species
exemplified by frequency (lef) and means of cover (right) in particular seasons: a study of plot 1
In study plot 4 in particular vegetation season positive correlation was observed with Galium
odoratum in first year and two negative ones with Hieracium murorum in the next year and in the
last year with seedlings of Acer pseudoplatanus (Tab. 38, Fig. 77). In total five species negatively
correlated with Impatiens parviflora in the study plot 5. These are: Fagus sylvatica, Galeobdolon
luteum, Oxalis acetosella, Pulmonaria obscura, Viola reichenbachiana. Other as cover of Hedera
helix at the beginning was positively correlated with cover of small balsam but in the last year
correlation became negative. With slightly increasing mean cover and frequency the participation
of Fagus sylvatica seedlings was decreasing. The possible competition with Viola reichenbachiana
did not result in displacement of any species. The similar situation concern Galeobdolon luteum
(Tab. 39; Fig. 78).
Tab. 38. The relations between cover of I. parviflora and selected native species in particular
seasons: a study of plot 4. In the table Spearman rank correlation coefficient are given
Acer pseudoplatanus c
Galium odoratum
Hieracium murorum
1
ns
0.22*
ns
2
Ns
Ns
-0.25*
3
ns
ns
ns
4
ns
ns
ns
5
ns
ns
ns
6
ns
ns
ns
7
ns
ns
ns
8
-0.23*
ns
ns
Abbreviations: * – p<0.01, ** – p<0.01, *** – p<0.001, ns – non-significant
157
Fig. 77. Comparison of dynamics tendencies between I. parviflora and selected native species
exemplified by frequency (lef) and means of cover (right) in particular seasons: a study of plot 4
In study plot 6 there is clear example of decreasing role of small balsam and increasing
contribution of Fagus sylvatica associated with appearance of seedlings (Tab. 40, Fig. 79). Other
species which increased in frequency within the area and in abundance was Athyrium filix-femina.
Tab. 39. The relations between cover of I. parviflora and selected native species in particular
seasons: a study of plot 5. In the table Spearman rank correlation coefficient are given
1
Fagus sylvatica c
Galeobdolon luteum
Hedera helix
Oxalis acetosella
Pulmonaria obscura
Ribes nigrum c
Viola reichenbachiana
-0.20*
-0.27**
ns
ns
ns
ns
-0.26**
2
-0.23*
ns
0.24*
ns
ns
0.27**
ns
3
ns
ns
ns
ns
-0.20*
ns
ns
4
ns
ns
ns
ns
ns
ns
ns
5
ns
ns
ns
ns
ns
ns
ns
6
-0.33***
ns
ns
ns
ns
ns
ns
7
ns
-0.26**
ns
-0.20*
ns
ns
ns
8
-0.03
-0.29**
ns
-0.23*
ns
-0.27**
Abbreviations: * – p<0.01, ** – p<0.01, *** – p<0.001, ns – non-significant
158
Fig. 78. Comparison of dynamics tendencies between I. parviflora and selected native species
exemplified by frequency (lef) and means of cover (right) in particular seasons: a study of plot 5
The other significant negative correlations with Hedera helix, Maianthemum bifolium and Mycelis
muralis at the beginning of research can be result of quite high frequency in subplots and higher
mean cover of small balsam (Tab. 40).
Tab. 40. The relations between cover of I. parviflora and selected native species in particular
seasons: a study of plot 6. In the table Spearman rank correlation coefficient are given
Athyrium filix-femina
Fagus sylvatica c
Hedera helix
Maianthemum bifolium
Mycelis muralis
Vaccinium myrtillus
1
ns
-0.28**
ns
-0.20*
ns
ns
2
ns
-0.35***
-0.20*
ns
-0.24*
ns
3
ns
ns
ns
ns
ns
ns
4
ns
ns
ns
ns
ns
-0.26**
5
ns
ns
ns
ns
ns
ns
6
ns
ns
ns
ns
ns
ns
7
-0.28**
ns
ns
ns
ns
ns
8
-0.26**
-0.35***
ns
ns
ns
ns
Abbreviations: * – p<0.01, ** – p<0.01, *** – p<0.001, ns – non-significant.
Fig. 79. Comparison of dynamics tendencies between I. parviflora and selected native species
exemplified by frequency (lef) and means of cover (right) in particular seasons: a study of plot 6
159
In study plot 7 frequency and cover of I. parviflora underwent fluctuations but generally it did not
change markedly. Stachys sylvatica correlated negatively and its frequency a little increased,
whereas mean cover decreased (Tab. 41). That could be consequence of exclusion. Native balsam
touch me not Impatiens noli-tangere decreased in profoundly in frequency and its mean cover. In
the last year this species disappeared (Fig. 80).
Tab. 41. The relations between cover of I. parviflora and selected native species in particular
seasons: a study of plot 7. In the table Spearman rank correlation coefficient are given
Carex sylvatica
Fagus sylvatica c
Galeopsis pubescens
Geranium robertianum
Hedera helix
Impatiens noli-tangere
Sanicula europaea
Stachys sylvatica
Viola reichenbachiana
1
-0.32***
ns
ns
ns
-0.38***
0.26**
-0.25*
0.35***
0.29**
2
-0.47***
-0.20*
0.22*
0.24*
-0.39***
0.46***
ns
0.55***
0.41***
3
ns
ns
ns
-0.38***
0.37***
-0.25*
0.39***
0.32**
4
-0.26**
ns
ns
-0.29**
0.27**
ns
0.42***
ns
5
ns
ns
ns
-0.40***
ns
ns
0.39***
ns
6
ns
ns
0.23*
-0.36***
ns
-0.20*
0.42***
ns
7
ns
ns
ns
-0.34***
ns
ns
0.39***
ns
8
ns
ns
ns
-0.29**
ns
0.26**
0.13
Abbreviations: * – p<0.01, ** – p<0.01, *** – p<0.001, ns – non-significant
Fig. 80. Comparison of dynamics tendencies between I. parviflora and selected native species
exemplified by frequency (lef) and means of cover (right) in particular seasons: a study of plot 7
There was significant negative relationship with Hedera helix but it seemed that within study plots
both species tended to occur in different sites. None of both species was displaced. Other negative
correlations are effect of fluctuating dominance of one of the species.
The mean cover of I. parviflora weakly increased in study plot 8 what could be reflected by
negative correlation with decreasing cover of Galeobdolon luteum in the first two and last year
160
(Tab. 42, Fig. 81). The increasing trend in mean cover of Oxalis acetosella was observed what
could lead to higher competition or avoidance in microsites occupancy (Tab. 42, Fig. 81).
Tab. 42. The relations between cover of I. parviflora and selected native species in particular
seasons: a study of plot 8. In the table Spearman rank correlation coefficient are given
Acer pseudoplatanus c
Aegopodium podagraria
Galeobdolon luteum
Oxalis acetosella
Sorbus aucuparia c
1
ns
ns
0.21*
ns
ns
2
ns
ns
-0.21*
ns
ns
3
0.21*
-0.24*
ns
ns
ns
4
ns
ns
ns
-0.20*
-0.21*
5
ns
ns
ns
ns
ns
6
ns
ns
ns
ns
ns
7
ns
ns
ns
ns
ns
8
ns
ns
0.22*
-0.25*
ns
Abbreviations: * – p<0.01, ** – p<0.01, *** – p<0.001, ns – non-significant
On study plot 9 the increase in the frequency and mean cover by I. parviflora led to negative
correlations with five coexisting species in the fifth year of the study (Tab. 43, Fig. 82). Despite
a very high increase in mean abundance on the subplots, no considerable decreases in other
coexisting species were observed.
Fig. 81. Comparison of dynamics tendencies between I. parviflora and selected native species
exemplified by frequency (lef) and means of cover (right) in particular seasons: a study of plot 8
On study plot 10 fifth vegetation season was critical because so many as five species negatively
correlated with I. parviflora and two species as Galeobdolon luteum and Galium odoratum were
positively correlated (Tab. 44).
161
Tab. 43 The relations between cover of I. parviflora and the selected native species in particular
seasons: a study of plot 9. The Spearman rank correlation coefficient is presented in the table
Fagus sylvatica c
Galeobdolon luteum
Galium odoratum
Hedera helix
Mycelis muralis
1
-0.27*8
ns
ns
-0.20*
ns
2
ns
ns
0.35***
-0.29**
ns
3
ns
ns
0.36***
-0.20*
ns
4
ns
ns
0.24*
-0.19
ns
5
-0.22*
0.32**
0.32**
-0.29**
0.27**
6
ns
ns
ns
ns
ns
7
ns
-0.20*
ns
ns
ns
8
ns
ns
0.37***
-0.33***
ns
Abbreviations: * – p<0.01, ** – p<0.01, *** – p<0.001, ns – non-significant
Fig. 82. Comparison of dynamics tendencies between I. parviflora and selected native species
exemplified by frequency (lef) and means of cover (right) in particular seasons: a study of plot 9.
Only Asarum europaeum was plant which weakly decreased in abundance in subplots along
increase of cover of small balsam (Fig 83), however, its frequency did not dropped. The increasing
role of Viola reichenbachiana resulted in appearance of negative correlation with I. parviflora but
probably both species tended to avoid each other.
Tab. 44. The relations between cover of I. parviflora and selected native species in particular
seasons: a study of plot 10. In the table Spearman rank correlation coefficient are given
Aegopodium podagraria
Asarum europaeum
Circaea lutetiana
Galeobdolon luteum
Galium odoratum
Maianthemum bifolium
Oxalis acetosella
Sanicula europaea
Viola reichenbachiana
1
ns
ns
-0.28**
ns
ns
ns
ns
ns
2
-0.39***
ns
ns
-0.35***
0.21***
-0.20*
ns
ns
ns
3
-0.30***
-0.36***
-0.34***
ns
ns
ns
ns
ns
ns
4
ns
ns
ns
ns
-0.31**
ns
ns
-0.22*
ns
5
ns
-0.25*
-0.17*
0.14*
0.22*
-0.26**
ns
-0.26**
-0.28**
6
ns
-0.34***
-0.32**
ns
0.20*
-0.34***
ns
ns
-0.29**
7
-0.21*
-0.27**
ns
-0.47***
-0.26**
-0.29**
ns
-0.29**
-0.15
Abbreviations: * – p<0.01, ** – p<0.01, *** – p<0.001, ns – non-significant
162
8
ns
ns
ns
ns
ns
ns
ns
ns
-0.24*
Fig. 83. Comparison of dynamics tendencies between I. parviflora and selected native species
exemplified by frequency (lef) and means of cover (right) in particular seasons: a study of plot 10
Relations with coexisting species in microhabitats of dead wood complex
The processes of coexistence in the microhabitats that had been distinguished were
associated with coarse woody debris as areas near log, logs under canopies, logs under canopy
opening, root plates, stumps, treefall disturbances and hollows went in different way. Positive
correlation was found for two time series with cover of I. parviflora with number of native species
occupying the microhabitats that had been distinguished. Only in April negative relationship with
total density of native plants was noted (Tab. 45).
Tab. 45. Relationships (Spearman rank correlation) between density of Impatiens parviflora and
species richness and total density of native species in all in microhabitats associated with coarse
woody debris complex combined
Date
16.04.07
18.05.07
14.06.07
17.07.07
17.08.07
22.09.07
Number of native species
ns
0.63***
ns
ns
0.68**
ns
Total density of native species
-0.42*
0.59***
0.39*
ns
0.76***
ns
Both the native cover and density of I. parviflora were several fold higher than the total density of
native species in all of the microhabitats that were analyzed except for root plates and snags (Fig.
88-89). The highest mean density of I. parviflora individuals in areas near logs occurred in July
when the native species were the least abundant (Fig. 84). A similar trend was recorded in hollows
when the mean density was the highest in June and July (Fig. 85).
163
The total density of plants on logs was low and was similar both under canopies and canopy
openings (Fig. 86-87) although in conditions with better light availability, the density of small
balsam did not decrease quickly and the mean native species survived until the last time series. The
highest mean density of small balsam, which exceeded 100 individuals/seedlings, was observed in
treefall disturbances although the total density of native plants was not high and was the largest in
mid-June (Fig. 90).
Fig. 84. Comparison of the mean number of shoots of Impatiens parviflora and the density of
native species and the number of native species during a particular time series in areas near logs.
Abbreviations: I – 16.04.07, II –18.05.07, III – 14.06.07, IV – 17.07.07, V – 17.08.07, VI –
22.09.2007
Fig. 85. Comparison of the mean number of shoots of Impatiens parviflora and the density of
native species and the number of native species during a particular time series in hollows.
Abbreviations: I – 16.04.07, II –18.05.07, III – 14.06.07, IV – 17.07.07, V – 17.08.07, VI –
22.09.2007
164
Fig. 86. Comparison of the mean number of shoots of Impatiens parviflora and the density of
native species and the number of native species during a particular time series on logs under
canopies. Abbreviations: I – 16.04.07, II –18.05.07, III – 14.06.07, IV – 17.07.07, V – 17.08.07, VI
– 22.09.2007
Fig. 87. Comparison of the mean number of shoots of Impatiens parviflora and the density of
native species and the number of native species during a particular time series on log under canopy
openings. Abbreviations: I – 16.04.07, II –18.05.07, III – 14.06.07, IV – 17.07.07, V – 17.08.07,
VI – 22.09.2007
165
Fig. 88. Comparison of the mean number of shoots of Impatiens parviflora and the density of
native species and the number of native species during a particular time series on root plates.
Abbreviations: I – 16.04.07, II –18.05.07, III – 14.06.07, IV – 17.07.07, V – 17.08.07, VI –
22.09.2007
Fig. 89. Comparison of the mean number of shoots of Impatiens parviflora and the density of
native species and the number of native species during a particular time series on stumps.
Abbreviations: I – 16.04.07, II –18.05.07, III – 14.06.07, IV – 17.07.07, V – 17.08.07, VI –
22.09.2007
166
Fig. 90. Comparison of the mean number of shoots of Impatiens parviflora and the density of
native species and the number of native species during a particular time series on treefall
disturbances. Abbreviations: I – 16.04.07, II –18.05.07, III – 14.06.07, IV – 17.07.07, V –
17.08.07, VI – 22.09.2007
One can find various reports about the possible negative impact of I. parviflora on other native
species or the negative inhibitory effect of native species on small balsam in the literature.
Results by Dobravolskaitė (2012) showed that species such as Maianthemum bifolium, Hepatica
nobilis and Galium aparine do not occur on sites where Impatiens parviflora is most abundant.
Only strong competitors such as Urtica dioica and Rubus idaeus occur when the coverage of
I. parviflora is high. In the oak-hornbeam forest of the Wielkopolski National Park, Ficaria verna,
Carpinus betulus self seedlings, Vaccinium myrtillus, Galium odoratum, Pteridium aquilinum and
tuft grasses were found to have a locally strong inhibitory effect on small balsam growth (Piskorz,
Klimko 2007). The same species with the exception of G. odoratum were observed by Hejda
(2012) in the forests of Central Bohemia. Other plants, namely, Aegopodium podagraria, Athyrium
filix-femina, Dryopteris filis-mas, Fragaria moschata, Luzula luzuloides and Poa nemoralis were
listed as those that can work as a biocenotic barrier that prevents invasions of small balsam.
According to Łysik (2008), after ten years Impatiens parviflora replaced four species: mainly
Galium odoratum, but also Asarum europaeum, Mercurialis perennis and Galeobdolon luteum. As
she claims the changes that were observed could be a result of natural succession. The species
composition had changed from a beech forest to an oak-hornbeam forest. Therefore, it is hard to
state that species turnover may be caused by the putative competition ability of small balsam.
Trepl (1984) indicated several species that could be affected by I. parviflora. These are: Impatiens
noli-tangere and perhaps Ficaria verna, Glechoma hederacea and Corydalis cava. Grass species
hinder the growth of the species, e.g., Dactylis polygama, Arrenatherum elatius and Avenella
flexuosa, all of which are present in mixed and coniferous forests. Some species are considered to
167
be indifferent, e.g., Urtica dioica, Oxalis acetosella and Pteridium aquilinum. Hegi (1965) also
believed that I. noli-tangere could be outcompeted, whereas others such as Moehringia trinervia,
Aegopodium podagraria, Geranium robertianum, Galeopsis speciosa and Viola riviniana do not
interact with small balsam.
As regards the native congener I. noli-tangere, theoretically small balsam shares the same
resources and natural enemies and enters into similar plant communities. The results of various
studies are ambiguous; for instance Godefroid and Koedam (2010) revealed a significant positive
correlation between the cover of small balsam and touch me not balsam, whereas Vervoort et al.
(2012) reported a significant negative correlation. Łysik (2008) noted an increase in I. noli-tangere
after ten years in spite of an increase in the abundance of I. parviflora. In the present study native
balsam had a lower cover that systematically decreased on one study plot and finally the species
disappeared although its cover was positively correlated with I. parviflora. This indicates that it
occupies the same microsites. This was also confirmed by IndVal analysis (Tab. 34). It is not
certain that the species was overgrown. Trepl (1984) believes that in moist habitats I. noli-tangere
keeps the area and dominance whereas can only be overgrown at its suboptimal conditions, i.e.
drier habitats. This study and an analysis of the literature confirm this statement.
Previous studies showed that Carex brizoides can outcompete I. parviflora (Chmura, Orczewska
2004) or that these two species simply avoid each other and differ in the microsites they occupy in
oak-hornbeam forests (Chmura, Sierka 2007). Small balsam is overgrown when they are both
present, which was shown by the present study where it was observed in the Silesian Upland
(Sierka, pers. comm).
It was demonstrated in an analysis of species interactions over time that perennial species, tree
seedlings, geophytes, chamaephytes and hemicryptophytes can constitute a barrier against the
development of I. parviflora. A decrease or displacement of small balsam on microsites depends
on fluctuations in the abundance of these plants. A positive cover with other annuals such as
Galeopsis pubescens or Geranium robertianum (annual but sometimes hemicryptophyte) can be
explained by the fact that these plants use gaps in the herb layer in the same way as I. parviflora
does. They have a similar Grime’s strategy: – CR and CSR, respectively.
It was previously reported that I. parviflora is one of the most common colonizing species on
dead beech wood (Chmura 2008c) and hornbeam Carpinus betulus, pine Pinus sylvestris (Piskorz
and Klimko 2001) and oaks (Quercus sp) (Nowińska et al. 2009). Staniaszek-Kik and Żarnowiec
(2012) found the species on 44 elements from among the many various structural elements of dead
wood (logs, snags, stumps, treefall disturbances) of Fagus sylvatica and Picea abies in the Sudeten
Mts., mainly on the logs and stumps of Fagus sylvatica. It was the most frequent alien species.
168
No competition was observed among plants that occur on dead logs and other types of similar
microhabitats during the vegetation season except for April, during which many of Impatiens
parviflora seedlings appeared. In some microhabitats such as dead logs and hollows, the number
and density of accompanying species was very low, which has an influence on correlation tests.
The conditions that occur in microhabitats that are associated with coarse dead wood are not
suitable for most plants that are encountered on forest floors due to a lack of moisture and
nutrients, thus species that represent a mixed S-R strategy such as I. parviflora can thrive in such
conditions.
169
4. Synthesis
4.1. Traits of invasive species
There have been a few attempts to characterize the plant traits that promote invasiveness or
weediness. One of the first and most widely cited was the proposal by Baker (1965, 1974), who
presented a set of twelve properties of an ideal invader. Later, Baker’s list of properties was
criticized by Fitter et al. (1990) and Perrins et al. (1992) for the vagueness of the plant traits that
had been defined. Perrins et al. (1993) gave more detailed criteria by which Baker’s characters
could be scored. These were, among others, 1) no chilling or specific light requirements are
required to break dormancy; 2) seed bank type 2 (short-term persistent) and 3 (long-term
persistent); 3) relative growth rate of least 1.9 per week, 4) not all meristems are devoted to
flowering, 5) neither exclusively inbreeding or outbreeding, 6) use the wind or a wide range of
insects for pollination 7), has an average of more than 2,500 seeds per plant and 12) can overgrow
other plants. Character 8 or as originally quoted after Baker (1974): “produces some seeds in
a wide range of environmental conditions: tolerant and plastic” was impossible to score, whereas
10 and 11 (not presented here), which are related to vegetative growth properties, are only
applicable to perennials.
Tab. 46. Set of plant traits considered to be beneficial for species invasiveness (Baker 1965,
Rejmánek 2000, Rejmánek et al. 2005) vs traits of Impatiens parviflora
Biological character of invading plant
Prediction
Individual fitness homoeostasis (=phenotypic plasticity), (=”general purpose
Yes
genotype”)
Small seed size
No
High relative growth rate of seedlings and large specific leaf area
Yes
Vertebrate dispersal
Yes
The size of native geographical range: (the larger, the more invasive the species)
No
Vegetative reproduction
No
No congener(s) in the new invaded area
No
The ability to utilize generalist mutualists
Yes
Efficient competition for limited resources (large height, long roots)
No
Characters that favor passive dispersal by humans
Yes
170
Based on these properties Perrins et al. (1993) attributed characters 3 and 5 to I. glandulifera,
which is more invasive in the British Isles. Small balsam, which is a less invasive species, was not
evaluated. Further studies summarized the data and reviewed other attempts to distinguish plant
traits that are responsible for invasiveness (Rejmánek 2000, Rejmánek et al. 2005). The compiled
properties are included in Table 46. Only those characters were included that can be estimated in
regard to the invasiveness of I. parviflora.
Since Baker (1965) it has been widely stressed that phenotypic plasticity (Tab. 45) plays a crucial
role in species invasiveness. As regards small balsam, Skálová et al. (2013) gave the results of
their research on competition among three Impatiens spp in a greenhouse experiment. Small
balsam showed the highest plasticity followed by I. capensis and I. glandulifera and I. nolitangere. When phenotypic plasticity was taken into account as an invasiveness predictor for
balsams, it turned out that I. glandulifera does not fully fit to these criteria as being the most
invasive in terms of the competition ability and impact. Indeed, as has been shown I. parviflora is
a highly diverse species morphologically. In the present study it was demonstrated that its
plasticity is nonequivalent in various habitats (Fig. 91). The following methods adapted by
Elemans (2004) and Skálová et al. (2013) showed that plasticity is the highest in populations that
grow along forest paths. The lowest variation was revealed in a floodplain forest in spite of the
quite high values of plant size and fecundity, although the individuals that grow there are quite
stable in relation to the features studied. The proportional phenotypical plasticity is much higher
than the one that was described by Skálová et al. (2013). It is the result of other traits that were
included into the computations. For instance, number of flowers and fruits were very variable
among and within a specific type of habitat. For the purpose of estimating the latter two characters,
only generative individuals were measured.
The next feature is a relatively small seed size (Tab. 46). Despite differences in the mean weight of
I. parviflora between the countries – Poland and Hungary that were described in Chapter 3.2.1., the
mean mass of I. parviflora was rather high. According to Csontos (2000), who gathered
information about the mass seeds for 1,676 species of Hungarian flora, small balsam was classified
into the sixth class (4.01-10.0 g per 1,000 seeds) among the eight that had been distinguished. Only
253 (ca. 15%) flora in Hungary belong to the upper two classes. It should be emphasized that the
flora pattern for forests is different. Trees and shrub species usually have seeds that are larger and
heavier. However, I. parviflora does not compete with those species at all. Other species that are
annuals and therophytes and that are found in forests, e.g., Geranium robertianum, have smaller
and lighter seeds (Fitter, Peat 1994).
171
80
60
40
0
20
Plasticity
DF
FA
FP
QP
TC
Fig. 91. Plasticity of Impatiens parviflora in various types of forest communities expressed as the
average ratio between the largest and the smallest value of plant height, length of leaves, width of
leaves, number of flowers and number of fruits. Only plants that had flowers and fruits were
included. Abbreviations: FA – Fraxino-Alnetum, DF – Dentario glandulosae-Fagetum, TC – TilioCarpinetum, QP – Querco robori-Pinetum, FP – forest path vegetation
Another important character is the behavior of seedlings (Tab. 46). The present study showed that
their ability to occur on different types of substrata is astonishing although their mortality was not
high. A comparison with other balsams (Skálová et al. 2012) indicated that the traits that are
associated with seedlings such as rate of growth and survival make I. parviflora a strong
competitor at this developmental stage especially when it grows together with I. noli-tangere.
Small balsam produces a relatively small biomass but at the seedling stage, individuals are larger
than its congeners. Despite some reports that crowded populations undergo self-thinning and that
the largest growing plants are isolated (Coombe 1956), the intra-specific is lower than the interspecific competition (Skálová et al. 2013).
Vertebral dispersal is still an open question (Tab. 46). To date information about zoochorous
dispersal is scarce except for Trepl (1984) and Graae (2002). There are no reports about
endozoochory, and it is probable that only epizoochory is possible in this species (Tanner 2008).
The quite frequent presence of the species inside hollows or even in the grooves of living trees,
which was confirmed in the present study, proves that birds transport seeds. Seeds are most likely
transported with dirt on their feet in the same manner as mammals, which was shown by Trepl
(1984).
The next three characters do not apply for small balsam at all (Tab. 46). The natural range of the
species is relatively small (Meusel et al. 1965) , which was stressed by Rejmánek et al. (2005) as
a noticeable exception. Its invasion success is not predicted by the natural range. The range of
172
adventives is much greater nowadays. As an annual and monocarpic plant, it reproduces only by
seeds. Its native congener I. noli-tangere is present in Europe; moreover, both species frequently
coexist.
Although small balsam is believed to have a positive impact on native fauna because it is a host for
Impatienticum asiaticum and a rich fauna of aphidophagous insects (Schmitz 1998), its
attractiveness for pollinators is low. It does use pollinators, mainly Syrphidae and perhaps also
Hymenoptera, but this requires further research. The research by Perez (2006) and Vervoort et al.
(2011) showed that Impatiens parviflora exhibits autonomous self-pollination reaching an 81.4%
fruit set that is linked to complete self-compatibility and no inbreeding depression was found for
this plant, which gives the species a great advantage to spread independent of the availability of
pollinators. The situation is similar with arbuscular mycorrhiza, which is not obligatory only
1500
1000
0
500
NRA [mol/g/dm/h]
2000
facultative, but when it is present it is beneficial for species (Chmura and Gucwa-Przepióra 2012).
FOREST
ROADSIDE
RUDERAL
Fig. 92. Activity of nitrate reductase (NRA) in Impatiens parviflora leaves in various types of
habitats (Chmura et al, in prep)
Small balsam is not an efficient competitor species (Tab. 46). As an annual species, which is on
average rather short lived and which has a shallow root system, it does not possess the mechanisms
that are necessary to compete with other species through shading or root competition. The only
possible weapon it possesses is allelopathy. Some of the studies that have reported on allelopathy
(Vrchotová et al. 2009; Csiszar, Bartha 2008; Csiszar et al. 2012) focused on laboratory
experiments. It is not known how to relate these finding to natural conditions.
Finally, the role of humans is not to be underestimated. Firstly, humans are the main vector of seed
transport (transport with vehicles, timber) and it facilitates long-distance dispersal. Secondly, due
173
to human activity artificial habitats – semi-natural and ruderal are more frequently colonized by
small balsam than natural ones.
Rejmánek (2000) when describing the biological attributes of invasive alien species mentioned
nitrogen fixation, which along with other ecophysiological traits can be useful in explaining
species invasiveness but we still do not have sufficient data. As far as nitrogen fixation in
I. parviflora is concerned, the first preliminary results indicate that the metabolism of nitrogen
compounds in this species varies among different habitats. The activity of nitrate reductase (NRA)
that was measured directly in the field was higher in ruderal habitats in which the nitrogen input
from the decomposition of biomass and wastes is higher than in other habitats (Fig. 92). Diekman
and Falkengren-Grerup (2002) believe that NRA has a lower value of prediction of plant responses
to atmospheric deposition and enhanced soil nitrogen levels; however, the prediction of the
nitrogen content in soil is dependent on the choice and accuracy of the methods that are applied
(Krywult, Bielec 2012). An analysis of NRA would probably be useful in predicting the degree of
the rate of nitrogen fixation among different habitats and among different species and in testing the
theory of fluctuating resources sensu Davis et al. (2000).
4.2. Causes of biological invasions
Among the various theories that seek to explain the invasions of plants, some seem to be
useful in the prediction of the success of I. parviflora. Others, which are based on previous
research, have failed. One of the first attempts chronologically was Darwin’s Naturalization
Hypothesis (Tab. 47). This corresponds with predicting the character “no congener(s) in the new
area” (Tab. 46). It not at all applicable in the case of invasions of Impatiens parviflora as well as
the entire genus Impatiens. According to this theory, exotic genera with native representatives
should be less successful because of an overlap in the use of resources and because they share the
same natural enemies. In Europe in addition to small balsam other balsams, other species such as
I. glandulifera, I. balfourii and I. capensis are both naturalized and invasive (Adamowski 2008;
Schmitz, Dericks 2010; Perrins et al. 1993). Moreover, some studies have suggested that native
congeneric I. noli-tangere can be displaced by exotics including small balsam (Daumann 1967;
Faliński 1998b).
There can be little doubt about the methods that were applied in these studies. Some insights are
given in the work by Skálova, Pyšek (2009), who in a germination experiment showed that
I. noli-tangere can be outcompeted by I. parviflora and I. glandulifera, which is the best
congeneric competitor.
174
Tab. 47. Major hypotheses explaining plant invasion (based on Hiero et al. 2005; Richardson et al. 2011) vs. hypotheses explaining invasion
success of Impatiens parviflora in introduced range
Hypothesis name
Darwin’s Naturalization
Hypothesis (DNH)
Enemy Release
Hypothesis (ERH)
Evolution of
Invasiveness
175
Empty Niche
Hypothesis
Fluctuating Resources
Theory Of Invasibility
Novel Weapons
Hypothesiss
Disturbance Hypothesis
Biotic Resistance
Hypothesis
Propagule Pressure
Hypothesis
Evolution Of Increased
Competitive Ability
Hypothesis ( EICA )
Competitive Release
Hypothesis
Definition
Alien species with close native relatives in
their introduced range may have reduced
chances of establishment and invasion
Exotics are released from natural enemies
that control their population growth
Exotics experience rapid genetic changes
linked to new selection pressures in the
novel environment
Exotics utilize resources unused by the
locals
Fluctuations in resource availability
enhances community invasibility
Exotics bring novel ways of biochemical
interaction to recipient communities
Exotics are adapted to disturbances’ type and
intensity that are novel to natives
Species-rich communities are more resistant
to invasion than species-poor communities
Variations in levels of invasion among
recipient communities are due to differences
in the number of exotics arriving in the
community
Exotics obtained increased competitive
ability through the relaxation of herbivore
pressure
Alien species may be released from
competition in habitats with novel
competitors or no competitors
Authors of hypothesis
De Candolle (1855),
Darwin (1859)
Darwin (1859), Williams
(1954), Elton (1958),
Keane, Crawley 2002
Colautti et al. 2004
Blossey and Nötzold
(1995), Lee (2002),
Stockwell et al. (2003)
Elton (1958), MacArthur
(1970)
Davis et al. (2000)
Callaway and Aschehoug
(2000), Bais et al. 2003
Gray (1879), Baker (1974)
Elton (1958), MacArthur
(1970, 1972)
di Castri (1989),
Williamson (1996),
Lonsdale (1999)
Prediction
No
References
Dostál et al. (2012), Dostál, Palečková
(2011)
No
Najberek et al. (2011), Dostál (2010)
No
Komosińska et al. (2006), Kupcinskiene
et al. (2013)
Yes
Trepl (1984), Piskorz, Klimko (2001),
Chmura (2008)
Eliáš (1999)
Yes
Yes(?)
Yes
Yes
Yes
Vrchotová et al. 2009; Csiszar, Bartha
2008; Csiszar et al. 2012
Obidziński, Symonides (2000), Chmura,
Sierka (2006)
Obidziński, Symonides (2000), Chmura,
Sierka (2006), Hejda (2012)
Coombe (1956), Trepl (1984)
Blossey and Nötzold (1995)
Maybe
Schmitz (1998)
Sorte et al. 2010
Maybe
-
In a pot experiment the growth and final biomass of I. parviflora was not affected by the presence
of native balsam (Dostál, Palečková 2011), but it was rather the other way round, in the absence of
an invasive competitor, I. noli-tangere displayed traits that were divergent with regard to the size
of plants, phenology and morphological plasticity (Dostál et al. 2012). These studies proved or no
effect of native representative of genus on behavior of I. parviflora or even its competition
capability.
Another theory that seems to be able to interpret the success of non-native plants is the Enemy
Release Hypothesis (Keane, Crawley 2002). It assumes that adventive species leave most of their
natural enemies (fungal pathogens, phytophagous insects, seed predators, etc.) at home. To date
there has been no study that compared the species composition of potential enemies in native and
invasive ranges. However, the predictability of the theory was tested in different ways as was
shown by Dostál (2010 and literature cited therein). He studied the survival rate of native and alien
species including I. noli-tangere and I. parviflora in relation to the effect of fungal pathogens, seed
predators and treatments with fungicide and seed predator exposure or both treatments in the
comparison with a control. His research partially supported the theory for other species taken into
analyses. Seed predator enclosures significantly increased the proportion of seeds that were
retrieved in both balsams, whereas the application of fungicide caused a decrease in I. parviflora.
A similar effect of seed predators on a native congener was revealed. In general, a larger number of
seeds of small balsam were retrieved in the control than was observed in the control of
I. noli-tangere, which might indicate that the seeds of I. parviflora were attacked less by enemies
in comparison with native balsam. However, the author did not present statistical analyses for this
result. Moreover, he emphasized that the enemies were limited to two groups – rodents and insects
and fungal pathogens and the lack of some significance in some analyses meant that these results
should be treated with caution. As regards the fungi samples, they were taken from mixed forests
for both balsam species. The type of habitat and soil has an impact on species composition of
fungi. Falencka and Grzywacz (1984) observed that pathogenous fungi predominate on seeds of
I. noli-tangere in oak-hornbeam forests whereas there are saprophytic fungi in floodplain forests.
Najberek et al. (2011, in prep) conducted a series of tests comparing the number of individuals,
invertebrate species and the amount of damage by fauna in pairs and triples of species. For a pair
of I. parviflora and I. noli-tangere only in 8% ERH was supported. Other analyses included
a comparison between two regions – lowland, mountain or the type of population – wildlife,
cultivation etc.
Two theories are similar, namely the Empty Niche Hypothesis (Elton (1958) and the
Fluctuating Resources Theory of Invasibility (Davis et al. 2002) both of which assume that non176
native species can penetrate an unoccupied niche or can utilize unused resources. The latter
additionally takes into account temporal changes in the availability of resources. A disturbance that
leads to the short- or long-term release of resources into the environment can be beneficial for
exotic species. For instance, thinning tree stands during forest management practices or
a windthrow of trees both contribute to more suitable light conditions on the forest floor. For
photophilous and also for shading-tolerant plants like small balsam such a disturbance promotes its
development. As some studies have demonstrated empty niches for I. parviflora are those that are
associated with coarse dead wood – dead logs, treefall disturbance sites, etc. Small balsam is a very
frequent species in such microsites. Those findings are also in accordance with the disturbance
hypothesis.
Disturbed habitats such as forest paths are corridors for the migration of species and can be
transitional habitats for the future invasion of a species into inner forest. Such a relationship
between small balsam and forest path environments, which was exemplified by a 4,300 ha forest
complex, was found by Godefroid and Koedam (2004b). In a much larger area, i.e. the Silesian
Upland (ca 4,000 km2), which is situated in Southern Poland, an inventory of 52 forest complexes
(Chmura 2004) showed that I. parviflora mostly occurred along forest roads and paths, followed by
its presence in forest interiors and along the forest edges that are associated with fire lines and
forest borders and ruderal sites (clear-cuts, dumping sites) (Fig. 93). The median cover along forest
borders was the highest but was only significantly higher than populations under a tree canopy.
The presence/absence of I. parviflora in a forest interior was under impact of abundance of the
species in non-forest habitats (Logistic regression, χ2 = 6.2042; p=0.0127).
The theory that assumes an evolution of invasiveness (Tab. 46) in terms of rapid genetic changes
also does not predict invasiveness of small balsam. When alien invasive species colonize a new
area, genetic variation is often lower than in the source population for two reasons – the founder
effect and self-compatibility (Novak, Welfley 1997; Amsellem et al. 2000). Even Coombe (1956)
claimed that the genetic variation in small balsam is rather low. The reason that was given is the
lower morphological variation in Europe than in its native region. Recent genetic studies in Poland
and Lithuania, respectively, (Komosińska et al. 2006; Kupcinskiene et al. 2013b) showed that
there are no significant differences in genetic variation among distinct populations and also showed
that small balsam was introduced many times, thus the founder effect did not occur and did not
lead to a genetic drift toward the newly adapted European population genetically.
177
Fig. 93. Frequency (A) and (B) differences in mean cover-abundance of Impatiens parviflora in the
four main groups of habitats within the forest areas of the Silesian Upland based on Chmura (2004)
but recalculated (Kruskal-Wallis test, χ2 = 8.2855, df = 3, p<0.05, Conover test as post-hoc
procedure)
This is somewhat congruent with Galera and Sudnik-Wójcikowska (2010) who pointed out many
escape events from botanical gardens at the beginning of invasions of I. parviflora.
The Novel Weapons Hypothesiss (Tab. 47) has been a relevant theory since the allelopathic
attributes of the species were recognized. Field tests are needed to estimate the effect of chemical
substances on the condition of coexisting species. The evolution of increased competitive ability
hypothesis (EICA) (Tab. 47) is unclear in relation to small balsam. On one hand, almost no
herbivores (except for roe deer) were recorded feeding on plants. On the other hand, a comparison
of the morphological variations between the adventive and native region did not indicate a higher
robustness of the species in invasive ranges, which could be attributed to a shift in biomass
178
allocation due to its release from the pressure of herbivores. Still there is a lack of data that present
morphometric studies of small balsam in its native East Asian and invasive ranges in Europe. The
scarce information in literature about biology and ecology of I. parviflora within its native range is
responsible for the lack of acceptance or rejection of the Competitive Release Hypothesis (Tab.
47). For all that is known about habitats occupied by small balsam in its native range, it can be
inferred that the species grows in the same or equivalent biotopes as I. noli-tangere in Europe.
Future comparative ecological studies in two ranges, native and adventive, will help to answer to
many questions. The need for conducting such studies was pointed out by Hierro et al. (2005).
4.3. Model of Impatiens parviflora invasion
Some efforts have been made to summarize the biological traits as well as the intrinsic
environmental and anthropogenic factors that enhance the success of an invasion of I. parviflora.
According to the general model of invasion proposed by Faliński (1968, 1998ab), like many alien
species I. parviflora underwent several phases of naturalization, which was called “neophytism”,
i.e. to become a neophyte species. The first phase is “epehemerophyte/ergasiophygophyte” =
“casual weed” – the occurrence in anthropogenic and semi-natural habitats such as parks, gardens.
The next phase is the stage of euneophyte in which the species occurs in natural habitats, i.e.
forests and forest margins. The last stage is a post-neophyte phase when a new xenospontaneous
community is formed instead of the natural community. In this phase more alien species are
capable of penetrating into the habitat and becoming a permanent element of the vegetation. This is
the equivalent of the invasion meltdown concept. During the process of neophytism, neophytes
have interactions with other resident species. Small balsam exhibits a suppletive action and in
a further phase a substitutive action by which it displaces I. noli-tangere (Faliński 1998b, KujawaPawlaczyk 1991). This model is rather general and descriptive. Few biological attributes of small
balsam are taken into account.
Eliáš (1999) introduced the abilities and limits of an invasion of I. parviflora into Europe.
Moreover, he listed the environmental factors (human-induced changes and the effects that are
related to them) that support an invasion of the species. The following characters were included for
abilities: 1) extreme plasticity in shade tolerance; 2) high efficiency of energy conversion; 3) high
reproductive capacity, even in deep shade; 4) lower nutrient demands; 5) very few parasites or
predators in Europe. In turn, the limits include: 1) droughts in summer; 2) low temperatures in
early spring; 3) high temperatures in spring and summer; 4) long-distance dispersal by humans and
animals.
179
Only human-induced changes were mentioned as environmental factors. These are: 1) the thinning
of the tree canopy, which creates canopy openings, increasing the irradiance near forest floor,
partial understory disturbance and seed dispersal; 2) the removal of the tree canopy, which leads to
clear-cut areas and forest margins, changes in the microclimate, large disturbances of the
undergrowth, substrate disturbances and seed dispersal and 3) trampling by humans and animals,
which results in partial disturbance of the undergrowth and seed dispersal.
The majority of the plant traits and factors that were mentioned are relevant. However, there
is a lack of information about mutualistic interactions and how natural environmental (disturbance)
factors might enhance the spread and invasion by this species. Some items such as a high
efficiency of energy conversion or a high reproductive capacity should be specified. These traits
varied considerably among the invaded habitats.
Tanner (2008) distinguished “risk and impact factors”. The main attributes of species
invasiveness were: fast growth, gregariousness, propagules that can remain viable for more than
one year, a high degree of mobility locally, being the pioneer in disturbed areas, proven invasive
outside its native range and tolerance of shade.
While the majority of these traits are valuable, there are exceptions. Small balsam does not create
viable seeds that last for more than one year. Single seeds were sometimes viable for more than
one season but there were too few to consider this variable as important for invasiveness. Indeed,
the species is invasive outside its native range. Perhaps, the idea was that if a species is invasive in
one area then after introduction into another area, it is also likely to be invasive, but this does not
explain anything about why it is invasive.
The putative “impact outcomes” were: modifications of the nutrient regime, formation of
a monoculture, negative impacts of forestry, reduced native biodiversity and threat to/ loss of
native species. These statements seem to be exaggerated. There is an absence of knowledge about
modifications of the nutrient regime in this species. It is unlikely that the species could affect
forestry. Reduction of native biodiversity and the threat to native species have not yet been proven
conclusively. The “impact mechanisms” that were listed include competition – monopolizing the
resources’ pest and disease transmission and rapid growth. These statements are accurate to some
extent. Small balsam is a very weak competitor and only in its seedling stage but it does not
monopolize resources – nutrients, light or space. It does indeed show rapid growth. It can transmit
some diseases; however, its pests and pathogens are very specific such as Puccinia komarovii and
attack small balsam almost exclusively.
180
Based on knowledge that has been obtained about the history of invasions and the properties
of the species, several schemes can be proposed to explain the pattern of an invasion by small
balsam (Fig. 94-96).
Impatiens parviflora is found more frequently along forest paths and forest margins rather
than inside forest interiors, which has been shown by many studies. As was postulated by Klimko
and Piskorz (2003), a network of paths, roads, firebreaks, etc. – any type of artificial habitats of an
anthropogenic origin – can be used by small balsam as migration routes into a forest. These
habitats are less vegetated and have better light conditions than sites under tree canopies and
therefore there is a greater chance for the long-distance dispersal for small balsam. Forest paths are
used by forestry vehicles and equipment more frequently and therefore they are more trampled
(Trepl 1984). It can be assumed that the rate of the spread of small balsam along a forest road
network is faster than within the dense herbaceous ground flora under tree canopies (Fig. 94). The
system of roads and paths also determines the direction of spread. At the beginning of an invasion,
small balsam quickly penetrates the system of linear shaped man-made biotopes and
simultaneously penetrates along the gradient forest road-forest interior where it runs more slowly.
Generally, the patches of individuals that grow along forest roads are denser and occupy larger
areas. In some cases these populations are quite extended, as much as several hundreds of meters
long. Kujawa-Pawlaczyk (1991) called them “ditch populations”. Such populations were observed
in the Silesian Upland and more rarely in the Jurassic Upland (Chmura 2004, unpublished). When
a species enters into a non-forest habitat within a forest complex, i.e. cutting areas, forest glades,
its abundance can increase.
Such a population can be the starting point for the further spread of the plant. Such a network of
forest paths, cutting areas and forest glades creates an anthropogenic pathway for an invasion by
I. parviflora, which is faster and more efficient for the spread and persistence of population. There
is a natural pathway as well (Fig. 94). Windthrow fallen trees and coarse dead wood are present in
the natural deciduous forests in lowlands due to natural disturbances. These create canopy gaps
that facilitate an invasion by small balsam.
The rate of penetration within a forest interior is slowed down through natural pathways (Fig. 94).
A population of small balsam can only become denser, more abundant and survive longer than
several seasons only under the canopy gaps that are associated with fallen trees.
Long-distance dispersal is probably a few-fold lower and hindered. Only birds and mammals (roe
deer, wild boars) can contribute to the propagation of I. parviflora within a forest interior.
181
Fig. 94. Model of the association between the rate of the spread of Impatiens parviflora and the
biocenotic resistance of vegetation and the type of habitat against the background of the spatial
organization of disturbed and natural forests
182
Fig. 95. Differences in the size and fecundity of individuals of Impatiens parviflora according to
the type of community/habitat
A common feature of both types of pathways is the greater abundance of populations that are
growing on disturbed sites (roads and paths vs. canopy gaps and fallen logs), which can be called
“disturbance populations”. These patches of individuals are characterized not only by their
abundance but also by other modified life history traits (Fig. 95).
183
Fig. 96. Model of the casual relationships between the factors and processes that influence species
invasiveness and the invasibility components of the invasion success of Impatiens parviflora.
Arrows indicate the direction of action; + increase, – decrease; ? no action or an unknown action, 1
– apart from favorable soil or substratum conditions, this includes beneficial (micro)climate
conditions. Some ideas were incorporated from Rejmánek et al. (2005) but modified
These plants are usually taller and have more flowers and fruits; they also set more seeds. Their
blooming phase starts earlier and sometimes lasts longer than that of individuals that grow in
“forest interior populations”. The propagule pressure of these disturbance populations is higher
than that of forest interior populations. Moreover, it is more likely that disturbance populations will
exhibit long-distance dispersal more often because plants are exposed to contact with more vectors
(man and animals) in these habitats. Forest interior populations differ in life traits in their
dependence on trophy (lower in mixed coniferous forests), light availability (lower in a dense,
compact treestand of broad-leaved forests). The third factor is a disturbance that promotes a larger
size and increased fecundity (Fig. 96).
The existence of both types of populations is similar to that of the dynamics of
a metapopulation on a local scale. Owing to short-distance dispersal (up to 3.4 m Trepl 1984) by
184
autochory, the distances between populations of small balsam within one forest area are relatively
long. The disturbance population can function as a source population because of a higher degree of
propagule pressure and forest interior populations tend to be a sink population type, which are
supplied by migrating individuals from disturbed sites. Klimko and Piskorz (2003) wrote that the
“ability of I. parviflora to invade a continuous forest complex depends on the intra-population
characteristics”. While this is a true, it can be added that among the habitats in a given forest
complex, a local metapopulation of I. parviflora always shows a high degree of morphological
plasticity. Indeed, its ability to spread is dependent on the characteristics of plants, which depend
on the forest complex and mosaic of habitats and microcosms that exist there. Plants differ in their
size, fecundity and phenology along the gradients of nutrients, light conditions and space, which
has been proven in many studies. Some environmental conditions are more suitable for plants of
small balsam which responds through faster growth and a modified allocation of biomass. Its
relatively high morphological plasticity, which also changes over time, as well as space are a major
drivers of species invasiveness.
The criteria that relate to the ability to become invasive and to achieve invasion success generally
belong to two categories – species invasiveness and a habitat’s invasibility (Fig. 96). The first
includes all of the biological characters that promote the spread and/or, competition, the measures
of which are generally theoretically derived from the abundance of the population on the site and
which are usually expressed in terms of cover or the degree of dominance. The ecological
background of invasibility relates to disturbance, the availability of resources (nutrients and
moisture) and interactions with resident biota (competition, mutualism and herbivory) (Richardson,
Pyšek 2012). Propagule pressure in terms of intentional or unintentional introduction by humans is
sometimes considered to be a third factor.
Some factors play a role at the beginning of invasion during the naturalization phase, while other
factors influence the process of ongoing invasion (Kueffer et al. 2013). The environmental
variables and biocenotic factors that currently influence invasions of I. parviflora are presented in
figure 96. During the first phase of an invasion, the character and number of subsequent
introductions are the most important for further spread. Close climate matching is also fundamental
for progression along the continuum in a novel environment (Richardson, Pyšek 2012).
Among other pivotal issues, residence time enhances naturalization and the rate of propagation of
the species as well. Residence time and all of the elements of propagule pressure, which
encompass the quantity, quality, composition, rate and other details of the supply of propagules,
are widely acknowledged as being positively associated with the “success”’ of N introduced
species (Richardson, Pyšek 2012). It is possible that alien species can be “forced” to be invasive
185
because of permanent frequent introductions. For instance, it was revealed and exemplified in
Florida that the probability of plants becoming naturalized increased significantly with the number
of years the plants were on sale as horticultural plants (Pemberton, Liu 2009). Impatiens parviflora
is a case of a naturalized and invasive plant that was mainly introduced unintentionally (Trepl
1984).
Propagule pressure is one of the key phenomena. The higher the number of seeds that are released,
the greater the probability of the establishment and founding of new populations. It is a stochastic
process that is influenced by residence time (Richardson, Pyšek 2012). Residence time matters
within the new invasive range of a nonnative species not only on a regional or global scale but also
important on a local scale. The degree of invasion of small balsam within a forest complex depends
on the residence time since the introduction on to the site. The model of invasion by I. parviflora
(Fig. 96) does not take into account residence time because all of these processes occur over time.
The main component that is associated with time is the continuous propagule pressure of small
balsam and therefore only it is presented.
Due to natural intra-population variation, propagule pressure has an indirect impact on
morphological variation and as a consequence, this results in morphological plasticity. Variations
in the biological traits in a species are determined genetically. The environment, in turn, selects and
causes the phenotypical responses of a species. The presence of various biotas (mutualists and
enemies) facilitates or hampers an invasion of plants. Human activity and animals are responsible
for long-distance dispersal. Some human actions can hinder the spread of a species. For instance,
managed coniferous forests from which coarse dead wood is removed, are less vulnerable to
invasion by small balsam than natural broad-leaved woods. A moderate level of
climatic/environmental matching is essential to enable an introduced species to establish, survive
and reproduce (Richardson, Pyšek 2012). Environmental factors remain fundamental during the
further spread of the species. The interplay of the input of resources and disturbances along with
habitat compatibility make habitats more vulnerable for the establishment, increase in abundance
and further spread of the introduced species.
Biocenotic resistance in terms of native species cover, species richness and the participation of
particular ecological groups (in the case of I. parviflora: tree seedlings, geophytes, plants of clonal
growth) is an effective barrier that protects a habitat from invasion or that simply slows down this
process.
186
5. Conclusions
Impatiens parviflora DC is a species with a very high degree of morphological plasticity that
is mainly induced by light availability followed by nutrient supply. This high degree of
variation in the morphometric features of the plant is manifested by differences among the
types of habitats/plant communities it occupies.
It is a species that is rather a “passenger” than a “driver” of disturbances. In well-developed
forest habitats, it penetrates sites that are unoccupied by other plants and utilizes any unused
resources. In disturbed forest habitats, which have higher amounts of unused resources, it
colonizes such sites with other species.
It has the ability to occupy very diverse microhabitats, which is one of the most important
aspects of its invasiveness. Despite the short seed dispersal spread of the species, it is very
effective. The presence of plants in some types of substrata, e.g., hollows; bark of living trees
and dead logs indicates the possibility of zoochorous spread.
The competitive ability of small balsam in the herbaceous layer is a rather low. An increase or
decrease in its abundance in the presence of other species, which was observed over time, is
rather a consequence of a shift in its biotopic requirements. Biocenotic resistance (species
richness and total species cover) are effective barriers and reduce the invasion success of the
species.
In specific microcosms (hollows, bark of living trees, dead logs, treefall disturbances),
I. parviflora is the most frequent colonizer and perhaps it responds efficiently as a stress
tolerator.
Mutualistic interactions seem to support an invasion by small balsam. Arbuscular
mycorrhization combined with some environmental factors is related to a higher robustness
and fecundity of plants.
The presence of I. parviflora in many openings and under canopy habitats promotes an
extension of the seed production period because of shifts in the phenology of plants.
187
A mosaic of disturbed (natural or man-made) and undisturbed habitats is a necessary factor for
the existence of sustainable populations of the species under forest conditions. A population of
small balsam in a forest complex exists as source-sink metapopulation. Source populations
exist on disturbed sites (canopy gaps, forest margins) in which individuals are characterized by
larger sizes and increased seed production. Forest interiors are habitats for sink populations
that are typified by a lower morphological plasticity and ability of seed setting.
The history of an invasion by I. parviflora in a specific area has an influence on the further
process of the dynamics of the species. Invaded areas differ in the invasion level by
I. parviflora and therefore the species increases in abundance on some sites while it declines
on others.
188
6. References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
Aber J.D., Nadelhoffer P., Steudler P., Mellilo J.M. 1989. Nitrogen saturation in Northern
forest ecosystems. Bioscience 39: 378–386.
Adamowski W. 2008. Balsams on the offensive: the role of planting in the invasion of
Impatiens species. In: Tokarska-Guzik B., Brock J.H., Brundu G., Child L., Daehler C.C.,
Pyšek P. (eds). Plant Invasions: Human perception, ecological impacts and management,
pp. 57–70.
Adamowski W., Keczyński A. 1998. Czynna ochrona zbiorowisk leśnych Białowieskiego
Parku Narodowego przed wkroczeniem Impatiens parviflora. Parki Narodowe Rezerwaty
Przyrody 17(1): 49–55.
Adamowski W., Keczyński A. 1999. Success of early eradication: the case of Impatiens
parviflora in Białowieża National Park (NE Poland). In: Proceedings 5th International
Conference on the Ecology of Invasive alien plants (13-16 October 1999 - La Maddalena –
Sardynia – Włochy): 3.
Adamska-Wachowiak E. 1983. Właściwości populacji inwazyjnego gatunku Impatiens
parviflora DC. w zależności od występowania w lesie grądowym. Praca magisterska,
Białowieska Stacja Geobotaniczna Uniwersytetu Warszawskiego (mscr).
Amsellem L., Noyer J.L., Le Bourgeois T., Hossaert-McKey M. 2000. Comparison of
genetic diversity of the invasive weed Rubus alceifolius Poir. (Rosaceae) in its native range
and in areas of introduction, using amplified fragment length polymorphism (AFLP)
markers. Molecular Ecology 9: 443–455.
Anioł-Kwiatkowska J., Świerkosz 1992. Flora i roślinność rezerwatu Ostrzyca
Proboszczowicka oraz jego otoczenia. Acta Universitatis Wratislaviensis 1358, Prace
Botaniczne 48: 45–115.
Arnold S. E., Le Comber S. C., Chittka L.2009. Flower color phenology in European
grassland and woodland habitats, through the eyes of pollinators. Israel Journal of Plant
Sciences 57(3): 211–230.
Babczyńska-Sendek A., Barć A., Urbisz A., Węgrzynek B. 2005. Characteristics of the
natural environment of the Cracow-Czestochowa Upland. Following the Jurrasic
strongholds. In: Tokarska-Guzik B., Babczyńska-Sendek B., Sierka E., Urbisz A (eds.).
Special characteristics of the Silesian Voivodship regions in focus. University of Silesia,
Katowice, pp. 13–32.
Bacigálová K., Eliáš P., Šrobarova A. 1998. Puccinia komarovii - a rust fungus on
Impatiens parviflora in Slovakia. Biologia, Bratislava 53(1): 7–14.
Bais H.P., Vepachedu R., Gilroy S., Callaway R.M.,Vivanco J.M. 2003. Allelopathy and
exotic plants: from genes to invasion. Science 301: 1377–1380.
Baker H.G. 1965. Characteristics and modes of origin of weeds. In: Baker H.G., Stebbins
G.L. (eds.). The genetics of colonizing species, New York: Academic Press, pp. 147–72.
Baker H.G. 1974. The evolution of weeds. Annual Review of Ecology and Systematics 5:
1–24.
Balkovič J., Kollár J., Šimonovič V. 2012. Experience with using Ellenberg’s R indicator
values in Slovakia: Oligotrophic and mesotrophic submontane broad-leaved forests.
Biologia 67(3): 474–482.
Barbour M.G. 1996. California Landscapes Before the Invaders. In: California Exotic Pest
Plant Council. Symposium Proceedings 1–6.
Barkworth M.E. 1973. Impatiens parviflora in British Columbia. Madrono 22: 1–24.
Berdau F. 1859. Flora Cracoviensis. Typis C.R. UJ Cracoviae, I-VIII.
189
18. Bergmann E., Bender J., Weigel H.J. 1999. Ozone threshold doses and exposure-response
relationships for the development ofozone injury symptoms in wild plant species. New
Phytologist 144: 423–435.
19. Blossey B., Nötzold R. 1995. Evolution of increased competitive ability in invasive
nonindigenous plants: a hypothesis. Journal of Ecology 83: 887–889.
20. Borhidi A. 1999. Az ismetelt tarsulasfelvetelezes buktatoi avagy megjegyzesek Horanszky
Andras cikkehez. Kitaibelia 4 (2): 357–366.
21. Borysiak J. 1984. Scales end ecological numbers of vascular plants of montane zones of
Babia Góra Massif (Western Beskids) using ecological profiles. Uniwersytet im. A.
Mickiewicza w Poznaniu. Seria Biologica 27: 1–148, Poznań.
22. BRC.
2008.
The
online
atlas
of
British
&
Irish
Flora.
URL:
http://www.brc.ac.uk/plantatlas/index.php?q=plant/impatiens-parviflora, date: April, 2014.
23. Brcak J. 1979. Isolates of cucumber mosaic virus from spontaneously infected plants of
Chelidonium majus and Impatiens parviflora. Biologia Plantarum 21 (3): 220–223.
24. Brothers T.S., Spingarn A., 1992. Forest Fragmentation and Alien Plant Invasion of Central
Indiana Old-Growth Forests. Conservation Biology 6: 91–100.
25. Brown N., Jennings S., Wheeler P., Nabe-Nielsen J. 2000. An improved method for the
rapid assessment of forest understorey light environments. Journal of Applied Ecology 37:
1044–1053.
26. Brzeg A., Wojterska M. 2001. Zespoły roślinne Wielkopolski, ich stan poznania i
zagrożenia. In: Wojterska M. (ed). Szata roślinna Wielkopolski i Pojezierza
Południowopomorskiego. Przewodnik sesji terenowych 52 Zjazdu PTB., Poznań.
27. Budziszewska J. 2006. Grzyby na nasionach niecierpka pospolitego (Impatiens noli-tangere
L.) i niecierpka drobnokwiatowego (Impatiens parviflora DC.) oraz ich potencjalny wpływ
na ekspansję jednego z gatunków. Praca licencjacka, Uniwersytet Warszawski (mscr).
28. Bütler R., Patty L., Le Bayon R. C., Guenat C., Schlaepfer R. 2007. Log decay of Picea
abies in the Swiss Jura Mountains of central Europe. Forest Ecology and Management
242(2): 791–799.
29. Buriánek V., Novotný R., Hellebrandová K., Šrámek V. 2013. Ground vegetation as an
important factor in the biodiversity of forest ecosystems and its evaluation in regard to
nitrogen deposition. Journal of Forest Science 59(6): 238–252.
30. Cabała S. 1990. Zróżnicowanie i rozmieszczenie zbiorowisk leśnych na Wyżynie Śląskiej.
Uniwersytet Śląski, Katowice.
31. Cáceres M. D., Legendre P. 2009. Associations between species and groups of sites: indices
and statistical inference. Ecology 90(12): 3566–3574.
32. Cáceres M.D, Legendre P., Moretti M. 2010. Improving indicator species analysis by
combining groups of sites. Oikos: 119(10): 1674–1684.
33. Callaway R.M., Aschehoug E.T. 2000. Invasive plants versus their new and old neighbors:
a mechanism for exotic invasion. Science 290: 521–523.
34. Causton D.R., Elias C.O., Hadley P. 1978. Biometrical studies of plant growth. I. The
Richards function and its application in analysing the effects of temperature on leaf growth.
Plant Cell and Environment 1: 163–184.
35. Celka Z., Żywica J. 2004. Flora naczyniowa wybranych cmentarzy Ostrowa
Wielkopolskiego i okolicy. Roczniki Akademii Rolniczej w Poznaniu 363, Botanika 7: 11–
31.
36. Chaudhry M.S., Rahman S.U., Ismaiel M.S., Sarwar G., Saeed B., Nasim F. 2009.
Coexistence of arbuscular mycorrhizae and dark septate endophytic fungi in an undisturbed
and a disturbed site of an arid ecosystem. Symbiosis 49: 19–28.
37. Chessel D. 1977. La description nom paramétrique de la dispersion spatiale des individuals
d’une espèce. Journee Francaise de Biometrie. 28 avril 1977
190
38. Chmura D. 2004. Penetration and naturalization of invasive alien plant (neophytes) in
woodlands of the Silesian Upland (Poland). Nature Conservation 60 (3): 3–11.
39. Chmura D. 2006. Wpływ gatunków inwazyjnych na różnorodność fitocenotyczną
rezerwatów Wyżyny Krakowsko-Częstochowskiej. Sprawozdanie merytoryczne (Grant nr 3
P04 G093 25). Instytut Ochrony Przyrody PAN, Kraków (mscr).
40. Chmura D. 2008a. Size variability in Impatiens parviflora DC. in natural and anthropogenic
habitats (S Poland). Thaiszia – Journal of Botany 18, Suppl. 1: 35–42.
41. Chmura D. 2008b. Slope aspect affects homogeneity and growth of ground vegetation in
deciduous forest. Polish Journal of Ecology 56(3): 463–470.
42. Chmura D. 2008c. The colonization of coarse woody debris of Fagus sylvatica by forest
herbs in Bukowica reserve (S Poland). Roczniki Akademii Rolniczej w Poznaniu 387,
Botanika Steciana 12: 3–7.
43. Chmura D. 2013. Impact of alien tree species Quercus rubra L. on understorey habitat and
flora: a study of the Silesian Upland (Southern Poland). Polish Journal of Ecology 61(3):
431–442.
44. Chmura D. 2014. Charakterystyka fitocenotyczna leśnych zbiorowisk zastępczych z
udziałem Quercus rubra L. na Wyżynie Śląskiej. Acta Botanica Silesiaca 10: 17-40.
45. Chmura D. Sierka E, Orczewska A. 2007. Autecology of Impatiens parviflora in natural
forest stands. Roczniki Akademii Rolniczej 386, Botanika Steciana. 11: 17–27
46. Chmura D., Csontos P., Sendek A. 2013. Seed mass variation in Central European
populations in invasive Impatiens glandulifera Royle. Polish Journal of Ecology 61(4):
747–751.
47. Chmura D., Gucwa-Przepióra E. 2012. Interactions between arbuscular mycorrhiza and the
growth of the invasive alien annual Impatiens parviflora DC: A study of forest type and soil
properties in nature reserves (S Poland). Applied Soil Ecology 62: 71–80.
48. Chmura D., Orczewska A. 2004. Udział Impatiens parviflora DC w zbiorowiskach leśnych
Wyżyny Śląskiej i Płaskowyżu Głubczyckiego. Archiwum Ochrony Środowiska 30 (3):
117–135.
49. Chmura D., Orczewska A., Sierka E. 2005. Differences in habitat preferences of Impatiens
parviflora DC in three adjacent regions in southern Poland. 8th International Conference
Ecology and Management of Alien Plant Invasions. University of Silesia, pp. 51.
50. Chmura D., Salachna A., Bochenek W. 2013. The changes in managed fir forest in Beskid
Mts due to forest management after 11 years. Journal of Ecological Engineering 14(1): 1–6.
51. Chmura D., Sierka E. 2006a. Relation between invasive plant Impatiens parviflora and
species richness of forest floor vegetation. Polish Journal Ecology 54(3): 417–428.
52. Chmura D., Sierka E. 2006b. The occurrence of invasive alien plant species in selected
forest nature reserves in southern Poland as a conservation problem. Nature Conservation
62(5): 3–11.
53. Chmura D., Sierka E. 2007. The invasibility of deciduous forest communities after
disturbance: A case study of Carex brizoides and Impatiens parviflora invasion. Forest
Ecology and Management 242(2–3): 487–495.
54. Chmura D., Urbisz A. 2005. Distribution and degree of naturalization of Impatiens
parviflora DC in the Silesian-Kraków Upland (Poland). Thaiszia – Journal of Botany,
Košice, 15, Suppl.1: 101–113.
55. Chytrý M., Maskell L. C., Pino J., Pyšek P., Vilà M., Font X., Smart S. M. 2008. Habitat
invasions by alien plants: a quantitative comparison among Mediterranean, subcontinental
and oceanic regions of Europe. Journal of Applied Ecology 45(2): 448–458.
56. Ciosek M.T., Bzdon G. 2003. Zmiany we florze Wyszkowa (woj. mazowieckie) na
przestrzeni ostatnich 25 lat. In: Korczyński M. (ed.) Flora miast. pp. 45–58.
191
57. Colautti R.I. , Grigorovich I.A., MacIsaac H.J. 2006. Propagule pressure: a null model for
biological invasions. Biological Invasions 8: 1023–1037.
58. Colautti R.I., Ricciardi A., Grigorovich I.A., MacIsaac H.J., 2004. Is invasion success
explained by the enemy release hypothesis? Ecology Letters 7: 721–733.
59. Coombe D.E. 1956. Impatiens parviflora. Journal of Ecology 44: 701–712.
60. Csiszar Á. 2004. A kisvirágú nebáncsvirág (Impatiens parviflora DC.) és a keresztlapu
(Erechtites hieracifolia RAF. ex DC.) terjedési stratégiáinak vizsgálata. NyugatMagyarországi Egyetem, Erdészeti és Vadgazdálkodási Tudományok Doktori Iskola,
Sopron, mscr.
61. Csiszár Á. Korda M., Schmidt D., Šporcic D., Teleki B., Tiborcz V., Zagyvai G., Bartha D.
2012. Study on allelopathic potential of some invasive and potentially invasive neophytes.
International Scientific Conference March 26-27 2012 on Sustainable Development &
Ecological Footprint. 1–6.
62. Csiszar Á., Bartha D. 2008. Small balsam (Impatiens parviflora DC.). In: Botta-Dukat Z,
Balogh L. (eds). The most important invasive plants in Hungary, pp. 139–149.
63. Csontos P. 1984. Az Impatiens parviflora DC. Vadallokovi (Polis) allomanyanak
conologiat es okologiai vizsgalata. Abstracta Botanica 8: 15–34.
64. Csontos P. 1986. Dispersal and establishment of Impatiens parviflora, an introduced plant,
in a hardwood forest. Abstracta Botanica 10: 341–348.
65. Csontos P. 2002. A magyar flóra ezermagsúly-adatbázisának bemutatása, alkalmazási
példákkal. Seed weight database of the Hungarian flora, with examples for its use. 5th
Hungarian Ecological Congress, Part 1, Abstracts of the lectures and posters. Acta
Biologica Debrecina Oecologia Hungarica 11(1): 51.
66. Csontos P., Chmura D., Sandor A. 2012. Seed mass of biogeographically distinct
populations of Impatiens parviflora DC. Tájökológiai Lapok 10 (1): 1–7.
67. Čuda J., Skálová H., Janovský Z., Pyšek P. 2014. Habitat requirements, short-term
population dynamics and coexistence of native and invasive Impatiens species: a field
study. Biological Invasions 16: 177–190.
68. Ćwikliński E. 1978. Die Einwanderung der synanthropen Art Impatiens parviflora DC. in
die naturlichen Pflanzengesellschaften. Acta Botanica Slovanica A; Taxon. Geobot. 3: 17–
34.
69. Dajdok Z., Wuczyński A. 2008. Alien plants in field margins and field of southwestern
Poland. Biodiversity Research and Conservation 9-10: 19–33.
70. Dańko B. 2009. Analiza występowania niecierpka drobnokwiatowego Impatiens parviflora
DC. w Lasach Komunalnych w Lądku Zdroju. Praca dyplomowa, SGGW Warszawa,
(mscr).
71. Darwin C. 1859. On the Origin of Species by Means of Natural Selection or the
Preservation of Favoured Races in the Struggle for Life. John Murray, London.
72. Daumann E. 1967. Zue Bestäubungs-und Verbreitungsökologie dreier Impatiens-Arten.
Preslia 39: 43–58.
73. Davis M.A., Grime J.P., Thompson K. 2000. Fluctuating resources in plant communities: a
general theory of invasibility. Journal of Ecology 88: 528–534.
74. De Candolle A.P. 1855. Géographie botanique raisonné, vol 2, V. Paris: Masson.
75. De Groot M., Kleijn D., Jogan N. 2007. Species groups occupying different trophic levels
respond differently to the invasion of semi-natural vegetation by Solidago canadensis.
Biological Conservation 136(4): 612–617.
76. di Castri F. 1989. History of biological invasions with special emphasis on the old world.
Biological Invasions: a Global Perspective In: Drake J.A., Mooney H.A., Castri F. di,
Groves R.H., Kruger F.J., Rejmánek M., Williamson M. (eds.), pp. 1–30. John Wiley &
Sons, New York.
192
77. Diekmann 1995. Use and improvement of Ellenberg’s indicator values in deciduous fforests
oft he Boreo-nemoral zone in Sweden. Ecography 18: 178–189.
78. Diekmann M., Falkengren-Grerup U. 2002. Prediction of species response to atmospheric
nitrogen deposition by means of ecological measures and life history traits. Journal of
Ecology 90(1): 108–120.
79. Dierschke H. 2006. Long-term dynamics in the herb layer of a calcareous beech:
investigations of permanent plots, 1981-2001. In: Holeksa J., Mirek Z., Wójcicki J.J. (eds.).
Polish Botanical Studies, pp. 165–172.
80. Dobosz M. 2004. Wspomagana komputerowo statystyczna analiza wyników badań.
Akademicka Oficyna Wydawnicza EXIT, Warszawa.
81. Dobravolskaitė R. 2012. Alien species Impatiens parviflora invasion into forest
communities of Lithuania. Botanica Lithuanica 18(1): 3–12.
82. Dostál P. 2010. Post-dispersal seed mortality of exotic and native species: effects of fungal
pathogens and seed predators. Basic and Applied Ecology 11: 676–684.
83. Dostál P., Palečková M. 2011. Does relatedness of natives used for soil conditioning
influence plant-soil feedback of exotics? Biological Invasions 13(2): 331–340.
84. Dostál P., Weiser M., Koubek T. 2012. Native jewelweed, but not other native species,
displays post-invasion trait divergence. Oikos 121(11): 1849-1859.
85. Drobnik J. 2007. Zielnik i zielnikoznastwo. Wydawnictwo Naukowe PWN.
86. Duchaufour P. 1970. Précis de Pédologie, Masson et Cie éditeurs, pp. 488.
87. Düll R., Kutzelnigg H., 1988. Botanisch-ökologisches Exkursionstaschenbuch, 3
Heidelberg, Wiesbaden, Germany: Quelle & Meyer.
88. Dufręne M., Legendre P. 1997. Species assemblages and indicator species: the need for a
flexible asymetrical approach. Ecological Monographs 67: 345–366.
89. Durak T. 2009. Changes in rich fir-forests in the Beskid Niski Moutains during 30 years. In:
Holeksa J., Babczyńska-Sendek B., Wika S. (eds). The role of geobotany in biodiversity
conservation, university of Silesia, Katowice: 141–148.
90. Dyguś K.H. 2008. Wykorzystanie metody fitoindykacji do oceny stanu siedliska leśnej
oczyszczalni ścieków przemysłu krochmalniczego. In: B. Kołwzan i K. Grabas (eds.),
Materiały z II konferencji naukowej "Ekotoksykologia w ochronie środowiska", Szklarska
Poręba 25-27 września 2008, pp. 83–92.
91. Dzwonko Z. 2001. Assessment of light and soil conditions in ancient and recent woodlands
by Ellenberg indicator values. Journal of Applied Ecology 38(5): 942–951.
92. Dzwonko Z., Loster S. 2000. Testing of Ellenberg and Zarzycki indicator values as
predictors of soil and light conditions in woodlands. Fragmenta Floristica and Geobotanica
45(1): 49–62.
93. Dzwonko Z., Loster S. 2001. Wskaźnikowe gatunki roślin starych lasów i ich znaczenie dla
ochrony przyrody i kartografii roślinności. Typologia zbiorowisk i kartografia roślinności w
Polsce. Prace Geograficzne 178: 119–132.
94. Elemans M. 2004. Light, nutrients and the growth of herbaceous forest species. Acta
Oecologica 26: 197–202.
95. Elias C.O., Causton D.R. 1975. Temperature and the growth of Impatiens parviflora DC.
New Phytologist 75(3): 495–505.
96. Eliáš P. 1992. Vertical structure, biomass allocation and size inequality in an ecotonal
community of an invasive annual (Impatiens parviflora DC.) on a clearing in SW Slovakia.,
Ekologia(CSFR) 11: 299–313.
97. Eliáš P. 1995. Stem fungi disease (Puccinia komarovii) on Impatiens parviflora in
Slovakia: effects on population dynamics and its role in regulation of plant populations.
Carinthia II, Sonderheft 53: 14–16.
193
98. Eliáš P. 1999. Biological and ecological causes of invasion of Impatiens parviflora DC. into
forest communities in Central Europe. Acta Horticulturae and Regiotecturae 1: 1–3.
99. Eliášova M. 2011. The phenological synchrony between alien aphid Impatientum asiaticum
Nevsky and its host – alien plant Impatiens parviflora DC. In: Šiška B., Hauptvogl M.,
Eliašová M. (eds.). Bioclimate: Source and Limit of Social Development International
Scientific Conference, 6th – 9th September 2011, Topoľčianky, Slovakia.
100. Ellenberg H., Weber H.E., Düll R., Wirth V., Werner W. and Paulißen D. 1992. Zeigwerte
von Pflanzen in Mitteleuropa. Scripta Geobot. 18. Erich Goltze, Göttingen.
101. Elton C.S. 1958. The ecology of invasions by animals and plants. Methuen, London.
102. EPPO data sheet on Invasive Plants Impatiens parviflora, File: 05-11832 DS Impatiens
parviflora.
URL:
http://www.eppo.int/QUARANTINE/Pest_Risk_Analysis/PRAdocs_plants/draftds/0511832%20DS%20Impatiens%20parviflora.doc, date: April, 2014.
103. Erkamo V. 1952. Pienikukkaisesta häpykannuksesta, Impatiens parviflora DC. Suomessa,
Arch. Soc. Zool. Bot. Fenn., Vanamo 6(1): 87–94.
104. Ertsen A. C. D., Alkemade J. R. M., Wassen, M. J. 1998. Calibrating Ellenberg indicator
values for moisture, acidity, nutrient availability and salinity in the Netherlands. Plant
Ecology 135(1): 113–124.
105. Evans G.C., Coombe D.E. 1959. Hemisperical and Woodland Canopy Photography and the
Light Climate. Journal of Ecology 47(1): 103–113.
106. Evans G.C., Hughes A.P. 1961. Plant growth and the aerial environment. I. Effect of
artificial shading on Impatiens parviflora. The New Phytologist 60: 130-179.
107. Falencka M., Grzywacz A. 1984. Composition of seed mycoflora - an agent modifying the
number of Impatiens noli-tangere L. population. Annales Warsaw Agricultural University
SSGW-AR. For and Wood Technol. 32: 49–52.
108. Falińska K. 1990. Osobnik, populacja, biocenoza. PWN, Warszawa.
109. Falińska K. 1991. Sukcesja jako efekt procesów demograficznych roślin. Phytocoenosis 3
(N.S.) Sem. Geobot 1, Warszawa-Białowieża, pp. 43–67.
110. Falińska K. 2004. Ekologia roślin. Wydawnctwo Naukowe PWN, Warszawa.
111. Faliński J.B. 1966. Degeneracja zbiorowisk roślinnych lasu miejskiego w Iławie. Materiały
Zakładu Fitosocjologii Stosowanej U.W. Warszawa-Białowieża 13: 1–13 + mapa.
112. Faliński J.B. 1968. Stadia neofityzmu i stosunek neofitów do innych komponentów
zbiorowiska. Materiały Zakładu Fitosocjologii Stosowanej U.W., Warszawa-Białowieża,
25: 15–29.
113. Faliński J.B. 1986. Vegetation dynamics in temperate lowland primeval forests. Ecological
studies in Białowieża Forest. Geobotany 8: 1-537. Dr W. JUnk Publishers,
Dordrecht/Boston/Lancaster.
114. Faliński J.B. 1993. Atlas fenologiczny Białowieskiego Parku Narodowego. Cz. I. Rytmika
sezonowa zbiorowisk leśnych i powiązane z nią zjawiska i procesy ekologiczne.
Białowieska Stacja Geobotaniczna UW (mscr).
115. Faliński J.B. 1998a. Invasive alien plants and vegetation dynamics. In: U. Starfinger K.,
Edwards I., Kowarik M., Williamson (eds.). Plant Invasions: Ecological Mechanisms and
Human responses, pp. 3-21.- Backhyus Publishers, Leiden, The Netherlands.
116. Faliński J.B. 1998b. Invasive alien plants, vegetation dynamics and neophytism.
Phytocoenosis (N.S.) 10, Supplementum Cartographiae Geobotanicae 9: 163–187.
117. Fitter A.H., Peat H.J. 1994. The Ecological Flora Database. Journal of Ecology 82: 415–
425. http://www.ecoflora.co.uk
118. Fitter A.H., Perrins J., Williamson M. 1990. Weed probability challenged. Bio/Technology
8: 473.
194
119. Fotyma M., Mercik S., Faber A. 1987. Chemiczne podstawy żyzności gleb i nawożenia.
PWRiL, Warszawa.
120. Franiel I. 2012. The biology and ecology of Betula pendula Roth on post-industrial waste
dumping grounds: the variability range of life history traits. Wydawnictwo Uniwersytetu
Śląskiego, Katowice.
121. Frank D., Klotz S., 1990. Biologisch-ökologische Daten zur Flora der DDR. Martin-LutherUniversität Halle-Wittenberg Wissenschaftliche Beiträge.
122. Galera H., Sudnik-Wójcikowska. 2010. Central European gardens as centres of dispersal of
alien plants. Acta Societatis Botanicorum Poloniae 79(2): 147–156.
123. Gerber E., Krebs C., Murrell C., Moretti M., Rocklin R., Schaffner U. 2008. Exotic
invasive knotweeds (Fallopia spp.) negatively affect native plant and invertebrate
assemblages in European riparian habitats. Biological Conservation 141: 646–654.
124. Godefroid S., Koedam N. 2004a. Interspecific variation in soil compaction sensitivity
among forest floor species. Biological Conservation 119(2): 207–217.
125. Godefroid S., Koedam N. 2004b. The impact of forest paths upon adjacent vegetation:
effects of the path surfacing material on the species composition and soil compaction.
Biological Conservation 119: 405–419.
126. Godefroid S., Koedam N. 2010. Comparative ecology and coexistence of introduced and
native congeneric forest herbs: Impatiens parviflora and I. noli-tangere. Plant Ecology and
Evolution 143(2): 119–127.
127. Godefroid S., Massant W., Weyembergh G., Koedam N. 2003. Impact of fencing on the
recovery of the ground flora on heavily eroded slopes of a deciduous forest. Environmental
management 32(1): 62–76.
128. Godefroid S., Monbaliu D., Massant W., Van der Aa B., De Vos B., Quivy V., Koedam N.
2007. Effects of soil mechanical treatments combined with bramble and bracken control on
the restoration of degraded understory in an ancient beech forest. Annals of Forest Science
64(3): 321–331.
129. Godefroid S., Phartyal S. S., Koedam N. 2011. Natural versus artificial cold-stratification:
which one allows better estimates of soil seed banks in a forest ecosystem? Current Science
101(2): 215–218.
130. Godefroid S., Phartyal S. S., Weyembergh G. Koedam N. 2005. Ecological factors
controlling the abundance of non-native invasive black cherry (Prunus serotina) in
deciduous forest understory in Belgium. Forest Ecology and Management 210: 91–105.
131. Godefroid S., Rucquoij S., Koedam N. 2006. Spatial variability of summer microclimates
and plant species response along transects within clearcuts in a beech forest. Plant Ecology
185(1): 107–121.
132. Godzik S. 1996. Ozone in upper Silesia region—concentration and effects on plants. In:
Bytnerowicz A., Arbough M.J., Schilling S.L. (eds.), Proceedings of the International
Symposium on Air Pollution and Climate Change Effects in Forest Ecosystems. USDA.
133. Golivets M. 2013. Native vs. non-native plant diversity across different spatial scales: a
case-study of the urban forest of Kyiv. In: Proceeding of: Challenging Problems in
Ecology, pp. 3–4. At Grodno, Belorus.
134. Graae B.J. 2002. The role of epizoochorous seed dispersal of forest plant species in a
fragmented landscape. Seed Science Research 12: 113–121.
135. Graae B.J., Hansen T., Sunde P.B. 2004. The importance of recruitment limitation in forest
plant species colonization: a seed sowing experiment. Flora 199: 263–270.
136. Gray A. 1879. The predominance and pertinacity of weeds. American Journal of Science
and Arts 118, 161–167.
137. Griebel C. 1948. Springkraut Impatiens parviflora DC als Wildgemüse. Zeitschrift für
Lebensmitteluntersuchung und -Forschung A 89(5): 411–415.
195
138. Grime 1979. Plant Strategies and Vegetation Processes. John Wiley, Chichester.
139. Gudžinskas Z., 1998. Conspectus of alien plant species of Lithuania, 8. Aceraceae,
Balsaminaceae Elaeagnaceae, Geraniaceae, Hippocastanaceae, Linaceae, Lytraceae,
Onagraceae, Oxalidaceae, Rutaceae, and Vitaceae. Botanica Lithuanica 4(4): 363–379.
140. Halarewicz A. 2011. Przyczyny i skutki inwazji czeremchy amerykańskiej Prunus serotina
w ekosystemach leśnych. Forest Research Papers 72(3): 267–272.
141. Hammer O., Harper D.A.T., Ryan P.D. 2001. PAST: Paleontological Statistics software
package for education and data analysis. Paleontologia Electronica 4(1): 1–9.
142. Harley J.L., Harley E.L., 1987. A checklist of mycorrhiza in the British flora. New
Phytologist (Suppl) 105: 1–102.
143. Hegi G. 1965. Illustrierte Flora von Mitteleuropa. Munich, Germany.
144. Heinken T., Raudnitschka D. 2002. Do wild ungulates contribute to the dispersal of
vascular plants in central European forests by epizoochory? A case study in NE Germany.
Forstwissenschaftliches Centralblatt 121(4): 179–194.
145. Hejda M. 2012. What is the impact of Impatiens parviflora on diversity and composition of
herbal layer communities of temperate forests?" PloS One 7.6: e39571.
146. Hermy M., Honnay O., Firbank L., Grashof-Bokdam C., Lawesson J.E. 1999. An
ecological comparison between ancient and other forest plant species of Europe, and the
implications for forest conservation. Biological conservation, 91(1): 9-22.
147. Hertel S. 1994. Populationsbiologische Untersuchungen an Impatiens parviflora de
Candolle 1824. Diplomarbeit. Ludwig-Maximilians-Universitat, Munchen, 272 pp.
148. Hierro J.L., Maron J.L., Callaway R.M. 2005. A biogeographical approach to plant
invasions: the importance of studying exotics in their introduced and native range. Journal
of Ecology 93: 5–15.
149. Hill M.O., Mountford J.O., Roy D.B., Bunce R.G.H. 1999. Ellenberg’s indicator values for
British plants. ECOFACT volume 2, Technical Annex. Huntingdon, Institute of Terrestrial
Ecology.
150. Höck F. 1990. Ankommlinge in der Pflanzenwelt Mitteleuropa währand des letzten halben
Jahrhunderts. Beih. Bot. Centralblb. 2.
151. Holeksa J. 2001. Coarse woody debris in a Carpathian subalpine spruce forest.
Forstwissenschaftliches Centralblatt vereinigt mit Tharandter forstliches Jahrbuch 120(1-6):
256–270.
152. Holeksa J., Woźniak G. 2005. Biased vegetation patterns and detection of vegetation
changes using phytosociological databases. A case study in the forests of the Babia Góra
National Park (the West Carpathians, Poland). Phytocoenologia 35(1): 1–18.
153. Holm S. 1979. A simple sequentially rejective multiple test procedure. Scandinavian
Journal of Statistics 6: 65–70.
154. Hood G. M. 2011. PopTools version 3.2.5. Available on the internet. URL
http://www.poptools.org.
155. Hubbel S. 2001. The Unified Neutral Theory of Biodiversity and Biogeography. Princeton
University Press
156. Hughes A.P. 1959. Effects of the environment on leaf development in Impatiens parviflora.
The Journal of Linnean Society Botany 56(366): 161–165
157. Hughes A.P. 1965. Plant growth and the aeral environment. IX. A synopsis of the
autecology of Impatiens parviflora. New Phytologist 64(3): 339–413.
158. Hughes A.P. 1965. Plant growth and the aerial environment. VI. The apparent efficiency of
conversion of light energy of different spectral compositions by Impatiens parviflora. New
Phytologist 64(1): 48–54.
159. Hughes A.P. 1965. Plant growth and the aerial environment. VII. The growth of Impatiens
parviflora in very low light intensities. New Phytologist 64(1): 55–64.
196
160. Hughes A.P. 1965. Plant growth and the aerial environment. VIII. The effect of (a) blue
light and (b) low temperature on Impatiens parviflora. New Phytologist 64(2): 323–329.
161. Hughes A.P. 1966. The importance of light compared with other factors affecting plant
growth. Light as an Ecological Factor. British Ecological Society Symposium 12: 121–146.
162. Hughes A.P., Evans G.C. 1961. Plant growth and aerial environment. I. Effects of artificial
shading on Impatiens parviflora. New Phytologist 60(2): 150–180.
163. Hughes A.P., Evans G.C. 1962. Plant growth and aerial environment. II. Effects of light
intensity on Impatiens parviflora. New Phytologist 61(2): 154–174.
164. Hughes A.P., Evans G.C. 1962. Plant growth and aerial environment. III. On the
computation of unit leaf rate. New Phytologist 61(3): 322–327.
165. Hughes A.P., Evans G.C. 1963. Plant growth and aerial environment. IV. Effects of
daylenght on Impatiens parviflora. New Phytologist 62(3): 367–388.
166. Hughes A.P., Evans G.C. 1964. Plant growth and the aerial environment. V. The effects of
(a) rooting conditions and (b) red light on Impatiens parviflora. New Phytologist 63(2):
194–202.
167. Hultén E., Fries M., 1986. Atlas of North European vascular plants: north of the Tropic of
Cancer. Königstein, Federal Republic of Germany: Koeltz Scientific Books.
168. Hurlbert S.H. 1984. Pseudoreplication and the design of ecological field experiments.
Ecological Monographs 54(2): 187–211.
169. Hylander N. 1971. Prima loca plantarum vascularium Sueciae. Första litteraturuppgift för
Sveriges vildväxande kärlväxter jämte uppgifter om första svenska fynd. Förvildade eller i
senare tid inkomna växter. Svensk Botanisk Tidskrift 64: 1–332.
170. Jackowiak B. 1999. Modele ekspansji roślin synantropijnych i transgenicznych.
Phytocoenosis 11 (N.S.) Sem. Geobot.6: 1–24.
171. Jacquemyn H., Butaye J., Hermy M., 2001. Forest plant species richness in small,
fragmented mixed deciduous forest patches: the role of area, time and dispersal limitation.
Journal of Biogeography 28: 101–112.
172. Janssens S., Geuten K., Yong-Ming Yuan Y.-M., Song Y., Kupfer P, Smets E. 2006.
Phylogenetics of Impatiens and Hydrocera (Balsaminaceae) using chloroplast atpB-rbcL
spacer sequences. Systematic Botany 31(1): 171–180.
173. Janssens S.B., Knox E.B., Huysmans S., Smets E.F., Merckx V.S.F.T. 2009. Rapid
radiation of Impatiens (Balsaminaceae) during Pliocene and Pleistocene: Result of a global
climate change. Molecular Phylogenetics and Evolution 52: 806–824.
174. Janulionienė R., Žukauskienė J., Jodinskienė M., Ambrasienė D., Paulauskas A.,
Kupčinskienė, E. 2011. Selection of RAPD primers for geographically most distinct
Lithuanian populations of Impatiens parviflora. Biologija 57(3): 130–136.
175. Jończyk M. 2007. Zmienność czasu spoczynku nasion i wczesnych faz wzrostu roślin
pomiędzy wybranymi populacjami niecierpka drobnokwiatowego (Impatiens parviflora
DC.). Praca magisterska, Uniwersytet Warszawski (mscr).
176. Jørgensen C.A. 1927. Impatiens parviflora D.C. i Danmark. Botanisk Tidsskrift 39: 385–
389.
177. Jouret M.-F. 1974a. Maturation et dormance des graines d'impatiens parviflora DC.
Bulletin de la Société Royale de Botanique de Belgique / Bulletin van de Koninklijke
Belgische Botanische Vereniging 107: 343–351.
178. Jouret M.-F. 1974b. Quelques aspects ecologiques de la dormance et de la germination chez
Impatiens parviflora DC. Bulletin de la Société Royale de Botanique de Belgique / Bulletin
van de Koninklijke Belgische Botanische Vereniging 107: 323–341.
179. Jouret M.-F. 1976. Écologie de la dormance séminale et de la germination chez diverses
espèces du genre Impatiens L. Bulletin de la Société Royale de Botanique de Belgique /
Bulletin van de Koninklijke Belgische Botanische Vereniging 109(2): 213–225
197
180. Jouret M.-F. 1977a. Gibbérelline et levée de la dormance séminale chez Impatiens
parviflora DC., I. glandulifera ROYLE et I. balfourii hook. F. Bulletin de la Société Royale
de Botanique de Belgique / Bulletin van de Koninklijke Belgische Botanische Vereniging
110 (1/2): 113–118
181. Jouret M.-F. 1977b. Relation entre la dormance séminale et la chorologie de diverses
espèces du genre Impatiens L. Bulletin de la Société Royale de Botanique de Belgique /
Bulletin van de Koninklijke Belgische Botanische Vereniging 110 (1/2): 119–128.
182. Jumpponen A. 2001. Dark septate endophytes — Are they mycorrhizal? Mycorrhiza, 11:
207–211.
183. Kamieński F. 1884. Nowy nabytek flory polskiej. Pamiętnik Fizjograficzny 4: 266–271.
184. Kapusta P. 2004. Metody geostatystyczne w ekologii. Wiadomości Ekologiczne 50: 171–
191.
185. Keane R. M., Crawley M. J. 2002. Exotic plant invasions and the enemy release hypothesis.
Trends in Ecology & Evolution 17(4): 164–170.
186. Kennedy T. A., Naeem S., Howe K. M., Knops J. M. H., Tilman D., Reich P. 2002.
Biodiversity as a barrier to ecological invasion. Nature 417: 636–638.
187. Kiedrzyński M., Zielińska K., Grzelak P. 2011. Transformation of forest vegetation after 40
years of protection in the Tomczyce nature reserve (Central Poland). Folia Biologica et
Oecologica 7(1): 207–227.
188. Kinzel W. 1927. Neue Tabellen zu Frost und Licht als beeinflussende Kräfte bei der
Samenkeimung. Stuttgart.
189. Klama H. 2001. Distribution patterns of liverworts in natural forest communities.
University of Bielsko-Biała.
190. Klimko M., Antkowiak M., Nowińska R. 2009. The influence of habitat conditions on
anatomical structure of Impatiens parviflora DC. (Balsaminaceae). Roczniki Akademii
Rolniczej w Poznaniu 388, Ser. Botanika - Steciana 13: 191–202.
191. Klimko M., Bozio A. 2003. Flora synantropijna Nowego Tomyśla. Roczniki Akademii
Rolniczej w Poznaniu 354, Ser. Botanika Steciana 6: 73–91.
192. Klimko M., Piskorz R. 2003. The usefulness of retrospective research along the road-forest
border zone for reconstruction of invasion of oak-hombeam forest communities by
Impatiens parviflora DC. Ecological Questions. An International Journal on Controversial
Problems of Ecology 3/2003: 63–69.
193. Klinggraeff C.J. 1866. Vegetationsverhaltnisse der Provinz Preuβen. Marienwerder.
194. Komarov W.L. (ed). 1934-1964. Flora SSSR. 1-30. Izdat. A. N. SSSR, Leningrad.
195. Komosińska E. 2008. Zróżnicowanie wybranych populacji Impatiens parviflora DC. a
środowiskowe efekty matczyne. Praca doktorska, Uniwersytet Warszawski (mscr).
196. Komosińska E., Wódkiewicz M., Jarzyna I., Jarochowska E., Milanowski R.,
Chwedorzewska K., Wyszomirski T. 2006. Some attempts to detect genetic differences
between populations of small balsam (Impatiens parviflora DC.). Biodiversity Research
and Conservation 3-4: 245–247.
197. Kondracki J. 1998. Geografia regionalna Polski. Wyd Nauk PWN, Warszawa.
198. Kornaś J. 1972. Rozmieszczenie i ekologia rozsiewania się chwastów w zespołach polnych
w Gorcach. Acta Agrobotanica 25: 5–67.
199. Kowarik I. 2003. Biologische Invasionen: Neophyten und Neozoen in Mitteleuropa.
Stuttgart, Germany: Ulmer.
200. Kowarik I., Pyšek P. 2012. The first steps towards unifying concepts in invasion ecology
were made one hundred years ago: revisiting the work of the Swiss botanist Albert
Thellung. Diversity and Distributions 18: 1243–1252
198
201. Kozłowska J. 2013. Rozmieszczenie i dynamika Impatiens parviflora DC. na powierzchni
badawczo-dydaktycznej w Czarnym Bryńsku. Praca magisterska, Uniw Mikołaja
Kopernika (mscr).
202. Kozyreva E.A., Mazaeva O., Molenda T., Rzętała M.A., Rzętała R., Trzhtsinsky Y.B.,
2004. Geomorphological processes in conditions of human impact. University of Silesia,
Sosnowiec.
203. Kraszewski M. 2007. Zróżnicowanie wybranych cech biometrycznych między osobnikami
z dwu populacji niecierpka drobnokwiatowego (Impatiens parviflora DC) uprawianymi we
wspólnych warunkach. Praca licencjacka, Uniwersytet Warszawski (mscr).
204. Krywult M., Bielec D. 2013. Measurement of nitrate reductase activity in a field conditions
– methodology. Journal of Ecological Engineering 14(1): 7–11.
205. Kueffer C., Pyšek P. Richardson D. M. 2013. Integrative invasion science: model systems,
multi-site studies, focused meta-analysis, and invasion syndromes. New Phytologist 200:
615–633.
206. Kühn I., Brandenburg M., Klotz S. 2004. Why do alien plant species that reproduce occur
in natural habitats occur more frequently? Diversity and Distributions 10: 417–425.
207. Kujawa-Pawlaczyk J. 1990. Ekspansja niecierpka drobnokwiatowego (Impatiens parviflora
DC.) w Puszczy Białowieskiej i jej związek ze stanem zachowania zbiorowisk leśnych.
Praca magisterska wykonana w Akademii Rolniczej, Poznań i Białowieskiej Stacji
Geobotanicznej Uniwersytet Warszawski, Białowieża (mscr).
208. Kujawa-Pawlaczyk J. 1991. Rozprzestrzenianie się i neofityzm Impatiens parviflora DC. w
Puszczy Białowieskiej. Phytocoenosis N.S. 3, Sem. Geobot.1: 213–222.
209. Kupcinskiene E, Sliumpaite I, Zybartaite L, Janulioniene R, Paulauskas A. 2013a. Appetite
of alien Impatiens spp. growing in the Baltic Region. 12th Reunion On Ecology and
Management Of Alien Plant Invasions. 22-26 September, 2013. Pirenópolis, Goiás, Brazil.,
p. 62.
210. Kupcinskiene E., Zybartaite L., Janulioniene R., Zukauskiene J., Paulauskas A. 2013b.
Molecular diversity of small balsam populations in relation to site characteristics. Central
European Journal of Biology 8(10): 1048–1061.
211. Kuźnicki F., Białousz S., Składkowski P. 1979. Podstawy gleboznawstwa z elementami
kartografii i ochrony gleb. PWN, Warszawa, p. 509.
212. Łaska G. 2001. The disturbance and vegetation dynamics: a review and an alternative
framework. Plant Ecology 157: 77–99.
213. Lee C.E. 2002. Evolutionary genetics of invasive species. Trends in Ecology and Evolution,
17: 386–391
214. Legg C. 2004a. TriGraph ver. 0.2. The University of Edinburgh School of Geosciences.
http://www.geos.ed.ac.uk/abs/research/plants/software
215. Legg C. 2004b. Analysis of Contingency Tables. Free software for ecologists. URL:
www.geos.ed.ac.uk/abs/research/plants/software.
216. Lemmerer K. 2010. Vegetationsökologische Einnischung von Impatiens parviflora DC. im
Gebiet des Wienerwaldes, Österreich. Masterarbeit, Universität Wien. Fakultät für
Lebenswissenschaften.
217. Ling K.A., Ashmore M.R. 1999. Influence of tree health on ground flora in the Chiltern
Beechwoods, England. Forest Ecology and Management 119: 77–88.
218. Lipińska A. 2011. Znaczenie interakcji międzygatunkowych w inwazji niecierpka
drobnokwiatowego Impatiens parviflora DC. na terenie Mazurskiego Parku
Krajobrazowego. Praca doktorska, Uniwersytet Warszawski (mscr).
219. Lityński T., Jurkowska H., Gorlach E. 1976. Analiza chemiczno-rolnicza PWN, Warszawa.
220. Lonsdale W. M. 1999. Global patterns of plant invasions and the concept of invasibility.
Ecology 80: 1522–36.
199
221. Łuczaj Ł. 2002 Dzikie rośliny jadalne Polski. Przewodnik survivalowy. Chemigrafia,
Krosno.
222. Lunau K., Maier E.J. 1995. Innate colour preferences of flower visitors. Journal of
Comparative Physiology 177: 1–19.
223. Łysik M. 2008. Ten years of change in ground-layer vegetation of European beech forest in
the protected area (Ojcow national park, south Poland). Polish Journal of Ecology 56(1):
17-31.
224. Łysik M. 2009. A 13-year change in ground-layer vegetation of Carpathian beech forests.
Polish Journal of Ecology 57(1): 47–61.
225. MacArthur, R.H. 1970. Species packing and competitive equilibrium for many species.
Theoretical Population Biology 1: 1–11.
226. MacArthur, R.H. 1972. Geographical Ecology: Patterns in the Distribution of Species.
Harper & Row, New York.
227. MacDougall A.S., Turkington R. 2005. Are invasive species the drivers or passengers of
change in degraded ecosystems? Ecology 86(1): 42–55.
228. Macková D. 2012. Invázne rastliny a riziká vyplývajúce z ich výskytu na vybraných
chránených územiach severovýchodného Slovenska. Dizertačná práca. Prešovská
Univerzita v Prešove, Prešov, Slovensko (mscr).
229. Majewski T. 1979. Flora Polska Rośliny zarodnikowe Polski i ziem ościennych. Grzyby
(Mycota) Tom XI: 65–66.
230. Manning W.J., Godzik B. 2004. Bioindicator plants for ambient ozone in Central and
Eastern Europe. Environmental Pollution 130: 33–39.
231. Manning W.J., Godzik B., Musselman R. 2002. Potential bioindicator plant species for
ambient ozone in forested mountain areas of central Europe. Environmental Pollution 119:
283–290.
232. Maringer J., Wohlgemuth T., Neff C., Pezzatti G. B., Conedera M. 2012. Post-fire spread of
alien plant species in a mixed broad-leaved forest of the Insubric region. Flora-Morphology,
Distribution, Functional Ecology of Plants 207(1): 19–29.
233. Matuszkiewicz J.M. 2008. Potential natural vegetation of Poland. IGiPZ PAN, Warszawa,
http://www.igipz.pan.pl.
234. Matuszkiewicz W. 2011. Przewodnik do oznaczania zbiorowisk roślinnych Polski. PWN,
Warszawa.
235. Meusel H., Jager E., Weinert R. 1965. Verieichende chorologie der Zentraleuropaischen
Flora, 2 (Text, Karten), Jena G. Fischer Verl, 418: 259–421.
236. Michalik J. 2005. Zależne od warunków proporcje międzygenotypowej i
wewnątrzgenotypowej zmienności sadzonek niecierpka drobnokwiatowego (Impatiens
parviflora). Praca licencjacka, Zakład Ekologii Roślin i Ochrony Przyrody Uniwersytet
Warszawski, mscr.
237. Mirek Z., Piękoś-Mirkowa H., Zając A., Zając M. 2002. Flowering plants and pteridophytes
of Poland a checklist. W. Szafer Institute of Botany, PAN.
238. Moran M. D. 2003. Arguments for rejecting the sequential Bonferroni in ecological studies.
Oikos 100(2): 403–405.
239. Moravcová L., Pyšek P., Jarošik V., Havličková V., Zakravský P. 2010. Reproductive
characteristics of neophytes in the Czech Republic: traits of invasive and non-invasive
species. Preslia 82: 365–390.
240. Muthukumar T., Vediyappan S. 2010. Comparison of arbuscular mycorrhizal and dark
septate endophyte fungal associations in soils irrigated with pulp and paper mill effluent
and well-water. European Journal of Soil Biology, 46(2): 157–167.
200
241. Najberek K., Solarz W. 2011. Inwazje biologiczne w polskich parkach narodowych i
krajobrazowych. In: Głowaciński Z., Okarma H., Pawłowski J., Solarz W. (eds.) Księga
gatunków obcych inwazyjnych w faunie Polski, pp. 624–639.
242. Najberek K., Solarz W., Chmura D., Walusiak E., Król W. (in prep.) Verification of Enemy
Release Hypothesis on the selected alien and native Impatiens spp. in the flora of Poland.
243. Najberek K., Solarz W., Walusiak E., Chmura D., Król W. 2011. Z deszczu pod rynnę?
Weryfikacja hipotezy uwolnienia od wrogów (Enemy Release Hypothesis) na wybranych
gatunkach roślin rodzimych i obcych we florze Polski. Uniwersytet Wrocławski, I
Ogólnopolska Konferencja Naukowa "Synantropizacja w dobie zmian różnorodności
biologicznej", materiały konferencyjne pp. 62.
244. NOBANIS 2013. Impatiens parviflora http://www.nobanis.org/speciesInfo.asp?taxaID=844
245. Noble I.R., 1989. Attributes of Invaders and the Invading Process: Terrestrial and Vascular
Plants. In: Drake J.A., Mooney H.A., Di Castri F., Groves R.H., Kruger F.J. Reymanek M.,
Wiliamson M. (eds.) Biological Invasions: A Global Perspective, John Wiley and Sons Ltd,
Chichester, pp. 301–313.
246. Novak S.J., Welfley A.Y. 1997. Genetic diversity in the introduced clonal grass Poa
bulbosa (Bulbous bluegrass). Northwest Science 71: 271–280.
247. Nowińska R. 2010. Reactions of the herb and moss layer, tree saplings and the shrub layer
to tree deaths in forests of the Wielkopolska National Park (Western Poland). Biologia
65(2): 265–272.
248. Nowińska R., Urbański P., Szewczyk W. 2009. Species diversity of plants and fungi on
logs of fallen trees of different species in oak-hornbeam forests. Roczniki Akademii
Rolniczej w Poznaniu 388, Botanika-Steciana. 13: 109–124.
249. Nyrek A., 1975. Gospodarka leśna na Górnym Śląsku. Prace Wrocławskiego Towarzystwa
Naukowego S.A., Wrocław.
250. Obidziński A., Głogowski R. 2005. Changes of forest flora composition in vicinity of dens
of red fox and setts of euroasian badger. Polish Journal of Ecology 53(2): 197–213.
251. Obidziński A., Kiełtyk P. 2006. Changes in ground vegetation around badger setts and fox
dens in the Białowieża Forest, Poland. In: Polish Botanical Studies 22: 407–416.
252. Obidziński T., Symonides E. 2000. The influence of the groundlayer structure on the
invasion of small balsam (Impatiens parviflora DC.) to natural and degraded forests. Acta
Societatis Botanicorum Poloniae 69(4): 311–318.
253. Obmiński Z. 1977. Ekologia lasu. PWN, Warszawa.
254. Olaczek R., 1974. Kierunki degeneracji fitocenoz leśnych i metody ich badania.
Phytocoenosis. 3(3/4): 179–190.
255. Ostrowska A., Gawliński S., Szczubiałka Z., 1991. Metody analizy i oceny właściwości
gleb i roślin. Instytut Ochrony Środowiska, Warszawa.
256. Otręba A., Ferchmin M. 2007. Obce gatunki drzew miarą przekształcenia przyrody
Kampinoskiego Parku Narodowego In: Anderwald D. (ed). Siedliska i gatunki
wskaźnikowe w lasach, (ed.).Studia i Materiały Centrum Edukacji Przyrodniczo-Leśnej) –
Rogów, 2/3 (16), pp. 234–244.
257. Parker F.D., Simberloff D., Lonsdale W.M., Goodell K., Wonham M., Kareiva P.M.,
Wiliamson M.H., Von Holle B., Moyle P.B., Byers J.E., Goldwasser L. 1999. Impact:
towards a framework for understanding the ecological effect of invaders. Biological
Invasions 1: 3–19.
258. Pavela R., Vrhotova N., Šera B. 2009. Repellency and toxicity of three Impatiens species
(Balsaminaceae) extracts on Myzus persicae Sulzer (Homoptera: Aphididae). Journal of
Biopesticides 2(1): 48–51.
259. Peace W.J.H., Grubb P.J. 1982. Interaction of light and mineral nutrient supply in the
growth of Impatiens parviflora. New Phytologist 90: 127–150.
201
260. Pemberton, R. W., Liu H. 2009. Marketing time predicts naturalization of horticultural
plants. Ecology 90(1): 69-80.
261. Perez N. 2006. Biologie de la reproduction et caracterisation de la niche ecologique d’une
espece envahissante, Impatiens parviflora DC. Université catholique de Lovain. B.Sc thesis
(mscr).
262. Perglová, I., Pergl, J., Skalova, H., Moravcova, L., Jarošík, V., Pyšek, P. 2009. Differences
in germination and seedling establishment of alien and native Impatiens species. Preslia
81(4): 357–375.
263. Perrins J., Fitter A., Williamson M. 1993. Population biology and rates of invasion of three
introduced Impatiens species in the British Isles. Journal of Biogeography 20: 33–44.
264. Perrins J., Williamson M., Fitter A. 1992. A survey of differing views of weed
classification: implications for regulation of introductions. Biological Conservation 60(1):
47–56.
265. Petrášová M., Jarolímek I., Medvecká J. 2013. Neophytes in Pannonian hardwood
floodplain forests–History, present situation and trends. Forest Ecology and Management
308: 31–39.
266. Phillips J.M., Hayman D.S. 1970. Improved procedures for clearing roots and staining
parasitic and vesicular-arbuscular mycorrhizal fungi for rapid assessment of infection.
Transactions of the British Mycological Society 55: 158–160.
267. Piękoś-Mirkowa H., Mirek Z., Mieczówka A. 1996. Endemic vascular plants in the Polish
Tatra Mts. Distribution and ecology. Polish Botanical Studies. 12: 1–107.
268. Pilková I. 2013. Zhodnotenie fenologických fáz inváznej byliny Impatiens parviflora DC.
V Bábskom Lese. Acta Universitatis Matthiae Belii Sekcia Environmentálne manažérstvo
15(1): 40–48.
269. Pirożnikow E. 1983. Seed bank in the soil of the stabilized ecosystem of a deciduous (TilioCarpinetum) in the Bialowieża National Park. Ekologia Polska 31(1): 145–172.
270. Pirożnikow E. 1991. Struktura populacji runa stabilnego zbiorowiska leśnego (TilioCarpinetum) w Białowieskim Parku Narodowym. Phytocoenosis 3 (N.S.) Sem. Geobot. 1
Warszawa-Białowieża, pp. 127–139.
271. Piskorz R. 2004. Biologia niecierpka drobnokwiatowego (Impatiens parviflora DC.) w
fitocenozach Galio Silvatici-Carpinetum na terenie Wielkopolskiego Parku Narodowego.
Praca doktorska, Akademia Rolnicza w Poznaniu (mscr).
272. Piskorz R. 2005. The effect of oak-hornbeam diversity on flowering and fruiting of
Impatiens parviflora. Roczniki Akademii Rolniczej w Poznaniu 373, Botanica Steciana:
187–196.
273. Piskorz R., Klimko M. 2001. Kolonizacja powalonych drzew i buchtowisk dzików przez
Impatiens parviflora DC. W zbiorowiskach Galio silvatici-Carpinetum wybranych
rezerwatow Wielkopolskiego Parku Narodowego. Roczniki Akademii Rolniczej w
Poznaniu 334 Botanika Steciana 4: 151–163.
274. Piskorz R., Klimko M. 2002. Fenologia Impatiens parviflora DC. w silnie przerzedzonym
grądzie środkowopolskimin na lokalnym stanowisku w Wielkopolskim Parku Narodowym.
Roczniki Akademii Rolniczej w Poznaniu 348, Botanika Steciana 5: 135–144.
275. Piskorz R., Klimko M. 2006. The effect of Puccinia komarovii Tranzsch. infection on
characters of Impatiens parviflora DC. in Galio sylvatici-Carpinetum (R. Tx. 1937) Oberd.
1957 forest association. Acta Societatis Botanicorum Poloniae 75(1): 51–60.
276. Piskorz R., Klimko M. 2007. Współwystępowanie Impatiens parviflora i wybranych roślin
lasu dębowo-grabowego w Wielkopolskim Parku Narodowym. Sylwan 2: 43–58.
277. Piskorz R., Urbańska M. 2007. Utilization of the invasive plant Impatiens parviflora DC.
by the snail Columella edentula (Draparnaud) in oak-hornbeam forests. Acta Societatis
Botanicorum Poloniae 76(1): 61–67.
202
278. Podlaska M. 2010. Zróżnicowanie wartości użytkowej runi nieużytkowanych łąk
pobagiennych niżowej części Dolnego Śląska. Woda-Środowisko-Obszary wiejskie 10(4):
149–163.
279. Polak Z. 1967. Impatiens parviflora DC. – a natural host of cabbage black ringspot and
cucumber mosaic viruses. Biologia Plantarum 9(5): 354–359.
280. Poorter H., Niklas K. J., Reich P. B., Oleksyn J., Poot P., Mommer L. 2012. Biomass
allocation to leaves, stems and roots: meta-analyses of interspecific variation and
environmental control. New Phytologist 193(1): 30–50.
281. Probst R. 1949. Solothurner Flora. Solothurn.
282. Pyšek P., Danihelka J., Sádlo J., Chrtek Jr J., Chytrý M., Jarošík V., Krahulec F. 2012.
Catalogue of alien plants of the Czech Republic: checklist update, taxonomic diversity and
invasion patterns. Preslia 84(2): 155–255.
283. R Core Team 2012. R: A language and environment for statistical computing. R Foundation
for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.Rproject.org/.
284. Rahmonov O., Czylok A., Orczewska A., Majgier L., Parusel T. 2014. Chemical
composition of the leaves of Reynoutria japonica Houtt. and soil features in polluted areas.
Central European Journal of Biology 9(3): 320–300.
285. RBG 2008. Royal Botanic Gardens Kew, Seed Information Database (SID). Version 7.1.
Available from: http://data.kew.org/sid/ Date: May 2008.
286. Rejmánek M. 1995. What makes a species invasive? In: Pyšek P, Prach K, M. Rejmánek
M, Wade M (eds.), Plant invasions – General Aspects and Special Problems. Academic
Publishing, Amsterdam, The Netherlands, pp. 3–13.
287. Rejmánek M. 1996. A theory of seed plant invasiveness: the first sketch. Biological
Conservation 78: 171–181.
288. Rejmánek M. 2000. Invasive plants: approaches and predictions. Austral Ecology 25: 497–
506.
289. Rejmánek M., Richardson D.M., Pyšek P. 2005. Plant invasions and invasibility of plant
communities. In: Maarel Van der (ed.) Vegetation ecology, Blackwell Science, Oxford, pp
332–255.
290. Richardson D. M. Pyšek P., Carlton, J. T. 2011. A compendium of essential concepts and
terminology in invasion ecology. In: Richardson D.M. (ed.) Fifty years of invasion ecology
The legacy of Charles Elton. Oxford: Wiley-Blackwell, pp. 409–420.
291. Richardson D. M., Pyšek P. 2012. Naturalization of introduced plants: ecological drivers of
biogeographic pattern. New Phytologist 196: 383–396.
292. Richardson D.M., Pyšek P. 2008. Fifty years of invasion ecology – the legacy of Charles
Elton. Diversity and Distributions 14: 161–168.
293. Ripka G., Csiszar Á. 2008. Impatientinum asiaticum Nevsky, 1929, new for the Hungarian
fauna from Impatiens parviflora. Folia Entomologica Hungarica: 15-18
294. Rostański A. 2006. Spontaniczne kształtowanie się pokrywy roślinnej na zwałowiskach po
górnictwie węgla kamiennego na Górnym Śląsku. Wyd. Nauk UŚ, Katowice.
295. Rostański K. 1977. Flora i roślinność synantropijna w Karkonoskim Parku Narodowym. Pr.
Karkon. Tow. Nauk. Jelenia Góra 9 1–29
296. Różański H. 2009. Niedoceniane ziele: niecierpek. Porady na zdrowie 11: 16–17.
297. Sádlo J., Chytrý M., Pyšek P. 2007. Regional species pools of vascular plants in habitats of
the Czech Republic. Preslia 79: 303–321.
298. Sadowska A. 2001. Gatunki synantropijne we florze roślin naczyniowych torfowisk
wysokich i przejściowych Pojezierza i Pobrzeża Kaszubskiego (Polska północna). Acta
Botanica Silesiaca 7: 79–95.
203
299. Salachna A., Chmura D. 2012. Wykorzystanie przyrządu Levy'ego i kanopiskopu w
nauczaniu metodyki badań fitosocjologicznych. W: XVIII Międzynarodowa Konferencja
Naukowo-Techniczna Zapobieganie zanieczyszczeniu, przekształcaniu i degradacji
środowiska. Szczyrk 15-16 listopada 2012 r. pp. 61.
300. Sax D.F. 2002. Native and naturalized plant diversity are positively correlated in scrub
communities of California and Chile. Diversity and Distributions 8: 193–210.
301. Schaddach A. 2008. Germination ecology of the invasive plant species Impatiens
parviflora. Thesis, University of Hamburg (mscr).
302. Schmitz G. 1998. Impatiens parviflora D.C. (Balsaminaceae) als Neophyt in
mitteleuropaische Waldern und Forsten: Eine biozonotische Analyse. Z. Okologie u.
Naturschutz 7: 193–206.
303. Schmitz, U., Dericks, G. 2010. Spread of alien invasive Impatiens balfourii in Europe and
its temperature, light and soil moisture demands. Flora-Morphology, Distribution,
Functional Ecology of Plants 205(11): 772–776.
304. Sebald O, Seybold S, Philippi G, Wörz A, 1998. Die Farn- und Blütenpflanzen BadenWürttembergs. Ulmer, Stuttgart.
305. Semenova G.V., van der Maarel E. 2000. Plant functional types—a strategic perspective.
Journal of Vegetation Science 11: 917–922.
306. Sierka E., Chmura D. 2005. Role of Calamagrostis epigeios (L.) Roth in forest vegetation
of Silesian Upland. In: Frey L (ed.) Biology of grasses, pp. 343–353.
307. Sierka E., Parusel B., Chmura D. 2009. Występowanie i dynamika Impatiens parviflora w
rezerwacie Ochojec (Wyżyna Śląska). In: Parusel J. B. (ed.) Rezerwat przyrody "Ochojec"
w Katowicach (Górny Śląsk). Monografia naukowo-dydaktyczna. Katowice, pp. 314–320.
308. Simberloff D. 2003. Confronting introduced species: a form of xenophobia? Biological
Invasions 5: 179–192.
309. Skálová H., Havlíčkova V., Pyšek P. 2012. Seedling traits, plasticity and local
differentiation as strategies of invasive species of Impatiens in central Europe. Annals of
Botany 110(7): 1429-1438.
310. Skálová H., Jarošík V., Dvořáčková Š., Pyšek P. 2013. Effect of intra- and interspecific
competition on the performance of native and invasive species of Impatiens under varying
levels of shade and moisture. PLoS One 85: e62842.
311. Skálová H., Pyšek P. 2009. Germination and establishment of invasive and native Impatiens
species in species-specific microsites. Neobiota 8: 101–109.
312. Skelly J.M., Innes J.L., Savage J.E., Snyder K.R., VanderHeyden D., Zhang J., Sanz M.J.,
1999. Observation and confirmation of foliar ozone injury symptoms of native plant species
of Switzerland and southern Spain. Water, Air and Soil Pollution 116: 227–234.
313. Small C.J., McCarthy B.C. 2003. Spatial and temporal variability of herbaceous vegetation
in an eastern deciduous forest. Plant Ecology 164(1): 37–42.
314. Sokal R. R., Rohlf F. J. 1995. Biometry New York. NY: WH Freeman & Co.
315. Solarz W., Tokarska-Guzik B., Zając K., Chmura D., Cierlik G., Król W. 2005. Zasady
postępowania z gatunkami roślin i zwierząt obcymi rodzimej faunie i florze. Instytut
Ochrony Przyrody, PAN.
316. Sorte C.J.B., Williams S.L. Carlton J.T. 2010. Marine range shifts and species
introductions: comparative spread rates and community impacts. Global Ecology and
Biogeography 19: 303 – 316.
317. Sowa R., Warcholińska A.U. 1980. Flora synantropijna Bełchatowa. Łódzkie Towarzystwo
Naukowe, Sprawozdania z Czynności i Posiedzeń naukowych 43: 1–7.
318. Štajerová K., Šmilauerova M., Šmilauer P. 2009. Arbuscular mycorrhizal symbiosis of
herbaceous invasive neophytes in the Czech Republic. Preslia 81: 341-355.
204
319. Staniaszek-Kik M, Żarnowiec J 2013. Invasive alien plants on decaying wood and on treefall disturbances in forests in the Karkonosze Mts (Sudeten, Sw Poland). Inżynieria
Ekologiczna 32: 155–163.
320. Stary P., Laska P. 1999. Adaptation of native syrphid flies to new exotic plant (Impatiens
spp.) - aphid-ant associations is Central Europe (Dip., Syrphidae; Hom., Aphididae; Hym.,
Formicidae). Anzeiger für Schädlingskunde-Journal of Pest Science 72: 72–75.
321. Stockwell, C.A., Hendry, A.P. Kinnison, M.T. (2003) Contemporary evolution meets
conservation biology. Trends in Ecology and Evolution 18: 94–101.
322. Stohlgren T. J., Barnett D. T., Kartesz J. T. 2003. The rich get richer: patterns of plant
invasions in the United States. Frontiers in Ecology and the Environment 1(1): 11–14.
323. Stohlgren T.J., Binkley D., Chong G.W., Kalkhan M.A., Schell L.D., Bull K.A., Otsuki Y.,
Newman G., Bashkin M., Son Y. 1999. Exotic plant species invade hot spots of native plant
diversity. Ecological Monographs 9: 25–46.
324. Stohlgren T.J., Jarnevich C., Chong G.W., Evangelista P. 2006. Scale and plant invasions: a
theory of biotic acceptance. Preslia 78: 405–426.
325. Strzemski M., Witek T. 1978. Gleby. In: Leszczycki S (Ed.) Narodowy atlas Polski, PAN,
Wrocław - Warszawa - Kraków – Gdańsk.
326. Sudnik-Wójcikowska B, Kożniewska B., 1988. Słownik z zakresu szaty roślinnej.
Wydawnictwa Uniwersytetu Warszawskiego, Warszawa.
327. Sukopp H. 1962. Neophyten in natürlichen Pflanzengesellschaften Mitteleuropas. Berichte
der Deutschen Botanischen Gesellschaf 75(6): 193-205.
328. Świerkosz K., Reczyńska K., Kulpiński K., Tyc A. 2011. Udział gatunków obcych w
zbiorowiskach z klasy Asplenietea trichomanis (Br.-Bl. in Meier & Br.-Bl. 1934) Oberd.
1977 w poludniowej Polsce. In: Kącki Z., Stefańska-Krzaczek. (eds.) Synantropizacja w
dobie zmian różnorodności biologicznej. Acta Botanica Silesiaca 6: 97–114.
329. Symonides E. 1988a. Population dynamics of annual plants. In: Davy A.J., Hutchings M.J.,
Watkinson A.R. (eds) Plant population ecology, Blackwell Scientific Publications, pp. 221248.
330. Symonides E. 1988b. On the ecology and evolution of annual plants in disturbed
environments. Vegetatio 77: 21–31.
331. Szary A., Michalik S. 1998. Zbiorowiska roślinne rezerwatu krajobrazowego Dolina
Racławki. Prądnik. Prace i Materiały Muzeum im. Prof. Władysława Szafera 11-12: 167–
182.
332. Szmajda P. 1974. Flora synantropijna Stargardu Szczecińskiego i Pyrzyc. PTPN, Bad. Fizj.
Pol. Zach. 27, Seria B Botanika: 227–261.
333. Szwed W. 1986. Ecological scale of chosen vascular plants of subalpine zones in Babia
Góra massif. Poznańskie Towarzystwo Przyjaciół Nauk, Wydział MatematycznoPrzyrodniczy. Prace Komisji Biologicznej 69: 1–91.
334. Tabak N.M., Wettberg E. 2008. Native and introduced jewelweeds of the Northeast.
Northeastern Naturalist 15(2): 159–176.
335. Tanner R. 2008. Datasheets: Impatiens parviflora (small balsam). CAB Europe - UK,
Bakeham Lane, Egham, Surrey TW20 9TY, UK, online: http://www.cabi.org, date: April
2014.
336. Ter Braak C.J.F., Šmilauer P. 2002. Canoco reference and CanoDraw for Windows User’s
guide: Software for Canonical Community Ordination (version 4.5) – Microcomputer
Power (Ithaca, NY, USA), 500 p.
337. Tichý L. 2002. JUICE, software for vegetation classification. Journal of Vegetation Science
13: 451-453.
338. Tilman D. 1997. Community invasibility, recruitment limitation, and grassland biodiversity.
Ecology 78: 81–92.
205
339. Tokarska-Guzik B. 2005a. Invasive ability of kenophytes occurrig in Poland: a tenative
assesment. In: Nentwig W. et al. (eds.). Biological Invasions from ecology to control.
NEOBIOTA 6: 47–65.
340. Tokarska-Guzik B. 2005b. The establishment and spread of alien plant species
(kenophytes) in the flora of Poland, Uniwersytet Śląski, Katowice.
341. Tokarska-Guzik B., Dajdok Z., Zając M., Zając A., Urbisz A., Danielewicz W., Hołdyński
C. 2012. Rośliny obcego pochodzenia w Polsce ze szczególnym uwzględnieniem gatunków
inwazyjnych. GDOIŚ, Warszawa.
342. Tokarska-Guzik B., Urbisz Al., Urbisz An., Węgrzynek B., Nowak T. and Pasierbiński A.
2007. Regional scale assessment of alien plant invasions: a case study for the Silesian
Upland (southern Poland). In: Tokarska-Guzik B., Brock J.H., Brundu G., Child L., Daehler
C.C., Pyšek P. (eds.), Plant Invasions: Human perception, ecological impacts and
management, Backhuys Publishers, Leiden, the Netherlands, pp. 171–188.
343. Tóthová E. 2011. Bionómia invázneho druhu Impatiens parviflora D.C. a ohrozenosť NPR
Patianska cerina jeho rozšírením. Technická univerzita vo Zvolene, Fakulta ekológie a
environmentalistiky, Katedra environmentálneho inžinierstva, diplomova práce, Zvolen,
(mscr).
344. Towpasz K., Tumidajowicz 1989. Fluctuations of the dominant herb layer species in an
oak-hornbeam forest (Tilio-Carpinetum typicum) in the Wieliczka Foothills (Pogórze
Wielickie) near Bochnia (Southern Poland). In: Kornaś J. Prace Botaniczne. Zeszyty
Naukowe Uniwersytetu Jagiellońskiego: 113–126.
345. Trepl L. 1984. Über Impatiens parviflora DC. als Agriophyt in Mitteleuropa. J. Cramer,
Vaduz.
346. Trouvelot A., Kough J.L., Gianinazzi-Pearson V. 1986. Mesure du taux de mycorhization
VA d’un systeme radiculaire. Recherche de methodes d’estimation ayant une signification
fonctionnelle. In: Gianinazzi-Pearson, V., Gianinazzi, S. (eds.), Physiological and Genetical
Aspects of Mycorrhizae. INRA, Paris, pp. 217–221.
347. Truszkowska W. 1953. Mykotrofizm olesów Białowiskiego Parku Narodowego i
Domaszyna pod Wrocławiem. Acta Societatis Botanicorum Poloniae 22: 737–752.
348. Turnau K., Mitka J., Kedzierska A. 1992. Mycorrhizal status of herb-layer plants in a
fertilized oak-pine forest. Plant and soil. 143(1): 148–152.
349. Ugoletti P., Stout J.C., Jones M.B. 2011. Ecophysiological traits of invasive and noninvasive introduced Impatiens species. Biology And Environment: Proceedings Of The
Royal Irish Academy 111(3): 1–14.
350. Uherčiková E., Eliáš P. 1987. Standing crop, dominance, and species diversity of tall-herb
communities in the Male Karpaty Mts., Western Slovakia. Ekologia 6(2): 147–163.
351. Urban M. 2009. Mapování výskytu netykavek (Impatiens glandulifera a I. parviflora) na
vybraném úseku řeky Želivky. Faculty of Biological Sciences, University of South
Bohemia, Ceske Budejovice, Czech Republic (mscr).
352. Urbisz A. 2004. Konspekt flory roślin naczyniowych Wyżyny KrakowskoCzęstochowskiej. Wyd. Uniwersytetu Śląskiego. Katowice.
353. USDA-ARS, 2008. Germplasm Resources Information Network (GRIN). Online Database.
Beltsville, USA: National Germplasm Resources Laboratory. http://www.ars-grin.gov/cgibin/npgs/html/tax_search.pl
354. Van Holle B., Delcourt H.R., Simberloff D. 2003. The importance of biological inertia in
plant community resistance to invasion. Journal of Vegetation Science 14: 425–432.
355. Verloove F. 2006. Catalogue of the Neophytes in Belgium (1800-2005). Scripta Botanica
Belgica 39, 89.
206
356. Vervoort A. 2011. Écologie comparative de 3 espèces du genre Impatiens. Une espèce
indigène, I. noli-tangere L., et deux espèces exotiques envahissantes, I. parviflora DC. et I.
glandulifera ROYLE. Université catholique de Louvain.
357. Vervoort, A., Cawoy, V., Jacquemart, A. L. 2011. Comparative reproductive biology in cooccurring invasive and native Impatiens species. International Journal of Plant Sciences,
172(3): 366–377.
358. Vervoort, A., Jacquemart, A. L. 2012. Habitat overlap of the invasive Impatiens parviflora
DC with its native congener I. noli-tangere L. Phytocoenologia 42(3-4): 249–257.
359. Vogel S. 1943.Monographie über dre Impatiens-Arten in Deutschland (mscr).
360. Vrchotová N., Šerá B., Krejčová J. 2011. Allelopathic activity of extracts from Impatiens
species. Plant Soil and Environment 57 (2): 57–60.
361. Wamelink G. W. W., Joosten V., Dobben H. V., Berendse F. 2002. Validity of Ellenberg
indicator values judged from physic-chemical field measurements. Journal of Vegetation
Science 13(2): 269–278.
362. Wang B., QiuY.-L. 2006. Phylogenetic distribution and evolution of mycorrhizas in land
plants. Mycorrhiza 16: 299–363.
363. Wang D., Maughan M. W., Sun J., Feng X., Miguez F., Lee D., Dietze M. C. 2012. Impact
of nitrogen allocation on growth and photosynthesis of Miscanthus (Miscanthus×
giganteus). GCB Bioenergy 4(6): 688–697.
364. Warcholińska A.U. 2005. Flora roślin naczyniowych terenów kolejowych Sieradza. PTPN,
Badania Fizjograficzne nad Polską Zachodnią, seria B, Botanika 54: 113–124.
365. Weber E. 2003. Invasive plant species of the world: a reference guide to environmental
weeds. – CAB International Publ., Wallingford.
366. Weber, E.F. 1997. The alien flora of Europe: a taxonomic and biogeographic review.
Journal of Vegetation Science 8: 565–72.
367. Węglarski K. 1991. Amplitudy ekologiczne wybranych gatunków roślin naczyniowych
Wielkopolskiego Parku Narodowego. Uniwersytet im. A. Mickiewicza w Poznaniu,
Poznań.
368. Whitelam G.C., Johnson C.B. 1982. Photomorphogenesis in Impatiens parviflora and other
plant species under simulated natural canopy radiations. New Phytologist 90(4): 611–618.
369. Wika S., Błońska A., Wesoły S. 2002. Zbiorowiska roślinne na obszarze nekropolii
Katowic. Kształtowanie środowiska geograficznego i ochrona przyrody na obszarach
uprzemysłowionych i zurbanizowanych, WBIOŚ WNoZ UŚ, Katowice-Sosnowiec 33: 45–
54.
370. Wilgen B.W.V., Richardson D.M., Maitre D.C.L., Marais C., Magadlela D. 2001. The
economic consequences of alien plant invasions: examples of impacts and approaches to
sustainable management in South Africa. Environment, Development and Sustainability 3:
145–168.
371. Williams J.R. 1954. The biological control of weeds. Report of the Sixth Commonwealth
Entomological Congress, London, United Kingdom, pp. 95–98.
372. Williamson M. 1996. Biological invasions. Chapman & Hall, London.
373. Williamson M. H., Brown K. C., Holdgate M. W., Kornberg H., Southwood R., Mollison
D. 1986. The analysis and modelling of British invasions [and discussion]. Philosophical
Transactions of the Royal Society of London. B, Biological Sciences 314(1167): 505-522.
374. Wittig R. 1977. Agriophyten in Westfalien. Natur und Heimat 37: 13–23.
375. Wójcik P. 2007. Międzypokoleniowe utrzymywanie się różnic wielkości kwiatów
pomiędzy populacjami niecierpka drobnokwiatowego (Impatiens parviflora DC) w
ujednoliconych warunkach uprawy. Praca licencjacka, Uniwersytet Warszawski (mscr).
376. Woźniak G., Błońska A. 2009. Bioindicative applicability of vegetation on post coal mine
sites. Polish Journal of Environmental Studies 6: 127–131.
207
377. Wulf M. 2003. Preference of plant species for woodlands with differing habitat continuities.
Flora 198: 444-460.
378. Young J.E. 1981a. Light quality and stem growth in Impatiens parviflora DC. New
Phytologist 89: 47–59.
379. Young J.E. 1981b. The use of canonical correlation analysis in the investigation of
relationships between plant growth and environmental factors. Annales of Botany 48: 811–
825.
380. Yurkonis K.A., Meiners S.J. 2004. Invasion impacts local species turnover in a successional
system. Ecology Letters 7: 764–769.
381. Załuski T. 1974. Materiały do flory naczyniowej okolic Sieradza. Zeszyty Naukowe
Uniwersytetu Łódzkiego seria II 54: 185–201.
382. Zarzycki K. 1976. Ecodiagrams of common vascular plants in the Pieniny Mountains (the
Polsih Werstern Carpathians). Part I. Ecodiagrams of selected woodland species. Fragmenta
Floristica et Geobotanica 22(4): 479–497.
383. Zarzycki K. 1984. Ekologiczne liczby wskaźnikowe roślin naczyniowych. Instytut Botaniki
PAN, Kraków.
384. Zarzycki K., Trzcińska-Tacik H., Różański W., Szeląg Z., Wołek J., Korzeniak U. 2002.
Ecological indicator values of vascular plant of Poland. W. Szafer Institute of Botany,
PAN.
385. Zelenchuk T.K. 1961. The content of viable seed in meadow peaty soils of the L’vov
region. Byull. Mosk. Obshch. Ispyt. Prir., Otdel. Biol. 66: 77-92
386. Zielonka T., Piątek G. 2004. The herb layer and dwarf shrubs colonization of decaying logs
in subalpine forest in the Polish Tatra Mountains. Plant Ecology 172: 63–72.
387. Zitter T. A., Murphy J.F. 2009. Cucumber mosaic. The Plant Health Instructor. DOI:
10.1094/PHI-I-2009-0518-01.
Available:
http://www.apsnet.org/edcenter/intropp/lessons/viruses/Pages/Cucumbermosaic.aspx
208
Appendix 1. Phytocoenotic spectrum of neophytes in forest communities of Silesian Upland. Constancy and median non-zero cover are
included in the table. Only relevés with participation of neophytes are presented
Name of plant association
L-P
M-P
Q-P
C-P
PI
C-Q
T-C
L-F
EF
FU
FA
ZA
RA
209
Number of relevés in the table
80
8
128
16
11
31
94
36
19
12
28
17
5
Impatiens parviflora
I(3)
II(2)
III(13)
II(2)
V(63)
III(3)
IV(13)
III(3)
IV(3)
IV(3)
IV(3)
II(7)
III(13)
Acer negundo a
.
.
.
.
.
.
I(3)
.
.
.
.
.
.
Acer negundo b
.
.
I(2)
.
.
.
.
.
.
.
.
I(2)
.
Acer negundo c
.
.
.
.
.
.
I(2)
.
.
.
.
.
.
Digitalis purpurea
.
.
.
.
.
I(2)
.
.
.
.
.
.
.
Epilobium adenocaulon
.
.
.
.
.
I(2)
.
.
.
.
.
.
.
Erechtites hieracifolia
I(2)
.
.
.
.
.
.
.
.
.
.
.
.
Erigeron annuus
.
.
I(2)
.
.
.
.
.
.
I(13)
.
.
.
Galinsoga ciliata
.
.
.
.
.
I(2)
.
.
.
.
.
.
.
Galinsoga parviflora
.
.
.
.
.
.
I(2)
.
.
.
.
.
.
Impatiens glandulifera
.
.
.
I(2)
.
.
I(3)
.
.
.
.
.
.
Juncus tenuis
.
.
.
I(2)
.
.
.
.
.
.
.
.
.
Lupinus polyphyllus
I(3)
.
I(2)
.
.
.
.
.
.
.
I(3)
.
.
Oxalis fontana
.
I(3)
.
.
.
I(2)
I(2)
I(2)
.
.
.
.
.
Padus serotina a
I(13)
.
I(3)
.
I(38)
.
I(2)
.
.
.
I(13)
I(3)
I(13)
Padus serotina b
II(13)
I(13)
II(13)
I(13)
I(38)
I(8)
I(8)
I(3)
I(3)
.
I(7)
III(3)
I(13)
Padus serotina c
III(3)
III(2)
II(3)
I(2)
.
I(13)
I(3)
II(2)
I(2)
.
I(3)
II(2)
.
Quercus rubra a
I(13)
.
II(38)
III(3)
I(3)
II(8)
I(13)
II(3)
.
.
.
.
.
Quercus rubra b
II(3)
I(3)
II(3)
II(3)
I(13)
II(3)
I(3)
I(13)
I(8)
.
I(7)
I(2)
.
Quercus rubra c
III(2)
.
II(3)
II(2)
II(13)
II(2)
II(3)
II(3)
I(3)
.
I(2)
II(3)
.
Reynoutria japonica
I(3)
.
.
.
.
I(13)
.
.
I(13)
.
.
.
.
Robinia pseudacacia a
I(3)
.
I(2)
.
.
.
I(2)
.
.
I(2)
.
I(8)
.
Robinia pseudacacia b
I(3)
.
I(2)
.
.
.
.
.
.
.
.
I(8)
.
Robinia pseudacacia c
I(3)
.
I(3)
.
.
.
.
.
.
I(2)
.
I(2)
.
Solidago canadensis
.
.
I(2)
.
.
.
I(3)
.
.
.
I(3)
I(3)
.
Solidago gigantea
I(2)
.
I(3)
I(3)
I(3)
I(3)
I(13)
.
I(2)
.
I(13)
I(3)
.
Explanations: L-P- Leucobryo-Pinetum; M-P- Molinio-Pinetum; Q-P- Querco roboris-Pinetum; C-P- Calamagrostio villosae-Pinetum; PI anthropogenic community with P.
sylvestris; C-Q – Calamagrostio-Quercetum; T-C – Tilio-Carpinetum; L-F- Luzulo pilosae- Fagetum; EF- Fagenion; F-U- Ficario-Ulmetum; F-A- Fraxino-Alnetum; ZA –
anthropogenic community of Querco-Fagetea class; R-A- Ribeso nigri-Alnetum.
Appendix 2.
Floristic diversity of accompanying species Impatiens parviflora in nature
reserves in the Jurassic Upland (KC) and the Silesian Upland (SU). In the table percent percentage
of occupied subplots by species are given. The order is given according to the decreasing
frequency
Species
KC
SU
Species
KC
SU
Oxalis acetosella
23.73
7.68 Agrostis canina
0.03 0.45
Galeobdolon luteum
19.73
1.13 Ranunculus repens
0.57 0.00
Fagus sylvatica
15.00
3.71 Deschampsia caespitosa
0.00 0.42
Aegopodium podagraria
17.17
1.68 Moneses uniflora
0.00 0.39
Galium odoratum
13.57
1.95 Carex elongata
0.00 0.37
6.97 Anthriscus silvestris
0.43 0.00
Carex brizoides
6.33
Asarum europaeum
13.03
0.45 Brachypodium sylvaticum
0.40 0.03
Acer pseudoplatanus
13.10
0.26 Galium aparine
0.33 0.08
Viola reichenbachiana
11.37
1.26 Orthilia secunda
0.00 0.34
Vaccinium myrtillus
2.17
6.95 Padus serotina
0.00 0.34
Athyrium filix-femina
6.73
2.58 Primola elatior
0.43 0.00
Maianthemum bifolium
4.00
4.42 Cardamine amara
0.00 0.32
Hedera helix
9.20
0.00 Agrostis capillaris
0.00 0.29
Pulmonaria obscura
7.57
0.00 Crataegus monogyna
0.20 0.13
Mycelis muralis
5.50
1.08 Dentaria glandulosa
0.20 0.13
Pteridium aquilinum
0.00
4.76 Ranunculus lanuginosus
0.37 0.00
Quercus robur
2.07
3.11 Solanum dulcamara
0.00 0.29
Rubus idaeus
0.50
4.11 Tussilago farfara
0.37 0.00
Anemone nemorosa
4.33
0.97 Abies alba
0.00 0.26
Impatiens noli-tangere
5.10
0.34 Fraxinus excelsior
0.33 0.00
Urtica dioica
3.70
1.39 Myosotis palustris
0.33 0.00
Circaea lutetiana
3.90
0.66 Scirpus sylvaticus
0.00 0.26
Mercurialis perennis
4.13
0.39 Veronica chamaedrys
0.33 0.00
Sanicula europaea
4.63
0.00 Alnus glutinosa
0.03 0.21
Geranium robertianum
4.50
0.08 Anthoxanthum odoratum
0.00 0.24
Acer platanoides
4.33
0.18 Campanula trachelium
0.17 0.11
Rubus hirtus
2.03
1.97 Equisetum arvense
0.00 0.24
Sambucus nigra
4.50
0.03 Eupatorium cannabinum
0.00 0.24
Pinus sylvestris
0.10
3.47 Leucobryum glaucum d
0.00 0.24
Festuca gigantea
1.40
2.26 Campanula rapunculoides
0.27 0.00
Dryopteris filix-mas
1.80
1.89 Lycopus europeus
0.00 0.21
Galeopsis pubescens
3.67
0.34 Vaccinium vitis-idaea
0.00 0.21
Geum urbanum
3.60
0.37 Veronica officinalis
0.27 0.00
Convallaria majalis
0.07
2.71 Calamagrostis canescens
0.00 0.18
Trientalis europaea
0.03
2.63 Dicranella heteromalla d
0.23 0.00
210
Species
KC
SU
Species
Luzula pilosa
0.30
2.34 Glyceria maxima
0.00 0.18
Carpinus betulus
0.83
1.89 Calamagrostis arundinacea
0.00 0.16
Tilia cordata
0.10
2.29 Epilobium montanum
0.20 0.00
Sorbus aucuparia
1.57
1.11 Juncus effusus
0.00 0.16
Dryopteris carthusiana
1.20
1.37 Pleurozium schreberi d
0.00 0.16
Melica nutans
2.10
0.61 Quercus petraea
0.20 0.00
Ajuga reptans
2.17
0.53 Rhamnus catharticus
0.20 0.00
Carex sylvatica
2.33
0.39 Melampyrum nemorosum
0.00 0.13
Geranium phaeum
2.77
0.00 Poa trivialis
0.00 0.13
Stachys sylvatica
2.50
0.18 Veratrum lobelianum
0.17 0.00
Deschampsia flexuosa
0.00
2.08 Acer campestre
0.13 0.00
Chaerophyllum aromaticum
2.47
0.00 Galeopsis speciosa
0.00 0.11
Poa nemoralis
1.70
0.47 Lonicera xylosteum
0.13 0.00
Allaria officinalis
2.00
0.00 Monotropa hypopitys
0.13 0.00
Euonymus europaeus
1.77
0.11 Padus avium
0.13 0.00
Fragaria vesca
0.03
1.47 Senecio rivularis
0.00 0.11
Rubus caesius
0.00
1.50 Viola mirabilis
0.13 0.00
Cruciata glabra
1.57
0.24 Alnus incana
0.00 0.08
Epipactis helleborine
1.83
0.00 Campanula persicifolia
0.10 0.00
Paris quadrifolia
1.13
0.50 Dactylis glomerata
0.00 0.08
Scrophularia nodosa
1.53
0.11 Daphne mezereum
0.10 0.00
Hieracium murorum
1.33
0.13 Lamium purpureum
0.00 0.08
Ulmus glabra
1.47
0.00 Plantago major
0.00 0.08
Calamagrostis epigejos
0.00
1.05 Poa annua
0.00 0.08
Picea abies
0.00
1.05 Pyrola chlorantha
0.00 0.08
Ribes nigrum
1.07
0.16 Salix cinerea
0.00 0.08
Quercus rubra c
0.00
0.95 Taraxacum officinale
0.00 0.08
Chrysosplenium alternifolium
1.17
0.00 Cirsium arvense
0.07 0.00
Lysimachia vulgaris
0.00
0.89 Equisetum fluviatile
0.00 0.05
Populus tremula
0.00
0.87 Euphorbia amygdaloides
0.00 0.05
Actaea spicata
1.07
0.00 Gymnocarpium dryopteris
0.00 0.05
Chaerophyllum hirsutum
0.00
0.84 Humulus lupulus
0.00 0.05
Stellaria holostea
0.27
0.63 Luzula luzuloides
0.07 0.00
Hepatica nobilis
0.73
0.18 Melandrium album
0.07 0.00
Milium effusum
0.13
0.66 Oxalis stricta
0.07 0.00
Glechoma hederacea
0.43
0.39 Viburnum opulus
0.07 0.00
Polytrichastrum formosum
0.93
0.00 Aconitum moldavicum
0.03 0.00
Astrantia major
0.73
0.13 Alium ursinum
0.03 0.00
Frangula alnus
0.00
0.71 Cirsium palustre
0.00 0.03
211
KC
SU
Species
KC
SU
Species
Ficaria verna
0.87
0.00 Equisetum palustre
0.00 0.03
Lamium maculatum
0.80
0.00 Holcus lanatus
0.00 0.03
Lilium martagon
0.80
0.00 Knautia arvensis
0.03 0.00
Lysimachia nummularis
0.77
0.00 Lamium album
0.03 0.00
Poa palustris
0.77
0.00 Lapsana communis
0.03 0.00
Lathyrus vernus
0.73
0.00 Lathyrus niger
0.03 0.00
Betula pendula
0.00
0.55 Orthodicranum montanum
0.03 0.00
Corylus avellana
0.03
0.53 Polygonatum verticillatum
0.03 0.00
Polygonatum multiflorum
0.40
0.24 Sambucus racemosa
0.00 0.03
Rubus plicatus
0.00
0.50 Viola palustris
0.00 0.03
Senecio nemorensis
0.10
0.42
212
KC
SU
Appendix 3. Frequency of species accompanying I. parviflora in microhabitats associated
with coarse woody debris. Abbreviations: ANL – area near log, H – hollow, LUC – logs under
canopies, LUCO – logs under canopy openings, RP – root plate, S – snag, TFD – treefall
disturbance, TF – Total frequency
ANL
N (number of species)
14(19)
Acer platanoides
4
Acer pseudoplatanus
1
Aegopodium podagraria
2
Ajuga reptans
1
Anemone nemorosa
1
Asplenium sp
0
Athyrium felix-femina
0
Betula pendula seedling
0
Brachypodium sylvaticum
1
Carex brizoides
0
Carex digitata
0
Carex silvatica
1
Chamaenerion angustifolium
0
Chelidonium majus
0
Circaea lutetiana
3
Convallaria maialis
0
Cruciata glabra
1
Dryopteris sp.
0
Epipactis helleborine
1
Eupatorium cannabinum
0
Euphorbia cyparissias
0
Fagus sylvatica
7
Galeobdolon luteum
0
Galeopsis pubescens
0
Galium odoratum
9
Geranium robertianum
0
Hedera helix
1
Impatiens parviflora
13
Lathyrus vernus
1
Luzula pilosa
1
Maianthemum bifolium
0
Melica nutans
1
Mercurialis perennis
0
Moehringia trinervia
0
Mycelis muralis
2
Pinus sylvestris
0
Poa nemoralis
1
Rubus sp.
0
Sambucus nigra
0
Sambucus racemosa
0
Solidago canadensis
0
Taraxacum officinale
0
Tussilago farfara
0
H
3(5)
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
0
0
0
0
0
2
0
0
0
0
2
0
0
0
0
LUC
LUCO
15(6)
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
0
2
0
0
2
0
0
3
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
213
20(13)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
9
0
1
2
1
1
9
1
0
0
0
8
0
1
0
1
0
0
0
0
1
0
RP
S
26(32) 20(21)
0
0
0
0
3
1
3
1
1
0
0
0
3
1
0
0
2
0
0
1
1
0
0
1
3
0
1
0
1
1
2
0
3
0
0
1
0
0
0
1
1
0
8
3
1
0
0
1
6
4
2
0
1
1
18
12
0
0
1
1
1
0
0
2
5
2
0
1
3
2
1
0
3
2
2
4
1
0
1
0
1
0
4
0
3
0
TFD
12(28)
1
1
1
0
0
0
2
0
0
0
0
0
0
1
2
0
1
2
0
0
0
6
1
1
2
4
1
12
1
0
1
1
2
0
5
0
2
1
1
0
0
4
1
TF
121
5
2
7
5
2
1
6
1
3
1
1
2
3
2
7
2
5
5
1
1
1
35
2
3
25
7
5
70
3
3
2
4
19
3
13
1
9
7
4
1
1
9
4
Urtica dioica
Veronica chamaedrys
Veronica officinalis
Viola reichenbachiana
ANL
0
0
0
0
H
0
0
0
1
LUC
0
0
0
0
LUCO
0
0
0
1
214
RP
0
0
2
2
S
0
0
0
2
TFD
1
1
0
2
TF
1
1
2
8
Streszczenie (Polish summary)
Damian Chmura
Biologia i ekologia inwazji Impatiens parviflora DC na naturalnych
i połnaturalnych siedliskach
W ostatnich dekadach byliśmy świadkami narodzin nowej dziedziny w obrębie nauk
biologicznych – ekologii inwazji. Zjawisko rozprzestrzeniania się i zadamawiania obcych
gatunków w nowej ojczyźnie stanowi, jako drugie po niszczeniu siedlisk, największe zagrożenie
dla rodzimej różnorodności biotycznej. Stąd też nie dziwi zainteresowanie zagadnieniem
rozprzestrzeniania się gatunków obcych u botaników, zoologów, ekologów jak i praktyków
ochrony przyrody.
Niecierpek drobnokwiatowy Impatiens parviflora DC, rodzimy dla Wschodniej Azji we
florze Polski ma status kenofita (holoagriofita). Można go uznać za „prawdziwie” inwazyjny
gatunek obcego pochodzenia, ponieważ od momentu introdukcji proces jego rozprzestrzeniania
się, zadamawiania przebiegał niemalże w sposób spontaniczny bez udziału człowieka. Obecnie to
gatunek szeroko rozpowszechniony w lasach, wzdłuż dróg leśnych i na siedliskach ruderalnych.
Jednakże opinie co do wpływu tego gatunku na rodzimą florę, pewne aspekty biologii,
ekologii, które mają go czynić skuteczną rośliną inwazyjną są podzielone. Doniesienia z różnych
części Europy przedstawiają odmienne a czasem przeciwstawne wyniki badań.
Głównym celem badań miało być podsumowanie i przedyskutowanie mechanizmów
inwazji niecierpka drobnokwiatowego w oparciu o wyniki własne i doświadczenia innych autorów.
Badania w niniejszej pracy mieszczą się w zakresie ekologii populacji, ekologii gatunku,
fitosocjologii, ekologii zbiorowisk roślinnych i biocenologii. Studia nad gatunkiem
przeprowadzone były w latach 2005-2012 lecz użyty materiał fitosocjologiczny pochodzi z okresu
1998-2012 (485 zdjęć). Prace terenowe wykonano na obszarze Wyżyny Śląskiej oraz na Wyżynie
(Jurze) Krakowsko-Częstochowskiej. W tych dwóch mezoregionach, na terenie wybranych
rezerwatów przyrody, założono łącznie 68 powierzchni badawczych o boku 10 m x 10 m
podzielonych na 100 poletek w różnych zbiorowiskach leśnych celem badań nad ekologią
zbiorowisk roślinnych z udziałem niecierpka. Na 10 z nich później prowadzono 8-letnie
obserwacje. Dokonano również obserwacji na kilkunastu typach podłoży głównie związanych
z różnym typem martwego drewna, gdzie śledzono losy oznakowanych osobników. Materiał
(nasiona) oprócz wyżej wymienionych regionów pozyskano także z obszaru Węgier do badań nad
biologią kiełkowania. Pobrano 30 prób do badań mikologicznych oraz ponad 150 prób gleby.
Przeprowadzono badania morfologiczne, fenologiczne, glebowe, zbadano cechy historii życia life
history traits, mykologiczne, fitosocjologiczne oraz długoterminowe badania ekologiczne. Wyniki
prac terenowych oraz laboratoryjnych poddano różnym analizom statystycznym.
Potwierdzono, że Impatiens parviflora jest gatunkiem o wysokiej plastyczności
morfologicznej, która ma związek z dużą dostępnością światła i nutrientów w podłożu. To wysokie
zróżnicowanie utrzymuje się w różnicach w cechach morfometrycznych między zbiorowiskami
i typami siedlisk, w których gatunek się pojawia.
Gatunek jest raczej beneficjentem (ang. passenger) niż czynnikiem sprawczym zaburzeń
(ang. driver). W dobrze zachowanych leśnych siedliskach wykorzystuje wolną przestrzeń i zasoby
nieużytkowane przez inne rosliny. W bardziej zaburzonych leśnych siedliskach kolonizuje takie
miejsca z innymi gatunkami.
215
Roślina ta ma wielką zdolność zasiedlania bardzo różnorodnych mikrosiedlisk, co jest
jednym z najważniejszych aspektów jej inwazyjności. Pomimo krótkodystansowej dyspersji
wynikającej z autochorii, rozprzestrzenianie się niecierpka jest bardzo efektywne.
Obecność osobników niecierpka na takich siedliskach jak dziuple, kora żywych drzew
świadczy o możliwościach zoochorii głównie przez ptaki.
Zdolność do konkurencji u tego gatunku jest niewielka zwłaszcza w warunkach warstwy runa.
Wzrost lub zmniejszenie się udziału niecierpka w obecności innych gatunków jest raczej efektem
różnic w wymaganiach siedliskowych niż mechanizmów konkurencji. Biocenotyczna odporność
(bogactwo gatunkowe, pokrycie roślin) jest skuteczną barierą uniemożliwiającą lub redukującą
sukces inwazji tego gatunku. W pewnych mikrosiedliskach (dziuple, kora żywych i martwych
drzew, kłody, dołki powykrotowe) I. parviflora jest najczęstszym kolonizatorem i nalepiej toleruje
stres środowiskowy spośród wszystkich obecnych gatunków roślin naczyniowych.
Interakcje mutualistyczne wydają się sprzyjać inwazji tego gatunku. Mikoryza arbuskularna
w powiązaniu z pewnymi czynnikami siedliskowymi występuje u największych i najpłodniejszych
osobników.
Obecność I. parviflora pod lukami w drzewostanie i pod okapem drzew powoduje wydłużenie
czasu produkcji nasion wskutek przesunięcia się faz fenologicznych.
Mozaika zaburzonych (naturalnie i antropogenicznie) i niezaburzonych siedlisk jest koniecznym
czynnikiem dla występowania trwałych populacji w warunkach leśnych. Populacja niecierpka
drobnokwiatowego w leśnym kompleksie egzystuje jako metapopulacja typu „źródło-ujście”.
Populacje typu źródła występują na miejscach zaburzonych (luki, obrzeża lasu), w których
osobniki charakteryzują się większymi rozmiarami i zwiększoną produkcją nasion. Z kolei
siedliska typowo leśne, niezaburzone, tj.wnętrza lasu są siedliskami dla populacji typu ujście, gdzie
rośliny odznaczają się mniejszą plastycznością morfologiczną i produkcją nasion.
Historia inwazji I. parviflora w danym miejscu ma wpływ na dalszą dynamikę tego gatunku.
Obszary występowania gatunku różnią się stopniem inwazji stąd też można zaobserwować wzrost
udziału niecierpka drobnokwiatowego w niektórych miejscach a w innych jego wycofywanie się.
216
View publication stats