Synchronous palynofloristic extinction and recovery after the end-Permian event in the Prince Charles Mountains, Antarctica: Implications for palynofloristic turnover across Gondwana

Sofie Lindström; Stephen McLoughlin

In the Prince Charles Mountains (PCMs) the conformable Permian–Triassic (P–T) succession is characterised by an abrupt transition from coal-bearing to coal-lacking strata, which coincides with the demise of the Permian Glossopteris-dominated flora. About 32% of the typical Permian spores and pollen are registered for the last time in the uppermost coal. Throughout the earliest Triassic an additional 34% of the lingering Permian taxa disappear, while pioneering typical Triassic taxa appear. This interval of contemporaneous stepwise extinction and recovery resulted in an actual increase in spore-pollen taxa diversity during the earliest Triassic. The estimated average sedimentation rate indicates that the 24 m sampling gap that separates the last Permian assemblage from the first Triassic one represents ca 96 000 years, and that the continued stepwise extinction and recovery lasted for ca 325 000 years. In the aftermath of the end-Permian crisis only 27% of the typical Permian spores and pollen, that were present from the lower McKinnon Member in the Prince Charles Mountains survived to the late Induan, but by then the biodiversity had only decreased by less than 10%. Comparisons of Gondwanan palynological and lithological data indicate that intense global warming had already begun in the Permian, and that high latitude Gondwana areas such as the PCMs, were affected later than areas to the north and west. They also suggest that the end-Permian crisis affected the various Gondwana regions in different ways, but that the end result appears to have been a more equable, sub-humid to semi-arid, and less seasonal climate across southern Gondwana.

Review of Palaeobotany and Palynology 145 (2007) 89 – 122 www.elsevier.com/locate/revpalbo Synchronous palynof loristic extinction and recovery after the endPermian event in the Prince Charles Mountains, Antarctica: Implications for palynof loristic turnover across Gondwana Sofie Lindström a,⁎, Stephen McLoughlin b b a Department of Geology, Geobiosphere Science Centre, Lund University, Sölvegatan 12, SE-223 62 Lund, Sweden School of Natural Resource Sciences, Queensland University of Technology, PO Box 2434 Brisbane, Q 4001, Australia Received 21 December 2005; received in revised form 31 August 2006; accepted 6 September 2006 Available online 24 October 2006 Abstract In the Prince Charles Mountains (PCMs) the conformable Permian–Triassic (P–T) succession is characterised by an abrupt transition from coal-bearing to coal-lacking strata, which coincides with the demise of the Permian Glossopteris-dominated flora. About 32% of the typical Permian spores and pollen are registered for the last time in the uppermost coal. Throughout the earliest Triassic an additional 34% of the lingering Permian taxa disappear, while pioneering typical Triassic taxa appear. This interval of contemporaneous stepwise extinction and recovery resulted in an actual increase in spore-pollen taxa diversity during the earliest Triassic. The estimated average sedimentation rate indicates that the 24 m sampling gap that separates the last Permian assemblage from the first Triassic one represents ca 96 000 years, and that the continued stepwise extinction and recovery lasted for ca 325 000 years. In the aftermath of the end-Permian crisis only 27% of the typical Permian spores and pollen, that were present from the lower McKinnon Member in the Prince Charles Mountains survived to the late Induan, but by then the biodiversity had only decreased by less than 10%. Comparisons of Gondwanan palynological and lithological data indicate that intense global warming had already begun in the Permian, and that high latitude Gondwana areas such as the PCMs, were affected later than areas to the north and west. They also suggest that the end-Permian crisis affected the various Gondwana regions in different ways, but that the end result appears to have been a more equable, sub-humid to semi-arid, and less seasonal climate across southern Gondwana. © 2006 Elsevier B.V. All rights reserved. Keywords: Permian–Triassic transition; Antarctica; palynostratigraphy; palaeobiogeography; extinction; recovery 1. Introduction The end-Permian extinction event is known to have been rapid in a geological context, lasting b500 000 years (Bowring et al., 1998) and perhaps as little as 40 000 years (Twitchett et al., 2001), and it marked the demise of as ⁎ Corresponding author. Tel.: +46 46 2227875. E-mail addresses: sofie.lindstrom@geol.lu.se (S. Lindström), s.mcloughlin@qut.edu.au (S. McLoughlin). 0034-6667/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.revpalbo.2006.09.002 much as 95% of all species on Earth (Benton and Twitchett, 2003). The recovery of global biodiversity to pre-extinction family levels is estimated to have taken 100 Ma (Hallam and Wignall, 1997). At the Permian– Triassic (P–T) boundary type locality near Meishan in China, the extinction pattern for Permian marine fossil species is threefold: 1) a stepwise disappearance of species during some 3 Ma immediately below the boundary with an extinction rate of 33% or less, 2) a sudden dramatic 94% loss of species at the boundary, followed by 90 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 3) a gradual loss of a few species that persisted into the lowermost Triassic (Jin et al., 2000). However, following the formal definition of the base of the Triassic by the first appearance of the conodont element Hindeodus parvus at the base of Bed 27c of the Meishan section in China (Jin et al., 2000), the main phase of the marine faunal extinction and the negative 13Ccarb excursion occurs in Bed 25, some hundred thousand years before the Permian–Triassic boundary. Data from an independently dated section in East Greenland clearly demonstrate that the terrestrial ecosystem collapse and the subsequent extinction of the typical Late Permian Subangaran gymnosperms precedes the P–T boundary (Looy et al., 2001; Twitchett et al., 2001), but also show that the floristic turnover was not instantaneous. The disappearance of the extensive peat deposits that had characterized the northern and southern humid climatic zones of Pangea was caused by dieback of woody swamp-forest vegetation at the end of the Permian (Looy et al., 1999). Land plant recovery and diversification after the end-Permian extinction event is considered to have been relatively slow, taking about 4 Ma (Eshet et al., 1995; Looy et al., 1999). In the Southern Hemisphere, complex coal-forming communities did not reappear until the Middle Triassic and were not extensively developed until the Carnian, i.e. some 23 Ma after the end-Permian crisis (Retallack et al., 1996; Anderson et al., 1999). In order to resolve the ecodynamic forces that caused the global collapse of the terrestrial ecosystem at the P–T transition, it is important to analyze and compare palynological data from different P–T sections with respect to the climatic and depositional conditions and the floristic diversity that characterized each area at that time. In Gondwana, the gymnospermous glossopterids that proliferated during the Middle and Late Permian were the most notable terrestrial casualties of the end-Permian extinction event. Several problems confront palaeontological analysis of Permian–Triassic transitional sequences of Gondwana. One is that most of the P–T sections were deposited in non-marine environments and, in the absence of radiometric data, definition of the P–T boundary is based solely on plant and/or tetrapod fossils. The second problem is that in Gondwanan Permian– Triassic transitional sequences the palynofloral turnover is typically associated with either a palynologically barren interval or a sampling gap straddling the P–T boundary, thus obscuring the signal of short-term ecological changes that took place at that time. A third problem is that although many Gondwanan P–T transitions have been investigated palynologically, detailed reports on taxon appearances, extinctions and changes in abundance are scarce. Additionally, contin- uous P–T sections are uncommon, and in some cases it is possible that sections are punctuated by hiatuses. A conformable sequence of P–T sedimentary rocks crops out in the Prince Charles Mountains, East Antarctica. During the Late Palaeozoic to Early Mesozoic the Prince Charles Mountains area was situated in the centre of southeastern Gondwana, surrounded by the rest of the Antarctic landmass, Australia, India, Madagascar and Africa. Its central position affords the Prince Charles Mountains special importance for correlation of P–T successions across Gondwana. This paper describes the palynofloral turnover across the Permian–Triassic transition in the Prince Charles Mountains, and compares it to other gondwanan successions, in order to evaluate geographic and temporal patterns in palynomorph turnover within southeastern Gondwana. 2. Geological setting and lithostratigraphy 2.1. Tectonic setting The Permian–Triassic sedimentary rocks of the Amery Group that crop out in the northern Prince Charles Mountains are preserved in a narrow faultbounded depression called the Lambert Graben (Fig. 1). Exposures are constrained to the west by the Amery Fault and by ice cover in other directions. The Lambert Graben appears to be a half-graben as gravity studies indicate that only the western side (i.e. the Amery Fault) of the Lambert rift is faulted (Mishra et al., 1999). The traditional view is that the Lambert Graben developed as a part of a major late Palaeozoic–early Mesozoic failed rift system (Stagg, 1985) that was continuous with the Son-Mahanadi Graben of India prior to the breakup of Gondwana (Fedorov et al., 1982; Stagg, 1985). In the elongate but narrow Lambert Graben, basin fill commenced with accumulation of alluvial fan deposits constituting the Radok Conglomerate (Figs. 1 and 2). This unit consists of conglomerates and coarse-grained sandstones, siltstones and minor coal; the sediments being derived predominantly from the uplifted area to the west of Amery Fault and transported easterly into the basin (Fielding and Webb, 1995). The succeeding Bainmedart Coal Measures incorporate repetetive fining-upward cycles of sandstone, siltstone and coal that were deposited predominantly in northerly to northeasterly flowing high-energy braided rivers alternating with lowenergy forest mires and floodplains (Figs. 1 and 2; Fielding and Webb, 1996; McLoughlin and Drinnan, 1997a; McLoughlin et al., 1997). The cessation of coal deposition marks the boundary between the Bainmedart Coal Measures and the Flagstone Bench Formation S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 91 Fig. 1. Map of the western side of Beaver Lake, Prince Charles Mountains, Antarctica, showing the distribution of units in the Amery Group. An enlargement of the Ritchie Point area shows the position of the Permian–Triassic boundary and the location of palynological samples with respect to the measured sections of McLoughlin and Drinnan (1997a,b). (McLoughlin and Drinnan, 1997a,b; McLoughlin et al., 1997). According to McLoughlin et al. (1997), the transition from the coal-bearing Bainmedart Coal Measures to the coal-lacking Flagstone Bench Formation marks the transition between the Permian and Triassic (Figs. 1 and 2). The lower part of the Flagstone Bench Formation, represented by the Ritchie Member, comprises sandstones and siltstones that were deposited with persisting cyclicity (albeit lacking coal) by predominantly northerly directed rivers under the influ- ence of increasing aridity (McLoughlin and Drinnan, 1997b; McLoughlin et al., 1997; Holdgate et al., 2005). The succeeding Jetty Member is represented by typical red-beds deposited in alluvial fans by easterly directed, episodic flows (Figs. 1 and 2). This unit exhibits a grossly fining-upwards succession of conglomerates, thin and discontinuous massive sandstones, and extensive iron-stained mudstones indicative of semi-arid conditions (Webb and Fielding, 1993; McLoughlin and Drinnan, 1997b; McLoughlin et al., 1997; Holdgate 92 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 basin contiguous with the Son-Mahanadi Graben of India has lately been disputed. Boger and Wilson (2003) suggested that all major faulting of the Lambert Graben took place during the Cretaceous, and that the Amery Group sediments were not deposited in a narrow faultbounded depression, but must have been deposited in one of the many sag basins that formed around the palaeo-highland of east Antarctica (Tewari and Veevers, 1993; Veevers et al., 1996). Additionally, Holdgate et al. (2005) indicated that the petrology and geochemistry of the Permian coals of the Prince Charles Mountains are more similar to those of the Godavari Basin than the Son-Mahanadi Basin of India. Harrowfield et al. (2005) rejected the notion that the Lambert Graben is a primarily Cretaceous feature. Building on the argument by Holdgate et al. (2005), they suggested that the Lambert and Godavari basins developed in the Permian and were juxtaposed across a broad intragondwanan rift that later (in the Cretaceous) was reactivated to complete separation of Antarctica and India. 2.2. Latest Permian — McKinnon Member Fig. 2. Stratigraphy and depositional environments of the Amery Group. D.F. Mbr = Dart Fields Member; D.T. Mbr = Dragons Teeth Member. et al., 2005). Palyno- and macrofloral data from the uppermost Flagstone Bench Formation demonstrate a return to more moist conditions in the Norian (Foster et al., 1994; Cantrill and Drinnan, 1994; Cantrill et al., 1995; McLoughlin et al., 1997), when sandstones and minor carbonaceous siltstones of the McKelvey Member were deposited by northerly directed rivers (McLoughlin and Drinnan, 1997b; McLoughlin et al., 1997). The traditional scenario that the Lambert Graben was formed during the Permian as a narrow fault-bounded The McKinnon Member is the uppermost unit of the Bainmedart Coal Measures, first named and described by McLoughlin and Drinnan (1997a). It conformably overlies the Grainger Member and is succeeded conformably by the Flagstone Bench Formation (McLoughlin and Drinnan, 1997a). The McKinnon Member is an approximately 530 m thick sequence of sandstones, siltstones, shales and coals. Deposition took place within alluvial settings, where low-sinuosity rivers transported the sediments in north to northeasterly directions (McLoughlin and Drinnan, 1997a). The lithologies are basically the same as those in underlying members of the Bainmedart Coal Measures, but at least in the lower and middle parts of the McKinnon Member the coal seams are thicker and more abundant (McLoughlin and Drinnan, 1997a). This indicates that extended periods of high water tables with low sediment input must have prevailed during deposition of that part of the member (McLoughlin and Drinnan, 1997a). Within the upper 100 m of the McKinnon Member the coalseams become progressively thinner and less abundant (McLoughlin and Drinnan, 1997a). Together with increasing ratios of silica and aluminium oxides in the coals towards the top of the Bainmedart Coal Measures, this suggests increased weathering and climatic drying towards the end of the Permian (Holdgate et al., 2005). Fifteen palynologically productive samples from the McKinnon Member were investigated. The uppermost sample 95/04 comes from the uppermost coal seam of S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 93 Fig. 3. Composite lithostratigraphic column of the Permian to Early Triassic sequence in the Prince Charles Mountains, with selected important palynoevents. Dotted line I refers to minimum upper boundary of the Lunatisporites pellucidus Zone based on the absence of true Aratrisporites in the samples below. Dotted line II shows the position of a smaller but first extinction phase prior to the disappearance of the coal. Dotted line III shows the position of a level with minor accelerated extinction rate and floral change. Asterisk ⁎ refers to independently dated palynozones (Foster and Archbold, 2001). DTM = Dragons Teeth Member. Australian palynozonations mainly after: 1. Mory and Backhouse (1997), Helby et al. (1987); 2. Price (1997). 94 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 the Bainmedart Coal Measures (Fig. 3), just below the upper boundary of the McKinnon Member. 2.3. Latest Permian to earliest Triassic — Ritchie Member South of Ritchie Point on the western side of Beaver Lake (Fig. 1) the lowermost member of the Flagstone Bench Formation, the Ritchie Member, conformably overlies the McKinnon Member. The Ritchie Member was named and described by McLoughlin and Drinnan (1997b), and corresponds to “the lower Flagstone Bench Formation” of Webb and Fielding (1993). The Ritchie Member is distinguished from the preceding McKinnon Member mainly by the lack of coals (McLoughlin and Drinnan, 1997b). The unit is estimated to be more than 550 m thick. Medium- to very coarse grained, subfeldsarenites are the dominant lithology. Thin carbonaceous siltstones are present in the lowermost part, but higher in the succession these are replaced by variegated, highly ferruginous siltstones (McLoughlin and Drinnan, 1997b; McLoughlin et al., 1997). The sandstone units are thick, laterally extensive, multistorey, interdigitating, and exhibiting trough and planar cross-bedding. Many beds contain botryoidal ferruginous concretions and ferruginous laminae (McLoughlin and Drinnan, 1997b). The Ritchie Member was deposited by northwesterly to northeasterly directed braided rivers on an alluvial plain (McLoughlin and Drinnan, 1997b). Four palynologically productive samples were obtained from the lower part of the Ritchie Member south of Ritchie Point. The lowermost sample, 95/01, comes from a thin siltstone 24 m above the lower boundary of the unit, i.e. the top of the last coal in the McKinnon Member (Fig. 3). Other samples from Ritchie Member sediments exposed on Flagstone Bench were analysed for palynomorphs by McLoughlin et al. (1997). Plate I. Selected Permian taxa from the McKinnon Member, illustrated at ×625, with sample and slide number and England Finder coordinates, and LO number. Scale bar = 40 μm. (see plate on page 95) a). b). c). d). e). f). g). h). i). j). k). l). m). n). o). Leiotriletes directus 95/10:2 S34/2, LO 9894 Microbaculispora tentula 95/34:1 O30/3, LO 9895 Indospora clara 95/28:1 G19/4, LO 9896 Camptotriletes warchianus 95/26B:2 E22/2, LO 9897 Didecitriletes ericianus 95/34:1 J29/3, LO 9898 Laevigatosporites colliensis 95/12:1 Q27/2, LO 9899 Marsupipollenites triradiatus 95/12:1 P31/1, LO 9900 Guttulapollenites hannonicus 95/11:1 W38/1, LO 9901 Protohaploxypinus amplus 95/06:2 K26/3, LO 9902 Striatopodocarpidites cancellatus 95/06:2 Y33/4, LO 9903 Gondisporites raniganjensis 95/04:2 S33/4, LO 9904 Protohaploxypinus rugatus 95/11:1 U34/1, LO 9905 Striatopodocarpidites fusus 95/06:2 Q24/2, LO 9906 Praecolpatites sinuosus 95/26B:1 J29/2, LO 9907 Scheuringipollenites ovatus 95/22:2 X34/4, LO 9908 Plate II. Selected Triassic taxa from the Ritchie Member, illustrated at ×625, with sample and slide number and England Finder coordinates, and LO number. Scale bar = 40 μm. (see plate on page 96) a). b). c). d). e). f). g). h). i). j). k). l). m). n). o). Rugulatisporites trisinus 95/05:1 G26/3, LO 9909 Triplexisporites playfordii 95/02:2 X40/2, LO 9910 Densoisporites nejburgii 95/02:1 K39/4, LO 9911 Uvaesporites verrucosus 95/02:1 U29/2, LO 9912 Densoisporites playfordii 95/02:1 H25/3, LO 9913 Falcisporites australis 95/05:1 Q32/1, LO 9914 Guttulapollenites hannonicus 95/02:1 F35/4, LO 9915 Lundbladispora sp. 95/02:1 K36/4, LO 9916 Protohaploxypinus microcorpus 95/01:1 K36/1, LO 9917 Lunatispoprites noviaulensis 95/02:1 W28/3, LO 9918 Klausipollenites schaubergeri 95/02:1 Q42/3, LO 9919 Protohaploxypinus samoilovichii 95/02:1 V33/3, LO 9920 Lunatisporites pellucidus 95/02:1 P29/4, LO 9921 Maculatasporites sp. 95/02:1 N34/4, LO 9922 Playfordiaspora cancellosa 95/02:2 X40/2, LO 9923 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 Plate I (caption on page 94 ). 95 96 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 Plate II (caption on page 94 ). S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 3. Palynological investigation 3.1. Materials and methods Reconnaissance sampling of the P–T transition near Ritchie Point was undertaken during a sedimentological and stratigraphic investigation of the Amery Group during the Austral summer of 1994–1995 (by Drs. S. McLoughlin and A.N. Drinnan). Previously, the succession near Ritchie Point was considered to be entirely Permian in age. The samples were processed using standard palynological preparation techniques involving HF, HCl, and HNO3. Two to three strew slides were mounted from each sample. In some cases the samples were subsequently treated with Schulze's solution, and an additional set of slides prepared. Quantitative investigation involved 500 counts of the total organic content for each sample, dividing it into coal and black phytoclasts, wood (brown and black), non-woody plant tissue, cuticles, amorphous organic matter (AOM), and palynomorph taxa. The palynomorph taxa/palynodebris ratio was noted, then counting of palynomorphs (or in some cases the palynodebris) continued until reaching 500 specimens. The remaining strew slides were examined and additional specimens, not included in the count, were recorded. Specimens illustrated are identified with LO + number, and will be housed at the Department of Geology, Lund University. 3.2. Palynostratigraphy Australian Permian and Triassic strata have been the focus of intensive palynological investigations for many years. Hence, the well-established Australian palynozonation has become the “de facto” standard to which many other Gondwanan assemblages are compared (Fig. 3). However, there were regional differences in palynofloral composition within Gondwana during the Middle and Late Permian. For example, the fern spore genus Dulhuntyispora contains many key-species for the Middle to Late Permian, but outside Australia and Timor (Basil Balme, pers. comm. 2005) only D. granulata has been recognised in situ and it is only represented by a few specimens in South Africa (Anderson, 1977; Backhouse, 1991) and the PCMs (Lindström, unpublished data). Despite this provincialism, palynofloral investigations of the entire Amery Group have now provided a more fully resolved biostratigraphic framework for correlation of the PCM sedimentary succession (Fig. 3). The PCM Permian succession is dominated by longranging taeniate bisaccate pollen, mainly Protohaplox- 97 ypinus and Striatopodocarpidites, and non-taenitae bisaccates assigned to Scheuringipollenites. Fern spores are generally scarce and this renders correlation with the Australian palynozonation difficult since many of those zones are based on the first appearance datum (FAD) of specific fern taxa. Didecitriletes ericianus is one of the Australian index species (Backhouse, 1991; Price, 1997) that also appears to have a synchronous inception in Antarctica (Lindström, 1995a), Africa (Anderson, 1977) and India (Tiwari and Tripathi, 1992). The first appearance of this fern spore defines the lower boundary of the D. ericianus Zone of Western Australia (Backhouse, 1991), and APP4.2 Zone of Price (1997). In the Bainmedart Coal Measures, D. ericianus has its FAD in the Toploje Member (Fig. 3). Another fern spore, Camptotriletes warchianus (Plate I, d), first appears in the Dragons Teeth Member. This species was originally described by Balme (1970) from the Salt Range, West Pakistan, where it is a rare component of the Amb, Wargal and Chhidru Formations. In Australia, this taxon has its FAD in the D. parvithola Zone of Western Australia (Mory and Backhouse, 1997), and in the Upper Stage 5 of Eastern Australia (Foster, 1982) or APP5 of Price (1997). Other taxa important in the Australian zonation include Triplexisporites playfordii and Playfordiaspora cancellosa. In eastern Australia the FAD of these taxa marks the lower boundary of the P. cancellosa Zone (Foster, 1982) or APP6 (Price, 1997), which can be correlated with the independently dated Early Changhsingian uppermost Chhidru Formation (“White Sandstone” unit) in Pakistan (Foster et al., 1997). The P. cancellosa Zone is succeeded by the Protohaploxypinus microcorpus Zone, but according to Price (1997) the transition between these zones differs from area to area and they lack a clearly defined boundary. They are, therefore, considered subunits of the APP6 Zone (Price, 1997). Neither the P. cancellosa nor P. microcorpus zones are recognized in the Prince Charles Mountains sequence. In this succession, both T. playfordii and P. cancellosa have their FADs in the lowermost sample of the Ritchie Member, which correlates with the Australian Lunatisporites pellucidus Zone (Fig. 3). The palynostratigraphic definition of the Permian– Triassic boundary in Gondwana has long been debated. The first appearance of pleuromeiacean monolete cavate spores assigned to Aratrisporites is favoured by some authors as a key-taxon for the P–T boundary (Foster et al., 1998), and in Australia Aratrisporites first appears in the Protohaploxypinus samoilovichii Zone (Foster et al., 1998). Because the pleuromeiacean parent plants of Aratrisporites may have been strongly facies 98 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 dependent (Retallack, 1975, 1977), some authors instead consider the first appearance of gymnospermous Lunatisporites pellucidus pollen to mark the P–T boundary (Price, 1997) and this is also favoured herein. In the Prince Charles Mountains a few small, inconspicuous, generally non-spinose Aratrisporites-type spores occur in the Ritchie Member samples from Ritchie Point, but as noted by Price (1997) such specimens may instead be small aberrant Lundbladispora spores. In the Prince Charles Mountains a single specimen of L. pellucidus is registered in the uppermost coal sample 95/04, but it is consistently present in the succeeding Ritchie Member samples (Fig. 6). The association of taxa, particularly L. pellucidus (Plate II, m), Falcisporites spp., Densoisporites nejburghii (Plate II, c), D. playfordii (Plate II, e), Lundbladispora spp. (e.g. Plate II, h), Playfordiaspora cancellosa (Plate II, o), P. samoilovichii (Plate II, l), P. microcorpus (Plate II, i), and Triplexisporites playfordii (Plate II, b), found in the Ritchie Member samples at Ritchie Point allows correlation with the L. pellucidus Zone or APT1 of Price (1997). The eastern Australian L. pellucidus Zone can be subdivided based on the FAD of the fern spore Rugulatisporites trisinus (Price, 1997). In the Prince Charles Mountains spores assigned to R. trisinus (see remarks under Appendix 1) are rare constituents in samples 95/02 to 05 (Fig. 6). True members of Aratrisporites have only been recovered in samples from the Ritchie Member on Flagstone Bench (Lindström, unpublished data). Those were originally considered to be roughly correlative to the upper two samples (95/07 and 95/05) from Ritchie Point (McLoughlin et al., Fig. 4. Relative abundance of palynodebris in the investigated samples from Ritchie Point in the Prince Charles Mountains. S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 1997), but are now regarded to be somewhat younger and equivalent to the P. samoilovichii Zone (Lindström, unpublished data). The boundary between the L. pellucidus and P. samoilovichii zones is independantly dated as late early Griesbachian (Foster and Archbold, 2001). 3.3. Sedimentation rate Calculating the average sedimentation rate for the different units of the Amery Group is difficult. Initial sedimentation in the Lambert Graben was probably first generated and controlled by faulting and mass-wasting of the basin flanks, as is indicated by the alluvial fan deposits of the Radok Conglomerate. The later clastic sedimentation of the Bainmedart Coal Measures was principally governed by axial drainage systems, influenced by rainfall, linked to orbital climatic forcing, as suggested by Fielding and Webb (1996), and by Michaelsen and Henderson (2000) for the Late Permian coal measures of the Bowen Basin, eastern Australia. The coal seams within the Bainmedart Coal Measures are laterally extensive and the alternation of peat-mire systems and broad sandy fluvial tracts indicates strong cyclic variation in the supply of clastic material. The McKinnon Member coals are of sub-bituminous rank suggesting a compaction ratio of around 6:1 from the original peat beds. The strongly differential compaction between peat and sandy beds and the variation in coal seam abundance and thickness means that only a broad estimate of sedimentation rates can be provided for this succession. The only useful age constraints available are palynomorph taxa. In the Prince Charles Mountains, the fern spore Didecitriletes ericianus is first recorded in the Toploje Member, 1874 m below the top of the uppermost coalseam, i.e. the interpreted Permian–Triassic boundary at 251 Ma. Using the known FAD of D. ericianus in the Late Roadian, ca 269 Ma, indicates an average depositional rate for the Bainmedart Coal Measures of 104 m/Ma. In comparison, the average sedimentation rate of the Upper Permian Blackwater Group in the northern Bowen Basin is 130 m/Ma in the depocentre, and 70 m/Ma in the more marginal parts of the basin (Michaelsen et al., 2001; Michaelsen, 2002). An average sedimentation rate of 104 m/Ma for the Bainmedart Coal Measures suggests that the entire McKinnon Member is late Wuchiapingian to Changhsingian in age. Calculating the sedimentation rate for the Triassic part of the Amery Group is much more difficult. The estimated minimum thickness of the Triassic Flagstone Bench Formation is 760 m (McLoughlin and Drinnan, 99 1997b). The N 72 m thick McKelvey Member is palynologically dated as Norian (Foster et al., 1994), leaving a minimum of 688 m for the pre-Norian Triassic, and with the Carnian/Norian boundary at 216.5 Ma this equals a sedimentation rate of ca 20 m/ Ma. This very slow sedimentation rate implies a 1.2 Ma duration for the 24 m sampling gap at the P–T transition. However, this is considered an under estimate of the sedimentation rate as part of the Flagstone Bench succession is concealed by ice, and because the silty red beds of the Jetty Member probably represent a long interval of very slow and episodic deposition in semiarid environments. There are no definite age constraints for the continuous section at Ritchie Point, and the sedimentation in the Ritchie Member is thought to have been subjected to strongly fluctuating clastic discharge (McLoughlin and Drinnan, 1997b). If the sedimentation rate for the Ritchie Member is equal to that of the preceding McKinnon Member, the 24 m sampling gap between the last coal sample (95/04) and the first Ritchie Member sample (95/01) would represent ca 230 000 years. However, the upper boundary of the Lunatisporites pellucidus Zone is independently dated as mid-Induan (ca 250.3 Ma) in Australia (upper lower Griesbachian of Foster and Archbold, 2001). The absence of genuine Aratrisporites from the Ritchie Member samples at Ritchie Point suggests that the entire 183 m sampled section can be assigned to the L. pellucidus Zone. In that case, the maximum sedimentation rate for that part of the section is 261 m/Ma, indicating that the 24 m sampling gap represents only ca 92 000 years. 4. Palynofloral turnover 4.1. The Late Permian stable ecosystem Palynoassemblages of the McKinnon Member indicate that the Late Permian vegetation was quite stable in this area, and not fundamentally different from that represented in the preceding members of the Bainmedart Coal Measures (Lindström, personal observations). This palynoflora is dominated by gymnospermous pollen (Fig. 3), primarily the taeniate glossopterid bisaccate pollen Protohaploxypinus and Striatopodocarpidites, and by the non-taeniate bisaccate pollen Scheuringipollenites (Fig. 5). Alisporites species, mainly A. splendens and A. tenuicorpus, are also present in some samples. Monosaccate pollen are rare, the most common forms being Densipollenites. Non-saccate monosulcate and polyplicate pollen assigned to Marsupipollenites striatus, M. triradiatus (Plate I, g), Praecolpatites 100 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 Fig. 5. a) Relative abundance of selected palynotaxa representing major plant groups in the investigated samples from Ritchie Point, Prince Charles Mountains. b) Continued from (a). The following palynofloral events are recognised in the section: (A) Level with a slightly elevated extinction rate 11% (4 samples from a 10 m thick stratigraphical interval, diversity = 75), proliferation of ferns and sphenophytes, common algae. (B) First extinction phase (3 samples from a 2 m thick stratigraphical interval) where 19% of the registered spore/pollen taxa disappear (Diversity = 69). Similar proliferation of ferns and sphenophytes as at level A. After this level the glossopterids appear to decline in diversity. (C) Second extinction phase (last coal sample) last appearance of 33% of the spore/pollen taxa registered in the sample. Demise of glossopterid dominated swamp forests (Diversity = 67). (D) Third extinction phase with loss of 14% of the registered spore/pollen taxa, and contemporaneous first major occurrence of pioneering taxa. Proliferation of peltasperms, corystosperms, ferns and lycophytes (diversity = 77). (E) Fourth extinction phase (2 samples from an 11 m thick stratigraphic interval) where 35% of the spore/pollen taxa are registered for the last time, and contemporaneous second major occurrence of pioneering taxa (diversity = 99). Continued proliferation of peltasperms, corystosperms and lycophytes, and also proliferation of probable bryophyte spores. (F) Increase in corystosperms and continued proliferation of lycophytes and probable bryophytes (diversity = 71). S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 101 Fig. 5 (continued ). sinuosus (Plate I, n) and Ephedripites, are also generally common. Brown wood is the dominant debris, whereas cuticle fragments are generally very rare (Fig. 4). A generally low spore/pollen ratio indicates that ferns, sphenophytes, bryophytes and herbaceous lycopsids played a subordinate role in the vegetation. Intervals with increased spore/pollen ratios are associated with the large coalseams (Fig. 4), where trilete fern spores Osmundacidites (and morphologically similar taxa), Horriditriletes and Lophotriletes increase in abundance (Fig. 5). Monolete Laevigatosporites spores also increase in abundance (Fig. 5), but these may represent sphenophytes (Balme, 1995). Several taxa, including Didecitriletes uncinatus, Granulatisporites absonus, Horriditriletes ramosus, Protohaploxypinus bharadwajii and P. pennatulus, 102 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 have their LADs within a 2 m interval a little less than 50 m below the top of the uppermost coal. Protohaploxypinus samoilovichii, Klausipollenites sp. A, Lueckisporites virkkiae, Indospora clara (Plate I, c), Lunatisporites obex and Chordasporites australiensis have their successive first appearances within the McKinnon Member (Fig. 6). Throughout this unit the taxonomic turnover rate is low, except for the sample from the uppermost coalseam in which 22 taxa, i.e. 33%, have their LADs. 4.2. The end-Permian crisis and the taxa that perished The most striking palynofloral change at the end of the Permian is a general decline in gymnospermous pollen (Fig. 5), and especially the dramatic decrease of glossopterid taeniate bisaccate pollen. Protohaploxypinus decreases from around 25% to b 1% relative abundance over the 24 m interval from the uppermost coal sample 95/04 to the lowest Ritche Member sample Fig. 6. a) Palynostratigraphic range chart for the Permian–Triassic transition at Ritchie Point in the Prince Charles Mountains. b) Continued from (a). c) Continued from (b): algal and acritach taxa. S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 103 } Fig. 6 (continued ). 95/01, and Striatopodocarpidites shows a similar pattern changing from ca 6% to b 1% (Fig. taxa groups). Glossopterid pollen do persist in the Ritchie Member, but always comprise b 1% of the assemblages. The same pattern in abundance is shown by the non-taeniate bisaccate pollen Scheuringipollenites. The parent plant of this commonly abundant and typically Permian bisaccate genus is unknown, but it may also be allied to the glossopterids. Many typical Permian Gondwanan taxa that are consistently present in the McKinnon Member have their LADs in the uppermost coal sample, e.g. the majority of species assigned to Protohaploxypinus and Striatopodocarpidites, together with Striatoabieites multistriatus, Densipollenites spp., Praecolpatites sinuosus, Florinites eremus, Microbaculispora tentula (Plate I, b) and Didecitriletes ericianus (Plate I, e). The virtual disappearance of glossopterid pollen can be directly linked to the cessation of coal formation. The disappearance of the peat-forming mire that hosted the glossopterids is a conspicuous feature of many Permian–Triassic transitions in southeastern Gondwana. The glossopterids first appeared during the late Palaeozoic glaciation, steadily diversified in the ameliorating post-glacial temperate 104 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 Fig. 6 (continued ). climate during the Early and Middle Permian, and markedly proliferated in the humid Late Permian climate. The glossopterids were middle- to high-latitude deciduous trees with roots that were adapted to semiaquatic conditions (Neish et al., 1993). The presence of well-developed growth rings in glossopterid wood from the PCMs shows that these plants were also subjected to strong seasonal variations (McLoughlin et al., 1997; Weaver et al., 1997). The narrow latewood and abrupt termination of rings suggests that growth of the Antarctic glossopterids was primarily controlled by seasonal photoperiod variation. Despite the fact that the glossopterids were deciduous trees, glossopterid cuticles are a relatively rare component in the McKinnon Member palynodebris. The maceral cutinite also constitutes a relatively minor portion (generally b 1%) of coals sampled from the Bainmedart Coal Measures (McLoughlin, unpublished data). Glossopterid cuticle S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 is typically thin and large sheets are normally difficult to prepare from compressed leaves. Thin and readily degraded cuticles may be a function of the deciduous nature of glossopterids, the prevailing humid, temperate climate of the Late Permian, and the plants' affinities for mire habitats, hence their little need for cuticular protection from desiccation. Glossopterids were overwhelmingly the dominant plants in the PCMs during the latest Permian, and were extremely well adapted to their environment. It seems plausible that other taxa that disappeared, or markedly decreased in number, at the P–T transition were also adapted to the peat-forming mires in which the glossopterids were the major constituents. 4.3. Taxa that proliferated after the end-Permian event The Ritchie Member palynoflora is also dominated by gymnospermous pollen, although not to the same degree as the McKinnon Member. In the Ritchie Member, the gymnosperms are represented by peltaspermous or corystospermous bisaccate pollen, such as the taeniate Lunatisporites spp., Protohaploxypinus microcorpus and P. samoilovichii, and the non-taeniate Falcisporites spp. and Alisporites spp. In the lowermost Ritchie Member sample (95/01) fern spores assigned to Brevitriletes spp. and Leiotriletes directus (Plate I, a), together with Osmundacidites spp. and Dictyophyllidites spp., are common, however, all but the last of these decrease in abundance in the succeeding samples. Another striking feature of the Ritchie Member palynoflora is the high diversity and abundance of lycophyte spores, mainly Densoisporites, Lundbladispora, Kraeuselisporites and Uvaesporites species. Probable bryophyte spores are minor constituents of the lowermost Ritchie Member assemblage, but increase in abundance in the succeeding samples. The spore/pollen ratio increases dramatically in the Ritchie Member indicating that spore-producing plants played a proportionately greater role in the earliest Triassic plant community. This is further indicated by the dramatic decrease of woody plant debris between sample 95/04 of the uppermost coal in the McKinnon Member and the lowermost sample of the Ritchie Member, 95/01. Non-woody plant remains, including cuticles, are the most common type of palynodebris in the Ritchie Member. Lycophyte sporangia and megaspores are also notably abundant in mesofossil (N 200 m) residues from this unit (McLoughlin et al., 1997). From the palynological data it is evident that many of the gymnospermous taxa that increase in abundance in the Early Triassic assemblages were already present in 105 low numbers in the Permian. They appear to have played a subordinate role in the glossopterid-dominated plant community, perhaps occupying drier sites where their macrofossils were less likely to be preserved. These opportunists proliferated once the glossopterids and their ecological associates were fading from the scene. So how did the surviving gymnosperms differ from those that perished? One of the most abundant pollen species in the earliest Triassic of the PCMs is Falcisporites australis (Plate II, f ). This non-taeniate bisaccate pollen has been found in association with the peltasperm Lepidopteris callipteroides (Carpentier) Retallack (2002) (Retallack, 2002), and small pinnules of Lepidopteris sp. are prominent in the earliest Triassic sample from the PCMs (McLoughlin et al., 1997). Retallack (2002) argued that the thick cuticle, low stomatal index and small-sized stomata of Early Triassic L. callipteroides leaves from the Sydney Basin indicated high atmospheric concentrations of CO 2 . According to Retallack (2002) L. callipteroides migrated southwards from northern Gondwana in the Early Triassic, but centres of origin and migration pathways are impossible to determine from the fossil record (Patterson, 1999). Nevertheless, this theory is supported by the recovery of an Upper Permian Falcisporites-dominated palynoflora and associated Dicroidium fossils in the Dead Sea region in Jordan (Kerp et al., 2006). The Dicroidiumbearing Permian flora of Jordan indicate that corystosperms developed in the Late Permian in an extrabasinal tropical lowland setting (Kerp et al., 2006). In fact, many of the gymnosperms that played minor rolls in the Permian of Gondwana may have been adapted to better drained and perhaps more elevated areas than the glossopterid dominated peat mires that developed in the Gondwanan basins. The PCM Permian–Triassic succession was deposited within the Lambert Graben, which is suggested to have been fed by a drainage system one-fifth the area of East Antarctica (Adamson and Darragh, 1991). Non-glossopterid gymnosperms may have been significant components of the vegetation through much of this upland region. In the aftermath of the end-Permian crisis, pleuromeian/isoetalean lycophytes appear to have diversified globally (Pigg, 1992; Kovach and Batten, 1993). Although it is not obvious from the quantitative analysis, small percentages of lycophytes (Gondisporites raniganjensis, Plate I, k; and Indotriradites spp.) are present in the McKinnon Member (Fig. 6). However, lycophytes definitely proliferate in the Ritchie Member where, e.g. Densoisporites nejburgii accounts for more than 5% of the assemblage in 95/01. 106 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 4.4. Taxa that lingered Several taxa already present, and locally common in the Permian, appear not to have been greatly affected by the Permian–Triassic crisis. Amongst these are the fern spores Lophotriletes and Osmundacidites. These genera have a slight decline in relative abundance in the Early Triassic, but they also vary in representation throughout the Late Permian. Leiotriletes directus is another fern spore typically common in the Permian. In the McKinnon Member this small, laevigate triangular spore is consistently present, but never reaches greater than 1.6% of any assemblage. In the lowermost Ritchie Member sample it suddenly flourishes constituting 3.8% of the palynoflora, but decreases dramatically in the succeeding samples. Guttulapollenites hannonicus (Plate I, h; Fig. 8) is another taxon that is traditionally associated with the Permian. This gymnosperm pollen rarely reaches 1% in the McKinnon Member, but reaches 1.0 to 2.4% in the Ritchie Member. 4.5. Reworking The aspect of reworking is critical when analysing taxon ranges and mass-extinction events based on microfossil data. Re-deposition of older material can severely affect the interpretations of an ecological crisis and its aftermath. As reworked material is quite commonly encountered in palynological investigations, it is necessary to estimate the degree of reworking in the PCM Permian–Triassic transition. In the PCMs, a few glossopterid pollen and some other typical Permian taxa linger on into the Early Triassic but disappear within a few hundreds of metres above the uppermost coal, e.g. Microbaculispora micronodosa, M. trisina, and M. villosa. It is tempting to suggest that these typical Permian spores are entirely reworked. However, it should be noted that they have also been encountered in small numbers in other Gondwanan Early Triassic sequences (see e.g. de Jersey, 1979). The likelihood of having been reworked is determined by the palynomorph's state of preservation, colour, abundance, and association with other contemporary taxa. Microbaculispora specimens encountered in the Ritchie Member show no obvious signs of reworking, as they do not differ in colour or state of preservation from the rest of the palynoassemblage. Quantitatively, they are more abundant than the few taeniate glossopterid pollen also present in these assemblages. There is no known Permian assemblage from the PCMs in which Microbaculispora outnumbers glossopterid pollen (Lindström, personal observation). If the lower Ritchie Member Microbaculispora specimens are reworked, then one would expect to find proportional representation of other typical Permian taxa such as glossopterid pollen. Differential preservation of these groups was probably not important because the durability of glossopterid taeniate pollen is demonstated by the fact that they are amongst the most commonly reworked palynomorphs in younger (Late Mesozoic–Cenozoic) strata from Seymour Island, west Antarctica (Askin and Elliot, 1982). In the shelf sediments off-shore from Prydz Bay and the Prince Charles Mountains, Permian palynomorphs in general are the rarest reworked elements (Kemp, 1972). 4.6. Diversity pattern It is important to remember that fossil spore- and pollen taxa do not necessarily equate to true plant taxa (Lindström et al., 1997). Nevertheless, the diversity of spore-pollen taxa across the P–T transition in the PCMs reveals a surprising pattern. The number of taxa registered in each sample of the McKinnon Member varies between a minimum of 31 and a maximum of 69 (Fig. 6), whereas in the Ritchie Member it varies between 70 and 92. There is a jump in diversity from 66 taxa in the last McKinnon Member sample (95/04) to 77 taxa in the lowest Ritchie member sample (95/01), and then to a maximum of 92 taxa in the second Ritchie Member sample (95/02). Species diversity declines slightly to 71 and 70 taxa respectively in the succeeding samples (95/07 and 95/05). If expected occurrences are taken into account, i.e. the local stratigraphical ranges of the different taxa are used instead of “de facto” registrations in each sample, the diversity in the lower and middle parts of the McKinnon member is very constant, vaying between 81 and 85 taxa. In the uppermost part of the McKinnon Member there is a drop in diversity to 75 taxa in samples 95/10 and 95/08, with a slight increase to 79 taxa in the uppermost McKinnon Member sample (95/04). This is followed by a marked increase to 90 taxa in the lowermost Ritchie member sample, with a continuing increase to 101 taxa in sample 95/02, followed by a gradual decrease to 82 and 70 in the two succeeding samples. Ninety-eight typical Permian taxa are present from the lower part of the McKinnon Member, or have been registered in the preceeding members. After a stepwise extinction pattern only 25 of these taxa are present in the uppermost Ritchie Member sample, i.e. only 26% of the typical Permian taxa survived up to this level in the aftermath of the end-Permian crisis. S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 4.7. Extinction pattern The extinction pattern of spores and pollen across the P–T boundary in the PCMs appears stepwise with a major extinction between the uppermost coal sample of the McKinnon Member and the lowermost Ritchie Member sample (Fig. 6). Through most of the McKinnon Member succession, few taxa have their last occurrences (Fig. 6). Exceptions to this pattern occur in the closely spaced samples 95/26B–95/29 and 95/11–95/8. At the lower level, a 10 m interval from 1963 to 1973 m, the combined LADs reach almost 11% and the combined FADs are almost 12%, both slightly higher than “normal”. At the higher level, a 5 m interval from 2225 to 2230 m, the combined LADs reach 19% and the combined FADs are only about 1%. However, few of the disappearing taxa at these levels occur regularly or in large numbers within the McKinnon Member, and their last occurrences at a certain level may be an artifact of incompleteness of the fossil record (Signor and Lipps, 1982). Here, we use the range extension method of Marshall (1995) to test the confidence levels of the palynostratigraphic ranges across the P–T transition in the PCMs. In accordance with improvements by Wang and Marshall (2004), 20% range extensions are used, and based on a minimum of four occurrence levels throughout the McKinnon and Ritchie members only about half of the taxa with LADs within this interval can be used. The results alter the stepwise extinction pattern very little (Fig. 6a, b). None of the taxa that has its LADs in the interval covering samples 95/11, 10 and 08 extends its range up to or above the topmost McKinnon Member sample (Fig. 6b). However, for this level the range extensions do show a more gradual extinction pattern stretching over a 25 m interval, than that of the true records. This 95/11 to 95/08 interval is taken to represent a first extinction phase, signalling the initiation of the collapse of the terrestrial ecosystem (Figs. 5a, b and 6a, b). Similarly, none of the taxa that has its LAD in the uppermost coal sample has its range extended up to or beyond the lowermost Ritchie Member sample. The range extensions indicate that the taxa with LADs in the uppermost McKinnon Member sample all disappear within an 8 m interval starting 9 m above the last coal (i.e. between 2283 and 2291 m). This is taken to represent the second and main phase of the end-Permian extinction in the PCMs (Fig. 6a, b). The third extinction phase is represented by the LADs in the lowermost Ritchie Member sample (Fig. 6a, b), the range extensions indicating that the taxa disappear within a 7 m interval starting 8 m above the sample level. All but one of the taxa that have 107 their LADs in sample 95/02 of the Ritchie Member have their ranges extended up to or beyond the next sample level. All the taxa with LADs at this level have their ranges extended within a 21 m interval starting 9 m above sample 95/02. In the next sample, 95/07, none of the taxa has its ranges extended up to or above the next sample level. In fact, all the taxa with LADs at 95/07, except one, have range extensions falling within the upper 8 m of the extension range interval from the previous sample. The LADs of samples 95/02 and 07 are taken to represent the fourth extinction phase in the PCMs (Fig. 6a, b). A 20% confidence level used on the FADs of Early Triassic taxa, based on a minimum of four occurrence levels, comprises all the taxa first appearing in the lowermost Ritchie Member sample, and gives a range extension downwards to 2286 m. The mean range extension level for the LADs in the uppermost coal is also 2286 m, i.e. the estimated position of the main endPermian extinction. 5. Patterns of change across the P–T boundary in Gondwana 5.1. Sedimentological changes across Gondwanan P–T transitions The most apparent lithological and sedimentological change across the P–T transition in the PCMs is the cessation of coal (Figs. 2 and 3), and this is also a wellknown feature of P–T continental boundary strata in several other southeastern Gondwana basins, notably the Bowen and Sydney basins, Australia and the Transantarctic Mountains, central Antarctica (Retallack et al., 1996; Taylor et al., 1989). In India, coals are widely represented in Upper Permian strata but disappear slightly below the P–T boundary. Upper Permian coals are also represented in the eastern Karoo Basin, South Africa, but they were replaced by redbeds well before the P–T transition (Ward et al., 2000). The P–T boundary has been less well studied in South America but uppermost Permian strata of the Parana Basin are dominated by red or varicoloured mudstones with scarce Glossopteris leaves (Rohn and Rösler, 1989). Apart from the disappearance of coal, McLoughlin and Drinnan (1997b) noted that the relative proportion of sandstone to siltstone increases through the upper part of the McKinnon Member and lower part of the Ritchie Member in the PCMs. There is a slight change in the direction of sediment transport across the P–T transition; from predominantly N to NE in the McKinnon Member to mainly NW to NNE in the Ritchie Member 108 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 (McLoughlin and Drinnan, 1997a,b). Sediment cyclicity (fining-up sandstone–siltstone–coal packages) is well developed in the lower members of the Bainmedart Coal Measures (cycles average ca. 9–10 m; Fielding and Webb, 1996), but is less well developed and less regular in the McKinnon Member (McLoughlin and Drinnan, 1997a). Sediment cyclicity persists in the Ritchie Member, but the cycles tend to be thinner (average ca. 5.6 m) compared to the preceding Bainmedart Coal Measures (McLoughlin and Drinnan, 1997b; Holdgate et al., 2005). The changes across the P–T transition in the PCMs are interpreted as a shift from large, low-sinuosity braided and minor meandering rivers and poorly drained forest-mire environments, to medium-sized, low-sinuosity braided rivers with progressively more episodic discharge (McLoughlin and Drinnan, 1997a,b). Similar changes in fluvial style across the Permian– Triassic transition have been reported from other Gondwana basins. In the Raniganj Basin, India, the P–T transition of the Banspetali section shows a drastic sedimentological change (Sarkar et al., 2003). The Upper Permian Raniganj Formation consists of alternating fine- to medium-grained white/grey, plagioclaserich sandstones, dark organic-rich shales and coal. The lithologies of the succeeding Lower Triassic Panchet Formation include sandstones rich in unaltered orthoclase, and grey to olive-green shales. The upper part of the Panchet Formation includes reddish strata, and they are succeeded by the highly immature, poorly sorted, red sandstones and conglomerates of the Supra-Panchet Formation (Sarkar et al., 2003). Tewari (1999) reported a similar change in fluvial style from Late Permian meandering river deposits to Early Triassic braided fluvial sediments in the Godavari Basin, India. However, there are local discrepancies. In the GAM-7 borehole in the Godavari Basin, the cessation of coals and carbonaceous shale occurs more than 100 m below the palynologically defined P–T boundary, and is succeeded by greenish gray sandstone and shale (Srivastava and Jha, 1990). Latest Permian sequences in the Karoo Basin, South Africa lack coal. Instead, the major sedimentological Fig. 7. Composite correlation chart of selected P–T transitions across Gondwana. S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 change across the P–T transition is a rapid, basin-wide change in fluvial style from meandering to braided river systems (Ward et al., 2000). Below the South African P– T boundary, sandstones deposited by large, high sinuosity meandering rivers are interbedded with olive grey and red mudstones (Ward et al., 2000; Steiner et al., 2003). The P–T boundary is associated with a several metres thick laminated sandstone-shale unit, under- and overlain by sandstone and conglomerate (Ward et al., 2000). Above the P–T transition, the Karoo Basin sequences are characterized by sediments deposited by braided river systems. The proportion of sandstone to shale is higher than in the Permian, and the silts and mudstones are predominantly maroon in colour instead of olive gray (Ward et al., 2000). The drastic change was interpreted as indicating increased sedimentation rates in the earliest Triassic, due to catastrophic die-back of the terrestrial vegetation that would normally prevent erosion of river banks and hill slopes (Ward et al., 2000). A similar scenario is reported from the northern Bowen Basin in eastern Australia where the Permian– Triassic boundary coincides with the lithostratigraphic boundary between the coal-rich Rangal Coal Measures of the Blackwater Group and the coal-lacking Sagittarius Sandstone of the Rewan Group (Michaelsen, 2002). Although palynological data indicates a gradational floristic change prior to the boundary, Michaelsen (2002) found no evidence of any lithological changes within the Rangal Coal Measures up to the boundary. Instead, a sharp change in the sedimentary regime is evidenced by Late Permian peat mire, sinuous braided river channels, extensive crevasse splay, and small lake deposits abruptly succeeded by high energy, braided river sediments in the Early Triassic (Michaelsen, 2002). The Permian–Triassic transition in the Transantarctic Mountains is also marked by the cessation of coal at or a few metres below the boundary, together with generally thicker packages of trough cross-bedded sandstones in the Triassic, and major changes in palaeosol composition (Retallack et al., 2005). 5.2. Palynofloral turnover in other Gondwanan P–T transitions Although there have been many studies of the Gondwanan P–T palynofloral transition, they typically report only general paterns of taxon turnover and include very little or no quantitative palynological data. The LADs and FADs of selected P–T transition taxa in key Gondwanan basins are briefly reviewed here (Fig. 7). Spores and pollen are very poorly preserved in the P–T transitional strata of the Karoo Basin, South Africa 109 (Anderson, 1977). However, Steiner et al. (2003) recently reported a 100% taxonomic turnover in the Permian–Triassic Carlton Heights boundary section in the Karoo Basin. The 1 m thick interval above the boundary is 100% dominated by the putative fungal palynomorph Reduviasporonites chalastus and woody plant remains. The assemblage was interpreted to represent proliferation of fungi upon large quantities of decaying plant material (Steiner et al., 2003; Fig. 7). This “fungal” spike separates 55.5 m of strata assigned to the latest Permian Klausipollenites schaubergerii zone (equivalent to the Australian APP6 or Protohaploxypinus microcorpus Zone), from a b0.5 m thick interval assigned to the Early Triassic Kraeuselisporites–Lunatisporites zone (equivalent of the L. pellucidus Zone); these strata succeeded by at least 12 m of palynologically barren sandstones (Steiner et al., 2003, Fig. 5). None of the constituents of the Permian palynoflora in this succession, including Densoisporites playfordii, Triplexisporites playfordii and Playfordiaspora cancellosa was registered above the “fungal” spike, where instead typical Early Triassic taxa e.g. L. pellucidus, Kraeuselisporites cuspidus, and Lundbladispora brevicula were found (Steiner et al., 2003). Despite the apparent high resolution detection of the P–T section at Carlton Heights, some uncertainties remain regarding the palynological signature in the Karoo Basin. Palynomorph yields were apparently low and several key taxa have patchy records in the range charts provided by Steiner et al. (2003). The thick barren intervals may also preclude identification of the full ranges of key taxa. Furthermore, the “fungal” spike at the Carlton Heights section is located 17 m above the P–T boundary beds as defined by Retallack et al. (2003), and according to Ward et al. (2005) well above the top of the Permian based on carbon isotope data. However, it should be noted that there are several typical Permian taxa that appear to have their last occurrences between 19 and 24 m below the “fungal spike” zone (Steiner et al., 2003; Fig. 5), possibly corresponding to an initial extinction level. The palynofloristic turnover at the P–T transition within the Maji Ya Chumvi Formation, Kenya, is characterised by the disappearance of 27% of latest Permian taxa. It is accompanied by a decrease in the number of acavate trilete spores, a slight decrease in the abundance of cavate trilete spores, and a major increase in the abundance of taeniate bisaccate pollen (Hankel, 1992). The latest Permian assemblage is equivalent to the Australian APP6 or Protohaploxypinus microcorpus Zone, and contains 38% cavate and 32% acavate spores, 18% taeniate bisaccate pollen, 7% monosulcate pollen 110 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 and only 5% non-taeniate bisaccate pollen (Hankel, 1992). It is separated from an earliest Triassic assemblage by a 24.2 m palynologically barren interval spanning the P–T boundary (Hankel, 1992). The earliest Triassic palynoflora is represented in 18 samples over an 8.2 m thick interval and is dominated by taeniate bisaccate pollen (37–56%). Acavate spores are less frequent, 8–32%, and cavate spores vary between 12% and 37%. It was correlated with the Lunatisporites pellucidus Zone of Australia (Hankel, 1992). The putative fungal palynomorph Reduviasporonites chalastus constitutes 24% of the latest Permian assemblage and is also present but less common (4–10%) in the earliest Triassic. Lunatisporites pellucidus first occurs in the earliest Triassic, constituting 5–9% of the microflora. Among the 22 Permian taxa that disappeared at the P–T transition were the important indices P. microcorpus, Lundbladispora willmottii, P. crenulata (Wilson) Foster (1979) and T. playfordii (Hankel, 1992). Of 34 taxa identified in the younger assemblage, 53% have their FADs in the earliest Triassic, e.g. D. playfordii and Kraeuselisporites cuspidus. However, at least two of the taxa that last occurred in the Permian (i.e. P. crenulata = P. cancellosa and T. playfordii) are present in an even younger assemblage from the Lower Mariakani Formation (correlated with the Australian P. samoilovichii Zone; Hankel, 1991), and could be considered “Lazarus taxa” (Fig. 7). According to Wright and Askin (1987) the boundary between the Lower and Middle Sakamena Group in Madagascar approximates the Permian– Triassic boundary. The latest Permian palynofloral assemblages are dominated by Guttulapollenites hannonicus, Weylandites spp. and Lueckisporites virkkiae. Glossopterid taeniate bisaccates assigned to Protohaploxypinus and Striatopodocarpidites are generally common. The assemblages include rare specimens of Protohaploxypinus microcorpus (Wright and Askin, 1987). Non-taeniate bisaccates are also common and include Platysaccus spp., Alisporites spp., Falcisporites spp., Scheuringipollenites spp., and Klausipollenites schaubergerii (Wright and Askin, 1987). Pteridophyte spores are generally rare. The first occurrence of rare Lunatisporites pellucidus is registered in the uppermost Lower Sakamena outcrop sample (Wright and Askin, 1987). Thirty-five percent of the taxa in this latest Permian assemblage disappear at the P–T transition. In contrast, the Early Triassic assemblages are characterised by common to abundant L. pellucidus. Striatopodocarpidites pantii and P. microcorpus increase in abundance. Some typical Permian taxa persist into the lowermost Middle Sakamena, e.g. Densipollenites indicus, Protohaploxypinus limpidus, Striatopodocarpidites rarus and G. hannonicus (Wright and Askin, 1987). In the earliest Triassic assemblage 42% of the taxa appear for the first time, including Densoisporites playfordii (Fig. 7). A cored section, GAM-7, of the lower and middle members of the Kamthi Formation in the Godavari Graben, India, preserves a palynological succession across the Permian–Triassic boundary, listed as assemblages I to V in ascending order (Srivastava and Jha, 1990). The Late Permian assemblages (I–IV) are dominated by glossopterid bisaccate pollen, constituting around 70–50% relative abundance. Other common and important taxa are Scheuringipollenites, Densipollenites and Osmundacidites, whereas Horriditriletes, Lophotriletes and Weylandites are locally common. Both Densipollenites and Scheuringipollenites have last appearances in assemblage III. The uppermost Permian coal seam and carbonaceous shales were recorded within the interval associated with assemblage II. Guttulapollenites is a rare component in the Permian assemblages, except in IV where it increases to 29% (Srivastava and Jha, 1990). Srivastava and Jha (1990) correlated assemblage IV with the palynoflora from the Chhidru Formation of the Salt Range (Balme, 1970) and the Australian Protohaploxypinus microcorpus Zone (latest Permian), on the basis of the common presence of i.e. Triquitrites proratus, Playfordiaspora cancellosa, L. noviaulensis, Falcisporites stabilis and P. microcorpus. Assemblage V, from a sample a little more than 12 m above that of assemblage IV (Srivastava and Jha, 1990), is correlated with the Lunatisporites pellucidus Zone and is dominated by Lunatisporites (32%) including L. pellucidus, and fern spores assigned to Verrucosisporites (10%). Other common elements are Lundbladispora, P. cancellosa, Limatulasporites, K. schaubergerii and Alisporites. Glossopterid pollen are apparently still present, but markedly decreased (Fig. 7). Several papers have dealt with the palynology of the P–T transition in the Bowen Basin in eastern Australia (e.g. de Jersey, 1979; Foster, 1982), and these were reviewed by Price (1997). In the GSQ Taroom8 borecore of Denison Trough in the western Bowen Basin, the first appearance of Lunatisporites pellucidus is registered about 50 m above the uppermost coal of the Bandanna Formation, in the lower part of the Rewan Group (de Jersey, 1979). The earliest Triassic assemblage is separated from the preceeding productive sample by a ca 42 m barren interval (de Jersey, 1979). Triplexisporites playfordii first appears within a few metres above the uppermost coal in the Bandanna Formation, thus placing the P–T boundary somewhere S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 111 Fig. 8. Palaeobiogeographic distributions of Guttulapollenites hannonicus, Triplexisporites playfordii and Playfordiaspora cancellosa. Palaeogeographic maps after Scotese (2001). Star denotes presence of taxon. Localities and occurrences from: 1. Oklahoma (Wilson, 1962); 2. Israel (Eshet, 1990); 3. Argentina (Ottone and Garcia, 1991; Zavattieri and Batten, 1996); 4. Pakistan (Balme, 1970); 5. Kenya (Hankel, 1991, 1992); 6. Tanzania (Hankel, 1987); 7. Zimbabwe (Falcon, 1973); 8. South Africa (Anderson, 1977; Steiner et al., 2003); 9. Madagascar (Hankel, 1993; Wright and Askin, 1987); 10. India, Godavari Graben (Srivastava and Jha, 1990); 11. India, Damodar and Rajmahal basins (Tiwari and Tripathi, 1992); 12. Western Australia, Perth and Collie basins (Backhouse, 1991, 1993); 13. Western Australia, Bonaparte Basin (Helby, 1977, unpublished report); 14. East Australia, Bowen Basin (de Jersey, 1979; Foster, 1979); 15. East Australia, Sydney Basin (Helby, 1973); 16. New Zealand (de Jersey and Raine, 1990); 17. Antarctica, South Victoria Land (Kyle, 1977; Kyle and Schopf, 1982); 18. Antarctica, Dronning Maud Land (Lindström, 1996); 19. Antarctica, Prince Charles Mountain (This paper; McLoughlin et al., 1997); 20. Zambia (Utting, 1979; Nyambe and Utting, 1997). P. cancellosa has also been registered in the Middle Permian of Spain (Broutin, 1986). in between the FADs of these two taxa (de Jersey, 1979). Only 4% of the latest Permian taxa disappear by the uppermost Permian sample. However, 25% of taxa disappear by the earliest Triassic assemblage and 19% of taxa are registered for the first time. In nearby GSQ Springsure-1 bore (Fig. 6), L. pellucidus has its FAD some 12 m above the last coal of the Bandanna Formation, just below the boundary between the Bandanna Formation and the Rewan Group (de Jersey, 1979). This assemblage is separated from the previous productive sample by a barren interval of ca 11 m, and T. playfordii first appears within the uppermost part of 112 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 the youngest coalseam. In this well, only 4% of the Permian taxa are present for the last time in the uppermost productive sample of the Bandanna Formation. In the earliest Triassic assemblage 19% of the taxa appear for the first time. On the basis of previous studies, Michaelsen (2002) argued that palynofloral change across the P–T boundary in the Bowen Basin is gradual (Fig. 7) but, as noted above, the significant sampling gap may overlook any dramatic turnover. High-resolution palynological sampling across the P–T boundary in the high-palaeolatitude Transantarctic Mountains has not yet been undertaken. However, McManus et al. (2002) recently reported sparse glossopterid leaves several metres above the traditional placement of the P–T boundary at the Buckley– Fremouw formation transition. Scattered reports of sparse earliest Triassic Glossopteris leaves elsewhere in Gondwana (Pant and Pant, 1987) are consistent with the PCM palynological evidence that a few elements of the Glossopteris flora persisted into the very earliest Triassic, presumably in isolated humid refugia. 5.3. Palaeobiogeographic patterns Several taxa show interesting palaeogeographic and biostratigraphic distributions in the Late Permian and Early Triassic (Fig. 8). Guttulapollenites hannonicus (Fig. 8) has been registered in Middle to Late Permian sequences from Africa (Anderson, 1977; Falcon, 1973; Utting, 1979), Madagascar (Wright and Askin, 1987), Pakistan (Balme, 1970), India (Srivastava and Jha, 1990; Tiwari and Tripathi, 1992; Tiwari, 1999), Antarctica (Balme and Playford, 1967; Lindström, 1995a, 1996; McLoughlin et al., 1997), and Australia (Backhouse, 1993). In Madagascar (Wright and Askin, 1987) and India (Srivastava and Jha, 1990; Tiwari, 1999) G. hannonicus is particularly abundant in the latest Permian. In the PCMs it is a rare but consistent component in the Late Permian, but it is more common in the Early Triassic, especially in the lowermost Ritchie Member sample. There are only two other areas where G. hannonicus has been recorded in the Early Triassic, namely in the Salt Range of West Pakistan (Balme, 1970) and in Madagascar (Wright and Askin, 1987), and together with the PCMs these areas appear to have acted as the last refuges for the parent plant of G. hannonicus. In many parts of Gondwana Triplexisporites playfordii and Playfordiaspora cancellosa occur together (Fig. 8). Both have their FADs in APP6 (latest Permian) microfloras of Australia (Price, 1997), in the early Changhsingian uppermost Chhidru Formation in Pakistan (Balme, 1970; Foster et al., 1998), and in equivalent microfloras in Kenya (Hankel, 1992) and South Africa (Steiner et al., 2003). In India, P. cancellosa is also known from the upper Raniganj Formation (Srivastava and Jha, 1990; Tiwari and Tripathi, 1992), but T. playfordii first occurs in the Early Triassic (Tiwari and Tripathi, 1992). In the PCMs (this paper) and Madagascar (Wright and Askin, 1987) T. playfordii and P. cancellosa are not registered in the Late Permian, but first appear in the Early Triassic. In South Africa there are no records of T. playfordii or P. cancellosa in the Early Triassic (Steiner et al., 2003). In Kenya T. playfordii and P. cancellosa are not recorded in the earliest Triassic assemblages, but reappear in a younger assemblage (Hankel, 1991, 1992). Triplexisporites playfordii shows a similar pattern in Pakistan, whereas P. cancellosa is registered there also in the Early Triassic (Balme, 1970). Playfordiaspora cancellosa has also been registered in a late Early Triassic assemblage from the mid-Zambesi Valley in southern Zambia (Nyambe and Utting, 1997), and in the late Middle Triassic of Tanzania (Hankel, 1987), and Argentina (Zavattieri and Batten, 1996). In New Zealand T. playfordii and P. cancellosa are first registered in the late Early Triassic (de Jersey and Raine, 1990), and in South Victoria Land both taxa are first registered in the Middle Triassic (Kyle, 1977). In the Middle to Late Permian both these taxa have only been registered outside Gondwana (Fig. 8). Playfordiaspora cancellosa (or closely related forms) are present at locations within ± 20° of the equator in, e.g. Oklahoma (Wilson, 1962), Southern Europe (Broutin, 1986) and Israel (Eshet, 1990), and Triplexisporites playfordii in Israel alone situated 20°S (Eshet, 1990). The distribution patterns suggest that during the latest Permian both taxa migrated southwards, spreading along a narrow belt between 65° and 45°S ranging from South Africa in the west to eastern Australia (Fig. 8). In the earliest Triassic they both appear to retreat eastwards with T. playfordii still retaining a very narrow field of distribution between 45° and 60°S, whereas P. cancellosa exhibits a wider distibution pattern from 35–60°S. In the Middle Triassic the distribution patterns for these taxa expand significantly. Triplexisporites playfordii then occupied an area stretching from Kenya in the west to eastern Australia, and between 35° and 70°S. Playfordiaspora cancellosa had a similar distribution pattern in eastern Gondwana, but had also expanded its range further westwards to central and southern Africa and South America (Fig. 8). Vijaya (1995) also noticed the palaeobiogeographical pattern of P. cancellosa and closely related species, suggesting that the parent plant(s) characterized cool climates. S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 However, this study indicates that the parent plant(s) were adapted to perhaps seasonally dry conditions, and that they migrated southwards during the Permian– Triassic as the semi-arid belt continued to expand further polewards. In Australia, Triplexisporites playfordii is a consistent and characteristic component of Triassic palynofloras, and it is abundant in the redbeds of both western and eastern Australia (Foster and Archbold, 2001). However, in eastern Australia it is also a common constituent in the latest Permian coal measures (Foster and Archbold, 2001), e.g. in assemblages from the Rangal Coal Measures (Michaelsen et al., 1999; Michaelsen, 2002). Foster and Archbold (2001) suggested that the parent plant of T. playfordii was part of the Late Permian swamp flora. However, the palaeobiogeographical distribution of this taxon through the Middle Permian to Middle Triassic suggests that it was adapted to dry conditions, and that its presence in South Africa, Kenya, Pakistan and Australia by the latest Permian signals the on-set of drier climate in those areas. 5.4. Contrasting palaeoecological trends across the P–T transition One interesting palaeoecological aspect is the quantitative changes of palynofloral groups across the P–T boundary in Gondwana. Some quantitative changes, e.g. the demise of glossopterid pollen, are useful biostratigraphic and palaeogeographic markers on a regional scale. But many other quantitative changes are prominent only locally, as they mirror narrow geographic environmental and climatic changes. The Late Permian assemblages of the PCMs, overwhelmingly dominated by glossopterid gymnosperms, were replaced in the Early Triassic by assemblages containing higher proportions of spores from ferns and lycophytes. The exact opposite scenario occurred in Kenya where the latest Permian assemblage is dominated by acavate and cavate trilete spores, and the Early Triassic palynoflora is enriched in taeniate (although non-glossopterid) bisaccate pollen (Hankel, 1992). In terrestrial P–T boundary sections in South China, spores are also dominant (85% relative abundance) in the latest Permian assemblages whereas in the earliest Triassic gymnospermous pollen are most abundant with 60% (Peng et al., 2005, in press). 5.5. 13 C signal across the P–T transition In terrestrial P–T sections from Gondwana there are discrepancies between the palynologically inferred P–T boundary and the negative δ13C excursion that is com- 113 monly used as a proxy for the boundary in the absence of biostratigraphic or radiometric data. δ13C analyses from the Bowen Basin, Australia, show a negative shift within the Protohaploxypinus microcorpus Zone or APP6, but with maximum negative values within the lowermost part of the Lunatisporites pellucidus Zone or APT1 (Morante, 1996; Hansen et al., 2000). In Madagascar, de Wit et al. (2002) recorded a sharp negative δ13C spike a few metres above the palynologically defined P–T boundary in the Morondava Basin, but also showed that the negative excursion began some 8 m below the boundary. In the Banspetali section in the Raniganj Basin, India, where the uppermost Permian coal seam occurs ca 17 m below the palynologically defined P–T boundary, Sarkar et al. (2003) reported a ∼9‰ drop in organic carbon δ13C 8 m above the palynologically inferred P–T boundary. However, in the GAM-7 borehole from Godavari Basin de Wit et al. (2002) found a large sharp negative δ13C spike of ∼8‰ ca 10 m below the palynologically inferred P–T boundary as defined by the FAD of L. pellucidus, followed by a sharp reversal of ∼14‰. The large negative spike is preceeded by a weak negative trend that appears to start some 85 m below the boundary. In the GAM-7 borehole, the cessation of strata containing coal and carbonaceous shale occurs a little more than 100 m below the P–T boundary (Srivastava and Jha, 1990). No δ13C analyses were carried out on the Carlton Heights section in South Africa but at Bethulie, also in the Karoo Basin, the initial negative δ13C excursion coincides with the first occurrence of Lystrosaurus at the base of the laminated maroon mudstone beds (MacLeod et al., 2000; Smith and Ward, 2001; Steiner et al., 2003). According to Steiner et al. (2003) this equates to 20 m below the fungal spike layer at Carlton Heights, but 17 m below according to Retallack et al. (2003). No δ13C values are yet available from the PCM succession. The primary source of the organic matter has a strong influence on the isotopic values of organic carbon (Foster et al., 1997). Wood-derived kerogen is isotopically heavier (− 24‰) than an assemblage dominated by spinose acritarchs (− 30‰), so bulk analyses of organic shale rich in wood debris always yield isotopically light signatures (Foster et al., 1997). This emphasises the importance of conducting throrough palynological and palynofacies studies on the same samples from which bulk δ13Corg analyses are carried out. 5.6. P–T Gondwanan palaeogeography, sea levels and palaeoclimate Palaeogeographic reconstructions of Pangea place the Prince Charles Mountains at 60°S around 250 Ma 114 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 (Torsvik and Van der Voo, 2002), more or less at the centre of the Gondwanan part of Pangea. The same palaeogeographic reconstructions place the Bowen Basin at 60–65°S, Karoo Basin at 60–50°S, Kenya and Madagascar around 50–40°S, and the Indian Godavari and Son-Mahanadi Basins at 55°S. During the Middle to Late Permian a gradual warming trend is evident from the western to the eastern parts of Gondwana. In southern and central Africa, coal deposition ended during the Middle Permian (Cairncross, 2001), and warm, semi-arid conditions reigned from the Capitanian (Visser, 1995). A gradual warming trend through the Permian is also evident by the reduction in ice-rafted dropstones and other cryogenic features in eastern Australian basins (Draper, 1983). The decline in number and thickness of coal seams in the PCMs, together with increasing ratios of silica and aluminium oxides in the uppermost coals, suggest increased weathering and climatic drying towards the end of the Permian (Holdgate et al., 2005). The Late Permian coals in the PCMs were deposited during consistently moist and cool conditions under a strongly seasonal light regime (Weaver et al., 1997; McLoughlin and Drinnan, 1997a). Similarly, coal deposition continued more or less to the end of the Permian in the Perth, Bowen, and Sydney basins. High-latitude coal forest swamps developed in Gondwana during the Late Permian due to global warming. However, by the end of the Permian the intensifying greenhouse conditions in combination with the strongly seasonal light regime could no longer sustain the remaining coal forest swamps even in polar latitudes (Kidder and Worsley, 2004). Although the end-Permian has traditionally been considered to correspond to a major sea-level lowstand (Hallam, 1984; Ross and Ross, 1987), Hallam and Wignall (1999) claimed that the P–T boundary corresponds to a phase of rising sea levels. Most Gondwana basins are characterized by continental sediments in the latest Permian, which favours the traditional eustatic models. However, there is widespread evidence for marine transgression in the Early Triassic (Wignall et al., 1996). Marine spinose acritarchs, mainly Veryhachium and Micrhystridium spp., have been encountered in Early Triassic assemblages from Pakistan (Balme, 1970), Madagascar (Wright and Askin, 1987; Hankel, 1993), and Western Australia (Dolby and Balme, 1976; Thomas et al., 2004). The earliest Triassic oil source rocks (Kockatea Shale) in the Perth Basin were deposited during a transgressive phase, either under strongly anoxic conditions, or as a result of coastal upwelling, and which lasted until the Dienerian, i.e. upper Induan (Thomas et al., 2004). The basal beds of the Kockatea Shale are characterized by a very low diversity marine fauna and extensive stromatolitic layers. Tripathi (1997) recorded marine acritarchs in the latest Permian of South Rewa, Rajmahal and Damodar basins, and in the earliest Triassic of South Rewa and Damodar basins. Following the model proposed by Harrowfield et al. (2005), marine inundation in the Early Triassic may have extended deep into the supercontinent along a pre-existing (Permian) intra-Gondwanan rift. An accurate eustatic signal may be difficult to resolve in the absence of a well-developed, tectonically undisturbed passive margin succession in Gondwana. The Early Triassic climate of the PCMs was less seasonal with increasing aridity as indicated by the initiation of red-bed deposition (McLoughlin and Drinnan, 1997b; McLoughlin et al., 1997). However, Retallack et al. (2003) argued that the increased ratio of alumina in the Early Triassic palaeosols compared to those of the Permian, and variations in distribution of calcareous nodules in the palaeosols indicate that the Early Triassic climate of the Karoo Basin was less seasonal, and more humid (semi-arid to sub-humid) than the strongly seasonal arid palaeoclimate of the Late Permian. It appears that the humid and strongly seasonal areas to the east became drier and less seasonal across the P–T transition, while the arid and strongly seasonal regions to the west also became less seasonal, but more humid. Thus, in the aftermath of the end-Permian event a generally warmer and less seasonal climate appears to have prevailed in southern Gondwana than during the Late Permian. 6. Implications for possible causes of the end-Permian extinction The cause of the end-Permian extinction event is conjectural; proposed scenarios including 1) an asteroid impact (Basu et al., 2003; Becker et al., 2004), 2) flood basalt volcanism (Renne et al., 1995; Courtillot and Renne, 2003), 3) release of methane from clathrates (Ryskin, 2003), and most recently 4) extreme global warming during the Permian initiated by the waning of the Alleghenian/Variscan/Hercynian orogeny and further intensified by 2) and 3) (Kidder and Worsley, 2004). Two large pulses of continental flood basalts occurred in the latest Permian: the Emeishan basalts in South China that appear to be synchronous with the end-Guadalupian extinction (Lo et al., 2002), and emplacement of the Siberian Traps was coeval with the end-Permian extinction (Mundil et al., 2004). Ecosystem recovery after the end-Permian extinction is known to have been unusually slow, with a duration at least twice those following other major extinctions S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 (Hallam, 1991; Erwin, 1998a,b). Several large fluctuations of both organic and carbonate δ13C occurred during the Early Triassic (MacLeod et al., 2000; de Wit et al., 2002; Payne et al., 2004). These perturbations may in part be linked to release of volcanigenic CO2 or methane hydrates but they may also incorporate a biological signature. As global marine biodiversity first began to rise in the Smithian, i.e. early Olenekian (Payne et al., 2004), these fluctuations coincided with the prolonged marine ecosystem recovery after the end-Permian crisis. The PCM palynological record indicates that the collapse of the terrestrial ecosystem was initiated already prior to the deposition of the last coal. A first extinction phase can be recognized within a 5 m interval starting 49 m below the last coal. The major extinction in the latest Permian is directly associated with the last coal, and was followed by continued stepwise extinction over a stratigraphic interval of 85 m. Using the calculated average sedimentation rate for the Ritchie Member of 261 m/Ma the extinction event may have lasted ca 325 000 years although high-resolution sampling will be necessary to constrain the finer details of floristic turnover. The PCM data also show a similar stepwise floristic recovery of the terrestrial ecosystem, where each level of extinction corresponds to the appearance of a suite of new taxa. The increase in spore-pollen diversity in the Early Triassic of the PCMs is, thus, a direct effect of continued stepwise extinction offset by simultaneous floristic recovery. These stepwise changes in the flora are consistent with the turnover of terrestrial vertebrates through the Permian– Triassic transition in the Karoo Basin, South Africa, reported by Smith and Botha (2005). The environmental changes that took place at the end of the Permian were dramatic enough to eliminate the glossopterid dominated ecosystem of southern Gondwana. In high latitude areas above 60°S, such as the PCMs and Bowen Basin, coal deposition continued throughout the Late Permian, while in other areas to the west and north coal deposition ceased earlier. This supports the theory that extreme global warming was occurring during the Permian. Stepwise extinction of taxa typically associated with the glossopterid flora continued for a short interval beyond the initial biotic crisis. Contemporaneous stepwise introduction of new taxa, and Gondwana-wide re-organisation of the terrestrial ecosystem show that the effects of the end-Permian crisis were continuing to affect the biota at least until the Olenekian. 7. Conclusions All samples from the McKinnon Member of the Bainmedart Coal Measures, Prince Charles Mountains, 115 including one from the uppermost coalseam (representing the top of the member), yielded typical Late Permian assemblages dominated by glossopterid pollen. There are minor quantitative variations in the palynoflora, but the samples contain essentially equivalent assemblages demonstrating that the Late Permian terrestrial ecosystem was quite stable in this area. The sample from the uppermost coal seam yielded the last typically Permian glossopterid-dominated palynoflora. The succeeding sample, collected from the lower Ritche Member of the Flagstone Bench Formation, 24 m above the uppermost coal, contains a fundamentally different palynoflora of earliest Triassic aspect demonstrating that the terrestrial ecosystem underwent a dramatic change through that interval. However, rather than a simple abrupt turnover, it is evident that the immediate post-crisis phase was a period of constant floristic change. After the disappearance of a large number (33%) of typical Permian taxa, an initial increase in diversity of taxa of earliest Triassic aspect occurred simultaneously with continued extinction of lingering Permian taxa. Only later in the Early Triassic do diversity levels appear to become more constant as the number of FADs decrease. A similar initial diversity increase instead of a decrease was also described by Looy et al. (2001) from East Greenland. In the aftermath of the end-Permian crisis only 26% of the typical Permian taxa present from the lower McKinnon Member appear to have survived until late Induan times. The average sedimentation rate indicates that the extinction event lasted ca 325000 years. This can be compared with the vertebrate extinction data for the Karoo Basin, which show that 69% of the vertebrate fauna disappeared over a period of 300 000 years, followed by a lesser extinction phase wiping out the remaining 31% 160000 years later (Smith and Botha, 2005). The palynological pattern is matched by the sedimentological record of the PCMs. Coal seams show gradual diminution in thickness and spacing below the Permian–Triassic transition, unlike that reported from the Bowen Basin (Michaelsen, 2002). The immediately overlying succession is dominated by thick sandstone packages interspersed with sparse carbonaceous shales. Higher in the Lower Triassic succession carbonaceous beds disappear and are replaced by thin red or mottled shales. Contemporaneous changes in fluvial style from meandering rivers or coal-rich braided systems, to braided river systems with episodic discharge have been recorded in different Gondwanan basins across the P–T transition, and have been attributed to increased sediment load due to loss of vegetation cover. 116 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 Comparisons of the palynology of P–T transitions from different parts of Gondwana also show that there were major differences in the composition of the regional palynofloras in the latest Permian, but that these became more similar following the initial biotic crisis. In humid areas, e.g. the PCMs, gymnospermous pollen were overwhelmingly dominant in the latest Permian, and following the mass-extinction lycophyte spores proliferated. In semi-arid areas, e.g. Kenya, lycophyte spores were already prominent constituents of the latest Permian palynoflora. Instead, gymnosperms became more dominant after the initial crisis. Humid areas became drier and at least some dry areas more humid. This Gondwana-wide re-organisation of the terrestrial ecosystem indicates that dramatic changes of the atmospheric cells took place during the latest Permian to earliest Triassic, as suggested by Kidder and Worsley (2004). In the terrestrial setting this resulted in apparently more equable, sub-humid to semi-arid conditions across southern Gondwana. The 24 m sampling gap between the uppermost Bainmedart Coal Measures sample and the lowermost Flagstone Bench Formation sample in the PCMs currently prohibits analysis of the short term changes associated with the end-Permian extinction. However, the palynological record from the PCMs shows that after the end-Permian crisis the terrestrial ecosystem was already on its way to recovery in the Induan. Acknowledgements This study was funded by a Swedish Research Council grant to SL and an Australian Research Council Australian Research Fellowship to SM. The Australian Antarctic Division provided financial and logistical support during the expedition to Prince Charles Mountains during the Austral summer of 1994–1995. Editor Henk Visscher and the reviewers John Backhouse and Clinton B. Foster are gratefully acknowledged for valuable comments that improved the manuscript. Appendix A. Alphabetical list of taxa identified in this study Alisporites asansolensis Maheshwari and Banerji 1975 Alisporites splendens (Leschik) Foster, 1979 Alisporites tenuicorpus Balme, 1970 Alisporites spp. Apiculatisporis clematisi de Jersey, 1968 Aratrisporites spp. Baculate sporomorph indet. Baculatisporites bharadwaji Hart, 1963 Baculatisporites spp. Barakarites rotatus (Balme and Hennelly) Bharadwaj and Tiwari, 1964 Bascanisporites undosus Balme and Hennelly 1956 Botryococcus sp. Brazilea scissa (Balme and Hennelly) Foster, 1975 Brevitriletes hennellyi Foster, 1979 Brevitriletes levis (Balme and Hennelly) Bharadwaj and Srivastava 1969 Calamospora tener (Leschik) de Jersey 1962 Camptotriletes warchianus Balme 1970 Cannanoropollis bilateralis (Tiwari) Lindström, 1995 Cannanoropollis janakii Potonié and Sah, 1960 Chordasporites australiensis de Jersey, 1962 Circulisporites parvus de Jersey, 1962 Clavatisporites spp. Concavissimisporites grumulus Foster, 1979 Converrucosisporites cameronii (de Jersey) Playford and Dettmann, 1965 Converrucosiporites sp. A Converrucosisporites spp. Convolutispora spp. Corisaccites alutas Venkatachala and Kar, 1966 Crustaesporites spp. Cycadopites follicularis Wilson and Webster, 1946 Cyclogranisporites sp. A Cyclogranisporites spp. Deltoidospora australis (Couper) Pocock, 1970 Deltoidospora breviradiatus (Helby) Densipollenites indicus Bharadwaj, 1962 D. invisus Bharadwaj and Salujha, 1964 Densoisporites complicatus Balme 1970 Densoisporites nejburghii (Schulz) Balme 1970 Densoisporites playfordii (Balme) Dettmann, 1963 Densoisporites psilatus (de Jersey) Raine and de Jersey in Raine et al. 1988 Dictyophyllidites mortonii (de Jersey) Playford and Dettmann, 1965 Dictyotidium spp. Dictyotriletes sp. A Didecitriletes ericianus (Balme and Hennlly) Venkatachala and Kar, 1965 Didecitriletes uncinatus (Balme and Hennelly) Venkatachala and Kar, 1965 Distriatites dettmannae (Segroves) Foster, 1979 Distriatites insolitus Bharadwaj and Salujha, 1964 Dulhuntyispora granulata Price, 1983 Ellipsovelatisporites sp. Enzonalasporites vigens Leschik, 1955 Ephedripites sp. (large) Equisetosporites steevesiae (Jansonius) de Jersey, 1962 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 Falcisporites australis (de Jersey) Stevens, 1981 Falcisporites stabilis Balme 1970 Florinites eremus Balme and Hennelly, 1955 Fungal hyphae cf. Palaeancistrus spp. Gnetaceaepollenites bulbiger Anderson 1977 Gondisporites raniganjensis Bharadwaj, 1962 Gondisporites sp. A Goubinispora morondavensis (Goubin) Tiwari and Rana, 1981 Granulatisporites absonus Foster, 1979 Granulatisporites sp. Grebespora concentrica Jansonius, 1962 Guttatisporites sp. Guttulapollenites hannonicus Goubin, 1965 Horriditriletes filiformis (Balme and Hennelly) Backhouse 1991 Horriditriletes ramosus (Balme and Hennelly) Bharadwaj and Salujha 1964 Horriditriletes tereteangulatus (Balme and Hennelly) Backhouse 1991 Inaperturopollenites nebulosus Balme 1970 Inaperturopollenites sp. Indospora clara Bharadwaj, 1962 Indospora laevigata Bharadwaj and Salujha emend. Foster, 1979 Indotriradites niger (Segroves) Backhouse 1991 Indotriradites rallus (Balme) Foster 1979 Indotriradites sp. cf. I. reidii Foster, 1979 Interradispora daedala Foster, 1979 Interradispora versus Price, 1979 Klausipollenites schaubergeri (Potonié and Klaus) Jansonius, 1962 Klausipollenites sp. A Kraeuselisporites cuspidus Balme, 1963 Kraeuselisporites saeptatus Balme, 1963 Kraeuselisporites verrucifer de Jersey and Hamilton, 1967 Kraeuselisporites spp.Leiotriletes virkkii Tiwari, 1965 Laevigate sporomorph indet. Laevigatosporites colliensis (Balme and Hennelly) Venkatachala and Kar, 1968 Laevigatosporites spp. Leiotriletes directus Balme and Hennelly, 1955 Limatulasporites fossulatus (Balme) Helby and Foster 1979 in Foster, 1979 L. limatulus (Balme) Helby and Foster, 1979 in Foster, 1979 Lophotriletes novicus Singh, 1964 Lueckisporites virkkiae Potonié and Klaus, 1954 Lueckisporites spp. Lunatisporites acutus Leschik, 1955 Lunatisporites noviaulensis (Leschik) Foster, 1979 117 L. sp. cf. L. noviaulensis (Leschik) Foster, 1979 Lunatisporites obex (Balme) de Jersey 1979 Lunatisporites pellucidus (Goubin) Helby, 1972 Lunatisporites spp. Lundbladispora brevicula Balme, 1963 Lundbladispora willmottii Balme, 1963 Lundbladispora spp. Maculatasporites spp. Marsupipollenites striatus (Balme and Hennelly) Foster, 1979 Marsupipollenites triradiatus Balme and Hennelly, 1956 Mehlisphaeridium regulare Anderson 1977 Microbaculispora micronodosa (Balme and Hennelly) Anderson 1977 Microbaculispora tentula Tiwari, 1965 Microbaculispora trisina (Balme and Hennelly) Anderson 1977 Microbaculispora villosa (Balme and Hennelly) Bharadwaj, 1962 Minutosaccus sp. Monosulcites spp. Osmundacidites fissus (Leschik) Playford, 1965 Osmundacidites senectus Balme, 1963 Osmundacidites wellmanii Couper, 1953 Ovalipollis sp. cf. Ovalipollis sp. Peltacystia monile Balme and Segroves, 1966 Peltacystia venosa Balme and Segroves 1966 Platysaccus leschikii Hart, 1960 Platysaccus queenslandi de Jersey, 1962 Platysaccus spp. Playfordiaspora cancellosa (Playford and Dettmann) Maheshwari and Banerji, 1975 Polypodiidites sp. sensu Balme 1970 Polypodiisporites mutabilis Balme 1970 Potonieisporites balmei (Hart) Segroves, 1969 Potonieisporites novicus Bharadwaj, 1954 Praecolpatites sinuosus (Balme and Hennelly) Bharadwaj and Srivastava, 1969 Protohaploxypinus amplus (Balme and Hennelly) Hart, 1964 P. bharadwajii Foster, 1979 Protohaploxypinus jacobii (Jansonius) Hart, 1964 Protohaploxypinus limpidus (Balme and Hennelly) Balme and Playford 1967 Protohaploxypinus microcorpus (Schaarschmidt) Clarke, 1965 P. pennatulus (Andreyeva) Hart, 1964 Protohaploxypinus perexiguus (Bharadwaj and Salujha) Foster, 1979 Protohaploxypinus rugatus Segroves, 1969 118 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 Protohaploxypinus samoilovichii (Jansonius) Hart, 1964 Protohaploxypinus spp. Pteruchipollenites gracilis (Segroves) Foster, 1979 Punctatisporites fungosus Balme, 1963 Punctatisporites spp. Punctatosporites walkomii de Jersey, 1962 Punctatosporites sp. Quadrisporites horridus Hennelly ex Potonié and Lele, 1961 Reduviasporonites chalastus (Foster) Elsik, 1999 Retusotriletes clipeata Helby, 1966 Retusotriletes junior de Jersey and Hamilton, 1967 Retusotriletes nigritellus (Luber) Foster, 1979 Retusotriletes ”radiatus” sensu Helby 1973 Rewanispora foveolata de Jersey, 1970 Rugulatisporites trisinus de Jersey and Hamilton, 1967 Remarks: Specimens herein assigned to R. trisinus differ slightly from the figured holotype in that the rugulate ornamentation is generally somewhat denser and finer, but they still conform with the original description for the species. Rugulatisporites spp. Sahnites sp. Scheuringipollenites maximus (Hart) Tiwari, 1973 Scheuringipollenites ovatus (Balme and Hennelly) Foster, 1975 Schizopollis disaccoides Venkatachala and Kar, 1964 Schizopollis woodhousei Venkatachala and Kar, 1964 Semiretisporis sp. cf. S. denmeadii (de Jersey) de Jersey, 1970 Small scabrate thin-walled sporomorphs indet. Spinate sporomorph indet. Striatoabieites multistriatus (Balme and Hennelly) Hart, 1964 Striatopodocarpidites cancellatus (Balme and Hennelly) Hart, 1964 Striatopodocarpidites fusus (Balme and Hennelly) Potonié, 1956 S. rarus (Bharadwaj and Salujha) Balme 1970 Striatopodocarpidites solitus (Bharadwaj and Salujha) Foster, 1979 Striatopodocarpidites spp. Striomonosaccites brevis Bose and Kar, 1966 Striomonosaccites sp. Sulcosaccispora alaticonformis (Malyavkina) de Jersey, 1968 Tetraporina tetragona (Pant and Mehtra) Anderson 1977 Thick-walled rugulate/scabrate sporomorphs indet. Thymospora cicatricosa (Balme and Hennelly) Hart, 1965 Triadispora sp. cf. T. epigona Klaus, 1964 Triplexisporites playfordii (de Jersey and Hamilton) Foster, 1979 Triquitrites proratus Balme 1970 Tuberculatosporites aberdarensis de Jersey, 1962 Uvaesporites verrucosus (de Jersey) Helby in de Jersey, 1971 Uvaesporites sp. Verrucosisporites surangei Maheshwari and Banerji, 1975 Verrucosisporites sp. cf. V. trisecatus Balme and Hennelly, 1956 Verrucosisporites spp. Vitreisporites bjuvensis Nilsson, 1958 Vitreisporites pallidus (Reissinger) Nilsson, 1958 Weylandites lucifer (Bharadwaj and Salujha) Foster, 1975 Weylandites magmus (Bose and Kar) Backhouse 1991 References Adamson, D., Darragh, A., 1991. Field evidence on Cainozoic history and landforms in the northern Prince Charles Mountains, East Antarctica. In: Gillieson, D., Fitzsimons, S. (Eds.), Quaternary Research in Australian Antarctica: Future Directions. Special Publication, vol. 3. Department of Geography and Oceanography, University College, The Australian Defence Force Academy, Canberra, pp. 5–14. Anderson, J.M., 1977. The biostratigraphy of the Permian and Triassic. Part 3. A review of Gondwana palynology with particular reference to the northern Karoo Basin. Mem. Bot. Surv. S. Afr. 41, 1–188. Anderson, J.M., Anderson, H.M., Archangelsky, S., Bamford, M., Chandra, S., Dettmann, M., Hill, R., McLoughlin, S., Rösler, O., 1999. Patterns of Gondwana plant colonisation and diversification. J. Afr. Earth Sci. 28, 145–167. Askin, R.A., Elliot, D.H., 1982. Geologic implications of recycled Permian and Triassic palynomorphs in Tertiary rocks of Seymour Island, Antarctic peninsula. Geology 10, 547–551. Backhouse, J., 1991. Permian palynostratigraphy of the Collie Basin, Western Australia. Rev. Palaeobot. Palynol. 67, 237–314. Backhouse, J., 1993. Palynology and correlations of Permian sediments in the Perth, Collie, and Officer Basins, Western Australia. Geol. Surv. West. Aust., Rep. 34, 111–128. Balme, B.E., 1963. Plant microfossils from the Lower Triassic of Western Australia. Palaeontology 6, 1–40. Balme, B.E., 1970. Palynology of the Permian and Triassic strata in the Salt Range and Surghar Range, West Pakistan. In: Kummel, B., Teichert, C. (Eds.), Stratigraphic Boundary Problems; Permian and Triassic of West Pakistan. Spec. Publ., vol. 4. Kansas University, Department of Geology, pp. 305–453. Balme, B.E., 1995. Fossil in situ spores and pollen grains; an annotated catalogue. Rev. Palaeobot. Palynol. 87, 81–322. Balme, B.E., Playford, G., 1967. Late Permian plant microfossils from the Prince Charles Mountains, Antarctica. Rev. Micropaléontol. 3, 179–192. S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 Basu, A.R., Petaev, M.I., Poreda, R.J., Jacobsen, S.B., Becker, L., 2003. Chondritic meteorite fragments associated with the Permian–Triassic boundary in Antarctica. Science 302, 1388–1392. Becker, L., Poreda, R.J., Basu, A.R., Pope, K.O., Harrison, T.M., Nicholson, C., Iasky, R., 2004. Bedout: A possible end-Permian impact crater offshore northwestern Australia. Science 304, 1469–1476. Benton, M.J., Twitchett, R.J., 2003. How to kill (almost) all life: the end-Permian extinction event. Trends Ecol. Evol. 18, 358–365. Boger, S.D., Wilson, C.J.L., 2003. Brittle faulting in the Prince Charles Mountains, East Antarctica: Cretaceous transtensional tectonics related to the break-up of Gondwana. Tectonophysics 367, 173–186. Bowring, S.A., Erwin, D.H., Jin, Y., Martin, M.W., Davidek, K.L., Wei, W., 1998. U/Pb zircon geochronology and tempo of the endPermian mass extinction. Science 280, 1039–1045. Broutin, J., 1986. Étude paléobotanique et palynologique du passage Carbonifére Permien dans le Sud-Ouest de la péninsule Ibérique. Cahiers de paléontologie, section Vertebres. Editions du Centre National de la Rechersche Scientifique, Paris, France. 165 pp. Cairncross, B., 2001. An overview of the Permian (Karoo) coal deposits of southern Africa. J. Afr. Earth Sci. 33, 529–562. Cantrill, D.J., Drinnan, A.N., 1994. Late Triassic megaspores from the Amery Group, Prince Charles Mountains, East Antarcticad. Alcheringa 18, 71h–78h. Cantrill, D.J., Drinnan, A.N., Webb, J.A., 1995. Late Triassic plant fossils from the Prince Charles Mountains, East Antarctica. Antarct. Sci. 7, 51–62. Courtillot, V.E., Renne, P.R., 2003. On the ages of flood basalt events. C. R. Geosci. 335, 113–140. de Jersey, N.J., 1979. Palynology of the Permian–Triassic transition in the western Bowen Basin. Geol. Surv. Qd 374, Palaeontol. pap. 46, 1–61. de Jersey, N.J., Raine, J.I., 1990. Triassic and earliest Jurassic miospores from the Murihiku Supergroup, New Zealand. N.Z. Geol. Surv. Bull. 62, 1–164. de Wit, M.J., Ghosh, J.G., de Villiers, S., Rakotosolofo, N., Alexander, J., Tripathi, A., Looy, C., 2002. Multiple organic Carbon isotope reversals across the Permo-Triassic boundary of terrestrial Gondwana Sequences: clues to extinction pattern and delayed ecosystem recovery. J. Geol. 110, 227–240. Dolby, J.H., Balme, B.E., 1976. Triassic palynology of the Carnarvon Basin, Western Australia. Rev. Palaeobot. Palynol. 22, 105–168. Draper, J.J., 1983. Origin of pebbles in mudstones in the Denison Trough. Permian Geology of Queensland. Geological Society of Australia, Queensland Division, Brisbane, pp. 305–316. Erwin, D.H., 1998a. After the end: recovery from extinctions. Science 279, 1324–1325. Erwin, D.H., 1998b. The end and the beginning: recoveries from mass extinctions. TREE 13, 344–349. Eshet, Y., 1990. Palynological aspects of the Permo-Triassic succession in the subsurface of Israel. Bull. Geol. Surv. Israel 81, 1–42. Eshet, Y., Rampino, M.R., Visscher, H., 1995. Fungal event and palynological record of ecological crisis and recovery across the Permian–Triassic boundary. Geology 23, 967–970. Falcon, R., 1973. Palynology of the Middle Zambesi Basin. In: Bond, G. (Ed.), The Palaeontology of Rhodesia. Geol. Surv. Rhodesia, Bull., vol. 70, pp. 43–71. Fedorov, L.V., Grikurov, G.E., Kurinin, R.G., Masolov, V.N., 1982. Crustal structure of the Lambert Glacier area from geophysical data. In: Craddock, C. (Ed.), Antarctic Geoscience. University of Wisconsin Press, Madison, pp. 931–936. 119 Fielding, C.R., Webb, J.A., 1995. Sedimentology of the Permian Radok Conglomerate in the Beaver Lake area of MacRobertson Land, East Antarctica. Geol. Mag. 132, 51–63. Fielding, C.R., Webb, J.A., 1996. Facies and cyclicity of the late Permian Bainmedart Coal Measures in the northern Prince Charles Mountains, MacRobertson Land, Antarctica. Sedimentology 43, 295–322. Foster, C.B., 1979. Permian plant microfossils of the Blair Athol Coal Measures, Baralaba Coal Measures, and basal Rewan Formation of Queensland. Publ.— Geol. Surv. Qld. 372. Palaeontol. Pap. 45, 1–244. Foster, C.B., 1982. Spore-pollen assemblages of the Bowen Basin, Queensland (Australia): their relationship to the Permian/Triassic boundary. Rev. Palaeobot. Palynol. 36, 165–183. Foster, C.B., Archbold, N.W., 2001. Chronologic anchor points for the Permian and Early Triassic of the Eastern Australian Basins. In: Weiss, R.H. (Ed.), Contributions to Geology and Palaeontology of Gondwana in Honour of Helmut Wopfner. Geological Institute, University of Cologne, Germany, pp. 175–197. Foster, C.B., Balme, B.E., Helby, R., 1994. First record of Tethyan palynomorphs from the Late Triassic of East Antarctica. AGSO J. Aust. Geol. Geophys. 15, 239–246. Foster, C.B., Logan, G.A., Summons, R.E., Gorter, J.D., Edwards, D.S., 1997. Carbon isotopes, kerogen types and the Permian–Triassic boundary in Australia: Implications for exploration. APPEA J. 37, 472–489. Foster, C.B., Logan, G.A., Summons, R.E., 1998. The Permian– Triassic boundary in Australia: where is it and how is it expressed? Proc. R. Soc. Vic. 110, 247–266. Hallam, A., 1984. Pre-Quaternary sea-level changes. Annu. Rev. Earth Planet. Sci. 12, 205–243. Hallam, A., 1991. Why was there a delayed radiation after the endPalaeozoic extinctions? Hist. Biol. 5, 257–455. Hallam, A., Wignall, P.B., 1997. Mass Extinctions and Their Aftermath. Oxford University Press. Hallam, A., Wignall, P.B., 1999. Mass extinctions and sea-level changes. Earth-Sci. Rev. 48, 217–250. Hankel, O., 1987. Lithostratigraphic subdivision of the Karoo rocks of the Luwegu Basin (Tanzania) and their biostratigraphic classification based on microfloras, macrofloras, fossil wood and vertebrates. Geol. Rundsch. 76, 539–565. Hankel, O., 1991. Early Triassic plant microfossils from the Kavee Quarry section of the Lower Mariakani Formation, Kenya. Rev. Palaeobot. Palynol. 68, 127–145. Hankel, O., 1992. Late Permian to Early Triassic microfloral assemblages from the Maji ya Chumvi Formation, Kenya. Rev. Palaeobot. Palynol. 72, 129–147. Hankel, O., 1993. Early Triassic plant microfossils from Sakamena sediments of the Majunga Basin, Madagascar. Rev. Palaeobot. Palynol. 77, 213–233. Hansen, H.J., Lojen, S., Toft, P., Dolenec, T., Jinan, T., Michaelsen, P., Sarkar, A., 2000. Magnetic susceptibility and organic carbon isotopes of sediments across some marine and terrestrial PermoTriassic boundaries. In: Yin, H., Dickins, J.M., Shi, G.R., Tong, J. (Eds.), Permian–Triassic Evolution of the Tethys and Western Circum-Pacific. Elsevier, Amsterdam, pp. 271–289. Harrowfield, M., Holdgate, G., Wilson, C., McLoughlin, S., 2005. Tectonic significance of the Lambert Graben, East Antarctica: reconstructing the Gondwanan rift. Geology 33, 197–200. Helby, R., 1973. Review of Late Permian and Triassic palynology of New South Wales. Spec. Publ. — Geol. Soc. Aust. 4, 141–155. 120 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 Helby, R., 1977. Illustrations of selected spores and pollen from the Bonaparte Gulf Basin, Australia. 11 pp, 21 plates. Unpublished report. Helby, R., Morgan, R., Partridge, A.D., 1987. A palynological zonation of the Australian Mesozoic. Mem. Assoc. Australas. Palaeontol. 4, 1–94. Holdgate, G.R., McLoughlin, S., Drinnan, A.N., Finkelman, R.B., Willett, J.C., 2005. Inorganic chemistry, petrography and palaeobotany of Permian coals in the Prince Charles Mountains, East Antarctica. Int. J. Coal Geol. 63, 156–177. Jin, Y.G., Wang, Y., Wang, W., Shang, Q.H., Cao, C.Q., Erwin, D.H., 2000. Pattern of marine mass extinction near the Permian–Triassic boundary in South China. Science 289, 432–436. Kemp, E.M., 1972. Reworked palynomorphs from the West Ice Shelf area, East Antarctica, and their possible geological and palaeoclimatological significance. Mar. Geol. 13, 145–157. Kerp, H., Abu Hamad, A., Vörding, B., Bandel, K., 2006. Typical Triassic Gondwanan floral elements in the Upper Permian of the paleotropics. Geology 34, 265–268. Kidder, D.L., Worsley, T.R., 2004. Causes and consequences of extreme Permo-Triassic warming to global equable climate and relation to the Permo-Triassic extinction and recovery. Palaeogeogr. Palaeoclimatol. Palaeoecol. 203, 207–237. Kovach, W.L., Batten, D.J., 1993. Diversity changes in lycopsid and aquatic fern megaspores through geologic time. Paleobiology 19, 28–42. Kyle, R.A., 1977. Palynostratigraphy of the Victoria Group of South Victoria Land, Antarctica. N.Z. J. Geol. Geophys. 20, 1081–1102. Kyle, R.A., Schopf, J.M., 1982. Permian and Triassic palynostratigraphy of the Victoria Group, Transantarctic Mountains. In: Craddock, C. (Ed.), Antarctic Geoscience. The University of Wisconsin Press, Madison, pp. 649–659. Lindström, S., 1995a. Early Late Permian palynostratigraphy and palaeobiogeography of Vestfjella, Dronning Maud Land, Antarctica. Rev. Palaeobot. Palynol. 86, 157–173. Lindström, S., 1996. Late Permian palynology of Fossilryggen, Vestfjella, Dronning Maud Land, Antarctica. Palynology 20, 15–48. Lindström, S., McLoughlin, S., Drinnan, A.N., 1997. Intraspecific variation of taeniate bisaccate pollen within Permian glossopterid sporangia, from the Prince Charles Mountains, Antarctica. Int. J. Plant Sci. 158, 673–684. Lo, C.-H., Chung, S.-L., Lee, T.-Y., Wu, G., 2002. Age of the Emeishan flood magmatism and relations to Permian–Triassic boundary events. Earth Planet. Sci. Lett. 198, 449–458. Looy, C.V., Brugman, W.A., Dilcher, D.L., Visscher, H., 1999. The delayed resurgence of equatorial forests after the Permian–Triassic ecological crisis. Proc. Natl. Acad. Sci. U. S. A. 96, 13857–13862. Looy, C.V., Twitchett, R.J., Dilcher, D.L., Van Konijnenburg-Van Cittert, J.H.A., Visscher, H., 2001. Life in the end-Permian dead zone. PNAS 98, 7879–7883. MacLeod, K.G., Smith, R.M.H., Koch, P.L., Ward, P.D., 2000. Timing of mammal-like reptile extinctions across the Permian–Triassic boundary in South Africa. Geology 28, 1047–1050. Marshall, C.R., 1995. Distinguishing between sudden and gradual extinctions in the fossil record: predicting the position of the Cretaceous–Tertiary iridium anomaly using the ammonite fossil record on Seymour Island, Antarctica. Geology 23, 731–734. McLoughlin, S., Drinnan, A.N., 1997a. Revised stratigraphy of the Permian Bainmedart Coal Measures, northern Prince Charles Mountains, East Antarctica. Geol. Mag. 134, 335–353. McLoughlin, S., Drinnan, A.N., 1997b. Fluvial sedimentology and revised stratigraphy of the Triassic Flagstone Bench Formation, northern Prince Charles Mountains, East Antarctica. Geol. Mag. 134, 781–806. McLoughlin, S., Lindström, S., Drinnan, A.N., 1997. Gondwanan floristic and sedimentological trends during the Permian–Triassic transition: new evidence from the Amery Group, northern Prince Charles Mountains, East Antarctica. Antarct. Sci. 9, 281–298. McManus, H.A., Taylor, E.L., Taylor, T.N., Collinson, J.W., 2002. A petrified Glossopteris flora from Collinson Ridge, central Transantarctic Mountains; Late Permian or Early Triassic? Rev. Palaeobot. Palynol. 120, 233–246. Michaelsen, P., 2002. Mass extinction of peat-forming plants and the effect on fluvial styles across the Permian–Triassic boundary, northern Bowen Basin, Australia. Palaeogeogr. Palaeoclimatol. Palaeoecol. 179, 173–188. Michaelsen, P., Henderson, R.A., 2000. Facies relationships and cyclicity of high-latitude, Late Permian coal measures, Bowen Basin, Australia. Int. J. Coal Geol. 44, 19–48. Michaelsen, P., Foster, C.B., Henderson, R.A., 1999. Destabilasation and collapse of a long-lived (c. 9 Myr) peat mire ecosystem and dramatic changes of fluvial architecture: Permian–Triassic boundary, northern Bowen Basin, Australia. In: Yin, H., Tong, J. (Eds.), International Conference on Pangea and the Paleozoic–Mesozoic Transition. Wuhan, China, 9–11 March 1999, pp. 137–140. Michaelsen, P., Henderson, R.A., Crosdale, P.J., Fanning, C.M., 2001. Age and significance of the Platypus-Tuff Bed, a regional reference horizon in the Upper Permian Moranbah Coal Measures, north Bowen Basin. Aust. J. Earth Sci. 48, 183–192. Mishra, D.C., Chandra Sekhar, D.V., Venkata Raju, D.C., Vijaya Kumar, V., 1999. Crustal structure based on gravity-magnetic modelling constrained from seismic studies under Lambert Rift, Antarctica and Godavari and Mahanadi rifts, India and their interrelationship. Earth Planet. Sci. Lett. 172, 287–300. Morante, R., 1996. Permian and Early Triassic isotopic records of carbon and strontium in Australia and a scenario of events about the Permian–Triassic boundary. Hist. Biol. 11, 289–310. Mory, A.J., Backhouse, J., 1997. Permian Stratigraphy and Palynology of the Carnarvon Basin, Western Australia. Geol. Surv. West. Aust., Rep. 46, 1–101. Mundil, R., Ludwig, K.R., Metcalfe, I., Renne, P.R., 2004. Age and timing of the Permian mass extinctions: U/Pb dating of closedsystem zircons. Science 305, 1760–1763. Neish, P.G., Drinnan, A.N., Cantrill, D.J., 1993. Structure and ontogeny of Vertebraria from silicified Permian sediments in East Antarctica. Rev. Palaeobot. Palynol. 79, 221–244. Nyambe, I.A., Utting, J., 1997. Stratigraphy and palynostratigraphy, Karoo Supergroup (Permian and Triassic), mid-Zambesi Valley, southern Zambia. J. Afr. Earth Sci. 24, 563–583. Ottone, E.G., Garcia, G.B., 1991. A Lower Triassic miospore assemblage from the Puesto Viejo Formation, Argentina. Rev. Palaeobot. Palynol. 68, 217–232. Pant, D.D., Pant, R., 1987. Some Glossopteris leaves from Indian Triassic beds. Palaeontographica 205B, 165–178. Patterson, C., 1999. Evolution, (3rd. ed.). The Natural History Museum, London. Payne, J.L., Lehrmann, D.J., Wey, J., Orchard, M.J., Schrag, D.P., Knoll, A.H., 2004. Large perturbations of the carbon cycle during recovery from the end-Permian extinction. Science 305, 506–509. Peng, Y., Zhang, S., Yu, T., Yang, F., Gao, Y., Shi, G.R., 2005. Highresolution terrestrial Permian–Triassic eventostratigraphic boundary in western Guizhou and eastern Yunnan, southwestern China. Palaeogeogr. Palaeoclimatol. Palaeoecol. 215, 285–295. S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 Peng, Y., Yu, J., Gao, Y., Yang, F., in press. Palynological assemblages of non-marine rocks at the Permian–Triassic boundary, western Guizhou and eastern Yunnan, South China. J. Asian Earth Sci., 15 pp. (Published on-line). Pigg, K.B., 1992. Evolution of isoetalean lycopsids. Ann. Mo. Bot. Gard. 79, 589–612. Price, P.L., 1997. Permian to Jurassic palynostratigraphic nomenclature of the Bowen and Surat Basins. In: Green, P.M. (Ed.), The Surat and Bowen Basins, South-East Queensland. Queensland Department of Mines and Energy, pp. 137–178. Renne, P.R., Zhang, Z., Richards, M.A., Black, M.T., Basu, A.R., 1995. Synchrony and causal relations between Permian–Triassic boundary crises and Siberian flood volcanism. Science 269, 1413–1416. Retallack, G.J., 1975. The life and times of a Triassic lycopod. Alcheringa 1, 3–29. Retallack, G.J., 1977. Reconstructing Triassic vegetation of eastern Australia: a new approach for the biostratigraphy of Gondwanaland. Alcheringa 1, 247–278. Retallack, G.J., 2002. Lepidopteris callipteroides, an earliest Triassic seed fern from the Sydney Basin, southeastern Australia. Alcheringa 26, 475–500. Retallack, G.J., Veevers, J.J., Morante, R., 1996. Global coal-gap between Permian–Triassic extinction and Middle Triassic recovery of peat-forming plants. Geol. Soc. Amer. Bull. 108, 195–207. Retallack, G.J., Smith, R.M.H., Ward, P.D., 2003. Vertebrate extinction across the Permian–Triassic boundary in Karoo Basin, South Africa. GSA Bull. 115, 1133–1152. Retallack, G.J., Jahren, A.H., Sheldon, N.D., Chakrabarti, R., Metzger, C.A., Smith, R.M.H., 2005. The Permian–Triassic boundary in Antarctica. Antarct. Sci. 17, 241–258. Rohn, R., Rösler, O., 1989. Novas ocorrencias Glossopteris na Formação Rio do Rasto (Bacia do Paraná, Permiano Superior). Bol. IG-USP, Publ. Espec. 7, 101–125. Ross, C.A., Ross, J.R.P., 1987. Late Paleozoic sea levels and depositional sequences. Cushman Foundation for Foraminiferal Research. Special Publication, vol. 24, pp. 137–149. Ryskin, G., 2003. Methane-driven oceanic eruptions and massextinctions. Geology 31, 741–744. Sarkar, A., Yoshioka, H., Ebihara, M., Naraoka, H., 2003. Geochemical and organic carbon isotope studies across the continental PermoTriassic boundary of Raniganj Basin, eastern India. Palaeogeogr. Palaeoclimatol. Palaeoecol. 191, 1–14. Scotese, C.R., 2001. Atlas of Earth History 1, Palaeogeography. Paleomap Project, Arlington, Texas. 52 pp. Signor, P.W., Lipps, J.H., 1982. Sampling bias, gradual extinction patterns, and catastrophes in the fossil record. In: Silver, L.T., Schultz, P.H. (Eds.), Geological Implications of Large Asteroids and Comets on Earth. Geol. Soc. Amer. Spec. Pap., vol. 190, pp. 291–296. Smith, R., Botha, J., 2005. The recovery of terrestrial vertebrate diversity in the South African Karoo Basin after the end-Permian extinction. C. R. Palevol. 4, 623–636. Smith, R.M.H., Ward, P.D., 2001. Pattern of vertebrate extinctions across an event bed at the Permian–Triassic boundary in the Karoo Basin of South Africa. Geology 29, 1147–1150. Srivastava, S.C., Jha, N., 1990. Permian–Triassic palynofloral transition in Godavari Graben, Andra Pradesh. Palaeobotanist 38, 92–97. Stagg, H.M.J., 1985. The structure and origin of Prydz Bay and MacRobertson Shelf, East Antarctica. Tectonophysics 114, 315–340. Steiner, M.B., Eshet, Y., Rampino, M.R., Schwindt, D.M., 2003. Fungal abundance spike and the Permian–Triassic boundary in the 121 Karoo Supergroup (South Africa). Palaeogeogr. Palaeoclimatol. Palaeoecol. 194, 405–414. Taylor, E.L., Taylor, T.N., Collinson, J.W., 1989. Depositional setting and palaeobotany of Permian and Triassic permineralized peat from the central Transantarctic Mountains, Antarctica. Int. J. Coal Geol. 12, 657–679. Tewari, R.C., 1999. Sedimentary-tectonic status of Permian–Triassic boundary (250 Ma) in Gondwana stratigraphy of Peninsular India. Gondwana Res. 2, 185–189. Tewari, R.C., Veevers, J.J., 1993. Gondwana basins of India occupy the middle of a 7500 km sector of radial valleys and lobes in central eastern Gondwanaland. In: Findlay, R.H., Unrug, R., Banks, M.R., Veevers, J.J. (Eds.), Gondwana 8: Assembly, Evolution and Dispersal. A.A. Balkema Publishers, Rotterdam, pp. 507–512. Thomas, B.M., Willink, R.J., Grice, K., Twitchett, R.J., Purcell, R.R., Archbold, N.W., George, A.D., Tye, S., Alexander, R., Foster, C.B., Barber, C.J., 2004. Unique marine Permian–Triassic boundary section from Western Australia. Aust. J. Earth Sci. 51, 423–430. Tiwari, R.S., 1999. Paradigm of FAD, LAD and DOD of some miospore taxa in the Late Permian and Early Triassic successions on the Indian peninsula. J. Palaeontol. Soc. India 44, 69–89. Tiwari, R.S., Tripathi, A., 1992. Marker Assemblage-Zones of spore and pollen species through Gondwana Palaeozoic and Mesozoic sequences in India. Palaeobotanist 40, 194–236. Torsvik, T.H., Van der Voo, R., 2002. Refining Gondwana and Pangea palaeogeography: estimates of Phanerozoic non-dipole (octupole) fields. Geophys. J. Int. 151, 771–794. Tripathi, A., 1997. Representation of acritarchs across the Permian– Triassic boundary in India. Palaeobotanist 46, 112–117. Twitchett, R.J., Looy, C.V., Morante, R., Visscher, H., Wignall, P.B., 2001. Rapid and synchronous collapse of marine and terrestrial ecosystems during the end-Permian biotic crisis. Geology 29, 351–354. Utting, J., 1979. Pollen and spore assemblages from the Upper Permian of the North Luangwa Valley, Zambia. IV International Palynological Conference, Lucknow (1976–77), vol. 2, pp. 165–174. Veevers, J.J., Tewari, R.C., Mishra, H.K., 1996. Aspects of Late Triassic to Early Cretaceous disruption of the Gondwana coal bearing fan of east-central Gondwanaland. In: Guha, P.K.S, Sengupta, S., Ayyasami, K., Gosh, R.N. (Eds.), Gondwana Nine, Ninth International Gondwana Symposium. A.A. Balkema Publishers, Rotterdam, pp. 637–646. Vijaya, 1995. Revision of the Late Permian–Triassic pollen genus Playfordiaspora Maheshwari and Banerji 1975. Palaeobotanist 43, 54–67. Visser, J.N.J., 1995. Post-glacial Permian stratigraphy and geography of southern and central Africa: boundary conditions for climate modelling. Palaeogeogr. Palaeoclimatol. Palaeoecol. 118, 213–243. Wang, S.C., Marshall, C.R., 2004. Improved confidence intervals for estimating the position of a mass extinction boundary. Paleobiology 30, 5–18. Ward, P.D., Montgomery, D.R., Smith, R., 2000. Altered river morphology in South Africa related to the Permian–Triassic extinction. Science 289, 1740–1743. Ward, P.D., Botha, J., Buick, R., De Kock, M.O., Erwin, D.H., Garrison, G.H., Kirschvink, J.L., Smith, R., 2005. Abrupt and gradual extinction among Late Permian land vertebrates in the Karoo Basin, South Africa. Science 307, 709–714. Weaver, L., McLoughlin, S., Drinnan, A.N., 1997. Fossil woods from the Upper Permian Bainmedart Coal Measures, northern Prince Charles Mountains, East Antarctica. AGSO J. Aust. Geol. Geophys. 16, 655–676. 122 S. Lindström, S. McLoughlin / Review of Palaeobotany and Palynology 145 (2007) 89–122 Webb, J.A., Fielding, C.R., 1993. Permo-Triassic sedimentation within the Lambert Graben, northern Prince Charles Mountains, East Antarctica. In: Findlay, R.H., Unrug, R., Banks, M.R., Veevers, J.J. (Eds.), Gondwana Eight: Assembly, Evolution and Dispersal. Proc. 8th Gondwana Symposium. Balkema, Hobart, pp. 357–369. Wignall, P.B., Kozur, H., Hallam, A., 1996. On the timing of palaeoenvironmental changes at the Permo-Triassic boundary using conodont biostratigraphy. Hist. Biol. 12, 39–62. Wilson, L.R., 1962. Plant microfossils from Flowerpot Formation. Okla. Geol. Surv. Bull. 49, 5–40. Wright, R.P., Askin, R.A., 1987. The Permian–Triassic boundary in the Southern Morondava Basin of Madagascar as defined by plant microfossils. In: McKenzie, G.D. (Ed.), Gondwana Six: Stratigraphy, Sedimentology, and Palaeontology. Geophysical Monograph, vol. 41. American Geophysical Union, Washington, D.C., pp. 175–197. Zavattieri, A.M., Batten, D.J., 1996. Chapter 20B. Miospores from Argentinian Triassic deposits and their potential for intercontinental correlation. In: Jansonius, J., McGregor, D.C. (Eds.), Palynology; Principles and Applications, vol. 2. American Association of Stratigraphical Palynologists Foundation, pp. 767–778.

RELATED PAPERS

RELATED TOPICS

Log In

Synchronous palynof loristic extinction and recovery after the end- Permian event in the Prince Charles Mountains, Antarctica: Implications for palynof loristic turnover across Gondwana

Synchronous palynof loristic extinction and recovery after the end- Permian event in the Prince Charles Mountains, Antarctica: Implications for palynof loristic turnover across Gondwana

Related Papers

RELATED PAPERS

RELATED TOPICS