Geological Society, London, Special Publications

Cretaceous and Cenozoic vegetation of Antarctica integrating the fossil record

Imogen Poole and David J. Cantrill

Geological Society, London, Special Publications 2006; v. 258; p. 63-81 doi:10.1144/GSL.SP.2006.258.01.05

Email alerting click here to receive free email alerts when new articles cite this service article Permission click here to seek permission to re-use all or part of this article request Subscribe click here to subscribe to Geological Society, London, Special Publications or the Lyell Collection

Notes

Downloaded by Universiteitsbibliotheek Utrecht on 25 November 2008

© 2006 Geological Society of London Cretaceous and Cenozoic vegetation of Antarctica integrating the fossil wood record

IMOGEN POOLE 1,2,3 & DAVID J. CANTRILL 4 1Wood Anatomy Section, National Herbarium of the Netherlands, University of Utrecht Branch, P.O. Box 80102, 3585 CS Utrecht, The Netherlands (e-maik i.poole@geo, uu.nl) 2palaeontological Museum, Oslo University, P.O. Box 1172 Blidern, N-0318 Oslo, Norway 3present address: Faculty of Earth Sciences, Organic Geochemistry Group, University of Utrecht, P. O. Box 80021, 3508 TA, Utrecht, The Netherlands 4Swedish Museum of Natural History, Department of Palaeobotany, Box 50007, Stockholm 104 05, Sweden

Abstract: A compilation of data for Cretaceous and Cenozoic Antarctic fossil wood floras, predominantly from the James Ross Island Basin, provides a different perspective on floristic and vegetation change when compared with previous studies that have focused on macrofossils or palynology. The wood record provides a filtered view of -forming elements within the vegetation, something that cannot be achieved from studies focusing on regional palynology or leaf floras. Four phases of vegetation development in the over- storey are recognized in the Cretaceous and Cenozoic of the Antarctic Peninsula based on the distribution and taxonomic composition of wood floras: Aptian-Albian coniferous forests; ?Cenomanian-Santonian mixed angiosperm forests; Campanian-Maastrichtian southern temperate forests; and Palaeocene-Eocene reduced diversity Nothofagus forests. Comparisons between the wood record and information derived from palynological and leaf floras have important implications for our understanding of the spatial composition of the vegetation. There is no doubt that climate change during the Cretaceous and Tertiary influenced the vegetational composition, but evolving palaeoenvironments in the Antarc- tic Peninsula region were probably of equal, if not greater, importance.

James Eights (1833) reported the first fossil such that this unique and dynamic ecosystem wood from the Antarctic (South Shetland can be more fully understood. Islands, Antarctic Peninsula), a discovery that Four major types of fossil material pointed to a once vegetated landmass in what is provide information for past vegetation, these now an icy world. Today, as questions focus on include: leaf compressions and impressions; trying to understand the Earth system more wood; palynomorphs; and dispersed cuticular fully, it becomes more ever apparent that the material, all of which are present in the floras high latitudes played an important role in evolu- of the Antarctic Peninsula. The Antarctic tion of biotas, development of Southern Hemi- record of terrestrial vegetation is derived sphere biogeography (Drinnan & Crane 1990; predominantly from leaf and palynomorph Cantrill & Poole 2002) and mediating global records (e.g. Askin 1988, 1992). Abundant leaf climates (Upchurch et al. 1998). Indeed, the high floras have been described (e.g. Dus6n 1908; latitudes are most sensitive to climate fluctua- Zastawniak 1981, 1990, 1993, 1994; Li & Shen tion, yet there is no modern analogue today with 1990; Li 1994; Zhou & Li 1994; Zastawniak et which to compare the evidence from the past. al. 1995; Hayes 1999; Cantrill 2000; Dutra & An unparalleled record of life in the high Batten 2000; Hunt 2001), but these often lack southern latitudes is found on the Antarctic cuticles making systematic identification more Peninsula, which has, in recent years, furthered problematic. Flowers and fruits also occur, but our knowledge of the palaeoecology of this these are rare (Gandolfo et al. 1998; Eklund region through many systematic and climatolog- 2003) and have not to date been extensively ical studies. It is therefore timely to bring studied. By comparison, wood is abundant and together the works published over the past thus makes an important contribution to a decade, and revise and update previous concep- picture of biodiversity than would otherwise tions pertaining to the southern high latitudes result predominantly from leaf and microfossil

From: FRANCIS,J. E., PIRRIE, D. & CRAME,J. A. (eds) 2006. Cretaceous-TertiaryHigh-Latitude Palaeoenvironments, James Ross Basin, Antarctica. Geological Society, London, Special Publications, 258, 63-81.0305-8719/06/$15 © The Geological Society of London 2006. 64 I. POOLE & D. J. CANTRILL evidence. However, the paucity of solid taxo- nomic investigations has severely limited the ,oow LI ° °Wo KG I ability to utilize this source of information. This ,-, paper synthesizes the recent data on the Creta- ceous and Cenozoic of the Antarctic Peninsula region, based largely on wood records from the James Ross Island Basin (e.g. Gothan 1908; 1 Torres et al. 1994a, b; Poole & Francis 1999, 2000; Poole et al. 2000a-c; Poole 2002). To help complete the picture, the fossil wood record from elsewhere in the Antarctic Peninsula (e.g. Torres & Lemoigne 1988, 1989; Chapman & Smellie 1992; Falcon-Lang & Cantrill 2000, 65% 2001a; Poole & Cantrill 2001; Poole et al. 2001, 2003) has been used to fill the gaps in the James Adeliad Ross Island Basin biodiversity record.

Geological setting The Antarctic Peninsula is the remnant of a continental-margin magmatic arc of Mesozoic- Cenozoic age (Storey & Garrett 1985). Formed as a result of subduction of the palaeo-Pacific Plate beneath the western margin of the Antarc- tic Peninsula, it has good exposures of 70% magmatic-arc, accretionary complex, and fore- and back-arc regions. These provide infor- mation on a diversity of environmental settings, many of which contain records of fossil wood. Forearc deposits exposed on the west side of the Antarctic Peninsula include the famous standing forests and associated leaf floras from Alexander Island (Jefferson 1982; Falcon-Lang & Cantrill 2001a). These are distal to the arc, swort .an0 o and record braided and meandering fluvial environments on a narrow coastal plain (Cantrill & Nichols 1996; Nichols & Cantrill 2002) (Fig. 1). Further north, in the South Shetland Islands, Early Cretaceous (Aptian) intra-arc deposits record the burial of existing topography through the development of local calc-alkaline volcanic edifices (Hathway 1997) (Fig. 1). Fossil wood from palaeotopographic surfaces and entombed in ignimbrite flows record the interaction between vegetation and environmental processes (Falcon-Lang & Cantrill 2002). Similar volcanic-dominated

Fig. 1. (A) General locality map of the Antarctic Peninsula, with locations of places mentioned in the text. LI, Livingston Island; SI, Snow Island; KGI, King George Island; VI, Vega Island; S, Seymour Island; SNI, Snow Hill Island; JRI, James Ross Island; TN, Table Nunatak; AI, Alexander Island. (B) Reconstruction of Late Cretaceous (84 Ma) times showing the high-latitude setting of the palaeofloras, supplied by Dr R. A. Livermore. The dashed line indicates the polar circle. CRETACEOUS AND CENOZOIC ANTARCTIC FLORAS 65 environments extend from the Late Cretaceous and early Tertiary the Antarctic Peninsula to the ?Oligocene-Miocene in the South extended from approximately 68 to 750S, with Shetland Islands and record the development of Alexander Island lying at about 70°S and King vegetation under these conditions. In contrast, George Island in the South Shetlands at 62°S the east side of the Antarctic Peninsula was rifted (Fig. 1). At these palaeolatitudes plant growth from the margin of Gondwana during the was strongly influenced by the unique growing Jurassic. The Early Cretaceous (Aptian)-early conditions not found on Earth today: a strongly Late Cretaceous (Turonian) of the James Ross seasonal polar light regime coupled with green- Basin is characterized by deep submarine fan house conditions. deposits. A major pulse in arc magmatism in the early Coniacian lead to the development of a Floristic composition and turnover complex fan-delta, slope and base-of-slope environment (Hidden Lake Formation: Whitham Basal Cretaceous sedimentary rocks are lacking et al. 2006). Plant fossil material becomes increas- in the James Ross Island Basin, with the oldest ingly common in the Hidden Lake Formation, Cretaceous strata being Aptian or slightly older. and provides a record of vegetation from the The oldest strata include the Lagrelius Point eastern side of the Antarctic Peninsula. (Aptian: Riding et aL 1998), Kotick Point (early This unparalleled record of diverse palaeoen- Albian: Keating et al. 1992; Riding & Crame vironments and associated biota from the Early 2002) and Pedersen Nunatak (Aptian: Hathway Cretaceous to the Eocene was set in the high & Riding 2001) formations (Fig. 2). Unfortu- southern latitudes. During the Late Cretaceous nately the record of plant macrofossils from

Alexander Island Livingston Island King George Island I Larsen Basin

La Meseta Fm Finlandia Formation ~,, Elgar Formation ~a Cross ValleyFm

Sobral Fm ~[ Colbert Formation ~ Lopezde BertadanoFm Walton Formation Snow Hill island Fm StaccatoFormation Monteverdi Formation Santa Marta Fm

Point Beds ~ ~ HiddenLake F~

Whisky Bay Fm

Neptune GlacierFm Kotick Point F ~

Pluto GlacierFm LagreliusPoint Fm Cerro Negro Fm

Spartan GlacierFm Crabeater Point Beds

Chester Cone Fm

President Beaches Fm II

Fig. 2. Stratigraphic summary of key formations and groups for the Cretaceous and Palaeogene of the Antarctic Peninsula (East Antarctica not included). Units with fossil leaf material or wood material are marked. Note that the stratigraphic succession for King George Island is incomplete. 66 I. POOLE & D. J. CANTRILL these units is poor. Lower Cretaceous (pre- South Shetland Islands crops out on Livingston Aptian) sedimentary rocks occur in the Fossil Island (on Byers Peninsula) and Snow Island (at Bluff Group on Alexander Island (Himalia President Head) (Hathway 1997). This unit Ridge and Spartan Glacier formations: Butter- accumulated in a terrestrial intra-arc setting, worth et al. 1988), and on Byers Peninsula, and was dominated by volcanic processes. A Livingston Island (President Beaches and diverse leaf flora occurs in a lacustrine sequence Chester Cone formations: Crame et al. 1993; on President Head (Torres et al. 1997a, b: Duane 1996) (Fig. 2). These strata accumulated Cantril12000). The impression and compression in marine settings, generally lack wood, and floras from the lacustrine unit are rich in contain only sparse foliage remains. Con- Bennettitales, and other gymnosper- sequently, this discussion starts in the Aptian mous (e.g. Pachypteris Brongniart), with interval when fossil wood becomes more a minor fern and bryophyte component common and terrestrial settings are better docu- (Fig. 3C). The coniferous component is domi- mented. nated by podocarps and taxodiaceous forms with rare araucarian remains (Fig. 3D). This leaf Aptian and Albian interval flora is similar in composition to those found within the non-marine units on Byers Peninsula The basal Gustav Group, of the James Ross (Hernfindez & Azcarfirte 1971; Cesari et al. Island Basin, yields sparse fossil wood (Francis 1998,1999), except that the latter are usually less 1986; Ottone & Medina 1998; Francis & Poole diverse with just one or two taxa. In particular, 2002), and only wood from the early Albian Bennettitales (Ptilophyllurn Morris) (Fig. 3C) Kotick Point Formation has been formally are encountered more frequently and the floras described (Ottone & Medina 1998). This are richer in araucarian elements in Byers material was assigned to Agathoxylon Hartig, a Peninsula localities. member of the Araucariaceae. It is unclear if Palynofloras from the same units record only this wood is contemporaneous with deposition, minor differences between the lacustrine unit or reworked from older formations (Botany assemblage (Cantrill 2000) and those in more Bay Group). However, most of the wood fully terrestrial environments (Hathway et al. preserved in the Gustav Group and overlying 1999). The palynoflora of the lacustrine environ- Marambio and Seymour Island groups is ment is typically rich in pteridophyte spores, replaced by calcite, whereas Botany Bay Group with diversity greater than that seen in the wood is silicified. The limited number of macrofossil record (Cantrill 2000). Localities samples from the Aptian and Albian within the are often dominated by single taxa, in particular James Ross Island Basin precludes making firm Cyathidites Couper and less frequently inferences about forest composition, but, fortu- Cyatheacidites Cookson ex Potoni6 (Lopho- nately, good Aptian wood floras in the South soriaceae). This is thought to reflect local Shetland Islands (Cerro Negro Formation: abundance or colonization events, as the Torres et al. 1982,1997a; Falcon-Lang & Cantrill Lophosoriaceae is a colonizer of disturbed 2001b), and Albian wood floras from Alexander environments (Cantrill 1998). Lycophytes and Island (Falcon-Lang & Cantrill 2001a) have bryophytes make up only a small component of been described. Both floras provide a picture of the flora, whereas conifers tend to be more the forest component of the vegetation to both abundant. pollen is plentiful and domi- the north and south of the James Ross Island nated by podocarps, although yields are lower in Basin. The general similarity in the composition the lacustrine units. This could be the result of of these floras enables us to assume that the taphonomic bias, as bisaccate grains tend to float vegetation within the James Ross Island Basin and so do not readily become incorporated into was comparable. Therefore, in order to set the lacustrine units. scene for ensuing vegetational changes, we Relative to the leaf floras, wood is more wide- consider the floras of these two regions during spread and locally abundant within the Cerro an interval that within the James Ross Island Negro Formation, which complements and Basin presently lacks data. supplements the vegetational picture obtained The Aptian Cerro Negro Formation in the from the palyno- and leaf floras. Palaeovalleys

Fig. 3. Aptian and Albian floras. (A) and (B) are late Albian Triton Point Formation flora, Alexander Island. (A) Standing tree with a pseudomonopodial habit. (B) Tetragleichenites acuta Nagalingum et Cantrill a gleicheniaceous fern. Scale bar shows 1 mm divisions, KG. 2817.75a/76a. (C) and (D) are Aptian flora from Snow Island, South Shetland Islands. (C) Ptilophyllum Menendez Cantrill. Scale bar shows 1 cm divisions, R 2501.1. (D) Elatocladus sp., a podocarpaceous conifer. Scale bar shows 1 cm, R 2501.21a. I i

I 68 I. POOLE & D. J. CANTRILL contain incipient soils, and trunk wood is scat- wood. Alternatively, the best-preserved forest tered across this surface (Hathway 1997; Falcon- horizons may be in sedimentary environments Lang & Cantrill 2001b). The that forested that lack araucarians and thus account for the this region were undoubtedly substantial, as deficit of araucarians in the wood flora. Typi- evidenced by trunks up to 1.5 m in diameter. cally, the araucarians are found in more Ignimbrite flows on Byers Peninsula contain proximal settings, in contrast to the podocarps entrained charred logs up to 5 m in length repre- and taxodiaceous conifers (Cantrill & Falcon- senting the remains of forest stands that grew on Lang 2001). The lack of angiosperm wood in the flanks of a volcanic edifice and which later Albian wood floras from Alexander Island became entombed during a subsequent (Falcon-Lang & Cantrill 2000) and the James eruption. Similar processes have been observed Ross Island Basin (Ottone & Medina 1998) also today in (Clarkson et aL 1988). supports a more shrubby, herbaceous habit of The leaf floras suggest that the Araucariaceae these first angiosperms. were important, but the wood record points to Inferences pertaining to the physical structure forest communities rich in podocarps (61%) of the vegetation that dominated the high with subsidiary araucarians (27%) alongside southern latitudes at this time have been drawn extinct groups such as Sahnioxylon Bose et Sah from in situ 'forests' using stump diameter and (12%) (Falcon-Lang & Cantrill 2001b; n = 33). density, and growth-ring sequences (e.g. Angiosperms first appear as pollen records in Chaloner & Creber 1989; Falcon-Lang & early Albian strata on the eastern side of the Cantrill 2000). Alexander Island provides an Antarctic Peninsula (Dettmann & Thomson ideal case study for such a fossil forest and has 1987), but do not appear in the leaf floras until been studied for many decades (e.g. Jefferson the late Albian (Cantrill & Nichols 1996). The 1981, 1982; Cantrill & Nichols 1996; Falcon- early pollen record is of a low-diversity Lang et al. 2001; Howe & Cantrill 2001). This angiosperm component (mostly Clavatipollen- forest comprised stumps and standing trees of ites Couper) indicative of a shrubby habit up to 8 m tall that have been observed in cliff (Dettmann 1989). The late Albian leaf floras are sections (Cantrill 2001a). Using the allometric also characterized by herbaceous (e.g. Hydro- approach of Niklas (1994), it has been suggested cotyllophyllum Teixeria) and shrubby (e.g. that actual heights of 29 m were attained Dicotylophyllum Bandulska) forms, but a few (Falcon-Lang & Cantrill 2000) by these trees leaf types probably represent more substantial which reached ages of more than 180 years plants, perhaps understorey trees (e.g. Araliae- (Chapman 1994). The habit of these trees do not phyllurn Ettingshausen, Ficophyllum Fontaine) appear to be cone-shaped (Brodribb & Hill (Cantrill & Nichols 1996). 2004), as suggested in earlier publications (e.g. The late Albian flora of Alexander Island, Chaloner & Creber 1989), as no evidence of situated at a palaeolatitude of 75°S, represents whorled branch insertion have been found at one of the most complete and most southerly this time (Cantrill 2001a). Where branching has forests known to date (Fig. 3A). Within the leaf been observed it appears to be towards a flora, gymnosperms dominate with arboreal pseudomonopodial habit and this is more elements such as Araucariaceae, subsidiary consistent with the habit of extant podocarps. Podocarpaceae, minor Taxodiaceae, and an Stand density ranges from a median of 568 understorey composed of ginkopsids, stems per hectare (perhaps representing taeniopterids, bennettitaleans, equisetites, ferns colonization stands: Falcon-Lang et al. 2001) to (Fig. 3B) and liverworts (Cantril12001a; Falcon- dispersed clumps of individuals similar to those Lang et al. 2001; Howe & Cantrill 2001). This in open woodland today (Cantrill 2001a). Such flora also records the first appearance of a maximum density would have ensured the angiosperms in the leaf record (Cantritl & minimization of mutual shading due to the low- Nichols 1996). The famous standing forest angle radiation. The productivity of these horizons contain abundant wood and, although forests has been thought to be as high as approx- often poorly preserved, are rich in podocarps imately 17.65 m 3 ha q a q (Creber & Francis (85.3%), with fewer araucarians (13.2%) and 1999) based on Jefferson's (1981) data, but more taxodiaceous (1.5%) conifers (Falcon-Lang & recent studies based on revised data and Cantril12000; n = 69). The discrepancy concern- additional fieldwork suggest that this is a drastic ing araucarian dominance may be due to the over-representation with high productivity robust nature of araucarian foliage and repro- probably being closer to 5.62-7.33 m 3 ha q a -I, ductive structures (cone scales) having survived similar to the warm temperate araucarian- destructive taphonomic processes better than podocarp stands in North Island, New Zealand podocarps, and thus tending to be over- today (Falcon-Lang et al. 2001). represented in the leaf record relative to the The question of evergreen v. deciduous CRETACEOUS AND CENOZOIC ANTARCTIC FLORAS 69 habit for late Albian Antarctic conifers was outcrop (Brandy Bay-Whisky Bay area and Gin investigated in detail by Falcon-Lang & Cantrill Cove-Rum Cove area). However, the complex (2001b). Based on five independent techniques, lateral variation makes it difficult to correlate they concluded that the canopy-forming veg- between outcrops and between the threefold etation that included conifers, ginkgos and subdivision in different parts of the basin taeniopterids was predominantly evergreen. (Riding & Crame 2002). Furthermore, consider- Araucarians and podocarps, which dominated able problems still exist in identifying and the vegetation, held on to their for at defining Cenomanian strata within the Whisky least 5-13 years, whereas some of the rarer Bay Formation (Riding & Crame 2002). Marine taxodiaceous conifers were evergreen but with invertebrate assemblages of Cenomanian age much shorter leaf retention times. Other taxo- co-occur with palynofloras that are best diaceous conifers, ginkgos and taeniopterids assigned to the late Albian (Riding & Crame were all deciduous along with the fern and 2002). No detailed collecting for fossil wood has angiosperm components of the understorey yet taken place, but a few terrestrial microfloras (Cantrill & Nichols 1996). (e.g. Dettmann & Thomson 1987; D 3057.3) Without preserved in situ stumps, estimates of point to low-angiosperm diversity (Dettmann general density, productivity and height of the 1989; Askin 1992). In the Cenomanian the vegetation cannot be obtained. Therefore, angiosperm component of the vegetation extrapolation from the conclusions drawn by probably still comprised largely shrubby, herb- Falcon-Lang et al. (2001) suggest that the earlier aceous forms, as evidenced by the deficit of mid-Cretaceous forests bordering the James angiosperm wood in deposits older than approx- Ross Island Basin had an estimated density and imately 90 Ma (Fig. 4) (Cantrill & Poole 2002). productivity similar to the Alexander Island The overlying Coniacian Hidden Lake forest. Nevertheless, with the increase in the Formation is rich in angiosperm leaf material abundance of the angiosperm component the (Hayes 1996, 1999; Hayes et al. 2006) and density, productivity and forest heights would contains angiospermous fossil wood, attesting have changed due to different growth rates to increasing dominance and radiation into between angiosperms and conifers, and the canopy niches by this time. The exact timing of influences they exerted on one another. Using the radiation into the canopy is problematic growth-ring and palaeotemperature determina- because of the paucity of data from the Ceno- tions, Specht et al. (1992) concluded that the manian and Turonian. Nevertheless, the fossil productivity of the angiosperm-dominated record documents an increase in angiosperm vegetation of the Peninsula region during the floral diversity throughout the mid-Late Creta- Late Cretaceous would have remained high. ceous at the expense of ferns and lycophytes In summary, plant fossil evidence from wood, initially, and subsequently bryophytes/hepato- leaf and pollen floras indicates the Aptian and phytes, bennettites and other gymnosperms Albian overstorey was dominated by conifers, (Cantrill & Poole 2002). Evidence suggests that and the understorey dominated by ginkopsids, initially these early angiosperms were coloniz- taeniopterids, bennettitaleans, equisetites, ferns ers occupying under- and middle-storey niches and liverworts with a very minor angiospermous and only later expanding to become more component. Such open-canopied, evergreen arborescent and invading the overstorey. This araucarian-podocapaceous conifer forests were invasion would have drastically changed the characteristic of the mid-Cretaceous Pacific prevailing landscape and thus the ecosystem. margins of the Gondwanan continent (Falcon- For example, this change enabled ferns to Lang & Cantrill 2000) extending from Alexan- colonize the new niches created by the der Island, probably across the James Ross angiosperms and thus began a fern rediversifi- Island Basin, to the South Shetland Islands and cation (Cantrill & Poole 2002). Important as far north as and as far east as New palaeobotanical information on the vegetation Zealand. during the Coniacian is derived from the Williams Point Beds on Livingston Island, Cenornanian-Santonian interval dated as 88 Ma (R. Hunt pers. comm. 1999), lying at a palaeolatitude of approximately 62°S The plant fossil record from the Cenomanian (Grunow et al. 1991). The wood, leaves and and Turonian are poorly known in the Antarc- palynofloras are well preserved, unlike many tic Peninsula. The Whisky Bay Formation (late sites around the Antarctic Peninsula where the Albian-latest Turonian: Riding & Crame 2002) palynomorphs are highly degraded or is the only unit where strata of this age occur in destroyed and the leaves are merely impres- the James Ross Basin, where it is divided into sions and lack cuticle (Chapman & Smellie three members in each of the two main areas of 1992). Both macrofossil and microfossil 70 I. POOLE & D. J. CANTRILL

° ~ o ~

o

n ~

~ ~ ~ ~.~ ~ =: ~ .

E E Eocene ~- E [ II l'l, E Paleocene m ii~i~:~!J::~2~!~i!;~!~!ii~!~::~!~:~:i;~;~!~:~!~`~::!~i~!:~`~'i!!~:!~i!~i~;i~!:@!~!!!i~!!~ii::i:~i:iii!ii;i!i~!i;!i!iiii~!i!:iii:~i~!: i;ii:i:iii%:iiii!iiii~i~i~!:~ii~i~!;iii~ii~i~;~!iiiii;!i~;;!:!i!i~ilil iii~!i:!~!¸ !i':i::;i~;iii?:!;i ¸ ~i: !;5i ¸ A?:II!L/'?:i:~iii:iii~iiiii~;~!!i:::::!i i;- i:.ii~i::!i;i:i!i;i ~!i;iii!i~;:;ii ¸ :i~i~!!~ili:;i!~iii::il;;i;!:ii~i:iii Maastrichtian I, I II1' Campanian II

Santonian Coniacian IIIII Turonian 7 7 7 7 7 7 7

Cenomanian

Fig. 4. Biostratigraphic range chart for angiosperm wood taxa (adapted from Cantrill & Poole 2005). evidence from the Williams Point Beds suggest would have occupied moist shady banks, damp that the vegetation comprised conifer forest rocks, and trunks and branches of trees (Poole with a diverse angiosperm component. It still & Cantril12001). Unfortunately, this flora is still remains uncertain as to which group of plants in need of a thorough revision as it was initially dominated the canopy (Chapman & Smellie assigned a Triassic age (e.g. Orlando 1968), and 1992; Poole & Cantrill 2001). Preliminary esti- the diverse leaf flora, representing many groups mates of wood abundance from this locality of non-woody plants, were assigned to Triassic suggest that between 67 and 81% of the organ taxa. A revision of the conifer and samples are coniferous (Philippe et al. 1993; n angiosperm wood component of the flora has = 270). Ferns, represented by leaves and petri- already been undertaken by Poole & Cantrill fied rachides, are abundant and diverse with (2001). They found that the size of the material large tree ferns and ?bennettites forming part indicates that the woody angiosperm taxa were of the arboreal component (Chapman & from large trees. The coniferous element Smellie 1992; Poole & Cantrill 2001). Under- comprised of Araucariaceae and Podo- storey fern taxa and Equisetites (Lacey & carpaceae, whereas the angiosperms include Lucas 1981) probably inhabited the moist forest two assigned to the Monimiaceae and floor and banks of streams. Thalloid liverworts Cunoniaceae and four other taxa with affinities (Lacey & Lucas 1981) and epiphytic ferns that lie with the 'Magnoliidae, Hamamelidae CRETACEOUS AND CENOZOIC ANTARCTIC FLORAS 71 and Rosidae' and were assigned to Antarctoxy- tion. This is supported by the lack of wood from lon Poole et Cantrill, an organ for fossil these groups. of equivocal taxonomic affinity (Poole & Cantrill 2001). The early woody-arboreal Campanian-Maastrichtian interval angiosperm component shows no evidence of distinct growth rings, suggesting that these Between the Santonian and Campanian there is plants may have been evergreen. With an an unmistakable turnover in pollen taxa that is increase in the abundance of distinct growth also reflected in the wood (Fig. 4) and other rings after this time, it is probable that the macro-floras (Dettmann 1989; Askin 1992). angiosperms only later adapted to the seasonal Within the angiosperm wood flora the more Antarctic environment in becoming predomi- 'primitive' wood types with affinities to the nantly deciduous (Cantrill & Poole 2005). 'Magnoliidae, Hamamelidae and Rosidae' Another slightly younger Coniacian leaf flora (Poole & Cantril12001) do not continue through occurs in the Hidden Lake Formation on James to the Campanian. The only primitive types of Ross Island (Fig. 5E-G). Although no cuticles wood occurring in the Coniacian flora, that also are preserved, thus hindering systematic place- appear in the Campanian, are the cunoniaceous ment, Hayes (1996, 1999) concluded that the Weinmannioxylon Petriella and the monimia- dominant leaf form in this flora shows great ceous Hedycaryoxylon Stiss (but these xylo- similarity to the Magnoliales with a strong types, W. ackamoides Poole et Cantrill and H. component of sterculiaceous (Fig. 5E) and tambourissoides Poole et Gottwald, both disap- lauralean (Fig. 5F) forms, indicating that pear during the Campanian). Hedycaryoxylon is angiosperm diversity was well underway by this replaced by other anatomically different, time. monimiaceous taxa, whereas the Weinman- Indeed, angiosperms had become dominant nioxylon nordenskjoeldii Poole et al. xylotype in the Antarctic macrofloras by the Coniacian continues through until the Eocene (Fig. 4). (Cantrill & Poole 2002). The increased abun- Perhaps the most dramatic change is the dance of wood fragments suggests that the increase in importance of the Nothfagaceae. The angiosperms were no longer herbaceous, Nothofagaceae is a typical, and one of the most shrubby understorey elements but had become important elements, of the relictual Southern a more important component of the canopy. The Hemisphere temperate floras today (Manos Santonian leaf floras still suggest a strong ster- 1997). Nothofagaceae in the Antarctic pollen culiaceous and lauralean component (Hayes floras is first represented by the ancestral 1996, 1999), alongside woods with lauraceous Nothofagidites senectus Dettmann et Playford, (Poole et al. 2000c), cunoniaceous (Poole et al. notably distinct from extant Nothofagus Blume 2000a), illiciaceous (Poole et al. 2000b), athero- pollen (Dettmann et al. 1990). Importantly, spermataceous (Poole & Francis 1999; Poole & Nothofagus pollen is extremely common and its Gottwald 2001) and winteraceous (Poole & absence can be regarded as evidence for the lack Francis 2000) affinity (Fig. 4). All these of the genus in areas where it has not been elements, with the exception of the illiciaceous, found (Swenson & Hill 2001). This family is only lauraceous and Sassafrasoxylon Poole et al., rarely recorded prior to the Campanian, as have adisjunct distribution between North pollen analyses from Santonian strata (Baldoni America and Asia, and are characteristic of & Medina 1989; Keating 1992). Despite the cooler temperate biomes. The presence of the sparse occurrences in the Santonian no wood is illiciaceous and sassafrasaceous elements could known until the Campanian when this group suggest a somewhat warmer temperate biome, becomes important in both the fossil wood and or perhaps a more recent adaptation of these pollen record (Poole 2002). These changes are plants to warmer latitudes. supported by an increase in abundance and/or Precursors to the changes in vegetation seen diversity of other taxa including pollen of in the Campanian-Maastrichtian are present in Dacrydium Sol. Ex G. Forst., Proteaceae and the Coniacian and Santonian. The first records Myrtaceae, and a variety of other angiosperms of Nothofagidites Erdtman ex Potoni6 (Baldoni also evidence the changes occurring at, or just & Medina 1989; Keating 1992), Proteacidites prior to, the mid-Campanian (Dettmann & Cookson ex Couper (Dettmann & Thomson Thomson 1987). 1987; Baldoni & Medina 1989; Barrera et al. In the Maastrichtian of Vega and Seymour 1999) and Myrtaceae (Dettmann & Thomson islands on the Antarctic Peninsula the oldest 1987; Baldoni & Medina 1989) occur, although known occurrences of Nothofagus subgenera the rare occurrence of these grains indicates the Fuscospora, Lophozonia and Brassospora have plants were of minor importance in the vegeta- been found in the pollen record (Dettmann et © © CRETACEOUS AND CENOZOIC ANTARCTIC FLORAS 73 al. 1990) alongside wood of the Fuscospora type Manchester 2002). The woods probably resided, (Poole 2002) (Fig. 4). Manos (1997) and for a time at least, in a damp, but not water- Swenson et al. (2000) recognize the subgenus logged, aerobic environment. Moist conditions Lophozonia as the most basal clade within probably prevailed on the magmatic arc Nothofagaceae, with the pollen record extend- adjacent to the James Ross Island Basin, ing back to the late Campanian in Antarctica encouraging fungal growth within the woody (Dettmann et al. 1990) and appearing only debris covering the forest floor. This woody slightly later in the wood flora (Poole 2002). By debris was probably then transported, by the Maastrichtian all but the Brassospora streams or rivers during times of storms or flash subgenus are represented in the wood flora, floods, to the coast where the material was whereas all four subgenera are represented in finally deposited offshore in the shallow-marine the pollen flora. The lack of wood allied to Bras- environment. sospora is not surprising as this group, although At this time podocarps and angiosperms were a copious pollen producer today, is often the main canopy elements of the perhumid, tall confined to montane areas. We can confidently open forests (Askin 1988; Specht et al. 1992). assume that by the mid-Campanian the Palynofloras suggest the Myrtaceae were Nothofagaceae had now become an important present along with diverse species of Notho- and diverse component of the Antarctic ecosys- fagus, Gunnera L., Proteaceae, Aquifoliaceae, tem having substantially changed the face of the Olacaceae, Loranthaceae and Sapindaceae flora relative to the ancestral Coniacian- accompanied by a rich fern component includ- Santonian vegetation. ing the Osmundaceae and Gleicheniaceae (e.g. By the Maastrichtian the wood floras Askin 1989,1990). At the end of the Maastricht- continued to increase in abundance and diver- ian the warmer temperate elements, Sassafra- sity, with other angiospermous taxa appearing soxylon and Illiciaceae, disappeared (Fig. 4), and Nothofagaceae diversifying further (Poole whilst the abundance of fern taxa also decreased et al. 2003). Wood anatomical characters, (Askin 1988, 1990), possibly in response to a coupled with good preservation of these now cooling in the climate, leaving a floral composi- prolific angiosperms, provide us with important tion very similar to that which occurs today in indications pertaining to the local environment, the cool temperate rainforests of South as well as the vegetational composition and America. In previous publications (Poole et al. dynamics. The Antarctic material is comprised 2001, 2003) the vegetational composition of the predominantly of aerial organs (branches or Antarctic Peninsula during the Late Cretaceous trunks - only one specimen could be confidently and early Tertiary has been likened to the extant identified as having a root origin), possibly low- to mid-altitudinal Valdivian rainforests dominated by sap-wood remains (cf. Wheeler & under the 'Valdivian model'. Indeed the similar- Manchester 2002 and their inferences drawn ities in terms of environmental dynamics in from the Eocene Clarno Formation woods). The addition to vegetational composition are dominance of sap wood is evidenced from remarkably similar. Along the Andean margin preserved fungal hyphae associated with the ray of South America stratovolcanic activity is a parenchyma and vessel lumens or tracheids, major source of disturbance along with associ- suggesting that these fungi are sap-staining ated events such as landslides, earthquakes and fungi restricted to sap wood. The detachment the flooding of lake systems. Glacial erosion and from the parent plant may have occurred during deposition contribute to the general disturbance the Antarctic polar winter or at least during the especially at altitude. The Antarctic Peninsula dormant phase of the year as many of the region during the Cretaceous and Tertiary was angiosperm woods have vessels filled with well- a vegetated active volcanic arc and thus would developed tyloses, but with no evidence of bud- have been subject to similar ecological distur- like outgrowths of parenchyma into the vessel bances. There is no doubt that climate change lumen, which mark the beginning of tylose influenced vegetational composition, although formation (cf. Chattaway 1949; Wheeler & evolving palaeoenvironments in the Antarctic

Fig. 5. Late Cretaceous and Cenozoic floras. (A)-(D) Palaeogene plant fossils from the James Ross Island Basin. (A) Araucaria marenssi Cantrill et Poole, a petrified araucarian trunk from the Eocene La Meseta Formation, DJ.1057.53. (B)-(D) Leaf fossil from the Palaeocene Cross Valley Formation. (B) Araucaria nathorstii, DJ.1111.14. (C) Angiosperm leaf, D.523.1. (D) Fern frond, DJ.1113.134. (E)-(G) Coniacian leaf fossil from the Hidden Lake Formation. (E) sterculiaceous taxon, D.8754.8.1a. (F) lauraceous taxon, D.8754.8.57a. (G) Unidentified angiosperm, D.8754.45a. Scale bar divisions are 1 cm. 74 I. POOLE & D. J. CANTRILL

Peninsula region probably contributed in equal, Higher up the sequence and in more distal if not greater, importance. One good example of locations floral diversity increases with pterido- environmental dynamics, rather than climate, phytes (such as Cyatheaceae, Dicksoniaceae accounting for vegetation change can be seen in and Osmundaceae), conifers (including Arau- Late Cretaceous-Palaeocene strata on King cariaceae, and Podocarpaceae) George Island lying to the north of the James and angiosperms (e.g. Nothofagaceae, Ross Island Basin. The Late Cretaceous-Early Proteaceae, Myrtaceae, ?Araliaceae, Anacar- Palaeocene strata is lava-dominated, suggesting diaceae and Cunoniaceae) representing low that stratovolcanic activity was high and the levels of disturbance, i.e. climax or pre-eruption centre of activity was proximal (Smellie et al. vegetation. Interestingly, there are no records 1984; Shen 1994, 1999; Orton 1996). By the of Chusquea Kunth. (bamboo), a characteristic Palaeocene epiclastic deposits including tuffa- understorey element of the Valdivian ecosys- ceous rocks indicate that the volcanic dynamics tem today. Poole et al. (2001) have suggested had changed, lowering the rate of disturbance. that ferns, ubiquitous at this time (Zhou & Li Late Cretaceous floras of this region also 1994; Mohr 2001), may have filled this niche in document ecological disturbances: low-density these Antarctic forests. By the Late Eocene, vegetation is dominated by ferns such as Thyr- angiosperm leaf floras at a palaeolatitude of sopteris Kunze and Dacrydium, and a relatively approximately 62°S show evidence of both small percentage of angiosperms with notho- deciduous and possible evergreen habits (Hunt fagaceous, sterculiaceous, lauraceous and myrta- & Poole 2003), whereas the wood floras ceous affinity (Dutra & Batten 2000; Poole et al. suggest an overriding deciduous habit (Poole et 2001), whereas post-disturbance vegetation in al. 2001). proximal volcanic settings is characterized by Although this Valdivian-type ecosystem may gymnosperms such as cycads and podocarps have persisted for longer, into the Late Eocene, (Cao 1992; Dutra & Batten 2000; Poole et al. the sudden widespread glaciation of Antarctica 2001). and the associated shift towards cooler temper- atures at the Eocene-Oligocene boundary (c. 34 Palaeocene-Eocene interval Ma) (DeConto & Pollard 2003) would have had a detrimental effect on the prevailing vegeta- The Valdivian-type ecosystem continues over tion. The cool temperate rainforests would have the Cretaceous-Tertiary boundary, with no become less diverse, possibly becoming more suggestion of catastrophic environmental analogous to the extant Magellanic subpolar devastation, into the Eocene where the last Nothofagus- (evergreen and deciduous) domi- evidence for this Valdivian-type ecosystem can nated forests mixed with conifers (Podocarpus be found in the palynological record of L'Her. Ex Pers., Pilgerodendron Florin) growing Seymour Island (Askin 1997) and the wood south of approximately 47°S across the southern record of King George Island (Poole et al. Aysdn and Magallanes regions of and 2001). Palaeocene and Eocene volcanic activity in (Moore 1983; was still prevalent across the Antarctic Penin- Veblen et al. 1996). Here permanent snow, ice sula with terrestrially deposited fossil floras caps and glaciers are present at altitude, cold (Fig. 5B, D) providing evidence for ecological temperate conditions (MAT 3-6 °C) prevail disturbance similar to that observed in the with high levels of precipitation (MAP Valdivian region today (Hunt 2001; Poole et al. 1000-4000 ram) and strong permanent winds 2001 and references therein; Hunt & Poole (Hoffmann 1975). Indeed, Late Eocene Antarc- 2003). Frequent disturbance in the lower part tic fossil assemblages from Seymour Island and of the sequence on King George Island is char- McMurdo Sound consist of podocarpaceous acterized by low-diversity flora consisting only and araucariaceous conifers, Nothofagus (both of Nothofagus and podocarps overlying a deciduous and evergreen) with at least three coarse volcanic debris flow. Floras more distal other angiosperm types and ferns (Dusdn 1908; from the centre of volcanic activity probably Case 1988; Doktor et al. 1996; Askin 1997, 2000; experienced lower levels of disturbance and Cantrill 2001b; Francis 2000; Pole et al. 2000). this is again supported by the floral composi- tion. The presence of myrtaceous, eucryphia- Neogene ceous and nothofagaceous angiosperms alongside podocarps and Cupressaceae suggest The cooling climate across the Eocene- that the flora was relatively diverse and had Oligocene boundary (e.g. Zachos et al. 2001) is experienced low or relatively moderate levels attested to both by a more fragmentary Antarc- of disturbance (Askin 1997; Poole et al. 2001). tic plant record and also shifts in leaf size classes CRETACEOUS AND CENOZOIC ANTARCTIC FLORAS 75

(Cantril12001b). These changes are also seen in dominated by Nothofagus (N. antarctica (G. the palynological record, with a dramatic Forst.) Oerst. and N. betuloides (Mirb.) Blume), decrease in diversity, occurrence and abundance Drimys J.R. Forst. Et G. Forst., Pilgerodendron of terrestrial palynomorphs (Raine 1998; Askin and occasional Maytenus Molina, elements that & Raine 2000; Raine & Askin 2001). Within the have been associated with past Antarctic vege- James Ross Basin post-Eocene floras are tation. At higher elevations a belt of low lacking, due in part to a lack of strata from the 'krumholz' vegetation comprising Nothofagus latest Eocene until the Miocene (Hobbs Glacier occurs (Moore 1983; Veblen et al. 1996). Formation: Pirrie et a11997; Jonkers et al. 2002). Changes to the Antarctic Neogene vegetation The overlying James Ross Island Volcanic due to cooling of the continent probably Group (late Miocene-Pliocene) and the resulted in a transition of the vegetation from Pliocene Cockburn Island Formation (Jonkers low forest to krumholz forms and ultimately to 1998a, b; Jonkers & Kelley 1998) also lack plant tundra. The exact timing of these transitions is material, and a terrestrial palynoflora has not unclear, but the terrestrial succession from Cape been documented. Elsewhere in the Antarctic Roberts (Raine 1998; Askin & Raine 2000; Peninsula the post-Eocene record of Raine & Askin 2001) shows a general decline in macroplants is contentious. In the South diversity and abundance through the Oligocene Shetland islands the age of a series of Oligocene and into the Miocene, with fewer abundance glacial and interglacials (Birkenmajer 1987, spikes in Nothofagaceae. This suggests that, 1990, 1997) has been questioned (Dingle et al. although the vegetation survived successive 1997; Dingle & Lavelle 1998; Hunt 2001) so that glaciations presumably in refugia, the ability to all the fossil floras are now considered to be recover progressively diminished. The Late Eocene in age (Hunt 2001). What is clear from Pliocene Meyer Desert Formation (Sirius the Oligocene and Miocene strata on King Group) from East Antarctica contains prostrate, George Island (e.g. Polonez Cove Formation gnarled deciduous Nothofagus (as evidenced and Cape Melville Formation), and the Miocene from leaves, wood and pollen from glacial sedi- and Pliocene units in the James Ross Basin ments of the Sirius Formation c. 1.8-5.3 Ma: (Hobbs Glacier Formation, James Ross Island Carlquist 1987; Francis & Hill 1996). Some of Volcanic Group and Cockburn Island Forma- the fossil woods show evidence of traumatic tion), is that conditions for plant growth were events and scarring (Francis & Hill 1996), which extremely unfavourable and a cyclical is also a common feature in other prostrate glacial-interglacial environment prevailed. species, such as the magellanic Nothofagus, to The impact of this climate regime on plant which the habit of these Nothofagus fossils can communities is best seen in a recent series of be likened. Nothofagus probably had a decidu- cores from offshore Cape Roberts in the Ross ous habit with single seasonal leaf falls attested Sea (Raine 1998; Askin & Raine 2000; Raine & to by the strong plicate vernation of fossil Askin 2001). Although sedimentation patterns Nothofagus leaves coupled with dense accumu- have influenced the accumulation of paly- lation of leaves within a single layer (Hill et al. nomorphs, the pattern is one of low diversity 1996). More recent investigations of this flora and abundance interspersed with periods of reveal a surprising floral diversity including higher diversity and abundance presumably in conifers, cushion plants, ranunculids, possible response to cold and warm climate periods. By grasses and chenopods, and a variety of moss the Early Oligocene the had taxa (some cushion forming) growing in a become species depauperate with Nothofagus subglacial tundra-like environment (Hill & and podocarpaceous conifers probably domi- Truswell 1993; Ashworth & Cantril12004). Fossil nating the canopy, and with lycophytes and plant material suggests mean annual tempera- some bryophytes, and ferns contributing to the tures of approximately -20 °C, but possibly understorey (Askin 2000; Pole et al. 2000; colder without snow cover to protect the Cantril12001b; Mohr 2001). The flora has analo- dormant plants, and growing seasons of up to 3 gies with today's Magellanic tundra (c. 48°S and months with temperatures reaching only about the southern tip of Tierra del Fuego and across 5 °C in the growing season (Francis & Hill 1996; the Chilean archipelago) which supports Hill & Jordan 1996). This evidence lends further predominantly bog communities (Moore 1983; support to a magellanic subpolar forest-tundra- Ruthsatz & Villagran 1991) under a regime of type ecosystem. high precipitation (2000-6000 mm year -1) and The change from true magellanic subpolar low temperatures (MAT 5-6 °C). In some shel- forest to true tundra in Antarctica at the end of tered, less unfavourable areas of this tundra the Tertiary would probably have been small stands of evergreen forest can be found punctuated rather than gradual with the flora 76 I. POOLE & D. J. CANTRILL responding to pulses of glaciations, surviving in similarity with the Vadivian rainforests of refugia during the severest of climates and only Chile was beginning to develop, although making a transitory comeback during the inter- the Antarctic vegetation retained some glacials. Finally, these last remaining plants also more warm temperate taxa such as the succumbed to the deteriorating climates Illiciaceae and Sassafras-like plants. The brought about by declining atmospheric CO2, expansion of Nothofagus and the replace- continental separation and widescale glaciation ment of the warm temperate elements in (Exon et al. 2000; DeConto & Pollard 2003) such the Campanian changed the vegetation that today only two species of , once more. It became similar to the cool Colobanthus quietensis (Kunth) Bartl. and temperate rainforests of Valdivia in South Deschampsia antarctica Desv., along with a few America today (Poole et al. 2001, 2003). species of lichens and mosses, survive in today's The similarities to the Valdivian rainforests extreme southern high-latitude environments. strengthened through the Maastrichtian to become dominant up until the Eocene. Summary Vegetation changes reflected environ- mental disturbances due to volcanic activity Four stages of vegetation development associ- that were occurring at this time. Climate ated with taxonomic turnover and canopy deterioration associated with the onset of development are recognized from the Aptian glaciation in Antarctica led to a further through to the Tertiary across much of the change in the vegetation. Antarctic Peninsula. Diversity decreased towards the Eocene- Oligocene boundary until the forests • Early-mid-Cretaceous vegetation was probably became more similar to the conifer dominated, ranging from forests Magellanic subpolar forests of southern with lower storeys comprising ginkopsids, South America today where communities spehnopsids, ferns, liverworts alongside are dominated by Nothofagus mixed with extinct taxa, such as bennettites and conifers. The flora became more depauper- taeniopterids, to dispersed clumps of indi- ate during the Oligocene and Miocene, such viduals more akin to open woodland. The that by the Miocene-Pliocene it was forests had stand densities of approxi- probably more similar to the Magellanic mately 600 stems per hectare, with pro- tundra of southern South America, support- ductivity estimates and taxon composition ing local stands of Nothofagus and a few similar to the warm temperate araucarian- other angiosperm and conifer taxa. podocarp vegetation of North Island, New Nothofagus decreased in stature as a result Zealand (Falcon-Lang et al. 2001; Falcon- of the deteriorating climate conditions. It is Lang & Cantrill 2001a). The Alexander likely that the vegetation went through a Island floras are the only known in situ krumholtz phase in habit prior to becoming fossil forests to be found in or near the low prostrate shrubs in the Pliocene tundra. James Ross Island Basin and therefore are By the end of the Tertiary, with glaciation taken as a general guide for the ensuing widespread and temperatures plummeting, angiosperm-dominated vegetation of the widespread vegetation had been wiped Peninsula region. from Antarctica such that only two species • Middle-early Late Cretaceous floras reflect of vascular plants occur there today. the arrival of the angiosperms at the expense of the bryophytes/hepatophytes, We are indebted to BAS for the loan of material, and gymnosperms such as bennettites, and, the opportunity for I. Poole and D. J. Cantrill to initially, ferns and lycophytes. The initially undertake fieldwork in Antarctica. They thank herbaceous angiosperms had become Dr R. Hunt and Professor J. Francis for their help arboreal by the Coniacian and invaded the with collecting the fossil material; Dr R Rudell and conifer-dominated canopy, giving rise to Professor D. Cutler for continued access to the slide new understorey niches that ferns and liver- collection at the Jodrell Laboratory, Royal Botanic worts subsequently exploited. By the Gardens Kew; D. Makaham for sectioning modern Santonian angiosperms were an important material for comparisons; and M. Tabecki (BAS) and O. Stiekima (Utrecht University) for sectioning the component of the overstorey, and families fossil material used in this study. Much of original including the Sterculiaceae, Lauraceae, work presented here was made possible through Cunoniaceae, Monimiaceae, Atherosper- NERC funding (grant number GR3/11088) and mataceae and Winteraceae progressively continued with funding from NWO (grant number replaced more 'primitive' forms. Taxonomic ALW/809.32.004). CRETACEOUS AND CENOZOIC ANTARCTIC FLORAS 77

References Shetland Islands (West Antarctica). Zentralblatt for Geologie und Paltiontologie, 1, 141-151. ASHWORTH, A. & CANTRILL,D.J. 2004. Neogene vege- BIRKENMAJER, K. 1997. Tertiary glacial/interglacial tation of the Meyer Desert Formation (Sirius palaeoenvrionments and sea level changes, King Group) Transantarctic Mountains, Antarctica. George Island, West Antarctica. An overview. Palaeogeography, Palaeoclimatology, Palaeoecol- Bulletin of the Polish Academy of Sciences, Earth ogy, 213, 65-82. Sciences, 44, 157-181. ASK~N, R.A. 1988. Campanian to Paleocene palyno- BRODRIBB, T. & HILL, R.S. 2004. The rise and fall of logical succession of Seymour Island and adjacent the Podocarpaceae in Australia - a physiological islands, northeastern Antarctic Peninsula. In: explanation. In: HEMSLEY,A.R. & POOLE, I. (eds) FELDMANN, R.M. & WOODBURNE, M.O. (eds) The Evolution of Plant Physiology. Harcourt, Geology and Paleontology of Seymour Island, London, 381-399. Antarctica Peninsula. Geological Society of BUTTERWORTH, P.J., fRAME, J.A., HOWLETT, P.J. & America, Memoirs, 169, 131-153. MACDONALD, D.I.M. 1988. Lithostratigraphy of ASKIN, R.A. 1989. Endemism and heterochroneity in Upper Jurassic-Lower Cretaceous strata of the Late Cretaceous (Campanian) to Paleocene eastern Alexander Island, Antarctica. Cretaceous palynofloras of Seymour Island, Antarctica: Research, 9, 249-264. implications for origins, dispersal and palaeo- CAYrnmL, D.J. 1998. Early Cretaceous fern foliage climates of southern floras. In: fRAME, J.A. (ed.) from President Head, Snow Island, Antarctica. Origins and Evolution of the Antarctic Biota. AIcheringa, 22, 241-258. Geological Society, London, Special Publications, CANTRILL, D.J. 2000. A Cretaceous macroflora from a 47, 107-119. freshwater lake deposit, President Head, Snow ASKIN, R.A. 1990. Campanian to Paleocene spore and Island, Antarctica. Palaeontographica Abteilung B, pollen assemblages from the upper Campanian 253, 153-191. and Maastrichtian of Seymour Island. Review of CANTmLL, D.J. 2001a. Cretaceous High-latitude Terres- Palaeobotany and Palynology, 65, 105-113. trial Ecosystems: An Example From Alexander ASK1N, R.A. 1992. Late Cretaceous-early Tertiary Island, Antarctica. Asociaci6n Paleontol6gica Antarctic outcrop evidence for past vegetation Argentina, Publicaci6n Especial, 7, 39-44. and climates. In: KENNETr, J.E & WARNKE, D.A. CANTRILL, D.J. 200lb. Early Oligocene Nothofagus (eds) The Antarctic Palaeoenvironment: A Perspec- from CRP-3, Antarctica: Implications for the vege- tive on Global Change. American Geophysical tation history. Terra Antarctica, 8, 401-406. Union, Antarctic Research Series, 56, 61-73. CANTRILL, D.J. & FALCON-LANG, H.J. 2001. Cret- ASKIN, R.A. 1997. Eocene-?earliest Oligocene terres- aceous (Late Albian) Coniferales of Alexander trial palynology of Seymour Island. In: RIcch C.A. Island, Antarctica. 2: Leaves, reproductive struc- (ed.) The Antarctic Region: Geological Evolution tures and roots. Review of Palaeobotany and Paly- and Processes. Terra Antartica Publication, Siena, nology, 115, 119-145. 993-996. CANTRILL, D.J. & NICHOkS, G.J. 1996. and ASKIN, R.A. 2000. Spores and pollen from the palaeoecology of Early Cretaceous (Late Albian) McMurdo Sound erratics, Antarctica In: STmWELL, angiosperm leaves from Alexander Island, Antarc- J.D. & FELOMANN, R.M. (eds) Palaeobiology and tica. Review of Palaeobotany and Palynology, 92, Palaeoenvironments of Eocene Fossiliferous 1-28. Erratics, McMurdo Sound, East Antarctica. CANTRILL, D.J. & POOLE, I. 2002. Cretaceous to American Geophysical Union, Antarctic Research Tertiary patterns of diversity change in the Antarc- Series, 76, 161-181. tic Peninsula. In: OWEN,A.W. & CRAME, J.A. (eds) ASKIN, R.A. & RAINE, J.I. 2000. Oligocene and Early Palaeobiogeography and Biodiversity Change: A Miocene terrestrial palynologyof Cape Roberts Comparison of the Ordovician and Mesozoic- Drillhole CRP-2/2A, Victoria Land basin, Antarc- Cenozoic Radiations. Geological Society, London, tica. Terra Antarctica, 7, 493-501. Special Publications, 194, 141-152. BALDONI, A.M. & MEDrNA, F. 1989. Fauna y CANTRILL D.J. & POOLE, I. 2005. Taxonomic turnover microflora del Cretficico, en bahfa Brandy, isla in wood floras from the Cretaceous and Tertiary of James Ross, Antfirtida. Instituto Antdtrtico Chileno Antarctica: implications for changes in forest Serie Cientifica, 39, 43-58. ecology. Palaeogeography, Palaeoecology, Palaeo- BARRERA, V., PALAMARCZUK, S. & MEDINA, E 1999. climatology, 215,205-219. Palinologfa de la formaci6n Hidden Lake CAO, L. 1992. Late Cretaceous and Eocene palynoflo- (Coniaciano-Santoniano), isla James Ross, Antfir- ras from Fildes Peninsula, King George Island tida. Rev&ta EspanOla de Micropaleontologia, 31, (South Shetland Islands), Antarctica. In: YOSH~DA, 53-72. Y. (ed.) Recent Progress in Antarctic Earth Science. BmKENMAJER, K. 1987. Oligocene-Miocene glacio- Terra Scientific (Terrapus), Tokyo, 363-369. marine sequences of King George Island (South CASE J. 1988. Paleogene floras from Seymour Island, Shetland Islands) Antarctica. Palaeontologia Antarctic Peninsula. In: FELDMANN, R.M. & Polonica, 49, 9-36. WOODI3URNE, M.O. (eds) Geology and Paleontol- BmKENMAJER, K. 1990. Geochronology and ogy of Seymour Island, Antarctica Peninsula. climatostratigraphy of Tertiary glacial and inter- Geological Society of America, Memoirs, 169, glacial successions on King George Island, South 523-530. 78 I. POOLE & D. J. CANTRILL

CARLQUIST, S. 1987. Pliocene Nothofagus wood from stratigraphy. Journal of the Geological Society, the Transantarctic Mountains. Aliso, 11, 571-583. London, 154, 257-264. CESARI, S.N., PARCIA, C.A., REMESAL, M.B. & SALANI, DOKTOR, M., GAZDZICKI, A., JERMANSKA, A., EN. 1999. First evidence of Pentoxylales in Antarc- POREBSKI, S.J. & ZASTAWNIAK,E. 1996. A plant and tica. Cretaceous Research, 19, 733-743. fish assemblage from the Eocene La Meseta CESARI, S.N., PARCIA, C.A., REMESAL,M.B. & SALANI, Formation of Seymour Island (Antarctic Penin- EN. 1999. Paleoflora del Cret~icico Inferior de la sula) and its environmental implications. Palaeon- peninsula Byers, islas Shetland del Sur, Ant~irtida. tologia Polonica, 55, 127-146. Ameghiniana, 36, 3-22. DUANE, A.M. 1996. Palynology of the Byers Group CHALONER, W.G. & CREBER, G.T. 1989. The phenom- (Late Jurassic-Early Cretaceous) of Livingston enon of forest growth in Antarctica: a review. In: and Snow islands, Antarctic Peninsula: its bio- CRANE, J.A. (ed.) Origins and Evolution of the stratigraphical and palaeoenvironmental signifi- Antarctic Biota. Geological Society, London, cance. Review of Palaeobotany and Palynology, 91, Special Publications, 47, 85-88. 241-281. CHAPMAN,J.L. 1994. Distinguishing internal develop- DUS~N, R 1908. Ober die terti~ire Flora der Seymour- mental characteristics from external palaeoen- Insel. Wissenscha[tliche ergebnisse Der Schwedi- vironmental effects in fossil wood. Review of schen Si~dpolar-expedition 1901-1903, 3, (3), 1-27. Palaeobotany and Palynology, 92, 1-28. DRINNAN, A.N. & CRANE, P.R. 1990. Cretaceous Pale- CHAPMAN, J.L. & SMELLIE,J.L. 1992. Cretaceous fossil obotany and its bearing on the biogeography of wood and palynomorphs from Williams Point, austral Angiosperms. In: TAYLOR, T.N. & TAYLOR, Livingston Island, Antarctic Peninsula. Review of E.L. (eds) Antarctic Paleobiology: Its Role in the Palaeobotany and Palynology, 74, 163-192. Reconstruction of Gondwana. Springer, New York, CHATTAWAY,M.M. 1949. The development of tyloses 192-219. and secretion of gum in heartwood formation. DUTRA, T. & BATTEN, D. 2000. Upper Cretaceous Australian Journal of Science, 2, 227-240. floras of King George Island, West Antarctica and CLARKSON, B.R., PATEL, R.N. & CLARKSON,B.D. 1988. their palaeoenvironmental and phytogeographic Composition and structure of a forest overwhelmed implications. Cretaceous Research, 21, 181-209. at Pureora, central North Island, New Zealand, EIGHTS, J. 1833. Description of new crustaceous during the Taupo eruption (c. AD 130). Journal of animal found on the shores of South Shetland the Royal Society of New Zealand, 18, 417-436. Islands, with remarks on their natural history. CRANE, J.A., PIRRIE, D., CRAMPTON, J.S. & DUANE, Transactions of the Albany Institute, 2, 53-69. A.M. 1993. Stratigraphy and regional significance EKLUND, H. 2003. First Cretaceous flowers from of the Upper Jurassic-Lower Cretaceous Byers Antarctica. Review of Palaeobotany and Palynol- Group, Livingston Island, Antarctica. Journal of ogy, 256, 1-31. the Geological Society, London, 150, 1075-1087. EXON, N., KENNETT, J., MALONE, M. & THE LEG 189 CREBER, G.T. & FRANCIS, J.g. 1999. Fossil tree-ring SHIPBOARD SCIENTIFIC PARTY. 2000. The opening analysis: palaeodendrology. In: JONES, T.E & of the Tasmanian gateway drove global Cenozoic ROWE, N.R (eds) Fossil Plants and Spores Modern paleoclimatic and paleoceanographic changes: Techniques. Geological Society, London, 241-244. Results of Leg 189. JOIDES Journal, 26, 11-17. DECONTO, R.M. & POLLARD,D. 2003. Rapid Cenozoic FALCON-LANG, H.J. & CANTRILL, D.J. 2000. Cret- glaciation of Antarctica induced by declining aceous (Late Albian) Coniferales of Alexander atmospheric CO2. Nature, 421, 245-249. Island, Antarctica. 1: Wood taxonomy: a quantita- DETTMAN, M.E. 1989. Antarctica: Cretaceous cradle tive approach. Review of Palaeobotany and Paly- of austral temperate rainforests? In: CRANE, J.A. nology, 111, 1-17. (ed.) Origins and Evolution of Antarctic Biota. FALCON-LANG, H.J. & CANTRILL, D.J. 2001a. Gymno- Geological Society, London, Special Publications, sperm woods from the Cretaceous (mid-Aptian) 47, 89-105. Cerro Negro Formation, Byers Peninsula, DETTMANN, M.E. & THOMSON, M.R.A. 1987. Cret- Livingston Island, Antarctica: the aborescent vege- aceous palynomorphs from the James Ross Island tation of a volcanic arc. Cretaceous Research, 22, area, Antarctica - a pilot study. British Antarctic 227-293. Survey Bulletin, 77, 13-59. FALCON-LANG, N.J. & CANTRILL, D.J. 200lb. Leaf DETTMANN, M.E., POCKNALL, D.T., ROMERO, E.J. & phenology of some mid-Cretaceous polar forests, ZAMOLA, M.D.C. 1990. Nothofagidites Erdtman ex Alexander Island, Antarctica. Geological Potonie, 1960; a catalogue of species with notes on Magazine, 138, 39-52. the palaeogeographic distribution of Nothofagus FALCON-LANG, H.J. & CANTRILL,D.J. 2002. Terrestrial B1. (Southern Beech). New Zealand Geological palaeoecology of an Early Cretaceous volcanic Survey Bulletin, 60, 1-79. archipelago, Byers Peninsula and President Head, DINGLE, R.V. & LAVELLE, M. 1998. Antarctic Penin- South Shetlands Islands, Antarctica. Palaios, 17, sula cryosphere: Early Oligocene (c. 30 Ma) initi- 535-549. ation and revised glacial chronology. Journal of the FALCON-LANG, H.J. CANTRILL, D.J. & NICHOLS, G.J. Geological Society, London, 155, 433-437. 2001. Biodiversity and terrestrial ecology of a mid- DINGLE, R.V., MCARTHUR, J.M. & VROON, P. 1997. Cretaceous, high latitude floodplain, Alexander Oligocene and Pliocene interglacial events in the Island, Antarctica. Journal of the Geological Antarctic Peninsula dated using strontium isotope Society, London, 158, 709-725. CRETACEOUS AND CENOZOIC ANTARCTIC FLORAS 79

FRANCIS, J.E. 1986. Growth rings in Cretaceous and Island, Antarctica. In: FRANCIS, J.E., PIRRIE, D. & Tertiary wood from Antarctica and their palaeo- CRAME, J.A. (eds) Cretaceous-Tertiary High- climatic implications. Palaeontology, 29, 665-684. latitude Palaeoenvironments: James Ross Basin, FRANCIS, J.E. 2000. Fossil wood from Eocene high Antarctica. Geological Society, London, Special latitude forests McMurdo Sound Antarctica. In: Publications, 258, 49-62. STILWELL, J.D. t~ FELDMANN, R.M. (eds) Palaeobi- HERNANDEZ, P.J. & AZCARARTE, V. 1971. Estudio ology and Palaeoenvironments of Eocene Fossili- paleobot~inico preliminar sobre restos de una ferous Erratics, McMurdo Sound, East Antarctica. tafoflora de la Peninsula Byers (Cerro Negro), Isla American Geophysical Union, Antarctic Research Livingston, Islas Shetland del Sur, Ant~irtica. Insti- Series, 76, 253-260. tuto Antdrctico Chileno Serie Cientifica, 2, 15-50. FRANCIS, .I.E. & HILL, R.S. 1996. Fossil plants from the HILL, R.S. (~ JORDAN, G.J. 1996. Macrofossils as indi- Pliocene Sirius Group, Transantarctic Mountains: cators of Plio-Pleistocene climates in Tasmania evidence for climate from growth rings and fossil and Antartica. Papers and Proceedings of the leaves. Palaios, 11, 389-396. Royal Society of Tasmania, 130, 9-15. FRANCIS, J.E. & POOLE, I. 2002. Cretaceous and HILL, R.S. & TRUSWELL, E.M. 1993. Nothofagus fossils Tertiary climates of Antarctica: evidence from in the Sirius Group Transantarctic Mountains. In: fossil wood. Palaeogeography, Palaeoclimatology, KENNETT, J.P. & WARNKE, D. (eds) The Antarctic Palaeoecology, 182, 47-64. Paleoenvironment: A Perspective on Global GANDOLFO, M.A., Hoc, E, SANTILLANA, S. (~ Change. American Geophysical Union, Antarctic MARENSSI, S. 1998. Una flor f6sil morpho- Research Series, 60, 67-73. loicamente afin alas Grossulariaceae (orden HILL, R.S., HARWOOD, D.M. & WEBB, EN. 1996. Rosales) de la formacidn La Meseta (Eoceno Nothofagus beardmorensis (Nothofagaceae), a Medio) Isla Marambio, Antdrtida. Asociaci6n new species based on leaves from the Pliocene Paleontoldgica Argentina, Publicacidn Especial, 5, Sirius Group, Transantarctic Mountains, Antarc- 147-153. tica. Review of Palaeobotany and Palynology, 94, GOTHAN, W. 1908. Die fossilen H/Slzer yon der 11-24. Seymour und Snow Hill-Insel. Wissenschaftliche HOFFMANN, J.A.J. 1975. Atlas climdtico de AmOrica del ergebnisse Der Schwedischen Si~dpolar-expedition Sur Mapas de temperatura y precipitaciones 1901-1903, 3, (8), 1-33. medias. WMO, UNESCO, Geneve. GRUNOW, A.M., KENT, D.V. & DALZIEL, I.W.D. 1991. HowE, J. & CANTRILL, D.J. 2001. Palaeoecology and New palaeomagnetic data from Thurston Island: taxonomy of Pentoxylales from the Albian of implications for the tectonics of West Antarctica Antarctica. Cretaceous Research, 23, 779-793. and Weddell Sea opening. Journal of Geophysical HUNT, R.H. 2001. Biodiversity and palaeoecology of Research, 96, 17 935-17 954. Tertiary fossil floras in Antarctica. PhD Thesis, HATHWAY, B. 1997. Nonmarine sedimentation in an University of Leeds. Early Cretaceous extensional continental-margin HUNT, R.H. & POOLE, I. 2003. Revising Palaeogene arc, Byers Peninsula, Livingston Island, South West Antarctic climate and vegetation history in Shetland Islands. Journal of Sedimentary Research, light of new data from King George Island. In: 67, 686-697. WING, S.L., GINGERICH, ED., SCHMITZ, B. & HATHWAY, B. (~z RIDING, J.B. 2001. Stratigraphy and THOMAS, E. (eds) Causes and Consequences of age of the Lower Cretaceous Pederson Formation, Globally Warm Climates in the Early Paleogene. northern Antarctic Peninsula. Antarctic Science, Geological Society of America, Special Papers, 13, 67-74. 369, 395-412. HATHWAY,B., DUANE,A.M., CANTRILL,D.J. & KELLEY, JEFFERSON,T.H. 1981. Palaeobotanical contributions to S.E 1999.40Ar/39Ar geochronology and palynol- the geology of Alexander Island, Antarctica. PhD ogy of the Cerro Negro Formation, South Shetland Thesis, University of Cambridge. Islands, Antarctica: a new radiometric tie for JEFFERSON, T.H. 1982. Fossil forests from the Lower Cretaceous terrestrial biostratigraphy in the Cretaceous of Alexander Island, Antarctica. . Australian Journal of Earth Palaeontology, 25, 681-708. Sciences, 46, 593-606. JONKERS, H.A. 1998a. The Cockburn Island Forma- HAYES, P. 1996. The Coniacian-Santonian (Late tion; Late Pliocene interglacial sedimentation in Cretaceous) flora of the Hidden Lake Formation, the James Ross Basin, northern Antarctic Penin- James Ross Basin, Antarctic Peninsula. In: sula. Newsletters on Stratigraphy, 36, 63-76. CORSETTII, E (~ TIFFNEY, B.H. (eds) Fifth Quadren- JONKERS, H.A. 1998b. Stratigraphy of Antarctic late nial Conference of the International Organisation Cenozoic pectinid-bearing deposits. Antarctic of Palaeobotany Abstract Volume, 41. International Science, 10, 161-170. Organisation of Palaeobotany and Department of JONKERS, H.A. & KELLEY,S.P. 1998. A reassessment of Geological Sciences, University of California, the age of the Cockburn Island Formation, Santa Barbara, USA. northern Antarctic Peninsula, and its palaeo- HAYES, P. 1999. Cretaceous angiosperm floras of climatic implications. Journal of the Geological Antarctica. PhD Thesis, University of Leeds. Society, London, 155, 737-740. HAYES, P., FRANCIS,J.E., CANTRILL,D.J. • CRAME, J.A. JONKERS, H.A., LIRIO, J.M., DEE VALLE, R.A. & 2006. Palaeoclimate analysis of the Late KELLEY, S.P. 2002. Age and environment of Cretaceous angiosperm leaf floras, James Ross Miocene-Pliocene glaciomarine deposits, James 80 I. POOLE & D. J. CANTRILL

Ross Island, Antarctica. Geological Magazine, 139, plant macrofossils from erratics, McMurdo Sound, 577-594. Antarctica. In: STILWELL,J.D. & FELDMANN, R.M. KEATING, J.M. 1992. Palynology of the Lachman Crags (eds) Palaeobiology and Palaeoenvironments of Member, Santa Marta Formation (upper Cret- Eocene Fossiliferous Erratics, McMurdo Sound, aceous) of north-west James Ross Island. Antarc- East Antarctica. American Geophysical Union, tic Science, 4, 293-304. Antarctic Research Series, 76, 243-251. KEATING, J.M., SPENCER-JONES, M. & NEWHAM, S. POOLE, I. 2002. Systematics of Cretaceous and Tertiary 1992. The stratigraphical palynology of the Kotick Nothofagoxylon: Implications for Southern Hemi- Point and Whisky Bay formations, Gustav Group sphere biogeography and evolution of the (Cretaceous), James Ross Island. Antarctic Nothofagaceae. Australian Systematic Botany, 15, Science, 4, 279-292. 247-276. LACEY, W.S. & LUCAS, R.C. 1981. The Triassic flora of POOLE, I. & CANTRILL, D.J. 2001. Fossil woods from Livingston Island, South Shetland Islands. British Williams Point Beds, Livingston Island, Antarctica: Antarctic Survey Bulletin, 53, 157-173. a Late Cretaceous southern high latitude flora. LI, H. 1994. Early Tertiary Fossil Hill flora from Fildes Palaeontology, 44, 1081-1112. Peninsula of King George Island, Antarctica. In: POOLE, I. & FRANCIS, J.E. 1999. The first record of SHEN, Y. (ed.) Stratigraphy and Palaeontology of fossil atherospermataceous wood from the upper Fildes Peninsula, King George Island, Antarctica. Cretaceous of Antarctica. Review of Palaeobotany State Antarctic Committee Monograph, 3, and Palynology, 107, 97-107. 133-171. POOLE, I. & FRANCIS, J.E. 2000. The first record of Ll, H. & SHEN, Y. 1990. A primary study of Fossil Hill fossil wood of Winteraceae from the Upper Creta- floras from Fildes Peninsula of King George ceous of Antarctica. Annals of Botany, 85,307-315. Island, Antarctica. A cta Palaeontologica Sinica, 29, POOLE, I. & GOTTWALD, H. 2001. Monimiaceae sensu 147-153. lato, an element of Gondwanan polar forests: MANOS, P.S. 1997. Systematics of Nothofagus Evidence from the Late Cretaceous-early Tertiary (Nothofagaceae) based on rDNA spacer wood flora of Antarctica. Australian Systematic sequences (ITS): taxonomic congruence with Botany, 14, 207-230. morphologyy and plastid sequences. American POOLE, I., CANTRILL, D.J., HAYES, P. & FRANCIS, J.E. Journal of Botany, 84, 1137-1155. 2000a. The fossil record of Cunoniaceae: new MOHR, B.A.R. 2001. The development of Antarctic evidence from Late Cretaceous wood of Antarc- fern floras during the Tertiary, and palaeoclimatic tica. Review of Palaeobotany and Palynology, 111, and palaeobiogeographic implications. Palaeonto- 127-144. graphica Abteilung B, 259, 167-208. POOLE, I., GOTTWALD, H. & FRANCIS, J.E. 2000b. Illi- MOORE, D.M. 1983. Flora of Tierra del Fuego. ciaceae, an element of Gondwanan polar forests? Anthony Nelson, Shropshire. Late Cretaceous and Early Tertiary woods of NICHOLS, G.J. & CANTRILL, D.J. 2002. Tectonic and Antarctica. Annals of Botany, 86, 421-432. climatic controls on a Mesozoic forearc basin POOLE, I., HUNT, R.J. & CANTRILL,D.J. 2001. A fossil succession, Alexander Island, Antarctica. Geo- wood flora from King George Island: ecological logical Magazine, 139, 313-330. implications for an Antarctic Eocene vegetation. NIKLAS, K.J. 1994. Predicting the height of fossil plant Annals of Botany, 88, 33-54. remains: an allometric approach to an old POOLE, I., MENNEGA,A.M.W. & CANTRILL, D.J. 2003. problem. American Journal of Botany, 81, Valdivian ecosystems in the late Cretaceous and 1235-1243. early Tertiary of Antarctica as evidenced from ORLANDO, H.A. 1968. A new Triassic flora from fossil wood. Review of Palaeobotany and Palynol- Livingston Island, South Shetland Islands. British ogy, 124, 9-27. Antarctic Survey Bulletin, 16, 1-13. POOLE, I., RICHTER, H.G. & FRANCIS, J.E. 2000c. ORTON, G.J. 1996. Volcanic environments. In: Evidence for Gondwanan origins for Sassafras READING, H.G. (ed.) Sedimentary Environments: (Lauraceae)? Late Cretaceous fossil wood of Processes, Facies and Stratigraphy. Blackwell Antarctica. International Association of Wood Science, Oxford, 485-567. Anatomists Journal, 21, 463-475. OTrONE, E.G. & MEDINA,CA. 1998. A wood from the RAINE, J.I. 1998. Terrestrial palynomorphs from Cape Early Cretaceous of James Ross Island. Ameghini- Roberts Project drillhole CRP-1, Ross Sea, aria, 35, 291-298. Antarctica. Terra Antartica, 5, 539-548. PHILIPPE, M., BARALE, G., TORRES, T. & COVACEVICH, RAINE, J.I. & ASKIN, R.A. 2001. Terrestrial palynology V. 1993. First study of in situ fossil woods from the of Cape Roberts Project drillhole CRP-3, Victoria Upper Cretaceous of Livingston Island, South Land Basin, Antarctica. Terra Antartica, 8, Shetland Islands, Antarctica: paaleoecological 389-400. implications. Compte Rendu Academie Science RIDING, J.B. & CRAME, J.A. 2002. Aptian to Coniacian Paris Series H, 317, 103-108. (Early-Late Cretaceous) palynostratigraphy of PIRRIE, D., fRAME, J.A., RIDING, J.B., BUTCHER, A.R. the Gustav Group, James Ross Basin, Antarctica. & TAYLOR, RD. 1997. Miocene glaciomarine sedi- Cretaceous Research, 23, 739-760. mentation in the northern Antarctic Peninsula RIDING, J.B., CRAME, J.A., DETrMANN, M.E. & region: the stratigraphy and sedimentology of the CANTRILL, D.J. 1998. The age of the base of the Hobbs Glacier Formation, James Ross Island. Gustav Group in the James Ross Basin, Antarc- Geological Magazine, 136, 745-762. tica. Cretaceous Research, 19, 87-105. POLE, M., HILL, R.S. & HARWOOD, D. 2000. Eocene RUTHSATZ, B. & VILLAGRAN, C. 1991. Vegetation CRETACEOUS AND CENOZOIC ANTARCTIC FLORAS 81

pattern and soil nutrients of a magellanic Seymour, Ant~irtica. Serie Cientifica del Instituto moorland on the Cordillera-de-Piuchue, Chiloe Antfrtico Chileno, 44, 17-38. Island, Chile. Revista Chilena de Historia Natural, TORRES, T.G., VALENZUELA, E.A. & GONZALEZ, I.M. 64, 461-478. 1982. Paleoxilologia de Peninsula Byers, Isla SHEN, Y. 1994. Subdivision and correlation of Cret- Livingston, Ant~irtica. In: Actas 3rd Congresso aceous to Paleogene volcano-sedimentary Geologico Chileno: Concepci6n, Chile, sequence from Fildes Peninsula, King George A321-A342. Island, Antarctica. In: SnEN Y. (ed.) Stratigraphy UPCHURCH, G.R., JR, OTTO-BLIESNER, B.L. & and Palaeontology of Fildes Peninsula, King SCOTESE, C. 1998. Vegetation-atmosphere inter- George Island, Antarctica. State Antarctic actions and their role in global warming during the Committee Monograph, 3, 1-35. latest Cretaceous. Philosophical Transactions of SHEN, Y. 1999. Subdivision and correlation of the the Royal Society of London, B353, 97-112. Eocene Fossil Hill Formation from King George VEBLEN, T.T., DONOSO, C., KITZBER~ER, T. & Island, West Antarctica. Korean Journal of Polar REBERTUS, A.J. 1996. Ecology of southern Chilean Research, 10, 91-95. and Argentinian Nothofagus forests. In: VEBLEN, SMELLIE, J.L., PANKHURST, R.J., THOMSON, M.R.A. & T.T., HILL, R.S. & READ, J. (eds) The Ecology and DAVIES, R.E.S. 1984. The Geology of the South Biogeography of Nothofagus Forests. Yale Uni- Shetland Islands: VI. Stratigraphy, Geochemistry versity Press, New Haven, CT, 293-353. and Evolution. British Antarctic Survey, Scientific WHEELER, E.A. & MANCHESTER, S. 2002. Woods of Report, 87, 1-85. the Eocene Nut Beds Flora Clarno Formation, SPECHT, R.L., DETTMANN, M.E. & JARZEN, D.M. 1992. Oregon, USA. International Association of Wood Community associations and structure in the Late Anatomists Journal Supplement, 3, 1-188. Cretaceous vegetation of southeast Australasia WHITHAM, A.G., INESON, J.R. & PIRRIE, D. 2006. and Antarctica. Palaeogeography, Palaeoclimatol- Marine volcaniclastics of the Hidden Lake Forma- ogy, Palaeoecology, 94, 283-309. tion (Coniacian) of James Ross Island, Antarctica: STOREY, B.C. d(¢ GARRETT, S.W. 1985. Crustal growth an enigmatic element in the history of the back-arc of the Antarctic Penisula by accretion, magmatism basin. In: FRANCIS, J.E., PIRRIE, D. & CRAME, J.A. and extension. Geological Magazone, 122, 5-14. (eds) Cretaceous-Tertiary High-latitude Pulaeo- SWENSON, U. & HILL, R.S. 2001. Most parsimonious environments: James Ross Basin, Antarctica. areagrams versus fossils: the case of Nothofagus Geological Society, London, Special Publications, (Nothofagaceae). Australian Journal qf Botany, 49, 258, 21-47. 367-376. ZACHOS, J., PAGANI, M., SLOAN, L., THOMAS, E. & SWENSON, U., HILL, R.S. & McLOUGHLIN, S. 2000. BILLUPS, K. 2006. Trends, rhyths and aberrations in Ancestral area analysis of Nothofagus (Notho- global climate 65 Ma to present. Science, 292, fagaceae) and its congruence with the fossil 686--693. record. Australian Systematic Botany, 13, 469-478. ZASTAWNIAK, E. 1981. Tertiary leaf flora from the TORRES, T. & LEMOIGNE, Y. 1988. Maderas f6siles Point Hennequin Group of King George Island terciarias de la Formacion Caleta Arctowski, Isla (South Shetland Islands, Antarctica). Preliminary Rey Jorge, Antfirtica. lnstituto Ant6rtico Chileno report. Studia Geologica Polonica, 72, 97-108. Serie Cientifica, 37, 69-107. ZASTAWNIAK, g. 1990. Late Cretaceous leaf flora of TORRES, T. & LEMOIGNE,Y. 1989. Fossil wood findings King George Island, West Antarctica. In: of angiosperms and gymnosperms of the Upper KONBLOCH, E. & KVA~EK,Z. (eds) Proceedings of Cretaceous at Williams Point, Livingston Island, the Symposium: Palaeofloristic and Palaeoclimatic South Shetland Islands, Antarctica. Instituto Antdr- Changes in the Cretaceous and Tertiary. Geological rico Chileno Serie Cientifica, 39, 9-26. Survey, Prague, 81-86. TORRES, Z., BARALE, G., MI~ON, H., PHILIPPE, M. & ZASTAWNIAK,E. 1993. Macroscopic plant remains from THt~VENARD, F. 1997a. Cretaceous floras from Upper Cretaceous and Tertiary of King George Snow Island (South Shetland Islands, Antarctica) Island, (South Shetland Islands, West Antarctica). and their biostratigraphic significance. In: RICCI, Wiadomogci Botanicczne, 37, 217-219. C.A. (ed.) The Antarctic Region: Geological Evolu- ZASTAWNIAK, E. 1994. Upper Cretaceous leaf flora tion and Processes. Terra Antarctica Publication, from the Blaszyk Moraine (Zamek Formation), Siena, 1023-1028. King George Island, South Sheltland Islands, West TORRES, T., BARALE, G., THEVERNARD,E, PHILIPPE, M. Antarctica. Acta Palaeobotanica, 34, 119-163. & GALLEGUILLOS, H. 1997b. Morfologfa y ZASTAWNIAK, E., WRONA, R., GAZDZICKI, A. & sistemfitica de la flora del Cretficico Inferior de BIRKENMAJER, K. 1995. Plant remains from the top President Head, isla Snow, archip61ago de las part of the Point Hennequin Group (Upper Shetland del Sur, Antfirtica. Serie Cientifica del Oligocene), King George Island (South Shetland Instituto Antdrtico Chileno, 2, 15-50. Islands, Antarctica). Studia Geologica Polonica, 81, TORRES, T., MARENSSI, S. & SANTILLANA, S. 1994a. 143-164. Maderas f6siles de la isla Seymour, Formacfon La ZHOU, Z. & LI, H. 1994. Early Tertiary ferns from Meseta, Ant~irtica. Instituto Antdrtico Chileno Fildes Peninsula, King George Island, Antarctica. Serie Cientifica, 39, 43-58. In: SHEN, Y. (ed.) Stratigraphy and Palaeontology TORRES, T., MARRENSSI, S. • SANTILLANA, S. 1994b. of Fildes Peninsula, King George Island. State Fossil wood of La Meseta Formation, isla Antarctic Committee Monograph, 3, 173-189.