Australian Paleogene vegetation and environments: evidence for palaeo-Gondwanic elements in the fossil records of Lauraceae and Proteaceae.

David R Greenwood

The prehistory of the Australian vegetation is summarized – from the earliest traces of life in the Precambrian to the events of the Cenozoic that shaped the continent’s modern flora. Australia’s oldest land plants appear in the Silurian with the rise of the lycophyte-dominated Baragwanathia flora. A succession of late Palaeozoic and Mesozoic pteridophyte- and gymnosperm-dominated floras followed, successive floras being punctuated by phases of major climate change and/or global mass-extinction events. For much of its Phenerozoic history, Australia has been part of the supercontinent Gondwana, thus it has shared broadly similar floras with the other Southern Hemisphere continents. The rise of the angiosperms in the Early Cretaceous was coincident with or slightly preceded the later stages of Gondwanan breakup. The modern angiosperm-dominated flora, therefore, developed during progressive isolation of Australia and its eventual collision with landmasses of southeast Asia. This history of isolation and amalgamation, together with the great climatic changes that have affected the continent over the past 100 million years, has set up intriguing scenarios for explaining the past and present distribution of a broad range of Australia’s plant taxa.

ABSTRACT--Charting the broad patterns of vascular plant evolution for Gondwana against the major global environmental shifts and events is attempted here for the first time. This is based on the analysis of the major vascular plant-bearing formations of the southern continents (plus India) correlated against the standard geological time-scale. Australia, followed closely by South America, are shown to yield by far the most complete sequences of productive strata. Ten seminal turnover pulses in the unfolding evolutionary picture are identified and seen to be linked to continental drift, climate change and mass global extinctions. The rise of vascular plants along the tropical belt, for instance, followed closely after the end-Ordovician warming and extinction. Equally remarkable is that the Late Devonian extinction may have caused both the terrestrialisation of the vertebrates and the origin of the true gymnosperms. The end-Permian extinction, closure of lapetus, together with warming, appears to have set in motion an unparalleled, explosive, gymnosperm radiation; whilst the Late Triassic extinction dramatically curtailed it. It is suggested that the latitudinal diversity gradient clearly recognised today, where species richness increases towards the tropics, may have been partly reversed during phases of Hot House climate. Evidence hints at this being particularly so at the heyday of the gymnosperms in the Late Triassic super-Hot House world. As for the origin of terrestrial, vascular, plant life, the angiosperms seem closely linked to a phase of marked shift from Ice House to Hot House. Insect and tetrapod evolutionary patterns are discussed in the context of the plants providing the base of the ever-changing ecosystems. Intimate co-evolution is often evident. This isn't always the case, for example the non-linkage between the dominant, giant, long-necked, herbivorous sauropod dinosaurs and the dramatic radiation of the flowering plants in the Mid-Cretaceous. RÉSUMÉ--Pour la première fois, une tentative de reconnaissance des modes principaux de l'evolution de la plante vasculaire du Gondwana en fonction des variations d'environnements et des événements globaux majeurs est présentée ici. Elle se fonde sur l'analyse des grandes formations à plantes vasculaires des continentaux méridionaux (plus l'Inde) corrélées avec les niveaux d'étage en fonction de l'échelle standard des temps géologiques. L'Australie, suivie de près par l'Amérique du Sud, montre les séquences de strates productives de loin les plus complètes. Dix épisodes de bouleversement sont identifiés dans l'établissement du tableau de l'évolution et sont liés de façons variées à la dérive des continents, aux changements climatiques et aux extinctions globales en masse. Par exemple, la progression des plantes vasculaires dans la zone tropicale a suivi de près le réchauffement et l'extinction de la fin de l'Ordovicien. Il est également remarquable que l'extinction du Dévonien tardif a pu déclencher à la fois la sortie des eaux des Vertébrés et l'origine des vrais Gymnospermes. L'extinction de la fin du Permien, et la fermeture de l'Ocean Iapetus, avec le réchauffement, semble avoir mis en mouvement une expansion explosive, sans parallèle, des Gymnospermes que l'extinction du Trias tardif a stoppée de façon brutale. Il est suggéré que le gradient de diversité en fonction de la latitude, clairement reconnu aujourd'hui, la richesse en nombre d'espèces augmentant vers les tropiques, a pu s'inverser en partie lors des phases de climat à effet de serre. L'apogée des Gymnospermes au Trias tardif sous un climat de super-effet de serre constitue un argument. Comme dans le cas de l'origine des plantes vasculaires terrestres, l'origine des Angiospermes semble étroitement liée à une phase de transition marquée d'un climat glaciaire à un climat chaud. Les modes d'évolution des Insectes et des Tétrapodes, quoique non reportés, sont discutés dans le contexte des végétaux constituant la base des écosystèmes toujours en réorganisation. Des évolutions intimement mêlées sont souvent évidentes. Ce n'est pas toujours le cas, par exemple l'absence de lien entre les Dinosaures Sauropodes herbivores à long cou dominants et l'expansion extraordinaire des plantes à fleurs au cours du Crétacé moyen.

In this chapter we use the plant macrofossil record as evidence of past ecosystems, and in particular the record from south-eastern Australia. We provide background on the evolution of Australian tropical rainforests over the Tertiary Period (65 to 1.8 million years ago), and we consider the origin of Australian tropical rainforests, how they responded to climatic and other environmental factors over geological time, and the ecological processes that have shaped them in the crucible of continental drift and climatic deterioration.

Australian Paleogenevegetationand environments:evidencefor palaeo-Gondwananelementsin the fossil recordsof Lauraceae and Proteaceae A.J. Vadala& D.R. Greenwood School of Lfe Sciences and Technologt, Victoria University of Technologt, PO Box 14428, Melbourne City MC, Victoria 8001, Australia ABSTRACT: Tropical rainforestsin the northeastof Australia have been interpretedas being either communities largely comprising taxa that 'invaded' newly available environmentsfrom Sundalandduring and after the Middle Miocene collision of the Australian plate and the Sunda plate, or refugia for humid-mesothermal Gondwanantaxa. Recent biogeographicanalyseshave suggestedfour 'tracks' (areas of endemism) that potentially account for some previously hypothesisedfloristic 'elements' defined by 'tropical' or Malesianorigins. Early Cenozoic(Paleogene) macrofloral records of Lauraceaeand Proteaceaeare informative on these issues.Unequivocal macrofossil evidence for Lauraceae and Proteaceaeoccurs at least from the Early Paleocene(- 65 million years,Ma) in Australia. This evidenceappearscontrary to the suggestion of their past dispersalfrom Malesia to the Australianplate. Tropical floristic 'elements' defined by Malesian origins are not appropriatefor elucidatingcurrent biogeographicpatternsof thesefamilies in Australia. I INTRODUCTION This paper reviews and providesmacrofossilevidencesupportingthe ancientnature and Gondwanan origin of extant floristic elementsof the Wet Tropics region of northeasternAustralia. Traditional descriptive phytogeographicanalysesof the extant Australian flora had identified three floristic elements,defined by hypothesisedorigin (sensuCrisp et al. 1999): l) a Gondwananelement that comprisesa rainforestflora with centresof diversity in the temperate south and humid tropical northeastand sharesgenerawith, or has closely relatedgenera in, other Austral landmasses; 2) an autochthonous element characterisedby high endemism and represented by the sclerophyllous and dry-climate adapted vegetation of much of Australia (particularly the southwestof WesternAustralia); 3) a tropical elementcomposedof taxa sharedwith southeastAsia, largely centredin the humid tropics and monsoonaltropics (Herbert 1932, 1967; Burbidge 1960; Barlow l98l; Schodde 1989;Crispet al. 1999). Rainforestsof the Wet Tropics region of northeasternQueenslandmainly comprisetaxa belonging to the third of these elements.They have been consideredas a vegetationtype largely comprising taxa that 'invaded' newly available environmentsfrom Sundalandduring and after the Miocene collision of the Australian and Eurasianplates. Alternatively, they have been regarded as refugia for humid mesothermalGondwanan taxa. These narrative analyseshave emphasisedthe role of either continental drift or long-distance dispersalof plant propagulesin shaping the modern flora of Australia. However, more recent biogeographicanalyseshave stressedthe 'autochthonous'characterof much of the flora of the Wet Tropics (e.g. Webb et al. 1984; Webb et al. 1986; Truswell et al. 1987) that may reflect an ancientGondwananheritage that was also proposedby Barlow (1981).Crisp et al. (1999) concludedthat the lack of success in identiffing generalisedbiogeographic tracks (sharply differentiated areas of endemism) 202 A.J. VnpnLA ANDD.R. GneeNwooD within the Australian craton (Australia and New Guinea) may reflect the lack of significant barriers to dispersal and consequentlypossible range expansion for some taxa acrossAustralia. Analyses of Australian fossil microfloras have demonstrated that many of the 'tropical' elements of the extant flora of the Wet Tropics region were present in Australia prior to its final separation from the remainder of Gondwana (Truswell et al. 1987; Drinnan & Crane 1990). These floral elements were once presented as descendants of immigrant Malesian or tropical floral elements (see Crisp et al. 1999). Truswell et al. (1987) concluded that some exchange of taxa between the Ausfalian craton and lands to the northwest of Australia had occurred, but had not resulted in any major alteration to the structure or compostion of Australian forests. Webb e/ al. (1984) also suggestedthat the Middle Miocene was the beginning of arid periods that would not have favoured the spread of rainforest immigrants. The evidence presented here consists of a review of the Australian macrofossil record of Lauraceae and Proteaceae from the Paleogene (- 65-23.3 million years (Ma): Fig. 3). New data from a Late Paleocene leaf macroflora from Cambalong Creek in the Southern Highlands of New South Wales are also presented. This macroflora may include the oldest known Australian taxa of tribes Laureae and Cryptocaryeae of Lauraceae and of tribes Banksieae, Oriteae, Stenocarpinae, Helicieae and lfuightieae (Grevilleoideae) of Proteaceae.These data are used as supporting evidence for the presence of Lauraceae and Proteaceae in eastern Gondwana prior to separation of some Austral landmasses,and clearly precede the Miocene 'contact phase' (Hall 1996, 1997) between Australia and those parts of Malesia from which tropical floral elements may have 'invaded'. Southeastern Australia is important in the evolution of the modern flora due to the interactions between vegetation and the physical environment. The area underwent significant tectonism in the Paleogene during uplift of the Eastern Highlands (Wilford & Brown 1994). Coeval subsidence along a failed rift saw formation of the Gippsland Basin, and the area experienced major sea-level changes associated with initial Antarctic glaciation and the opening of Bass Strait as Australia and Antarctica rifted apart (Crook l98l; Kemp l98l; Powell et al. l98l;Blackburn & Sluiter 1994; Wilford & Brown 1994). l.l The plant macrofossil record Evidence of the Cenozoic vegetation is abundant throughout southeastern Australia. Microfloras and macrofloras have been recorded from numerous localities spanning the Paleogeneand Neogene (Carpenter et al. 1994; Christophel1994:' Macphail et al. 1994; Greenwood et al. in press, and references therein). Paleogene macrofloras in particular constitute a record of the vegetation of Australia prior to its final separation from East Antarctica and preceding the Miocene collision between the Australian craton and the Sunda Arcs. The most detailed hlpotheses regarding palaeovegetation and phytogeography for the Late Cretaceous and Cenozoic of southeastern Australia are currently based on palynology (e.g., Kemp l98l; Martin 1981, 1991, 1994, 1998; Truswell et al. 1987:'Drinnan & Crane 1990; Truswell 1990, 1993; Kershaw et al. 1994; Macphail et al. 1994). These have been complemented by systematic taxonomic research on leaf macrofossils (e.g., Christophel 1981, 1989, 1994; Hill 1983, 1992a & b, 1994; Hill & Jordan 1993; Hill & Pole 1992; Hill & Carpenter I 99 I ; Carpenter & Jordan 1997; Jordan et al . 1998; Hill et al . 1999). The Murray and Gippsland Basins provide virtually continuous sequencesof Late Cretaceous to Pliocene-Pleistocenepalynofloras and the highly detailed dinoflagellate, foraminiferal and sequence stratigraphy for these basins acts as an independent control for spore-pollen zonation (e.g. Holdgate & Sluiter l99l; Macphail et al. 1994). However, reconstructionsof palaeovegetationbased on palynology have significant limitations. For example, the family Lauraceae is a major component of southeastern Australian (particularly Victorian) Cenozoic macrofloras, yet it is absent from the palynological record because the thin sporopollenin exine of the pollen preservespoorly (Truswell et al. 1987; Drinnan et al. 1990; Martin 19941'Hill et al. 1999; although see Macphail 1980). Definitive cuticular morphological characters have been identified for and within many significant families in the fossil and extant floras of Australia (e.g., Hill 1986, 1990, 1991,1992b, 1994; Hill & Carpenter l99l; Hill & Christophel 1988; Hill & Read l99l; Carpenteret al. 1994 and references therein; Christophel & Rowett 1996; Hill & Christophel 1996; Carpenter & Jordan 1997; Jordan et al. 1998). Such a high degree of taxonomic resolution (to generic level for instance) is rarely available from palaeopalynological analyses(Macphail et al. 1994). PeleocnNe 203 vEcETATtoN AND ENVIRoNMENTS Podocarpaceae Elaeocarpaceae Eucryphiaceae as Figure l: Compositionof the Late Paleocenemacroflora recoveredfrom CambalongCreek indicated percentageof 173 total specimens. Extantlineag€ Boilschmiectia Cryptocarya Australia non-RF RF o o o o |r s Ursea o Fiilt"ifif,8"r o o Bt"TJ[!fo18*r o o llfJiEfr€B"in""r Endiandn Fl€lrclnae, d?;8" c"l[?6niaz$flfr0 ,fi%*!" MaresiaSE Asia o o o o o o o o o o o o o o o o o o o India Sri Lanka Japan o o o O o o o Aftica o o o o I o lfo"sh\fiss"r Odteae o o o relativesof l-ate PaleoceneLauraceaeand ProteaceaemacFigure 2: Extant distributions of closestliving-cr"y circle for Endiandra indicatesthat some taxa grow in rofossils recoveredfrom CambalongCreek. that one gallery and temperateforests rathei than tropical rainforest. Grey circle for Litsea indicates th" KimberTerritory Northern northern coastaf throughout Queensland, ryq gtatnisa) iaxon'(L. -glact !ro*, circles for Heliciiae indicatethe distribution of Helici4 grey cirley region of westeri Australia. to small areas cles for this tribe indicate distribution of Hollandaea,a genusof trvo speciesboth_restricted for DarlinKnightieae-is for shown bistribution northeastern of region Tropics Queensland. in the Wet northeastern of region Queensland' Tropics Wet the to restricted gia, agenusofiwo rp?"i"tufro 204 A.J. VepnLA ANDD.R. GnerNwooD E .gET Es = ; g s ss E c E, EEfi B$ s eE i 65Es EgeEF$sggs$$$e f" iEgE rfid I L LrC I 0 I o c t" 0 0 0 t z tO :- 0 c 0 o : (- >= o E : sX F o .9 € < s f ; ge 6 0 fl .9 -9 : .s.g € 9 : F |o g (D= c 0 c 0 o 0 .9 0 N 0 0 0 0 o o lrl 0 c 0 o 0 oc 0 0 (U 0o t! I Fl N o o ll, o 0 6 0. .9 :a CEb. Fsa2P E3"Efii g E{€FtE' < ea c o u, $ a 0- Gd ;gEE gEEEE .eE:gu* e€#tc 93sES 0 c 0 0 o It #fiii ! 0 -9 c 0 rO ct gH€E :E$F " $g; E;.!: eEd i* E?$I ( , G g ggEE ggEg "EgFac €;Egg gF€ PeLeoceNe vEGETATToNAND ENvtRoNMENTs 205 1.2 Lauraceae and Proteaceae as targets for research Proteaceae and l.auraceae are comparatively well represented temporally and spatially in Australian macrofloras, where oldest fossils are recorded from the early Paleogene (Fig. 3). Proteaceae and Lauraceae leaf macrofossils with cuticular preservation are particularly amenable to identification with high levels of taxonomic resolution due to relatively recent research into extant tribal and generic limits. The tribal taxonomy of extant Proteaceae was established with the detailed and extensive work of Johnson & Briggs (1963, 1975) and revised by Douglas (1995). General cuticular morphological characteristics of the family have been described from macrollogs spanning the Paleogeneand Neogene (Blackburn 1985; Hill & Merrifield |993;Carpenter & Pole 1995; Carpenter & Jordan 1997; Jordan et al. 1998). Cuticular features of fossil extant taxa have been described at subfamilial, generic, tribal and subtribal levels (e.g. Cookson & Duigan 1950; Lange 1978; Hill & Christophel 1988; Hill & Merrifield 1993; Carpenter t994; Carpenter et al. 1994; Carpenter & Pole 1995; Jordan 1995; Carpenter & Jordan 1997; Jordan et al.1998; Vadala & Drinnan 1998). By contrast, intrafamilial relationships within Lauraceae are poorly understood (Eklund 1999). Suprageneric taxonomy of extant Lauraceae is less well established than for Proteaceae and consequently numerous suprageneric classifications for the family exist (see van der Werff & Richter 1996). Bandulksa (1926, 1928) recognized the strong similarities in cuticular morphology between some extant and fossil genera of Lauraceae.However, the lack of good cuticular morphological data for extant and fossil Australian genera of Lauraceae limited the usefulness of this record for many years (see Hill 1986). The published Cenozoic record of the family in Australia is substantial (Hill 1986, 1988a, b) and dates from the Early Eocene (Table I and Fig. 3). Abundant leaf impressions with ascending acrodromous venation led to the gradual development of the concept of a'Cinnamomum flora' (reviewed by Duigan l95l). Most of the early identifications of these Paleogeneand Neogene leaves withCinnamomum were based only on gross leaf morphology. However, most of the earliest described leaves could not even be accepted as Lauraceae without cuticular morphological evidence (Hill 1988a). Hill (1986, 1988a) reassigned some of these taxa with adequate cuticular preservation to the genus Laurophyllum and the concept of the 'Cinnamomumf\ora'was graduallyrejected. Nevertheless,the fossil record of Lauraceae is impressive compared with that of other taxa that are significant in the modern flora of Australia, such as Acacia, Eucalyptus and Casuarina. Many Paleogene and Neogene Lauraceae from Australia have been described as Laurophyllum (e.g. Hill 1996; Carpenter & Pole 1995), a genus indicating only general affinity with Lauraceae and limiting the systematic or phylogentic use of the record. Subsequentresearch has demonstrated that leaf venation and shape (Christophel & Hyland 1993) and cuticular morphology (Christophel & Rowett 1996) can be used as reliable characters in the taxonomy of extant Australian Lauraceae. The potential utility of the Australian macrofossil record of Lauraceae has been increased since the first critical work on the Australian taxa (Hill 1986) by refinement of generic and suprageneric concepts in Lauraceae. This has been based on a wide suite of characters (van der Werff & Richter 1996) including cuticular morphology (Christophel et al. 1996) and foliar morphology (Klucking 1987; Cristophel & Hyland 1993), and a major revision of the arborescent Australian genera of the family (Hyland 1989). Table l. Publishedrecordsof Paleogeneand NeogeneLauraceaemacrofossilsfrom Australia.The table does not include some fossils of possiblelauraceousaffinity that were included in the survey of Hill (1988b) and which were describedwithout cuticulardetail on the basisof leaf morphologyonly. Specimenslistedwith extantaffinity as '?Lauraceae'shouldbe considereddoubtful(seetext). Locality (Aee) Fossiltaxon Extantaffinitv CobungraRiver' Cinnamomumpolymorphoides' ?Lauraceae (?EarlyEocene) Nerriga (Early/Middle Laurop hy Cryptocarya (C. bellendenkerana, ' I lum ac ro c ry p t oc ary Eocene)'" oidesh C. grandis)'" Nerriga ( Early/Ir4iddle Laurop hyl lum acrodromum ?Endiandra:E. pubens group* Eocene)'' L. conspicuum L. acuminatum ?Endiandra : E. jonesii group* 206 A.J. VaoeLA ANDD.R. GneENwooD L. aculum L. angulosum L. squamatum L. lanceolatum Nelly Creek (Middle Eocene) Angleseaclay lenses (late Middle Eocene) Angleseaclay lenses (late Middle Eocene) Lefroy paleodrainage (Middle/LateEocene)re Hasties(Late Eocene) JungleCreek(LateEocene) Kojonup Sandstone (Late Eocene) PallinupSiltstone (Late Eocene) VegetableCreek (Late Eocene) GoldenGrove(Eocene)2 PascoeVale (Late Eocene/Early Oligocene) SedanCoalfield (Oligocene- Micoeneor Late Eocene- Miocene) Narracan(Early Oligocene) ?Endiandra : E.jonesii or E. pubens group+ ?Cryptocarya : C. pleurosperma groupt ?Cryptocarya : C. pleurosperma group* ?Neolitseadealbata or Litsea : L. fawcettiana group* ?Endiandra: E.jonesii group* ?Endiandra : E. jonesii group* L. brochidodromum L. intramarginatum L. sinuatum L. pubescens L. arcuatum Parataxon 53 ?Endiandra : E.jonesii group* ?Endiandra: E. pubens group* Lauraceaeo Parataxon 416 Lauraceael6 Lauraceae Il7 Lauraceae IIIT Lauraceae IIIrT Lauraceae IVIT Endiandra muelleri, ?LitseatT NeolitseadealbatatT ?Cinnamomuml? Cryptocarya" cur-L-00lre cur-L-002'e cur-L-003re cur-L-004'e cur-L-005re cur-L-006'e cur-L-007'e cur-L-009'e cur-L-009'e cur-L-OlOre ?Litsea bennettii groupr Laurophyllum cf. L. arcuatumT ?Endiandra:E. pubens group* ?Endiandra:E. pubens group* ?Endiandra:E.jonesii group* ?Lindera* ?Lindera* ?Endiandra:E. pubens groupt ?Endiandra:E. pubens group* Cryptocaryorylon gippslandi- Laurophyllum arcuatum (alsoZ. acuminatum)1 Cryptocaryaoblatat\ cum'" cf. Lauraceaes Lauraceae5 wAM.P88.2l s Lauraceaes Cinnamomum nuytsii23 (= Lauro p hy I lum nuy tsi i22) C i n namo mum po lymo rp ho i dese Acrodromous primary venation taxon' Pinnate primary ve.nation taxon2 Ltnnamomum sp.'- Parataxon AA 007: 'aff. Endiandra'l Parataxon AG 005: 'aff. Cryptocarya'l Parataxon AA 006: 'aff. Cryptocarya'l C in na mo mu m po lymo r p ho i dese C. polymorph;id;se' t2' Cryptocarya australistz Dalton (Late Oligocene) C i n na mo mu m p o lymorp ho idese Morwell Open Cut C innamomu m praev irenst I (Late Oligocene) Darlimurla (Late Oligocene) Cryptoc arya prae o bovatat2 ?Lauraceae ?Lauraceae ?Lauraceae ?Lauraceae ?Endiandral ?Cryptocaryal ?Cryptocaryal ?Lauraceae fossil Cinnamomumburmannit2 Cryptocarya australis/C. murrayil C. mackinnoniana" ?Lauraceae Cinnamomumvirens. C. oliveritl Cryptocaryaobovatat2 PeLeocENe vEcETATToN AND ENvTRoNMENTS Tambellup Siltstone (?Oligocene) West Dale (?Oligocene) Berwick Quarry (Late Oligocene/ Early Miocene) Newstead(: Elsmore); (Oli goceneilr4iocene) Maddingley(Miocene) Pitfield (Miocene) Werribee Ck./Lyalls Ck. (Miocene) Werribee Ck./Lyalls Ck. 207 'Laurophyllum's Laurophyl lum striatuma Laurophyllun sp.'sinuous'la Laurophyllurnsp.'thick' ra Laurophyllum sp.'butterfly"o Laurophyllun sp.'smooth'la Cinnamomumleichhardtiie Laurophyllum arcuatumlL. brochidodromuma ?Lauraceae Crinnamomum polymorphoidese' Cinnamomumpolymorphumts cinnamomumpolymorphoidese ?Lauraceae Cinnamomumpolymorphoidese' Cinnamomumpolymorphumtt 13 Laurus werribeensist3 ?Lauraceae (Miocene) Regatta Point Laurophyllum aus tralum6 Cryptocarya novae-anglical C. sp.nov (Early/?MiddlePleistocene) (Mt. BellendenKer, Queensland)u Mount Bischoff, Tasmania Laurus sprentiis ?Lauraceae Travertine lake depositsnear Cinnamomumwoodwardito ?Lauraceae Hobart William Creek,SouthAusCinnamomumsp.ro ?Lauraceae tralia *PossibleextantaffinitiesofLaurophyllumspecimensfromttreIproduced by keying parataxausing the key to extant Australian generaof Christophel & Rowett lteeO; where possible from the descriptionsand illustrationsprovided in Carpenter& Pole (1995) and Hill (1e86). rRowett(1991); 2Christophel & Greenwood(1987); 3christophel, Scriven& Greenwood(1992): 4ttill & -(lgg7): ttole 8loirnston.(1886); Merrifield (1993);.JMcloughlin & Hill (1996); 6Jordan -11992b;; 'Chapma,1 (1921); "Deane(1925);r2Paterson (1935);rsMccoy(1876);tapoleet al. 0926);'uChapman rsDouglas(^1967);r5Rowett (^1993);''Douglas & Christop^hel(1990); rTChristophel,Harris & Syber (1987); 'ol-eisman(1986); ''Carpenter ''Hill (t986): 22Hiit 22Hill n& Pole (1995); 'uConran& Christophel(1998);2'Hitt 11lAO); ( I 88s;. ( I 988a);2rEttingshausen 2 MACROFOSSI EVIDENCE FOR LAURACEAE AND PROTEACEAE 2.1 Lauraceae - Laurasian records Recent phylogenies based on DNA sequence analyses have emphasised the antiquity of Laurales (Qiu et al. 1999). Drinnan & Crane (1990) indicated that Lauraceae had diffirentiated early in angiosperm evolution, likely by the Albian, - I l0 Ma. Indeed, fossils of definite lauraceous affinity are known from the early Cenomanian (- 97 Ma: Drinnan et al. 1990; Eklund & Kvacek 1998). Monosulcate pollen typical of the magnoliid dicotyledon (including Laurales, Winterales and Chloranthaceae) and monocotyledon grade first appear in the fossil record around the Hauterivian (Hughes & McDougall 1987), preceding the appearance of triaperturate pollen typical of the non-magnoliid dicotyledon clade (Crane 1987; Mcloughlin et al. igqS). Supporting this antiquity is a good mid-Cretaceous macrofossil record of I-auraceae (Eklund & Kvacek 1998 and references therein). Early Cenomanian (- 97 Ma) inflorescences and flowers of Mauldinia mirabilis were described from the Potomac Group, eastern North America (Drinnan et al. 1990), and Cenomanian M. bohemica inflorescences are known from the PerucKorycany Formation in the Czeck Republic (Eklund & Kvacek 1998). Macrofossils of Lauraceae are common across middle and low palaeolatitudes in the Northern Hemisphere in the Maastrichtian Northern Gondwana and Normapolles Provinces (Crane l9S7). Pole (1992a) suggested lauraceous affinity for several leaves with pinnate margins, acrodromous primary venition and percurrent secondary venation from the Upper Cretaceous Taratu Formation, Otago, New Zealand. The Late Cretaceous (Santonian/ Campanian; - 83 Ma) coincided with increased 208 A.J. VnoeLA ANDD.R. GneeNwooD seafloorspreadingbetweenAustralia and Antarctica (Veeverset al. l99l), separationof India from Australia (Powell et al. l98l; Wilford & Brown 1994),and initial spreadingto form the TasmanSea(Crook l98l; Veeverset al. l99l). Audley-Charles (1987)suggested that dispersal of land plantsbetweenmainlandAsia and Australiawould havebeenpossiblearoundthat time (- 90 Ma). However, more recent tectonic reconstnxctions of the region indicate severalthousandkilometresstill separatedthe Eurasiancontinentalmargin and the leadingedgeof the Australian cratonat the Early/Ir4iddleEocene,- 50 Ma (Hall 1996,1997). Diversity within the Lauraceaepersistedin the Paleogenewith leaves,wood and reproductive structuresbeing abundantand diverse in macroflorasfrom most parts of the world (Eklund & Kvacek 1998). lndeed, PaleocenemacroflorasspanningnorthernAmerica, central and eastern Europe(the SouthernLaurasianfloristic provinceof the CenomanianandNormapollesProvince of the Santonian{ampanian) are typically dominatedby palms, Euphorbiaceaeand Laurales (Crane 1987). The london Clay flora containseight taxa of Beilschmiedia(somepossibly attributable to the closely relatedgenusEndiandra), two taxa of Cinnamomum,onetaxon of Litsea, one taxon of Crowella, five taxa of Laurocalyx and 30 taxa of Laurocarpzz (Chandler 1964).This indicatesthe existenceof two of the three extanttribes of Lauraceaerecognisedby van der Werff & Richter (1996): Perseeae(as Cinnamomum)and Cryptocaryeae(as Beilschmiedia,Cryptocarya and Endiandra) in the North Atlantic/Europeanfloristic Provinceduring the Paleocene(sensuCrane 1986).PaleoceneLauraceaeleavesare alsorepresentedin macrofloras preserved in nine other Lower Eocene sedimentarybeds from southern England (Chandler 1964), and were a dominant componentof the vegetatio'npreservedin both the BournemouthBeds (Bandulska 1928) and london Clay (Chandler 1964).Exact generic relationshipsof l^auraceaeleaf macrofossilsfrom the EoceneLondonand BournemouthClay floras and from [,ate Cretaceousand Early Paleogenelocalities in the Unites Statedare uncertainbeyond the more general 'Cinnamomum' and'Laurophyllum' types.The presenceof Lauraceae macrofossilsin theseareasstronglysuggestthat lauraceaewas a prominentcomponentof early angiospermplant communitiesin Laurasia,and substantiallypredatesthe earliestknown records of the family from Gondwana. 2.2 Lauraceae- Australian records PublishedAustralian macrofossilrecordsof Lauraceaeextendfrom the Early Eocene(Table I and Fig. 3). This extensiverecord connotesthe prominenceof Lauraceaein the easternAustralian sector of Gondwanabefore the formation of a deep marine strait betweenTasmaniaand Antarctica in the early Late Eocene/EarlyOligoceneand the ensuingdevelopmentof CircumAntarcticoceaniccirculation(Kemp 1981;Martin l99l; Veeverset al. l99l; Wilford & Brown 1994). Theseeventsclearly predatethe 'contact phase' betweenthe Australian craton and the Eurasianplate in the Miocene(Powell et al. 1981;Truswell et al. 1987;Metcalfe 1990;Hall 1996,1997:Fig. 3). Many of the older publishedrecordsin Table I shouldbe considereddoubtful. For example Early Eoceneand I-ate Oligocene/Miocenerecordsof Cinnamomum(McCoy 1876; Chapman 1921,1926;Deane1925),Cryptocarya(Paterson1935),andLaunts (McCoy 1876)were made on the basis of leaf morphology,and lack the cuticular morphologicalinformation neededfor definite placementin Lauraceae(Hill 1988a).Publishedmacrofossilswith cuticularpreservation enabling secureassignmentto Lauraceaedate from the Early/Middle Eocene,with Hill (1986)and Conran& Christophel(1998)describinga total of l3 taxaof Laurophyllumfromthe Nerriga locality in New SouthWales(Table I and Fig. 3). Conran& Christophel(1998) indicatedthe fossil taxon L. acrocryptocaryoidesfrom Nerriga had a combinationof cuticular morphological characterssuch as wide, butterfly-shapedcuticular scalesand rounded epidermal cells (Conran& Christophel1988,figs.2C,2D) characteristic of extantCryptocarya(Christophel & Rowett 1996).This reiteratesthe presenceof l,auraceaesimilar to extant Tribe Cryptocaryeaein southeastern Australia in the Paleogene(early Middle Eocene;-50 Ma). Hill (1986) described12 taxa of Lauraceaefrom Nerriga on the basisof micromorphological charactersand assignedall to Laurophyllum(Table l). The taxa describedby Hill (1986)may include Endiandra, Cryptocarya andeither Neolitseaor Litsea using the key of Christophel& Rowett (1996; Table l). Laurophyllumacrodromum,L. brochidodromum,L. intramarginatum, L. acuminatum,L. acutumandL. arcuatumdescibedby Hill (1986)all appearto havecombina- PnLeoceNE vEcETATtoN AND ENvTRoNMENTS 209 tions of micromorphologicalcharacterstypical of extant Endiandra (Christophel & Rowett 1996).Theseincludeepidermalcellswith inegularlythickenedand/orgranulatepericlinalwalls and cuticularscalesthat appeardouble(Hill 1986,figs. 7C and 7F,,l3C and l3E, 14D,9C, lOC, l7C). L. acrodromum,L. brochidodromum,L. intramarginatum and L. arcuatum also have guardcellswith polar extensions or rods(Hill 1996,figs. 7E,I3C,l4D, l7D) while L. acuminetum, L. acutum andL. intramarginatumhave mostly angularadaxial cell wall outlines (Hill 1986,figs. 98, l0A, l4E). Thesecharacters arealsotypicalof extantEndiandrain combination with the othercharactersdescribedabove(Christophel& Rowett 1996).The fossil describedas L. lanceolatumbyHill (1986)has cuticularledgesthat appearsingle,thin and mainly straight (Hill 1986,figs. 9G and 9H), which aretypicalof extantLitsea(Christophel& Rowett 1996).L. angulosumandL. squamatumfrom Nerriga appearto have wide, butterfly-like cuticular scales (Hill 1986,figs. I lF, l2E). This impliesa closerelationshipwith extantCryptocarya(Christophel & Rowett 1996). TheseNerriga fossils suggestthe presencein southeasternAustralia of Tribes l-aureae(as Litsea) and Cryptocaryeae(as CryptocaryaandEndiandra;van der Werff & Richter 1996)in the Early/MiddleEocene(- 50 Ma). Lauraceaealso feature prominently in the Middle/Late Eocene(- 50-35 Ma) of southern Australia (Table l), at which time therewas a substantialseawayin the Indian Ocean,Southern Oceanand TasmanSea(Fig. 3; Veeverset al. l99l), and the Australiancratonwas still distant from Sundalandand the Eurasianplate (Hall 1996, 1997).Christophelet al. (1987) described four late Middle Eocenetaxa of Lauraceaefrom the Angleseaclay lensesin southernAustralia (Victoria). Thesemacrofossilshave strongaffinites with extantEndiandralLitsea,Neolitseaand (Christophelet al. 1987),reflectinga similar rangeof diversityin CinnamomumlCrytpocarya Lauraceaeas may havebeenpresentearlier at Nerriga. Carpenter& Pole (1995) describedten taxa of Laurophyllumfrom the Middle/LateEocene Lefroy Palaeodrainage(Pidinga Formation)in WesternAustralia, some of which may reperesentLitsea, Lindera andEndiandra (Table l). Four of the fossil cuticle qpes describedby Carpenter& Pole (1995)havecombinationsof charactersincluding'double' cuticularscales(Carpenter& Pole 1995,figs. 25, 37-38,40,42-44,54,56)typicalof extantEndiandra(Christophel & Rowett 1996).Fossil type I has a combinationof charactersincluding granularpericlinal walls; single, thin and straightcuticular ledgesand prominentabaxialpapillae(Carpenter& Pole 1995,figs. 24, 28, 25) that are typicalof extantLitsea(Christophel& Rowett 1996).Fossil cuticle We 7 has highly sinuousabaxialcell outlinesand heavily 'beaded'abaxialanticlinal walls (Carpenter& Pole, 1995figs. 47,45) that are typicalof the extantgenusLindera (Christophel& Roweff 1996). (- 58-60 Ma) at CambalongCreek(Taylor from the Late Paleocene Lauraceaemacrofosssils from India and New Zeaet al. 1990)grew at a time when Australiahad alreadybeenseparated IndianOceanand TasmanSea(Poweller land by a long periodof developmentof the southeast al. l98l; Wilford & Brown 1994),althoughAustraliaand Antarcticawere still partiallyjoined (Veeverset al. l99l). Australia was also still distant from Sundaland,the continentalblock forming southeast Asia, at approximately60 Ma (Powellet al. l98l; Hall 1996,1997). Lauraceoustaxa from the CambalongCreekmacroflorawill be formally describedin a future publication (Vadala & Drinnan, in prep.). Lauraceaecompriseapproximately20Yoof total taxa in the macroflora,as eight speciesin four genera(Fig. l: A.J. Vadala,unpubl.).Thesefossils predateall publishedrecordsof Lauraceaefrom Australia(Table I and Fig. 3) and havebeen identified as extant generausing morphologicalcharacterspreservedby the mummified leaf cuwasachievedby runningthe fossil ticules.Primaryclassificationand sortingof fossil specimens specimensthrough the key to Australian genera of Lauraceaedevelopedby Christophel & Rowett (1996).This identifiedfossilsbelongingtoBeilschmiedia,Cryptocarya,Endiandraand Litsea. Stepsin the key were usedto derivea binary (qualitative)characterset (28-33 characfor the fossilsspecimensand for ters)and a continuous(numerical)characterset (7 characters) 25 extant Australian speciesof Beilschmiedia,Endiandra and Cryptocarya.The two character setswere used in conjunctionand separatelyin patternanalysesfor the threegenerato determine further taxonomic divisions between specimens.These analysesapplied sequentialstrategies usingdissimilaritymetrics(Belbin 1987)to agglomerative-heirarchical-combinatorial the datasets. Australiaof taxasimiThe CambalongCreekLauraceaeindicatethe presencein southeastern lar to the extant tribes Laureae(sensuvan der werff & Richter 1996;'as Litsea) and Crypto- 2to A.J. VnoeLA ANDD.R. GneeNwooD caryeae(as Beilschm.iedia,Cryptocarya and,Endiandra) before the breakup of the last fragmentsof Gondwana(Powell et al.l98l; Veeverset al. i99l). This precedejtheMiocenecontact betweenthe Australiancontinentand Sundaland by - qOMra1i.ig.:; powell et al.lggl; Truswell et al. 1987;Metcalfe 1990).The extantdistributionsof the n--earest living relativesof thesetaxa are shownin Fig. 2 andTable3. Extant Endiandra consistsof approli-m-ately100 species(Table 3; Hyland 1989) occurring from Australia throughNew Guineato Malesii and acrossbroadersoutheastAsia (Fig. 2). Most of the Australian speciesgrow in rainforest(Table 3). Beitschmiediaconsistsof ZOO-ZSO extant species(Table 3; Hyland 1989)found in Africa, Australia,SouthAmerica,New Zealandand New Guinea,throughMalesiaand broadersoutheast Asia to India (Fig. 2). The I I extantAustralian taxa of Beilschmiedia are all restrictedto rainforest habitais lHytanA 1989: Table 3). Cryptocaryaconsistsof 200-250extantspecies(Hyland 1989)found in Australia, SouthAmerica, Africa, New Guinea,Malesiaand broadersout-heast Asia (fig. 2). All46 Ausiralianspecies are restrictedto rainforest(Fig.2 and Table 3; Hyland 1989),wittr haUitatsvaryingfrornnorth Queenslandseasonalrainforestswith Agathis to diier rainforestsof northernNSW ind southern and central Queensland,to monsoonforestsin northernQueensland,NorthernTerritory and the K_ilber-leyregion of WesternAustralia(Hyland 1989).Litsea consistsof 100 species(Hyland 1989) found in Australia, New Zealand,South America, New Guinea,throughMalesia and southeastAsia to -J"pul Gig. 2). All but one of the I I Australianrp.ri.r are rainforesttrees (Fig.2 andTable3; Hyland1989). Five speciesof Endiandra have been identified from sedimentsat CambalongCreek. The fossil taxa haveroyld3d abaxialepidermalcell outlines,granularinnerpericlinalialls (Figs.4 and 5, p) and 'double' cuticularscalesconsistingof a narrowinner andbuter ridge (Figi . iand 5; o, i). Thesecharactersare typical of most ofthe 38 extantspeciesof Australian Endiandra (Christophel& Rowett 1996).The Late Paleocenespeciesof En4iandraare most closelyrelated to the 'E. pubens groupl of Christoplel & Rowett (1996) and some comparevery favourably with exta_nt E. glolosa, R. woWeiand,E. cowleyana(cf. Fig. 4 and Fig. 5). One of the fossiltaxafrom CambalongCreekhasbeenidentifiedis lieilschmiedia.Thefossil taxonhasthick (> 2.5 pm) epidermalanticlinalwalls with buttressed thickenings(Fig. 6, b) and prominent inner stomatalledges.Thesecharactersare typical of most of the I extint Australian taxa of Beilschmiedia(Christophel& Rowett 1996i.The fossil is similar in cuticle morphology to extant B. tooram and 8. recurva (cf. Fig. 6 and Fig. 7) from rainforestsof northern Queensland(Hyland 1989).Crisp et al. (1999)desciibedBeilsihmiediaasoneof the generaexhibiting,an 'Equatorialtrack', with an Afro-Indo-Malesiandistribution(Fig. 2). Mimbers of this track had been describedas the 'tropical element' of the Australianito.u Uy Burbidge (1960)and as the 'Irian Element'by Schodde(1989).Crispet al. (1999)howeveralludedto tf,e more likely Late Cretaceous, Gondwananoriginsof taxa Lxhibiting this track. A more ancient origin is supportedby the presenceof Beilschmiediafossilsin the iate Paleoceneof southeastern Australia. Twenty-two fossil cuticle fragmentsfrom CambalongCreekhavebeenidentified as one species of Cryptocarya.The fossilshave roundedepidermalanticlinalwalls and wide, butterflyshapedcuticularscales(Fig. 8, s) that are charactiristicof mostof the 46 extantAustralianspecies of Cryptocarya(Fig. 9, s; Christophel& Rowett 1996).The fossilsare similar to extaniC. bidwillii, c. clarksoniana(cf. Fig. 8 and Fig. 9) and c. cunninghamii. Three cuticle fragmentsfrom CambalongCreekhave beenidentifiedas a singlespeciesof Litsea. FossilcuticlesCMB 4-22/35/2c-28featureprominentthickenedrings of-cutiile encircling the outer surfaceof the stomates,and papillaeon the outerabaxialsuiface(Fig. 10, pa). -& These characterscompare favourably with the 'L. bennetlif group' of Christophel noweit (1996),particularlyL. connorsii(cf. Fig. l0 and Fig. I l), a taion lhat grows in rainforestsand forest marginsin northernQueenslandover an altitudinalrangeof ObO-IZOO m asl (Hyland 1 9 8 9) . PaleocENE vEGETATIoN AND ENVIRoNMENTS 2tl Figs. 4-13: electron micrographs offossil cuticles from CambalongCreek and extant Lauraceaeand Procuticles; all scab 6ari indicate l0 pm. Fossils are indicated by the prefix CMB; extant taxa folrci.o lowed by acc"*iion numbe,tsin pare'ntheses:MEL = Royal Botanic Gardens Melboume. Figs.4-5, inne-r surfacaof stomates;i = inner stomatal ledge, o = outer stomatal ledge, p = granular inner pedclinal,wall. Fig. 4: CMB2c-21; Fig. 5: Endiandra cowleyana(MELI604225). Figs 6-7, inner adaxialsurfacesof nonn"I*"outse cells. Figl 6: CMB4-9, b = buttressed irregular thickening on anticlinal wall; Fig. 1.Beil; schmiedia recurva (trAnI,tOOeZOt). Figs. 8-9, inner surfaces of stomates; s = cuticular scale. Fig. 8: CMB4-23; Fig.9: iryptocarya clarksoniana(M8L1605606). Figs l0-ll, inner_abaxialsurfacesof nonvein-course cells witli papillae; pa = outline of baseof papilla. Fig. l0: CMB4-35; Fig. I l: Litsea connor= sii (MELl6042l9). Fi!s.-12-13;ouEr surface$of stomates;su = outer surfaceof subsidiarycell, g outer suri'aceof guard cell. Fig. 12: CMB21-24 Fig. 13: Stenocarpus verticis (MEL669961). 2r2 A.J. VeonLA ANDD.R. GneeNwooD 3 PROTEACEAE 3.1 Proteaceaein Gondwana Proteaceaehave a lolg macrofossilrecord from Australia, with earliestpublishedrecordsfrom the Late Paleocene(Table 2; Carpenteret al. 1994;Vadala & Drinnan ffAy. However the palynological evidencefor the family in Australia is diverse and extendsto tiie Turonian (- 90 Ma; Dettmann & Jarzen1990; Dettmann 1994;Hillet al. 1999).This precedesthe Miocene contactphasebetweenAustralia and Eurasiaby - 75 Ma (Fig. 3). The rainforest ancestorto the extant subfamilies ('Proto-Proteaceae')was hlpothesisedby lq$ ol & Briggs (1963) to have existedsomewherein northernGondwana(Johnion & Briggs 1975; Dettmann 1989; Hill et al. 1995)prior to the Late Cretaceousand the separationof iie gondwananlandmasses(Johnson& Briggs 1975). Dettrnann& Jarzen(lgglisuggested the phylogenyand ecogeographyof extant Proteaceaeimplied evolution and diveriificit'ion of the faryily during the mid{retaceous in Gondwana.Johnson& Briggs (1981) proposedthat the earliest^palynologicalrecordsof Proteaceaeactually post-date thievolution'and first appearance of the precunors of extant subfamilies.Despitaproblemsregardingthe identification of somefossil proteaceouspollen with extantgenera(Martin 1973;Uirtin tggZ; Truswell & Harris 1982; Hill et al. 1995) fossil pollen belonging to Beauprea, Macadamia, GevuinaHicl<sbeachiaand Knightia has been reliably identified from Campanian-Maastrichtian (- 74 Ma) sedimentsof southeasternAustralia, New Zealandand Antaictica (Dettmann & Jarzen 1990, l99l). The oldest palynologicalrecord of the family in New Zealand,datesfrom the Campanianor possiblySantonian(Pole 1998).Thesepollen datareiteratethe existenceof subfamilies Proteoideae (tribe Conospermae)and Grevilleoideae (tribes Macadamieae and Knightieae) in the southeastAustralian/Antarcticsectorof Gondwanafrom at least the Senonian. Indeed,thesegeneraof Proteaceaealong with conifers including Araucaria, Dacrydium, Lagarostrobusand Podocarpusprobably constitutedoverstoreyelementsof southeasternAustralian forestsin the Late Cretaceous(Spechtet at. 1992).Carnartonra (subfamilyCarnarvonioideae),Telopea(subfamily Grevilleoideae,tribe Embothrieae)and Persoonialsubfamity lersoonioideae,tribe Persoonieae)probably formed part of the forest understorey,with Stiiltngia (subfamily Proteoideae,tribe Conospermae), Adenanthos(tribe Franklandieae)and Beauprei in scleromorphiccommunitieson the fringes of theseforests(Dettmann& Jarzenl99l ; Dettmann 1994). Forest and scleromorphiccommunitiesin southernGondwanaduring the Late Cretaceous,therefore,includedrepresentatives of four of the extantsevensubfamilies,includingthe two largestsubfamilies,Proteoideae andGrevilleoideae. _ The palynological record demonstratesthe presenceof Proteaceaein Australia during the Cretaceousand prior to any contactwith southeastAsia (Fig. 3). Indeed,Antarctica and southeasternAustralia have been integral to hlpothesesregardingthe evolution and dispersalof the extant tribes of Proteaceaeand their progenitors.NorthernGondwanahasbeenpostulatedas the origin of someof theseancestors(Johnson& Briggs 1975,1981;Dettmann1989,1994;Dettmann & Jarzen1990, l99l; Hill et al. 1995; Hill et al. 1999)and the openingof the early SouthernOceanand concomitanthabitatchangescoincidedwith diversifiCationlf the family (Dettmann1989,1994;Hill er al.1999). Table2. Published records of Paleogene andNeogene Proteaceae macrofossils fromAustralia, excluding Banlcsieaephyllum andBanksieaeformis (for tableof theseseeVadala& Drinnan1998).Records of New ZealandProteaceaeare limited to thosedescribedby pole 0 Fossiltaxon Mount Somerscoal mine, CUT-P-OI3 New Zealand(Paleocene) RegattaPoint (Early Eocene) Unidentified Proteaceae (at least8 taxa)2E Brooker (Early Eocene) Euproteaciphyllum brookeren- Extant affrni Proteaceae2s ?Lomatia28 PeleoceNe sts28 E. tasmanicum2s Buckland(Early Eocene) E. cf. brookerensis2e Livingstone, North Otago, Grevilleoideae cf. Orites NZ (Early/lvliddle Eocene) excelsa2e GoldenGrove (Middle EoProteaceae aff . Neoritess cene) Musgraveinanthuss Maslin Bay (Middle Eocene) Maslin Bay cf. ProteaceaeIII2 Maslin Bay cf. ProteaceaeII2 Maslin Bay cf. Proteaceae IV2 Maslin Bay (Middle Eocene) Mas I in ia grev i I I eo i des3 Nelly Creek(Middle EoParataxon 2'' '7,9,l2t4 cene) Parataxa Cowan and Lefroy PaBanksieae2o leodrainages (Middle/Late Eocene) Hasties (Middle/Late Eocene) Merlinleigh Sandstone, KennedyRange,WA (Middle/Late Eocene) Anglesea (late Middle Eocene) Kalgoorlie (late Middle Eocene, -39 my) Anglesea(Late Eocene) Kojonup Sandstone (Late Eocene) Lake Lefroy (Late Eocene) Nelly Creek(Eocene) Cethana(Early Oligocene) Cethana(Early Oligocene) Glencoe (mid Early Oligocene) Lea River (Early Oligocene) Lemonthyme (Early Oligocene) Leven River 2r3 vEcETATIoN AND ENvTRoNMENTS Proteaceae28 ?Lomatia2s Orites excelsa2e Darl ingia cf . Ferruginea2o Lomatiafraxinifolilo Cenarrhenes nitidazs Neor it es (immature leaves)8 Musgravea fruits8 D a r I i ng i a (? Kn ig h t i a)2 Helicia (?Darlingia)' ?F insc h iaz/ He I i cia / G rev i I I ea2 Grevillea3 ?Grevilleata Unknownla Banksieae2o Banksieae: Mus gravea2e Banksieae: Mus graveaze Telopeazo D a r t i ng i a fe r ru g i n eazo Lomatiafraxinifulia2o C en a r r h en es - B ea up r e a28 Banl<sia archaeocarpa infructescencea Series Spic igerae; Banl<sia attenuata (Series Crytostytis)a Parataxon l2lo (=Type I 2,'Lobed Prteaceae'7) ?Banksieae'o ?Darlingiats. ?Gevuininaett Lomatiatg Bivalved fruits Follicle5 Leavess Mu sgrave i nanthus a lcoe ns isa Deeolv-dissected leafts Fructihcation2s Lake Lefroy cf. Proteaceae 12 Proteaceae leaves (toothed) ta L. fraxinifofiae'te'21 Lomatia xeromorphae'te Eup ro t eac ip hy I!1 m I o ma t io des27 E. tridacnoides'' E. gevuininoideszT E. cethanicum2T E. Iinearis2T E. rugulatum2T-E. atlenualum'' E. ornamentalis2T E. integrifoliumzT _ E. microphyllum'' Telopea truncata2T lTi I kinson i a gl enc o ens istt'24 (syn. A t h er t on ia gl enc oesis)| r'24 E. papillosum'o Orites scleromorphazs O. milliganoideszs Euproteaciphyllum oolvmorohum'o 'E. ilcroiobiu^z8 E. falcatum2s E. serratumzs O. excelsioideszT Oritesto cur-P-00220'2e cur-P-00320'2e cur-P-00420 Unknownls Unknownls Unknownls Lomatiafraxinifuliats Cardwellia' ?Grevilleas Orites/Darlingias ( infl orescence) 5 Mus gravei r^r-ae ?Grevillea" Proteaceae25 Synaphea2 l4 Unspecified'rainforest taxa' e'27 Lo ma t i a fr ax i n i fo I i ae't Lomatia tin"toiio'''o cf. Lomatia2s cf. Lomatia2s ?G ewi nina e-H i cks b ea c h i a2E Grevilleoideae2E Grevilleoideae2s Grevilleoideae2s Grevilleoideae28 cf. Darlingia21'28 Grevilleoideae28 cf. OriteszT'28 Telopea lruncatazT At h erton i a (? He I ic iopsis) endocarp I l'2a Lomatia polymorlthazs Orites millieaniit" O. millisanli2s GrevillJoideae2s Grevilleoideae2s Grevilleoideae2E Grevilleoideae28 O. excelsa2T 2r4 (Early Oligocene) West Dale (?Oligocene) Yallourn Open Cut (Oligocene) Moonpeelyata (Late Oligocene/ Early Miocene) Morwell Open Cut (Oligocene/Ir4iocene) SedanCoalfield (Oligocene/Miocene) Yallourn and Morwell (Oligocene/Miocene) Manuherikia Group, NZ (Early Micoene) A.J. VaoeLA AND D.R. GnnENwooD Proteaceae cf. Stenocarpus'7 Proteaceaespp. l-617 Proteaceae aff . Conospermumu Alloxylon (918.,A. wickhamii, A. pinnata)".Stenocarpus" " Unknown Conospermum6 Proteaceaesp. l2E Grevilleoideae2s Proteaceae aff. Darlingia6 Proteaceae aff. Oriteso Proteaceae aff. Stenocarpus or Oreocallis6 Parataxon Sl 001 'Banl<sieaephyllum aff. B. laeve't3 Parataxon LC 004: '8. aff. B. Darlingia6 Orites" Stenocarpus salignus6 Proteaceae cf . AlloxylontT fastigatum'ts Parataxon LC0l2: '8. aff. B. obovatum'ts Proteaceaetaxa 58, 59, 60, 616 cur-P-0172e cur-P-0032e Mangonui Formation, North Island,NZ (Late Miocene) Melville Island(Van Diemen Sandstone) (?LatePliocene) RegattaPoint (Early Pleistocene) Marionoak Formation (Early Pleistocene) RegattaPoint B.fastigatumts B. obovatumts Unknown6 llilkins onia bilaminatal Macadamie ae; Mac adam i aze Gevuininae -H ic ks beac h ia (? Euplassa)2e Banksieae: Musgravea2e Athertonia2a Wi I kinso n i a b i I am inat al Athertoniaza ll/ i I kinso n i a b i I am inat aza 6 l|ri I kinso n ia b i lam inat at Athertonia2a Athertoniat6 ?Euplassate'21 e'2| ?Gevuinina elH i c lesb eac h i at Macadamiate'21 ?7. papuanate'2| | ? Turr i I I ia b I easda I e ite'2 Macadamia ternifolial M. tetraphvllate'rl Embothrieae2e Helicieae: Heliciaze cur-P-olg2e Beneree (early/mid Miocene) Gulgong (early/mid Miocene) Gulgong(mid Miocene) Yallourn Formation (mid Miocene) New Zealand(Miocene) Banlcsieaephyt tum laevet3 cur-P-o152e cur-P-Ot62e Grevilleasp.A ('deeply lobed')'o Grevillea sp. B ('serrate')26 Grevillea.sp. C ('deeply serrate')'o Proteaceae cf. Dilobia2a Agastachys odooratazz Banksia kingii"".^ B. strahanensis" Cenanhenes nitidazz Hakea sp.22 cf. Lomitiazz Orites revoluta22 O. truncatazz Proteaceae cf . Lomatiaz2 Telopea truncataz2 Telopea cf . mongaens is22 TeI op ea s t raha ie ns is22 Orites revoluta" O. aciculariszz O. revolutaz3 Grevillea whilianal G. pteridifu lial G. dryandrilG. rubicunda'" G rei i I I ea I o ng ifu I i a26 G r ev i I I ea d ry op hy t!926 Dilobeia thouarsii'" Agastachys odorataz2 Banl<sia saxicola - B. canei22 B. spinulosatz Ceiarrhenes nitidaz2 Hakea22 ?Lomatiaz2 Orites revolutazz O. d ive rs ifo I ia I O. mi I I igan i i22 Lomat ial Knight ial Oriles" Telopea lruncala" Telopea mongaensis22 Telopea" O. revoluta" O. acicularis2z O. revoluta23 PnleoceNn 2t5 vEGETATIoN AND ENvtRoNMENTs (RPA: Early/Middle Pleistocene) Henty lignites (?Early Pleistocene) RegencyFormation (Middle Pleistocene) Melaleuka Inlet (Late Pleistocene) O. milliganii2s Telopea tntncata" Banlcsia sp.28 O. milliganii2s T. lruncata" Banlcsia2s Agastachys odoratan Cenarrhenes nitida'o, Agastachys odorata2l Banlcsia kingii'" Hakea sp.'o Lomat ii aff . tasmanicazs A. odoratazs C. nitida2t Agastachys odorata2E Banksia saxicola - B. canei'" Hakea2t Lomatia tasmanicazE 'vonMueller(1883);"Lange "McNamara (1978);'Blackburn (1981); & Scott(1983);'Christophel "Christoph.el (198a);oBlackburn (1985);'Christophel (1987)vCarpenter et al..(1987); & Greenwood & (1990);"Rozefelds (1990);''Jordan Hill (1988);'T.owett & Christophel & Hill (1991);''Rowett(1991); ttChristophel (1992\;r6Rozefelds (1992'l;rTHill et al. (1992\;tsRowett & Merrifield(1993);rtCarpenter etat.(1999;]lCarpenter&Pole(1995);ztryjlletal.(1995);22Jordan(!995);23Jordan Qg9{;reCarqenter (1995);"McLoughlin& Hill (1996);'oPole et al. (1995);"Rozefelds & Bowman(1996);"Carpenter & Jordan(1997);'oJordan et al. (1998);"Pole(1998). 3.2 Proteaceaeafter the isolation of Australia The palynologicalrecordalso indicatesa high abundanceand diversity of Proteaceae during the (- 56 Ma: Martin 1978;Martin 1982; Paleogene,particularly during the Late Paleocene/Eocene Hill et a/. 1985). The macrofossilrecord corroboratesthesepollen data (Table 2), and fossils with cuticular preservationprovide the most reliable and unambiguousevidencefor proteaceous affinity (Carpenter & Jordan 1997; Jordan et al. 1998). Records of tribes Grevilleeae and Banksieae,which now dominatethe sclerophyllousflora of Australia, are abundantin the Paleogeneand Neogenemacrofossilrecordsof cuticle (Table 2). The oldest macrofossilsable to be attributedwith confidenceto Proteaceae havebeendescribedfrom the Late Paleoceneof the SouthernHighlandsof New SouthWales (Fig. 3). Carpenteret al. (1994) describedcuticles of Banlrsiseaephyllumtaylorii from Lake Bungarby and Vadala & Drinnan (1998) describedB. praefastigatumfrom nearbyCambalongCreek.Thesetaxa have simple leaveswith serratemargins, superficial stomates,epidermal cells with inegularly thickenedanticlinal walls and trichome baseswith somedegreeof thickening(Carpenteret al. 1994;Vadala & Drinnan 1997). These charactersare qpical of extant Banksia and Dryandra, which are indistinguishableon thesecriteria alone, and the fossils are consequentlyattributedto the proteaceousfossil genus Banlrsieaephyllum(tribe Banksieae;Cookson& Duigan 1950). The publishedfossil record of Proteaceaeis extensivethroughoutthe Cenozoicof southeastern Australia, and comprisesmainly subfamily Grevilleoideae(Table 2 and Fig. 3: Hill et al. 1995; Carpenter& Jordan 1997; Jordanet al. 1998; Vadala & Drinnan 1998).The Paleocene and Early Eocene(- 65-35 Ma) macrofossilrecord corespondsto an apparentincreasein di(Martin 1978; versity and abundanceof Proteaceaepollen during the Late Paleocene/Eocene Martin 1982;Hill et al. 1995),prior to the completeseparationof the Australian continentfrom Antarctica in the Late Eocene/EarlyOligocene(Crook l98l; Veeverset al. l99l; Wilford & Brown 1994).This precededthe Australiancrator/SundaArc collision in the l^atelMiddleMiocene(Crookl98l; Powellet al.l98l; Hall 1996,1997)by- 20-15 Ma (Fig.3). The Paleogenemacrofossilrecord of tribe Banksieae(consistingof subtribesBanksiinaeand Musgraveinae)is in concordwith the size and diversity of the tribe in the modernflora of Australia (Table 3). The record for Banksiinae(consistingof Banksia andDryandra) is extensive: 27 taxaof BanksieaephyllumandBanksieaeformishavebeendescribedfrom WesternAustralia, South Australia, Victoria and Tasmania,althoughthe identification of severalmay be doubtful (Carpenter& Jordan 1997;Jordanet al. 1998).Theserangein age from Late Paleocene(- 60 Ma) to Early Micoene (- 23 Ma; Fig. 3; alsoreviewedin Vadala& Drinnan 1998).Dettmann& Jarzen(1991) hlpothesized that both rainforest and sclerophyll membersof Proteaceaehad evolved by the Campanian/Maastrichtian(- 74 Ma'1. Indeed, the oldest describedtaxa of Banlrsieaephyllumexhibit either sclerophyllousor more mesic charactersby the Paleogene (Carpenteret al.19941'Hillet al. 1995;Vadala& Drinnan1998).The recordof Musgraveinae is alsoimpressive(Fig. 3 andTable2). Christophel& Greenwood(1987)recordedMiddle Ecoene Musgraveaflowers from Golden Grove, and at leastthree taxa of Banksieaehave been recov- 2r6 A.J. VaoeLA ANDD.R. GneeNwooD ered from the Middle/Late EoceneLefroy and Cowan Palaeodrainages in western Australia (Carpenter& Pole 1995).Two of thesetixa have been a.r..iu.a ur-rlrrrgvea (pole l99g). Mid- to late Middle Eoceneinflorescences of Musgrariih^r"also beenreJoveredfrom Angleseain Victoria and GoldenGrovein SouthAustralTa Git. 3 and Table2; Christophell9g4). Table 3: Australian and worldwide abundanceof extantgeneraand tribes to which l?uraceousand proteac fossils frqm cambalqgct"Et ur" -ost similar. Total species SpecieJl;---Endemic Tribe Laureae:Litsea ' TribeCryptocaryeae:Beilschmiediat Cryptocarya' Endiandra ' Proteaceae Tribe Oriteae:Orites' Stenocarpus3 llfe _S_tepcarpinae: Tribe Helicieae Heliciinae:Heliciaa Hollandaeinae:Hottandaeas Tribe Knightieae Knightiinae:Darlingiau Tribe Banksieae Banksiinae:Banlrsia' - 100 2OO_2SO 200-250 - 100 ll ll 46 38 g - 25 7 9 4* 7** ;no Z t*** 2 2 z 16 75 76 % s 3 * 4 taxa **2 taxaextendto NewGuineaandAru *t*l taxonextendsto NewGuineas Is.3 2.q9o1ee & Hyland(199s);3Foreman( (!?9q); t99sa);o Foreman (1995b); I fJvlano 'Hyland 5Hylind(t9-9Sb); ?c"oig" (reeeaj.-d5(1995a); e (1999b). The Cenozoicmacrofossilrecord of tribe Knightieaeis lessextensivethan that of Banksieae (Fig. 3 and Table 2). Lange(1978) identifiedDaingn/xnightia(tribe trr,ighti.ur) fr; 141;al. EoceneMaslin Bay and Carpenter& Pole (1995) de"scribedbarligia from'the Middle/Late Eocene Cowan and Lefroy Palaeodrainages in Weitern Australia. Lat-eEoceneleaveswith a combination of characterssimilar to eithei Darlingia or Orites were also described no- errgffiu (Christophel1984). The oldest Australian macrofossilrecord of tribe Oriteae is Middle Eocene Neorites from GoldenGrove in South.-Ausfialia (Christophel& Greenwood1987),althoughpofe (rqggi d;scribedleaveswith similarities to extant ornes excelsafromEarlyfviiaJi" po."ne sedimentsin of orites(Rowen & Christo\ey z9{and (Table2 andFig.3). LateMiddle Eocenespecimens phel 1990) and leavesof either Orites or Darlingia (Christophel l98a) have been described from Angleseain Victoria. Tribe Embothrieaeis preservedin the macrofossilrecord mainly as Lomatia, although the oldest fossil may be a Late Paleocenetaxon from CambalongCreek,*itrt umnites to Stenocar4as (s1e above). Jordanet al. (1998) describedtwo Early Eo."n. taxa of Euproteaciphyllutm from Tasmaniathat micromorphologicalsimilaritieswith extantLomatia(Table 2). !1v_9sjrong_ Carpenter& Pole (1995) describeda vtidOlelt ale EoceneLomatiamacrofossilfrom the Cowan and Lefroy Palaeodrainages as indistinguishablefrom the extantnortheasteueenslandrainforest anotherlate Middle EoceneLomatia fossil fto- fuftoorlie probablyhas ?*on L. fryiniftlia; the sameaffinities (Table 2; Carpenter1994). Other tribes of Grevilleoideaeare lessextensivelyrepresented in the paleoceneand Eocene macrofossilrecords(Fig. 3). Blackburn(1981)iaeniineAMasliniagreviitiordes from the Middle EoceneMaslin Bay localityas closelyrelaiedto extantGrevittei (tribe Grevilleea.; Fig. fj Many other tribes are representedin the record by fossils with suggestedor implied umiiti., (Table 2). For example,Lange(197S)identifiedMiddle Eoceneleaveswith possible affinities to Helicia (tribe Helicieae)andGrevillea(tribe Grevilleeae)from Maslin Bay (Table 2 and Fig. 3). Fossil leaveswith possibleaffinities to Grevillea havealso beendescribedfiodgl;; PnLeocrNe vEGETATIoN AND ENVTRoNMENTS 2t7 (Christophel 1984) and the Late Eocene(- 36 Ma) Kojonup Sandstonein WesternAustralia (Table 2 and Fig. 3; Mcloughlin & Hill 1996).The only publishedtaxon of subfamilyProteoideaefrom the Paleoceneor Eoceneis Middle/Late EoceneCenarrhenesnitida (tribe Conospermeae)from Hastiesin Tasmania(Table2 and Fig. 3; Jordanet al. 1998).TheAustralian macrofossilrecord of Proteaceaefrom the Oligocene(- 35-23 Ma) and the Neogene(- 23-l Ma; Table 2) indicatesthe family was equallydiversefollowing the separationof the Australian and Antarctic continentsin the Late Eocene/EarlyOligoceneand the collision of the Australian platewith the SundaArcs in the Middle Miocene(Fig. 3). Proteaceaemacrofossilsfrom CambalongCreek other than Banksieaephyllun(see above) will be formally describedelsewhere,but likely representthe earliestmacrofossilsof tribes Embothrieae,Helicieae and Oriteae describedfrom Australia (subfamily Grevilleoideae;A.J. Vadala, unpubl.). These fossil taxa have cuticular charactersdescribedby Carpenter& Jordan (1997) as typical of Grevilleioideae.Theseinclude granulationon the inner cuticle surface,hypostomaty,with stomatesaligned randomlyover the cuticle ratherthan parallel to the long axis of the leaf, and most havetrichomebasesassociatedwith at leastone epidermalbasalcell. Most of the fossil taxa are too fragmentaryto allow reliable comparisonsof leaf morphologywith extant taxa, and most lack enoughdistinctive cuticular morphologicalfeaturesto be placed with confidence in any extant genus of Proteaceae.However, the fossils all have brachlparacytic stomatesand most have trichome basesoverlying one or several epidermal cells, characters typical of the fossil proteaceousgenusEuproteaciphyllum(Carpenter& Jordan 1997; Jordanet al.1998). Each fossil taxon from CambalongCreek has a suite of cuticular charactersenablingit to be compared favourably with taxa in either one of the extant tribes Banksieae,Embothrieae, Helicieae,Knightieaeor Oriteaein subfamilyGrevilleoideae.Distributionsof the nearestliving relatives of the fossils are shown in Fig. 2 and Table 3. Carpenter(1994) describedtribal and generic characteristicsof extant taxa growing in the Wet Tropics region of north Queensland; certain of theseextant taxa sharemany similaritieswith proteaceousmacrofossilsfrom Cambalong Creek,and contributeto a solid taxonomicframeworkfor the fossils. Fossil proteaceouscuticles from CambalongCreek were comparedwith those of 19 extant taxa representingtwo subfamilies,six subtribesand five tribes on the basisof a data set of 32 micromorphologicalcharacters.Patternanalysesof the datasetswere conductedas for the Lauraceaefossils from the locality (describedabove).Two fossil taxa comparefavourablywith extant taxa in tribe Embothrieae(including the genusStenocarpus;Table l) on the basisof thickenedbandsof cuticle over the outersurfaceof guardcells and subsidiarycells (Fig. 12; g, su), which are characteristicof most extant taxa in the tribe (cf. Fig. 12 and Fig. 13; Carpenter 1994).Thesefossils sharemany micromorphologicalcharacterswith extantStenocarpussinualzs and S. verticis, including striatedor rugulatedouter abaxial surfaces,superficial stomates, prominentinner cuticular ledgesand epidermalanticlinal walls with irregularthickenings. One fossil taxon from CambalongCreekis very similar in cuticular morphologyto two extant taxa in tribe Helicieae: Hollandaea riparia (subtribe Hollandaeinae)and Helicia glabriJlora (subtribeHeliciinae:cf. Figs. 14 and 16with Figs. 15 and l7). The fossilhashighly granularinner periclinalsurfaces(Fig. 16,p), typicalof extantHelicia andHollandaea(Fig. 17,p; Carpenter 1994),and all threetaxa havevein coursesmarkedby elongatecuticular striations.The fossil taxon has a thickenedring of cuticle over the outer surfaceof the guardcells and striatedthickening over subsidiarycells (Fig. 14; g, su), which are qpical of extantsubtribeHeliciinae,includingHollandaeariparia (Fig. 15; g, su: Carpenter1994).The innerstomatalstructureof the fossil is similar to that of H. riparia in termsof prominentinner cuticularledges(Fig. 16; i), granularpericlinal walls over the guardcells and thickened,heavily granularpericlinal walls over the subsidiarycells.The fossiltaxondiffers from extantHelicieaeby not havingtrichomes, which are large and characteristicin extant taxa of the tribe, though very rare in H. riparia (Carpenter1994). A fourth fossil taxon from CambalongCreek appearsclosely related to extant Darlingia (tribe Knightieae),a genusof two speciesendemicto rainforestsof north-eastQueensland(Table 3 and Fig. 2; Hyland 1995b).Epidermalcell outersurfacesof the fossilare coveredin intricatestriations(Figs l8 and 20, st) that are tlpical of extantDarlingia (Figs. 19 and 21, st),Eucarpha andKnightia in subtribeKnightiinae(Carpenter1994;Carpenter& Pole 1995). 2t8 A.J. VapnlA ANDD.R. GnneNwooD Figs. 14-23: electron micrographs of fossil cuticles from CambalongCreek and extant lauraceae and Proteaceaecuticles; all scale bars indicate l0 pm. Fossils arp indicated by the prefix CMB; extant taxa followed by accession numbers in parentheses:MEL = Royal Botanic Gardens Melbourne; AWD = A.W. Douglas.Figs.l4-15, outer surfacesofstornates,labelledas for Figs. 12-13.Fig.14: CMB4-62a; Fig. l5: Hollandaea riparia (MEL712266). Figs 16-17, inner surfaces ofstomates; i = inner stornatal ledge, p = granular inner periclinal walls. Fig. 16: CMB4-62a; Fig. 17: Helicia glabriflora (M8A32949). Figs 1819, outer surfaces ofstomates; g = outer surface ofguard cell, st = cuticular striations. Fig. 18: CMB2c29; Fig. 19: Darlingia darlingiana (AWD629). Figs. 20-21, outer surfaces ofuichome bases; c = thickenedcollar, st - cuticular striations.Fig. 20: CMB2c-39; Fig. 2l: D. darlingiana (AWD629). Figs.22-23, outer surfacesof stomates;g = outer surface of guard cells.Fig.22: CMB4-14; Fig. 23: Orites diversdolia (MEL593824). Pnl-EoceNE vEcETATIoN AND ENVTRoNMENTS 2r9 The thickened ring of cuticle over the outer surface of the guard cells of the fossil (Fig. 18, g) is also characteristic of extant Knightiinae (Fig. 19, g; Carpenter 1994; Carpenter & Pole 1995). The fossil has large trichome bases with a raised ring (collar) of cuticle around the insertion point of the foot cell (Fig.20, c), and pronounced radiating striations (Fig. 20, st). These characters are also typical of extant Knightiinae (Fig. 2l1, c, st: Carpenter 1994). However, the fossil has cuticular thickenings at the poles of the guard cells, which are not present in extant lfuightiinae (Carpenter 1994). This fossil may indicate the existence of either Darlingia in particular, Knightiinae generally, or a close relative in the Late Paleocenein the Southern Highlands. This correlates with the Cretaceousoccurrence in southeasternAustralia of pollen similar to that produced by extant Knightia (Dettmann & Jarzen l99l; Specht et al. 1992). A fifth fossil taxon of Proteaceaefrom Cambalong Creek (Fig.22) bears many similarities to extant Orites and Neorites, and may represent a Late Paleocene relative of Oriteae, providing some support for the hlpothesis of Johnson & Briggs (1975) that Orites must have evolved by the Paleocene. This fossil is characterised by heavy thickening over the outer surface of the guard cells (Fig. 22, g), abaxial trichomes associatedwith 2-3 basal cells and striations over the vein-courses only. These charactersare typical of extant Oriteae (e.g. Orites diversifulia: cf. Fig. 22 and Fig. 23; Carpenter 1994). The Late Paleocene fossil is similar to three taxa endemic to north Queensland montane rainforests, O. megacarpa, O.excelsa, and O. fragrans in terms of granular inner cuticular surfaces and abaxial trichome structure (Carpenter 1994). The oldest described example of tribe Oriteae (consisting of extant genera Orites and Neorites) dates from the Early Oligocene (- 35 Ma; Carpenter & Jordan 1997), although Christophel et al. (1987) suggesteda late Middle Eocene (- 38 Ma) fossil from Anglesea in Victoria may have been related to extant Orites. The proteaceous taxa from Cambalong Creek described above are significant macrofossil evidence for taxa related to tribes Banksieae, Embothrieae, Helicieae, Knightieae and Oriteae in south-easternAustralia approximately 60 Ma. This closes the wide temporal gap that has existed between the earliest occurrence of these tribes in the macrofossil record and the palynological record. The oldest published macrofossils are Middle Eocene taxa possibly related to Darlingia (tribe lfuightieae: Table 2; Lange 1978), while the palynological record dates from the Campanian-Maastrichtian, including forms with affinities to extant Adenanthos, Beauprea, Stirlingia (Proteoideae), Persoonia (Persoonioideae), Carnarvonia (Carnarvonioideae) and Grevilleoideae including Gevuina-Hiclrsbeachia, Grevillea, Knightia, Macadamia, Telopea, and possibly Embothrium (Dettmann 1989; Dettmann & Jarzen 1990, l99l; Specht et al. 1992). 4 DISCUSSION The nature of the palaeobotanical record, particularly the macrofossil record with identifications based on cuticular characters,reiterates the ancient nature of Lauraceaeand Proteaceaein Australia. Johnson & Briggs (1981) hypothesisedthat most of the tribes and subtribes of Proteaceae had evolved by the beginning of the Late Cretaceous, well before any known fossils with the characteristics of extant Proteaceae.The published macrofossil records (Table 2) correspond on a more general level with the hypothesis of Late Cretaceous diversification of Proteaceae in southern high latitudes (Dettmann & Jarzen l99l). Macrofossil and palynological evidence support an ancient presence of Proteaceaein Gondwana (- 90 Ma; Dettmann 1989, 1992, 1994; Dettmann & Jarzen l99l; Hrll et al. 1995), notwithstanding the unpublished Late Paleocene taxa from Cambalong Creek. The palaeobotanical record is consistent with the hypothesis of Dettmann & Jarzen(1990) that evolution and initial diversification of several clades within Proteaceae (e.g. subfamilies Proteoideae and Grevilleoideae) occurred in the region of the embryonic Southern Ocean before the separation of Australia from Antarctica. The palynological and macrofossil records are consistent with the hypothesis of Johnson & Briggs (1975) that Proteaceae originated before the Middle Cretaceous as part of a mesic, moist forest flora and dispersed with the breakup of Gondwana. This is reflected in biogeographic data superimposed onto phylogenies based on Proteaceaechloroplast sequences(atpB gene and the atpB - rbcL intergenic spacer), which evince divergence of the major groups in the family prior to or during the break-up of Gondwana (Hoot & Douglas 1998). By contrast, many published Cenozoic lauraceous fossils from Australia are younger than those at Cambalong Creek, Nerriga and Angle- 220 A.J. VnonLA ANDD.R. GneENwooD sea described abole (Table l). The more reliable of these descriptions, such as Eocene/Oligocene and Miocene Cryptocarya (Leisman 1986; Rowett l99l;'Pole et al. 1993) and Endiandra (Rowett l99l), all nevertheless precede the 'contact phase; or collision between Australia and southeastAsia in the Miocene (Fig. 3; Truswell et at. lggT; Metcalfe 1990; Hall t996, t997). _ The long Australian fossil record of Lauraceae and Proteaceae demonstrates that the preCenozoic distribution of these plant taxa throughout the Australian region was significant to the current blogeography of these groups, as reviewed for the Austral landmassesly Drinnan & Crane (1990). These distributions arose before the Cenozoic, and prior to any floristic ex-distribution changes between the Australian Craton and Malesia. The fossil record and extant of Proteaceae and Lauraceae consequently imply neither family was introduced into the Australian Craton via Malesia subsequent to the Miocene contact phase. Australia and South America are centres of diversity of extant Proteaceae. Forty-six of 79 genera and 1100 of approximately 1700 species of Proteaceaeare found in virnrally all excepithe most arid habitats in Australia (Douglas 1995). Most taxa of the sclerophyllous subtribe Banksiinae (the fossil record of which extends to the Late Paleocene, - 58-60 Ma; Fig. 3) are endemic to Australia, and most are restricted in distribution to the Southwest botanical province of Western Australia (George 1999a, b). This high degree of diversity and endemism reiterates a long evolutionary histo[ for the Proteaceaeon the Australian landmass. 'floristic elements' _ Martin (1981) recognised a disjunction between traditional concepts of defined on the basis of extant distributions, and the Cenozoic fossil record, which indicates that 'tropical' many extant northern Australian taxa with distributions have been in Australia since the Paleogene. Martin (1981) also proposed that the Cenozoic fossil record of such taxa exemplified continual floral evolution closely linked to climatic change since the Cretaceous,rather than indicating recent migrations from the Malesian region. The temporal and spatial extent of the macrofossil record reviewed here indicates this could be the case lor genera with a long fossil history in Australia and even longer records in Europe and North AmJrica, specifically-n"its_chmiedia,Cryptocarya and Endiandra, for example. Beilschmiedia, Cryptocarya, Endiandra, Litsea and Neolitsea are all typical tree components of tropical and subtropical forests in Australia, with all but Beilschmiedia also typical of warm temperate forests (Specht l98l). These closed forests and sclerophyll communities of Australia contain a flora that-specht (19-81)suggested must be regarded as of ancient origin. The palynological record indicates that.these closed forests were present over most of southern and central Australia in the Paleogene;however, there is no extant equivalent vegetation for comparison with most of these foisil assemblages (Martin l98l). The macrofossil record for Lauraceae and Proteaceae presented is contrary to the thesis of Herbert (1932) that the rainforests of north Queensland are 'essentially Maliysian'. The 'palaeotropic element' was defined on the basis of presumed Malaysian or iropicai origin (Herbert l_932, 1967). Proteaceae and Lauraceae constitute part of this northeast Queensland rainforest flora, but Proteaceae first appeared in the Australian region soon after the early diversification of the family in Middle or Late Cretaceous.The flora ofAustralia at that time was probably the result of a gradual shift in floristic composition of plant communities from the Neocbmian to the Senonian, with mostly deciduous gymnosperm communities dominated by forms with cosmopolitan Jurassic affinities and with no modern analogues(Hill er al. 1995) disappearing and bein_greplaced by angiosperms (Mcloughlin et al. 1995). The Australian palaeobotanicil record of Lauraceae is not as ancient, but neverthelessindicates the presenceof ihe family in southeast Australia many millions of years prior to contact between the Australian craton and Malesia (Fig.3). The paleobotanical evidence presented in this review reiterates in a specific sensethe importance of the Cretaceous flora of Gondwana to the biogeography of some extant Austral angiosperns, discussedin detail by Drinnan & Crane (1990). Barlow (1981) suggestedthat the temperate and subtropical rainforests of easternAustralia were derived from an ancient Gondwanan flora. The Gondwanan origins of at least some extant rainforest taxa in northeastern Australia previously considered to be of Indo-Malayan or Malesian origin suggestedby the fossil record presented here was also indicated by the detailed ecological data of Webb et al. (198a). This plant macrofossil record complicates earlier concepts of the extant flora of Australia being com- ANDENvTRoNMENTS Pnr-EoceNevEGETATToN 221 posed of discrete 'elements', some of which were defined by distinct Malesian, Antarctic or 'autochthonous' origins. ACKNOWLEDGEMENTS The authorsare grateful to AssociateProfessorIan Metcalfe for the opportunityto contributeto this volume, to AssociateProfessorAndrew Drinnan and Dr StephenMcloughlin for valuable commentsand suggestionson the manuscript,to ProfessorJamesRossand the staff of the Herbarium, Royal Botanic GardensMelbournefor use of the extant Lauraceaeand ProteaceaecolFossildata lections,and to Dr Andrew Douglasfor accessto his collectionof extantProteaceae. from CambalongCreek presentedin this report was gatheredwhilst AJV was a postgraduate studentin the School of Botany, The University of Melbourne.Preparationof this report was fundedby AustralianResearchCouncilLargeGrant439802019to DRG. REFERENCES to evolutionof theangiosperms. In Audley-Charles, M.G. 1987.Dispersal of Gondwanaland: relevance T.C. Whitmore (ed), Biogeographical Evolution of the Malay Archipelago: 5-25. Oxford: Clarendon Press. Bandulska,H. 1926. On the cuticles of somefossil and recent Lauraceae.Botanical Journal of the Linnean Society47 : 383425. Bandulska,H. 1928.A Cinnamonfrom the BournemouthEocene.Botanical Journal of the Linnean Society 48: 139-147. Barlow, B.A. l98l . The Australianflora: its origin andevolution.In A.S. George(ed'1,Flora of Australia, Volume I : 25-7 6. Canberra:Australian GovernmentPublishingService. Belbin, L. 1987. PATN Pattern Analysis Package: ReferenceManual. Canberra: CSIRO Division of Wildlife and RangelandsResearch. Berggren,W.A., & Aubry, M.-P. 1996.A Late Paleocene-EarlyEoceneNW Europeanand North Sea magnetobiochronological correlationnetwork. In R.W.O'B Knox, R.M. Confield, and R.E Dunay (eds), Correlation of the Early Paleogenein Northwest Europe. Geological Society Special Publication No. l0l: 309-352. Blackburn,D.T. 1981.Tertiary macrofossilflora of Maslin Bay, SouthAustralia:numericaltaxonomic study of selectedleaves. Alcheringa 5: 9-28. Blackburn, D.T. 1985. Palaeobotany of the Yallourn and Morwell coal seams:SECV Palaeobotanical Project, Report No. 3. Melbourne: StateElectricity Commissionof Victoria. Unpublishedreport. Australia.In R.S. Blackburn,D.T., & Sluiter,I.R.K. 1994.The Oligo-Miocenecoal florasof southeastern Hill (ed), History of the Australian Vegetation: Cretaceousto Recent; 328-367. Cambridge: Cambridge University Press. Burbidge, N.T. 1960.The phytogeographyof the Australian region.Australian Journal of Botany 8: 75209. Carpenter,R.J. 1994.Cuticularmorphologyand aspectsof the ecologyand fossil historyof North QueenBotanicalJournal of the LinneanSocietyl16:249-303. slandrainforestProeaceae. macrofossilsfromTasmania, Carpenter,R.J., & Hill, R. S. (1988). EarlyTertiary Lomatia (Proteaceae) Australia.Reviewof Palaeobotanyand Palynologt 56: l4l-150. Carpenter,R.J., Hill, R.S. & Jordan,G.J. 1994.Cainozoicvegetationin Tasmania:macrofossilevidence. In R.S. Hill (ed), History of the Australian Vegetation:Cretaceousto Recent:276-298.Cambridge: CambridgeUniversityPress. from Tasmania.Australian Carpenter,R.J.& Jordan,G.l. 1997.Early Tertiarymacrofossilsof Proteaceae SystematicBotany l0: 533-563. from the Late Pataylorii (Proteaceae) Carpenter,R.J.,Jordan,G.J.,& Hill, R.S. 1994.Banl<sieaephyllum leoceneof New SouthWalesand its relevanceto the origin of Australia'sscleromorphicflora. Australian SystematicBotany 7:385-392. Carpenter,R.J., & Pole, M. 1995. Eoceneplant fossils from the Lefroy and Cowan Palaeodrainages, WesternAustralia.AustralianSystematicBotany8: I107-l154. Chandler,M.E.J. 1964. The Lower Tertiary Floras of SouthernEngland IV: a Summaryand Survey of Findings in the Light of Recent Botanical Observations.London: Trusteesof the British Museum (NaturalHistory). 222 A.J. VeoeLA ANDD.R. GnerNwooD Chapman,F. 1921. { 9!e1chof the geologicalhistory of Australian plants: the Cainozoic flora. Victorian Naturalist 37: 127-133. Chapman,F- 1926.New or little-knownfossilsin the NationalMuseum.part XXIX. On somepaleogene and Neogeneplant remainsfrom Narracan,SouthGippsland.Proceedingsof the Royal Societyof-i, toria (newseries)38: 183-191. Christophel,D.C. l98l' Tertiary megafossilfloras of Australiaas indicatorsof floristic associations and palaeoclimate.In A. Keast (ed), Ecological Biogeographyof Australia, Volume l: 379-390. The Hague:Dr. W. Junk Publishers. Christophel,D.C- 1984.Early,lertiary Proteaceae: the first floral evidencefor the Musgraveinae. Australian Journal of Botany 32: 177-186. christophel, D.c. 1989. Evolution of the Australian flora throughthe Tertiary. plant systematics and Evolution 162:63-78. Christophel,D.C. 1994.The early Tertiarymacroflorasof continentalAustralia.In R.S.Hill (ed),History of The Australian Vegetation: Cretaceous to Recent; 262-275. Cambridge: Cambridge University Press. Christophel,D.C. & Greenwood,D.R. 1987.A megafossilflora from the Eoceneof GoldenGrove,South Australia.Transactionsof the Royal Societyof SouthAustralia I I l: 155-162. Christophel,D.C., Hlni,sr W.K. & Syber,A.K. 1987.The Eoceneflora of the Anglesealocality, Victoria. AlcheringaI l: 303-323. Christophel, D.C. & Hyland, B.P.M. 1993. Leaf Atlas of Australian Tropical Rainforest Trees. Melbourne:CSIRO Australia. Christophel,D.C., Kerrigan,R. & Rowett,A.l. 1996.The useof cuticularfeaturesin the taxonomyof the Lauraceae.Annals of the Missouri Botanical Gardens g3:419432, Christophel,D.C. & Rowett,A.I. 1996.Leaf and CuticleAtlas of AustralianLeafy Lauraceae.Canberra: AustralianBiological ResourcesSrudy. Christophel,D.C., Scriven,LJ. & Greenwood,D.R. 1992.AnEocenemegafossilflora from Nelly Creek, SouthAustralia. Transactionsof the Royal Societyof SouthAustralia lrc: eS-lO. Conran,J.G. & Christophel,D.C. 1998.A new speciesof triplinerved Laurophyllum from the Eoceneof Nerriga, New SouthWales.Alcheringa 22: 343-348. Cookson,I.C. & Duigan, S.L. 1950.Fossil Banksieaefrom Yallourn, Victoria, with noteson the morphology and anatomyof living species.Australian Journal of ScientitficResearch, SeriesB (Biotogical Sciences) 3: 133-165. Crane,P.R. 1987.Vegetationalconsequences of the angiospermdiversification.In E.M. Friis, W.G. Chaloner & P.R. Crane.(eds), The Origins of Angiospermsind Their Biological Consequences:lO7-144. Cambridge:CambridgeUniversitypress. Crisp, M.D., West, J.G. & Linder, H.P. 1999.Biogeographyof the terrestrialflora. In A.E. Orchard& H.S. Thompson(eds), Flora of Australia VolumeI; Iniroduaion. SecondEdition: 321-367. Melbourne:ABRS/CSIROAustralia. Crook, K.A-W. 1981.The break-upof the Australian-Antarcticsegmentof Gondwanaland.In A. Keast (ed), Ecological Biogeographyof Australra, Volume l: 3-14. ]-hi hague:Dr. W. Junk publishers. Deane,H. 1925.Fossilleavesfrom the OpenCut, StateBrown Coal Miie, Morwell. Recordsof the Geological Surveyof Victoria 4: 492498. Dettmann,M.E. 1989.Antarctica:Cretaceous cradleof australtemperaterainforests?In J.A. Crame(ed), Origins and Evolution of the Antarctic Biota, GeologySocietySpecialPublicationNo.47: Sq-iOj. London:The GeologicalSociety:London. Dettmann,M.E. 1992.Structureand floristics of Cretaceous vegetationof southernGondwana:implications for angiospermbiogeography.Palaeobotanist 4l : 224-233. Dettmann,M.E. 1994.Cretaceous vegetation:the microfossilrecord.In R.S. Hill (ed) History of the Aus_ tralian Vegetation:Cretaceousto Recent:143-170.Cambridge:CambridgeUniversitypress. Dettmann,ME. & Jarzen,D.M. 1990. The Antarctic/Australianrift valleyl Late Creticeouscradle of northeastern Australasianrelicts?Reviewof Palaeobotanyand Palynologt 65: l3l-144. Dettmann,M.E. & Jarzen,D.M. 1991.PollenevidenceforLate Creticeouidifferentiationof proteaceae in southernpolar forests.canadian Journal of Botany69: 901-906. Douglas, A.W. 1995. Morphological features.In P. McCarthy (ed) Flora of Australia Volume 16: EIaeagnaceae,Pro teaceae I : | 4-20. Melbourne:ABRS/GSIRoAustralia. Drinnan,A.N. & Crane,P.R. 1990.Cretaceous palaeobotany and its bearingon the biogeographyof Austral angiosperms.In T.N. Taylor & E.L. Taylor (eds)Antarctic Paleobiologt: its no6 in-tie Reconstruction of Gondwana:192-219.(New York: Springer-Verlag New york Inc. Drinnan, A.N., Crane,P.R., Friis, E.M. & Pedersen,K.R. 1990.Lauraceousflowers from the potomac Group (Mid-Cretaceous)of easternNorth America.BotanicalGazettel5l : 370-384. Duigan, S.L. 195l. A catalogueof the Australian Tertiary flora. Proceedingsof the Royal Societyof Victoria 63:41-56. PeleoceNe vEGETATIoN AND ENvtRoNMENTs 223 Eklund, H. 1999.Big Survivorswith Small Flowers: FossilHistory and Evolution of Laurales and Chloranthaceae.ComprehensiveSummariesof UppsalaDissertationsfrom the faculty of Scienceand Technology495. Uppsala:Acta UniversitatisUpsaliensis. Eklund, H. & Kvacek, J. 1998. Lauraceousinflorescencesand flowers from the Cenomanianof Bohemia (Czeck Republic, Central Europe).International Journal of Plant Sciences159: 668-686. Ettingshausen,C. von 1888. Contributions to the Tertiaryflora of Australia. Memoirs of the Geological survey of New Southwales: Palaeontology No. 2.Sydney:Departmentof Mines. Foreman,D.B. 1995a. Stenocarpus.In A.E. Orchard (ed), Flora of Australia Volume 16, Eleagnaceae, ProteaceaeI : 3 63-369. Melbourne:ABRS/CSIRO Australia. Foreman,D.B. 1995b.Helicia. In A.E. Orchard(ed), Flora of Australia Volume16, Eleagnaceae,Proteaceae I : 393-399. Melbourne: ABRS/CSIRO Australia. George,A.S. 1999a.Banksia.In A.E. Orchard,H.S. Thompson& P.M. McCarthy(eds),Flora of Australia Volumel7B, Proteaceae3, Hakea to Dryandra: 175-251.Melbourne:ABRS/CSIRO Australia. George,A.S. 1999b.Dryandra.In A.E. Orchard,H.S. Thompson& P.M. McCarthy (eds),Flora of Australia VolumeI78, Proteaceae3, Hakea to Dryandra:251-363. Melbourne:ABRS/CSIRO Australia. Ceorge,A.S. & Hyland,B.P.M. 1995.Orites.tn A.E. Orchard(ed),Flora of AustraliaVolume 16,Eleagnaceae,ProteaceaeI : 346-352. Melbourne:ABRS/CSIRO Australia. Greenwood,D.R., Vadala, A.J. & Douglas, J.G. 2000. Victorian Paleogeneand Neogenemacrofloras: a conspectus.Proceedingsof the Royal Societyof Victoria l12:65-92. Hall, R. 1996.ReconstructingCenozoicSE Asia. In R. Hall & D. Blundell (eds),TectonicEvolution of SoutheastAsia, GeologicalSocietySpecialPublicationNo. 106: 153-184. Hall. R. 1997.Cenozoicplate tectonicreconstructions of SE Asia. In A.J. Fraser,S.J.Matthews& R.W. Murphy (eds),PetroleumGeologt of SoutheastAsia, GeologicalSocietySpecialPublicationNo. 126: tt-23. Harland,w.B., Armstrong,R.L., cox, A.v., craig,L.E., smith, A.G. & Smith,D.G. 1990.A Geologic Time Scale 1989.Cambridge:CambridgeUniveristy Press. Herbert, D.A. 1932. The relationships of the Queenslandflora. Proceedings of the Royal Society of Queensland44: l-23. Herbert, D.A. 1967.Ecological segregationand Australian phytogeographicelements.Proceedingsof the RoyalSocietyof QueenslandTS: l0l-l I l. Hill, R.S. 1983.Evolutionof Norftoftsut cunninghamiiand its relationshipto N. moorei as inferredfrom Tasmanianmacrofossils.AustralianJournal of Botany 3l: 453465. Hill, R.S. 1986.Lauraceousleavesfrom the Eoceneof Nerriga,New SouthWales.Alcheringa l0:327352. Hill, R.S. 1988a.A re-investigationof Nothofagusmuelleri (Ett.) Patersonand Cinnamomumnuytsii Ett. from the Late Eoceneof VegetableCreek.Alcheringa 12:221-231. Hill, R.S. 1988b.AustralianTertiary angiospermand gymnospermleaf remains- an updatedcatalogue. Alcheringa 12: 207-219. Hill, R.S. 1990. Tertiary Proteaceaein Australia: a re-investigationof Banlcsiaadunca andDryandra urniformis. Proceedingsof the Royal Societyof Victoria 102: 23-28. Hill, R.S. 1991.Leavesof Eucryphia(Eucryphiaceae) from Tertiarysedimentsin south-eastern Australia. Australian SystematicBotany 4: 481497. Hill, R.S. 1992a.Nothofagus:evolution from a Southernperspective.Trendsin Ecologt and Evolution 7: t90-194. Hill, R.S. 1992b.Australian vegetationduring the Tertiary: macrofossilevidence.The Beagle,Recordsof the Northern Teruitory Museumof Arts and Sciences9: l-10. Hill, R.S. 1994.The historyof selectedAustraliantaxa.[n R.S. Hill (ed),History of TheAustralian Vegetation: Cretaceousto Recent;390_4.20. Cambridge:CambridgeUniversityPress. Hill, R.S. & Carpenter,R.J. 1991.Evolutionof Acmopyle andDacrycarpus(Podocarpaceae) foliage as inferred from macrofossilsin south-easternAustralia.Australian SystematicBotany 4:449479. Hill, R.S. & Christophel,D.C. 1988. Tertiary leavesof the tribe Banksieae(Proteaceae)from southeasternAustralia.BotanicalJournal of the LinneanSociety97:205-227. Hill, R.S. & Jordan,G.J. 1993.The evolutionaryhistoryof Nothofagus(Nothofagaceae). Australian SystematicBotany 6: lll-126. Hill, R.S.& Merrifield, H.E. 1993.An Early Tertiarymacroflorafrom West Dale,southwestemAustralia. Alcheringa 17: 285-326. Hill, R.S. & Pole, M. 1992. Leaf and shoot morphologyof extantAfrocarpus, Nageia andRetrophyllum (Podocarpaceae)species, and specieswith similar leaf arrangement,from Paleogeneand Neogene sedimentsin Australasia,Australian SystematicBotany5: 337-358. Hill, R.S. & Read,J. 1991. A revisedinfragenericclassificationof Nothofagas(Fagaceae).Botanical Journal of the LinneanSociety105:37-72. 224 A.J. VnonLA ANDD.R. GnpeNwooD Hill, R.S., Scriven,L.J.& Jordan,C.J. 1995.The fossil recordof the AustralianProteaceae. In P. McCarthy (ed), Flora of Australia Volume I6: Elaeagnaceae, Proteaceae I: 2l-30. Melbourne: ABRS/CSIROAustralia. Hill, R.S., Truswell, 8.M., Mcloughlin, S. & Dettmann,M.E. 1999.Evolution of the Australianflora: fossil evidence.In A.E. Orchard & H.S. Thompson(eds),Flora of Australia, Volumel: Introduction. SecondEditi on: 25 | -320. Melbourne: AB RS/CSIRO Australia. Holdgate,G.R. & Sluiter,I.R.K. 1991.Oligocene-Miocenemarineincursionsin the LaTrobeValley depression,onshoreGippslandBasin: evidence,faciesrelationshipsand chronology.In M.A.J. Williams, P. DeDeckker& A.P. Kershaw (eds), The Cainozoicin Australia: a Re-Appraisalof the Evidence. GeologicalSocietyof Australia SpecialPublicationNo. l8: pp. 137-157.Sydney:GeologicalSociety of Australia. Hoot, S.B. & Douglas,A.W. 1998.Phylogenyof the Proteaceae basedon atpB andatpB-rbcl intergenic spacerregionsequences. AustralianSystematicBotany I l: 301-320. Hughes,N.F. & McDougall,A.B. 1987.Recordsof angiospermidpollen entry into the EnglishEarly Cretaceoussuccession.Reviewof Palaeobotanyand Palynologt 50:255-272. Hyland, B.P.M. 1989.A revisionof Lauraceaein Australia(excludingCassytha).Australian Systematic Botany 2: 135-367. Hyland, B.P.M. 1995a.Hollandaea. In A. E. Orchard (ed), Flora of Australia Volume16, Eleagnaceae, ProteaceaeI : 39 l -393. Melbourne: ABRS/CSIRO Australia. Hyland, B.P.M. 1995b.Darlingia. In A. E. Orchard(ed), Flora of Australia Volume16, Eleagnaceae, ProteaceaeI : 356-357.Melbourne:ABRS/CSIROAustralia. Johnson,L.A.S. & Briggs, B.G. 1963.Evolutionin the Proteaceae. AustralianJournal of Botany I l: 2l61. - the evolutionand classificationof a southern Johnson,L.A.S. & Briggs, B.G. 1975.On the Proteaceae family. BotanicalJournal of the LinneanSociety70: 83-182. Johnson,L.A.S. & Briggs, B.G. 1981.Three old southemfamilies - Myrtaceae,Proteaceae and Restionaceae.In A. Keast (ed), Ecological Biogeographyof Austalia, VolumeI:429-469. The Hague: Dr. W. Junk Publishers. Johnston,R.M. 1886. Descriptionsof new speciesof fossil leavesfrom the Tertiary depositsof Mount Bischoff belonging to the generaEucalyptus,Lantrus,Quercus,Cycadites,Etc. Papers and Proceedings of the Royal Societyof Tasmania 1885:322-325. Jordan,G.J. 1995. Early-Middle Pleistoceneleavesof extinct and extantProteaceae from westernTasmania,Australia.BotanicalJournal of the LinneanSocietyI l8: l9-35. Jordan,G.J. 1997. Evidenceof Pleistoceneplant extinctionand diversity from RegattaPoint, western Tasmania,Australia.BotanicalJournal of the LinneanSociety123:45-71. Jordan,G.J., Carpenter,R.J. & Hill, R.S. 1998.The macrofossilrecordof Proteaceae in Tasmania:a review with new species . Australian SystematicBotany I l: 465-501. Jordan,G.J. & Hill, R.S. 1991.Two new Banl<siaspeciesfrom Pleistocene sedimentsin WesternAustralia. Australian SystematicBotany 4: 499-511. Jordan,G.J., Macphail,M.K., Barnes,R. & Hill, R.S. 1995.An early to middle Pleistocene flora of subalpineaffinities in lowland westernTasmania.Australian Journal of Botany 43: 231-242. Kemp, E.M. 1981.Tertiary palaeogeography and the evolutionof Australianclimate. In A. Keast (ed), Ecological Biogeographyof Australia, Volume1: 3 l-50. The Hague:Dr. W. Junk Publishers. Kershaw,A.P., Martin, H.A. & McEwenMason,J.R. 1994.The Neogene:a periodof transition.In R.S. Hill (ed), History of the Australian Vegetation: Cretaceousto Recent: 299-327. Cambridge: Cambridge UniversityPress. Klucking, E.P. 1987.Lauraceae Volume2: Leaf VenationPatterns.Berlin: J. Cramer,inderGebruder BorntraegerVerlagsbuchhandlung. Lange,R.T. 1978.SomeEoceneleaf fragmentscomparableto Proteaceae. Journal of the Royal Societyof llesternAustralia60: 107-l14. Leisman,G.A. 1986.CryptocaryorylongippslandicumEen.et sp. nov. from the Tertiaryof easternVictoria. Alcheringa l0: 225-34. Macphail, M.K. 1980.Fossil and modernBeilschmiedia(Lauraceae)pollen in New Zealand.New Zealand Journal of Botany 18: 453457 . Macphail,M.K., Alley, N.F., Truswell,E.M. & Sluiter,I.R.K. 1994.Early Teetiaryvegetation: evidence from sporesand pollen. In R.S. Hill (ed), History of the Ausnalian Vegetation:Cretaceousto Recent: | 89-261. Cambridge:CambridgeUniversityPress. Martin, A.R.H. 1982. Proteaceaeand the early differentiationof the central Australian flora. In W.R. Barker & P.J.M. Greenslade(eds),Evolutionof the Flora and Fauna of Arid Australia: 77-83. South Australia:PeacockPublications. Martin, H.A. 1973.The palynologyof someTertiary Pleistocenedeposits,Lachlan River Valley, New South Wales.Australian Journal of Botany SupplementarySeries,SupplementNo. 6: l-57. PnLEoceNEvEcETATToN ANDENvTRoNMENTs 225 Martin, H.A. 1978.Evolution of the Australianflora and vegetationthroughthe Tertiary:evidencefrom pollen.Alcheringa 2: l8l-202. Martin, H.A. 1981.The Tertiary flora. In A. Keast (ed), Ecotogical Biogeographyof Australia: Volumel: 391406. The Hague:Dr. W. Junk Publishers. Martin, H.A. 1991. Tertiary stratigraphicpalynologyand palaeoclimateof the inland river systemsin New Southwales.tn M.A.J. Williams, P. DeDeckker& A.P. Kershaw(eds'5, The Cainozoicin Auatralia: a Re-Appraisalof the Evidence.GeologicalSocietyof AustraliaSpecialPublicationNo. l8: l8l194. Sydney:GeologicalSocietyof Australia. Martin, H.A. 1994.Australian Tertiary phytogeography:evidencefrom palynology.In R.S. Hill (ed), f/istory of the Australian Vegetation:Cretaceousto Recent:104-142. Cambridge:CambridgeUniversity Press. Martin, H.A. 1998. Tertiary climatic evolution and the developmentof aridity in Australia. Proceedings of the LinneanSocietyof New South lltales I l9: I l5-136. McCoy, F. 1876.Prodromusof Palaeontologyof Victoria, DecadeIY. Geological Surveyof Victoria Special Publication 1876, 29-32. Mcloughlin, S., Drinnan,A.N. & Rozefelds,A.C.F. 1995.A Cenomanianflora from the Winton Formation, EromangaBasin,Queensland, Australia.Memoirsof the QueenslandMuseum38:273-313. Mcloughlin, S. & Hill, R.S. 1996.The succession of WesternAustralianPhanerozoic terrestrialfloras.In S.D. Hopper,J.A. Chappill, M.S. Harvey& A.S. George(eds),GondwananHeritage: Past,Present and Future of the llestern AustralianBiota:61-80. ChippingNorton:SurreyBeattyand Sons. McNamara,K.J. & Scott,J.K. 1983.A new speciesof Banksia(Proteaceae) from the EoceneMerlinteigh sandstoneof the KennedyRange,westernAustralia.Alcheringa7: 185-193. Metcalfe, I. 1990. Allochthonous terraneprocessesin SoutheastAsia. Philosophical Transactionsof the Royal Societyof London A 331:625-640. Mueller, F. von 1883. Observationson New VegetableFossi/sof the Auriferous Drifts. Melbourne:Geological Surveyof Victoria. Paterson,H.T. 1935.Notes on plant remainsfrom Narracanand Darlimurla,SouthGippsland.Proceedings of the Royal Societyof Victoria (new series) 48:67-74. Pole, M.S. 1992a.Cretaceousmacroflorasof EasternOtago, New Zealand:angiosperms.Australian Journal of Botany 40: 169-206. Pole, M. S. 1992b.Eocenevegetationfrom Hasties,north-eastern Tasmania.Austalian SystematicBota n y5 : 4 3 1 4 7 5 . Pole,M.S. 1998.The Proteaceae recordin New Zealand.AustralianSystematicBotany ll:343-372. Pole,M.S. & Bowman,D.M.J.S. 1996.Tertiaryplant fossilsfrom Australia's 'Top End.' Australian SystematicBotany9: l13-126. Pole, M.S., Hill, R.S., Green,N. & Macphail,M.K. 1993.The OligoceneBerwick quarr),flora - rainforestin a drying environment.AustralianSystematicBotany6:399427. Powell, c. McA., Johnson,B.D. & Veevers,J.J. 1981.The Early Cretaceousbreak-upof easternGondwanaland,the separationof Australiaand India, and their interactionwith southeastAsia. In A. Keast (ed), Ecological Biogeographyof Australia, VolumeI: 15-29. The Hague:Dr. W. Junk Publishers. F., Soltis,D.8., Soltis,P.S.,Zanis,M., Zimmer,8.A., Chen, Qiu, Y.-L., Lee, J., Bernasconi-Quadroni, 2., Savolainen,V. & Chase,M.W. 1999.The earliestangiosperms: evidencefrom mitochondrial,plastid and nucleargenomes.Nature 402:404407. Rowett, A.L l99l. The dispersedcuticularfloras of SouthAustralianTertiary coalfields,part l: Sedan, Transactionsof the Royal Societyof SouthAustralia ll5:21-36. Rowett, A.l. 1992. Dispersedcuticular floras of South AustralianTertiary coalfields,Part 2: Lochiel. Transactionsof the Royal Societyof SouthAustralia I l6: 95-107. Rowett,A.I. & Christophel,D.C. 1990.The dispersedcuticleprofile of the EoceneAngleseaclay lenses. In J.G. Douglas& D.C. Christophel(eds),Proceedingsof the Third International Organisation of PalaeobotanyConference:I l5-121. Melbourne:A-Z Printers. Rozefelds,A.C.F. 1990.A Mid-Tertiary rainforestflora from Capella,CentralQueensland. ln J.G. Douglas & D.C. Christophel (eds), Proceedingsof the Third International Organisation of Palaeobotany Conference, MeI bourne: 123-136. Melbourne: A-Z Printers. Rozefelds,A.C.F. 1992.The subtribeHicksbeachiinae (Proteaceae) in the AustralianTertiary. Memoirs of the QueenslandMuseum32: 195-202. Rozefelds,A.C.F. 1995.Miocene ll/ilkinsoniafruits (Hicksbeachiinae, Proteaceae) from the baseof the Yallourn Formation, LaTrobe Valley, Victoria. Papers and Proceedingsof the Royal Society of Tasmania 129:59-62. Schodde,R. 1989. Origins, radiationsand sifting in the Australasianbiota - changingconceptsfrom new dataand old. AustralianSystematicBotanySocietyNewsletter60: 2-l l. Specht,R.L. 1981.Evolutionof the Australianflora: somegeneralizations. In A. Keast (ed\, Ecological Biogeographyof Australia:783-805. The Hague:Dr. W. Junk Publishers. 226 A.J. VnonLA ANDD.R. GneeNwooD Specht,R.L., Dettmann,M.E. & Jarzen,D.M. 1992.Communityassociationsand structurein the Late Cretaceousvegetationof southeastAustralasiaand Antarctica. Palaeogeography,Palaeoclimatologi, PaIaeoecoIogt 94 : 283-309. Taylor, G., Truswell, 8.M., McQueen,K.G. & Brown, M.C. 1990.Early Tertiarypaleogeography, landform evolution, and paleoclimatesof the SouthernMonaro, N.S.W., Australia. eobeogeogiophy, falaeocliamtologt, Palaeoecologt 78: 109-134. Truswell, E.M. 1990. Australian rainforests:the 100 million year record. In L.J. Webb & J. Kikkawa (eds),Australian Tropical Rainforests:Science-Value-Meaning:7-22. Melbourne:CSIRO Australia. Truswell, E.M. 1993. changesin the AustralianTertiary in responseto climatic and phyto_Vegetation geographicforcing factors.Australian SystematicBotany 6: 533-557. Truswell, E.M. & Harris, W.K. 1982.The Cainozoicpalaeobotanical recordin arid Australia:fossil evid:nT for the _olgins of-an arid-adaptedflora. In W.R. Barker & P.J.M. Greenslade(eds),Evolution of the Flora and Fauna of Arid Australia:67-76. SouthAustralia: PeacockPublications. Truswell,E.M., Kershaw,A.P. & Sluiter,I.R.K. 1987.The Australian-South-east Asian connection:evidencefrom the palaeobotanicalrecord. In T.C. Whitmore (ed), Biogeographical Evolution of the MaIay Archipelago: 3242. Oxford: ClarendonPress. Vadala,A.J. & Drinnan, A.N. 1998. Elaboratingthe fossil history of the Banksiinae:a new speciesof Banlrsieaephyllum(Proteaceae)from the Late Paleoceneof New South Wales.Australian Sistematic Botany ll:439463. van der Werff, H. & Richter,H.G. 1996.Toward an improvedclassificationof Lauraceae. Annals of the Missouri Botanical Gardens83: 409-418. Veevers,J.J., Powell, C.McA. & Roots, S.R. 1991. Review of seafloorspreadingaroundAustralia. I. Synthesisof the patternsof spreading.AustralianJournal of Earth Sciences38: 373-389. Webb, L.J., Tracey,J.G. & Jessup,L.W. 1985.Recentevidencefor autochthonyof Australiantropica! and subtropicalrainforestfloristic elements.Telopea2: 575-589. Webb, L.J., Tracey,J.G. & Williams, W.T. 1984.A floristic frameworkof Australianrainforests.Australian Journal of Ecologt 9: 169-198. Wilford, G.E. & Brown, P.J. 1994.Maps of late Mesozoic-CenozoicGondwanabreak-up:somepalaeogeographicalimplications. In R.S. Hill (ed), History of the Australian Vegetation:Cretaceoui to Recent: 5-13. Cambridge:CambridgeUniversityPress.

Log In

Australian Paleogene vegetation and environments: evidence for palaeo-Gondwanic elements in the fossil records of Lauraceae and Proteaceae.

Australian Paleogene vegetation and environments: evidence for palaeo-Gondwanic elements in the fossil records of Lauraceae and Proteaceae.

Related Papers

RELATED PAPERS