Palynological Evidence for Tertiary Plant Dispersals in the SE Asian Region in Relation to Plate Tectonics and Climate
Total Page：16
File Type：pdf, Size：1020Kb
Tertiary plant dispersals in SE Asia 211 Palynological evidence for Tertiary plant dispersals in the SE Asian region in relation to plate tectonics and climate Robert J Morley PALYNOVA, 1 Mow Fen Road, Littleport, nr Ely, Cambs CB6 1PY, UK Key words: SE Asia, Tertiary, palynology, plants, dispersal, plate tectonics, climate Abstract cial, high sea level periods coinciding with times of rain forest expansion, and glacial, low sea levels coincided with Geological evidence for plant dispersals in SE Asia is re- periods of more strongly seasonal climates, accompanied by viewed by reference to both published, and previously un- the expansion of forests adapted to seasonal climates (such published, evidence from the time of first appearance of as pine forests) and savannah angiosperms until the Quaternary It is concluded that A major montane connection existed in South and East angiosperms did not originate in the SE Asian region, but Asia through both the Tertiary and late Cretaceous, from the dispersed into the area from West Gondwanaland Many equator to 60oN, allowing Laurasian mountain plants to dis- African plant species dispersed into India as the Indian plate perse to and from the equator throughout this period The drifted past Madagascar in the Cenomanian/Turonian, and survival of representatives of many primitive northern, many of their descendants subsequently dispersed into SE angiosperm families in lower montane forests within the SE Asia following the collision of the Indian plate with Asia in Asian region is thought to be due to the continuous pres- the middle Eocene Prior to this time, the SE Asian flora ence of this unbroken mountain belt, rather than an origin appears to have developed in some degree of isolation in SE Asia In contrast, the New Guinea mountains were There is no palynological evidence for dispersals from the formed only in the middle Miocene, from which time many Australian plate in the Cretaceous, and minimal evidence Gondwanan taxa dispersed into this area from the south for such dispersals in the Paleocene/Eocene Those well adapted to dispersal, such as Podocarpus The Sundanian Eocene flora stretched as far east as the imbricatus and Phyllocladus subsequently dispersed widely South arm of Sulawesi, and subsequent to the opening of into SE Asia, whereas those poorly adapted to dispersal, the Makassar Straits in the late Eocene, a part of this flora such as Nothofagus, never reached beyond New Guinea became stranded to the east of Wallaces Line, and probably formed a major source for other areas to the east of Wallaces Line of taxa of Sundanian and Asiatic affinity, as islands rose above sea level during the Miocene, negating Introduction the need for wholesale Miocene dispersal eastward A small number of plant taxa have dispersed westward across Plant geographers have long recognised that the Wallaces Line since the beginning of the Miocene, at 17, 14, SE Asian flora has become enriched through the 95, 35 and about 1 Ma All of the taxa involved were well dispersal of taxa from other continental regions adapted to dispersal, and emphasise that Wallaces Line has been a substantial barrier to plant dispersal from the There is clear biogeographical evidence for the Oligocene onward dispersal of mountain plants into the region Since the Eocene, plant dispersals to and from the Sunda along three trackways, from the Himalayan re- region have largely been controlled by climate The gion, East Asia, and Australasia (Steenis, Oligocene and earliest Miocene were moisture deficient over much of the region, with ever-wet rain forest climates 1934a,b, 1936) and also for lowland plants, es- first becoming widespread at about 20 Ma in the early pecially those requiring a strong dry season, Miocene, subsequent to which they have repeatedly ex- both from Asia and Australasia Proposals have panded and contracted The greatest latitudinal expansion also been made for massive dispersals from the of tropical rain forests occurred at the beginning of the mid- Sunda region to the east of Wallaces Line, fol- dle Miocene, at which time they extended northward as far as Japan Fluctuations between wetter and drier climates lowing the mid-Miocene collision between the became more pronounced in the Quaternary, with intergla- Australian and Sunda plates, although the scale Biogeography and Geological Evolution of SE Asia, pp 211-234 Edited by Robert Hall and Jeremy D Holloway © 1998 Backhuys Publishers, Leiden, The Netherlands 212 R J Morley of such migrations is debatable Plant distribu- PALAEOLATITUDE tions have also been used to suggest that there 60S 40 20 0 20 40 60N was a pre-mid Miocene contact between Aus- CENOMANIAN tralia and SE Asia, with the suggestion that this LT contact was pre-Tertiary (Steenis, 1962) Claims that the angiosperms actually evolved in the SE ALBIAN M Asian region also continue to be made, despite EY an absence of fossil evidence (Takhtajan, 1987) Biogeographical hypotheses such as these APTIAN can only find confirmation when they are based BARREMIAN on a foundation of historical geology This dis- cussion attempts to review geological evidence Fig1 Poleward migration of angiosperms during the mid for plant dispersals by examining the fossil pol- Cretaceous; latitude versus age for Barremian to len and spore record for the Tertiary (and Late Cenomanian monosulcate (¡) and tricolp(or)ate (l) pollen Cretaceous) of the SE Asian region, paying par- records (from Hickey and Doyle, 1977) ticular attention to the times of appearance in the region of pollen types exhibiting clear affini- ties with other continental regions The review is based on both published and unpublished data from both outcrops and boreholes, and records discovered so far for Australia post-date uses the plate tectonic reconstruction of Hall those from western Gondwana by 10 Ma, and (1995) for SE Asia and Daly et al (1987) for the also post-date the timing of separation of such Indian plate microcontinents from Gondwanaland The old- Geological evidence for the dispersal of low- est record for Australian monosulcate pollen is land plants and those of the uplands, are dis- from the latest Barremian or earliest Aptian cussed separately, although it must be appreci- (Burger, 1991), considerably later than its first ated that through most of the Tertiary, it is not appearance in the western hemisphere Strong always easy to determine from which of these evidence to suggest that angiosperms originated sources each pollen type is derived at tropical palaeolatitudes is provided from glo- In bringing to attention names of fossil pollen, bal plots of earliest appearances against latitude a simple convention has been followed In cases (Fig 1) which indicate parallel, slow adaptation where a fossil pollen type has been adequately to cooler, or more seasonal, climates at higher described according to the botanical code, the latitudes in both hemispheres during the course fossil name is used, generally followed, if appro- of the mid-Cretaceous The parallel diversifica- priate, by an indication of the botanical affinity tion of both pollen and macrofossils in the mid- of the parent plant, if this is known In cases Cretaceous (Hickey and Doyle, 1977) suggests where a pollen type remains inadequately de- that this pattern reflects the radiation of the early scribed according to the code, but the parent angiosperm flora, and not simply the develop- plant is known, the name of the parent plant ment of those angiosperm groups with recognis- alone is used In cases where such a pollen type able pollen might be derived from two or more plant taxa, Fossil evidence of the representation of the the taxon name is also followed by the word earliest angiosperms in the SE Asian region is type still very meagre In the past, this was due to the lack of studies from the region Recent palynological analyses of thick fluvial Lower Initial stages of angiosperm radiation Cretaceous sediments from Thailand, ranging in age from Neocomian to Aptian, have, however, The current fossil record provides no evidence failed to yield a single angiosperm pollen grain to suggest that angiosperms actually originated (Racey et al, 1994), and probable Barremian in SE Asia, as proposed by Takhtajan (1969) It is sediments from the Malay peninsula have pro- more likely that they migrated into the region duced but a single tentative record of the Truswell et al (1987) dismissed the suggestion chloranthaceous Clavatipollenites (Shamsuddin of Takhtajan (1987) that they originated on an and Morley, 1994), among an assemblage domi- isolated Gondwanan microcontinent, which nated by pollen of Classopollis spp and fern subsequently became embedded in SE Asia, on spores Current evidence, therefore, suggests the grounds that the earliest angiosperm pollen that angiosperms were much less well repre- Tertiary plant dispersals in SE Asia 213 ✚ Fig2 Mid-Cretaceous migration routes along the southern shore of Tethys Elateropollenites africaensis (o) is recorded from the Turonian of Irian Jaya (Bates, unpublished) and Papua New Guinea (Lowe, pers comm, 1987), and has a late Albian to Turonian centre of abundance in Africa and South America (outlined area); Afropollis jardinus exhibits an identical distribution pattern in the Cenomanian of Pakistan (l) to that seen in West Africa (IEDS, unpublished); Constantinisporis, Victorisporis and Andreisporis (n) are very well represented in the Turonian to Campanian of sub(palaeo)equatorial Africa, but appear in the Senonian of Madagascar (Chen 1982) and India (Venkatachala,