Late Cretaceous and Cenozoic Vegetation in China, Emphasizing Their Connections With North America

Abstract

The close floristic relationship and disjunct occurrence of many plant taxa between eastern Asia and eastern North America has been given special attention by many botanists. Formerly some people thought that this close floristic link was a result of extensive migrations mainly through Beringia, but recently plate tectonic studies demonstrated that opening of North Atlantic Ocean was completed as late as middle Eocene and land bridge (Beringia) between Asia and North America was not available until Miocene. Thus extensive migrations via Beringia did not occur before Miocene. The present paper reviews Late Cretaceous and Cenozoic vegetation in China, enumerating plant megafossils of Late Cretaceous and Cenozoic Periods so far recorded there. One-half to two-thirds of Late Cretaceous plants and one-fourth to one-half of Late Eocene taxa seem similar (or nearly identical) to those of North America, and some plants are cosmopolitan in Laurasia. This indicates that these plants must have migrated directly between eastern Asia and eastern North America via Europe. So far as megafossils are concerned, after Eocene only a few Chinese species are common with those of North So most of isolated and disjunct genera of eastern Asia and eastern North America are remnants of ancient plants widely distributed all over Northern Hemisphere. Eastern Asia and eastern North America at present are two relic temperature centers of Northern Hemisphere. The recent flora of China, one of richest in world, is a mixture of northern temperate elements and tropical to (northern) subtropical elements. It is characterized by many relic genera and has a close floristic relationship with that of eastern North As to relicts, in ferns there are Angiopteris and Archangiopteris; in gymnosperms, Ginkgo, Metasequoia, Cephalotaxus, Amentotaxus, Torreya, Taxus, Pseudolarix, Tsuga, and Pseudotsuga; and in angiosperms, Magnolia, Trochodendron, Liriodendron, Saururus, Illicium, Schisandra, Nelumbo, Euptelea, Eucommia, Cercidiphyllum, Aralia, Nyssa, and Menispermum. Most of them are relic genera of Late Cretaceous. Fossils like Ginkgo and Angiopteris have been recorded from southwestern China in Late Triassic (Hsu et al., 1974); Torreya and Taxus were described from Jurassic of Europe by Florin (1958); Metasequoia was quite abundant in Late Cretaceous and Late Eocene of northeastern China. Metasequoia, Glyptostrobus, Cryptomeria, Cunninghamia, Cephalotaxus, and Amentotaxus were widely distributed in Eurasia and North America during Late Cretaceous (Florin, 1963). Now some primitive families of angiosperms, such as, Magnoliaceae, Calycanthaceae, Schisandraceae, Nyssaceae, Staphyleaceae, Platanaceae, Saururaceae, and Illiciaceae are widely distributed both in South, East, Central, and southwestern China on one hand and eastern North America on other. The genera Fagus, Lithocarpus, Castanea, Quercus, Trigonobalanus, Diphylleia, Lindera, Astilbe, Itea, Hamamelis, Cladrastis, Carya, Buckleya, Stylophorum, Penthorum, Decumaria, Gymmocladus, Ascyrum, Halesia, Chionanthus, Campsis, and Catalpa are examples of distribution pattern between these two regions (Li, 1971). Others like Wisteria, Apios, Pachysandra, Glaucidium, Gordonia, Nyssa, Chionanthus s.s., Stewartia, Panax, Pieris, Lyonia, Epigaea, Shortia, Gelsemium, Veronicastrum, Phryma, Mitchella, Triostreum, Zizania, Diarrhena, Croomia, Symplocarpus, Tipularia, Celastrus, and Diospyros are also disjunct between eastern Asia and eastern North America (Li, 1971, 1972). ' I am especially grateful to Dr. Peter H. Raven for his kind invitation and also very grateful to Dr. Daniel I. Axelrod for critical comments on manuscript. 2 Institute of Botany, Academia Sinica, Beijing, People's Republic of China. ANN. MISSOURI BOT. GARD. 70: 490-508. 1983. This content downloaded from 157.55.39.58 on Fri, 24 Jun 2016 05:20:15 UTC All use subject to http://about.jstor.org/terms 1983] HSU-CRETACEOUS AND CENOZOIC VEGETATION 491 These genera, formerly believed to number about 80, are now increased to more than 120, among which 112 are recorded from China (Wu, 1984). To these facts special attention has been paid by many phytogeographers since time of Asa Gray (1846) and later on by Good (1947), Li (1971, 1972), Wu (1984), and many others. The cause of this disjunction has been demonstrated by various authors from paleobotanical, ecological, and floristic points of view. Li (1971: 404405) thought that the present isolated and disjunct floras of eastern Asia and eastern North America appear to be remnants of a great mesophytic forest that extended over all northern hemisphere and reached arctic regions in Tertiary. Geological changes, including mountain elevation, submergence, climatic variations, glaciations, etc., have destroyed and changed floras of many lands so that this mesophytic forest of Tertiary in northern hemisphere survives in eastern Asia and eastern North America, with only relicts scattered in southeastern Europe, western Asia, and western North America. But Li did not mention route of migration. However, Takhtajan (1969, p. 175) wrote that during Late Cretaceous there existed extensive migrations and close links between Eurasia and North America, both by North Atlantic route (across landbridge that included what is now Greenland and Iceland) and by North Pacific route (through Beringia). He further suggested that link through Beringia was a particularly strong one and served as a bridge joining North America directly to eastern Asia. Anyway, judged from current paleomagnetic investigations, there was no land bridge in Bering area available between Asia and North America before Miocene. In present paper I review late Cretaceous and Cenozoic history of vegetation of China, trace successional stages of past floras in China, analyze possible causes for survival of such a large number of relics in China, and discuss relationship of Chinese past floras with those of eastern North America, particularly emphasizing migration routes between eastern Asia and North GEOLOGY AND PALEOGEOGRAPHY According to paleogeomagnetic data given by Dietz and Holden (1970) during late Paleozoic and early Mesozoic, North America, Greenland, and Eurasia were united together as a supercontinent, Laurasia. Even up to start of Tertiary, northern North America remained firmly attached to Eurasia, with Greenland sandwiched between them. Van der Linden (1975) proposed that Labrador Sea in North Atlantic opened in two stages, first in late Jurassic-early Cretaceous (about 138 to 110 million years ago) and second in early Tertiary (about 60 to 47 million years ago). By Late Cretaceous, southern Europe was situated at a lower latitude somewhere about 15 to 300N, eastern United States at 19 to 200N, and Bering area at a higher latitude at 750N (Dietz & Holden, 1970). Passing on to start of middle Eocene (about 49 million years ago) North America and Greenland began to separate from Eurasia and North Atlantic became a major ocean (Raven & Axelrod, 1974; Schuster, 1976). Europe had drifted some degrees northward (Dietz & Holden, 1970). According to these authors, India was cut free from east Africa and initiated its migration northward at beginning of Jurassic (about 180 million years ago). McElhinny (1970) presented data indicating that initial rifting took place only in Mid-Cretaceous, about 100 million years ago. So it is safe to say that Indian block migration started only after middle Cretaceous. This plate, drifting northward some 5,000 km at an average rate of 7.5 cm per year, with rates varying from 16 cm per year to less than 6 cm per year, reached Eurasia and joined up with northern Xizang (Tibet) by middle Eocene, about 40 to 45 million years ago, to become a subcontinent of Asia (Hsii, 1978). The suture of Indian plate and Eurasian plate lies along Yarlung-Zambo valleys. So southern part of Xizang is actually a part of Gondwanaland. At same time, Himalayan orogenic impulses affected by collision of Indian plate and Eurasian plate caused uplift of Himalayas and withdrawal of Tethys, Ob6 Sea in Central Asia, and Kachi Bay in western Sinjiang, during time of Oligocene (about 25 to 40 million years ago). By estimate of Deffeyes (1973), tectonic movement of Pacific plate was northwest until 60 million years ago. Until 25 million years ago, plate moved directly northward. Up to present it moved 1,500 km at an average rate of 25 cm per year. Today, for certain plates, migration rates of 16 cm per year are generally acThis content downloaded from 157.55.39.58 on Fri, 24 Jun 2016 05:20:15 UTC All use subject to http://about.jstor.org/terms 492 ANNALS OF THE MISSOURI BOTANICAL GARDEN [VOL. 70