Amphitropical Relationships in the Floras of North and South America

Abstract

Amphitropical distributions of vascular plants in the Western Hemisphere are divided into three groups: bipola or high-latitude, with about 30 species; temperate, with about 130 species; and desert, with a substancial number. Close relationships of this sort can be explained either by divergence from a common tropical ancestor or by origin on one side of the tropics and subsequent migration to the other. The latter is true for the great majority of groups discussed. These may have crossed the tropics in a single jump or by direct migration. The plant communities involved are relatively recent in origin. The amphitropical disjuncts are drawn from relatively few families and mostly are plants that occur in open habitats such as seacoast or seasonally moist places where establishment would be relatively easy. Woody plants and even herbs of closed communities are scarcely represented. Animals by and large do not have analogous amphitropical distributions as they would be expected to if the plants migrated by a land bridge or mountain chain that took them directly through the tropics. Nearly all of the plants are self-compatible; the arrival of a single seed in a suitable habitat is sufficient for the start of a new disjunct population. Animals, being predominatly bisexual, are not as likely to establish colonies in new areas to which they are rarely dispersed as are autogamous plants. Many of the disjunct patterns correspond with the migration routes of birds, which must occasionally carry seeds between one hemisphere and the other. The floras of extratropical North and South America have been distinct since at least the middle Cretaceous and are still very different at present; the few common or closely similar species are discordant elements superimposed on this pattern. Despite this, in at least 200 species and species-groups of plants, the relationships between North and South American populations are very close. Many of these are annual herbs of rapidly evolving groups in which chromosomal differences often accumulate quickly. Nevertheless, crosses between North and South American plants in several groups have demonstrated a relatively high degree of chromosome pairing and some fertility. The bipolar species mostly come from the Northern Hemisphere; their most likely time of dispersal seems to have ben the Pleistocene. About 85 per cent of the temperate species may likewise have come from the north, possibly in the late Pliocene and Pleistocene. A majority of the desert species may have come from the south, possibly as recently as post-Pleistocene time in some cases. Sporadic long-distance transtropical dispersal seems to be the best hypothesis to account for all these facts.