Development and Evolution of Fern Floras of Oceanic Islands

Ferns are especially useful for the study of certain biogeographic problems of migration and speciation because the species have a high and nearly equivalent capacity for spore dispersal by air transport. Dispersal can bring spores of many species to an island from a source area; the ones that become cstablished are not a random assortment of the source species. Island floras have a major element of widely distributed source area species, although these are a minor element in the source area. The endemics on islands are mostly related to narrowly distributed source species, although these are a minor element in the island flora. In an archipelago the endemics are nearly always related to source species rather than to other species of the insular flora. Establishment of a species on an island is an individual, rather than a population, phenomenon because dispersal cannot bring the variability of a source population to the island. Adaptability of the genotype-phenotype of the single spore to a new environment and sensitivity to selection are characteristics that dominate success in migration and evolutionary potential under geographic isolation. Widely and narrowly distributed source species differ in these characteristics, and these differences account for the relative proportions of the two source groups in the insular flora and their relations to insular endemics. Re-immigration can maintain gene-flow between islands and is a deterrent to the evolution of species-flocks in an archipelago. INTEREST IN ISLAND biogeography has increased with the modern orientation toward the biology, ecology, and evolution of insular biotas (Carlquist 1965, 1967; MacArthur and Wilson 1967). This Symposium is an example of the heightened interest in islands, and its main theme reflects the new perspective on island studies. The ferns are a partictularly important groutp of plants to consider in relation to insular problems because of their high dispersibility. They comprise the Class Filicopsida (Polypodiophyta, Filicineae) with 10 major farnilies, some 200 genera, and about 10,000 species. I have restricted my discussion to the smaller and more isolated islands because it is among these that migration and endemism, as they relate to dispersal, can be examined to the best advantage. Large islands such as New Guinea, Madagascar, and New Zealand present special problems because of their size, and the West Indies and smaller islands of the IndoMalayan region are so close that their fern floras do not have attribtutes of instularity. Figure 1 provides the location of the islands referred to in this paper. The degree of geographic isolation and the age of the islands have not been considered in the context of historical geology. The existence of previotus high islands, for cxample, in the Pacific (Menard and Hamilton 1963) provides an historical background for present distribtutions withotut altering their instular nature. The biota of the present Pacific islands shows evidence of its devclopment tunder strong geographic isolation. The ages of all of the islands disctussed are sufficient to allow angiosperm endemics to have differentiated (rmany of them also have fern endemics) which wotuld also be sufficient time for the operation of the processes discussed in this paper. Dispersal and migration are distinct but complcmcntary processes. Dispersal is the transport (in ferns) of spores from one place to another; migration is the successful establishment of a species in a new area after dispersal. Migration cannot take place without dispersal, btut dispersal may occur (sometimes with high frequency) without migra-