tropical refugia

tropical refugia Tropical refugia are patches of vegetation and their faunal inhabitants which remained unchanged while others were modified during Quaternary ice ages. The refuges were mainly areas of lowland rainforest that survived when savanna became more widely distributed in colder and drier episodes (see rainforests and ice ages). They were not places to which plants and animals retreated to shelter from hostile conditions.

Tropical refugia have been identified by means of a combination of biogeographical and palaeoecological data. Biogeographical studies have demonstrated the current existence of rainforest areas with high levels of diversity (richness) and endemism (localized occurrence) among their component species. Populations of these species are normally characterized by a marked degree of morphological similarity. Between these areas are contact zones containing fewer species, whose members exhibit more differences of form. Information about past ecology has come mainly from examination of the composition of superficial deposits beneath rainforest and the pollen content of Quaternary sediments.

At present, species diversity in areas considered to have been tropical refugia is perpetuated by a genial climate, varied environment, abundant biological productivity, and a substantial amount of competition and predation within the biota. These factors produce smaller niche breadths and higher niche overlaps in the species.

However, even if such mechanisms have been operational in the past, it is possible that current species richness had another contributory cause. This has been hypothesized as evolutionary activity associated with the persistence of refugia in the tropics during cold stages of the Quaternary. While the notion of such localities has been in existence for a considerable time, explanation of the Theory of Tropical Refugia is generally credited to Jürgen Haffer in 1969. This theory neither suggests that speciation has been confined to refuges, nor that the entire complement of present-day species has evolved during the past 2million years. However, it does propose that the Quaternary has witnessed the most recent, and possibly the most fruitful, of a series of evolutionary episodes which have occurred over geological time. The theory is underpinned by evidence that the climatic cycles of the Quaternary have influenced biogeography in the tropical as well as the temperate part of the Earth. As Ghillean Prance observed in 1982, the major point is to what extent did oscillations in the extent of tropical forest during the Quaternary affect speciation? He also pointed out that analogous families and genera of tropical rainforest plants to those of today had developed by the Early Tertiary (about 60 million years ago) and that over this time-span, episodes of continental drift and suturing have influenced both the distributional and evolutionary strategy of biota.

Haffer stated in 1982 that there are three potential results of a contraction of forest cover into refugia:(1) species will become extinct because they are unable to resist heightened competitive and predatory pressures in a reduced area;(2) they will continue to exist without change; or(3) they will survive and evolve into new subspecies and species by means of genotypic (genetic constitutional) and phenotypic (physical) modifications.A combination of outcomes is likely to occur in many cases.

The most prevalent method of speciation is that brought about by the presence of barriers to reproductive contact between members of the same plant and animal species. Such barriers are some kind of geographic feature which isolates parts of populations, and leads to allopatric (geographical) speciation. Refuge theory suggests the development of ephemeral habitat barriers which have caused breaks in the distribution (range) of species. Thereafter, separated parts of the populations evolved in isolation. Haffer pointed out that certain animals (small mammals, for example) can speciate in under 10 millennia. Such a period is considerably shorter than those that have been assigned to cold stages of the Quaternary, and would have enabled new species to emerge. Additionally, a certain amount of sympatric speciation (involving reproductive isolation between segments of one population) and parapatric speciation (whereby a population enters a new habitat but the barriers between it and its parent population are related to gene flow rather than physical conditions) probably took place in refugia, as did some evolution of different kinds of plants and animals arriving by dispersal from long distances away.

The theory also states that re-expansion of the tropical rainforest flora and fauna, mainly at the expenses of that of savanna, during intervening temperate stages of the Quaternary will have led to contact with populations of other refugia. The amount of species interaction in these contact zones will have varied, depending upon the degree of differentiation achieved by particular plants and animals. In this context, biota evolve at differing rates and the duration of episodes of isolation has been unequal. A sexually and ecologically discrete species which evolved in a refugium might extend its range over a considerable part of the newly enlarged habitat. During the next episode of inhospitable climate, both this distribution and that of its parent species would undergo additional fragmentation. As John Flenley has stated, the amalgamation of formerly separate patches of tropical rainforest in which genetic isolation had taken place would produce groups of closely related species. Such occurrences are frequent in this biome today.

It is worth noting that the effects of a colder and drier climate would have had different biogeographical and evolutionary consequences at altitude than over latitude. On mountains, the downward movement of life zones during the ice ages lessened the efficacy of the barriers between them, facilitated the dispersal of organisms, and militated against evolutionary change. More effective isolation will have been achieved during interglacials, when upward migration of zones took place and species differentiation was fostered.

In South America, rainforest seems to have survived cold episodes beside major rivers and in several limited areas of tropical lowland. Such localities have the most abundant precipitation today, and have relief that has endured with little change since the Late Tertiary. They are likely to have had the highest rainfall and most fertile soils during colder and drier phases. Thus they probably were recurrent refugia, a phenomenon that would have enhanced the differentiation of certain of the species populations within them. The floral diversity of Amazonia has been attributed to refugia by Prance, while Haffer has interpreted its avifaunal richness in the same light. Other animals (butterflies and lizards, for example) possess similar patterns of species richness and endemism, which can be explained in a similar way.

In Africa, sectors of coast in Upper Guinea, Cameroon and East Africa, together with parts of Gabon and Zaïre, are thought to have had rainforest refugia. Distributional data on amphibians, reptiles, mammals and birds demonstrate centres of species diversity and endemism. There is currently a distinct western and eastern distribution of rainforest in Africa. According to R. E. Moreau, links between the two areas during former periods with higher precipitation may afford an explanation for the pattern of bird speciation in Africa. It is interesting to note that Paul Richards has suggested that the diversity of African rainforest was reduced as a result of arid episodes.

Donald Walker has speculated that the diversity of Sunda (mainly Malaya, Sumatra and Borneo) and New Guinea rainforests in South-East Asia was not significantly increased during arid episodes of the Quaternary. In Sunda, niche subdivision, and in New Guinea environmental differences promoting rapid allopatric speciation, may be responsible for current diversity levels.

Doubts have been expressed concerning the validity of the refuge theory. Some of its opponents contend that loci of current species diversity in tropical rainforest could be explicable in terms of existing ecological (demographic and physical, for example) factors. Contact zones would then relate to gradients of these factors. Refugia should be assessed against the backdrop of this and other hypotheses regarding evolution in tropical lowlands. Alternatives include the island and river theories. The former states that modifications to the geography of the Earth during the Tertiary caused the separation of substantial populations which evolved into species that are extant. The latter propounds the view that the development of large rivers (such as the Amazon) early in the Quaternary led to the fragmentation of hitherto widely distributed populations, thereby bringing about speciation on opposite banks.

R. L. Jones

Bibliography

Bradley, R. S. (1985) Quaternary palaeoclimatology: methods of palaeoclimatic reconstruction. Allen and Unwin, Boston.
Brown, K. S. (1987) Conclusions, synthesis and alternative hypothesis. In Whitmore, T. C. and Prance, G. T. (eds) Biogeography and Quaternary history in tropical America, pp. 175–96. Clarendon Press, Oxford.
Flenley, J. R. (1979) The equatorial rain forest: a geological history. Butterworth, London.
Haffer, J. (1969) Speciation in Amazonian forest birds. Science, 165, 131–7.
Haffer, J. (1982) General aspects of the refuge theory. In Prance, G. T. (ed.) Biological diversification in the tropics, pp. 6–24. Columbia University Press, New York.
Moreau, R. E. (1966) The bird faunas of Africa and its islands. Academic Press, New York.
Prance, G. T. (ed.) (1982) Biological diversification in the tropics. Columbia University Press, New York.
Richards, P. W. (1973) Africa, the ‘odd man out’. In Meggers, B. J., Ayensu, E. S. and Duckworth, W. D. (eds) Tropical forest ecosystems in Africa and South America, pp. 21–6. Smithsonian Institution Press, Washington.