Portal de Periódicos da CAPES

Evolutionary Ecology: When Relatives Cannot Live Together

2006; Elsevier BV; Volume: 16; Issue: 16 Linguagem: Inglês

10.1016/j.cub.2006.07.030

1879-0445

T. Jonathan Davies,

Plant and animal studies

The importance of competition in determining species coexistence has been much debated. A phylogenetic analysis of sedges indicates that competitive exclusion may inhibit co-occurrence among closely related species, but not among more distant relatives. The importance of competition in determining species coexistence has been much debated. A phylogenetic analysis of sedges indicates that competitive exclusion may inhibit co-occurrence among closely related species, but not among more distant relatives. It is commonly accepted that there is a limit to the similarity of co-occurring species — the theory of 'limiting similarity' [1Hutchinson G.E. The paradox of the plankton.Am. Nat. 1961; 95: 137-145Crossref Google Scholar, 2MacArthur R. Levins R. The limiting similarity, convergence, and divergence of coexisting species.Am. Nat. 1967; 101: 377-385Crossref Google Scholar]. However, the deceptively simple question of how similar two species may be to each other before one competitively excludes the other has proven remarkably divisive [3Lewin R. Santa Rosalia was a goat.Science. 1983; 221: 636-639Crossref PubMed Scopus (84) Google Scholar, 4Dayan T. Simberloff D. Ecological and community-wide character displacement: the next generation.Ecol. Lett. 2005; 8: 875-894Crossref Scopus (410) Google Scholar]. One major obstacle to predicting when one species should displace another has been the difficulty in differentiating between the relative importance of two ecological forces: the fitness advantage a species gains by occupying a favourable habitat (habitat filtering), versus the cost of competition from species sharing similar resource requirements in that habitat (competitive exclusion). In the absence of competition, species with similar ecological requirements will have high fitness in similar environments. But as resources are divided among an increasing number of species, relative fitness will decline, potentially resulting in the displacement of the competitively inferior species. Phylogenetic methods offer a powerful approach for evaluating ecological patterns and, in particular, provide a robust null model for exploring the association between species distributions and ecological traits by controlling for the effects of shared ancestry [4Dayan T. Simberloff D. Ecological and community-wide character displacement: the next generation.Ecol. Lett. 2005; 8: 875-894Crossref Scopus (410) Google Scholar, 5Harvey P.H. Phylogenies for ecologists.J. Anim. Ecol. 1996; 65: 255-263Crossref Scopus (161) Google Scholar, 6Webb C.O. Ackerly D.D. McPeek M.A. Donoghue M.J. Phylogenies and community ecology. Annu. Rev. Ecol.Syst. 2002; 33: 475-505Google Scholar]. Competitive exclusion and habitat selectivity may lead to phylogenetic structure in co-occurring species [6Webb C.O. Ackerly D.D. McPeek M.A. Donoghue M.J. Phylogenies and community ecology. Annu. Rev. Ecol.Syst. 2002; 33: 475-505Google Scholar]. First, co-occurring species may be more closely related to each other than expected from the regional species pool — phylogenetic clustering. Second, co-occurring species may be less closely related than expected — phylogenetic overdispersion. Whether one pattern comes to dominate over the other will depend upon the relative strengths of interspecific competition, habitat filtering and the evolutionary flexibility of functional traits determining fit to the environment (Figure 1). If closely related species share similar functional traits, and the benefits in occupying a suitable habitat outweigh the potential increased cost of competing with close congeners, phylogenetic clumping is predicted. If the cost of competition is high, however, phylogenetic overdispersion may occur. Alternatively, if the traits determining habitat preference are evolutionary labile, such that more distantly related taxa share similar habitat preferences due to convergence, our predictions will change. Habitat filtering, in which species distributions are largely determined by their fit to the environment, will tend to result in phylogenetic overdispersion, whereas strong interspecific competition may remove any trend between phylogeny and co-occurrence. In a recent paper, Slingsby and Verboom [7Slingsby J.A. Verboom G.A. Phylogenetic relatedness limits co-occurrence at fine spatial scales: evidence from the Schoenoid sedges of the Cape Floristic Region, South Africa (Cyperaceae: Schoenaea).Am. Nat. 2006; 168: 14-27Crossref PubMed Scopus (119) Google Scholar] combine information from phylogenetic trees with fine scale distribution data to test for phylogenetic structure in schoenoid sedges (family Cyperaceae) in the Cape of South Africa, a global biodiversity hotspot. The floristic diversity of the Cape rivals that of the wet tropics, comprising around 9,000 species [8Goldblatt P. Manning J.C. Plant diversity of the Cape region of southern Africa. Ann. MO Bot.Gard. 2002; 89: 281-302Google Scholar]. Recent molecular evidence along with the high frequency of rare endemics in the region indicate that much of the Cape's diversity is a product of rapid in situ diversification within a few lineages [8Goldblatt P. Manning J.C. Plant diversity of the Cape region of southern Africa. Ann. MO Bot.Gard. 2002; 89: 281-302Google Scholar, 9Cowling R.M. Holmes P.M. Rebelo A.G. Plant diversity and endemism.in: Cowling R.M. The Ecology of Fynbos. Oxford University Press, Cape Town1992: 62-112Google Scholar, 10Linder H.P. Radiation of the Cape flora, southern Africa.Biol. Rev. 2003; 78: 597-638Crossref PubMed Scopus (407) Google Scholar, 11Linder H.P. Evolution of diversity: the Cape flora.Trends Plant Sci. 2005; 10: 536-541Abstract Full Text Full Text PDF PubMed Scopus (167) Google Scholar], resulting in a large pool of closely related species. Hence, understanding the geographical patterns of species richness within the Cape requires an understanding of the mechanisms that govern species co-occurrence among closely related species, where the influence of common ancestry will be strongest. Slingsby and Verboom [7Slingsby J.A. Verboom G.A. Phylogenetic relatedness limits co-occurrence at fine spatial scales: evidence from the Schoenoid sedges of the Cape Floristic Region, South Africa (Cyperaceae: Schoenaea).Am. Nat. 2006; 168: 14-27Crossref PubMed Scopus (119) Google Scholar] provide an elegant demonstration of phylogenetic overdispersion among the reticulate sheathed Tetraria, a monophyletic clade of Cape sedges. Using the evolutionary distance between species pairs from a molecular phylogenetic tree of sedges, the authors show that closely related sedges are less likely to co-occur than expected by chance, with phylogenetic relatedness setting an upper bound to co-occurrence (Figure 2). They also find greater than expected divergence in functional traits among co-occurring species. Because these traits were found to be evolutionary conserved, so that closely related Tetraria sedges are ecologically similar, Slingsby and Verboom [7Slingsby J.A. Verboom G.A. Phylogenetic relatedness limits co-occurrence at fine spatial scales: evidence from the Schoenoid sedges of the Cape Floristic Region, South Africa (Cyperaceae: Schoenaea).Am. Nat. 2006; 168: 14-27Crossref PubMed Scopus (119) Google Scholar] suggest the most parsimonious interpretation is that competitive exclusion, rather than habitat filtering, determines species co-occurrence within this clade. It is of interest that, when comparisons are extended to across all Cape schoenoid sedges, the pattern breaks down, suggesting phylogenetic scale sensitivity to the processes regulating species co-occurrence. Moreover, at even higher taxonomic levels, previous studies found evidence for phylogenetic clustering [12Webb C.O. Exploring the phylogenetic structure of ecological communities: an example for rainforest trees.Am. Nat. 2000; 156: 145-155Crossref PubMed Scopus (989) Google Scholar]. One explanation for variation with phylogenetic scale would be greater niche conservation in higher taxa, for example, if the traits defining the fundamental niche of a taxon were relatively invariable. Slingsby and Verboom [7Slingsby J.A. Verboom G.A. Phylogenetic relatedness limits co-occurrence at fine spatial scales: evidence from the Schoenoid sedges of the Cape Floristic Region, South Africa (Cyperaceae: Schoenaea).Am. Nat. 2006; 168: 14-27Crossref PubMed Scopus (119) Google Scholar] propose that the absence of evidence for increased trait conservatism at broader taxonomic levels within the sedges might imply that these traits are more important in terms of competitive interactions rather than habitat preferences. As evolutionary distance increases, species are likely to vary in an increasing number of traits, reducing the strength of competitive interactions. Therefore, competitive exclusion among close relatives would not preclude the possibility that habitat filtering influences community structure at broader taxonomic scales, but the important traits may differ. How does this new study enhance our understanding of Cape diversity? Strong competition among closely related species in a region such as the Cape, where the species pool is composed of many close relatives, will place limits on species richness, as local richness is restricted by competitive exclusion. This same mechanism, by forcing spatial divergence of closely related species, will also result in high turnover of species along spatial gradients. As Slingsby and Verboom [7Slingsby J.A. Verboom G.A. Phylogenetic relatedness limits co-occurrence at fine spatial scales: evidence from the Schoenoid sedges of the Cape Floristic Region, South Africa (Cyperaceae: Schoenaea).Am. Nat. 2006; 168: 14-27Crossref PubMed Scopus (119) Google Scholar] observe, this rather neatly fits with the observation that the high diversity in the Cape is best characterised in terms of exceptional beta diversity (species turnover), whilst alpha diversity (local species richness) remains similar to that found in other Mediterranean-type biomes [9Cowling R.M. Holmes P.M. Rebelo A.G. Plant diversity and endemism.in: Cowling R.M. The Ecology of Fynbos. Oxford University Press, Cape Town1992: 62-112Google Scholar, 10Linder H.P. Radiation of the Cape flora, southern Africa.Biol. Rev. 2003; 78: 597-638Crossref PubMed Scopus (407) Google Scholar]. However, reasons for the rapid rates of diversification observed in Cape lineages, such as the sedges, remain a matter for speculation. Might the same processes responsible for structuring ecological communities also drive speciation rates [13Schluter D. Experimental evidence that competition promotes divergence in adaptive radiation.Science. 1994; 266: 798-801Crossref PubMed Scopus (370) Google Scholar]? Slingsby and Verboom [7Slingsby J.A. Verboom G.A. Phylogenetic relatedness limits co-occurrence at fine spatial scales: evidence from the Schoenoid sedges of the Cape Floristic Region, South Africa (Cyperaceae: Schoenaea).Am. Nat. 2006; 168: 14-27Crossref PubMed Scopus (119) Google Scholar] present a convincing argument for phylogenetic structure in community membership, a likely product of competitive displacement of closely related species sharing similar functional traits. Measuring the maximum evolutionary distance between co-occurring species on a phylogenetic tree can therefore provide an estimate of limiting similarity. However, the web of competitive interactions is likely to be complex within any natural community. Considering only pair-wise interactions will tend to underestimate competitive load. For example, if the strength of competition scales with relatedness, a species co-occurring with a single close relative might experience the equivalent competitive pressure as a species co-occurring with two more-distant relatives, yet pair-wise comparisons will suggest the competitive load of the latter to be half that of the former. Generating a comprehensive model of species-co-occurrence will be challenging, requiring knowledge of phylogeny, biogeography, and ecomorphology for all species within a community.