Parallel evolution and ecological selection: replicated character displacement in spadefoot toads

Amber M. Rice () 1 Aaron R. Leichty 1 David W. Pfennig 1 0 Present address: Department of Animal Ecology, Evolutionary Biology Center, Uppsala University , Uppsala , Sweden 1 Department of Biology, University of North Carolina , Chapel Hill, NC , USA character displacement in spadefoot toads Articles on similar topics can be found in the following collections Receive free email alerts when new articles cite this article - sign up in the box at the top right-hand corner of the article or click here - Subject collections Email alerting service To subscribe to Proc. R. Soc. B go to: http://rspb.royalsocietypublishing.org/subscriptions Parallel evolution and ecological selection: replicated character displacement in spadefoot toads Ecological character displacementtrait evolution stemming from selection to lessen resource competition between speciesis most often inferred from a pattern in which species differ in resource-use traits in sympatry but not in allopatry, and in which sympatric populations within each species differ from conspecific allopatric populations. Yet, without information on population history, the presence of a divergent phenotype in multiple sympatric populations does not necessarily imply that there has been repeated evolution of character displacement. Instead, such a pattern may arise if there has been character displacement in a single ancestral population, followed by gene flow carrying the divergent phenotype into multiple, derived, sympatric populations. Here, we evaluate the likelihood of such historical events versus ongoing ecological selection in generating divergence in trophic morphology between multiple populations of spadefoot toad (Spea multiplicata) tadpoles that are in sympatry with a heterospecific and those that are in allopatry. We present both phylogenetic and population genetic evidence indicating that the same divergent trait, which minimizes resource competition with the heterospecific, has arisen independently in multiple sympatric populations. These data, therefore, provide strong indirect support for competitions role in divergent trait evolution. 1. INTRODUCTION Ecological character displacement comes about when individuals most dissimilar from the average resourceuse phenotype of another species are selectively favoured, thereby causing the species to diverge in resource use and associated traits (Slatkin 1980; Taper & Case 1992). Because this process has long been seen as crucial in explaining how closely related species are able to coexist in the same habitat (Brown 1995), considerable effort has gone into documenting character displacement in natural populations (Brown & Wilson 1956; Grant 1972; Schluter 2000a; Dayan & Simberloff 2005). Much of this evidence is indirect (but see Grant & Grant 2006), in that character displacement is often inferred from patterns of trait expression in different species and populations. For example, because selection to diverge from a heterospecific competitor acts only in areas where the two species actually co-occur (i.e. sympatry), character displacement should produce a distinctive pattern in which potential competitors are more dissimilar in sympatry than in allopatry, and in which sympatric populations within each species differ from conspecifics in allopatric populations (Brown & Wilson 1956; Grant 1972; Endler 1986; Schluter 2000a; but see Goldberg & Lande 2006). These patterns of geographical variation are widely used to infer that Electronic supplementary material is available at http://dx.doi.org/10. 1098/rspb.2009.1337 or via http://rspb.royalsocietypublishing.org. ecological character displacement has taken place (reviewed in Schluter 2000a; Dayan & Simberloff 2005). Some of the strongest indirect support for character displacement comes from studies showing that divergent phenotypes occur in multiple sympatric populations (Schluter & McPhail 1993). Such data are generally taken as evidence of parallel character displacement. Parallel character displacement is a special case of parallel evolution that occurs when a similar trait evolves repeatedly in closely related, independently evolving lineages. Parallel divergence is compelling evidence that selection has promoted trait evolution (Clarke 1975; Endler 1986; Schluter & Nagel 1995). Thus, data showing that a divergent trait has arisen repeatedly in closely related, independently evolving sympatric populations would provide strong support for competitions role in promoting divergence between species and populations. Demonstrating that alleged instances of parallel character displacement have occurred cannot be accomplished, however, by merely showing that a divergent phenotype is present in multiple sympatric populations. Such a pattern can arise for evolutionary reasons other than selection (Grant 1972; Arthur 1982; Losos 1992). In particular, gene flow may produce a similar pattern, but this possibility is rarely considered (for exceptions, see Hansen et al. 2000; Marko 2005; Matocq & Murphy 2007). For example, a divergent phenotype may originate in a single ancestral sympatric population, either because of selection to lessen resource competition (i.e. character displacement) or possibly even because of genetic drift. If high levels of gene flow connect A. M. Rice et al. Parallel character displacement numerous sympatric populations, or if colonization into sympatry is frequent, then multiple sympatric populations may come to exhibit the divergent phenotype (figure 1a c). However, in all of these derived sympatric populations, the presence of the divergent phenotype would be due to gene flow and not to selection promoting parallel evolution. Moreover, if the divergent phenotype arose in the ancestral sympatric population through genetic drift, then character displacement would have played no role in its evolution, despite the presence of this phenotype in multiple, sympatric populations. Thus, in the absence of information on population history, the presence of a divergent phenotype in multiple sympatric populations cannot necessarily be construed as evidence that selection has promoted the repeated evolution of divergent phenotypes in sympatric populations; i.e. that parallel character displacement has occurred. Indeed, as this example illustrates, non-selective processes may ultimately be more important than selection in explaining why species and populations differ. Here, we test for parallel character displacement in spadefoot toads (Spea multiplicata). In the San Simon Valley of southeastern Arizona and southwestern New Mexico, USA, this species potentially breeds in the same ponds with a congener, Spea bombifrons (figure 2). In this region, both species co-occur below 1350 m in elevation (hereafter termed sympatry). At higher elevations, by contrast, only S. multiplicata is present (hereafter termed allopatry; Pfennig et al. 2006). Larvae of both species may develop into either a smallheaded omnivore m (...truncated)