The relationship between sexual selection and speciation

Abstract

The intriguing relationship between sexual selection and speciation has sparked an increasing number of papers in recent years that have attempted to draw connections between the two processes. These include a myriad of theoretical and empirical studies, as well as several influential reviews (e.g. Price, 1998; Panhuis et al., 2001, Kirkpatrick and Ravigne, 2002; Ritchie, 2007; Maan and Seehausen, 2011). In particular, many researchers have speculated on whether sexual selection can lead to speciation, either alone, or in conjunction with ecological processes. Such speculation is tempting in part because of the obvious differences in secondary sexual characters that often exist between closely related species. Indeed one of the earliest and most influential of all sexual selection models, Lande’s (1981) paper “Models of speciation by sexual selection on a polygenic trait” centered its title on the role that sexual selection may play in speciation. What, however, is the evidence of a causal relationship between sexual selection and speciation? It turns out to be surprisingly ambiguous. Many have attempted to address a relationship by comparative studies, a process that can be problematic because of both methodological concerns and the fact that the presence of sexual selection is inferred by proxy (Kraaijeveld et al., 2011). Kraaijeveld et al. (2011) point out that these studies only sometimes support a correlation between sexual selection and speciation, and using a meta–analysis find only weak support for a relationship across multiple datasets. A large number of individual experimental studies have also examined various aspects of the intersection of sexual selection and speciation, but it is often unclear what would constitute a true test of the role of sexual selection in speciation, and unambiguous tests of this role are essentially absent (large scale literature analysis by Safran, Mendelson, Scordato and Symes, unpublished data). In particular, the relative roles of ecological factors and sexual selection often appear very intertwined and difficult to tease apart (case studies reviewed in Ritchie, 2007; Maan and Seehausen, 2011). Some studies have attempted to assess whether there is a difference between mate choice and species recognition (e.g. Kozak et al., 2009), but these have also found mixed results. Theoretical studies, which focus on the case of speciation with gene flow in particular, have also demonstrated a very mixed relationship between sexual selection and speciation. Many studies examining the evolution of premating isolation couch the language in terms of sexual selection, simply assuming that species recognition and sexual selection are one and the same in this regard. While this is strictly accurate by the definition of sexual selection as differential mating success, it does not address the spirit of the question of whether sexual selection processes within a species are ultimately responsible for speciation. Several theoretical studies have focused on whether sexual selection alone could account for sympatric speciation, but the general conclusion is that this is unlikely (e.g., van Doorn et al., 2004). Intriguingly, the theoretical literature has also in recent years developed an idea largely absent from the empirical literature, namely that the role of sexual selection in speciation with gene flow may often be inhibitory rather than driving (see discussion in Servedio and Kopp, 2012, this issue). Specifically, sexual selection may often eliminate variation necessary for speciation (e.g., Kirkpatrick and Nuismer, 2004), or lead to stabilizing selection that can counter divergent ecological selection to prevent speciation with gene flow (e.g., Matessi et al., 2001; Otto et al., 2008; Pennings et al., 2008).