Has human-mediated habitat fragmentation driven rapid genital divergence in Gambusia spp. inhabiting tidal creeks across The Bahamas?

Background/Question/Methods

Anthropogenic environmental modifications can alter selection in natural populations with potential evolutionary consequences. Male genital morphology may represent a trait particularly susceptible to such impacts. Genitalia evolve quickly and divergently in internally fertilizing organisms; postcopulatory sexual selection largely bears responsibility. At least two major questions remain unanswered: (1) Ecological variation commonly alters sexual selection, but can ecological divergence ultimately, though indirectly, drive genital evolution? (2) Genitalia are one of the most rapidly evolving animal traits, but how quickly do genitalia evolve in natural populations? We investigated both questions within the context of recent human-induced habitat fragmentation.

Over the last 40 years, road construction has fragmented many tidal creeks across Bahamian islands, restricting or eliminating hydrological connectivity with the ocean, resulting in a dramatic reduction of predatory fish. Recent work has demonstrated evolutionary divergence of male genitalia associated with different levels of predation risk in small, livebearing fishes, likely by altering the context of sexual selection. We employed geometric morphometric methods to investigate whether human-mediated environmental change has driven rapid divergence of male genitalia by comparing genital shape between fragmented and unfragmented tidal creeks within three closely related Gambusia species (family: Poeciliidae) across six Bahamian islands. 

Results/Conclusions

Overall patterns of male genital shape were largely associated with historical factors: island of origin, species, and allometry with body size, as expected. We did not observe consistent, overall effects of habitat fragmentation on genital morphology across species and islands. We did, however, find evidence for unique effects of habitat fragmentation across species and islands. Thus, human-induced habitat fragmentation may have driven rapid divergence of male genitalia in some cases, though genital shape did not diverge between fragmentation regimes in a consistent way across species and islands. Tidal creeks are dynamic systems and predation intensity is not the only factor that differs among tidal creeks (resource availability, salinity, water depth, and temperature can all differ). Thus, selection may not be acting uniformly within the same fragmentation regime. Forty years (40-80 generations) also represents a very short timeframe to observe significant evolutionary change, and some other phenotypic traits investigated within the context of tidal creek habitat fragmentation in Gambusia have yielded similarly mixed results. Future work should investigate whether observed genital divergence has a genetic basis, and attempt to disentangle how variation in different ecological factors may contribute to genital divergence.