Ecological drivers of species radiations in the genus Limia (Teleostei, Poeciliidae) in the West Indies

Abstract

The archipelagoes of the West Indies in the Caribbean represent geographically composite regions that provide ideal settings to study the evolution of a very distinctive biota. The combination of complex geological events and tropical climatic conditions has driven radiation processes that are not homogeneous for all groups in the Caribbean. In the course of my dissertation work, I examined the case of livebearing fishes of the genus Limia (Teleostei: Poeciliidae) that show some deviations from the positive species‐area relationship predicted by the theory of island biogeography. I particularly analyzed the potential role of some ecological drivers in explaining the lopsided distribution of Limia fishes in the Greater Antilles. Chapter one is an extensive review about the West Indian biogeography that highlights divergent diversification patterns observed in terrestrial versus aquatic groups. This review also sheds light on the unbalanced number of studies covering the biogeography of these groups of organisms in the West Indies. Chapter two is an assessment of the variation of tolerance to extreme temperatures (measured as critical thermal minimum (CTmin) and maximum (CTmax)) and also the thermal breadth for multiple species of Limia occurring in the West Indies and that occupy different altitudinal distributions. This study analyzes the role of temperature gradients across elevations in determining species distribution in the genus. Chapter three evaluates the diet and feeding specializations of Limia through a comparative analysis based on data of the gut content of eight species. This chapter explores the potential role of the feeding ecology in the radiation of the genus Limia in the Caribbean. Finally, chapter four uses laboratory experiments and also fieldwork data to test whether sexual selection or natural selection may explain the presence of color polymorphism in Limia vittata, an endemic species to Cuba. Overall, my results show that species richness in the genus Limia is positively correlated with island size but also with the presence of elevations showing that not only island area but also mountainous relief may be an important factor determining the number of freshwater species in the Greater Antilles. In addition, we found no association between thermal tolerance and the altitudinal distribution in Limia species, suggesting that biotic factors such as species interactions, diet specializations, and others should be taken into account when interpreting current distribution patterns. Even though our study showed that the majority of Limia species tend to be feeding generalists, some degree of diet specialization occurs in certain species from Hispaniola where the diversity of the genus Limia is remarkably higher compared to the rest of the Greater Antilles. Lastly, we revealed that natural selection determined by habitat variation in terms of salinity levels, is the most important factor in maintaining color polymorphism in L. vittata. Salinity levels could be indirectly responsible for maintaining different color morphs in this species, likely due to the regulatory effect of saline gradients on predation regimes.