Historical demography of four gecko species specializing in boulder cave habitat: Implications in the evolutionary dead end hypothesis and conservation

Specialization in narrow ecological niches may not only help species to survive in competitive or unique environments but also contribute to their extermination over evolutionary time. Although the “evolutionary dead end” hypothesis has long been debated, empirical evidence from species with detailed information on niche specialization and evolutionary history remains rare. Here we use a group of four closely related Cnemaspis gecko species that depend highly on granite boulder caves in the Mekong Delta to investigate the potential impact of ecological specialization on their evolution and population dynamics. Isolated by unsuitable floodplain habitats, these boulder-dwelling geckos are among the most narrowly distributed Squamata in the world. We applied several coalescence-based approaches combined with the RAD-seq technique to estimate their divergence times, gene flow and demographic fluctuations during the speciation and population differentiation processes. Our results reveal long-term population shrinkage in the four geckos and limited gene flow during their divergence. The results suggest that the erosion and fragmentation of the granite boulder hills have greatly impacted population divergence and declines. The habitat specialization of these geckos has led to fine-scaled speciation in these granite rocky hills; in contrast, specialization might also have pushed these species toward the edge of extinction. Our study also emphasizes the conservation urgency of these vulnerable, cave-dependent geckos.