Phylogenetics and the evolution of terrestriality in mudskippers (Gobiidae: Oxudercinae).

ABSTRACT

The initial vertebrate conquest of land by stegocephalians (Sarcopterygia) allowed access to new resources and exploitation of untapped niches precipitating a major phylogenetic diversification. However, a paucity of fossils has left considerable uncertainties about phylogenetic relationships and the eco-morphological stages in this key transition in Earth history. Among extant actinopterygians, three genera of mudskippers (Gobiidae Oxudercinae), Boleophthalmus, Periophthalmus and Periophthalmodon are the most terrestrialized, with vertebral, appendicular, locomotory, respiratory, and epithelial specializations enabling overland excursions up to 14 h. Unlike early stegocephalians, the ecologies and morphologies of the 45 species of oxudercines are well known, making them viable analogs for the initial vertebrate conquest of land. Nevertheless, they have received little phylogenetic attention. We compiled the largest molecular dataset to date, with 29 oxudercine species, and 5 nuclear and mitochondrial loci. Phylogenetic and comparative analyses revealed strong support for two independent terrestrial transitions, and a complex suit of ecomorphological forms in estuarine environments. Furthermore, neither Oxudercinae nor their presumed sister-group the eel gobies (Amblyopinae, a group of elongated gobies) were monophyletic with respect to each other, requiring a merging of these two subfamilies and revealing an expansion of phenotypic variation within the "mudskipper" clade. We did not find support for the expected linear model of ecomorphological and locomotory transition from fully aquatic, to mudswimming, to pectoral-aided mudswimming, to lobe-finned terrestrial locomotion proposed by earlier morphological studies. This high degree of convergent or parallel transitions to terrestriality, and apparent divergent directions of estuarine adaptation, promises even greater potential for this clade to illuminate the conquest of land. Future work should focus on these less-studied species with "transitional" and other mud-habitat specializations to fully resolve the dynamics of this diversification.