Bathygobius soporator

Tornabene, Luke & Pezold, Frank, 2011, Phylogenetic analysis of Western Atlantic Bathygobius (Teleostei: Gobiidae), Zootaxa 3042, pp. 27-36 : 32

Bathygobius soporator sublineages

Distinct mitochondrial sublineages within B. soporator were observed by Tornabene et al. (2010), as they recognized the presence of two genetic lineages of B. soporator in the western Atlantic. The current study confirms the presence of these sublineages. With the inclusion of West African material in our dataset, there are now three clades of B. soporator . Two of these clades contain individuals from the western Atlantic and the third clade consists of individuals from Guinea. One of the two western Atlantic clades consists of individuals from the Gulf of Mexico and Atlantic coast of Florida (“lineage 3” of Tornabene et al. 2010). The other western Atlantic clade has representatives from throughout the western Atlantic, but not from the Gulf of Mexico (“lineage 2” of Tornabene et al. 2010). The specimens of B. soporator from Guinea in this study are morphologically indistinguishable from specimens of the two western Atlantic clades of B. soporator . The western Atlantic lineages of B. soporator were not described as distinct species by Tornabene et al. (2010) despite their occurrence in sympatry in some locations, due to the possibility of deep coalescence, the lack of diagnostic morphological or pigmentation characters, and the lack of additional independent information from nuclear genes. Although the current study incorporates the nuclear gene RAG1, the level of polymorphism in this gene alone was not high enough to resolve the species that comprise the B. soporator group, much less the three smaller clades within B. soporator itself. Thus, the question of whether or not B. soporator (sensu Tornabene et al. 2010) represents several cryptic species remains unanswered.

Because the sublineages have only been observed in mtDNA thus far, the possibility of deep coalescence cannot be ruled out as an explanation for the observed pattern of divergence. On the other hand, the three clades of B. soporator may indeed represent genetically and evolutionarily distinct, reproductively isolated independent species. If we assume that B. andrei and its Atlantic counterpart (whether B. soporator , B. lacertus , or the common ancestor of the two) became separated roughly 2.8-3.1 mya by the closure of the Isthmus of Panama ( Lessios 2008), then the diversification between the two western Atlantic clades of B. soporator and the West African clade must have occurred significantly later than the closure of the Isthmus of Panama. Models of vicariance and/or dispersal that would explain this pattern of distribution are difficult to hypothesize because of our poor understanding of the effect of Pliocene glaciations on the trans-Atlantic ocean currents that would be responsible for facilitating or preventing gene flow between amphi-Atlantic populations. Although the finer details of surface currents during and since the Pliocene are not known, the overall surface patterns within the Atlantic may have been stabilized following the closure of the Isthmus of Panama ( Maier-Reimer et al. 1990; Haug and Tiedemann 1998), which may have contributed to the formation of semi-permiable barriers to gene flow and subsequent speciation within the Atlantic basin (e.g. Muss et al. 2001).

Several genera of tropical and subtropical fish have similar patterns of recent west-to-east Atlantic dispersal and diversification after the closure of the Isthmus of Panama ( Floeter et al. 2008: fig 10, scenario “e”). Some examples include seahorses of the Hippocampus erectus -group (Casey et al. 2004), blennies of the genus Ophioblennius ( Muss et al. 2001) , and wrasses of the genus Clepticus ( Heiser et al. 2000) . In these three examples the African members are most closely related to members throughout the western Atlantic ( Hippocampus erectus - group), South American members ( Clepticus ), or are separated from Caribbean members by a mid-Atlantic Ridge clade ( Ophioblennius ). A similar pattern is also seen in the goby Gnatholepis thompsoni , which has recently invaded the eastern Atlantic via a central Atlantic “stepping stone” population ( Rocha et al. 2005). Unlike our study however, the aforementioned examples do not exhibit separation between a Gulf of Mexico clade and Caribbean/South American clade. Bathygobius soporator does occur in the Cape Verde Islands in the central Atlantic, but specimens were not available for this study. A population genetic analysis of B. soporator with increased sample sizes from each lineage plus the central Atlantic, as well as a phylogenetic analysis using a more sensitive nuclear gene may further clarify the present biogeographic distribution and increase our understanding of the relationships between the three mitochondrial lineages of B. soporator .