Pre-existing biases for swords in mollies (Poecilia)

Pre-existing female biases are female preferences for a particular trait that evolved prior to the evolution of that trait. Phylogenies are needed to show when the preference and trait have originated. In several live-bearing fishes (Poeciliidae), females show pre-existing biases for male swords, a colorful extension of the caudal fin. Here, we investigated the pre-existing bias hypothesis by predicting preferences for a sword in several molly species, including 2 unusual species in the monophyletic subclade Mollienesia: the Amazon molly, Poecilia formosa, a sperm-dependent hybrid form, and the Tamesi molly, Poecilia latipunctata, a species in the long-fin molly clade, that has a short-fin morphology. Using published sequence data available for this family, behavioral approaches, robust phylogenetic analyses, and Bayesian ancestral state reconstructions, we tested whether the hybrid P. formosa has a preference similar to the maternal ancestor, Poecilia mexicana, or the paternal ancestor, Poecilia latipinna. Surprisingly, the preference shown by P. formosa was variable between populations and matched the preference found in the co-occurring host species. In P. latipunctata, we found a pre-existing bias for sworded males, suggesting that this represents an ancestral trait for the long-fin molly clade. On the basis of the combined evidence from multiple studies, it seems as if pre-existing biases for sworded males are relatively basal to poeciliids and that existing phylogenetic relationships allow us to predict sensory biases.