The evolution and the adaptive logic (if any) of female mate choice are sub
jects of lively debate [1]. Whereas most researchers believe that females h
ave evolved to recognize signs of male 'quality' (the ability to provide fe
males or their offspring with direct or indirect genetic or material benefi
ts), there is intriguing evidence that males can evolve to appeal to pre-ex
isting female preferences [2-6]. Evidence for these pre-existing biases is
often ambiguous because phylogenetic reconstructions have usually failed to
establish conclusively whether the female preference or the favored male t
raits evolved first [7-11]. This potential difficulty is minimal in the mos
quitofish genus Gambusia, none of whose 45 species appears to have a female
-choice mating system in the wild, and none of which shows the male behavio
ral and morphological traits that are characteristic of female choice [12-1
9]. Nevertheless, in an experimental situation in the laboratory, female Ga
mbusia holbrooki readily chose between models of males and demonstrated sig
nificant and reliable preferences for a variety of exaggerated male traits
that are not seen in their species or their genus. Other morphological alte
rations were not preferred. The latent willingness of females to choose tra
its in a genus without such traits and without evident female choice in the
wild is remarkable and may indicate a pre existing bias in females that is
ready to drive male evolution, should the social system or the ecological
variables that control it change.