Why are some male pygmy swordtails large?

" Swordtails in the genus Xiphophorus exhibit substantial variation in male body size influenced by genetic variation at the pituitary (P) locus on the Y-chromosome. However, males of Xiphophorus pygmaeus have historically been classified as uniformly small; they were thought to possess one of two P-alleles, both for small size. In 1988, large male X. pygmaeus were discovered at two sites in the Río Huichihuayán, Mexico. Surprisingly, females from these sites have lost the ancestral preference for large males although females from adjacent sites have not. This study investigates the heritability of large body size in male X. pygmaeus to understand its role in the evolution of male mating strategies and female preference in this species. We conducted breeding experiments with X. pygmaeus from two populations to determine whether large size is influenced by the same P-allele system found in other swordtails. Small fathers had a significant effect on the size at maturity of their sons but large fathers did not. All sons were small in size regardless of paternal size. Large fathers were no more successful than small fathers in producing offspring, and we found no significant relationship between age at maturity and size at maturity of the sons. In some cases, there was a significant difference in prematuration growth rate of sons between populations and in postmaturation growth rates within and between populations, yet all sons remained small in size. These results suggest large size is not paternally heritable through a P-allele in this species but could be a result of changes in pre- or postmaturation growth rates. The evolutionary persistence and spread of large male size may not follow current models of sexual selection on P-alleles and the loss of female preference in one population may be explained by the apparent lack of heritability of this male trai "