Development of the sexually dimorphic anal fin appendicular support of an internal fertilizing bony fish Gambusia affinis affinis was investigated by staining whole-mounted embryos, immature, and adult female and male G. a. affinis with alizarin red S and alcian blue. The tissue was examined histologically to assess development of the amphicelous centrum and to verify specificity of the stains. Our data confirm earlier claims about the development of the male and female characteristics in this species, and we provide for the first time direct embryonic evidence suggesting that development of the sexually dimorphic anal fin appendicular support is biphasic: (1) anteriorization of the most anterior caudal segments, and (2) growth and elongation of hemal arches of vertebrae 14-16. The first process involves a sequential homeotic transformation of hemal arches of vertebrae 11-13 through resorption of mineralized connective tissue, thus forming parapophyses that bear pleural ribs. This process begins in undifferentiated embryos and proceeds similarly in postnatal males and females. During the same period, the second process, likely induced by male gonadal hormones, causes the addition of mineralized connective tissue at the hemal arches of vertebrae 14-16. This second process, which occurs only in males, elongates the hemal arches of vertebrae 14-16 anteriorly. This elongation apparently translocates the anal fin appendicular support (including parts of the hemal spine of the hemal arch of vertebra 13) to the level of vertebra 11. It appears that the developmental programs of both female and male G. a. affinis create an area of 6 vertebrae which are markedly different from any vertebrae anterior to 11 and posterior to 16. We propose to term the area including these vertebrae and the associated anal fin, the genital area. We also propose that the first process, homeotic transformation of caudal into precaudal segments, is regulated by differential expression of control genes, such as homeobox genes, whereas the second process is regulated by gene expression under the control of male gonadal hormones. Conflicting data in the literature can be resolved with this model. Appropriate tests of the model are proposed.