Reproductive mode is predicted to influence the form of sexual selection. The viviparity-driven conflict hypothesis posits that a shift from lecithotrophic (yolk-nourished) to matrotrophic (mother-nourished or placental) viviparity drives a shift from precopulatory towards post-copulatory sexual selection. In lecithotrophic species, we predict that precopulatory sexual selection will manifest as males exhibiting a broad distribution of sizes, and small and large males exhibiting contrasting phenotypes (morphology and coloration); conversely, in matrotrophic species, an emphasis on post-copulatory sexual selection will preclude these patterns. We test these predictions by gathering data on male size, morphology and coloration for five sympatric Costa Rican poeciliid species that differ in reproductive mode (i.e. lecithotrophy vs. matrotrophy). We find tentative support for these predictions of the viviparity-driven conflict hypothesis, with some interesting caveats and subtleties. In particular, we find that the three lecithotrophic species tend to show a broader distribution of male sizes than matrotrophic species. Furthermore, large males of such species tend to exhibit proportionately large dorsal and caudal fins and short gonopodia relative to small males, while these patterns are expressed to a lesser extent in the two matrotrophic species. Finally, large males in some of the lecithotrophic species exhibit darker fins relative to small males, a pattern not evident in either matrotrophic species. One unexpected finding was that even in the matrotrophic species Poeciliopsis retropinna and Poeciliopsis paucimaculata, which lack courtship and dichromatic coloration, some morphological traits exhibit significant allometric relationships, suggesting that even in these species precopulatory sexual selection may be present and shaping size-specific male phenotypes in subtle ways.