The evolution of the placenta in poeciliid fishes.

How and why complex organs evolve is generally lost to history. The mammalian placenta, for example, was derived from a single common ancestor that lived over 100 million years ago.1-3 Therefore, the selective factors favoring this complex trait remain obscure. Species in the live-bearing fish family Poeciliidae have independently evolved placentas numerous times while retaining closely related non-placental sister species.4-7 This provides the raw material to test alternative hypotheses for the evolution of the placenta. We assemble an extensive species-level dataset on reproductive mode, life histories, and habitat, and then implement phylogenetic comparative methods to test adaptive hypotheses for the evolution of the placenta. We find no consistent family-wide associations between placentation and habitat. However, placental species exhibit significantly reduced reproductive allotment and have a higher likelihood of exhibiting superfetation (the ability to gestate multiple broods at different developmental stages). Both features potentially increase body streamlining and enhance locomotor performance during pregnancy, possibly providing selective advantage in performance-demanding environments such as those with high predation or fast water flow. Furthermore, we found significant interactions between body size and placentation for offspring size and fecundity. Relative to non-placental species, placentation is associated with higher fecundity and smaller offspring size in small-bodied species and lower fecundity and larger offspring size in large-bodied species. This pattern suggests that there may be two phenotypic adaptive peaks, corresponding to two selective optima, associated with placentation: one represented by small-bodied species that have fast life histories, and the second by large-bodied species with slow life histories.