Maternal provisioning of embryos in Gambusia (Poeciliidae) entails both production of large, yolky eggs and mother-to-embryo transfer of nutrients, the latter of which is readily quantified using injection of radiolabeled nutrients. We assayed patterns of nutrient transfer in broods of 26 Gambusia geiseri and 23 Gambusia affinis females, using injection of tritiated leucine. We examined maternal and embryo characteristics affecting the instantaneous rate of transfer and characterized the pattern of transfer to individual embryos within broods. Maternal (female size and condition) and brood characteristics (mean embryo size, developmental stage, brood size) did not predict the mean level of nutrient transfer to embryos in a brood for either species. Within broods, individual provisioning of embryos was not related to developmental stage, but was related to embryo mass in G. affinis with nutrient transfer higher to larger embryos. In addition, overall within-brood variation in nutrient transfer, measured as coefficient of variation in embryo radioactivity, was higher in G. affinis than in G. geiseri.

Theoretical models of the evolution of matrotrophy from a lecithotrophic ancestor suggest that resource availability plays a major role in selective scenarios favoring a change in offspring provisioning. We examined effects of feeding level on embryo provisioning in the livebearing fish Gambusia geiseri, a species with dual provisioning of embryos via both yolk sequestered in large eggs and post-fertilization mother-to-embryo nutrient transfer. Females were fed either once per day or once every three days for three months. Females fed daily had marginally larger brood size, significantly larger embryos, and a higher rate of nutrient transfer (assayed directly by injection of radiolabeled nutrients) than females fed every third day. There was no difference in the size of unfertilized eggs between the feeding treatments. Resource effects on matrotrophic provisioning in G. geiseri suggest that matrotrophy plays an important role in provisioning and allows females to adjust offspring size in response to resource availability.

Many conservation strategies focus on reducing the risk of road mortality. Turtles, which are highly susceptible to road mortality, exhibit life history traits that increase their vulnerability to population declines because of road mortality. Here, we use Akaike's Information Criterion to identify road mortality hot moments and hot spots for 5 turtle species (Chelydra serpentina, Pseudemys concinna, Terrapene carolina, Terrapene ornata, and Trachemys scripta). We tested 2 hypotheses: 1) Hot moments coincide with the breeding season of each species and 2) hot spots are associated with the habitat where each species is most commonly observed. As predicted, breeding season (May-Jun) explained most of the temporal variation in road mortality for all 5 species. Over 98% of all observed road mortality occurred during this 2-month span. For C. serpentina and T. ornata, spatial variation in road mortality was best explained by the distance to the nearest water body, whereas distance to the nearest forest was the best model for T. scripta. The null model ranked best for P. concinna and T. carolina, suggesting none of our spatial models sufficiently explained mortality patterns of these 2 species. These results, coupled with previous studies, suggest that hot moments may be consistently associated with the breeding season. However, hot spots may be species-specific, requiring conservation efforts to be targeted at certain species. ß 2012 The Wildlife Society.