Effects of captivity on behavior and morphology in the oldfield mouse, <italic>Peromyscus polionotus subgriseus</italic>.

Abstract

Long-term maintenance of captive populations, and release of captive animals into the wild, is one of many approaches to endangered species conservation. Reintroduced captive-bred populations, however, often suffer high mortality rates due to behavioral deficiencies in the released individuals. To test captivity's effects on behavior and morphology, I compared four populations of oldfield mouse <italic>Peromyscus polionotus subgriseus</italic> that had been collected from Ocala National Forest, Florida, and held in captivity for varying numbers of generations: 35, 14, 2, and 0 (wild-caught). To assess captivity's effects on behavior, I measured responses to a novel environment and a simulated predator (an owl silhouette). To measure morphological change, I used geometric morphometric techniques to compare cranial and mandibular size and shape. Results from components of these studies indicated that the population held in captivity for 35 generations was behaviorally and morphologically distinct from the other three populations. In the simulated predator experiments, behavioral variance increased with generations in captivity, thus indicating selective pressures on predator response behaviors were relaxed in the captive environment. Morphological data showed that the magnitude of change increased with generations in captivity, but the physical changes between populations were not cumulative or progressive. The observed changes were likely due to founder effects and relaxed selective pressures associated with captivity. Currently, many reintroduction programs do not consider the differences between captive-bred and wild animals when determining the size of a release population. Based on the results of these studies, I calculated a theoretical release ratio, which can be used to determine the number of release animals needed to compensate for increased variance in a captive-bred population. The calculations indicate that 150 captive-bred oldfield mice (from populations bred in captivity for 14 to 35 generations) should be released to obtain a target population of 100 animals. Release ratios, however, will vary among species and perhaps among different populations of the same species. Therefore, a release ratio should be calculated separately for each situation. This research significantly increases our understanding of behavioral and morphological changes in captive-bred animals and how to integrate those changes in reintroduction planning decisions.