GEOGRAPHIC RANGE SIZE AS A PREDICTOR OF DISPERSAL-DEPENDENT BEHAVIORAL TRAITS IN TWO CLADES OF A TERRESTRIAL SALAMANDER

Animal movement has the potential to affect diverse processes within ecology and evolution including range expansion, gene flow, adaptation, and speciation. Two aspects of animal personality that are germane to dispersal are exploratory and aggressive behavior. These behavioral categories may represent a trade-off such that energy invested in territorial defense leaves little energy for movement and dispersal. The Eastern Red-backed Salamander (Plethodon cinereus) is a wide ranging, dispersal limited, terrestrial salamander with well documented phylogeographic divisions. I examined dispersal-relevant behavioral traits within two clades of P. cinereus with disparate geographic ranges. The Northern Clade (NC) has a range extending from the mid-Atlantic coast upward into Canada, across the northern reaches of the Great Lakes and south into Indiana. The Ohio Clade (OC) is largely restricted to Ohio. Previous research suggests that differences in aggression and territorial behavior exists between the clades. The OC is known to be aggressive and territorial while some evidence suggests that the NC lacks territorial behavior. I therefore predicted that members of the NC would be more exploratory and less aggressive and that members of the OC would be less exploratory and more aggressive. Salamanders from each clade were tested in two separate laboratory experiments to assess 1) aggressive behavior and 2) exploratory behavior. In experiment 1, resident salamanders of both clades spent significantly more time in the behavior Look Toward compared to control trials when no intruder was present. This response suggests that salamanders recognized intruders as potential threats. However, residency status (resident or intruder) did not influence the aggressive or submissive behavior for either clade. Additionally, in contrast to my prediction, no differences in submissive or aggressive behaviors by residents were detected between clades when intruders were present. Intruders from each clade also behaved similarly when interacting with residents. 2 In experiment 2, clade members differed in their exploratory behavior, with members of the OC being more exploratory, counter to my predictions. Specifically, members of this clade were quicker to leave their territory, and they crossed over barriers more often. I compared aggressive and exploratory behavior from experiments 1 and 2 in a correlational analysis to investigate trade-offs between these behaviors. Only in the NC was aggressive behavior “All Trunk Raised” correlated with latency to the exploratory chamber’s final zone, suggesting that, in this clade, aggressive individuals were more cautious during exploration. My results suggest that there are clear exploratory differences between the NC and OC but that these differences are not easily explained by geographic range size and historical range expansions