Doctoral Dissertations

Date of Award


Degree Type


Degree Name

Doctor of Philosophy



Major Professor

Neil B. Greenberg


Field work was conducted in 1983-1986 to gather data on social behavior of the Florida red-bellied turtle, Pseudemys nelsoni, by direct, underwater observation (Rock Springs run, Apopka, Florida) and home range analysis using mark-recapture and telemetry techniques (Rock Springs run and Payne's Prairie, Gainesville, Florida). Laboratory observations and experiments were conducted at the University of Tennessee in 1982-1986. A behavior catalog of P. nelsoni was constructed. While not comprehensive, it indicates that the social behavior of this species is complex and adapted for conditions of poor visibility. Conspecific identification probably occurs prior to courtship and other social interactions. Aggressive behavior, mostly by melanistic males, was directed toward conspecifics of both sexes. Aggressive behavior directed towards other males consisted largely of threats while females were bit and tugged on. Basking phases were not clearly discernable in P. nelsoni although some postures and behaviors were associated with the length of time an animal had spent basking. An investigation of how juvenile turtles assembled behavior revealed that alternating pairs of behaviors were the most common pattern. Variability in the patterning of behavior among observation periods and individual variation likely obscures much of the organization of their behavior. The most valuable techniques for investigating behavior patterns in this study were 1) identifying "units" of behavior, 2) first order transition matrices, 3) information theory, and 4) auto- and cross-associations. Precocial courtship was a frequent social event of juveniles in the laboratory. While this behavior's function is not obvious, many of its attributes coincide with those given for play behavior suggesting that the function of juvenile courtship is similar to one or more of the possible functions of play. The results of two laboratory experiments, one using positive reinforcement and the second punishment, suggest that these animals can vii distinguish between a pair of turtles matched for size and pattern. To describe home range size and usage, the data were recast as a constrained nonlinear optimization problem and solved using a FORTRAN program. Home ranges were small at both the Apopka and Gainesville populations despite substantial habitat differences and the different techniques used to locate turtles. The entire home range could be traversed by an individual in one or two hours.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."