Biologists often trap animals to obtain information on them. If trapping is selective toward some animals, the information may be inaccurate. Most mammalogists know or suspect that their trapping techniques (reviewed by Hayne, 1949; and Stickel, 1954) contain sources of bias. Since trapping remains the only feasible way to obtain information on many animals, researchers have tried to discover sources of sampling error and refine their techniques. They have found that one major source of difficulty may lie in the behavior of animals. Individual animals seem to respond differently to trapping, both initially and through learning (Geis, 1955; Crowcroft and Jeffers, 1961; and others). However, researchers seldom observe the behavioral responses of animals to traps. They infer information from capture data. Perhaps an empirical approach would shed more light on the relationship between behavior and trapping. The present study is such an approach. The study concerns the behavioral responses of adult Uinta ground squirrels, Citellus armatus, to trapping. I based the study on the direct observations of known individuals in a wild population. My primary objective was to learn how animals respond to a trap, to capture, and to recapture. My approach was both that of a population ecologist interested in factors affecting trapping success and that of a behaviorist interested in the effect of trapping procedures on the behavior of animals. I conducted a broad ecological and behavioral study of the population (Balph and Stokes, 1963) before beginning the research on trap response, which helped me select parameters and develop procedures. I also conducted a pilot study on deer mice, Peromyscus maniculatus, in the laboratory to test some procedures and the design of the trap-response investigation.

Biologists often trap animals to obtain information on them. If trapping is selective toward some animals, the information may be inaccurate. Most mammalogists know or suspect that their trapping techniques (reviewed by Hayne, 1949; and Stickel, 1954) contain sources of bias. Since trapping remains the only feasible way to obtain information on many animals, researchers have tried to discover sources of sampling error and refine their techniques. They have found that one major source of difficulty may lie in the behavior of animals. Individual animals seem to respond differently to trapping, both initially and through learning (Geis, 1955; Crowcroft and Jeffers, 1961; and others). However, researchers seldom observe the behavioral responses of animals to traps. They infer information from capture data. Perhaps an empirical approach would shed more light on the relationship between behavior and trapping. The present study is such an approach. The study concerns the behavioral responses of adult Uinta ground squirrels, Citellus armatus, to trapping. I based the study on the direct observations of known individuals in a wild population. My primary objective was to learn how animals respond to a trap, to capture, and to recapture. My approach was both that of a population ecologist interested in factors affecting trapping success and that of a behaviorist interested in the effect of trapping procedures on the behavior of animals. I conducted a broad ecological and behavioral study of the population (Balph and Stokes, 1963) before beginning the research on trap response, which helped me select parameters and develop procedures. I also conducted a pilot study on deer mice, Peromyscus maniculatus, in the laboratory to test some procedures and the design of the trap-response investigation.