A natural population of painted turtles [Chrysemys picta (Schneider)] was studied in University Bay of Lake Mendota, Dane County, Wisconsin, from September 1960 to October 1962. Five methods were used to capture turtles for marking, measurement, and release. Each of the methods yielded a different size-class distribution and a sex ratio which, in four out of five cases, was significantly different from that of the total population estimate. Baited hoop nets yielded predominantly males and relatively few juveniles; basking traps tended to yield higher percentages of females than did nets, and large turtles were an important component of the sample. Hand capture resulted in a sample heavily weighted by juveniles and a 1: 1 sex ratio for adults. The influence of trapping method in the estimation of population structure in turtles must not be overlooked. This study indicates that even within a species, trapping method introduces considerable bias in the estimation of sex ratio and size-class distribution. INTRODUCTION The structure of turtle populations, as estiimated by different authors (Cagle, 1942, 1944, 1954; Pearse, 1923; Sexton, 1959), has varied considerably even within the same species. Although this variation may be due to inherent differences in the populations studied, or to various ecological factors affecting each population with different intensities, it is more probable that the disparity lies in the sampling method used by the investigators. Cagle and Chaney (1950) suggested that sampling methods are biased by the behavioral patterns of the size classes and sexes that make up the population, but we are unaware of any study that has compared the variation in estimation of population and its strutcture brought about by use of different sampling methods. A natural population of Chrysemys picta (Schneider) was studied in University Bay of Lake Mendota, Dane County, Wisconsin, from September 1960 to October 1962. The central painted turtle, Chrysemys picta marginata (Agassiz) and the western painted turtle, Chrysemys picta belli (Gray) intergrade in this region (Pope, 1949). Five methods were used to capture a maximum number of turtles for marking, measurement, and release. Variant results obtained from use of these methods were sufficiently striking to, warrant a comparison of the methods. Acknowledgments.-This study was done with the aid of two NSF Summer Fellowships for Teaching Assistants. Additional financial assistance for field equipment and limited field assistance was provided by a grant to Dr. W. G. Reeder from the Wisconsin Alumni Research Foundation through the University of Wisconsin Graduate School. Appreciation for encouragement and constructive criticism is given to Dr. W. G. Reeder, Dr. J. M. Legler, and Dr.