Natural Selection in Water Snakes (Natrix sipedon L.) on Islands in Lake Erie

Abstract

Although in recent years laboratory selection has become a commonplace, natural selection operating in wild populations has been quite difficult to document. Much of the published work has been inconclusive (see summary in Robson and Richards, 1936, for early work) or open to controversy (e.g., Lamotte, Cain and Sheppard, and Sedlmair on Cepaeca). The present study analyzes all available data on color pattern variation in the water snakes of the Lake Erie islands. These data appear to illustrate a situation in which migration and strong selection pressure combine to give a relatively clear-cut picture of differential elimination of color pattern types from a population. The uniform medium-gray color of the majority of adult water snakes inhabiting the islands in the western part of Lake Erie (as opposed to the normal dark banded type) was first noted by Morse (1904). In 1937 Conant and Clay described the island population as a separate subspecies Natrix sipedon insularum, differing from typical sipedon primarily in the large percentage of unbanded or weakly banded individuals. In 1954 Camin, Triplehorn and Walter compared the frequencies of various pattern types in wild-caught juveniles with those of the adult population and found a statistically significant decrease in the proportion of banded individuals from the juvenile to the adult population. In the present paper these previous data are reanalyzed and integrated with additional data, principally on frequencies of pattern types in