Wright (1948) pointed out that the conditions most favorable for rapid evolution exist when a species is divided into a large number of local populations, some of which are of small or medium effective size. Such conditions exist in the chorus frog, Pseudacris triseriata, in the mountains near Chambers Lake, Larimer County, Colorado. There are at least 21 breeding sites within a three mile radius among which exchange of individuals, and thus genetic interchange, occurs infrequently, if at all (Spencer, 1964). Eight phenotypes of dorsal coloration occur, and three loci, each with two alleles, are involved (Matthews and Pettus, 1966). Tordoff et al. (1976) demonstrated significant differences among seven of these populations in the frequencies of alleles controlling dorsal coloration. Tordoff and Pettus (1977) further demonstrated that phenotypic frequencies for any one of three of these populations remained nearly constant over a five year period of study, and that the populations had small to medium sizes, ranging from 37 to 580 adults estimated from recapture data using a modified Lincoln index (Bailey, 1952). Matthews (1971) and Hess (1969), working in montane and piedmont populations of chorus frogs respectively, concluded that changes in gene frequencies during the life cycle appear to be cyclic. One allele of a pair increases in frequency during the aquatic phase, and the other allele increases during the terrestrial phase of the life cycle. However, there appears to be little or no change from generation to generation. Tordoff and Pettus (1977) believed natural selection to be an important factor in maintaining the observed gene frequencies in these populations. Nevo (1973) drew similar conclusions for two other species of hvlid frogs, Acris crepitans and A. gryllus, and suggested that visual selective predation was one factor involved in maintaining the color polymorphism. In our montane populations, adults and recently metamorphosed young frogs are active during the day in mid and late summer and thus are susceptible to visual predation by diurnal predators. Gray jays, Perisoreus canadensis, are common in and around the meadows frequented by the frogs and they have been observed eating young frogs in these meadows. Visual selection of prey by various bird species has been demonstrated previously under both laboratory and field conditions. Isely (1938) and Dice (1947), in laboratory situations, demonstrated that various species of birds strongly selected against prey which contrasted with the soil color. Kettlewell (1955, 1956, 1965) observed different species of birds prey upon moths under natural conditions. In the majority of the cases, the birds selected the moth which contrasted with the background coloration. Research on mimicry in Lepidoptera has added greatly to our knowledge of visual selection by various bird species (Ford, 1964; Terhune, 1977). Cain and Sheppard (1954, 1961), Cameron (1969), Parkin (1971), and others have shown that birds selectively feed on land snails (Cepaea nemoralis, C. hortensis, Arianta arbustorum), with intensity of selection on different morphs being highly correlated with the degree of crypticity. The present study was undertaken to test, under laboratory conditions, whether gray jays would prey differentially on polymorphic chorus frogs exhibiting vary-
Read full abstract