Ecology Color Variation in a Butterfly and the Problem of "Protective Coloration"

Abstract

The distribution area of Oeneis chryxus Dbld. & Hew. (Lepidoptera: Satyridae) extends from the arctic regions of North America, south through the high ranges of the Rocky Mountains, the Sierra Nevada and the Great Basin as far south as New Mexico. The butterfly is not uniformly distributed over this area but is found only in the Arcticalpine life-zone and in the unforested ridges of the upper Hudsonian; hence, in the southern parts of its range, it is restricted to isolated islands on the higher peaks. The species is variable with respect to the predominant color on the upper surfaces of the wings, some individuals being a rather dark yellow-brown, others a medium yellow-brown and in extreme instances, yellow-white. The colors appear to be entirely quantitative in nature; individuals showing all different intermediate shades of color can be arranged from darkest to lightest. Within a restricted population, however, the individuals are remarkably uniform and only very occasionally are atypical examples found. Most populations consist of individuals of a rather dark yellowbrown color; occasional ones consist of extremely light or extremely dark forms. In California and western Nevada, the species is represented by two races, a very pale, yellow-white race (ivallda Mead, fig. 5B) and a very dark, yellow-brown race (stanislaus Hov., fig. 5D). The distribution area of each of these races is shown on the map (fig. 1); the localities where the white race has been found are shown by circles and those of the brown race by triangles. The contour line shown on the map approximately encloses the habitable territory of the species, this being the Arctic-alpine and upper Hudsonian life-zones. The color shade of individuals inhabiting the various localities is shown on the diagram (fig. 2). The horizontal scale indicates the geographical position of the locality while the vertical scale indicates the shade of color, darkest at the top and lightest at the bottom; the line connecting the points gives an index of the steepness of the variation gradient (cline as used by Huxley, '39) between the localities. At Sonora Pass, in the center of the range of the brown race, white or light-colored individuals are taken very rarely. Between Ebett and Echo Passes the populations all lighter in color, but entirely white individuals are as yet unknown. South of Sonora Pass to a point north of Tioga Pass where the brown and white races appear to interbreed, the population of the brown race is nevertheless dark, showing no variation gradient. In the very narrow transition zone, however, the population is mixed, showing the entire range of variation from the darkest to the lightest. These facts would seem to indicate that the differences between the brown and white races are not due to a single gene, but rather to a combination of several genetic factors, and that these factors or genes are widely distributed throughout the entire range of the species though with differential concentrations. As far as one can judge without genetical experimentation, the white races from the northern and southern sections are entirely identical even though the intervening area inhabited by the dark race is at least eighty miles in extent. The shortest means of communication between the two regions is through that occupied by the brown one (fig. 1). Furthermore, as the surrounding territory is of lower elevation and is in a different life-zone, it is ecologically unsuited for the existence of this particular species; it could hardly be a region through which migration could 371