Body size of diverse ectotherms is inversely related to developmental temperature in the laboratory. We monitored seasonal variation in wing length of two populations (Oregon, Washington) of D. subobscura, which was introduced in the Paci®c Northwest in the late 1970s. Wing length varied seasonally and was shortest in summer. In Washington, however, wing length was longest in spring, not winter. Wing length was inversely and curvilinearly related to mean ambient temperature, as in a few previous studies of drosophilids. Mid-winter D. subobscura might not be the largest either because extremely low temperatures depress size or because¯ies collected in winter were in fact born the previous autumn, when developmental temperatures were more moderate. 7