MORPHOMETRIC ADAPTATION OF THE HOUSEFLY, MUSCA DOMESTICA L., IN THE UNITED STATES

It has long been recognized that morphometric difference among geographic populations of a species can reveal modes of adaptation to environment and of eventual speciation. Most early studies of such variation were limited to few characters (e.g., Alpatov, 1929; Dobzhansky, 1933), and not until the advent of computers could large suites of characters be analyzed by appropriate techniques of covariation, such as principal component or factor analysis, to isolate and identify underlying patterns of morphogenesis (Blackith, 1960). The statistical revolution in geographic variation analysis, however, has not generated a similar revolution of insight into population dynamics, and for the most part, morphometric approaches have been abandoned by population biologists uninterested in systematics. A part of this problem is that few modern studies attempt to separate spatial from temporal aspects of variability or genetic causation from environmental induction of phenotype, typically present in early work (e.g., Goldschmidt, 1934; Stalker and Carson, 1947, 1948, 1949). This paper attempts to partition morphometric variation in the housefly, Musca domestica L., in the United States into temporal, spatial, and genetic components, and thereby elucidate the dynamics of adaptation to differing environments in this species. The lack of systematically recognized variatioi of the housefly in the United States offers reasonable assurance of common niche and habitat requirements of flies from widespread populations, so that sample partitions of variation can be compared to reveal trends of populational adaptation to changing environments. The essence of the methodology utilized here, paralleling an analysis of variance, was to collect flies from numerous localities throughout the United States, return to selected localities at the same times in subsequent years to assess year-to-year variation within sites, and then to follow at least one population throughout the year to determine a seasonal component of morphometric change. Thus variance components can be estimated at these various levels of potential variation, to discern underlying causes of morphometric adaptation. Comparisons among repeated collections within localities and between wild and laboratory reared flies were used to determine repeatability and consistency in patterns of covariation and to ascertain the extent of genetic commitment to morphometric response.