Sokal, R. R. (Division of Biological Sciences, State University of New York at Stony Brook, Stony Brook, New York 11790) 1974. The Species Problem Reconsidered. Syst. Zool. 22:360–374.—Four common, current species concepts are described and their strengths and weaknesses discussed. It is proposed that a review of population biology at the species level will lead to inferences regarding speciational mechanisms, which in turn may lead to a new synthesis of the species concept. Major advances in the study of variation in populations have come through multivariate methodology and biochemical techniques. The relation between genetic, phenetic, and allozymic variation is still unclear. Attention is drawn to the spatial structure of local populations which must play an important role in breeding patterns. Comparisons of patterns in phenetic hyperspace should yield clues to similarities of speciational patterns in sexually reproducing and apomictic populations. There is probably less reproductive connectedness between local population samples of putative species than is commonly assumed. Large-scale population structure will affect geographic variation patterns and reproductive connectedness among populations and needs to be analyzed by various techniques, among them graph theoretical methods. Any theory of species must take into account well-established facts at several levels of biological organization. The great diversity of kinds of species and patterns of speciation is stressed. In In addition to the classical theories of speciation, it is suggested that much local variation is due to the molding forces of local selection, little influenced by gene flow. Attention is directed to evidence from patterns of geographic variation of a single species examined over a series of localities, and from phenetic patterns of sympatric species.