Natural selection has always been assumed to be the major force of evolution, but its presence has been difficult to demonstrate. A review of the evidence for selective differences among genotypes for most human genetic polymorphisms indicates there is little of a direct nature. Indirect theoretical evidence, however, seems to support a major role for natural selection, and it does not seem to support the hypothesis that most amino acid substitutions within the human species are neutral. Among small isolates, most of the gene frequency differences are most likely due to genetic drift or the founder effect, and the principal counterbalancing force is gene flow or migration. But genetic differences among the major human subdivisions do not seem to be due to the same interacting forces. One reason for the inability to detect selection has been an oversimplified view of its operation, which assigns genotypes a constant fitness in every generation. Many recent theoretical developments of more complicated kinds of selection may lead to a resolution of the problem and suggest better interpretations of the enormous amount of data on human genetic variation that is rapidly accumulating.