The Darwinian explanation for evolution is that it is the outcome of the interaction between genetic variation and natural selection. There is now good evidence for both the existence of genetic variation and the occurrence of natural selection, the latter potentially at high intensities. The outcome should be rapid evolutionary change; yet in practice very little change is found. Most species are very stable, and in situations where evolution is observed in one species often none is found in others despite equivalent opportunity. Evolutionary failure is commonplace. Despite the occurrence of high levels of protein polymorphism, there is good evidence that the supply of variation making a major contribution to fitness is very limited. As a result it is argued that lack of evolution in most species may be due more to lack of appropriate variability than to other causes: a condition for which the term `genostasis' is proposed. In those situations where appropriate genetic variation is available for one reason or another, evolution is found to be very rapid. There are good theoretical and practical reasons for more attention being paid to the mechanisms of supply of new variation and to those situations where evolution appears not to be taking place.