The new discipline of the genetics of behaviour, to judge by some recent books, is caught in the dogmas of Mendelian genetics without regard to developments in modern genetics during the last ten years, and to modern experimental approaches to the genetic roots of behaviour. Books on the subject usually begin with an account of the principles of Mendelian genetics. The material on behaviour deals mainly with mutated animals and their observed changes in behaviour. That is exactly what genetic principles predict. If an important mutation should not be followed by a change in behaviour-then geneticists would have to worry about the validity of the principles.