The social behaviour which is the material of psychological observation at any given time is the product of a certain genetic constitution on the one hand, and of a complex process of social, educational, familial and, to some extent, uterine environment on the other. The social behaviour of human beings cannot be subjected to direct methods of genetic analysis. The genetic psychologist may be compared to the astronomer, in so far as he has to rely on Nature to perform the experiments which he himself is unable to undertake. To the category of such experiments belongs the phenomenon of twin production. In the production of monozygotic twins Nature provides us with individuals of the same genetic constitution. Such differences as they exhibit arise uniquely from the operation of differences in the environment in which they are placed. In ordinary circumstances they share many more features of environment than any other two individuals selected at random from the population. When reared together, the extent of their dissimilarity can only throw light on such differences of nurture as may operate on members of the same family unit and of the same age. If they are reared apart, the measure of resemblance exhibited by monozygotic twins may throw light on the influence of social environment in a less restricted sense. With a sufficiently large sample of twins reared apart, a satisfactory standard of comparison for assessing the extent of resemblance due to genetic constitution, plus the effects of a common uterine environment, would be provided by the degree of resemblance existing between their foster sibs in families into which they have been adopted. Such data cannot be collected easily. As yet they do not exist in sufficient quantity to yield satisfactory conclusions.