We have now considered in detail the approved personalities of each sex among three primitive peoples. We found the Arapesh-both men and women-displaying a personality that, out of our historically limited preoccupations, we would call maternal in its parental aspects, and feminine in its sexual aspects. We found men, as well as women, trained to be co-operative, unaggressive, responsive to the needs and demands of others. We found no idea that sex was a powerful driving force either for men or for women. In marked contrast to these attitudes, we found among the Mundugumor that both men and women developed as ruthless, aggressive, positively sexed individuals, with the maternal cherishing aspects of personality at a minimum. Both men and women approximated to a personality type that we in our culture would find only in an undisciplined and very violent male. Neither the Arapesh nor the Mundugumor profit by a contrast between the sexes; the Arapesh ideal is the mild, responsive man married to the mild, responsive woman; the Mundugumor ideal is the violent aggressive man married to the violent aggressive woman. In the third tribe, the Tchambuli, we found a genuine reversal of the sex attitudes of our own culture, with the woman the dominant, impersonal, managing partner, the man the less responsible and the emotionally dependent person. These three situations suggest, then, a very definite conclusion. If those temperamental attitudes which we have traditionally regarded as feminine-such as passivity, responsiveness, and a willingness to cherish childrencan so easily be set up as the masculine pattern in one tribe, and in another be outlawed for the majority of women as well as for the majority of men, we no longer have any basis for regarding such aspects of behaviour as sex-Iinked. And this conclusion becomes even stronger when we consider the actual reversal in Tchambuli of the position of dominance of the two sexes, in spite of the existence of formal patrilineal institutions.

The material suggests that we may say that many, if not all, of the personality traits which we have called masculine or feminine are as lightly linked to sex as are the clothing, the manners, and the form of head-dress that a society at a given period assigns to either sex. When we consider the behaviour of the typical Arapesh man or woman as contrasted with the behaviour of the typical Mundugumor man or woman, the evidence is overwhelmingly in favour of the strength of social conditioning. In no other way can we account for the almost complete uniformity with which Arapesh children develop into contented, passive, secure persons, while Mundugumor children develop as characteristically into violent, aggressive, insecure persons. Only to the impact of the whole of the integrated culture upon the growing child can we lay the formation of the contrasting types. There is