Before attempting to probe the intricacies of the mental side of the subject, we had better see what we can learn of the physical. Let us first remind ourselves of one or two facts gained from animal experimentation. In the first place, in mammals the activation of the sexual instincts of one or the other sex appears to be completely or almost completely under the control of the internal secretions of the reproductive organs. Steinach and others have taken new-born male guinea-pigs and have removed their testes and grafted ovaries in their place. The result has been an animal almost completely feminized both as regards body and mind. In some of the animals milk was secreted, and when this occurred they would act as foster-mothers to new-born guinea-pigs of other parents. The reverse operation, the masculinization of females, was equally successful, the animals growing large and showing all the instincts of a normal male and none of those of a normal female. A similar dependence of behaviour on gonad is seen in fowls. Here nature makes a number of experiments, which have recently been studied by Dr. Crew of Edinburgh. When the ovaries of a hen are affected by a certain type of tumour, the bird stops laying, her comb and wattles enlarge to the size of a cock’s, her spurs grow, she begins to crow, her plumage changes at the moult and becomes cock-like, and finally she becomes indistinguishable from a male. Indistinguishable, even in behaviour: her years of feminine routine in laying and brooding are forgotten: the secretion of the altered ovary now apparently resembles that of a testis and stimulates centres of the brain which would otherwise have remained permanently dormant. She struts and crows, fights and mates, and the memory of the previous part of her life is for all practical purposes lost, since the centres for female activity are no longer stimulated at all. Various workers have even experimentally produced a state of hermaphroditism in mammals by simultaneous grafting of portions of testes and ovary: the behaviour here oscillates between male and female.[34] It is quite clear from these and other facts that in higher vertebrates there are present in every individual of either sex the nervous connections which give the possibility of either male or female behaviour; but that normally only one of these two possibilities is realized, since for the potentiality of action given by the nervous connections to become actual as behaviour it is necessary for the nervous system to be activated by the secretion of one or other of the reproductive organs. Castrated animals fail to realize either possibility of normal sex-behaviour, although their nervous machinery is untouched.