4 6S SCIENCE PROGRESS 



develop when the gonads are removed. This is not a fact or 

 phenomenon of heredity, but of physiology ; and no facts or 

 theories of heredity can explain why the development of the 

 antlers of the stag or the comb of the cock should depend on the 

 chemical secretion produced by the testes. This fact was dis- 

 covered by the physiologists and by physiological methods, and 

 could never have been discovered by Mendelian experiments. 

 It is equally erroneous to assert that the Mendelian theory of sex 

 explains the phenomena of sexual dimorphism in ordinary cases, 

 for a similar reason : the theory admits that the male secondary 

 characters are present in both sexes, and their development in 

 the male and absence in the female (according to the Mendelian's 

 own account) is not due to the constitution of the zygote, but 

 to the influence of the gonads in the developed organism. 



It is quite possible, however, that there are sex-limited 

 characters which are not dependent on hormones, but whose 

 development is entirely due to the constitution of the zygotes. 

 In fact, in the present state of knowledge we are forced to the 

 conclusion that there are two kinds of sexual dimorphism, 

 When I wrote my paper (8) on Hormones and Heredity, I was 

 puzzled by the apparent difficulty of reconciling my conclusions 

 concerning the significance of the effects of castration with the 

 entire absence of such effects in certain Lepidoptera. Oudemans 

 Kellog, and Meisenheimer destroyed or removed the gonads in 

 the caterpillars of Bombyx niori and Ocneria dispar, and found 

 that the sexual characters of the moths reared from the larvae 

 were unaltered. There is the same difficulty in the case of 

 congenital defects limited to one sex in man. Prof. Bateson has 

 overlooked the possibility of the distinction which I am here 

 pointing out. He evidently regards all sex-limited characters 

 as belonging to the same category, for he remarks {Menders 

 Principles of Heredity, 1909, p. 174) : "Just as disease or removal 

 of the ovaries may lead to the appearance of male characters, 

 we should expect that such disease might lead to the occasional 

 appearance of colour-blindness in females. I do not, however, 

 know of such a case." Now we do not know anything of the 

 effect of castration in preventing the development of colour- 

 blindness in men ; but we know one fact which indicates that 

 this defect, although usually confined to the male sex, is not 

 a sexual character of the same kind as the beard. Colour- 

 blindness is present in young boys at an early age. It does not, 



