SEX AND SEXUAL CHARACTERS 459 



ment and functional activity of the mammary glands in mammals 

 took place after their connections with the nervous system had 

 been severed, and Prof. Starling and Miss Lane Claypon (9) 

 showed that some development of the glands was caused in 

 virgin rabbits by the injection of extract of foetuses from other 

 rabbits ; some special chemical compound must therefore be 

 present in the fcetal tissue which, when absorbed into the blood, 

 acts as a stimulus to the development of the gland. Shattock 

 and Seligmann (3) found in their experiments on castration 

 that some cocks on which the operation had been performed 

 developed the male secondary characters — especially the comb, 

 wattles and hackle feathers— to the same degree as normal 

 specimens, and on dissection found that in these cases fragments 

 of the testes, detached in the operation, had been left in the 

 abdominal cavity, and had grafted themselves to the peritoneum 

 in some new position, even sometimes on the outer surface of 

 the intestine. In these grafts spermatogenesis took place, and 

 ripe spermatozoa were present. Here again the original 

 nervous connections were all severed, and the only reasonable 

 conclusion is that some specific chemical substance is produced 

 in the process of spermatogenesis which acts as a'stimulus for the 

 development of the somatic structures characteristic of the male 

 individual. This remarkable peculiarity of secondary or somatic 

 sexual characters is entirely beyond the scope of Mendelian 

 conceptions and Mendelian experiments. Mendelism regards 

 all characters as represented by certain units in the gametes. 

 These units may be segregated in the formation of the gametes, 

 and may be combined in the union of the gametes in the process 

 of fertilisation. Some units may have no effect in the develop- 

 ment of the organism, remaining latent or recessive, because 

 others are dominant ; but Mendelism deals only with the 

 characters determined at fertilisation — in fact, with heredity- 

 while the development of secondary sex-characters (although 

 of course partly due to heredity) is not determined at fertilisa- 

 tion, but at a later stage of life by a chemical stimulus. 



From the terms of Mr. Doncaster's statement on the subject 

 it might be supposed that the development of male characters 

 was a regular or frequent consequence of the removal of the 

 ovaries in the female, and that the appearance of the female 

 characters would be expected after castration in the male. 

 Bateson is more guarded in his words — referring only to in- 



