56 INTERNAL SECRETIONS 



characteristic of the sex, although they develop in an organism 

 deprived of the sexual gland. Also the antennae and the shape 

 of the abdomen are normal in the "castrate." It is the same 

 with the sexual instinct and the instinct for progeny rearing. 

 Copulation is performed by the male in a normal manner, and 

 both normal and castrated males copulate also with castrated 

 females. The castrated female of Lymantria dispar prepares 

 for the reception of the eggs in the usual manner after copula- 

 tion, although no eggs can be laid. 



From these results the authors unanimously concluded, as 

 already mentioned, that the sexual characters in butterflies 

 do not depend upon the sexual glands, and from these results 

 they drew conclusions also upon the Vertebrata. We have 

 seen that some sexual characters in man and mammals re- 

 main more or less unaltered after castration, and this fact 

 suggests the possibility that the sexual characters may be 

 to some extent independent of the sexual glands, as was 

 suggested more particularly by Halban. It is possible, how- 

 ever, to raise objections to the conclusions reached about 

 insects, and these we may now consider. 



Firstly, one might object that the castration was done too 

 late, at a time when the sexual characters were already fixed 

 by the influence of the sexual glands; further, such a fixation 

 might be a latent one. We know from the experiments with 

 man and mammals that the sexual characters show more 

 independence of the sexual glands both in regard to their 

 formation and their preservation, the later the castration is 

 performed. Herhst (1901, p. 80) has raised this objection 

 against the experiments of Oudemans. Meisenheimer and 

 Kopec, however, castrated their caterpillars at a very early 

 stage of development. Furthermore, the objection is weakened 

 also by the experiments of Meisenheimer mentioned above, 

 in which together with the sexual glands the wing rudiments 

 were removed. The regenerating wdngs were not subjected 

 at all to the influence of the sexual glands during their develop- 

 ment; nevertheless they showed the characteristic features of 

 normal male and female wings. 



A second objection has been discussed by Kammerer. He 

 argues that the castration was possibly not a complete one. 

 We shall see later on how important such an objection might 

 be when considering the question of the formative action of the 



