126 " Gynandromorphlfnii" and Kindred Problems 



ceased the groups of cells formed by subsequent divisions will lead to 

 the mosaic arrangement of sex glands, secondary sexual apparatus, and 

 somatic structures, which I have already referred to and laid such stress 

 upon. Naturally nuclear migration is usually slight and so most insect 

 gynandromorphs approximate to the halved condition. The weak point 

 of this theory is that nuclear migration has not been observed to occur 

 and its possibility is denied by some ; but it explains better than any 

 other how in heterochroic gynandromorphs the areas occupied by the 

 two colours and two sexes are identical. 



The fourth theory which has been advanced to explain the somatic 

 peculiarities of unisexual gynandromorphs is that they are due to 

 trophic changes in post-embryonic (larval and pupal) life. This has 

 been suggested by Wittenhagen and Menzel. 



In the light of our present knowledge the alteration produced must 

 be one in the chromatin of the cells. It does not seem improbable that 

 such changes may be jJroduced, but since all the cells are equally 

 affected the change must be a uniform one. It cannot explain the case 

 of the Hertfordshire coriduii, where the alteration is almost invariably 

 confined to one side of the insect. In any case it only explains some 

 instances of gynandromorphisra, the .symmetrical unisexual ones, which 

 form a very small minority of those described. 



It seems, however, to afford a very good explanation for the sym- 

 metrical changes produced by heat in the female of Rhodocera rhamni, 

 Saturnia caecigena, and Lymantria dispar, and by the presence of 

 Stylops in the female of Andrena lahialis and ^1. chrysosceles. 



The change produced in the male Andrena is merely due to im- 

 perfect development of a character present in every male, and not to the 

 acquisition of a female character. It is not comparable to the change 

 produced in the female Andrena. 



The acquisition of male characters by females of these insects, 

 and the non-acquisition of female characters by the males, are ex- 

 plained by the fact that the females are heterozygous for sex, the 

 males homozygous. 



In conclusion I should like to express my thanks to those who have 

 helped me with information or material, especially Messrs W. Bateson, 

 R. C. Punnett, L. W. Newman, H. B. Williams, F. N. Pierce, C. P. Pickett, 

 R. Tait, Junr., G. T. Porritt, and the Revs. C. R. N. Burrows, George 

 Wheeler and G. H. Raynor. 



