338 Dr. E. A. Cockayne on the relation between 



to assume in some degree the colour and structure charac- 

 teristic of the males, and the males, though to a less 

 extent, can be made to approach the females in structure 

 and colour. 



Merrifleld by means of heat obtained females of Gone- 

 pteryx rhamni having the yellow colour of the males, and 

 Standfuss obtained females of Peris omena (Saturnia) 

 coecigena with feathered antennae, though these are simple 

 in the normal female. Kosminsky by subjecting pupae of 

 Lymantria dispar to cold produced in some males an 

 alteration of the colour and shape of the scales towards 

 those found in the females, and in females he produced a 

 colour nearly as dark as that of the males, a shape in the 

 scales very like that found in the opposite sex, and shorter 

 feathered instead of longer simple antennae. The testes 

 and ovaries were smaller or unaltered in size, and the 

 ova were always infertile. He considered that the mal- 

 development of the sex-glands was not associated with 

 the alteration in colour and structure, because those with 

 the most normal glands were sometimes those with most 

 marked alteration in secondary sexual characters. 



The parasite Stylops in the bees Andrena labialis and 

 chrysosceles may produce in some females an approach to 

 the male in colour and to a smaller extent in structure, 

 and in some males a similar but less marked approach to 

 the female. In other specimens no effect is produced. 

 The ovaries are reduced to about one-quarter their normal 

 size. In other bees no effect on the secondary sexual 

 characters is produced, though the ovaries are equally 

 reduced in size. These observations, published by Geoffrey 

 Smith and Hamm, like the temperature experiments, fail 

 to show a direct relationship between the gonads and the 

 secondary sexual characters. 



The conditions found in gynandromorphs afford us still 

 more definite evidence. In halved gynandromorphs there 

 is perfect development of the secondary sexual characters 

 of the male on one side and of the female on the other; 

 yet the most varied arrangement of gonads and accessory 

 glands are met with. In the normal male there are two 

 testes and glandulae accessoriae ; in the female two ovaries, 

 two cement-glands and the receptaculum seminis or sper- 

 matheca, probably also a secretory gland. In gynandro- 

 morphs secondary sexual characters of the male may be 

 perfectly developed not only in the absence of the testes 



