E. A. Cockayne 125 



every respect except that of cdloiir-lilindniss. Thr proof that they were 

 identical is incontrovertible. 



The gynandromorphous hybrids Injbridus and operosa can be ex- 

 phiined ecjually well on either Goldschmidt's hypothesis or by the one 

 just put forward. But if Goldschmidt's is the more likely for the 

 ordinary fine mosaic hybrids, the other must be invoked to explain 

 the very rare completely halved gynandromorphous hybrids ^1. populi 

 X ocellatus, and for the other instances of halvcMl hybrid gynandro- 

 morphs. In this connexion it is noteworthy that no fine mosaic 

 gynandromorphs occur in the heterochroic group or those containing 

 a sex-limited colour. This does suggest that fine mosaic gynandro- 

 morphs are really ditferent and arise by a mutation, produced according 

 to Goldschmidt by inbreeding, or like other mutations arising in a 

 manner not yet understood. After the sudden change in the sex factors 

 has arisen in this way, it may be propagated on Mendeliari lines or die 

 out, owing to coincident loss of sexual instinct. 



Familial and hereditary cases can be explained as due to the 

 inheritance of some abnormality of the chromosomes. Doncaster has 

 found that the females (jf a race of Abraxas grossulariata, which 

 produces nearly all females, have a chromosome fewer than the normal 

 females and even than the females of the same race, which produce 

 normal families. The occurrence of a race of Agriades coridon, which 

 produces a similar excess of females and a number of gynandromorphs 

 all of a special predominantly female type, suggests that there is some 

 chromosome defect in this isolated colony. 



There is some evidence to show that a race of gj'nandniiiiorphs in 

 which the condition is familial or hereditary tends to produce gynandro- 

 morphs of the same type, as can be seen by referring to the details 

 already given. 



The imperfect nature <jf the halving and the remarkable mosaic 

 arrangement, which so markedly differentiates many insect gynandro- 

 morphs from the gynandromorphous birds and some gynandrouKjrphous 

 Crustacea, is explained by Wheeler. In birds each of the first two 

 cleavage cells gives rise to one half of the body and .so we get a perfectly 

 halved bird, with .sex organs and somatic characters entirely male on 

 one side and female on the other. In insects according to this author 

 there is a syncytial ariangement of the ovum in the cleavage and 

 preblastodermic stages, which allows of a considerable degree of nuclear 

 migration. Hence we get nuclei with different chromatin content 

 arranged in almost any imaginable way, and when migration has finally 



