196 



Dr. L. Doncaster. Gametogenesis and 



Summary. 



Iii Neuroterus lentimlaris there are two generations in the year, agamic 

 females appearing in early spring, and sexual females and males in early 

 summer. Previous work had shown that any individual agamic female has 

 onlv male or only female offspring, and the object of the present work was 

 to discover the nature of the difference between these two classes of agamic 

 females. 



Experiment showed that any individual sexual female has grandchildren 

 exclusively or almost exclusively of one sex. The galls produced by sexual 

 females were sleeved, each sleeve containing galls derived from one female 

 parent, and in sleeves of galls derived from six females 4235 males and 

 83 females were bred, while in sleeves of galls from another six females there 

 were 5139 females and 117 males. About half tlie sleeves contained no 

 exceptions, and reasons are given for believing that the exceptions were due 

 to eggs of wild flies laid through the meshes of the sleeves. 



Two possible cytological causes might account for the fact that some 

 sexual females produce only male-producing offspring, whilst others produce 

 only fern ale -producing. If each fly pairs only once, the difference might 

 depend on the existence of two kinds of males, or it might arise through 

 differences in the maturation-processes of eggs laid by the two classes of 

 sexual female. No cytological differences in the spermatogenesis of different 

 males could be detected. The maturation phenomena of the eggs (about 300) 

 of 15 separate females have been examined, and while they seem to fall into 

 two rather distinct types, the differences are not sufficiently considerable to 

 correlate them with the sex-phenomena with any confidence. 



The maturation-processes of the eggs are remarkable, and if, as seems 

 probable, the peculiar figures are not due to methods of preservation, they 

 differ widely from the ordinary type of mitosis. The first division takes 

 place by the drawing out of threads (probably double) on each side of the 

 nucleus ; the reticulum becomes absorbed in these threads, which form two 

 groups of parallel chromosomes on a spindle. These chromosomes then 

 divide, probably longitudinally, giving rise to the group which forms the 

 egg-nucleus and three groups of polar chromosomes. 



[PostscrijJt, February 19, 1916. — While the preceding paper was in the press 

 I received E. W. Hegner's paper " Protoplasmic Diffeientiation in the Oocytes 

 of Certain Hyinenoptera."* He describes phenomena in the parasitic 

 Hymenopteran Copidosovia, and less fully in the Cynipid Andricus (the latter 



* ' Journ. of Morphology,' vol. 26, p. 495 (1915). 



