658 H. GRAHAM CANNON 



some specimens of Haematopinus consobrinus from 

 a monkey that died in the Zoolof^ical (lardens. 



As with Pechcukis, the material was found to be extremely 

 refractory with re^^ard to fixation. The best fixation generally 

 was obtained with Flennning and Flemming without acetic. 

 For the various stages of the mitosome Mann's corrosive osmic 

 fixative gave excellent results. Kopsch's method of pro- 

 longed fixation in osmic w-as found to be useless — the unstained 

 sections showing nothing impregnated. For staining — as with 

 Pediculus — the only stain of any considerable use was Heiden- 

 hain's iron haematoxylin. Altmann's methyl green-fuchsin was 

 used for some of the material but did not give results of much 

 value. 



The most important points in the spermatogenesis of Pedi- 

 culus as described by Doncaster and myself are as follows : 

 (1) While the somatic chromosome numl)er is t^\elve the 

 spermatogonia] figures show^ only six. This apparently haploid 

 numl)er of chromosomes in the spermatogonia we ascribed to 

 premature pairing. (2) There is only one spermatocyte division, 

 which is extremely unequal, leading to the separation of 

 a minute ' polar body-like ' cell which degenerates. (3) The 

 centrosome of the spermatid is double and from each half 

 an axial filament grows out so that the developing spermatozoa 

 have two conspicuous axial filaments. 



The account given by Miss Foot differs from our description 

 in the following points : (1) The somatic number of chromo- 

 somes is stated to be ten, and in the few spermatogonial groups 

 found it is also stated to be ten. (2) There is always an unequal 

 bivalent in the first spermatocytes. (3) Although the division 

 of the second spermatocytes was not observed, it is assumed 

 that this division is similar to that in corresponding mitoses 

 in ' other species of Hemiptera '. 



With regard to the chromosome number. Miss Foot states that 

 the chromosomes are so small and so frequently constricted that 

 the estimated number can always be questioned. The chromo- 

 somes certainly are very small, but in equatorial plates of 

 follicle cells and of cells that we called spermatogonia we 



