STUDIES ON INSECT SPERMATOGENESIS. 401 



Respecting the general proportion between cytoplasm and nucleus 

 in the large- and smaller-celled generations, Montgomery was of the 

 opinion that the increase in size was due primarily to added cyto- 

 plasm and in a lesser degree to the volume of the nucleus. My own 

 observations, especialh^ on Arvelius, do not bear out these conclusions 

 entirely, and while the nucleo-plasmic ratio seems to be slightly re- 

 duced in the large spermatocytes, I do not think that the reduction is a 

 particularly striking one. (Compare especially Text-figs. 2-111^ 

 and C and Figures 1 and 2.) More exact measurements of the nucleo- 

 plasmic ratios would be of interest, and they could readily be obtained 

 by computing the proper areas on many camera lucida drawings by 

 means of a planimeter. 



As to the contents of the nucleus, I agree with Montgomery that the 

 karyolymph, linin and plasmosome are increased in A'olume in the 

 large cells, but I can not corroborate his statement that, " the chroma- 

 tin nucleolus" (X and Y chromosomes) "is ... larger in cells of the 

 large than in those of the small generation." This statement seems 

 to me doubtful. The plasmosome or true nucleolus is often stained 

 quite differently in the large and small cells, the result apparently of 

 differences in the rapidity of extraction (Figs. 1 and 2). With respect 

 to the most important constituent of the nucleus, the chromatin, 

 Montgomery pointed out that its amount was the same in all nuclei 

 regardless of size. This is indicated in part by the diiTerence in 

 appearance of the large and small nuclei during the height of the growth 

 period, the former being relatively poorer in chromatin as indicated by 

 their 'clearness' (compare Figs. 1 and 2). As several workers have 

 shown, the diplotene threads resulting from synapsis, become spread 

 out in a " confused" way during the growth period, the separate chro- 

 mosomes being thus lost from view for a time. In the large cells this 

 confused state reaches a very advanced stage, while in the smallest 

 nuclei it scarcely gets a start. Indeed, it is possible that in such nuclei 

 (in Arvelius, for example) the threads could be followed through with- 

 out a break to the prophases of the first maturation division. The 

 unusual diffusion in the confused period in the large cells has brought 

 to light a very interesting phenomenon in connection with the be- 

 havior of the sex chromosome nucleoli (thus far I have studied this in 

 Euschistus only) during the earlier part of this stage. These bodies, 

 which are usually stated to retain their definite, compact form (and 

 were so described by Montgomery '11), become broken up into groups 

 (of more or less definite individuality) of granular masses. These 

 may be arranged in a chain formation, or simply massed in an in- 



