STUDIES ON CHROMOSOMES. 247 



nuclei seem to proceed to further division. This anomalous 

 division, in so far as I have been able to follow it, seems to be 

 confined to the spermatids which have received the four chromo- 

 somes. Photo 69 shows in the upper field a telophase of a 

 secondary spermatocyte which is dividing as a four-chromosomed 

 cell to form two spermatids which will obviously contain four 

 chromosomes each. Below it, however, is such a division of a 

 four-chromosomed spermatid as I have just mentioned. The 

 size relations of the two are obvious. Figs. 193 and 201 show 

 similar conditions. I am inclined to believe that even further 

 divisions of such four-groups occur inasmuch as it is not un- 

 common to find still smaller cells with four small chromosomes 

 in process of division. Photos 65, 66 and Figs. 196 and 197 

 represent such a type. They may become so small as to make it 

 difficult to decide whether one is dealing with 4 small chromo- 

 somes or a quadripartite chromosome. On the other hand, one 

 may from time to time encounter division figures in which, 

 instead of four chromosomes, only two appear at the equator of 

 the spindle (Fig. 198, photo 68) suggesting that the fours have 

 again paired. 



Such unaccountable behavior on the part of the many four- 

 chromosomed cells suggests that these cells have run riot, as it 

 were, and are to be looked upon as degenerating. This condition, 

 together with the fact that there are many cells among the sper- 

 matids which look as if they were not normal (Fig. 204) makes it 

 appear probable that the one class of spermatids does not trans- 

 form into adult spermatozoa. Moreover, careful measurements 

 of spermatozoa from the vas deferens fail to reveal more than 

 one type. 



Fig. 204 represents forms which occur very commonly among 

 the spermatids. Instead of an elongation of the nucleus and a 

 rearrangement of the chromatic bodies into the form charac- 

 teristic of what may be interpreted as the normal course of 

 development (Figs. 200, 203), the chromatin compacts into a 

 single dense round or very irregular mass. Occasionally what 

 looks like an incipient axial filament appears but does not de- 

 velop far. At least, I would so interpret the many instances 

 which occur as represented in Fig. 204. Not infrequently also 



