172 PROCEEDINGS OF THE ACADEMY OF [Feb-. 



maturation mitoses could not be definitely settled, since it could not be 

 satisfactorily distinguished from other chromosomes of about the same 

 size. But there is some probability, as was pointed out, that it may be 

 the chromosome marked y in Plate IX, figs. 32 and 33; and in very 

 early stages of the spermatocytes (figs. 15, 16) it showed a longitudinal 

 splitting which soon after seemed to disappear. For these reasons of its 

 proved bivalence and its longitudinal splitting, in conjunction with 

 the fact that each chromosome divides in each maturation mitosis, it 

 becomes most probable that it undergoes a reduction division in the 

 first mitosis, and an equation division in the second. For since it is 

 formed and has essentially the same constitution as the other chromo- 

 somes, there would be all reason to expect it to divide like them ; and 

 a more trenchant reason is this, that of the ten chromosomes of a second 

 spermatocyte there is no particular one which from any peculiarity of 

 structure could be regarded as bivalent. This is, of course, only cir- 

 cumstantial evidence of its undergoing first a reductional and then an 

 equation division, but the probability of this contention is obvious; 

 there is no doubt that it undergoes two divisions. 



In the monaster stage of the spermatogonia (Plate IX, figs. 7, 8) cer- 

 tain chromosome pairs could be rccogm'zed bj'' peculiarities in form and 

 size, namely, those lettered in these figures. It is corroborative evidence 

 of the persistence of the individuality of the chromosomes, if, indeed, any 

 further proof of this idea is needed to-day, that the same differences 

 are observable in later stages. So among the ten univalent chromo- 

 somes of a second spermatocyte (Plate X, figs. 37, 38) are found three 

 notably larger than the rest and three markedly smaller. So in the 

 figures one marked F (/) would correspond either to F or f in Plate 

 IX, figs. 7, 8; A (a) to either A or a of figs. 7 and 8; and so on for the 

 others. And even in the spermatid (Plate X, fig. 40) there are the 

 same size relations; the ten pairs of chromosomes of a spermatogo- 

 nium could be obtained by putting together the ten chromosomes 

 from each of two second spermatocytes derived from the same first 

 spermatocyte ; but the ten single chromosomes of a second sperma- 

 tocyte could be reestablished only by bringing together the ten semi- 

 valent chromosomes from each of the two spermatids resulting from 

 such a spermatocyte. The first maturation mitosis separates from 

 each other the two univalent chromosomes that compose a pair of dis- 

 associated ones in the spermatogonium, and a conjugated pair in the 

 first spermatocyte. 



Finally, a word as to the behavior of certain cellular structures other 

 than chromosomes — only a brief statement, for the present results are 



