158 X. M. STEVICN-. 



certainty; tin- smaller one i- one of the -ma'le-t chromosomes, 

 perhaps y, and .v which seems to have no mate may be i he larger. 



I iKsi SPKKM VTOCY1 i-:s. 



'I'lu- inn-lei of the young <ermatoeyte- arc r-imilar to those 



of tin- spermatogonia Fig. 2.1. clumps of chromatin and line 

 thread-. Before the clumps have disappeared the chromatin 

 t lire. ill- are more or less massed at one side of the nucleus (Fij 

 in a syni/e-i- stage, which resolves it -el I" into an irregular bouquet 

 e ; .Fig. iji \\ith loop-, much coarser than the threads of the 

 prc\iov- st.ige. Frequently one dense chromatin ma-- i- seen 

 half hidden among the bases of the loops. Some of the loop- 

 look as though formed by telosynapsis. but if this is so, the halves 

 of the bivalent loops must later become parallel, for the bouquet 

 stage passes over into a parasynaptic stage (Fig. 10 and Fig. u) 

 which is followed by a -pitvme stage (Fig. 12) in which then' is 

 no longer any e\idence of synapsis. In the spireme stage there 

 always small masses of dense chromatin-like material lying 

 against tin- nuclear membrane, and staining as deeply with 

 thionin as the chromosomes in mitosis. A large heterochromo- 

 some -v is always present , and one or more plasmosdmes can 

 u>ually be found in sections. In early propha-e- (Figs. i $ and 

 14) the chromatin appears as though gathering together about 

 definite centers along the spireme, leaving linin threads between. 

 Sometime- a longitudinal split is visible in sonic of the larger 

 ( -hrom-ome- i Fig. 14.) In several cases at this stage the two 

 heterochromosomes (xy) have been clearly separated as in Fi. 

 13 and 14. hi Fig. 13 a pale plasmosome is also present. 



In most first spermatoeyte spindl< - I \\ o chp>mo->ome.-> are 

 conspicuous (Fig- 15). One of these (.vv) is I he unequal lietero- 

 chromosome, usually bilobed at the larger end (.Y). The other 

 (a) is larger than an \ o! i he ot her- and u it lion t doubt is composed 

 of the two long rods of tin- spermatogonia. As a means of 

 determining the reduced number of chromosomes, a number of 

 spindles in the best material were -elected, .md an attempt 

 was made to draw all of the chromosomes \\hich were divided 

 between two sections. T \\enty-eight were found in a number 

 of the clearest cases. The-e are all -ho\\n I'm- the spindle of 



