256 C. W. METZ. 



the testis. It seems certain, therefore, that no important stage 

 has been overlooked in this region. 



Figs, ii and 12 represent entire, or almost entire, nuclei show- 

 ing the condition maintained by the chromosomes up to the end of 

 the growth period. In each of the three large pairs of chromo- 

 somes the two components diverge medially in the form of a loop ; 

 while in the shorter pair they are usually closely applied through- 

 out their length, with only occasionally a small loop visible. In 

 Fig. ii the entire contents of the nucleus are drawn in position; 

 in Fig. 12 the pairs are transposed to facilitate examination, but 

 are all taken from the same nucleus. 



During the late prophase, as the chromosomes are about to go 

 on the spindle, they become so condensed and shortened that in 

 many cases the loops become closed (Fig. 13). The line of sep- 

 aration between the two chromosomes, however, is still evident, 

 and it is practically certain that no intimate association occurs here. 



It appears, then, that in the case of the three large pairs of 

 chromosomes synapsis has been incomplete, unless it occurred at 

 a very early stage in the growth period and was followed by a 

 secondary opening out to form the loops. The evidence from the 

 early stages may now be examined from this point of view. 



THE EARLY GROWTH PERIOD. 



The early growth stages resemble those of Asilus sericeus (Metz 

 and Nonidez, '21) in a general way, but the details differ mate- 

 rially. In Asilus homologous chromosomes become closely asso- 

 ciated in the final spermatogonial telophase and remain thus as 

 they draw out into threads. Just before they draw out (stage &) 

 the pairs look like single chromosomes relatively clear cut in out- 

 line. In Dasyllfs the chromosomes are likewise paired ; but the 

 pairs form loose, irregularly granulated aggregates, giving little or 

 no indication of the intimate association seen in Asilus. 



The aggregates become more loose and irregular in structure as 

 growth proceeds, and then draw out into irregular, granulated 

 threads. Only two bodies remain condensed : one the X-chromo- 

 some and the other the " dense body " attached to the large pair 

 of chromosomes. As the aggregates spin out into threads a more 

 intimate association becomes possible. Unfortunately as the spin- 



