84: KANSAS UNIVERSITY QUARTERLY. 



Figs. 24 and 25 show the metaphase of the second spermatocyte 

 division. The diad character of the chromosomes is here plainly ex- 

 hibited, and there is no obliteration of the boundaries between the 

 chromatids, which is so characteristic an appearance of the elements 

 in the first spermatocyte. The chromatids are superimposed upon 

 each other in the plane of the spindle axis so that their separation in 

 metakinesis is a very simple matter. The union of the diad elements 

 seems to be a slight one. their only point of contact, usually, being 

 that at which the fibers of the first spermatocyte division were attached. 



The character of the spindle, also, throws some light upon the na- 

 ture of the chromosomes. It is small and weak compared with that 

 of the first spermatocyte division, showing that the force necessary 

 for the separation of the diad elements is slight. It is also more un- 

 stable and transient than that of the preceding cell generation. Figs. 

 26 and 27 show anaphases of the second spermatocyte. It is notice- 

 able that, with the polar movement of the chromosomes, the spindle 

 is also elongated, so that it stretches throughout most of the length 

 of the cell. As a result of this longitudinal extension, the spindle 

 becomes almost non-existent at the end of the anaphase, being rejore- 

 sented only by a few masses of archoplasmic substance. 



The telophases are brief, and soon from the second spermatocyte 

 are formed the spermatids which transform directly into spermatozoa. 

 The changes involved in this process are not connected with the sper- 

 matocyte divisions, and so will be left for a subsequent paper, where 

 they may be considered in detail. 



Only casual mention has as yet been made of the interesting struc- 

 ture described under the name of "accessory chromosome" in a pre- 

 vious i^aper (23). The material for the earlier work was derived 

 exclusively from the Locustidte, and it was surmised that, although 

 this element might be found in all insects, there would be more 

 £)r less difference in its form and behavior in various species. 



These surmises have been verified by subsequent study. A great 

 many specimens, drawn from the different orders of insects, have been 

 examined, and in no case was the absence of the accessory chromosome 

 noted. On the other hand, considerable differences were found to 

 exist in the general character of the element. Since, however, the 

 material for the present paper was taken from the Orthoptera, com- 

 parisons will not be instituted except among the subdivisions of this 

 order. 



Some broad differences with respect to this nuclear element ex- 

 ist between the Locustidse and the Acrididse. In the former family, 

 as previously described, the accessory chromosome arises some time 

 before the last of the spermatogonial divisions. It may be observed 



