$ McCLUNG. 



from the spireme thread, it commences to shorten and grows into 

 the form of a horseshoe, and is finally to be distinguished from 

 the chromatic rings only by its deeper staining quality and by 

 the smoothness of its outline. In the formation of the mitotic 

 figure of the first spermatocyte division, it assumes its position on 

 the outside of the group of chromosomes as it did in the sper- 

 matogonial division, and again has the boomerang shape that 

 marked its appearance in the early figures. When the chromatin 

 separates and moves to the two poles, the accessory chromosome 

 divides longitudinally and presents the appearance of two horse- 

 shoes with their rounded ends in contact. In the second sper- 

 matocyte division apparently the same process is followed. 



43. "The recently formed spermatids possess a nucleus in 

 which the ordinary chromatin is extremely scant and very weak 

 in staining power, while the accessory chromosome shows as 

 prominently as ever and stains in the same uniform manner. It 

 is not easy to trace out the part that the different elements of the 

 nucleus take in the formation of the spermatozoon, but in the 

 light of present knowledge it appears as if the accessory chro- 

 mosome was prominently concerned in the formation of the head. 



44. "In seeking to point out the features that characterize this 

 peculiar nuclear element, perhaps the most striking thing to be 

 noticed, is the almost uniform staining power exhibited. 



45. "Thus there seems to be no reason to suppose that the 

 accessory chromosome of Xiphidium arises by the direct trans- 

 formation of one of the ordinary ones, although such a change 

 may be possible. This does not argue against the chromatic 

 origin of the body, however, for it is almost certainly modified 

 chromatin, but in Xiphidium it arises during the resting stage and 

 may represent derivative substance from one or all the chromo- 

 somes." 



46. A later, more detailed, study of various members of the 

 Acrididae ('oo) gave me a much better insight into the nature of 

 the accessory chromosome and the result of my observations 

 upon its behavior during the spermatocyte divisions are summa- 

 rized in the following words : 



47 "As a result of the last spermatogonial division, the much 

 reduced daughter cells are each provided with the somatic num- 



