B. F. LU I.MAN. 



There is no evidence that it has disappeared as tin- nucleolus 

 usually does; it seems simply to have become a tetrad. The 

 elaborate formation and dividing of the tetrad would argue 

 against this disappearance also. This chromatin nucleolus can 

 be traced no further. During the equatorial plate stage of divi- 

 sion the chromosomes all lie in one plane and it is impossible to 

 identify any particular one as the transformed nucleolus. Nei- 

 ther does any one lag behind in divisions in the metaphase nor 

 in the movement toward the poles in anaphase (Figs. 26 and 32). 

 If this body forms a chromosome, as it undoubtedly does, that 

 chromosome behaves exactly like all the others. 



The number of chromosomes is of great interest here if this is 

 a true accessory chromosome. According to cither the McClung 

 or the Wilson type of an accessory chromosome, or the Wilson 

 type of a hcterotypic one, there should appear an odd number of 

 chromosomes plus this additional one; or as McClung has found 

 in Orchesticus sixteen chromosomes and the accessory one. 



In all the counts made in Platyphylax, however, the number of 

 chromosomes for both the reduction divisions \vas found to be 

 thirty. This is the result of repeated trials. These countings 

 are as easy to make as of dots on a piece of paper (Figs. 27, 28, 30) 

 as the polar plate views are numerous and the chromosomes are 

 short. There is some variation in size in the chromosomes in 

 polar view but it is impossible to pick out one of them as the 

 special structure that has been followed. 1 



It will be seen from this description that while this body re- 

 sembles the accessory chromosome of McClung in many respects 

 still it differs from it in one very essential one. It apparently 

 forms a tetrad that divides in both divisions and so each sperm 

 would receive one fourth of it. This w r ould make it impossible 

 for it to serve as a sex-determinant, for all the sperms would 

 receive a part of it, and not half of them, as would happen if we 

 credit the observations of McClung or of Miss Wallace (18). 

 This body in that case could not be a sex-determinant. 



'In cutting abdomens to get the development of the testes I have cut and stained 

 as many, or more, females than I have males. The divisions in the former can be 

 readily obser% r cd here as they are much larger than in the testes. The nucleolus 

 undergoes a similar lengthening out, and then forms a part of the pireme. M.n- 

 shall (9) in his paper on the development of the ovary also shows several figures 

 that strongly suggests this. 



