Spermatogenesis in spiders. 543 



found in the young- spermatocytes was nine. A study of later 

 stages makes it seem probable that this is approxiamtely the total 

 number in the cell. In Fig-. 75, we may count seven small chromo- 

 somes on one side of the cell and ten on the other, the side which 

 receives the accessory having three more of these bodies than the 

 other. In Fig-. 77, there are five small chromosomes on one side of 

 the cell plate and eight on the other. Here ag-ain the side bearing- 

 the accessory receives a larger number of chromosomes. Since, as 

 pointed out, the ctetosomes seem to lose their attachment to the 

 accessory chromosome, after the latter body has reached one pole, 

 as in Figs. 75 and 77, we should count three of the small bodies 

 on the side with the accessory chromosome as being ctetosomes. 

 In the case of the cell shown in Fig. 75, there should be present 

 then three ctetosomes and fourteen planosomes, seven of which are 

 going to one daughter spermatocyte and seven to the other. Assuming 

 that the planosomes have divided, there would be in tlie primary 

 spermatocyte cells three ctetosomes and seven planosomes. As 

 already noted, as many as nine small nucleoli have been found in 

 the rest period (Fig. 68). 



The ultimate distribution of the planosomes is subject to much 

 iiTegularity in the first maturation division. Because the time of 

 division varies and the planosomes are apt to be hidden by some 

 overlying autosomes, it is not possible to determine the exact 

 distribution for any given cell. On the whole the daughter second 

 spermatocytes seem to receive about the same number, although 

 there are some striking exceptions. In the cell shown in Fig. 76, 

 we find that most of the planosomes are going to one side. Such 

 case are rare, however, occuring in 4 out of 51 counts. On the 

 other hand slight irrégularités are very common. 



Second maturation division. 



After the telophase of the foregoing division the daughter 

 second spermatocytes pass into a resting stage, during which the auto- 

 somes lose their compact form and their affinity for the stain. The 

 hétérochromosome elements persist as deeply staining nucleoli. 

 A careful examination of cells at this stage has been made in order 

 to determine if any of the small bodies which were found in the 

 spindle were left out in the cytoplasm when the nuclear wall was 

 formed. There is no evidence that this has happened. On the other 

 hand, we find lying free in the reticulum of this resting stage, very 



36* 



