Spermatogenesis in spiders. 529 



The accessory chromosomes and the ctetosomes, during the 

 maturation period are so closely associated that we can discuss their 

 behavior together. After the telophase of the last spermatogonial 

 division the autosomes enter into a rest period during which the 

 hétérochromosome elements retain their affinity for the stain. Wallace 

 has followed the behavior of the accessory chromosome elements in 

 great detail for Agalena naevia, during this period, and my own 

 observations conflrm Wallace in every point. 



The rest period interests us particularly as we have here the 

 first direct evidence that the primary spermatocytes are of two 

 kinds, as we were led to suppose from the behavior of the cteto- 

 somes in the earlier division. During the rest stage, the nucleoli 

 fuse more or less completely, but towards the end of the period the 

 single nucleolus thus formed breaks up. This has been interpreted 

 as meaning that the fusion is only superficial and does not involve 

 an interchange of nuclear material. In the later stages the accessory 

 chromosome retains its compact form and its affinity for the stain, 

 but I have not been able to find any trace of the ctetosomes. 



After the second contraction, the chromosomes lie scattered 

 about the nuclear cavity and the accessory chromosome may be 

 identified by its size and shape. It appears in two forms, shown 

 in Figs. B and C. In part of the cells the accessory has the simple 

 two rod form, as shown in Fig. B, and in part of the cells it has 

 a quadripartite form shown in Fig. C. When the bivalent form is 

 present in the cell, we fail to find a quadripartite body and vice 

 versa. It may be doubted whether or not it is posssible to identify 

 the chromosomes certainly. Much time has been spent in studying 

 the chromosomes of this period, and I feel sure I have not mistaken 

 some autosome for the accessory body. This is because no autosome 

 of this period has the short two rod form so characteristic for the 

 sex-chromosome, and furthermore the relative size of the accessory 

 would be sufficient to separate it from the autosomes. In Fig. D, 

 we have the accessory just before it is drawn into the spindle of 

 the first division. The inequality of length shown by the two rods 

 cannot be entirely explained by foreshortening, as it has been seen 

 in many cases. 



During the first division the accessory chromosome shows a 

 variation in form. In Figs. 33 and 36, we have the simple bivalent 

 form. In Figs. 34 and 35, the accessory is made up of four elements 

 while, in Fig. 37, there is an inequality in the size of the two 



