The Spermato genesis of Aplopus may en. 23 



lent chromosomes of the equatorial plates of the first maturation mitosis 

 and pairs of chromosomes of the spermatogonial stages can be found, but 

 these are hardly of sufficient precision to seem convincing in support of the 

 theory of the individuality of the chromosomes or to add anything con- 

 firmatory of the selective character of synapsis. 



The accessory chromosome in Aplopus, it will have been noticed, passes 

 undivided to one of the poles of the first maturation spindle. In the late 

 telophase and during the stages when the daughter-nuclei are formed and 

 the ordinary chromosomes pass into the nuclear reticulum, the accessory 

 becomes more or less bipartite, but always retains its sharp contour and 

 deep-staining reaction and its usual position in close connection with the 

 nuclear wall (figs. 92, 93, 94, 95, and 96). The accessory chromosome 

 still retains these distinguishing characteristics throughout the resting stage 

 (fig. 97), which is interpolated between the two maturation mitoses, as well 

 as during the early prophase of the ensuing division. At this stage the cyto- 

 plasm of the secondary spermatocyte also frequently contains several larger 

 or smaller masses of eliminated chromatin. Obviously only one-half of the 

 secondary spermatocytes resulting from the previous division can have the 

 accessory chromosome. Study of many sections shows without a doubt that 

 only about one-half contain the chromatic body (or any body that reacts to a 

 selective chromatin stain) which we have identified as the accessory chromo- 

 some. Figures 101 and 102 show two secondary spermatocytes side by side, 

 one with the accessory chromosome and the other without it. 



According to the several investigators, there is variation among the 

 Arthropoda in regard to the time when the accessory chromosome divides, 

 i. e., in the first or second mitoses. McClung (1900 and 1902^), in the case 

 of several Orthoptera, has traced the accessory back into the spermatogonial 

 rest-stages, and finds that it subsequently divides only in the first sperm- 

 atocyte division. Baumgartner (1904) in Gryllus domesticus, Stevens 

 (1905) in Stenopelmatus and Blatella gcrmanica, and Otte (1906) in 

 Locust a viridissima find that this chromosome divides in the second division 

 instead of the first. Moore and Robinson (1905) claim that the accessory 

 in Periplaneta americana is only a plasmosome that dissolves before each 

 division and is reconstructed after it. This can not be the case in Aplopus. 

 No indication of a true nucleolus (plasmosome) can be demonstrated by 

 any of the several staining methods employed in any of the cells in the 

 line of the spermatogenesis. 



SECONDARY SPERMATOCYTE. 



The prophase stages of the second maturation division present nothing 

 extraordinary. The succession of events is similar to that of the sperm- 

 atogonial division and an ordinary homeotypic mitosis, with the exception 

 of the presence of the accessory chromosome. The latter never passes into 



