MARK AND COPELAND. — SPERMATOGENESIS OF THE HONEY BEE. 105 



arrangement of the spermatocytes. These become grouped in the cyst 

 in a radial manner with their interzonal bodies clustered at the centre, 

 very much as the spermatogonia are arranged in their follicles, l^his 

 constant orientation of the cells in relation to the cluster as a whole 

 makes it possible to use without ambiguity the terms proximal and 

 distal. 



The nucleus, which hitherto has been near the centre of the sper- 

 matocyte, now changes its position, migrating to the distal end of the 

 cell ; it generally comes to lie close to the corresponding centrosome. 

 Coarse fibres, which stain deeply in iron haematoxylin, now make their 

 appearance, extending from the distal centrosome around the nucleus 

 to the region of the proximal centrosome. At this time the finger-like 

 process terminating in the proximal centrosome may be reduced to 

 a very narrow cytoplasmic stalk (Figure 4), so that it is difficult to 

 observe whether the fibres at this stage actually reach the centrosome 

 or not. It is, however, probable that they do, as they may be traced 

 for some distance into the base of the finger-like projection. 



Changes, which we shall not attempt to describe here in detail, have 

 meantime taken place within the nucleus. The chromatin passes 

 through a spireme stage and gives rise — apparently by segmentation 

 and a concentration of the segments — to sixteen double, dumb-bell 

 shaped chromosomes (Figure 5), which lie in rather scattered positions. 

 The bivalent structure of the chromosomes becomes less conspicuous 

 when, a little later, the nucleus enters on the metaphase. 



At about this stage the nucleus elongates, first in the direction of 

 the distal end of the cell (Figure 7), and later at the opposite end, until 

 finally it becomes spindle shaped, extending more or less completely 

 from pole to pole (Figure 8). At the time of nuclear elongation 

 a few thick, deeply staining intranuclear-spindle fibres are seen con- 

 necting the chromosomes with the distal centrosome, and later, as the 

 nucleus increases in length, similar fibres also run to the proximal 

 pole of the cell. These fibres are clearly separate from one another 

 in the region of the chromosomes, but as they approach the centro- 

 somes they seem to coalesce, until it is impossible to make out 

 separate elements. In the cell represented in Figure 8, a bundle of 

 fibres is traceable fi'om the chromosomes to the distal centrosome, and 

 on the opposite side of the chromosomes a bundle may be followed for 

 some distance into the base of the finger-like process which terminates 

 in the proximal centrosome. In other cases the nucleus, although 

 much attenuated at the poles, has been clearly seen to stretch from one 

 centrosome to the other, the chromosomes being midway between 

 the two centrosomes. I find nothing to corroborate the condition 



