SPERMATOGENESIS OF LEPISMA DOMESTICA 405 



The acrosome 



Although a number of the older writers, notably Platner ('89), 

 Niessing ( 7 96), Field ('95), and Moore (94'), have described the 

 acrosome as arising from the centrosome, Wilson ('06) comes to 

 the conclusion that the work of Henking ('91), Wilcox ('96), and 

 Paulmier ('99) show conclusively that in the insects the acro- 

 some is derived from the nebenkern. 



That this is not the case in Lepisma domestica can be easily 

 proved, for in every stage from the telophase of the second 

 division until the oldest transformation stage in which it was 

 possible to identify structures, the centrosome rod and its change 

 into the acrosome can be followed. 



It might perhaps be argued that the granule is really the 

 centrosome and the rod only a product of the centrosome, formed 

 in somewhat the same manner as the 'battonet' or rodlet in the 

 spermatid of the Pribilof fur seal. As described by Oliver ('13), 

 it arises as a prolongation from the anterior centrosome. In 

 any event, the acrosome owes its origin either directly or indi- 

 rectly to the centrosome. 



Goldsmith ('19) describes in the tiger-beetle a condition, which 

 in view of my own work on the acrosome and middle-piece, is 

 very suggestive. He figures an extra nuclear plate or middle- 

 piece which is formed at the point of junction of the axial filament 

 and the nucleus. It contains several chromatin-staining bodies 

 to which the axial filament is attached. These chromatin 

 bodies move to one side and then toward the anterior end of the 

 nucleus, the filament coming at last to lie against the elongated 

 mitochondrial body (nebenkern) and the bodies to assume a 

 bivalent appearance at the anterior end of the nucleus. The 

 middle-piece becomes drawn out into a granular thread con- 

 tinuous with the axial filament, while the acrosome appears 

 later and fuses with the other two bodies. 



It would appear that the chromatin-staining bodies, to which 

 the axial filament is attached, must be the centrosomes, and that 

 their change in position, due to the rotation of the nucleus, is 

 exactly parallel with what occurs in Lepisma domestica. 



