192 



ROBERT H. BOWEN. 



ordinarily interesting to know whether the vacuole primordia, 

 more particularly their enclosing membranes, play any joint role 

 in the formation of the limosphere, and might thus be brought 

 into line with the Golgi apparatus. But thus far there is not 

 the slightest evidence of such a cooperation, indeed at present 

 no vacuome has been demonstrated in the androcytes, and at 

 the moment I am inclined to the opinion that no such cooperation 

 occurs. 



The particular mistake of my earlier account, which I wish 

 now to make clear, has to do with an error in interpreting a 

 stage like Fig. 14. It was my belief, in the absence of an extensive 

 series of successful Golgi preparations, that the limosphere in 

 early androcytes was an early stage in the history of the chon- 

 driome. The blackened granule and its enclosing vacuole gave 

 to it an appearance exactly like that of the nebenkern in an 

 insect sperm. I believed that the limosphere proper appeared 

 only at a later stage. The whole picture (Fig. 14) is so strikingly 

 like an insect spermatid that the error was not suspected by 

 myself or by others who were familiar with osmic acid prepara- 

 tions. As a matter of fact, the moss androcyte does possess a 

 remarkable similarity to a generalized animal spermatid, though 

 not completely in the manner I suggested at first. 



This clears the ground for a final disposition of the plastidome 

 and pseudochondriome in the developing sperm. As regards the 

 plastidome I have been as yet unable to follow its history, and 

 its identification at any stage in the androcytes is still uncertain. 

 It is possible that it degenerates, but it seems more probable 

 that it either remains distinct and is disposed in some charac- 

 teristic place in the sperm, or merges inextricably with the 

 pseudochondriome. In the first case it would be practically 

 necessary to consider the plastidome as an independent cyto- 

 plasmic component equal in rank with the chondriome and 

 Golgi apparatus. In the second case the conclusion would seem 

 justified that the plastidome is merely a specially differentiated 

 portion of the pseudochondriome, as has long been contended by 

 Guilliermond. Confusion between these two components in the 

 spermatid led me to identify the plastidome with the pseudo- 

 chondriome in my earlier report, and to draw the conclusion 



