STRUCTURAL ELEMENTS OF THE CYTOPLASM. 



upon seeing preparations of the osmiophilic platelets and of the 

 pseudochondriome have asked upon what basis any separation can 

 be made. It is true that the one class of bodies is composed of 

 (flat) discs, the other of spherical granules, but it was thought 

 that possibly the morphological differences were due mainly to 

 the different technical treatment. The first suspicion in my 

 own mind of the distinction between these two components came 

 from a study of the moss androgonia, and I eventually spent 

 several months during the past fall trying to get critical evidence 

 on this point. This evidence I believe is now conclusive, and the 

 pseudochondriome and osmiophilic platelets must be considered 

 as entirely distinct kinds of bodies. 



The pseudochondriome consists of a larger or smaller number 

 of granules scattered throughout the cell and showing no im- 

 portant relations to the spindle in cell division (in root-tips at 

 least). These bodies rarely blacken with osmic acid, an exception 

 being the hyacinth, where they are finely demonstrated as jet 

 black granules on a clear background (Fig. 3). They are 

 occasionally stained by Benda's method or Fe-hematoxylin, but 

 by far the best method is that of Kull. In good Kull preparations 

 (Fig. 4) each granule shows a typically vesicular structure, 

 with a brightly stained periphery and a light center a com- 

 position quite common in the true chondriosomes of molluscan 

 and lepidopteran sperm cells, for example. In older cells these 

 granules often show signs of elongation, etc., but their typical 

 form in the earliest meristem is that of vesicular granules. My 

 results accordingly agree in a general way with Guilliermond's 

 findings, and the facts also indicate that his interpretation of 

 these bodies as chondriosomes is also correct. But it must 

 again be emphasized that there is nothing in root-tip cells 

 themselves upon which to base such a decisive conclusion as to 

 their homologies with the animal chondriome. 



Finally as to the distinction which I have drawn between 

 these bodies and the osmiophilic platelets, three lines of sup- 

 porting evidence combine, it seems to me, to render it secure. 

 In the first place, there are the obvious differences in morphology. 

 The pseudochondriosomes are spherical, and thus, while pre- 

 senting the appearance after proper staining of rings, these 



