LIFE-CYCLE OF " CYSTUBLl " iaREC4ULAUIS (mINOH.). 47 



And^ lastly, h and i i^epreseiit, indubitably, earlier stages in 

 this centrosomic division. 



Disintegration of tlie sterile nuclei. — Turning 

 again to the large or sterile nuclei, most of these have 

 increased greatly in size (all the nuclei are drawn to the 

 same magnification) and have noAV very large and apparent 

 centrosomes in connection with them. Of these, there are 

 generally three or four in relation with each nucleus, but 

 they are not always equally closely attached. In k, for 

 instance, while one is touching the nuclear membrane and 

 appears to have pulled it out more or less, another at the 

 left side is not actually in contact. Many of these centro- 

 somes are about double the size of the solitary ones, as if 

 they had not divided for some time. I am inclined to think 

 that the attraction-spheres in connection with these large 

 somatic nuclei do not originate from the latter, in the manner 

 in which the others are formed at the periphery of the sexual 

 nuclei, for I have never seen any, as it were, commencing to 

 be developed with a small centrosomic portion (compare figs. 

 49 and 50) . It is more likely that they correspond to some 

 of the loose division-centres in the cytoplasm which have 

 come into relation secondarily with the large nuclei, in order 

 to bring about their break-up and ultimate dissemination. 



For there is another variety of nuclear figure to be met 

 with, different aspects of which are seen in fig. 52 ; these 

 represent stages in the further alteration of the sterile nuclei, 

 leading to their complete resolution into chromatoid grains 

 and granules, which eventually become scattered throughout 

 the cytoplasm. The granular chromatin or chromatoid matter 

 becomes more abundant and dispersed through the nuclear 

 sap, which is thus divided up into a number of vacuole-like 

 meshes often containing one or more chromatic lumps. These 

 vacuolar meshes (which somewhat resemble a number of vesi- 

 cular nuclei closely aggregated) are at first variable or large in 

 size (a and part of h), but become subdivided into smaller and 

 more uniform ones (c), until they are gradually obliterated, 

 leaving only the granular chromatic material, as in the other 



