March, 1908.] On the Cytology of Synch ytrium. 281 



trosomes. This is particularly true of the stage represented by 

 fig. 14, where the chromatin is in spireni in every one of the nuclei 

 in the three cysts observed although the aster has apparently run 

 its course. This suggests that we have not reached a full under- 

 standing of the phenomena as yet. Nevertheless in any case it 

 seems certain that the resting nucleus, with its chromatin all or 

 nearly all concentrated in the single globular karyosome (figs. 

 12-13) must be connected with the chromosomes of anaphase 

 by a series of stages not far divergent from these here described. 



The cvcle of astral activity which is the main interest of the 

 present paper may now be taken up. When at their maximum 

 size the rays are relatively few, long, and so thick as to be 

 clearly visible under a low magnification. They are not always 

 straight (fig. 7) and do not always center exactly in the focus 

 of the aster (fig. 12). Quite often they have thickenings or 

 granules along their length (figs. 2, 4. 5, 12). These are apt 

 to be located at the intersections of the rays with the strands 

 of the cytoreticulum (fig. 5). At later stages the central 

 deep staining granule gives place to a larger diftuse gran- 

 ular area in which there may be still some deeper staining granules 

 but thev are more minute than those which preceeded them. 

 The ravs at the same time become finer and more numerous till 

 thev resemble the spindle fibres in ordinary mitosis. The granu- 

 lar area then appears to enlarge while the rays disappear and 

 finally the centrosome seems to become simply a densely granu- 

 lar mass of cvtoplasm which does not stain more deeply than the 

 general reticulum (fig. 14). In this stage the centrosome re- 

 sembles greatly the so-called attraction spheres of some animal 

 cells, e. g., some stages of Ascaris. This mass is then dissipated 

 into the general cytoplasm by imperceptable stages thus leaving 

 the nucleus without centrosomes as it began. As before indicated 

 the condition of the nucleus at this time throws some doubt upon 

 this sequence so that the history of the centrosome, like that of 

 the nucleus, may be subject to some reidsion. But it appears sure 

 that the centrosome arises de novo out of the cytoplasm and dis- 

 integrates into cytoplasm again whether the sequence of events 

 be exactly that given or not. 



The remarkable feature of the aster, however, is the relation 

 of the ravs of the centrosome to the reconstruction of the nucleus. 

 At an early stage a vacuole appears around the chrom.osomes. 

 This is at first quite without a membrane (fig. 2) but very soon 

 those astral rays which are nearby come to form a cone enclosing 

 it (figs. 3,4). Soon the ends of the rays bend around the vacuole 

 and enclose it forming the unclear membrane (figs. 5,*{), 7). These 

 membrane-forming rays may be observed to taper greatly from 

 the center toward the' curved ends (figs. 4, 7). At their thicker 

 ends they are very heavy indeed and stain deeply so that they 



