levine: somatic and reduction divisions in DROsEiLrV 131 



undoubtedly like the prochromosomes of Rosenberg ('04) and 

 Overton ('05). With careful study their number can be fixed 

 at about twenty. In the case of D. longifolia, which is the species 

 of special interest in view of Rosenberg's contention that its so- 

 matic chromosome number is forty, I have been unable so far by 

 a study of the prochromosomes in the somatic cells to verify his 

 conclusion. In my preparations, however, the prochromosomes 

 were favorably shown in only a few cells. l-The prochromosomes stain 

 sharply with gentian-violet but other smaller granules which stain 

 faintly are also present. The nucleole stains a ruby red. Not infre- 

 quently one can find a cell with two or three nucleoles. The cyto- 

 plasm stains a faint orange. In the cells of the anther wall the 

 cytoplasm is not at all dense and a number of large vacuoles appear, 

 while in the immature pollen mother-cell the cytoplasm is densely 

 granular with few or no visible vacuoles. The prochromosomes 

 appear to spin out and form a spireme which fills the entire cavity of 

 the nucleus. The band is narrow and seems to be continuous. The 

 nucleole disappears, the spireme segments and twenty more or 

 less angular chromosomes can be counted before a bipolar spindle 

 can be recognized. In this stage also I had relatively little 

 material of D. longifolia and considerable dififtculty was encoun- 

 tered in making a definite count. The nuclear membrane 

 disappears and the chromosomes are seen in equatorial plate 

 stage (Fig. 2). The spindles are like those described by Rosen- 

 berg ('99) for Drosera rotundifolia — some may be pointed while 

 others are broad-poled. In the early telophases of the somatic 

 divisions a rudimentary cell plate (Figs. 5, 6) is developed but the 

 incipient stages of the process were not observed. The fibers at 

 the center of the spindle shorten as shown by Strasburger ('82), 

 Timberlake ('00) and others. New fibers also appear which make 

 the spindle seem to bulge out until the diameter of the spindle 

 reaches across the width of the cell (Figs. 6-7). At the same 

 time the nuclear masses seem to approach each other (Fig. 6). 

 The cell plate continues to grow peripherally until it forms a 

 diaphragm cutting the cell in two (Fig. 8). In the later telophases 

 the chromosomes begin to anastomose and there seem to be lines 

 of flowage of material from each chromosome. The stain is taken 

 more deeply at the center of the mass while its intensity decreases 

 as the distance from the center increases. The flowage progresses 



