282 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



concave, and two flattened; no cell-plate Is shown. In the text he 

 only says " the cell-walls are formed between these four nuclei, pro- 

 ducing the normal tetrad." Figure 55 shows an instance of biparti- 

 tion of the pollen-mother-cell. In Daphne he ( 5 Oa) shows a marked 

 furrow on two sides of the mother-cell, which is about to undergo 

 tetrahedral quadripartition. There are also indications of cell- 

 plates here. The mother-wall is about one ninth of the diameter 

 of the cell. 



Rosenberg (57a) shows a schematic drawing, text figure III., 

 of successive bipartition in Hieracium venosum. In text figure V. 

 he has a tetranucleate mother-cell which is rectangular, and without 

 a thickened wall. He says (p. 149), "The following division stages 

 correspond with, for instance, those of Tanacetum or Taraxacum, 

 and the development of the pollen cells does not offer anything 

 extraordinary." In no instance does he show a cell-plate. Of 

 Drosera he shows several figures of quadripartition. Figure 22 

 (576) shows the four nuclei in one plane in the form of a rhomb, 

 with four spindles of equal size, and a fifth small spindle between 

 the two opposite nuclei which are in closest proximity. No 

 indication of a cell-plate is shown. Figure 21 shows a similar 

 stage with tetrahedral arrangement of the nuclei. In figure 2 the 

 mother-wall is shown much thickened and a cleavage furrow is 

 beginning in the equator at the metaphase of the homeotypic 

 division. No cell-plates are likewise shown here. During the course 

 of my investigation on Nicotiana and other forms described below, 

 Levine (39) in his studies on the germ cells of Drosera observed 

 both bipartition and quadripartition in which the mother-cell-wall 

 is much thinner and more angular, as in Asdepias, Ceratophyllum, 

 etc. His figures show no broad constriction furrow, and at the 

 stage when partition has apparently just been completed, the 

 appearance is as if the mother-cell had divided by cell-plate forma- 

 tion. Drosera is one of the plants in which the four pollen grains 

 thus formed never separate. Whether this has anything to do 

 with the mode of cytokinesis in microspore formation is a problem 

 for further investigation. 



In her study of Parnassia Miss Pace (52) presents a figure, 54, 

 which shows tetrahedral (luadrijiartition with the invagination of 

 the thickened mother-wall along the eciuaturs. Fibers are shown 

 crossing the equator, but there is no indication of a cell-plate. 



