136 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



apices. In the case where four nuclei are visible (Fig. 25) no 

 tetrahedron is formed; the nuclei lying opposite each other form 

 a pair and one pair usually lies at a slightly higher plane than 

 the other. Where two nuclei alone are visible in a pollen mother- 

 cell this may be due to the fact that the four-nucleated pollen 

 mother-cell is so situated as to expose only two nuclei. It is 

 likewise possible, although not so common, that the two nuclei 

 are the results of cell division after the heterotypic division followed 

 by nuclear division. From such figures it is accordingly clear 

 that successive division or bipartition of the pollen mother-cell takes 

 place as described below. 



In the late telophase of the second, as in the first division, 

 chromosome-like bodies appear on the spindle fibers and in the 

 cytoplasm. In FiG. 24 three nuclei appear in a cell with their 

 spindle fibers. One of the nuclei is intact while the other two 

 are cut through in sectioning. 



On the spindle fibers three bodies appear like those seen in the 

 first division (see Fig. 23). In this figure in the complete nucleus 

 there are unquestionably ten more or less spherical chromosomes, 

 while in the preceding sections of the same cell ten chromosomes 

 likewise appear in the other nuclei. In a slightly older stage, as 

 shown in Fig. 25, of Drosera filiformis, two large bodies are seen 

 on one of the central spindles with a number of smaller ones 

 distributed through the cytoplasm. In another section of a late 

 telophase stage of the homoeotypic division shown in Fig. 35 I 

 found cytoplasmic bodies which stain faintly and were promiscu- 

 ously scattered through the cytoplasm and on the spindle. In 

 practically all cases where chromosome-like bodies appear, ten 

 distinct chromosomes may also be clearly seen. What the sub- 

 sequent history of these bodies is after the second division I have 

 been unable to learn. It is quite clear that they do not form 

 extra nuclei or dwarf nuclei as shown by Rosenberg, Tischler, 

 Juel, Gates, and others. It is not wholly impossible, as held by 

 Juel and Tischler, that these bodies may be the forerunners of 

 abnormalities which eventually lead to the disintegration of the 

 pollen mother-cell, but I have been unable to prove this. The 

 formation of the normal pollen grains follows. The nuclei are 

 reconstructed as they are after the heterotypic division. The 

 central spindle fibers do not disappear but remain as prominent 



