302 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



this fact, together with the large amount of collateral evidence 

 presented in the review of the literature given, it seems quite 

 probable that in a majority if not in all of the dicotyledons in 

 which quadripartition of the pollen-mother-cells occurs, there is 

 a division of the cytoplasm by constriction furrows, as above 

 described for Nicotiana, without the development of a true cell- 

 plate. 



In Helianthus there is a greater number of blue-stained granules 

 irregularly distributed throughout the cell at about the time of 

 the beginning of the invagination of the mother-wall. They 

 also frequently appear to be in chains, but these may lie either in 

 or out of the equatorial plane. No evidence was found of these 

 granules being arranged in the form of a plate. The same thick- 

 ening of the wall occurs as in Nicotiana. 



In Chrysanthemum frutescens a large number of irregularities 

 occur as to number of pollen-grains formed from a single mother- 

 cell. There sometimes appears to be only one, no division of the 

 mother-nucleus taking place. In other cases, two, three, four, or 

 five nuclei may be present and as many spores may be formed, 

 respectively. When five are present they may all lie in one plane 

 with one nucleus in the center and the other four at equal distances 

 about it. In such cases there are eight spindles. The tetrahedral 

 arrangement is, on the whole, less frequent in Chrysanthemum 

 than in Nicotiana. The size of the respective nuclei in the mother- 

 cell also varies. Sometimes it appears as if a nucleus was organized 

 from a single chromosome, and several instances of very small 

 pollen-grains have suggested that these might be later stages of 

 such an aberrant development as Juel found was the case in 

 Hemerocallis. A study of nuclear division in Chrysanthemum 

 would doubtless lead to interesting results. The cell-division of 

 the pollen-mother-cells appears to be as described above for 

 Nicotiana. This last statement may also be made for the other 

 member of the Compositae that was studied, namely Ambrosia 

 artemisiifolia. 



The author is able to confirm the statement of Miss Digby that 

 constriction occurs in Primula. No evidence of a cell-plate was 

 found. The isthmus between the nuclei in the later stages takes 

 the form of a narrow neck, the daughter cells being farther 

 apart than in Nicotiana. With iH's^anl lo Trnpacolum, mother- 



