The individuality of the chromosomes and their serial arrangement, etc. 127 



of growth such as is seen in the prophases of the somatic divisions. The 

 long axes of the homoeotypic spindle figures may lie at any angle with 

 reference to each other and they may lie in almost any position with 

 reference to the apex of the pollen mother-cell. Figure 30 shows a 

 cell in which the two spindles are at right angles to each other with one 

 lying nearer the apex of the mother cell. This figure shows both a lateral 

 and a polar view of the equatorial plate. The former shows an edge view 

 of a rather flat plate. The polar view makes it possible to count the 

 thirty-seven individual oval chromosomes and there is evidence that 

 here again they are connected together in a series. In such views as 

 are shown in figure 30 the series can be followed throughout most of 

 its length. The behavior of the chromosomes in these homoeotypic 

 divisions is quite similar to that which I have described for the somatic- 

 divisions. The spindles are about one-half the size of the heterotypic 

 spindle figure. In the telophases the remains of the central spindles 

 are still present, but not strongly developed, nor do they persist and 

 form cell plates. No central spindle figure or cell plates between the 

 granddaughter nuclei appear in my preparations. 



The Development of the Pollen. 



Three of the four daughter nuclei, as described by Juel for Carex 

 acuta, now migrate toward the apex of the spore mother-cell and become 

 crowded together. As shown in figure 31 these nuclei are small. They 

 contain for a time conspicuous nucleoles and the chromosomes are clearly 

 defined oval masses distributed about the periphery. The fourth nucleus 

 lies near the center of the general cytoplasmic mass. It is larger than 

 the others, especially as it prepares for the division which forms the 

 vegetative and the generative nuclei. The masses of chromatin which 

 are present at this stage are difficult to count with absolute exactness, 

 but the number is never larger than that of the haploid chromosomes. 



Juel (1900) describes for Carex acuta the development of the gene- 

 rative cell by a sort of free cell formation which, however, judging from 

 his figure is quite different from the method of free cell formation in 

 the embryo sac of Ephedra altissima as described by Strasburger (1880). 

 In Juel's figure the oell boundary is on the inside of an apparently fibril- 

 lar zone. The method of formation of the generative cell, as described 

 by Juel, is furthermore at variance with what described by Strasburger 

 (1884) for Heleocharis, and with what is considered to be the common 

 method in Angiosperms. Friemann (1910) has recently studied the 

 development of the generative cell in a number of monocots (Fritillaria 



