71 8 Beer.' — Studies in Spore Development. If. 
the two species of Crepis very obvious vacuolization of the chromosome 
bodies occurs at the same time. 
Moreover, the vacuoles tend much more regularly than in the previous 
division to occupy a central position, and thus to divide the chromosomes 
into two lateral halves. Figs. 53, 54, and 55 show these facts In the cases of 
Tragopogon , Crepis taraxacifolia , and Crepis virens . 
It will be observed from these figures that by the time that the division 
is completed and the daughter nuclei entirely reconstructed, the substance of 
the chromosomes has become more or less distributed through the nuclear 
cavity. As in Matricaria , there is some variation in the degree to which 
this dispersal of the chromatin attains, and In both cases the state of 
chromatin aggregation no doubt depends upon the metabolic activity of 
the cell. 
No definite prochromosomes are, however, to be seen In the young 
pollen-grains, I have illustrated these nuclei in the young tetrads of Crepis 
taraxacifolia (Figs. 57-9). 
From the time of the telophase of the first meiotic division until the 
second division is completed and the pollen-tetrads formed, the cytoplasm 
contains a number of deeply staining granules or droplets in Crepis taraxa- 
cifolia (Figs. 56-8), and also in Crepis virens (Fig. 55). In Matricaria 
Chamomilla deeply staining particles are also very obvious in the cytoplasm 
during, and for a short time after, the heterotype telophase, but In the later 
stages these seem to disappear in this plant (Figs. 28 and 33 a). The presence 
of chromatic granules in the cytoplasm during these stages of mitosis has 
often been described in other plants, and in one case at any rate the origin 
of these particles in the cytoplasm has been ascertained. In Polypodium 
vulgare Professor Farmer and Miss Digby (8) succeeded in tracing at least 
some of these droplets through the nuclear membrane into the cytoplasm. 
In the two species of Crepis I have spent much time in the attempt to 
trace the origin of these granules lying in the cytoplasm during the later 
stages of the meiotic phase. Granules or droplets of chromatin often appear 
to be passing through the nuclear membrane, but in the great majority of 
these cases a careful change of focus shows that the granules are really lying 
against one side of the membrane and are not actually passing through. 
A few rare instances were, however, met with In which even with the most 
careful focusing the chromatin droplet still gave the impression of being In 
the act of passing through the nuclear membrane. I have drawn such 
a case in Fig. 60. In Matricaria several cases more or less similar to the 
one represented in Fig. 33 a were met with. Here the chromatin droplet is 
on the outside of the nuclear membrane, but it strongly suggests that it Is 
derived from the nuclear chromatin, perhaps before the nuclear wall has 
been deposited, and that it is passing further into the cytoplasm of the cell. 
From these facts I believe that the chromatin granules lying in the cyto- 
