A STUDY IN CHROMOSOME REDUCTION. 609 
or nearly so—through the nucleus. It reveals some features of considerable import- 
ance. First, it will be noted that as the development of the chromosomes proceeds, 
by the condensation of the chromatin composing the spiremes, the longitudinal 
fission becomes less evident and is only made out with difficulty. Second, it should 
be noticed from this figure (15) that the number of spiremes is large—that the real 
act of reduction has not yet occurred. It shows that throughout the entire period 
between the stage represented in fig. 1 and the stage represented in fig. 15, we 
_ have had the diploid number of chromatin threads. The important thing that has 
occurred during this period is not the fusion or blending of filaments, or the 
interchange of chromatin substance or influence between them, but the actual 
longitudinal fission of somatic chromosomes, which becomes evident at a very early 
period. Longitudinal fission has been observed so frequently and in so many types 
of plants, that there is now no room for doubt as to its general occurrence in the 
spiremes of all somatic mitoses. I am quite convinced that this longitudinal fission 
in the early meiotic prophase corresponds to the longitudinal fission which occurs in 
ordinary somatic divisions. This view has also been expressed by Gates (1911) 
and one or two other writers. If this interpretation is the correct one, then we have, 
up to the stage represented in fig. 15, what are really the preliminary stages of an 
ordinary somatic mitosis. If the two longitudinal halves of each chromosome were 
allowed to separate from one another and pass to opposite ends of the cell, an 
ordinary somatic division would result with the diploid number of chromosomes. 
This, however, in the meiotic prophase is prevented by a series of events which 
bring about the haploid condition. 
Before going into a description of the details of these events, | would call 
attention to an important physiological aspect of the cells in which the meiotic 
phenomena occur. I have already mentioned the fact in another paper (Lawson, 
1911) that these mother cells—in which this peculiar form of nuclear activity 
occurs—are really temporary storage cells. A mere superficial examination will suffice 
to convince anyone that “storage cells” is the correct designation for these structures. 
In the first place, the presence of large quantities of dense granular food substances 
suspended in the cytoplasm points to this conclusion. The cytoplasm is so charged 
with food substances, that vacuoles as we ordinarily have them in vegetative cells 
are not present. Another feature which characterises these cells and leaves no 
room for doubt as to their storage function, is to be seen in the thickness of the cell 
walls. The gradual thickening of the cell wall may be easily followed in the series 
of stages here figured. For instance, in figs. 8 to 25 and from 28 to 43 the thickening 
of the wall progresses with other activities until, as shown in figs. 41, 42, and 48, it 
becomes enormously thick. There is no doubt in my mind that these thick walls are used 
as reserve food. The wall becomes completely dissolved after the second meiotic division, 
and is no doubt absorbed by the developing microspores (figs. 43, 44, 55, 57, 58).* 
* Sacus, J. (1875), Text-Book of Botany (English translation. Clarendon Press, Oxford), p. 485. 
TRANS. ROY. SOC. EDIN., VOL. XLVIII., PART III. (NO, 25). 89 
