A STUDY IN CHROMOSOME REDUCTION. 619 
call attention to one or two features which I consider of prime importance as character- 
ising this peculiar form of merismatic activity. 
In the first place, I would again mention the high state of nutrition that prevails 
before and during this second division. An examination of the stages represented in 
figs. 41 and 42 make it perfectly plain that an exceptionally large quantity of reserve 
food is stored in these cells. This is evident not only in the cytoplasm, which is very 
dense and granular and devoid of vacuoles, but also in the enormous thickness of the 
cell-walls (figs. 41 and 42). This latter becomes absorbed by the developing micro- 
spores before they separate from one another (figs. 43, 44, and 58). 
I believe this high state of nutrition has a direct bearing on another important 
feature of meiosis—namely, that there appears to be little or no pause between the first 
and second meiotic divisions. The merismatic activity here shown is obviously not of 
the ordinary kind. It manifests itself by two nuclear divisions that follow one another 
very rapidly. I believe it may well be considered an accelerated form of mitosis—the 
acceleration being due to the high state of nutrition. 
Other points of importance to be noted in this connection are the changes that 
occur in the chromatin in preparation for the second division. ‘This division has been 
regarded as a somatic division, and I do not doubt the accuracy of that interpretation. 
Not enough attention has been given, however, to one striking feature that distin- 
guishes it at once from an ordinary somatic mitosis. A study of the stages represented 
in figs. 39 and 40 and also figs. 55, 56, and 57, will prove quite clearly that the chromo- 
somes become alvevlised to a high degree. After this they extend out and appear as 
long thick spiremes (fig. 40); but the point of interest is that they never become so 
finely divided as in ordinary somatic mitoses, where we always find a finely divided 
reticulum stage. I have never been able to find a reticulum stage between the first 
and second meiotic divisions. Back in the very early prophase of the first division, as 
shown in fig. 1, we always find the chromatin in the finely divided condition forming 
a definite reticulum. That condition does not occur again until after the second division, 
when the microspores are formed as shown in figs. 44 and 58. 
I attach considerable significance to this fact because I believe it is further evidence 
in support of the view, already expressed, that we have here in meiosis an accelerated 
form of merismatic activity. 
THEORETICAL CONSIDERATIONS. 
As ordinarily practised, cytology concerns itself with the interpretation of artifact. 
It isa makeshift substitute for the interpretation of changes, physical and chemical, 
which are believed to occur, but which cannot be satisfactorily observed in the living 
cell under present methods. With the methods at our disposal, it is impossible to inter- 
pret the minute structure, configuration, and changes that characterise the protoplasmic 
bodies of a living cell with any degree of exactness. We are dependent upon methods 
which, at most, can yield but approximate results. The underlying principle of these 
