Notes . 
577 
the Diatomeae the formation of auxospores follows only as a conse- 
quence of that organization of the cell which has been described after 
a number of divisions which may be almost mathematically defined, 
while the external conditions play no definite part in the process. 
But, meanwhile, we ought not to forget that this theory is in much 
need of further confirmation. The main point in the whole question 
is whether the cell- wall of both cells after division really does undergo 
no increase by growth in length to even a very small degree. Pfitzer 
himself has noted that such growth does occur in certain species, 
though this would be of no account for making up the loss of size 
which accompanies division. It ought to be distinctly proved, by 
direct and exact measurements, whether a growth in length takes 
place or not ; above all, we ought to know precisely what influence 
the various external conditions exert upon the life of the Diatomeae. 
It may possibly be that under certain circumstances growth takes 
place, under others not. The fact, brought forward by Miquel and 
others, that it is by no means always the smallest cells which form 
auxospores, but also those of middle size, deserves consideration. 
Still more important are the statements made by Karsten, that Melosira 
nummuloides can be brought to the formation of auxospores simply by 
change of water, that in Achnanthes longipes the impending conjugation 
does not take place, but is replaced by vegetative growth when the 
cells are exposed to a cool temperature. If we assume that definite 
external conditions induce growth in length during division, others 
encourage auxospore-formation, the earlier observations on the small- 
ness of the cells which form auxospores may thus be explained. 
In many Thallophytes the rule holds that in the stage of preparation 
for the higher fruit-form growth diminishes, or ceases, but division 
is still continued, so that, for instance, in the Desmidiaceae, Spirogyra , 
and Chlamydomonas , it is always the smallest cells which conjugate. 
This may also be the case on the formation of the auxospores of 
the Diatomeae, and the smallness of the cells would then be less 
the cause of auxospore-formation than the result of those external 
conditions which occasion this process. I bring this possibility 
forward in order to draw attention to the pressing need for accurately 
defining the conditions of the events in the life of the Diatomeae 
by the help of pure cultures, and by the use of physiological methods. 
Whichever way the decision may fall, the life-history of the Diatomeae 
gives no explanation of the wholly different alternation of generations 
