Cell Structure, Growth and Division in the Antheridia of Polytrichum etc. 159 
Similar coarse Strands, representing fiber groups, were observed in the 
root tip of the onion by Timberlake (1900), who thought that the crowd- 
ing together of the finer fibers may sometimes result in an actual fusion, 
as had previously been suggested by Guignard (1891). 
Figure 109 shows a cell containing rather conspicuous polar radiations 
— a very uncommon appearance in this as in the immediately preceding 
stages of division. 
The further development of the spindle and the formation of the 
cell plate (Figs. 115 — 120) proceed, so far as I have been able to observe, 
in exactly the same fashion in this as in the earlier mitoses. A cell wall 
is formed between the daughter cells (Figs. 123, 124), as the staining 
reactions demonstrate; but, as already noted, this wall remains for some 
time very thin. Later, of course, like the other partition walls within the 
antheridium, it swells and is finally dissolved. 
Discussion. 
The Organization of the Cells. 
It is well established that certain thallophytes (notably some phae- 
ophyceae and various ascomycetes) are characterized by a polarity of 
cell Organization which persists in each case at least during several cell 
generations, and which is related to, or evidenced by, the presence of a 
distinctly delimited central body. Such a body will be generally referred 
to in the present discussion as a centrosome, using this commonly 
accepted term without any intended implication as to the permanence 
or the homologies of the structures to which it is applied. Kecent accounts 
by Marquette (1907, 1908) of the phenomena accompanying nuclear 
division in leaf cells of Isoetes and spore mother cells of Marsilia have 
called attention to the possible possession of a persistent polar Organiza- 
tion by cells which contain nothing of the nature of a centrosome. In these 
cells, groups of starch grains (each group, in Isoetes , surrounded by a 
membrane) behave in a somewhat centrosome-like fashion, a single group 
dividing into two, which separate and place themselves at the respective 
poles of the future spindle 1 ). Similar starch groups appear, according to 
Marquette, in the spore mother cells of Equisetum, but their relation 
to cell division has not been dctermined. Marquette has found scattered 
through the literature numerous descriptions and references which, taken 
!) In the spore mother cells of Marsilia, the starch mass, without dividing, 
takes a position between the daughter nuclei. In the division of the cell follow- 
ing the second nuclear division, the starch grains are distributed in four groups 
to the respective spores. 
