396 BOTANICAL GAZETTE [DECEMBER 
solution as used by Mottier (’97). A number of stains were 
used, but the best general results were obtained by the triple 
method of aniline-safranin, gentian-violet, and orange G. The 
stains were varied both in kind and length of time of use, 
in accordance with the structures or parts of the cell which were 
to be brought out. 
The cells from the more rapidly growing stalk of the arche- 
goniophore are in general cubical, and a longitudinal section of 
the stalk will almost invariably result in meridional sections of 
the spindles. The resting nuclei are round and contain large 
conspicuous nucleoli. The nucleolus seems to lie in the cavity 
or vacuole. A careful examination reveals also a very delicate 
network of linin threads. These are clearly seen only in success- 
ful preparations when sufficiently magnified. 
In the granular cytoplasm becoming vacuolate in older cells 
are situated many oval or rounded chloroplasts, together with 
metaplasmic inclusions which stain variously. The chloroplasts 
and metaplasmic inclusions generally collect about the poles of 
the nuclear spindle, thus obscuring the finer details and render- 
ing observation difficult. 
The first noticeable evidence of nuclear division is the pulling 
out of the nuclear membrane, generally in the direction of the 
long axis of the cell, into two points which are opposite each 
other and around which the chloroplasts tend to collect. A 
careful examination reveals a minute body at each point 
from which extend conspicuous radiations. These bodies with 
their radiations are the centrospheres. They seem undoubtedly 
to exert a great attractive force from the manner in which 
certain of the cell contents are drawn to them (jigs. 2, 4): The 
way in which the nuclear membrane is often drawn out into very 
long points would seem to indicate further that the centrosphere 
exerts a pulling force upon it. The exact time of the appeat 
ance of the centrosome cannot be stated with certainty, because 
in the absence of the radiations it is impossible to distinguish the 
centrosome from other bodies appearing quite like it. From 
what is known of their behavior in other plants it is reasonable to 
