68 Smith.—On Direct Nuclear Divisions in the 
the strands are fewer in number, although the density of each is not greatly 
lessened (Fig. 3). 
The constant upward streaming seems to cause a tension or strain 
within the semi-liquid cytoplasm. This strain is apparently transmitted 
throughout the length of the hypha, with the result that in the less dense 
portions at the base the strain is the same as in the apex. As, however, the 
cytoplasm is denser in the apex, the effect on each individual strand is not so 
great here. In the basal portions the strands are fewer, with the result that 
each is in a state of greater tension. The nuclear membranes are in close 
contact with the cytoplasm, so that the tension in the cytoplasm is trans¬ 
mitted to the nuclei, which, being of a plastic nature, respond to the strain by 
becoming elongated. In the hyphal tips where the cytoplasm is evenly 
distributed, the tension is equal on all sides, with the result that the nucleus 
is spherical. Where the cytoplasm is in a state of greater longitudinal 
strain the nuclei respond by assuming an elongated form, the extent of the 
elongation being proportional to the tension, and in some cases a length of 
6 [jl with a breadth of 2 ^ is attained. 
The central nuclein body responds in a similar way, assuming a shape 
varying from spherical to oval according to the strain on the outer mem¬ 
brane ; but the change is not so marked, and in some cases the nuclein 
retains its spherical shape even in elongated nuclei. It is probable that the 
alteration in the shape of the chromoblast is caused by the linin threads 
which connect it to the outer membrane, so that in cases where the linin 
is not well developed, or where it is c6ncentrated towards the middle, the 
spherical form is retained. 
Amitosis. 
No nuclear divisions take place in the zoosporangium, nor in the region 
just below the septum, but they are found chiefly in the region of the 
middle of the hypha, and below this. 
In every case the nuclei in the vegetative hyphae were found to divide 
by a direct division. In the oogonium, mitotic divisions have been observed 
by Hartog, Trow, Davis, and Claussen, but the writer has not discovered 
anything comparable to spindle fibres or chromosomes in the vegetative 
nuclei. 
The division begins with an elongation of the chromoblast, the mem¬ 
brane remaining intact until the end of the division. The chromatin 
granules on the membrane and the linin threads are visible throughout the 
whole process. After the chromoblast has elongated slightly the nuclear 
membrane becomes larger. A median constriction then appears in the 
chromoblast, and after a short time a similar constriction begins in the mem¬ 
brane (Fig. 6). Elongation of both parts continues throughout the whole 
process, with the result that the constrictions become gradually more 
