114 
as also the cross walls where they meet the outer wall. In making this 
statement he refers to the episporium and endosporium as constituting two 
layers of one wall. In some instances the endosporium is clearly seen to 
extend between, and to separate, the protoplasts (Fig. 11, b), while in other 
cases the protoplasts appear to touch each other (Fig. 11, a), yet when the 
conidial contents are pushed from a crushed conidium there is always a 
line, though sometimes it is very thin, separating the protoplasts. Since 
the protoplasts are distinct from each other, and are thus separated by the 
endosporium, it seems justifiable to assume that this second cell-wall forms 
the septa, sometimes obvious though very thin, between the protoplasts. 
Treated with concentrated sulfuric acid the conidial endosporium dissolves 
rapidly, and by the generated pressure the episporium is ruptured, invari- 
ably at the basal end first, this often opening trap-door-like, though fre- 
quently the pressure is sufficient to tear the wall of the conidium open 
throughout its length. With the solution of the endosporium the proto- 
plasts issue from the case of the conidium and appear to be unattached. 
The individual protoplasts vary much in shape, sometimes being nearly 
spherical; in other cases nearly cubical. - Each protoplast is surrounded 
by a differentiated layer which in some cases is so clear, distinct, and thick 
as to appear to be a third wall (Fig. 11). Perhaps it is. Under gentian 
violet and many other aniline stains, while the protoplast takes a strong 
stain this layer refuses to stain. Microtome cross-sections and longitudi- 
nal sections of conidia verify the foregoing conclusions (Fig. 11, f). In 
cross-sections (Fig. 11, g), with Fleming’s triple stain the protoplast stains 
as usual, but the second cell-wall refuses to stain; under Bismarck brown it 
takes a very faint stain. Under action of aniline blue, iodine, fuchsin, mal- 
achite green, Pianese, or chlor-zinc-iodine it remains unstained. In longi- 
tudinal sections cut so thin that two sides of the conidium have been cut 
away, mature conidia show no continuity of the protoplasts (Fig. 11, d). 
When plasmolized the protoplasts all shrink and lie quite separate from each 
other, and it is in such condition that the appearance of a third conidial 
wall is most evident (Fig. 11, a, 6). Previous to plasmolysis the proto- 
plasts are frequently seen to touch each other on the median longitudinal 
axis of the conidium, and a very faint line (plane) is observable extending 
across the conidium (Fig. 11, a, b). This probably represents the true 
septum, following nuclear division. The protoplast wall bears a small 
dot-like thickening (Fig. 11, e) adjacent to its sister protoplast, which may 
also be residual evidence of nuclear mitosis. 
The characters as here given for H. No. 1 are found also in such re- 
