290 BOTANICAL GAZETTE ——_ focrozrr 
(jig. 16), a portion of it lying in contact with the protoplast stains 
differently and marks the beginning of the two coats. This layer, 
almost as soon as it can be detected at all, shows a dual nature: It 
appears in section as'a shaded band not sharply marked off from 
the undifferentiated portion of the spore membrane. Its outer 
region stains more vividly than the inner, perhaps because it is 
denser. Gradually a difference in structure becomes apparent, 
which I have attémpted to show in fig. 17. 
The spores are so small, stain so intensely during this critical 
stage, and the changes in the métamorphosing spore membrane 
occur with’such rapidity that it has been with difficulty that I could 
demonstrate to my own satisfaction that one coat was not formed 
after the other, but both simultaneously. Frrrmne thinks that the 
endospore is formed after the first and arises de novo as a new forma- 
tion of the protoplast. The spore is much smaller, and the changes 
much more rapid in this form; consequently they are harder to follow 
than in S. rupestris. The outer denser region with its roughened 
surface becomes the exospore, the inner the endospore. The pro- 
toplast becomes a vesicle by the formation of one large central vacuole. 
It is at this time when there ate two coats of distinctly different struc 
ture that the peculiar phenomena occur which led FitTINé to draw 
the conclusion that the protoplast is not concerned in the conspicuous 
_ growth of the spore coats. ie 
As the spore grows larger, the outer coat increases in diameter 
and thickness much more rapidly than the inner, producing a strat 
in the viscid mass of which the two are concentric layers. FITTIN® 
thinks this results normally in a complete separation of the two, 
thus developing between them a comparatively large empty se 
In a like fashion, due to the fact that the protoplast is quiescent a? 
the endospore growing, a second “empty space” is formed a 
them, except at one point, thus severing all organic connection ok 
protoplast with the developing coats except at this point of pee 
There is no doubt that the majority of preparations have precisely 
appearance. Moreover, the living spore examined in a no 
: : ation of the coals: 
solution seems to confirm Firrrnc’s view of a separation 0! ne 
and on this fact he lays great stress. The spaces, pete an 
are afterward partially filled with a solution which has entered the sp 
