326 BOTANICAL GAZETTE [ocropeR 
shown in figs. 43 and gg. The hyphal coil may be traced for a 
longer time here than in the other genus. In figs. 45 and 47 it 
is shown somewhat straightened out and several times septate. 
For want of time the study of this form was not carried through 
to the end, but enough observations were made to furnish con- 
vincing proof that the course of development is essentially the 
Same as in Ceratostoma. The process is somewhat more com- 
plicated than in Teichospora, and may be summarized as fol- 
lows: 
1. The spores upon. germination send out polynucleated 
mycelial threads which become septate, branch, and form ci 
cular colonies. 
2. Upon the mycelium are borne short thick branches which 
become curved, or sometimes several times coiled, and function 
as archicarps. | 
3. Near these archicarps are usually found long slender 
branches, the antheridia. 
4. The antheridia intertwine with the archicarp, their tips 
meet and fuse. 
5. The archicarps in some cases appear to develop without 
fertilization. 
6. The archicarp by growth and division furnishes the cells 
which make up the interior of the perithecium. 
7. From certain of these cells of the interior the asci arise. 
8. In each young ascus there is a single primary nucleus. 
9. The primary nucleus divides karyokinetically and the 
daughter nuclei in the same manner, to furnish a nucleus for each 
spore, 
10. Nuclear division probably continues within the spore 
after the formation of the spore wall. Ae 
11. The wall of the perithecium is formed from surrounding 
filaments. 
The evidence furnished by the foregoing investigations es 
© corroborate the theory of De Bary that marked analogies ee : 
between the higher fungi and the Floridee. In the Pyren0™ 
