222 
sented. The young pod, according to Watson’s description, is 
‘stipitate glabrate, linear, acuminate, compressed, the sutures 
thick and nerve-like,” a set of characters that might be applied to 
almost any Cassia. The two detached pods on a sheet in the De- 
partment of Agriculture Herbarium, collected by Wheeler in 
Nevada, apparently do not belong to the genus at all, but as Mr. 
Coville has remarked, seem to be those of Acacia Greggit. 
Until this odd species can be more carefully examined, it has 
been thought best not to include it in the regular sequence. 
While belonging to the subgenus Sezma, it is not referable to any 
of Bentham’s sections. 
Steps toward a Revision of the linosporous Species of North American 
graminicolous Hypocreacee. | 
By Gero. F. ATKINSON. 
Certain of the linosporous species of Hypocreacee present such 
a similarity in the character of the asci and spores that some Uy 
certainty seems to exist regarding the generic relationship. This ry 
has led to the incorporation of some species under the same gene 
without regard to certain well-defined and important differential | 
characters of the stroma. In the study of several species cole ; 
lected in Alabama my attention has been called to these peculiar 
ities, and this step toward a revision of the species is offered here ~ 
as a contribution upon the subject. 
The genus Epichloé Fr. is here restricted to those charactets 
the stroma of which £. ¢yphina is the type. ‘ 
The genus Hyfocrella was erected by Saccardo to include oe 
cies with quite a variety of stroma but which in contradistinction 
from Zpichloé was disciform, and did not surround the host, oF sub- 4 
stratum, as E. typhina does. The new genus Dothichloé 1s hey 
proposed for those species of Hypocrella which possess a dothidia- ; 
ceous stroma and are represented by forms like Epichloe Hypory , 
fon Pk. : oe 
Hypocrea tubcriformis B. & Rav. with linear spores is made ee 
type of another genus, for which the name Echinodothis 'S ele : 
posed. ae 
ob 
