1898] THE DEVELOPMENT OF SOME ANTHRACNOSES 77 
elliptical, measuring 5-7 X 2.5-34. The short basidia, which 
soon become erumpent, spread over the spot and are not con- 
fined to a definite pustule. 
The fungus does not adapt itself readily to artificial culture, 
and considerable difficulty was experienced before obtaining a 
pure culture. The conidia are not easily distinguished from 
isolated yeast cells, and a fungus forming light colored masses 
on the anthracnosed spots, and producing conidia closely resem- 
bling the conidia of G. venetum, proved to be a yeast form. 
The conidia germinate readily in water or agar, but after the 
germ tube has attained a length of three or four times the length 
of the conidium, further development is very slow. In agar, 
numerous short branches are sent out from the primary germ 
tubes, which become closely septate; the cells, including the 
conidium, become somewhat swollen, so that the conidium itself 
can be distinguished with difficulty from the cells of the myce- 
lium. Separation is more or less difficult, since growth in acidi- 
fied agar is even less favorable than ina neutral medium, and 
bacterial and other foreign growth is liable to contaminate the 
cultures before the colonies can be seen in order to be separated. 
Many unsuccessful attempts were made to obtain a pure 
culture of the anthracnose of the raspberry. 
As late as November 29 fresh conidia were again obtained. 
With some of these a culture was made ina hanging-drop of 
water; dilution cultures were also made in acidified and neutral 
agar. On the following day the cultures were examined and 
Many conidia were found to be sending out slender germ tubes 
by which, in the crowded condition in the hanging-drop, the 
conidia were frequently fused; the tubes attaining three or four 
times the length of the conidium. Four days after sowing, the 
germ tubes in the Petri dish cultures had attained a length of 
24—-60u. They were divided into cells but slightly longer than 
broad, which were swollen so that the mycelium presented a 
moniliform appearance. On December 2 transfers of blocks of 
@gar containing germinating conidia were made to tubes of bean 
Stems. The growth at first seemed very unpromising. Small 
