34 BULLETIN 1727, U. S. DEPARTMENT OF AGRICULTURE. 
during the early part of July and a heavy rain on July 22, followed 
_by temperatures favorable to this fungus, should have induced an 
early development of anthracnose. Although there were frequent 
rains during the remainder of the season, all daily mean temperatures 
remained well below 75° F., the optimum for the anthracnose fungus. 
The mean monthly temperature for August was 67° F., and for the 
first week of September, 63° F. 
Anthracnose was first found on July 27 in field 1 and consisted at 
that time of eight infected plants scattered through four rows. On 
August 1 there was considerable anthracnose in 12 rows in the north- 
west corner of the field, in 3 rows in the southeast corner, and in a 
detached block at the southwest corner. This infection did not 
appear in typical original centers, but rather as scattered and com- 
paratively recently infected plants. On August 13 anthracnose was 
found in a total of 28 rows scattered through the field, but was in no 
case serious. Inspections made August 25 and September 8 showed 
that no extensive spread had occurred. This can not be attributed 
to lack of rain and is probably due to the low daily mean tomipet 
tures prevailing after the end of July, as noted above. 
In the other fields no anthracnose was found except in the Tsom 
field, and there not until August 13, when old lesions were noted on 
the alt leaves of two adjacent plants On September 8 an additional 
original center from which very little spread had occurred was found. 
Why these were so late in developing is not understood. 
LOCAL DISSEMINATION BY WATER AND OTHER AGENCIES. 
Consideration of the subject of dissemination falls very naturally 
into two general categories: One, the mode of spread of the disease 
in the field during the growing season; the other, the manner in 
which the disease is first introduced into new fields (21). The former 
may conveniently be considered as local dissemination. 
SPATTERING ACTION OF RAIN. 
That water is essential to the separation and dispersal of the spores 
of this type of fungus is generally recognized. Since the spores tend 
to remain in masses adherent to the acervuli until their matrix is 
dissolved by water, the importance of the latter as a prerequisite for 
spore dispersal is quite evident. But the mechanical action of rain 
is more important. 
The appearance of so-called centers in fields has been mentioned. 
The gradual enlargement of these foci of infection and the spread 
therefrom is the phase with which we are now concerned. Some of 
the experimental work on water dissemination has been previously — 
reported (21). 
Observations made in the Madison fields in 1916 showed that an 
enlargement of old centers of infection in all directions was to be 
—-e 4 
