ANTHRACNOSE OF CUCURBITS. 35 
noted within five to seven days after-each rain. A new ‘‘crop” of 
young lesions of uniform age, many within the old infected area in 
the row, many out around its periphery, appeared after each period 
of rainy weather. As these centers thus increased in size they ex- 
tended not only along the row on either side but also across into 
neighboring rows and up hill as well as down. On the other hand, 
after a period of dry weather very little new infection was to be 
found. While this type of gradual spread of the disease may occur 
during light rains, it is, of course, more marked after heavy rains. 
Considerable observational evidence relative to this question has 
been accumulated. In Princeton in 1915 abundant incipient infec- 
tion was noted about old centers on July 26, probably as a result of 
the rains of July 17, 19, and 21. By August 9 and 10 a second crop 
of incipient lesions appeared, quite evidently as a result of the rains 
of August 1 to 5. 
In 1916 an examination of the anthracnose centers in field 2 after 
a period of 10 days of dry weather revealed little or no incipient 
infection on new growth and only a few incipient lesions on old leaves. 
The latter may well be attributed to delayed germination or the 
penetration of spores actually disseminated during the earlier rains. 
Observations made in this field on August 8, 11, and 14 revealed 
abundant new infection as a result of the rains of August 3, 4, and 5. 
It is commonly held that this type of spatter infection takes place 
by the actual spattering of the spores from diseased to healthy leaves. 
The agency of the soil as an intervening depository has not been 
emphasized. It seems quite likely that much of this spatter infec- 
tion is accomplished by spores first washed to the soil in great num- 
bers and thence splashed on to neighboring leaves. 
To determine the validity of this hypothesis, a test was made of 
the soil under diseased leaves in field 2 five days after a light rain 
of August 10, 1916. Two grams of the soil was shaken up in 99 
c.c. of sterile water and 1 c.c. of this suspension was transferred to a 
second 99 c. c. of sterile water. From the latter dilution 1 c. ¢. and 
fractions thereof were taken as inoculum for poured plates in water 
agar plus 2 per cent dextrose. Two soil samples were thus tested 
with five plates each. No colonies of the anthracnose fungus ap- 
peared. | 
On August 26, while the soil was still wet from a rain of 0.31 inch, 
two more soil samples were collected under diseased foliage and in 
a small drainage channel in field 2. Dilution plates were poured as 
above, four from each sample. In each set of plates six colonies 
of this fungus were identified. This indicated, according to the 
dilutions used, a spore content of 15,937 per gram of wet soil in 
one sample and of 60,000 per gram in the other. 
