June, *18] BURGER AND SWAIN: WALNUT APHIS, FUNGUS 285 
were infected with the fungus. In the following two weeks these 
increased greatly, until on August 16 there were present about 132 
aphids per leaf. The percentage of infection had also increased, but 
not so rapidly, there being about 4| per cent of the aphids killed by the 
fungus. In the next two weeks the number of aphids per leaf had 
dropped to but slightly over 35, of which some 37 per cent were in¬ 
jected. Two weeks later, on September 12, only about 13 aphids per 
leaf remained, and over 93 per cent of these were infected with the 
fungus, leaving an average of a little less than one aphid per leaf 
alive. By September 20 the number had increased to almost 9 per 
leaf, but again the percentage of infection was high, being almost 
95 per cent, with the result that the aphids were practically eradicated. 
This almost complete eradication is accounted for by the action of 
the fungus together with other natural agencies. Which of these 
was the most important the writers cannot state, but the presence of 
the mummified bodies of aphids infected by the fungus permits the in¬ 
terpretation that this parasite was largely responsible for this great 
mortality. 
More convincing results are obtained from a study of Table 5. This 
is a record of counts made on the tree under tent 2, from August 1 to 
September 20. During observations covering a period of four months 
(June to September) no insect enemies of the walnut aphis were ever 
found in this tent. No specimens of Coccinellidoe, Syrphidce or Chryso - 
pidoe were seen, yet results parallel to those observed in the grove 
were found. On August 1 an average of 40 aphids per leaf was present 
and none were infected by the fungus. This tree had been sprayed in 
the spring before the buds opened with lime sulfur, with the result that 
no aphids appeared until very late. In fact it was at this time that 
the first real infestation was noticed. These probably were the suc¬ 
ceeding generations of some few that gained entrance accidentally from 
time to time during the summer. By August 9 they had increased, as 
might be expected, to about 50 per leaf, with a small percentage in¬ 
fected with the fungus. A week later the number per leaf was found 
to be about 92, with about one dead as a result of fungus attack. 
Approximately 102 aphids per leaf were present on August 30, but by 
this time the percentage of infection had increased to 25.71 per cent, 
with the result that only 76 per leaf remained alive. These increased 
to about 90 aphids per leaf in the next two weeks, the number infected 
increasing as well, although in greater proportion. Almost 85 per 
cent were so infected, leaving but 13.5 per leaf alive. On September 
20, when the last observation was made, there were found to be 70.5 
aphids per leaf on an average, of which 95.6 per cent were infected by 
the fungus, leaving only slightly more than 3 aphids per leaf alive. 
