37 
‘the rest of the body to hang limp. The illustration (Fig. 18, 
Plate 1) shows these caterpillars two days after inoculation 
still clinging to the leaf on which they fed. 
On Aug. 25 a dilution-culture was made from the body-juices 
and intestines. Six species of bacteria were found,and as it was 
impossible to say which of the six produced the disease, the 
following experiment was made: eight cages were prepared 
with food, and five apparently healthy caterpillars placed in 
each. The caterpillars in each of the six cages were infected 
with one of the species of bacteria obtained from the dilution- 
culture, in the following manner: an agar-culture was stirred up 
with a sterilized needle and a little distilled water was added; 
the caterpillars were then each dropped into the tube and 
shaken; they were then placed on the food in the cage. One 
cage was left uninfected as a check, and the caterpillars in the 
remaining cages were infected with material obtained from a 
dead caterpillar. After four days the greatest number of deaths 
were seen to have occurred among those caterpillars directly 
infected from the dead ones, and next to these were those in a 
cage infected with a species of bacteria which grew in plate- 
cultures in a large, lobulated form. The caterpillars in the 
other cages died as well, and showed the disease, (it now being 
impossible to obtain perfectly healthy material here at the sta- 
tion), but not quite as promptly or in as great numbers. The 
Species which produced the lobulated colonies seemed, there- 
fore, the one which was desired, so on Sep. 3 two more cages 
were prepared, each containing fifteen caterpillars. The con- 
tents of a tube containing a bouilion-culture of the suspected 
form was poured over the larve in one. After three days ten 
were dead in the infected cage, and in the check only five. In 
four more days all were dead in both cages. 
The disease had by this time become so widely distributed 
that it was impossible to obtain healthy caterpillers on which to 
experiment. It was thus impossible to say which caterpillar 
died from infection and which from having been accidentally 
infected by handling, or before capture. 
An attempt was made on September 11, to render the des- 
truction by the disease more complete by more evenly and 
thoroughly spreading the germs in the field. For this purpose 
about one hundred diseased caterpillars were collected and 
