190 REPORT 4, UNITED STATES ENTOMOLOGICAL COMMISSION. 
repeated occasions without being treated to beer inash. An incident 
connected with these experiments which we made is, however, well 
worthy of being mentioned ; because it shows how very easily single 
experiments may lead to false hopes and conclusions. A certain pro- 
portion of the last-named larvse— the proportion differing in the differ- 
ent lots treated — perished before or while transforming to the chrysalis 
state. They became flaccid and discolored, and after death were little 
more than a bag of black putrescent liquid. We should have at once 
concluded that the yeast remedy was a success had we not experienced 
the very same kind of mortality in previous rearing of this larva, and 
had we not, upon returning to the field from which the larvae in ques- 
tion were obtained, found a large proportion similarly dying there. 
Though from this experience we had little faith in the value of the 
proposed remedy as against the Cotton Worm, we nevertheless took 
pains to have it tested in 1870 both by Professor Smith in Alabama 
and Professor Willet in Georgia, and in 1880 by Messrs. Bailey, Hub- 
bard, Jones, Schwarz, and Stelle in various parts of the cotton belt. 
The experiments were made in each instance during the latter part of 
the season, when the vitality of the worms was already considerably 
lowered, a condition which, in our experience with fungus diseases in 
insects, was eminently favorable for satisfactory trial. But in spite of 
these favorable conditions the uniform result of all the experiments 
showed that there was no insecticide virtue in the ferment. For this 
reason we merely add it to the other substances just enumerated which 
proved to have no insecticide property. 
