Contagious Diseases of Insects. 321 



histolysis granules commencing to appear in the blood of 

 slightly affected larvae as early as the fourth day after infection. 

 Caterpillars thus attacked did not commence to die until the 

 sixth day, and most lived until the 15th. As in the case of the 

 silkworm jaundice with which this is compared, the bacterial 

 affection was less evident than in more rapid and pronounced 

 cases of disease, but the usual intestinal micrococci were always 

 present in varying numbers. 



The last infection experiment I had to report, began August 

 2, 1884, with the same fluid, applied to the food of the European 

 cabbage worm, was abandoned August 14 because the assistant 

 in charge was unable to account for all the larvae, some having 

 evidently been allowed to escape when the food was changed. 

 As far as carried, it tended to confirm the indications of the 

 preceding experiment, the blood of those dying up to the 7th 

 August being full of a large active Bacillus only, similar to 

 that used in the infection, and those perishing later containing 

 chiefly large micrococci together with mulberry cells and gran- 

 ules. Later the common flacker ie of the cabbage worm appeared 

 in the remnants of both the infected and check lots. 



Finally in a note on muscardine I have attributed largely 

 to this affection the disappearance of a vast host of the 

 forest tent caterpillar (Clisiocampa sylvatica] which devastated 

 the forests and orchards of a part of southern Illinois in 1883, 

 basing this conclusion upon the observed phenomena of the 

 disease appearing among them as compared with those accom- 

 panying the death of larvae of this species from the same local- 

 ities, perishing in our breeding cages the following year of 

 demonstrated muscardine. 



There now remains to me only the pleasing duty of 

 acknowledging my grateful obligations for aid in this work to 

 my first assistant, Mr. W. H. Garman, to whose faithful care 

 and unimpeachable accuracy of manipulation the larger part of 

 the bacterial cultures were due ; to Prof. T. J. Burrill, who has 

 had the kindness to examine many of my slides, giving me the 

 benefit of his extensive acquaintance with the bacteria ; and 

 to Dr. H. J. Detmers, now of the State University of Ohio, to 

 whom I owe, among many other favors of this character, the 

 excellent photographs of micrococci reproduced in the plate. 



