482 
Journal of Agricultural Research 
Vol. XXVI, No. to 
What seems to be a phagocytosis occurs in the infected worms. The 
presence of this phenomenon is suggested by the occurrence of groups, 
here and there in the blood spaces, of closely packed cells, which are not 
unlike those observed in infections in other insects. At summer tempera¬ 
ture at least the protection received in consequence of this and what 
other protective phenomena the larva may possess is, however, wholly 
inadequate to preserve the life of worms inoculated with Bacillus 
sphingidis by the puncture method. 
PATHOGENESIS 
Homworms in all instars are very susceptible to infection with pure 
cultures of Bacillus sphingidis when the puncture method of inoculation 
is followed. Larvae in the fifth instar, this stage being used most often 
for experimental purposes, become infected and die in about 100 per cent 
of the inoculations. When the feeding method is employed, however, 
a relatively small percentage of the worms die. 
The period from the inoculation of a homworm by puncture to its 
death varies considerably, depending chiefly on temperature. During 
the warmer days of summer, death takes place in one day or less, while 
this period is more often extended to two days or more during the cooler 
weather. At incubator temperature it may be less than twelve hours. 
The susceptibility of the larvae of the silkworm (Bombyx mori L.) to 
experimental infection with Bacillus sphingidis is about equal to that of 
homworms (PI. 1, F). The same is found to be true of the larvae of the 
catalpa moth, Ceratomia catalpae Bdv. (pi. 1, E). Cutworms inocu¬ 
lated died in abou tthree days (pi. 1, C). Grasshoppers also are sus¬ 
ceptible to experimental infection (pi. 1, D). All of these insects ex¬ 
cept the homworms and silkworms were tested by the puncture method 
alone. No species was found to be immune to puncture inoculations. 
Silkworms were inoculated each year from 1918 to 1921, inclusive, with 
a culture of Bacillus sphingidis, isolated in 1917 and kept on agar at room 
temperature, and no evidence of any change in virulence was observed. 
By puncture inoculation the culture produced septicemia and death in 
homworms as readily in 1921 as in 1917, when it was first isolated. The 
vims direct from the decaying tissues of recently dead worms seems to 
kill in slightly less time than do isolated cultures of the organism. 
A rabbit inoculated intravenously with pure cultures of the bacillus 
showed on the following day a tendency to anorexia, but soon recovered 
and for a month thereafter no further symptoms were noted. An autopsy 
on the etherized animal showed no lesions of note. 
COMPARISON OF BACIEEUS SPHINGIDIS AND B. ACRIDIORUM 
One of the specially interesting facts brought out by the study of the 
diseases of insects is that among them there is a large group in which a true 
septicemia occurs, the infecting organisms being in many respects similar. 
They are actively motile, gas-producing, and nonsporulating short bac- 
cilli which are often coccoid in form when grown on solid media. In the 
literature these have frequently been referred to as coccobacilli. Be¬ 
ginning about 1911, the study of this group was given a marked impetus 
by the work of the French investigator d’Herelle (3) on an infection in 
grasshoppers. D’Herelle encountered and described as the cause of the 
