Ik 



Paillot (1933) has found that the majority of cocco- 

 bacilli isolated from insects change their form more or 

 less according to the insect into which they are inoculated. 

 The "bacteria may undergo such minor changes as a slight 

 elongation of the cell or the changes may "be of a more 

 striking nature. Such variations have "been observed with 

 Bacterium pieris liquefaciens alpha , Bacterium melolonthae 

 liquefaciens gamma , and Bacterium lymantricola adiposus . 

 For instance Bacterium pieris liquefaciens alpha is in the 

 form of coccohacilli in the "blood of the larvae of the 

 cabbage butterly, Pieris brassicae . In the "blood of the 

 larvae of Vanessa urticae , there is no appreciable differ- 

 ence. In the larvae of Vanessa polychloros and Euproctis 

 chrysorrhoea the cells are considerably longer than in 

 the first two species. In the "blood of Lymantria dispar , 

 however, the elongation is so great that the "bacteria lose 

 all aspects of coccobacilli and are transformed into 

 sinuous filaments which may attain the length of ^0 or 50 

 microns. When inoculated "back into the general cavity of 

 the Pieris "brass icae larvae the cells return to their 

 normal form. 



Bacterium melolonthae liquefaciens gamma , which usually 

 appears in the form of a cocco"bacillus, "becomes elongated 

 and thicker when inoculated into the larvae of the gypsy 

 moth, Porthetria dispar . As the infection advances, a 

 certain number of the "bacteria show one or two median or 

 polar swellings. According to Paillot, these swellings 

 later "become detached from the "bacterial elongations and 

 float freely in the "blood though they are not actively 

 motile like the "bacilli from which they originated. These 

 forms resemble small true cells, possessing a central 

 portion which may "be taken for a true nucleus since it 

 takes nuclear stains. When the insect dies these forms 

 gradually disappear. 



Similar "growth forms" have "been observed in certain 

 strains of Bacterium lymantricola adiposus inoculated into 

 Porthetria dispar . In this case they originate from an 

 enlargement of the central portion of the "bacterium. 

 These enlargements continue to elongate and often give rise 

 to secondary elongations, the whole thing resembling a 

 kind of mycelium. In old strains the "bacteria and "growth 

 forms" may "be seen to possess a large central vacuole 

 which disappears when the elongations re -absorb themselves 

 into new rounded forms which float freely in the "blood. 

 Similar forms of this "bacterium have "been described as 

 developing in the "blood of the larvae of Agrotis segetum . 



