1 ( 18 MARSHALL. 



fashion, but the cholera vibrio did not lose its capacity to become ag- 

 glutinated, although it grew readily in the new medium. This line of 

 investigation was not carried any further. 



One interesting observation was made upon a strain which had been 

 transplanted nine times through successive serum bouillon cultures and 

 which had been allowed to remain in the last flask for a long period. The 

 vibrio was plated out and its biological characteristics determined. Among 

 other distinctive features the blood-agar-hauuolysis test was tried simul- 

 taneously with this bacterium and the stock culture from which it was 

 taken for the original transplant into serum. The original stock culture 

 showed no haemolysis within twenty-four hours and only a trace in forty- 

 eight hours, while the vibrio which had been grown in beef broth, con- 

 taining anticholera horse serum, showed a marked ability to produce 

 haemolysis of goat's blood within twenty-four hours. We have here a 

 definite variant produced by artificial laboratory means, the only dif- 

 ference between the stock culture and the variant being that the variant 

 grew in the presence of serum from a horse which had been immunized 

 against cholera. The factor which produced the haunolyzing reaction 

 was probably not the specific anticholera substance of the serum but those 

 constituents of it which are close to the ones of the red blood corpuscles. 



The question of variation was then attacked in a different way by Dr. 

 Philip K. Gilman, of the Philippine Medical School, and myself. We 

 created an artificial epidemic of cholera in guinea pigs in order to observe 

 what changes the vibrio would undergo both in virulence and in other 

 respects. A guinea pig received an inocidation intraperitoneally of 0.1 

 cubic centimeter of the material taken from the ileum of a man dead of 

 cholera, the diagnosis being based upon clinical history, autopsy appear- 

 ances and laboratory examinations. The guinea pig died within twenty- 

 four hours ; 0.1 cubic centimeter of the peritoneal contents from this ani- 

 mal, diluted with salt solution, was inoculated into the peritoneal cavity 

 of a second guinea pig, and upon its death, within twenty-four hours, 

 the peritoneal exudate was used for inoculating a third. Upon its death 

 within twenty-four hours, the exudate was diluted and a series of guinea, 

 pigs was inoculated with varying amounts of this material. The mini- 

 mum dose fatal within twenty-four hours was 0.02 cubic centimeter. 

 The exudate from the guinea pig killed by this amount was diluted and 

 a fresh series inoculated. On this occasion, although as small a dose 

 as 0.0005 cubic centimeter was inoculated; the minimal fatal dose was not 

 determined. The exudate from the animal killed by the lowest fatal 

 dose was used for inoculating a sixth series and in this 0.0001 cubic 

 centimeter was the minimal fatal dose. From the exudate obtained from 

 the animal which succumbed to this cpiantity a new series was inoculated 

 and 0.00002 cubic centimeter proved fatal. An interesting result 

 followed the next inoculation. The exudate was diluted and inoculated 



