REACTIVITY OF MALIGNANT NEOPLASMS 249 



mice. The subcutaneous and intraperitoneal inoculations of small 

 doses of this microorganism into normal mice gave septicemia 

 and invariably proved fatal. Eisen then injected tiie microorgan- 

 ism into Ehrlich carcinoma in a dose of 0.1 c.c. of a suspension 

 (an agar slant washed with 5 c.c. of saline) . Following this iir- 

 jection, the tumor appeared reddish or reddish-gray. Histologi- 

 cally, there were hyperemia and marked regressi\'e changes. In 

 tmtreated timiors regressive changes were limited to a central 

 portion ^vell demarcated from normal Avide peripheral zone of 

 active gro^vth. In treated tinnors there ^vas a decided increase 

 in the amount of regressive changes — centrally and peripherally 

 about the blood vessels. These changes involved the entire tumor. 

 The nuclei were pycnotic, the cells showing various stages of 

 degeneration. The mitotic hgures tended to disappear. There 

 was no demarcation of the necrotic tissue from the healthy tissue. 

 There was no change in the viridence of the microorganism after 

 se\en passages through timiors. 



RELATION OF MORTALITY TO HF.MORRHAGIC NECROSIS AND REGRES- 

 SION OF TUMORS ELICITED BY ACTIVE PRINCIPLES OF 

 THE PHENOMENON 



As pointed out before, the opinion has been voiced by Duran- 

 Reynals and Apitz that there exists a definite relation between 

 the destruction of the tumor and the mortality rate. My own 

 observations, however, suggested that the apparent susceptibility 

 of mice to the lethal effect of the active principles may be due to 

 a sensitization following secondary bacterial infections of the 

 tumors. It "was also observed on some occasions that certain active 

 bacterial principles may lose their lethal potency on prolonged 

 storage and yet retain the tumor-destroying j^roperty. The prob- 

 lem received special consideration in experiments comparing 

 the lethal effect of active preparations upon normal and timior- 

 bearing mice, and in studies on the relation of the tumor-destroy- 

 ing factors of filtrates to their lethal potency. 



The phenomenon-producing potency of the bacterial filtrates 

 was titrated in rabbits shortly before use. The exj^eriments were 

 carried out in mice bearing twelve day old timiors. The injections 

 were given either intravenously or intraperitoneally. 



Meningococcus "agar washings" filtrates were found to contain 

 tinnor-destructive principles of high potency. Prompt hemor- 

 rhages leading to complete regression in most of the sin"\'iving 



