PRODUCTION OF ANTIPNEUMOCOCCIC SERUM 525 



killed avirulent pneumococci, Yoshioka (1925) 1565 found that only 

 weak serums were produced. During investigations on the smooth 

 and rough forms of pneumococci, Griffith (1923) 560 concluded that 

 only smooth strains should be employed for the production of se- 

 rum. Using Type III pneumococci, Levy-Bruhl (1927) 804 obtained 

 agglutinating and precipitating serums readily from horses, but 

 protective antibody was obtained only when recourse was had to 

 the injection of exceptionally virulent strains. Meyer and Sukneff 

 (1928) 898 reported that the administration of sodium taurocholate 

 solutions of virulent strains yielded serum with protective and 

 some cross-protective powers, but the injection of solutions of 

 avirulent strains evoked no protective antibodies. The observations 

 are consistent with the earlier reports of Avery and Neill 59 and of 

 Avery and Heidelberger 49 on the antigenic action of constituents 

 of the pneumococcal cell. In studies on the response of rabbits to 

 inhalation of pneumococci, Stillman (1930) 1334 concluded that the 

 type-specific response of the animals, as evidenced by the appear- 

 ance of specific protective antibodies and agglutinins in the serum, 

 varies in direct proportion to the virulence of the culture used. 

 There seems to be little dissent, therefore, from the opinion that 

 only cultures of maximal virulence furnish satisfactory antigenic 

 material for the production of potent antipneumococcic serum. In 

 the opinion of the authors of the present volume, the best results 

 are obtained by choosing cultures of exalted virulence and strong 

 antigenicity and by maintaining high virulence by passing the 

 strains through mice daily for at least five days a week. The vari- 

 ous methods of maintaining the virulence of cultures in routine use 

 have been discussed in Chapter VI. 



In a study of the immunizing properties of the four groups of 

 pneumococci classified by Dochez and Gillespie, 322 Dochez and 

 Avery 318 found the order of virulence of the types of pneumococci 

 for human beings to be III, II, I, and Group IV, and stated that 

 "the degree of protective power developed in the sera of animals 

 immunized against members of these groups varies inversely with 



