184 



In the systematic testing of agglutination all that is required 

 therefore, is to first put up the test in 1:50 and 1:100 dilutions 

 and then titrate those samples further which give a reaction 

 in 1:100. 



In this way agglutination tests were made from November 

 1919 — February 1920 with all the 14 sera against a considerable 

 number of strains of Pfeiffer's bacillus, about 15 strains at a 

 time being tested against all the sera. As far as possible enough 

 suspension of each culture was prepared for the further titra- 

 tion of the positive reactions. 



The collection of reactions reported in the table below is there- 

 fore in reality composed of a large number of experiments each 

 consisting of about 200 single reactions of 2 tubes each, and 'dso 

 of the further titration of the positive reactions. The principal 

 result of the whole investigation, the great differences in aggluti- 

 nation between the individual strains, cannot be due to variations 

 in the experimental conditions from day to day, because they are 

 also fully developed in tests performed simultaneously. It was thought 

 unnecessary therefore to indicate which experiments were earned 

 out at the same time. The strains are arranged in the table according 

 to their origin (influenza, whooping-cough, etc.), but the experiments 

 were made in a rather different order. 



From considerations of space the strength of the reactions is 

 given by the number of dilutions giving complete agglutination. For 

 example, „2" signifies complete agglutination, -f- or + ~f"> * : 50 and 

 1 : 100, ,,2-f " complete agglutination in these two tubes and in- 

 complete agglutination, (-{-), in 1:200. In some cases partial agglu- 

 tination in the first two or three tubes is recorded; this is denoted 

 by 1 and 1-j- respectively, which is not quite correct but was ren- 

 dered necessary on account of the statistical treatment. An error 

 of real significance is hardly introduced in this way. 



In the case of each strain it is stated whether it was „typical" 

 (t) or „atypicar' (at), and whether it formed indol (-(-) or not (0). 



The homologous reactions are given in heavy type. 



Sera 



