MODE OF ACTION OF BACTERICIDAL SERA. 121 



The same phenomenon was noticed by Leclainche and Morel l 

 in their work on the bacillus of malignant oedema, and these authors 

 had similar experiences with erysipelas of swine and with sympto- 

 matic anthrax. As a result of this they concluded that there was 

 a "dosis optima neutralisans " of the immune serum. 



Since we encountered the same phenomenon in bactericidal 

 test-tube experiments it seemed advisable to undertake a study of 

 these occurrences, especially because the question seemed to offer 

 points of vantage important both theoretically and practically. 

 None of the authors above mentioned has furnished an adequate 

 explanation of the phenomenon. 



In our experiments the bactericidal action was determined in two 

 ways, namely, with the aid of the bioscopic method previously 

 described by us, 2 and by means of plate countings. The methods 

 gave identical results even in parallel series. In order, therefore, 

 to facilitate looking over the results we shall here give only the 

 results obtained by the counting method. 



The method of procedure was generally as follows: Vsooo cc - f a 

 one-day bouillon culture of the bacterium in question was put into 

 each of a series of test-tubes. To this were added varying amounts 

 of immune serum inactivated at 56 C. and equal amounts of the 

 complementing active serum; or in another series, equal amounts 

 of immune serum and varying amounts of the complementing serum. 

 It was so arranged that all the tubes contained equal emounts of 

 fluid, usually 2.5 cc. Dilutions were made with 0.85% salt solution. 

 Furthermore three drops of bouillon were added to each tube, for 

 we had convinced ourselves that this assured a good growth in the 

 control tubes. Numerous control tests were necessary nevertheless, 

 even if only to test the sterility of the sera employed. The specimens 

 were kept at 37 C. for three hours and then plated on agar, using 

 five drops from pipettes of uniform size for each plate. The results 

 w r ere noted by comparison and estimation, somewhat after the 

 following scheme: 0, isolated, hundreds, thousands, infinite number. 



Omitting the very extensive preliminary tests the following 

 example is given to show the phenomenon studied by us. The 

 immune serum employed was obtained from a rabbit by treatment 



1 Leclainche and Morel, La Serotherapie de la septicemie gangreneuse, Annal.. 

 de 1'Inst. Pasteur, 1901, No. 1. 



2 Mi'mch. med. Wochenschr. , 1900, No. 37. 



