A GENERAL RESUME OF IMMUNITY. 505 



chloroform very well. Since the bacteria then are passive agents, 

 red blood cells may be similarly considered. I had further noted 

 (and this fact was likewise mentioned at about the same time by 

 Gruber and Durham), that normal sera (and particularly certain of 

 them such as horse serum), frequently show an agglutinating prop- 

 erty for various bacteria (for example, V. cholerae, B. typhosus, 

 B. tetani). This constitutes a second relation between bacteria and 

 red blood cells, inasmuch as the latter are also frequently aggluti- 

 nated by normal sera of alien species. It then occurred to me, 

 naturally enough, that if immunization against bacteria increases 

 the agglutinating property toward a given organism to a consider- 

 able extent over that in the normal animal, we might hope to obtain 

 a similar result with red blood cells. By immunizing an animal of 

 species A with the blood of species B, we should obtain in animal A, 

 a powerful agglutinin for the corpuscles of B. With this purpose in 

 mind I vaccinated guinea-pigs against rabbit blood. When I saw 

 that specific hemolysis also occurred under these conditions, the 

 analogy of which with bacteriolysis was most obvious, I had simply 

 to repeat the experiments that I had already performed with cholera 

 serum and cholera vibrios with hemolytic sera and corpuscles. The 

 law of two substances was completely verified, and its significance 

 and importance became more evident. 



My researches on "lacto serum," that is to say an immune serum 

 which precipitates milk, as well as the ideas which I obtained con- 

 cerning, first, the mechanism of agglutination, and later, the nature 

 of the reactions between antibodies and antigens, which, according 

 to my idea, are probably to be classed in a category of adsorption 

 phenomena, also began quite independently of any abstract concep- 

 tion. On examining test tubes which contained bacteria aggluti- 

 nated by sera, I had been struck with the resemblance between 

 their appearance and the flocking out of chemical precipitates, par- 

 ticularly of the insoluble salts coming from a double decomposition. 

 It is a current notion that flocculation is favored by electrolytes, and 

 the influence of the latter had been particularly studied in respect to 

 the sedimentation of clay and the agglutination of emulsions of mas- 

 tic. A study as to whether the agglutination of bacteria also neces- 

 sitated the presence of salts was obvious, and as we know, such a 

 necessity was experimentally proved. It was found that bacteria 



