142 On the Specificity of Precipitins 



All observers, including myself, who have worked with different 

 blood dilutions, have noted the fact that an antiserum will act upon 

 a higher dilution of homologous than of non-homologous blood. In view 

 of the labour involved, but very few figures have been obtained by this 

 method, employed comparatively. Linossier and Lemoine (8, III. 1902) 

 thus sought to express the reactions with various dilutions, as follows : 



Antisera added in the proportion of 15 : 100 of dilutions. 



Anti-ox serum gave reactions with Ox serum diluted 1 : 5000 

 Horse 1 : 300 



Man 1 : 50 



Reactions of "apparently equal intensity" were obtained when anti- 

 human serum was added to human serum diluted 1 : 1000, and to ox 

 serum diluted 1 : 20. In another similarly conducted test, anti-human 

 serum acted on human serum diluted 1 : 2500, but only on horse serum 

 diluted 1 : 20. As the authors state, these figures have no absolute 

 value, as they naturally will vary according to the antisera used being 

 different. Strube (12, vi. 1902) testing with anti-human serum, 

 obtained a reaction with human blood diluted 1 : 20,000, other bloods 

 reacting in dilutions of 1 : 100. He said the reactions might take place 

 after some hours, as already mentioned above, and he does not mention 

 anything with regard to the rate at which different bloods react in 

 a series, an omission made by most authors. I find the rate at which 

 a reaction takes place is a very fair index of the degree of relationship. 

 This rate would appear to be due chiefly to the differences in the 

 amount of matter precipitated ; where this is slight, a reaction would 

 only be registered when the particles of precipitum form agglomerations 

 which are visible to the naked eye. On the other hand there may 

 be a slower reaction actually taking place, due to a lesser degree 

 of avidity between the combining substances, but to prove this will 

 be a matter of some difficulty. Experiments made by progressively 

 diluting the antiserum which is added to a blood dilution, have further- 

 more been reported by Ewing (ill. 1902, p. 14), following a suggestion 

 made by Uhlenhuth, and Kister and Wolff. He does not state the 

 blood dilution used, but I presume it was 1 : 100. He found 



Anti-ox serum diluted 1 : 5 to act on dilutions of "several bloods." 



1 : 30 bloods of ox and goat, 



,, 1 : 50 ,, ,, ox blood alone. 



Similar observations were made with anti-human serum tested upon 

 the bloods of man, arid of the baboon, rhesus and Java monkeys. When 



