A STUDY OF FIXATION REACTIONS. 375 



and simultaneous tests were conducted by the precipitin and fixation 

 methods. The results were as follows : 



Precipitin limits of anti-Caucasian serum No. 3332. 



Monkey 1 to 1,000. Negro 3 1 to 800. 



Negrito 1 to 1,000. Chinese 1 to 2,000. 



Tagalog 1 to 2,000. Caucasian 1 to 3,000. 



Japanese 1 to 3,000. 



Precipitin limits of anti-Filipino serum No. 3167. 



Tagalog 1 to 2,400. Negro 1 to 1,800. 



Chinese 1 to 2,400. Negrito 1 to 1,800. 



Japanese 1 to 2,400. Caucasian 1 to 1,200. 



Monkey 1 to 1,000. 



The limits here recorded indicate the greatest dilution of the serum 

 which still gave a trace of reaction with the antiserums and, as it is 

 extremely difficult to determine with certainty the exact position of these 

 limits, it seems to us that the differences obtained can have but little, if 

 any, practical value. However, the anti-Caucasian serum does precipitate 

 Chinese, Japanese, Tagalog, and Caucasian serums at greater dilutions 

 than it does monkey, Negro, and Negrito and likewise the anti-Filipino 

 serum gives precipitation with greater dilutions of Filipino, Chinese, 

 and Japanese serums than with those of the Negro, Negrito, Caucasian, 

 and monkey. This last finding is especially interesting because it is 

 directly contradictory to the rather remarkable conclusion arrived at by 

 Brack ( 16 ) from his deflection experiments. He found that anti-Cau- 

 casian serum gave deflection with Caucasian serum 1 to 1,000, Arabian 

 1 to 900, Chinese 1 to 700, Malay 1 to 500'; that anti-Chinese serum 

 showed the same limits with Chinese and Caucasian serums, but required 

 more of the Malay serum; that with anti-Malay serum the same limits 

 were obtained for all three serums. He therefore concludes that the 

 protein of the Caucasian contains all the groups of the Chinese and 

 Malay proteins and in addition certain groups peculiar to itself ; that the 

 protein of the Chinese contains all the groups of the Malay and certain 

 other groups not contained in the Malay protein. 



In deflection of complement tests conducted with dilutions as close 

 together (600, 700, 800) as is indicated in Brack's tables, the difference 

 in the degree of haemolysis in the adjacent tubes is so slight that it 

 becomes very difficult to say where the limit of deflection lies and we 

 believe that Brack's tables indicating a sharp dividing line between the 

 blocking of haemolysis and haemolysis are misleading, although we 



3 As the Negro serum was cloudy, it was subjected to repeated centrifugation 

 during the course of these tests and a not appreciable amount of proteid matter 

 was thus removed, so that the dilution was probably really greater than 1 to 800. 



