THE PRECIPITIN REACTION. 77 



viz., animal relationships and evolution. It would seem perhaps 

 worth while to make a careful study of the reaction in the hope 

 of clearing away discordant results and of properly evaluating 

 the reaction as a research tool. And inasmuch as precipitins 

 are little known to zoologists in general it has seemed advisable 

 to explain the reaction more fully than would otherwise be 

 necessary. 



PURPOSE OF THIS INVESTIGATION. 



The purposes of this investigation are (i) to study the accuracy 

 and the dependability of the precipitin reaction and factors 

 which may increase these qualities, (2) to apply an improved 

 technique of the reaction with a known per cent, of error to the 

 study of animal relationships. 



TECHNIQUE. 



HISTORICAL REVIEW. 



As first performed by Kraus in 1897 the precipitin reaction 

 consisted in the mixing of a sterile antiserum (formed by in- 

 jecting goats with bacterial culture filtrates) with clear sterile 

 bacterial filtrates in small glass tubes which were then placed 

 in the incubator at 37 C. After twenty-four hours the tubes 

 were examined and those which were markedly cloudy and 

 showed flocculent precipitates were recorded as positive. The 

 "titer" or strength of any antiserum was measured by putting 

 into a series of similar tubes containing increasing dilutions of 

 the antigen (protein to be tested) in normal salt solution, a 

 definite amount of antiserum, then mixing, incubating, and 

 recording after a certain definite time the highest dilution in which 

 a precipitate occurred. As controls, tubes containing normal 

 blood serum and antigen dilution, and tubes containing anti- 

 serum plus salt solution alone, were used. 



This technique was used by many workers in the titration of 

 a great variety of antisera and antigens. It soon came to be 

 noticed, however, that the end point of this " flocculation " test 

 was exceedingly hard to determine, for a cloudiness might occur 

 in the tubes without a settled precipitate and therefore the 

 cloudiness of the experimental tubes graded off imperceptibly 



