THE PRECIPITIN REACTION. 8 1 



according to the method of Folin and Wright (1919). Total 

 nitrogen was determined on one sample and non-protein nitrogen 

 on another, the protein in the latter being removed by pre 

 cipitation by sodium tungstate (10 per cent. ; I cc.) and sulphuric 

 acid (2/3 N; i cc.). The protein nitrogen was then obtained by 

 difference and the protein calculated (protein N x 6.25) as grams 

 per 100 cc. serum. The titrations recorded hereinafter have all 

 been made with these standard antigen solutions of known 

 protein concentration. 



The ring test has been employed throughout, the observations 

 being made at room temperature. As suggested by Evans 

 (1922) buffered salt solutions have been made according to 

 Sorenson's tables given in Clark (1922). These buffered solu- 

 tions were made to have a P H of 7 determined colorimetrically 

 except in certain cases where the P H has been specially noted. 

 The successive increasing dilutions have been made according 

 to the usual Stern-Korte method. Thus a series of small clean 

 glass tubes is set up in a rack and to each tube is added | cc. of 

 buffered salt solution with a sterile 10 cc. pipette. To the first 

 tube of the series is added \ cc. of a standard antigen dilution 

 using a I cc. pipette. The fluids are mixed by sucking up into 

 this pipette several times. From this mixture (now \ as con- 

 centrated as the standard solution) \ cc. is transferred to the 

 next tube, etc. Thus each tube in the series contains a con- 

 centration of protein in buffered salt solution (.85 per cent. NaCl 

 for rabbit sera and usually 2.25 per cent, for chicken sera) one 

 half as great as the preceding tube. The antisera used were 

 generally of a strength of about i : 10,000 which requires eight 

 tubes beginning with a 2 per cent, standard solution. Now to 

 each of these dilutions beginning at the highest is added .1 cc. 

 of the antiserum to be titrated the pipette being placed in the 

 bottom of the tube and the antiserum carefully layered under 

 the antigen dilution. As a control .1 cc. of the same antiserum 

 is layered under f cc. of buffered salt solution. The tubes are 

 then allowed to stand at room temperature and read at 20, 40, 

 and 60 minutes by light from an electric bulb transmitted through 

 a narrow horizontal slit behind the tubes. 



When a suitable antiserum has finally been obtained pre- 



