go ALAX ARTHUR BOYDEX. 



experiments. It was noted, however, that certain chicken anti- 

 sera gave spontaneous rings in such salt solutions. It became 

 necessary then to inhibit this reaction in order to record ring 

 tests due to the presence of antigen in the salt solution. Fol- 

 lowing a suggestion of Hektoen's (1918) a more concentrated salt 

 solution was used. It was found that in some cases a 2.25 per 

 cent, salt solution was necessary to inhibit completely this spon- 

 taneous ring. The question then arose as to whether such a 

 heavier salt concentration would alter the group reactions of a 

 given antiserum. To settle this matter a chicken antiserum 

 which did not require a 2.25 per cent, saline was nevertheless 

 titrated with the same series of antigens in both .85 per cent, and 

 2.25 per cent, saline. The results are shown in Table VI. 



TABLE VI. 



THE EFFECT OF N.A.CL CONTENT ON TITER. 



Table VI. shows the distinct inhibition due to the heavier 

 salt solution. Furthermore the specificity of the antiserum has 

 been increased. As a matter of fact the range of reactions in 

 chicken antisera is more closely parallel with that of rabbit 

 antisera when this heavier salt solution is used (see Figs. 3 and 5). 



~ . l-'.JJ'ci't of Filtration on Titer. 



When an antiserum becomes contaminated it must be filtered 



at once for the tiler r.ipidly decreases during putrefaction. The 



possible effect -.1' MI. h till ration on the specificity of reaction of 



isera must then \- known. For this purpose parallel titra- 



