438 THE TECHNIC OF COMPLEMENT-FIXATION REACTIONS 



be more anticomplementary than the serum used in the complement 

 titration, and hence when used in the main test, with exactly one unit 

 of complement, mild degrees of inhibition of hemolysis will be secured 

 that may be interpreted as slightly positive reactions; or (2) the serum 

 used in the complement titration may contain more or less natural 

 hemolytic amboceptor than the patient's serum, and this factor exerts 

 an influence on the titration, so that the unit varies with different serums. 

 For these reasons I have included here a fourth method for using a 

 single unit of complement under conditions where the anticomple- 

 mentary action of each serum and the antigen are determined, in pref- 

 erence to titrating the complement with a serum and using this unit for 

 a number of other serums that are sure to differ from each other. 



In this titration, therefore, we determine the amount or unit of 

 complement necessary to produce hemolysis with fixed amounts of 

 amboceptor and corpuscles. It will be observed that I have used two 

 units of amboceptor, as determined by a previous titration. These two 

 units are equivalent to one dose, and the same would be true whether 

 three, four, five, or more units were 'used, because in this titration the 

 corpuscles and amboceptor are arbitrary and fixed constants, and are 

 used for determining the amount of complement necessary to bring 

 about complete hemolysis. 



In conducting the main tests the dose of corpuscles and that of 

 amboceptor are the same as those used in the complement titration, but 

 instead of using exactly one unit of complement, it is necessary to use 

 one and one-half or two units, to allow for the anticomplementary 

 action of antigen and patient's serum. 



As the result of a large number of comparative titrations and tests 

 I have found that 0.05 c.c. of complement serum (=1 c.c. of a 1:20 

 dilution) is a safe amount to use with 2.5 per cent, corpuscle suspension, 

 and equally good results are secured by adopting this amount as a fixed 

 dose in titrating the amboceptor before each day's work. If the pig 

 serum happens to be weaker or stronger in complement than it is ex- 

 pected to be, the differences are adjusted in the amboceptor titration. 

 Following the same rule, one and one-half or two units of amboceptor 

 are used in conducting the main test, the extra half, or one unit, over- 

 coming the anticomplementary action of antigen and patient's serum. 



It is important to remember that, in conducting this titration, 

 amboceptor, complement, and corpuscles are to be mixed at once; if 

 amboceptor and corpuscles are mixed and allowed to stand for ten 



