254 MEDICAL BACTERIOLOGY 



which otherwise would show a doubtful or negative reaction is more than offset 

 by the frequency of spurious positive reactions, obtained with non-syphilitic 

 sera by this method. This is the worst fault any test can have, hence it is un- 

 justifiable to make the test more sensitive by increasing the amount of antigen 

 used to more than one-fourth the minimum anticomplementary quantity. 



Antigen, in whatever quantity employed, to some degree exerts an inhibiting 

 effect on complement; human serum does likewise; when human serum and anti- 

 gen are combined the total inhibiting effect is greater than the sum of both 

 acting separately. The non-specific anticomplementary factor varies in the 

 sera of different individuals. Within certain limitations, described as the Neis- 

 ser-Wechsberg phase, a deficiency of complement can be made up for by an 

 excess of amboceptor; therefore, if two units of amboceptor are used in the 

 Wassermann test there is always sufficient compensation of the non-specific 

 deviation of complement by human serum and antigen and false positive reac- 

 tions are precluded. The disadvantage of this method of compensation is that 

 frequently serum under examination contains not quite enough syphilitic ambo- 

 ceptor to fix one unit of complement and the anticomplementary factor of the 

 human serum and antigen is low, so that the excess of amboceptor is sufficient 

 to bring about a doubtful or negative reaction although the serum does contain 

 a small amount of syphilitic amboceptor. 



The temptation to increase the sensitiveness of the test by using one unit of 

 amboceptor is too strong for many, but it is a baneful practice bound to elicit 

 numerous pseudopositive reactions. 



It is possible to measure the combined human serum and antigen non-specific 

 deviation of complement; to do so is a procedure that consumes an amount of 

 time and labor that forbids it as a routine procedure. In the comparatively 

 few cases where the use of two units of amboceptor give unsatisfactory results 

 and it is desirable to perform the test using one unit, the non-specific deviation 

 of complement should first be measured and compensated for by the addition of 

 the required amount of complement. 



The selection of rabbit serum containing hemolytic amboceptor is worthy 

 of note. In a previous chapter it has been stated that no serum should be used 

 if more than o.oi cc. was required to furnish a unit of amboceptor. It is not 

 difficult to produce a serum 10 or 20 times as strong, so that o.ooi cc. contains 

 a unit of amboceptor. The greater the amboceptor content the more desirable 

 is the serum because, when a rabbit is injected with sheep cells, precipitins and 

 agglutinins as well as lysins are formed; when the lysin (amboceptor) content is 

 low the agglutinin and precipitin content is proportionately very much greater 

 than when the lysin content is high; therefore, if a serum poor in lysins is em- 

 ployed precipitation and agglutination of the red cells occur and to some degree 

 inhibit hemolysis. Furthermore, rabbit serum, like all other sera, may nor- 

 mally contain substances which tend to inhibit hemolysis and the possibility of 

 such an occurrence is precluded when very small quantities are employed. 



Generally the quantity of antibodies per cubic centimeter in serum is in pro- 

 portion to the severity of an infection or the mass of inoculation; the greater the 



