200 THE PRINCIPLES OF IMMUNOLOGY 



Modifications of the Tests. Numerous modifications of the test 

 have been recommended. These are based on variations in syphilitic 

 antigen, various ways of treating the human serum, differences in 

 selection of the complement and in selection of the hemo- 

 lytic system. These are indicated in the chart on page 192. It 

 is our opinion that any method, to be acceptable, must per- 

 mit of accurate measurement of the reacting bodies. The possi- 

 bilities as to methods of preparing antigen and human serum have been 

 discussed. The use of human complement in the test interposes errors, 

 which we believe have not been overcome. The titration of human 

 complement must differ with different specimens and in the Gradwohl 

 method fails to take account of the variable content of natural hemo- 

 lytic amboceptor in human serum. Of the hemolytic systems recom- 

 mended, the most satisfactory are the sheep or goat and the human 

 systems. In most laboratories the sheep system appears to be most 

 accessible and the factor of error introduced by the presence of normal 

 anti-sheep amboceptors in human serum can usually be overcome by 

 absorption with sheep erythrocytes or can be controlled by the use of 

 one and one-half units of complement. The human hemolytic system 

 largely obviates this objection, but it is sometimes difficult to obtain 

 enough blood to immunize animals for the production of the specific 

 immune hemolysin. We also suggest the possibility that an unusually 

 strong natural iso-hemolysin in the tested serum may confuse the 

 results. Kolmer, in a recent study, has found that the human hemo- 

 lytic system considerably increases the delicacy of the reaction, espe- 

 cially when small amounts of the patients' sera are employed. In 

 positive cases he found 10 per cent, more positive reactions by the use of 

 the human system than with the sheep system. 



The Specificity of the Wassermann Reaction, Numerous studies 

 have been made as to the specificity of the test in the different stages 

 of syphilis. In evaluating such figures certain factors of error in the 

 actual performance of the test must be considered. Unless the worker 

 is familiar with the many factors which may influence the reaction 

 of hemolysis and the fixation of complement, as pointed out briefly 

 in the chapter on hemolysis and the discussion of complement fixation, 

 the results may be misleading. The type of antigen employed is also 

 of significance as influencing the results. Of no small importance is 

 the operator himself, for although the Wassermann test may properly 

 be regarded as a physico-chemical test rather than a strictly biological 

 or immunological reaction, nevertheless it requires a thorough under- 

 standing of immunological procedures. Tests made in the hands of 

 persons trained to perform this test, without broader training, are not 

 to be given the same value as tests in the hands of broadly-trained 

 immunologists. The subject of specificity of the test is closely bound 

 with the clinic, in which certain factors of error in clinical diagnosis 

 must be accepted. Until more satisfactory methods are provided for in 

 the post-mortem room, the factor of error there is almost as large as 

 in the clinic. Warthin, by particularly refined methods applied to 



