386 Mr. A. H. Caulfeild. Inhibitive and Fixation [Aug. 24, 



lessened or no evidence of inhibitin with - 3 c.c. Besides this apparent increase, 

 a serum without inhibitin may erroneously be credited with that reaction, as 

 the haemolysis can be shown to be due to the natural amboceptor, although 

 the control tube with antigen showed no haemolysis. This agrees with what 

 was earlier stated, namely, that anti-complementary strengths of antigen act 

 partially by suspending complement, and can be offset by increasing the 

 avidity of the corpuscle-amboceptor complex. 



Apart from the inhibitive reaction, this factor may hinder true fixation 

 results. This is best illustrated when - l c.c. serum shows no haemolysis, 

 while 0-3 c.c. gives a trace. Thus in a serum with strong natural ambo- 

 ceptor effects, specific fixation may be partially masked and the results 

 imperfectly interpreted. 



Again, in sera that have been classed as indifferent, all degrees of 

 haemolysis with the second and third dilutions of antigen have been observed. 

 After adsorption, the serum in nearly all instances shows no haemolysis with 

 the first three dilutions, the explanation being already given as serum 

 suspension in addition to antigen adsorption. In fact, in interpreting the 

 results obtained with sera untreated (beyond inactivation), regard must be 

 paid to the controls for both serum suspension and the natural amboceptor 

 for corpuscles. 



From many observations upon the presence and variation of natural ambo- 

 ceptor in the serum of patients in all stages of nutrition, the results have 

 suggested that a study of these bodies* might prove of value in digestion and 

 assimilation experiments. 



The technique that has been adopted for adsorption experiments with 

 small amounts of serum has been as follows : — ■ 



Corpuscle solution (1/20) 5 - c.c. + serum 2-0 c.c. are incubated for 

 1\ hours, centrifuged, and the supernatant fluid removed by a pipette. 

 Proportionately smaller amounts can be used where the amount of serum is 

 less. This serum-saline mixture contains nearly 0*15 c.c. serum in 0*5 c.c, 

 which corresponds to the unit and actual amount used. The serum treated 

 corpuscles are brought to twice the original amount of solution {i.e., 

 10 c.c.) and activated by 10 units of complement. Table VII illustrates the 

 variation that may obtain with both a partially inhibitive and a fixation 

 positive serum. Table VIII shows the effect of the natural haemolysin with 

 indifferently reacting sera, and several sera with varying amounts of 

 inhibitin. 



* McGowan (' Journ. Path, and Bact.,' 1911, vol. 15, No. 3), has shown that when 

 rabbits are fed on ox-blood a haemolysin, an agglutinin, and a precipitin against ox-blood 

 are produced. 



