156 PROPERTIES OF ANTI-SERUM TO SERUM 



when he says that the amount of complement fixed always 

 depends upon the amount of precipitate, as the results 

 above given show that the maxima of the two reactions 

 may not correspond. 



We may vary the conditions of experiment by keeping 

 the amount of serum fixed and varying the amount of 

 anti-serum. In this case Moreschi also found, as shown in 

 his table No. 3, that on increasing the amount of the latter 

 an optimum point was reached, beyond which additional 

 increase of anti-serum resulted in diminution in the amount 

 of complement taken up. We have made a large number 

 of observations on this point, but with varying results. 

 In one or two instances we found a slight diminution in the 

 amount of complement deviated, as the anti-serum was 

 increased, but this was never very marked ; whilst in the 

 majority of cases we found no such diminution, even when 

 as much as 0-3c.c. anti-serum was added. At present we 

 cannot give any explanation of this discrepancy. Certainly 

 the phenomena of optimum deviation do not occur in the 

 same striking manner as when the amount of anti-serum 

 is kept fixed and the amount of homologous serum is varied. 

 As stated above, however, we found when we continued to 

 increase the amount of anti-serum (the homologous serum 

 being kept fixed) that the amount of precipitate formed 

 might continue to increase, whilst the deviation of comple- 

 ment did not do so. 



GENERAL CONSIDERATIONS 



The deviation of complement by a serum plus its anti- 

 serum (presumably by a new compound formed) is one of 

 the most striking of serum reactions, and opens up many 

 questions of high theoretical importance. The relation of 

 the phenomenon to precipitation has already been discussed 

 at some length, and whilst there is a certain parallelism 

 between them, we cannot say that it is the precipitate which 



