DEVIATION OF COMPLEMENT 



153 



sweat. For this reason he advises the use of anti-sera 

 which give deviation with 0-000,01 c.c. of the homologous 

 serum as a minimum ; with a serum of this strength there 

 is no risk of error such as might arise from the material to 

 be tested being impregnated with sweat. 



6. ON THE QUANTITATIVE RELATIONS OF SERUM AND ANTI- 

 SERUM TO THE DEVIATION OF COMPLEMENT 



In the tables given above it appears that the amount 

 of deviation of complement increases, though not in arith- 

 metical ratio, with the amount of serum, when the amount 

 of anti-serum is kept constant. If, however, comparatively 

 large amounts of the homologous serum be used, the amount 

 of complement taken up again becomes diminished as the 

 amount of serum increases. We have carried out a number 

 of observations on this subject, one or two examples of 

 which will exemplify the phenomenon. The results are 

 confirmatory of those recently published by Moreschi, 

 though the details of experiment are somewhat different. 



Anti-serum, rabbit v. ox, 0-025 c.c. to each tube, with varying amounts 

 of ox's serum 55 C. 



Test for complement =1 c.c. suspension of ox's corpuscles treated with 

 immune-body ; the dose of complement for this is 0-0075 c.c. 



It is thus seen that with a given quantity of anti-serum, 

 0-025 c.c. in this case, there is an optimum amount of serum 

 which gives the maximum deviation, viz. 0-001 c.c., whilst 



