BY «. GREIG-SMITJI. 



391 



and 4 foi- excample. The tests in No.4 were made with thick 

 suspensions, and the curves should have risen above unity at the 

 half dilution. A variable factor must come into play in the 

 determinations. This cannot be the saline used in all cases; it 

 is probably not the bacteria, for the same race was used; the 

 time of incubation will not explain the variation. We are left 

 with the serum and the leucocytes. But, in order to determine 

 the variable factor as nearly as possible, all the constituents v/ere 

 tested for their influence upon the curves. 



It will be noted that the " hump " is well marked in these 

 curves; and it is possible that, had the dilutions of serum l^etween 

 the half and normal been tested, the I'atios might have risen 

 above unity, like the bioken curve in the Hgure. 



The iitjiuence of the serum. — The sera of three individuals were 

 tested with the same suspensions of leucocytes and bacteria. 

 The incubation was for 15 minutes, and the dilutions, etc., were 

 made with 0'85 "o saline. The phagocytic indices of these were 

 17'1, 15-9, and 16o respectively. 



There is a steady fall fi-om unity of the average curve, which 

 is a stiaight line between the y\- and ^ dilution. As there was 

 little ditFerence between the individual curves, it is evident that 

 the serum does not cjive rise to the variations which have been 

 observed in the curves. 



TJie infiii.ence of the bacterial sKspf'i/siotf!^. — Three races of 

 Micrococcus aureus, which had been isolated at different times. 



