BY K. GREIG-SMITH. 



389 



We have here a great exaggeration of a break or "hump" at 

 the quarter dilution. The main point that is to be inferred from 



Fig. 3. — Salted serum, with thin and thick bacterial suspensions, 

 the experiment is that tlie serum does not contain a deficiency of 

 salt but an excess, and this is probably the chief reason for the 

 rise of the curve upon dilution. 



The low indices obtained with the undiluted serum are a 

 direct effect of the added salt. A preliminary experiment showed 

 that the addition of salt lowered the index to one-fourth. 



The effect of adding distilled ivater to the serum. — The last 

 experiment led to testing the effect of adding distilled water to 

 the serum in varying proportions. The corpuscles and staphylo- 

 cocci were suspended in 0*85 % saline. 



Vols, of serum 987654. 3 21 



Vols, of distilled water 123456789 



Cocci per leucocyte 167 19-8 21-8 189 17-9 157 14-6 12 2 113 



The maximum phagocytosis was obtained when seven parts of 

 serum were mixed with three parts of distilled water. 



