362 BACTERIOLOGY. 



lates the existence of a hypothetical " im- 

 munizing proteid " (p. 332), which is a combi- 

 nation of such active leucocytic proteid with 

 bacterial proteid. The formation of such 

 substances, however, is advantageous only 

 when kept within certain limits. If too many 

 leucocytes are broken up so that a cloudy so- 

 lution, rich in eosinophilous granules arises, 

 then such a serum, which when originally 

 possessed of a merely moderately alkaline reac- 

 tion and intact leucocytes could kill bacteria, 

 may become through the steady increase in 

 alkalinity a favorable nutrient medium for 

 bacteria (Fischel). Hence we may find con- 

 nected with the state of the serum all degrees 

 of bodily condition from perfect immunity to 

 a high predisposition to disease.- During the 

 interchange of substances in cell nutrition, 

 other active proteid bodies may pass over into 

 the serum from the cells and tissues of the 

 body, and so by the indirect participation of 

 those cell territories which are concerned in 

 any special disease, these organic elements may 

 be a factor in the predisposition to disease, 

 without it being necessary to assume that any- 

 thing of a foreign nature is added to the body. 

 Within the body, however, blood or any 

 body-fluid, if devoid of cells exerts no defen- 

 sive action. MetschnikofI discovered, for ex- 



