THEORIES OF MECHANISM OF REACTION 95 



rations; that in such cases the patient has probably come very near to the 

 end of his tether in immunizing responses to the existing infection when 

 collateral immunization will accomplish the desired results. This he con- 

 siders very important, because it embraces such a large percentage of 

 cases and says: 



"We are, however, here considering primarily the question of principles, 

 and in connection with this what is of fundamental importance is, that 

 we should discard the confident dogmatic belief that immunization must 

 be strictly specific, and that we should in every case of failure endeavor 

 to make our immunization more and more strictly specific. We should 

 instead proceed on the principle that the best vaccine to employ will always 

 be the vaccine which gives on trial the best immunizing response against 

 the microbe we propose to combat. 



"I would point out that this would almost certainly not involve any 

 revolutionary changes in the accepted practice in either serum therapy or 

 in prophylactic or ordinary therapeutic inoculation. But it would mean 

 taking into account in cases which proved intractable to treatment with 

 the homologous vaccine the possibility of seeking for collateral immuniza- 

 tion by inoculating a microbe or mixture of microbes other than that which 

 the patient is infected." 



Wright's work emphasizing the importance of the tissue enzymes 

 in local resistance to bacterial infection is of interest, although we 

 are of the impression that the terms that have been coined by him 

 are superfluous and will merely add to the confusion produced by an 

 already overburdened nomenclature. It would seem much more 

 rational to determine the exact role of enzymes already known 

 and for which we possess adequate terms, in the reactions that are 

 involved in tissue resistance; only when we have exhausted this field 

 would it seem justified to bring in new terms to cover hypothetical 

 factors. The antibody reactions should not be confused with enzyme 

 reactions, no matter how much we may be tempted to do so. 



Ottenberg and Wallach, who have repeated some of Wright's ex- 

 periments on the production of nonspecific bactericidal substances by 

 methods described by him, failed to confirm his findings. 



Leukocytosis. The reactive leukocytosis that follows nonspecific 

 injections has been repeatedly studied and the suggestion was ad- 

 vanced quite early that the therapeutic effect was to be sought in 

 the artificial leukocytosis established. As noted elsewhere this leu- 

 kocytosis which follows a primary leukopenia, is largely polymorpho- 

 nuclear in character; occasionally an eosinophilia has been described 

 (Holler, Rohonyi), suggesting some relation to the anaphy lactic pic- 

 ture. It has, however, been the general experience that the therapeutic 

 effect need not parallel the leukocytosis that follows the injections: 

 certain cases respond very favorably as far as clinical results are 

 concerned but show relatively little increase in the white count; in 

 other cases the hyperleukocytosis will be very marked but the thera- 

 peutic effect absent. Gay in this country brought out the role of 



