878 PHAGOCYTES AND PHAGOCYTOSIS IN IMMUNITY 



examined. The average number of bacteria taken up by some arbitrary number of 

 cells was determined and compared with a control. In the control test he used physio- 

 logical salt solution in place of serum, or better, the pooled sera of a number of pre- 

 sumably healthy individuals. 



Wright determined the "opsonic index" of patients by dividing the average num- 

 ber of bacteria taken up when opsonized by the patient's serum by the average num- 

 ber phagocytized under the influence of normal serum. In a large variety of infections 

 he determined that the index was always below normal. In attempting to raise this 

 index in patients by therapeutic inoculation he found that immediately following the 

 injection of the antigen there was a preliminary drop in the patient's index (negative 

 phase) but that as antibodies were produced the index rose to its former level or even 

 higher. During the negative phase the disease condition became aggravated, but im- 

 provement followed upon the rise in the index. As Wright pointed out, this fall and 

 rise in antibody production had already been studied by Brieger and Ehrlich.' 



We have had to change our conception of what goes on during these in vitro ex- 

 periments on phagocytosis, and of their accuracy unless very definite precautions 

 are taken. Kite and Wherry^ showed that the phagocytosis is brought about by the 

 mechanical agitation of the mixture, adhesion between the bacteria and leukocytes 

 taking place when they meet in collision. It follows naturally that in order to insure 

 comparable results one must take precautions to use the same number of bacteria and 

 kind of leukocytes and subject them to the same amount of agitation. When such 

 precautions are taken, the final average number of bacteria found within leukocytes 

 in successive experiments made on the same mixture agree very closely. However, 

 the difficulty encountered in eliminating such sources of error makes impracticable 

 the control of therapy by the estimation of the opsonic activity of sera collected at 

 varying intervals. A quantitative estimation of the opsonin content of a serum is best 

 attained by using dilutions up to the point where there is no longer opsonic action. 

 This method is particularly useful when the serum to be tested has a lytic action.^ 



Another group of workers has sought a way out of the difficulty by advocating 

 non-specific protein therapy."* The rapid improvement which sometimes followed upon 

 even a single injection of a foreign protein^ especially when this produced fever or 

 shock, was noted as long ago as 1893 when Frankel treated cases of typhoid fever 

 with typhoid vaccine and Rumpf with B. pyocyaneus vaccine. The fever, the initial 

 leukopenia followed by leukocytosis, and the quickened circulation of whatever anti- 

 bodies the patient possesses through the blood and lymph are the chief physiological 

 factors which aid in combating infection. ^ 



It is certain that, in view of the specificity of antibodies, therapeutic immuniza- 

 tion must develop along specific lines. We have come to recognize that we must do 

 more than merely inject an antigen. It must be the right antigen, administered in the 

 proper dose at sufficiently frequent intervals; and even if suitable antibodies are 

 formed, we must resort to measures which will overcome the obstacles in the way of 



'Brieger, L., and Ehrlich, P.: Ztschr.f. Hyg. it. Injeldionskrankh., 13, 336. 1893. 



2 Kite, G. L., and Wherry, \V. B.: loc. cil. 



3 Klien, H.: Bull. Johns Hopkins Hasp., 18, 245. 1907. 



'' See chapter Ixxxii of this volume. s Mcintosh, J.: Lancet, 2, 889. 1926. 



