LOCKE AND HIRSCH 1053 



that not only do dififerent homologous immune sera diflfer greatly in their character 

 but that the constituent immune substances in a single serum differ in similar ways.' 

 The source of these differences is undoubtedly to be found, in part, in correlated differ- 

 ences in the conditions surrounding the origin of the immune substances in the living 

 tissues during the immunization procedure. 



The blood plasma normally contains a small amount of so-called "natural" 

 hemolysins which differ from the true immune hemolysins in that they are less specific 

 and more susceptible to the destructive action of heat.^ The content is variable and 

 so readily replaced and augmented after frequent bleedings or injections of sulistances 

 such as colloidal manganese^ as to suggest an adsorption equilibrium in which the 

 natural hemolysin free in the plasma may represent only a small percentage of a 

 larger amount fixed in the tissues. When sheep erythrocytes are injected into a non- 

 immune rabbit, they adsorb this natural hemolysin so rapidly as to produce a negative 

 phase during which its concentration in the plasma becomes too small to be detected. 

 Adsorption persists until the erythrocyte stroma are saturated and no longer anti- 

 genic nor capable of anchoring to the tissues or, in the excess of antigen, until the 

 animal's supply of pre-antibody substance is exhausted.^ The unwieldy antigen- 

 antibody aggregates are then withdrawn from the plasma and adsorbed upon the 

 surface of phagocytic cells. ^ 



The consequent ingestion and disintegration of the antigen-pre-immune-sub- 

 stance aggregates by the phagocytes'' may partially destroy the binding capacity of 

 the antigen in the aggregates, in much the same manner as is accomplished by ether 

 extraction,^ leaving the pre-immune substance in the resultant disintegration product 

 with a surplus of binding affinity, rendered specific by the residual inclusion of un- 

 digested antigen.9 



'See also Browning, C. H.: Iinniuno-chemical Studies. Wm. Wood & Co., 1925; Singer, E.: 

 Ztschr.f. Immunitdtsforsch.ii.exper. Therap., 3s, 191. 1922; Thiele, F. H., and Embleton, D.: ibid., 

 20, I. 1914; Miiller, P. Th.: Kraus-Levaditi's Handb. d. Tech. u. Method, d. Immunitatsforsch. (ist 

 suppl.), p. I. 191 1 ; Hecht-Johansen, A.: Typhoid-Paratyphoid Group and about Avidity. Copen- 

 hagen: Levin fij Munksgaard, 1923; Locke, A., Main, E. R., and Hirsch, E. F.: loc. cit. 



^ Kolmer, J. A.: /. Immunol., 4, 403. 1919. 



3 Mackie, T. J.: J. Hyg., 24, 176. 1925. -t lijima, T.: /. Path. 6° Bad., 28, 397. 1925. 



5 The duration of the adsorption period is proportional to the size of the animal and to the amount 

 of antigen injected, within certain restrictions (Armstrong, R. R.: Proc. Roy. Soc, B, 98, 525. 

 1925)- 



' Bloom, W.: Arch. Path. &° Lah. Med., 3, 608. 1927. During this negative or "inductive" phase, 

 when much of the animal's available antibody and pre-antibody substance may be combined 

 with antigen and adsorbed upon the phagocytic surfaces of the reticulo-endothelial system, particu- 

 larly in the liver and spleen, extraction of these organs with water and ether, or water and glycerin, 

 may yield a large amount of immune substance (cf. Cary, W. E.: J. Med. Research, 43, 399. 1922). 

 But the antibodies thus extracted do not necessarily have their entire origin in these tissues. The 

 source of the mobilized pre-antibody or natural immune substances is not yet known. Natural hemo- 

 lysin is thought to be a secretion of the phagocytic cells by A. Carrel (/. Exper. Med., 36, 655. 1922) . 

 F. Maltaner believes the hemolysis produced by normal serum to be associated with the mechanism 

 of the blood-coagulation process (/. Immunol., 6, 271. 1921). 



7 Carrel, A., and Ingebrigtsen, R.: J. Exper. Med., 15, 287. 191 2. 



8 See p. 1050, note 6. » Locke, A., Main, E. R, and Hirsch, E. F.: loc. cit. 



