160 IMMUNOLOGY 



Mention has already been made of the effect of zymin and ammonia 

 upon midpieee and endpiece, respectively. These results indicate 

 the necessity of using chemically clean glassware in all im- 

 munological work dealing with opsonins and complement fixation. 



Origin of Complement. — In regard to the origin of complement 

 nothing is known definitely, nor is it absolutely settled that it 

 exists as such in the circulating plasma. Buchner (1892), Hankins 

 (1892) and Metchnikoff suggested, quite early, that it was of leuco- 

 cyte origin. In connection with this, it is of interest to note that 

 when sensitized red cells are phagocytized by leucocytes, they seem 

 to be no more readily hemolyzed than unsensitized cells. A great 

 many have reported that the complement titer of serum is higher, 

 if it is allowed to stand on the clot in the refrigerator at 6 to 8° C, 

 for 24 hours. It has been argued that this increase in titer is due 

 to the leucocytes present. Regardless of theoretical considerations, 

 the observation that an increase in titer frequently occurs under 

 the conditions just mentioned is of practical importance in diag- 

 nostic serology. A great many have offered experimental evidence 

 that the liver is the source of complement. Sherwood, Smith and 

 West (1916) investigated this question and concluded that it is 

 neither the only, nor the principal source of complement. They 

 point out important sources of error in the work of Dick and of 

 Nolf and others who had concluded that complement is of hepatic 

 origin. 



Desirable Qualities of a Complement. — ^For use in complement 

 fixation, there are certain qualities that must be considered in 

 choosing complement. It should be readily fixed by a sensitized 

 antigen ; free from natural hemolysins, and agglutinins ; lysis 

 should be reasonably rapid, and it should be sufficiently stable 

 for use in overnight fixation at 6° to 8° C. 



Complement From Various Animals.^ — Guinea pig complement 

 most nearly meets these requirements, although human complement 

 is quite satisfactory. The complement titer varies from day to 

 day in the same pig, and may be nil in apparently healthy guinea 

 pigs. For these reasons it is recommended that pooled complement 

 (from 3 or 4 guinea pigs) be used in complement-fixation tests. 

 A satisfactory unit of guinea pig complement in the Kolmer- 

 Wa.ssermann technique should be from 0.25 to 0.35 c.e. of a 1 :30 



