Complement Fixation Method. 719 



termine the smallest quantity of serum that will prevent 

 hemolysis. The result is shown by the contrast between the 

 clear serum following complement fixation and the distinctly 

 red serum containing the hemoglobin in solution. Although 

 errors may occur in this method of diagnosis in those excep- 

 tional instances in which the serum of horses affected with 

 chronic glanders will not fix the complement and in the still 

 rarer cases in which an apparent specific fixation occurs in 

 healthy horses, these errors are considerably less frequent than 

 those, obtained by other sero-diagnostie methods. 



According to the official regulations of Prussia, which prescribe that the serum 

 of horses which will fix complement in maximum quantities of 0.1 c.cm. in the pre- 

 scribed method of procedure (see below) are regarded as afEected with glanders. 

 Basing their work upon this standard, the results of tests made by Miessner & 

 Trapp in the examination of 549 healthy horses showed that the serum of only 7 

 (1.27%) showed a complement fixation power equal to 0.1-0.01, while on the other 

 hand the sera of 69 glandered horses showed a fixing power of 0.2 in two cases and 

 a fixing power exceeding 0.2 in five cases. In one experimentally infected 

 horse the fixing power, as well as the agglutination, began to decline on the sixth 

 day, but in horses infected naturally with glanders the fixing power was retained 

 for a longer time; previous malleinization also tended to increase the fixing power 

 temporarily. Haan, de Bliek, and Valenti, as well as Klimmer & Kiessig, also 

 report favorably on the results obtained by this method, while Keyser found this 

 method applicable to cadavers also. Sera of horses affected with other diseases 

 gave no positive reaction. 



Technic of the Complement Fixation Method. This consists essentially in 

 mixing suitable amounts of glanders-bacillus-antigen and fresh guinea pig serum 

 containing complement with the inactivated serum to be tested and then 

 incubating for one hour, in order to hasten fixation. To this mixture 

 are then added the inactivated hemolytic serum and the corresponding 

 red blood corpuscles. This two-fold mixture, after again being incubated for 

 two hours, is ready for the determination of results. If complement fixation has 

 been complete all "of the red blood corpuscles will have settled to the bottom of 

 the test tube, while the supernatant slightly yellow fluid is perfectly clear. On the 

 other hand, if hemolysis was complete all blood corpuscles will have been dissolved, 

 so that the supernatant fluid, which is clear also in this case, is of an even red 

 or port wine color. Transition stages in which only a part of the red blood cor- 

 puscles are dissolved, may be observed between these two extremes, in which case 

 the fluid is only slightly reddened, or reddened only in the lower part of the test 

 tube, while a certain number of intact blood corpuscles, forming, ^ correspondingly 

 smaller mound, will have collected on the bottom of the vessel. For the sake of 

 convenience the different degrees of hemolysis are referred to as 0, trace, incom- 

 plete, nearly complete, and complete. 



This method was first applied practically to the diagnosis of glanders by 

 Sehiitz & Schubert, but subsequently perfected by Miessner & Trapp. According to 

 these investigators the test is conducted with the following reagents: 



(a) The antigen usually consists of a normal saline solution of sterilized 

 shake-extract, or of a suspension of dead glanders bacilli; antiformin extracts or 

 mallein are 'less suitable for this purpose, although in some cases the_ authors 

 obtained more pronounced resiilts with these than with the first named -antigens. 



(b) A series of successively decreasing amounts of the serum to be tested, 

 which must be inactivated by incubating for 30 minutes at 56° to 60° C. 



(c) Fresh guinea pig ferum, containing complement in the smallest amount 

 (usually 0.03-0.04 e. cm.), in which it will produce complete hemolysis in a hemolytic 



sys em.^ Hemolytic inactivated rabbit serum (from rabbits previously treated 

 with sheep blood corpuscles), in double the amount required to produce hemolysis. 



(e) Sheep blood corpuscles, 5% suspension in physiological salt solution. 



Reagents of which only small quantities are needed should be standardized by 

 the addition of normal salt solution, so that equal volumes of each reagent enter 

 into the mixture. By providing suitable control tubes we are enabled to determine 

 that neither the antigen nor the serum under test have alone been the cause of the 

 complement fixation, and also that the hemolytic system was properly formed. The 

 successive steps in this test are illustrated graphically m the tables on page 720. 



