132 CLINICAL BACTERIOLOGY AND H^MATOLOGY 



TO. No disease other than syphilis indigenous to this coun- 

 try gives a positive Wassermann reaction except occasionally 

 scarlet fever, in which it only persists for a short time. A 

 positive reaction is therefore practically conclusive. In 

 foreign countries there are a few other diseases which cause 

 it, such as leprosy and frambcesia, and these should be borne 

 in mind in the case of a patient who has travelled. 



11. The most conclusive evidence of the complete cure of 

 syphilis can be obtained by the non-occurrence of a " pro- 

 voked" reaction after an injection of salvarsan. A full dose 

 should be given, and the blood examined after twenty-four 

 hours, and also after a week, ten days, and a fortnight. If no 

 reaction occurs we are as safe as science can make us in con- 

 sidering the patient cured. If only one examination can be 

 made the best time is said to be ten days. 



12. In some late cases (never in early ones) the reaction 

 remains positive in spite of the most vigorous and prolonged 

 treatment, symptoms being absent : the spirochsetes being 

 probably in regions of the body to which drugs cannot gain 

 access. We have not yet enough experience to enable us to 

 say what will happen to these patients. Probably some at least 

 will develop general paralysis or tabes, whereas in others no 

 ill effects seem to arise. They should be closely watched, and 

 a short course of treatment each year seems to be indicated. 



MclNTOSH AND FlLDES's METHOD.* 



The essential feature of this method is the use of a mixture of anti- 

 gen and complement, in standardized amounts, to a fixed quantity of 

 which variable amounts of the serum to be tested (decomplemented 

 by heat) are added. After a first incubation to allow of the linking 

 up of complement, a mixture of sheep's corpuscles and amboceptor 

 are added, the mixtures reincubated, and the results read off. By 

 using the complement and antigen in one mixture and the corpuscles 

 and amboceptor in another, the five essentials in the reaction are prac- 

 tically reduced to three, and much measurement is avoided. The re- 

 actions are carried out in small test-tubes (about 4 by f inch), and the 

 measurements are made by means of a graduated pipette with an 

 indiarubber teat. 



The antigen is prepared by mincing ten parts of heart muscle 

 (human), grinding with sand, and shaking with 90 parts of absolute 

 alcohol for one and a half hours. A i per cent, solution of cholesterin 

 in absolute alcohol is also required. For use take 3 parts of the 



* Brain, 1913, vol. xxxvi., part ii. A slight modification of the method has 

 been introduced by the author (Lancet, 1918, vol. ii., pp. 547'55 2 )- 



