518 THE TECHNIC OF COMPLEMENT-FIXATION REACTIONS , 



losis with bacillary emulsion as antigen and obtained 33 per cent. 

 Turban I cases, 70 per cent, Turban II, and 62 per cent. Turban III 

 positive results. Laird 1 (1912) out of 84 tests in 34 cases obtained 24 

 positives in 4 cases, using watery emulsion of tubercle bacilli (which 

 he does not describe); his results were inconclusive. Hammer, 2 using 

 O. T. and extracted tuberculous nodules, obtained 97 per cent, positive 

 results in 46 tuberculous cases. Calmette and Massol, 3 using prepara- 

 tions made from tubercle bacilli by extracting with water and peptone, 

 obtained in 134 cases 92.5 per cent, fixation. Fraser 4 (1913), testing a 

 large variety of antigens, found that living bacilli gave no fixation in 

 96.6 per cent, of normal individuals, but gave positive reactions in 

 42.3 per cent, of tuberculous individuals. She states that the most 

 reliable antigen is prepared from living human bacilli, and that diag- 

 nostically the complement-fixation test with living bacilli is of more 

 value from the standpoint of positive results than any other reaction 

 discovered to date. She believes the absence of antibodies accounts 

 for the low percentage of results obtained. Dudgeon, Meek, and Weir 5 

 also tested a large number of antigens, and in 102 untreated cases 

 obtained 86 positive results, while all cases which had been treated 

 with tuberculin gave positive results. Products of the bacilli themselves 

 were found to be the most satisfactory as antigen. With an alcoholic 

 antigen 6 prepared from tubercle bacilli they obtained from a total of 

 234 cases, 209 (89.3 per cent.) positives, 194 of these on first examination, 

 11 (of the 15 negative) on second examination, and 4 more on third 

 examination. Besredka 7 (1913) prepared an antigen by growing 

 tubercle bacilli on egg broth, heating it, and filtering. With this antigen 

 Bronfenbrenner 8 (1914) obtained a very high percentage of positive 

 results, 93.8 per cent, in active cases, and 55.5 per cent, in convalescents, 

 while suspected cases gave 75 per cent, and syphilitic sera 24 per cent, 

 positive reactions. Inman 9 and Kuss, Leredde and Rubenstein 10 found 

 this antigen non-specific. Mclntosh, Fildes, and Radcliffe 11 (1914) 

 also justly criticized Besredka's antigen, and concluded, after testing a 

 large number of antigens, that the living bacillary emulsion was best, 



1 Jour. Med. Research, 1912, 27, 163. 



2 Miinchen. med. Wehnschr., 1912, 59, 1750. 



3 Compt. rend. Soc. de biol., 1912, 73, 120. 

 4 Ztschr. f. Immunitatsf., 1913, 20, 291. 



5 The Lancet, 1913, 184, 19. 6 Jour. Hyg., 1914, 14, 52, 72. 



7 Compt. rend. Acad. d. sc., 1913, 156, 1633. 

 8 .Arch. Int. Med., 1914, 14, 786. 

 9 Compt. rend. Soc. de biol., 1914, 76, 251. 

 10 Ibid., 244. " The Lancet, 1914, 185, 485. 



