350 



Scientific Proceedings (130) 



dinary broth. The first generation on ordinary broth showed 

 a tendency to agglutinate, but the reaction was much slower 

 than in the control test. The experimental production of serum 

 fast staphylococci suggests that the mechanism of serum fast- 

 ness may be due to a lack of wetting. This would then explain 

 why the tubercle bacillus which is so rich in fat-like substances 

 gives inconstant serum reactions. 



170 (2130) 



The precipitin test in the diagnosis of tuberculosis. 

 By W. P. LARSON, IRWIN A. MONTANK, and EDMOND NELSON. 



[From the Department of Bacteriology, University of Minnesota, 

 Minneapolis, Minnesota.] 



The precipitin test has been found to give reliable results in 

 the diagnosis of active tuberculosis. The antigen is prepared 

 by disrupting tubercle bacilli, preferable an old culture, with 

 carbon dioxide by the method described by Larson, Hartzell and 

 Diehl. 1 The disrupted bacteria are filtered through paper in 

 order to remove the shells. The clear filtrate is layered over 

 the serum to be tested, and the tubes incubated for a period up 

 to two hours. A definite cloudy ring at the interface of the two 

 fluids indicates a positive reaction. The cloud often appears 

 within the first five minutes. In the far advanced cases, how- 

 ever, the reaction develops more slowly, but is usually very 

 definite. Upon standing several hours the ring gradually be- 

 comes dispersed. 



Thus far the blood serum of 190 cases have been examined. 

 Of these, 100 were patients in the University Hospital and Dis- 

 pensary, but not in the tubercular clinics. Ninety cases, repre- 

 senting all stages of tuberculosis, were from a local sanatorium. 

 From the 100 cases not suspected of having tuberculosis, eleven 

 positive reactions were obtained. Six of these have since been 



i Jour. Inf. Dis. f 1918^ xxii, 271. 



