BACILLI IN CHRONIC INFECTIOUS DISEASES. 407 



for the purpose of washing the precipitate. This is again allowed to 

 settle and the procedure is repeated three or four times, after which 

 the precipitate is washed with absolute alcohol. It is then placed 

 upon a filter and dried in a vacuum exsiccator. 



An analysis of this purified tuberculin by Proskauer gave 18.46 

 per cent of ash, consisting almost entirely of potassium and magne- 

 sium phosphate. The elementary analysis gave 48.13 per cent of 

 carbon, 7.06 per cent of hydrogen, 14.46 per cent of nitrogen, and 

 1.17 per cent of sulphur. 



Tizzoni and Cattani (1892) have presented some experimental evi- 

 dence which indicates that injections of Koch's tuberculin into 

 guinea-pigs may produce in these animals a certain degree of im- 

 munity against tuberculosis ; and that this immunity depends upon 

 the presence of an anti-tuberculin formed in the body of the partially 

 immune animal. 



Numerous experiments made by veterinary surgeons upon tuber- 

 culous cows show that the injection of Koch/s tuberculin in these 

 animals, in doses of thirty to forty centigrammes, produces a rise of 

 temperature of from 1 to 3 C. The febrile reaction usually occurs 

 in from twelve to fifteen hours after the injection. Its duration and 

 intensity do not depend upon the extent of the tuberculous lesions, 

 but is even more marked when these are slight than in advanced 

 cases. In non-tuberculous animals no reaction occurs, and the ex- 

 periments made justify the suspicion that tuberculosis exists if an. 

 elevation in temperature of a degree or more occurs as a result of 

 the subcutaneous injection of the dose mentioned. 



When the number of tubercle bacilli in sputum is comparatively 

 small they may easily escape observation. Methods have therefore 

 been suggested for finding them under these circumstances. Ribbert 

 (1886) proposed the addition to the sputum of a two-per-cent solution 

 of caustic potash, and boiling the mixture. The tenacious mucus is 

 dissolved, and when the mixture is placed in a conical glass vessel 

 the bacilli are deposited at the bottom and may easily be found in 

 the sediment after removing the supernatant fluid. The same object 

 is accomplished by Stroschein (1889) by the addition to sputum of 

 three times its volume of a saturated solution of borax and boracic 

 acid in water. 



A method of estimating the number of bacilli in sputum has re- 

 cently been proposed by Nuttall, which appears to give sufficiently 

 accurate results and to be useful in judging of the progress of a 

 case or of the results of treatment. For the details of this method 

 we must refer to the author's paper (Johns Hopkins Hospital Bulle- 

 tin, vol. xi. ? No. 13, 1891). It consists essentially in first making 

 the sputum fluid by the addition of a solution of caustic potash ; in 



