TUBERCULOSIS. 213 



and the inoculation with substances containing tubercle bacilli was done in 

 the way just given, I invarjably succeeded in obtaining pure cultures. The 

 result was uncertain, on the contrary, when material from human corpses or 

 from cattle with perlsucht was used, as it was always impure on the surface, 

 and, moreover, was not always quite fresh when it reached me. In these 

 cases I first rinsed the surface of the object repeatedly with a solution of 

 corrosive sublimate ( i to 1000) and then cut away the upper parts in layers 

 with red-hot instruments, which were changed repeatedly ; finally, I took 

 the material for inoculation from a depth which justified me in concluding 

 that it would be free from the bacteria which had entered the tissue after 

 the death of the animal. In this way I generally succeeded in obtaining 

 pure cultures even from this kind of material, particularly from small 

 superficial pulmonary cavities, the outer wall of which, after treatment with 

 solution of corrosive sublimate, was removed with hot instruments. 



" After the inoculation of the solidified serum with material containing 

 bacilli has been accomplished, the vessels are place4 in the incubator and 

 kept constantly at a temperature of about 37° C. Every incubator is not 

 suitable for the culture of tubercle bacilli. Growth takes place but very 

 slowly, and the vessels must therefore remain in the incubator for weeks. 

 So that if the incubator is so constructed as to favour rapid evaporation of 

 liquids from the culture vessels, the serum gets dry before visible colonies 

 of tubercle bacilli have developed. For example, an apparatus cannot be 

 used in which the heat is unequally distributed, so that the vapour constantly 

 present condenses in the cooler parts, e.g. , on the glass cover, and has to be 

 continually replaced by moisture given off from the culture glasses. D'Ar- 

 sonval's thermostat is very convenient ; the warmth is equally distributed in 

 it, and the blood serum remains almost unchanged. 



" For the first few days no alteration is to be observed in the cultures in the 

 incubator. If, however, there is a change, and drops or spots of white or 

 other colour form on the surface of the serum, increase more or less rapidly 

 in size, render the fluid at the bottom of the glass turbid or cause the serum 

 to liquefy, it is a sign that the cultui^ is not pure, and that foreign bacteria 

 have entered and choked the growth of the tubercle bacilli. If these drops 

 or spots are examined they are found to consist of bacilli or micrococci 

 which, by Ehrlich's method of staining, assume a different colour from the 

 tubercle bacilli, and are distinct from them also in size and shape. In the 

 tubes free from these impurities, the first signs of the growing colonies of 

 tubercle bacilli are not visible to the naked eye for ten to fifteen days. 

 They then appear as whitish points and small spots lying on the surface of 

 the serum ; they have no lustre, and consequently stand out clearly from 

 their moist surroundings. They are best compared to tiny dry scales 

 adhering loosely to the surface of the serum. The number of the scales 

 and the extent of surface covered by them vary with the richness of the 

 infecting material in bacilli, and with the extent of the surface over which 

 it was rubbed or spread out. 



" The individual scales attain only a limited size, so that if few are present 

 they remain distinct ; but when numerous and closely packed, they coalesce 

 finally and form a very thin, greyish-white, lustreless covering on the serum. 

 After a fragment of the tubercular lung of a guinea-pig has been rubbed on 

 serum, small whitish colonies of tubercle bacilU appear close to the greyish- 

 red bit of lung, and also in its neighbourhood wherever it has been pushed 

 over or pressed on to the surface of the serum by means of the platinum 



