SOUECES J3F INFECTION. 35 



squeezing it out in bouillon, so as to liberate any germs that might 

 have been caught. This fluid was then injected into the peritoneal 

 cavity of guinea-pigs. If tubercle bacilli were present, the animal 

 invariably manifested symptoms of peritoneal tuberculosis three to 

 four weeks after inoculation. In 7 hospitals in Berlin, 21 wards 

 occupied by tubercular patients were examined. The sponge was 

 only used on surfaces which had never been exposed to direct con- 

 tact with the expectorations of the patients. In 15 out of the 21 

 wards tubercle bacilli were present, and of 94 animals which were 

 inoculated, 20 died of tuberculosis. In the homes of fifty -three 

 private patients, the presence of tubercle bacilli was demonstrated 

 by the same method in 20. 



That microbes in a dried state, adherent perhaps to particles of 

 dust, follow the currents of atmospheric air, has been demonstrated 

 by the history of endemics of malaria. 



Cadene and Malet (" Etude experimental de la transmission de 

 la mave par contagion mediate ou par infection," Revue de MSdicale, 

 t. viii. p. 227) have studied experimentally the transmission of the 

 bacilli of glanders through the atmospheric air, and have come to 

 the positive conclusion that the disease is not contracted in this 

 manner. They exposed animals which are known to be very sus- 

 ceptible to the disease, such as the ass, to the expired air of animals 

 suffering from glanders, but the results were always negative. 

 They next produced a bronchitis in animals by inhalations of 

 bromine before they were exposed to the expired air of animals 

 suffering from glanders, but the results remained the same. The 

 expired air of diseased animals was next mixed with steam, and 

 the latter was condensed, but no bacilli could be found, and inocu- 

 lation in guinea-pigs only yielded negative results. The same 

 negative results were obtained by passing the expired air over 

 water, as the latter was always found sterile. The air in the 

 stables occupied by the diseased animals was also found free from 

 bacilli. Pure cultures of the bacilli were injected into the trachea 

 of a number of animals, with the result that in only two was the 

 disease produced, while all animals inoculated in the same manner 

 which were suffering from bromine bronchitis contracted the disease 

 and died. 



It is also well known that microbes which have been carried 

 upward by currents of air descend by virtue of their own weight. 

 When the air is in a quiescent state the lower strata contain 

 microbes in greater abundance than the upper. Klebs found that 

 in the Campagna districts, near Kome, an elevation of three metres 

 afforded perfect immunity against malaria. Soyka made some 

 observations on currents of air as a vehicle for the diffusion of 

 microbes in Klebs's laboratory at Prague. He came to the conclu- 



