142 CLINICAL BACTERIOLOGY. 



they contain tubercle-bacilli or are sterile. While operation 

 for all other varieties of empyema, even in cases of tuber- 

 culosis, is attended with a relatively favorable prognosis, 

 this is unfavorable in cases of empyema associated with the 

 presence of tubercle-bacilli. Generally, healing does not take 

 place, a fistula remains, and the chronic suppuration leads 

 to death. Nevertheless, some surgeons recommend opera- 

 tion even in these cases, as, without doubt, recovery may 

 take place also under such conditions. 



Pneumothorax. Pneumothorax with perforation, which 

 occurs so commonly in tuberculosis, is, as a rule, followed 

 by purulent effusion. The exudate contains tubercle- 

 bacilli, demonstrable microscopically or by experiment on 

 animals, and, in consequence of mixed infection, one or 

 another of the pyogenic cocci, and, besides, at times, putre- 

 factive bacteria, especially the proteus. Only one case of 

 pneumothorax without perforation has been studied bacter- 

 iologically. In this the anaerobic, gas-forming bacillus 

 emphysematosus (p. 128) was found to be the exciting 

 agent. The method of examination is the same as in the 

 case of pleurisy. 



PNEUMONIA. 



Frankel' s pneumococcus maybe looked upon as the caus- 

 ative agent of lobar croupous pneumonia, as this organism 

 is demonstrable in more than three-fourths of all the cases 

 in the pulmonary tissues, which, normally, are entirely free 

 from bacteria. In the larger proportion of cases the pneu- 

 mococcus alone is present in the diseased tissue ; in the 

 smaller proportion, staphylococci and streptococci besides 

 are present. Streptococci alone, or in association with 

 staphylococci, rarely staphylococci alone, may, however, 

 be found in pneumonic foci. In addition, the bacillus of 

 Friedlander (Figs. 42, 47) is found in some cases. All of 

 these bacteria, as has been pointed out, exist in the upper air- 

 passages. They have, further, been demonstrated, though 

 less commonly, in the healthy larynx, but bacteria do 

 not occur in the bronchi and in the pulmonary tissues in 

 healthy persons. For the development of pneumonia, it is, 

 therefore, generally necessary that an accidental, indirect 

 cause (cold, traumatism, etc.) shall cooperate that renders 

 it possible for the bacterium to penetrate more deeply 

 into the air-passages, and there to give rise to inflammation. 



