140 CLINICAL BACTERIOLOGY. 



to pleurisies that arise in consequence of penetrating 

 wounds of the chest-wall. The putrid effusions contain, 

 in addition to the causative agents of suppuration, also 

 putrefactive bacteria, and generally the proteus. 



Method of Bacteriologic Investigation. A hypoder- 

 mic syringe with a capacity of from one to six cubic centi- 

 meters is kept filled for from six to twelve hours in five 

 per cent, carbolic acid, and is then carefully cleansed with 

 sterilized water in order to remove all of the disinfectant ; 

 or a Roux's syringe is sterilized by thorough boiling. The 

 point on the chest-wall where the exploratory puncture is 

 to be made is washed with soap, alcohol, mercuric-chlorid 

 solution (i : 1000), and ether, and the puncture, after 

 thorough disinfection of the hands, is made in the usual 

 way. The fluid obtained is received into a sterilized dish and 

 each of four or five agar-tubes successively, or an agar-plate, 

 is inoculated with a drop thereof, or a drop of the exudate 

 is permitted to flow directly from the syringe upon the sur- 

 face to be inoculated. The tubes are introduced into the 

 thermostat at a temperature of 37 C. (98.6 F.). At the 

 same time cover-slip preparations are made in the usual 

 manner, and examined for tubercle-bacilli and other bac- 

 teria. With the remainder of the fluid, if tuberculosis be 

 suspected, two or three guinea-pigs may be inoculated 

 through the peritoneum (see later). Serous effusions con- 

 tain few, if any, microorganisms. It is, therefore, a more 

 reliable procedure to remove considerable amounts of the 

 fluid, to centrifugate or sediment it, and to study the pre- 

 cipitate only. 



Diagnostic and Prognostic Significance of the Bac- 

 teriologic Findings. Bacteriologic investigation is not of 

 great importance in the diagnosis of serous effusions. The 

 large majority of serofibrinous pleurisies prove to be sterile. 

 The metapneumonic serous effusions contain at times the 

 diplococcus lanceolatus Frankel, and before, as well as 

 after, the crisis. The presence of pyogenic microbes in 

 serous pleural effusions, as has been repeatedly observed, 

 does not, in all instances, justify the conclusion that puru- 

 lent metamorphosis into an empyema will take place. Such 

 effusions may, under circumstances, recede completely. If 

 a decision is to be reached in doubtful cases of serofibrinous 

 pleurisy whether tuberculosis exists or not, it is advisable 

 to inject into the peritoneum of guinea-pigs some of the 



