520 



Messrs. Carnelley and Haldane. 



Experiments on the Efficiency of Water-traps. 



The means commonly employed for preventing the escape of sewer 

 air into houses is the ordinary water-trap. Since the experiments of 

 Nageli (' Die Niederen Pilze,' p. 109) it has been known that these 

 traps, when acting properly, absolutely prevent the passage of micro- 

 organisms. But it is evident that they cannot altogether prevent the 

 passage of volatile constituents of sewer air, and we thought it worth 

 while to make a few experiments on this point. We were not aware 

 that the matter had already been experimentally investigated by 

 Fergus ('The Sewage Question,' 1874), who employed methods simi- 

 lar to those used by us. As, however, the test substances used by us 

 were nearly all different from those employed by Fergus, it may be 

 well to give the results of our experiments. A leaden U -shaped trap, 

 A, B, C, 2 J inches in diameter, and with a seal, a h, of 3 inches in depth, 

 was closed at each end, A and C, with a sheet of india-rubber 

 stretched tightly over the mouth and fixed with wire, each sheet 



being perforated by a hole in the middle. The trap was then filled 

 with water, and a glass stopper placed in the aperture E, while the 

 neck of a flask D, containing the substance under investigation, was 

 fixed through the india-rubber sheet at A. The whole was then left at 

 rest, and observations made from time to time by removing the stopper 

 and ascertaining whether the smell of the substance in D could be 

 detected at E. In other cases a tightly fitting inverted test-tube, 

 containing litmus or other test-paper, was inserted at E, in place of 

 the glass stopper, and observations made as to when the test-paper 

 was first distinctly affected. The results obtained are given in the 

 following table : — 



