ISOLATION OF B. TYPHOSUS 719 



the organisms present may theoretically be collected in a few 

 cubic centimetres. Practically, however, a large proportion Of 

 the organisms are lost in the process : they probably get carried 

 into and remain in the superficial layers of the filter-candle, and 

 for this reason, though sometimes employed, this method has been 

 largely given up. 



2. Concentration. W. J. Wilson l devised the following method : 

 The water is placed in one or two Winchester quart bottles, and 

 10 c.c. of nutrient broth are added for every litre. The bottles 

 are placed in a water-bath maintained at 37-40 C., and are 

 connected by rubber corks and tubing with a condenser (at a 

 lower level) through which cold water continuously passes, and the 

 tube of the condenser is connected to a large bottle (at a still lower 

 level). This bottle is kept partially exhausted by means of a 

 filter-pump. The water evaporates and is thus concentrated, the 

 evaporated water being condensed and collected in the exhausted 

 bottle. It requires twenty-one to twenty-two hours to evaporate 

 a litre of water. The water remaining in the bottles, now con- 

 centrated to a few cubic centimetres, is then plated on Conradi- 

 Drigalski or malachite -green agar. 



3. Chemical precipitation. These methods depend on the 

 formation in the water of a fine, inert precipitate, which entangles, 

 and carries down with it a large proportion of the bacteria present. 

 Thus in the Vallet-Schiider 2 method, to 2 litres of the water are 

 added 20 c.c. of a 7-75 per cent, solution of sodium hyposulphite 

 and 20 c.c. of a 10 per cent, solution of lead nitrate. The pre- 

 cipitate is allowed to settle or is centrifuged off, and is dissolved in 

 a small volume of a saturated solution of the hyposulphite, from 

 which plates are made in suitable media. Ficker 3 uses ferrous 

 sulphate after making the water faintly alkaline with caustic soda : 

 the ferrous hydrate formed carries down the micro-organisms (this 

 must be a risky procedure, as the typhoid bacillus is very sensitive 

 to caustic alkalies). Iron oxychloride may also be used as the 

 precipitant. H. S. Willson (loc. cit.) employs alum. A stock 

 solution of alum is prepared, containing 10 grm. per 100 c.c., and 

 of this sufficient is added to the water to obtain 0-5 grm. to the 



1 Brit. Mcd. Journ., 1907, vol. i, p. 1176. 



2 Zeitschr. f. Hyg., xlii, No. 2, p. 317. 



3 Hyg. Rundschau, xiv, No. 1, 1904, p. 7. 



