1885.] The Removal of Micro- Organisms from Water. 



381 



its test-tube, and before removing the cotton- wool ping the latter is 

 burnt outside, so as to destroy any organisms that may be adhering 

 to the • exterior. A given number of drops of the water or other 

 liquid under examination are now introduced into the open test-tube 

 by means of a pipette, which has been previously sterilised by heating 

 it nearly to redness in a Bunsen flame. The water and liquid gelatine 

 are mixed by agitation, and then quickly poured out on to the steri- 

 lised plate, the glass cover being immediately replaced. The whole 

 operation is so managed that the time of exposure to the air is 

 reduced to a minimum. A solution of mercuric chloride is then 

 poured into the porcelain tray as already described, and when the 

 gelatine has set, which generally takes place in about ten minutes, 

 the dishes are placed in a cupboard, and maintained at a temperature 

 of 20—25° C. 



After a period of incubation, varying from 3*- — 6 days, the organisms 

 make their appearance in isolated colonies, which may be readily 

 counted with the assistance of a lens. The operation of counting is 

 greatly facilitated by placing the plate on a black ground, ruled in 

 squares. 



That when due precautions are taken in the execution of this pro- 

 cess little or no appreciable error is introduced by the unavoidable 

 but momentary contact of the gelatine with the air, was proved by 

 making blank experiments, in which all the abovementioned opera- 

 tions, excepting the addition of water, were performed, and in these 

 cases no organisms were found. Additional proof of this is also fur- 

 nished by the fact that on several occasions no organisms were found 

 in the course of the experiments to be described below. 



The experiments were in nearly all cases made in duplicate, the 

 concordance in the results of parallel experiments being, on the whole, 

 very satisfactory ; occasionally, however, wide discrepancies did occur, 

 but these could in most cases be accounted for through the water 

 under examination not having been rendered sufficiently homogeneous 

 by agitation. 



Experiments on the Filtering Power of Different Substances. 



The substances selected for experiment were natural greensand, 

 silver sand, powdered glass, brickdust, coke, animal charcoal, and 

 spongy iron. 



These substances were all obtained in a fine state of division by 

 powdering them in a mortar, and then passing them through a sieve 

 (40 meshes to the inch). The filters were constructed of pieces of 

 glass tubing (1 inch diameter), drawn off at one extremity to a small 

 aperture ; the latter was plugged with a small quantity of asbestos, 

 and upon this was placed a column 6 inches in height of the closely 

 packed filtering material, the surface of which was again protected by 



