152 Composition of Watei-s of Land-Drainafjc and of Rain. 



adapted his process ; but, as he makes no mention of it, we must 

 suppose that he did not accomplish its estimation, I shall 

 describe the process as it is practised in the examination of 

 waters ; the modifications necessary for other substances con- 

 tainino: nitrates, will readily occur to the reader. To prevent 

 confusion, it will be supposed, in the first instance, that the 

 water contains little or no organic matter in solution. A pint 

 of the water is introduced into a flask, and rendered alkaline by 

 a few drops of lime-water, so as to avoid all risk of the loss of 

 nitric acid in the subsequent evaporation ; the mouth of the flask 

 is then closed with a cork, furnished with a rather large glass 

 tube, and drawn out to a comparatively small opening, and by 

 means of a lamp it is evaporated to a small bulk. The object 

 of the tube is, by an abundant issue of steam, to prevent the 

 possibility of any circulation in the bottle of the air of the 

 laboratory, which might introduce nitric acid. A stream of 

 carbonic acid is now passed through the liquid, to remove the 

 excess of lime, which is objectionable in the later stages of the 

 process ; and the liquid, having been filtered, is transferred by 

 means of a small funnel to a small globular flask, which is figured 

 at a in the adjoining woodcut. This flask, as well as the rest 

 of the vessels used in the operation, must be able to support a 

 pressure from without of about one atmosphere, but it need nat 

 be at all thick for this purpose. The flasks which I employ are 

 globular, and of about 2i inches in diameter ; they are furnished 

 with necks about 2 inches in length, and -]%- of an inch external 

 diameter. They hold when full about 2000 grains of water, but 

 of course not more than two-thirds of this quantity is evaporated in 

 them at the time ; the flasks weigh, when empty, about 800 grains, 

 which will serve as a guide to their thickness. The concentrated 

 water is nov/ further evaporated in the flask a to perfect dryness ; 

 the last parts of the operation being, for the sake of precaution, 

 performed in an air bath, at a temperature not exceeding 350" 

 Fahrenheit. Into the flask is introduced a small quantity (gene- 

 rally for rain-water about 6 or 8 grains) of pure and dry iodide 

 of silver ; the quantity of this being of course increased in cases 

 where a larger amount of nitric acid is anticipated. The bottle is 

 now connected with the apparatus, as shown in the woodcut. 

 The little bottle, or tube, Z>, upon which it is convenient to have 

 a bulb, is partly filled with strong hydrochloric acid, which must 

 be free from chlorine ; the quantity employed may be from 150 

 to 200 grains. These flasks, as well as all the parts of the 

 apparatus, are connected by short India-rubber tubes, which, if of 

 proper size, are air-tight without being tied. In using them it 

 is necessary to bring the glass-tubes as nearly as possible into 

 contact, by which the India-rubber tube is prevented from collaps- 



