8 PROCEEDINGS OF THE AMERICAN ACADEMY 



" The solubility of the chloride is greatest when in the flaky state, 

 as precipitated in the cold from a sufficiently dilute solution of silver ; 

 the solubility diminishes as the flakes shrink when left to themselves, 

 or as they are rendered pulverulent by long agitation with water. 

 Flaky or pulverulent chloride of silver,, dissolved in water, pure or 

 acidified by nitric acid, is precipitated by the addition of a salt of silver, 

 or of hydrochloric acid, or of an alkaline chloride. . . . The solution 

 of the chloride is wholly effected by pure or acidified water, as the case 

 may be, and is not caused by the soluble salt formed simultaneously 

 with the chloride of silver. The presence of nitric acid in the water 

 does not affect the solubility of flaky chloride of silver ; but it increases 

 the solubility of the pulverulent chloride in proportion to the quantity 

 of acid present. The precipitation of the dissolved chloride is the ex- 

 clusive result of its insolubility in the solution formed by adding an 

 excess either of the silver salt or of the alkaline chloride." 



So also in Liebig and Kopp, Jahresbericht, 1871, 339: '"'Accord- 

 ing to Stas, the granular scaly and crystalline chloride is wholly insol- 

 uble in cold water: in boiling water the solubility is comparatively 

 great, but decreases rapidly with the temperature." 



In our own investigation of this subject we have at once confirmed 

 and extended these observations of Stas, and our results may be of 

 interest as showing that in the very familiar method of determining 

 chlorine by precipitation with nitrate of silver, which is generally sup- 

 posed to be extremely accurate, a sensible error may arise from the 

 solubility of the chloride of silver in the hot distilled water used in 

 washing the precipitate. It would be well for every analyst to make 

 the following very striking experiment, which will enable him to appre- 

 ciate the extent of the action in question. 



Take from five to ten cubic centimetres of pure hydrochloric acid, 

 and precipitate the chlorine in the usual way with nitrate of silver, 

 avoiding a large excess. After pouring off the supernatant liquid and 

 washing the precipitate once or twice with cold distilled water, pour 

 upon the white flaky chloride of silver a comparatively large volume of 

 boiling water. As soon as the precipitate settles, pour off the clear 

 hot water, dividing the solution between two precipitating jars. To 

 one of these add a few drops of a solution of nitrate of silver, and to 

 the other a few drops of hydrochloric acid. In both cases a precipi- 

 tate of chloride of silver will fall, and most chemists, certainly, w'ill be 

 surprised at the effect ; for it is not a mere turbidness that results, but 

 a well-defined precipitate, whose amount is easily estimated. Succes- 

 sive portions of boiling water poured upon the precipitate give the 



