122 J. Scully — On the Effects produced hy small quantities of [No. 2, 



the circumstance that some silver bullion, in the shape of English re- 

 fined bars of as high a fineness as 990 per mille, proved so brittle as to 

 be unfit for mintage. Attention was first attracted to this matter by 

 the peculiar behaviour under assay of the granulated samples taken 

 from this silver after melting. The appearances noticed under assay 

 will be referred to presently, but they led to the bullion being at once 

 tested for brittleness. A bar, about 21 inches long, 2*25 broad, and 1 

 inch thick, was hammered out at one end without cracking, but on 

 being passed through the rolls it cracked badly at the edges and was 

 pronounced to be " brittle," in the Mint sense of the term. The bullion 

 was then remelted in five plumbago pots, and a partial refinement of it 

 attempted in the ordinary way with nitre, about eight to ten pounds of 

 this salt being used for each pot. The resulting silver bars were not 

 appreciably improved by this treatment ; hammering again proved an 

 inconclusive test, but a bar of the size I have mentioned broke in two 

 by merely dropping on the floor of the melting room. 



In the meantime the assay had shown that the brittle bullion con- 

 tained bismuth, and that this was the only substance present likely to be 

 the cause of brittleness. The Indian process of assaying silver has been 

 described by Dr. Busteed in the Journal of the Asiatic Society of 

 Bengal (1870, Part II. p. 377) ; and a brief abstract of this paper is given 

 on p. 292 of Dr. Percy's work before mentioned. The main features of 

 the process may here be briefly recapitulated for the purpose I have in 

 view. A fixed weight of the silver ballion to be assayed is dissolved in an 

 assay bottle, by means of nitric acid aided by heat ; the solution is diluted 

 with water and an excess of hydrochloric acid is added, to precipitate all 

 the silver present as chloride. The silver chloride having been caused 

 to aggregate and settle by vigorous shaking, the bottle is filled up with 

 water and the supernatant fluid is subsequently syphoned off, to remove 

 all the now dissolved matter which may have been contained in the 

 bullion. Under these conditions of solution, precipitation and dilution 

 with water, chemists will readily understand that even a small trace 

 of bismuth, if it be in the silver, will reveal its presence by the 

 partial formation of insoluble oxychloride of bismuth. Now, in the 

 assay of the brittle bullion under consideration, solution in nitric 

 acid had been readily and completely effected by the aid of heat : 

 antimony and tin were consequently absent. After the addition of water 

 and hydrochloric acid, however, the solution in the assay bottles could 

 not be cleared by shaking ; the bulk of the silver chloride collected at 

 the bottom of the bottles, but the supernatant fluid remained turbid. 

 Tin and antimony being excluded, only two metals could produce this 

 result in the wet assay of silver, namely, mercury and bismuth. To 



