No. 1.] CHEMISTRY. 57 



by Sonstadt for the detection of the gold are exceedingly inter- 

 esting and ingenious. According to the first method, he operates 

 upon 150 or 200 cubic centimetres of water. Two or three 

 decigrammes of ferrous sulphate are dissolved in the water, which 

 is made acid by the addition of two or three drops of hydro- 

 chloric acid. The solution is then heated in a glazed porcelain 

 dish over a small flame, which is so arranged as to touch the 

 under part of the dish, but should not produce ebullition. By 

 this means a lustrous film of ferric oxide is deposited upon the 

 bottom of the dish. The heat is kept up as long as the film 

 increases, and the remaining liquid then poured ofi", the film 

 washed with a little water, and 50 c. c. of strong chlorine water 

 allowed to stand in the dish for an hour or two, and then evapo- 

 rated down to a few drops, a drop of dilute hydrochloric acid 

 being added towards the close of the evaporation. The liquid, 

 which should be nearly colourless, is now poured into a test-tube 

 containing a few drops of a solution of stannous chloride, and 

 after a few minutes the liquid takes a bluish or purplish tint. 

 The reaction is of course more distinct when larger quantities of 

 water are used. Sonstadt says that he has sometimes failed to 

 obtain the film of ferric oxide, but has been most successful when 

 after the addition of the ferrous sulphate and hydrochloric acid 

 to the water he has allowed the solution to stand for some hours 

 exposed to the air. 



In his second method he takes from half a litre to a litre of 

 sea-water, and after adding sufiicient barium chloride to produce 

 about a grain of precipitate, allows the whole to stand for a day 

 or two. The precipitate is then collected, dried, and after mix- 

 ing with borax and lead, treated before the blow-pipe on charcoal 

 and finally cupelled. In this way a yellowish-white button is 

 obtained, having about the colour of an alloy of 60 parts of gold 

 and 40 of silver. For the sake of confirmation, the button may 

 be dissolved in a few drops of aqua regia and the solution evapo- 

 rated nearly to dryness. A few drops of hydrochloric acid are 

 now added, and the solution again evaporated in order to destroy 

 the excess of nitric acid. When nearly to dryness a few drops of 

 water are added, the mixture warmed, and, as soon as the argentic 

 chloride has settled, a drop of solution of stannous chloride 

 allowed to run down the side of the tube into the liquid, when 

 the characteristic gold reaction is obtained. 



The precipitation of the gold by barium chloride is curious^ 



