488 Dr. Andrew Fyfe on the Detection of Arsenic. 



not only from non-volatile chlorides, but also from those metala 

 the chlorides of which are easily volatilized. That arsenic might 

 be volatilized in this way, and be thus detected, occurred to Dr. 

 Clarke, Professor of Chemistry in Marischal College, Aberdeen, 

 some years ago, but, so far as I know, he has not prosecuted the 

 subject. My attention was lately drawn to it more particularly, 

 by the perusal of a paper, in which processes are described 

 for the separation of metals from one another, in analysing 

 their ores and alloys*. The perusal of that paper induced me 

 to take up the consideration of the subject, and the results of 

 the experiments which I have made have confirmed the opinion 

 I had entertained, that the volatilization of arsenic, by the action 

 of chlorine, as suggested by Professor Clarke, would add another 

 to the many tests which we have of the presence of arsenic. 



The question at first occurred. How is arsenic, in the state of 

 arsenious acid, to be made to unite with chlorine ? This was sug- 

 gested by the results of the trials recorded in the paper referred to, 

 viz. causing the decomposition of a chloride in solution in the fluid 

 containing the arsenic. For this purpose the fluid is heated with 

 oil of vitriol in a flask, to which a bent tube is adapted, and by 

 which the product of distillation is conveyed into a cool receiver. 

 When the mixture is brought to the boiling-point, a little dried 

 sea salt is thrown in, and the distillation is continued for some 

 time. By this process the hydrochloric acid evolved yields its 

 hydrogen to the oxygen of the arsenious acid, while the liberated 

 arsenic and chlorine unite, and come over as chloride, which 

 is condensed in the receiver. 



(As 0^ + 3S0'^ HO -h 3C1 Na=3S03 NaO + 3H0 + As CF). 

 We thus obtain a transparent colourless liquid, in which arsenic 

 is easily detected by the transmission of sulphuretted hydrogen, 

 which gives the yellow sulphuret. By the addition of nitrate 

 of silver in excess to the product of distillation, a copious white 

 deposit of chloride of silver is formed. On filtering this fluid, 

 and then adding ammonia solution, continuously ^ or, which I pre- 

 fer, by holding over it a rod dipt in the ammoniacal solution, so 

 as to avoid excess, the yellow arsenite of silver is either precipi- 

 tated or appears on the surface as a yellow film, according to its 

 quantity. 



Though this method of detecting arsenic acts easily with pure 

 fluids in which arsenious acid only is dissolved, it was necessary to 

 ascertain whether it would act with mixed fluids, particularly 

 those containing organic matter. A number of these was ac- 



* So far as I remember, the paper alluded to appeared in one of the 

 Foreign Journals — I think the Comptes Rendus — two or three years ^o, 

 but, though I have again and again searched for it, I have not succeeded 

 in discovering it. 



