Davy — On Arsenical Co7npounds. 229 



fication, I have found that tartaric acid answers very "well. As the 

 presence of alkalies in solution do not interfere with the evolution of 

 the arsenical gas, this is itself a means of distinguishing the two metals, 

 arsenic and antimony. 



But it may be occasionally necessaiy to determine whether the 

 effects observed on the paper moistened with nitrate of silver are due 

 to arsenic or to antimony. There are different methods by which we may 

 determine this question ; but the one I have found the simplest and 

 on the whole, the most satisfactory, is to digest the paper stain in sul- 

 phide of ammonium, when the arsenic or antimony present will be con- 

 verted into a sulphide, and dissolved by the excess of the alkaline salt, 

 leaving the silver sulphide undissolved, and adhering pi-incipally to the 

 paper — the alkaUne solution, on being evaporated to dryness, will, in the 

 case of arsenic, leave a bright yellow residue almost insoluble in 

 hych'ochloric acid, whereas in the case of antimony, an orange one will 

 remain, which readily dissolves in that acid, at least on the application 

 of heat. 



Before concluding, I wish to observe, that according to some ex- 

 periments recently made by Dr. Eussell, it appears that hydi'ogcn alone 

 is capable of reducing solutions of nitrate of silver to the metallic 

 state ; but this action, even from his observations, is an exceedingly 

 slow one, and takes place to a very minute extent in dilute solutions. 

 On the other hand, M. H. Pellet maintains, that hydi-ogen carefully 

 fi'eed fi'om acid and arsenic, by passing it through solutions of soda and 

 of nitrate of silver, has no action on that salt at the ordinary tempera- 

 ture. But he states, that nitrate of silver which has been fused possesses 

 an alkaline reaction in solution, and that a slight precipitate is produced 

 in such by pure hydi'ogen ; if, however, he observes, a di-op or two of 

 nitric acid be added, then notlaing is precipitated. Be this as it may, as 

 regards the reducing action of pure hydrogen, I found in an experiment 

 I made, that hydrogen which had been passed tlu'ough solutions of 

 caustic soda, and of nitrate of silver, and was afterwards brought in 

 contact with a porcelain crucible cover, moistened with the dilute and 

 acidulated solution of nitrate of silver already noticed, produced only 

 the faintest possible effect, even after several hours' exposure to a 

 stream of this gas, and this very slight action might possibly be due to 

 the hydi'ogen not being perfectly freed from its impurities. Conse- 

 quently, it is veiy doubtful that any reduction of the silver salt from 

 the hydrogen alone will occur under the circumstances of the proposed 

 test. Finally, I must observe, that where paper moistened with the 

 silver solution is used to detect arsenic or antimony, we must bear in 

 mind that nitrate of silver will alone, after some time, blacken the 

 paper, especially if it is exposed to the light; but this gradual 

 change which is so produced is very unlike the rapid effect that takes 

 place where either arseniuretted or antimoniuretted hydrogen acts on 

 paper moistened with that silver salt. 



