88 



THE ALUMNI JOURNAL. 



to be best to use equal volumes of speci- 

 men and reagent, although the variable 

 amount of dark coatings of reduced ar- 

 senic adhering to the tin prevented an 

 accurate comparison. 



3. INFLUENCE OF THE USE OF METALLIC 



TIN TOGETHER WITH STANNOUS 



CHLORIDE. 



When metallic tin is heated with pure 

 concentrated hydrochloric acid, a copi- 

 ous evolution of hydrogen results. When 

 heated with the saturated solution of 

 stannous chloride in hydrochloric acid, 

 the evolution of hydrogen is but scant. 

 When arsenic is mixed with the solution, 

 the evolution of gas is perceptible, but 

 very feeble. Very little arsine appears to 

 be evolved, for a paper cap with a drop 

 of acidulated silver nitrate does not show 

 any evidence of reduction by As H 3 for 

 over 15 minutes, and even at the end 

 of an hour but a very faint arsenic reac- 

 tion is seen. To ascertain whether any 

 of the arsine formed, would react with the 

 stannous chloride and thus hasten the 

 reduction of As, I arranged a small ap- 

 paratus in which a copious current of 

 arsin mixed with hydrogen was generat- 

 ed, and, after passing through a bottle 

 filled with dry cotton, was permitted to 

 bubble through stannous chlorid solution 

 of the U- S P. After more than an hour's 

 time, not a trace of coloration could be 

 detected. 



So that after arsine has once been form- 

 ed, it is not again decomposed by stan- 

 nous chloride. Hence the prompt action 

 of metallic tin must depend upon the 

 nascent hydrogen, which aids the stan- 

 nous chloride in reducing the trioxide and 

 pentoxide. 



But however satisfactory the action of 

 metallic tin in accellerating and intensi- 

 fying the reduction ot arsenic, it cannot 

 be employed to detect arsenic in prepara- 

 tions of bismuth or antimony. For the 

 metallic tin reduces both of those metals 



and precipitates them from their solution 

 as black flocculi, which aggregate into 

 small granular lumps. There could be a 

 distinction made between the firm floc- 

 culi of Sb or Bi and the finely divided 

 brown particles of arsenic, but it would 

 hardly be safe to trust to this appearance 

 for a sufficient evidence of the presence 

 of arsenic, and the preparations of Bi and 

 Sb must be tested without the aid of me- 

 tallic tin. 



A number of specimens of pure bis- 

 muth and antimony salts were tested 

 with the U. S. P. and the German solu- 

 tion of stannous chloride in various pro- 

 portions. But neither at ordinary tem- 

 perature, nor when heated did the least 

 coloring occur. 



4. INFLUENCE OF THE PRESENCE OF 



OTHER SUBSTANCES IN THE 



SPECIMEN TESTED. 



The preceding experiments were made 

 with a solution of either arsenic trioxide 

 or sodium arsenite or arsenate in water, 

 other substances being absent. But in 

 testing various chemicals for traces of 

 arsenic there are different conditions, as 

 the arsenic forms only a very small por- 

 tion of the mixture. In the U. S. P. the 

 Bettendorf test is directed for 1 1 prepara- 

 tions : For hydrobromic, hydrochloric, 

 phosphoric and sulphuric acid, 1 Cc. of 

 the acid is to be tested by mixing with 1 

 Cc. of the reagent, adding a small piece 

 of tinfoil and heating. In case of mag- 

 nesium sulphate 1 Gm. of the dry sub- 

 stances is to be shaken with 3 Cc. of the 

 reagent, metallic tin is then to be added 

 and an hour allowed for the appearance 

 of the reaction ; sodium phosphate and 

 pyrophosphate are treated in like man- 

 ner, but heat applied and 15 minutes 

 time given. 



In case of antimony and potassium tar- 

 trate, antimony oxide, bismuth subcar- 

 bonate and bismuth subnitrate, it was 

 intended to test with stannous chloride 



