in the Presence of Bismuth, Antimony, eto. 143 



of the copper precipitated. Determination 12 of the table 

 was treated in this way, but as the tin present was much 

 more than equivalent to the copper, some precipitation of 

 sulphur probably took place, as the high result would indicate. 

 In all such cases it would be better to convert the tin into the 

 higher condition by some suitable oxidizing agent before pre- 

 cipitating the copper. Where the weight of copper present is 

 not desired, but merely a separation between the copper and 

 tin, this precaution is of course unnecessary. 



Experiments 13 and 14, with arsenic and antimony respec- 

 tively, show that for these metals the conditions which gave 

 good results with bismuth and tin are equally favorable. 



In experiments 15 and 16 arsenic, bismuth, antimony and 

 tin were all present to the extent of 0*1 grm. of each metal, 

 and the results for copper are still accurate. 



If cuprous sulphocyanide is precipitated in the presence of 

 a moderate amount of free acid, using a decided excess of 

 ammonium sulphocyanide, as was done in the experiments of 

 Table II, the tendency of the precipitate to pass through the 

 filter during the washing may be almost completely overcome. 

 It is, however, more likely to appear the more rapidly the pre- 

 cipitation has taken place, that is, the smaller the effective 

 amount of acid present. This tendency, in. the experience of 

 the writer, need not greatly interfere with the accuracy of the 

 determination. If only a light pressure of the filter pump is 

 used, the washing may generally be completed and the small 

 quantity of copper thus lost disregarded. A very distinct cloudi- 

 ness in the filtrate may be caused by an amount of copper not 

 greater than one or two tenths of a milligram, and the loss 

 from this cause will seldom exceed this limit. Although, on 

 the other hand, by precipitating in strongly acid solutions 

 cuprous sulphocyanide can be thrown down in a form that 

 filters well, the losses from incomplete precipitation may be 

 many times greater. 



The filtrates from determinations 15 and 16, the first of 

 which showed a distinct cloudiness, were treated with great 

 care for the recovery and estimation of the trace of copper 

 present. The colorimetric method described in a previous 

 article* was used in determining the copper, and the amounts 

 found were respectively 0*00034 and 0*00020 grm., these 

 figures representing the sum of the mechanical loss with that 

 due to the solubility of the precipitate in the acid solution. 



Under the conditions of Table II, the estimation of copper 

 in the presence of arsenic, bismuth, antimony and tin, either 

 separately or in any combination, is obviously possible. To 



* This Journal, vol. xiii, 22. 



