OF ARTS AND SCIENCES. 



179 



oxide in a similar manner, except that the sulphate was crystallized a 

 greater number of times. In this connection it may be mentioned that, 

 of many samples of German cupric oxide tested, not one was found 

 which did not contain a comparatively large amount of arsenic. In 

 the case of many samples, after several reductions with pure hydrogen 

 and oxidations, the arsenic will actually sublime off as arsenious oxide ; 

 and a quantity of the substance was collected in this manner. 



The method of cleaning the copper was similar to that previously 

 adopted ; it was treated in succession with dilute potassic hydrate, 

 dilute sulphuric acid, and a very large amount of water, and then dried 

 and reduced by pure hydrogen. 



The silver which was obtained by precipitation from the pure 

 argentic nitrate was first washed and dried at 150°, and weighed, as 

 before ; and was then heated to incipient redness, and weighed again. 

 The loss of weight by this process varied with the different samples 

 between three tenths of a milligram and one milligram. The Gooch 

 crucible and asbestos mat subjected to the same treatment did not 

 lose an appreciable quantity. The results were calculated for the 

 weight of silver both before and after ignition, and it will be noticed 

 that the first column of results corresponds almost exactly to the re- 

 sults given in the previous paper. The silver was in each case tested 

 for copper, and no trace was found. 



Results. 

 German Copper. 



