130 
Copper o 
of accomplishing my purpose* At length, I found a pro- 
cess of Pelletiers, which promised to succeed, and mine 
Is merely extending his idea a little further than he did 
himself. 
He was, I believe, employed by the French govern* 
merit to discover an easy way of separating the tin from 
copper on belbmetal, and the process he gave, is this. 
Upon the melted bell-metal project, black oxyd of man- 
ganese in powder, frequently stirring the metal till all 
die tin becomes oxydated by the manganese. He adds 
a caution, not to add too much manganese otherwise part 
of the copper also will be' destroyed. 
It immediately struck me, that in this way I might be 
able to oxydate the copper which alloys our silver, and 
*ipon making the trial I succeeded completely : I had 
some impure silver rolled out to about the thickness of a 
shilling, this I coiled up spirally, and put into a crucible, 
the bottom of which was covered with black oxyd of 
manganese. I then added more oxyd till the silver was 
covered, and all the space between the coils completely 
Med. A cover was then luted to the crucible, and a 
small hole left for the escape of oxygen gas. When this 
had been exposed for a quarter of an hour to a heat suffi- 
cient to melt silver, I Found the surface of the manganese 
brown from the loss of oxygen ; but, where the silver had 
been, the whole was one uniform black powder, without 
the least appearance of metallic lustre, so that I had no 
doubt, that even die silver was become an oxyd. 
I then put the whole contents of the first crucible into 
a second of a larger size, into the bottom of which I put a 
quantity of pounded green glass, about three times the 
bulk of the contents of the first crucible, and luted on a 
cover as before, to prevent the access of any inflammable 
substance* 
