1836.] On Assaying Silver. fg 



were thought necessary to produce accuracy of result. The 

 notation, &c. is the same as before. 



+ 2 



S 



— 1 



— 2 



— 2 



— 2 



— 1 



+ 6 



+ 1 



— i 



— 2 



— 1 



— 1 



S 



+ 1 



4- 3 

 — 1 



— 1 

 + 1 



S 



+ i 



+ 4 

 — 1 



S 

 + 1 



S 



+ 5 



+ 3 



— 1 



— I 



+ 6 



+ 4 



+11 



? 6 



+ 5 



+ 15 



+ 12 



+ 13 + 11 



? 



—11 



-12 



— 9 



—12 



— 9 



— 1 



— 7 



— 8 



— 1! 



z™ 



— 8 



—10 



— 7 



— 7 



— 7 



— 7 



— 7 



— 7 



— 7 



— 7 



— 3 



— 6 



— 6 



— 6 



S 



S 



S 



S 



S 



— 2 



— 1 



— 3 



s 



— 1 



S 



+ 7 







S 



s 



s 



+ 3 



Front. 



Front. 



10. The examination of these experiments will shew 

 Considerable discordancies s and as discordancies occur with 

 silver of known fineness, they will of course occur equally 

 with silver of unknown fineness. In these particular ex- 

 periments the discrepancies can be traced, because the 

 silver was all of the same known quality ; but it is evident 

 they cannot be so traced when bullion of unknown standard 

 is to be experimented on, and its true valile to be ascertain- 

 ed. It may therefore be asked how is the true quality of the 

 silver to be known, seeing the capricious results produced by 

 this mode of assay ? From the variable out-turn of the fur- 

 nace, its irregularities would indeed be fatal to all practical 

 precision were they not brought under some controul by a 

 practical expedient, which is, that every time an assay, or 

 set of assays, is made, one or two musters of standard silver 

 are placed in the muffle with the silver under examination, 

 and according as the standard or guide musters come out 

 above or below their true fineness, which is known, an allow- 

 ance is made for the other musters,' considering them to be 

 finer or coarser according to the indications of the standard 

 musters. And by this ingenious method of check the re- 



