ASSAYING. 131 



was, however, not proportional to the weight of the lead, as has been explained 

 before. The difference, however, between buttons weighing 300 grains and 

 those weighing 500 grains never exceeded 2 cents (0.000053 per cent), and 

 it is therefore safe to say that when the lead button did not vary more than 

 25 grains either way from 425 grains the possible difference could not ex- 

 ceed one cent (0.0000265 per cent). Allowing 2 cents (0.000053 per cent), 

 or plus or minus one cent for inaccuracies in weighing, the total amount of 

 all the possible inaccuracies can be reckoned at 3 cents (0.0000795 per 

 cent), or plus or minus 1 J cents. 



Resum* of errors. — The possible errors in the silver assay as it has been 

 described ai-e the following: Inaccuracies in weighing the "pulp"; imper- 

 fect fluxing; insufficiency of the time of melting; impurity of the litharge; 

 loss by cupellation; mechanical losses; and inaccuracies in weighing the 

 silver button. All these errors, with the exception of those caused by silver 

 in the litharge and the inaccuracies in weighing the silver burton, are so 

 infinitesimal, when the assay has been properly conducted, that they may be 

 neglected altogether. The other two sources of error, the litharge and the 

 balance together, cannot change the results more than three cents, and the 

 influence of the latter of these can be very much reduced by the substitution 

 of a microscope with a micrometer eye-piece for the balance. 



Estimation of gold. — The determination of the amount of gold in any country 

 rock where it is present in extremely small quantities is attended with great 

 difficulties. It is scarcely ever as much in value as the silver and is always 

 very much less in quantity. It is only by the concentration of a large 

 number of assays that it can be determined at all, and the results even of 

 this method are not always reliable. It is impossible to obtain litharge free 

 from gold as well as silver, and it is much more difficult to determine its 

 quantity or the effect that it has upon the assay. As nearly as could be 

 determined, the amount of gold in the litharge used was about one cent to 

 the ton of 2,000 pounds (0.0000016 per cent.). This result was obtained in 

 the following manner: Twenty assays of the ordinary flux were reduced in 

 the usual way, and the resulting lead buttons were separately cupelled until 

 the lead remaining would weigh about twenty grains. The cupels were then 

 removed from the fire and allowed to cool. The twenty lead buttons, which 



