72 MR. H. A. SMITH ON ARSENIC IN 



to the fault of the analysts. It was with this idea that I 

 worked out Part 2 of the same Table, which I think some- 

 what explains the fact. 



The difference in the numbers found is certainly large ; 

 but, as I said before, I find them borne out in a great 

 measure by other analyses which I will shortly mention. 



The purest pyrites is the Belgian, containing only 0*9437 

 per cent. As ; but this carried along with it the very great 

 inconvenience (which also became an expense) that, in 

 breaking for the kilns, a large amount of " smalls " accumu- 

 lated ; and it was therefore put aside so as to make room 

 for that containing the next smallest amount of arsenic, 

 the Hard Norwegian, which had the great advantage of 

 making little or no " smalls." This contained a decidedly 

 larger amount of arsenic than the last, namely 1*6490 per 

 cent. As. The increase found by laboratory analysis was 

 also corroborated by the practical working. 



In Table II. I have endeavoured to show how this 

 amount of arsenic is distributed among the separate pro- 

 ducts of manufacture. 



After this pyrites (Norwegian), containing originally 

 1*649 P er cent. As, has been burnt in the kilns, we find the 

 amount reduced to 0*465 per cent. As, leaving thus 1*184 

 per cent. As to be accounted for. As is naturally expected, 

 the greatest quantity of this finds its way into the sulphuric 

 acid, the amount there present being 1*051 per cent. As; 

 so that 0*133 per cent. As has been lost between the kilns 

 and the chamber. 



In a flue leading from the kilns to the lead chamber 

 a yellowish white, partly crystalline, deposit was found, 

 consisting of arsenious and arsenic acids mixed with 

 sulphur in a viscous state. This, on analysis, gave 46*360 

 per cent, of arsenic. Unluckily this number becomes 

 of very little use to us here, as this was a flue that had 

 been in use for a very long time ; so that we have an 



