1. When a mineral is subjected to analysis, whatever care we take to collect all the constituents, and to weigh them without losing any portion whatever, it is generally found that the sum of the constituents obtained fall a little short of the weight of the mineral employed in the analysis. Thus, if we take 100 grains of any mineral, and analyze it, the weights of all the constituents obtained added together will rarely make up 100 grains, but generally somewhat less; perhaps only 99, or even 98 grains. But some cases occur, when the analysis of 100 grains of a mineral gives us constituents that weigh, when added together, more than 100 grains; perhaps 105, or, in some rare cases, as much as 110. It was the custom with Bergman, and other analysts of his time, to consider this deficiency or surplus as owing to errors in the analysis, and therefore to slur it over in the statement of the analysis, by bringing the weight of the constituents, by calculation, to amount exactly to 100 grains. Klaproth introduced the method of stating the results exactly as he got them. He gives the weight of mineral employed in all his analyses, and the weight of each constituent extracted. These weights, added together, generally show a loss, varying from two per cent. to a half per cent. This improvement may appear at first sight trifling; yet I am persuaded that to it we are indebted for most of the subsequent improvements introduced into analytical chemistry. If the loss sustained was too great, it was obvious either that the analysis had been badly performed, or that the mineral contains some constituent which had been overlooked, and not obtained. This laid him under the necessity of repeating the analysis; and if the loss continued, he naturally looked out for some constituent which his analysis had not enabled him to obtain. It was in this way that he discovered the presence of potash in minerals; and Dr. Kennedy afterwards, by following out his processes, discovered soda as a constituent. It was in this way that water, phosphoric acid, arsenic acid, fluoric acid, boracic acid, &c., were also found to exist as constituents in various mineral bodies, which, but for the accurate mode of notation introduced by Klaproth, would have been overlooked and neglected.