Bergman had made a set of experiments to determine the proportional quantities of phlogiston contained in the different metals, by the relative quantity of each necessary to precipitate a given weight of another from its acid solution. It was the opinion at that time, that metals were compounds of their respective calces and phlogiston. When a metal dissolved in an acid, it was known to be in the state of calx, and therefore had parted with its phlogiston: when another metal was put into this solution it became a calx, and the dissolved metal was precipitated in the metallic state. It had therefore united with the phlogiston of the precipitating metal. It is obvious, that by determining the quantities of the two metals precipitated and dissolved, the relative proportion of phlogiston in each could be determined. Lavoisier saw that these experiments of Bergman would serve equally to determine the relative quantity of oxygen in the different oxides. Accordingly, in a paper inserted in the Memoirs of the Academy, for 1782, he enters into an elaborate examination of Bergman's experiments, with a view to determine this point. But it is unnecessary to state the deductions which he drew, because Bergman's experiments were not sufficiently accurate for the object in view. Indeed, as the mutual precipitation of the metals is a galvanic phenomenon, and as the precipitated metal is seldom quite pure, but an alloy of the precipitating and precipitated metal; and as it is very difficult to dry the more oxidizable metals, as copper and tin, without their absorbing oxygen when they are in a state of very minute division; this mode of experimenting is not precise enough for the object which Lavoisier had in view. Accordingly the table of the composition of the metallic oxides which Lavoisier has drawn up is so very defective, that it is not worth while to transcribe it. The same remark applies to the table of the affinities of oxygen which Lavoisier drew up and inserted in the Memoirs of the Academy, for the same year. His data were too imperfect, and his knowledge too limited, to put it in his power to draw up any such table with any approach to accuracy. I shall have occasion to resume the subject in a subsequent chapter. In the same volume of the Memoirs of the Academy, this indefatigable man inserted a paper in order to determine the quantity of oxygen which combines with iron. His method of proceeding was, to burn a given weight of iron in oxygen gas. It is well known that iron wire, under such circumstances, burns with considerable splendour, and that the oxide, by the heat, is fused into a black brittle matter, having somewhat of the metallic lustre. He burnt 145·6 grains of iron in this way, and found that, after combustion, the weight became 192 grains, and 97 French cubic inches of oxygen gas had been absorbed. From this experiment it follows, that the oxide of iron formed by burning iron in oxygen gas is a compound of Iron 3·5 Oxygen 1·11