" place of 72 parts sulphate of soda in effecting the decomposition of 130 parts nitrate of baryta. That the quantity of sulphuric acid necessary to produce by decomposition 116 parts of sulphate of baryta, would have produced 68 parts sulphate of lime; and carrying the examination further, a circular nexus of numeral alliances would have been rendered manifest. This discovery was reserved for Richter of Berlin; who, extending his investigations to numerous compounds, drew up tables of decomposing proportionals, and in his "Geometry of the Chemical Elements," published between the years 1792 and 1802, arranged a series of tables showing the weight of each base capable of saturating 100 parts of each acid, and also the weight of each acid capable of saturating 100 parts of each base. He observed that in all these tables the bases and the acids followed the same order; and further, that the numbers in each table constituted a series of which the numbers had the same ratio to each other. Thus Richter had succeeded in expanding to a considerable extent the first discovery of Wenzell, but he did not succeed in elevating it into a general law, involving the whole scope of chemical combination. For the list of substances operated upon, Richter demonstrated the existence of numeral decomposing proportionals, but he did not seem to have apprehended the universality of proportionalism which it has been the province of subsequent chemists to determine, and on the recognition of which hinge some of our most beautiful systems of chemical analysis. The proportionality of chemical composition and decomposition attracted but little notice from the time of Richter until between the years 1803* and 1808,† when Dalton, speculating on the rational consequences of the facts made known by chemical decomposition, first promulgated a theory of atoms, based upon arguments no less rational and intelligible than the theory of Epicurus was irrational and obscure:-a theory which carries with it so many elements of conviction, that a celebrated modern chemist has concluded his arguments for and against the existence of atoms by the remark, “that whether matter be atomic or not, thus much is certain, that granting it to be atomic, it would appear as it now does." The arguments adduced by Dalton in favour of the existence of atoms, were such as these:-seeing that all marked cases of chemical combination can be demonstrated always to take place in definite proportion, and that by inference, a similar proportionality may be supposed to extend to less marked cases-seeing that these definite proportions of bodies entering into combination are mutually proportional amongst themselves, it follows that such definite immutability, such mutual proportionality, should most rationally be considered as indicating a ponderable ratio between combining elements; and that the ratio never changing would seem to be indicative of elementary ponderable molecules of determinate *Manchester Phil. Trans. for 1803. + Dalton's New System of Chemical Philosophy. relative weight, unchanging, indivisible; qualities which will be recognised as fulfilling the definition of an atom. With the view of practically illustrating the train of reasoning by which Dalton arrived at the conclusion that matter should be atomic, it will be well to fix on the exemplification afforded by some particular combining series. For this purpose none is more convenient than the one furnished by the oxygen compounds with nitrogen, of which there are five, as displayed by the accompanying table.* A glance at this table will show the justice of the remark of M. Dumas,—that, granting matter to be * It would nevertheless be unjust to the late Mr. Higgins, of Dublin, were the circumstance not mentioned that he had entered upon certain trains of reasoning which, if persevered in, must have conducted him to the discovery of Dalton. Mr. Higgins, in his comparative views of the phlogistic and antiphlogistic theories, published in 1789, states, p. 36 and 37, that in volatile vitriolic acid, a single ultimate particle of sulphur is intimately united only to a single particle of dephlogisticated air; and thus, in perfect vitriolic acid, every single particle of sulphur is united to two of dephlogisticated air, being the quantity to saturation. A similar train of reasoning is applied by Mr. Higgins to the constitution of water, and the compounds of nitrogen and oxygen. atomic, it must necessarily combine as it is found to do. If, for instance, the ultimate molecule of nitrogen should weigh 14, that of oxygen weighing 8, then it should necessarily follow that, in a series of oxygen combinations with nitrogen, in which the amount of nitrogen remains fixed, the ratio of progressive increment of oxygen should be a ratio having for its arithmetical basis the number eight, of which every subsequent term should be a multiple by a whole number. When we come to examine the volume as well as the weight of bodies entering into chemical combination, a definite order of progression will again be recognisable. Taking for instance the same five compounds of nitrogen and oxygen, and expressing by a diagram the rates of their combining volumes, the following result will be manifested : From the inspection of this diagram it will be evident that a ratio of measure as well as a ratio of weight exists between the combinations of oxygen and nitrogen ; and the same remark applies to every combining series, = the elements of which can be obtained in the gaseous form, and their volumes estimated. It is evident that no indication afforded by the phenomena of definite chemical combination, can ever give us more than the ratio subsisting between the weight, and occasionally between the measure of bodies entering into chemical union. Granting, for the sake of argument, the postulate of the existence of atoms, they must be so exceedingly small that no one can hope to see them, or to have any other direct manifestation of their presence; hence the ratio only of their combining weight, and combining volumes, is all that phenomena of composition and decomposition will enable us to recognise. But the ratio subsisting between the dimensions of atoms involves some curious points of speculation. Referring to the preceding table, in which the ratio of combining volumes between compounds of oxygen and nitrogen is indicated, it will be observed that the assumed atomic size of oxygen is set down as half that of nitrogen; accordingly, starting with the assumption of the atomic size of oxygen being half that of nitrogen, the first member in the series of compounds of oxygen and nitrogen is considered to be a combination of atom to atom; but it would not be inconsistent with rational hypothesis, so far as facts of D |