In the year 1808 Mr. Dalton published the first volume of his New System of Chemical Philosophy. This volume consists chiefly of two chapters: the first, on heat, occupies 140 pages. In it he treats of all the effects of heat, and shows the same sagacity and originality which characterize all his writings. Even when his opinions on a subject are not correct, his reasoning is so ingenious and original, and the new facts which he contrives to bring forward so important, that we are always pleased and always instructed. The second chapter, on the constitution of bodies, occupies 70 pages. The chief object of it is to combat the peculiar notions respecting elastic fluids, which had been advanced by Berthollet, and supported by Dr. Murray, of Edinburgh. In the third chapter, on chemical synthesis, which occupies only a few pages, he gives us the outlines of the atomic theory, such as he had conceived it. In a plate at the end of the volume he exhibits the symbols and atomic weights of thirty-seven bodies, twenty of which were then considered as simple, and the other seventeen as compound. The following table shows the atomic weight of the simple bodies, as he at that time had determined them from the best analytical experiments that had been made:
| Weight of atom. | Weight of atom. | |||||
| Hydrogen | 1 | Strontian | 46 | |||
| Azote | 5 | Barytes | 68 | |||
| Carbon | 5 | Iron | 38 | |||
| Oxygen | 7 | Zinc | 56 | |||
| Phosphorus | 9 | Copper | 56 | |||
| Sulphur | 13 | Lead | 95 | |||
| Magnesia | 20 | Silver | 100 | |||
| Lime | 23 | Platinum | 100 | |||
| Soda | 28 | Gold | 140 | |||
| Potash | 42 | Mercury | 167 | |||
He had made choice of hydrogen for unity, because it is the lightest of all bodies. He was of opinion that the atomic weights of all other bodies are multiples of hydrogen; and, accordingly, they are all expressed in whole numbers. He had raised the atomic weight of oxygen from 6·5 to 7, from a more careful examination of the experiments on the component parts of water. Davy, from a more accurate set of experiments, soon after raised the number for oxygen to 7·5: and Dr. Prout, from a still more careful investigation of the relative specific gravities of oxygen and hydrogen, showed that if the atom of hydrogen be 1, that of oxygen must be 8. Every thing conspires to prove that this is the true ratio between the atomic weights of oxygen and hydrogen.
In 1810 appeared the second volume of Mr. Dalton's New System of Chemical Philosophy. In it he examines the elementary principles, or simple bodies, namely, oxygen, hydrogen, azote, carbon, sulphur, phosphorus, and the metals; and the compounds consisting of two elements, namely, the compounds of oxygen with hydrogen, azote, carbon, sulphur, phosphorus; of hydrogen with azote, carbon, sulphur, phosphorus. Finally he treats of the fixed alkalies and earths. All these combinations are treated of with infinite sagacity; and he endeavours to determine the atomic weights of the different elementary substances. Nothing can exceed the ingenuity of his reasoning. But unfortunately at that time very few accurate chemical analyses existed; and in chemistry no reasoning, however ingenious, can compensate for this indispensable datum. Accordingly his table of atomic weights at the end this second volume, though much more complete than that at the end of the first volume, is still exceedingly defective; indeed no one number can be considered as perfectly correct.
The third volume of the New System of Chemical Philosophy was only published in 1827; but the greatest part of it had been printed nearly ten years before. It treats of the metallic oxides, the sulphurets, phosphurets, carburets, and alloys. Doubtless many of the facts contained in it were new when the sheets were put to the press; but during the interval between the printing and publication, almost the whole of them had not merely been anticipated, but the subject carried much further. By far the most important part of the volume is the Appendix, consisting of about ninety pages, in which he discusses, with his usual sagacity, various important points connected with heat and vapour. In page 352 he gives a new table of the atomic weights of bodies, much more copious than those contained in the two preceding volumes; and into which he has introduced the corrections necessary from the numerous correct analyses which had been made in the interval. He still adheres to the ratio 1:7 as the correct difference between the weights of the atoms of hydrogen and oxygen. This shows very clearly that he has not attended to the new facts which have been brought forward on the subject. No person who has attended to the experiments made on the specific gravity of these two gases during the last twelve years, could admit that these specific gravities are to each other as 1 to 14. If 1 to 16 be not the exact ratio, it will surely be admitted on all hands that it is infinitely near it.
Mr. Dalton represented the weight of an atom of hydrogen by 1, because it is the lightest of bodies. In this he has been followed by the chemists of the Royal Institution, by Mr. Philips, Dr. Henry, and Dr. Turner, and perhaps some others whose names I do not at present recollect. Dr. Wollaston, in his paper on Chemical Equivalents, represented the atomic weight of oxygen by 1, because it enters into a greater number of combinations than any other substance; and this plan has been adopted by Berzelius, by myself, and by the greater number, if not the whole, of the chemists on the continent. Perhaps the advantage which Dr. Wollaston assigned for making the atom of oxygen unity will ultimately disappear: for there is no reason for believing that the other supporters of combustion are not capable of entering into as many compounds as oxygen. But, from the constitution of the atmosphere, it is obvious that the compounds into which oxygen enters will always be of more importance to us than any others; and in this point of view it may be attended with considerable convenience to have oxygen represented by 1. In the present state of the atomic theory there is another reason for making the atom of oxygen unity, which I think of considerable importance. Chemists are not yet agreed about the atom of hydrogen. Some consider water a compound of 1 atom of oxygen and 2 atoms of hydrogen; others, of 1 atom of oxygen and 1 atom of hydrogen. According to the first view, the atom of hydrogen is only 1-16th of the weight of an atom of oxygen; according to the second, it is 1-8th. If, therefore, we were to represent the atom of hydrogen by 1, the consequence would be, that two tables of atomic weights would be requisite—all the atoms in one being double the weight of the atoms in the other: whereas, if we make the atom of oxygen unity, it will be the atom of hydrogen only that will differ in the two tables. In the one table it will be 0·125, in the other it will be 0·0625: or, reckoning with Berzelius the atom of oxygen = 100, we have that of hydrogen = 12·5 or 6·25, according as we view water to be a compound of 1 atom of oxygen with 1 or 2 atoms of hydrogen.
In the year 1809 Gay-Lussac published in the second volume of the Mémoires d'Arcueil a paper on the union of the gaseous substances with each other. In this paper he shows that the proportions in which the gases unite with each other are of the simplest kind. One volume of one gas either combining with one volume of another, or with two volumes, or with half a volume. The atomic theory of Dalton had been opposed with considerable keenness by Berthollet in his Introduction to the French translation of my System of Chemistry. Nor was this opposition to be wondered at; because its admission would of course overturn all the opinions which Berthollet had laboured to establish in his Chemical Statics. The object of Gay-Lussac's paper was to confirm and establish the new atomic theory, by exhibiting it in a new point of view. Nothing can be more ingenious than his mode of treating the subject, or more complete than the proofs which he brings forward in support of it. It had been already established that water is formed by the union of one volume of oxygen and two volumes of hydrogen gas. Gay-Lussac found by experiment, that one volume of muriatic acid gas is just saturated by one volume of ammoniacal gas: the product is sal ammoniac. Fluoboric acid gas unites in two proportions with ammoniacal gas: the first compound consists of one volume of fluoboric gas, and one volume of ammoniacal; the second, of one volume of the acid gas, and two volumes of the alkaline. The first forms a neutral salt, the second an alkaline salt. He showed likewise, that carbonic acid and ammoniacal gas could combine also in two proportions; namely, one volume of the acid gas with one or two volumes of the alkaline gas.
M. Amédée Berthollet had proved that ammonia is a compound of one volume of azotic, and three volumes of hydrogen gas. Gay-Lussac himself had shown that sulphuric acid is composed of one volume sulphurous acid gas, and a half-volume of oxygen gas. He showed further, that the compounds of azote and oxygen were composed as follows:
| Azote. | Oxygen. | |
| Protoxide of azote | 1 volume | + ½ volume |
| Deutoxide of azote | 1 " | + 1 |
| Nitrous acid | 1 " | + 2 |
He showed also, that when the two gases after combining remained in the gaseous state, the diminution of volume was either 0, or ⅓, or ½.