Let us consider what knowledge must be acquired before the value to be assigned to the atomic weight of an element can be found.
Hydrogen was the element chosen as a standard by Dalton. He assumed that the atom of hydrogen weighs 1; the atomic weight of any other element is therefore a number which tells how many times the atom of that element is heavier than the atom of hydrogen. Thus, when Dalton said the atomic weight of oxygen is 8, he meant that the atom of oxygen is eight times heavier than that of hydrogen. How was this number obtained?
Accurate analyses of water show that in this liquid one part by weight of hydrogen is combined with eight parts by weight of oxygen; but (it is said) as the atom of hydrogen weighs 1, the atom of oxygen must weigh 8. In drawing this conclusion it is assumed that the atom, or smallest particle, of water is built up of one atom of hydrogen and one atom of oxygen. Let it be assumed that the atom of water contains two atoms of hydrogen and one of oxygen, then the latter atom must weigh sixteen times as much as each atom of hydrogen; let it be assumed that three atoms of hydrogen combine with one atom of oxygen to form an atom of water, then the weight of the oxygen atom must be twenty-four times that of the hydrogen atom. Any one of these assumptions will equally satisfy the figures obtained by analyzing water (1: 8 = 2: 16 = 3: 24). Now, had we any method whereby we could determine how many times an atom of water is heavier than an atom of hydrogen we should be able to determine which of the foregoing assumptions is correct, and therefore to determine the atomic weight of oxygen. Hence, before the atomic weight of an element can be determined, there must be found some method for determining the atomic weights of compounds of that element. Unless this can be done the atomic theory is of little avail in chemistry.
I conceive it to be one of the signal merits of Dalton that he so clearly lays down rules, the best which could be devised at his time, for determining the atomic weights of compounds, or, what is the same thing, for determining the number of elementary atoms in one atom of any compound. In his "New System" he says that he wishes to show the importance of ascertaining "the relative weights of the ultimate particles both of simple and compound bodies, the number of simple elementary particles which constitute one compound particle, and the number of less compound particles which enter into the formation of one more compound particle."
Considering compounds of two elements, he divides these into binary, ternary, quaternary, etc., according as the compound atom contains two, three, four, etc., atoms of the elements. He then proceeds thus—
"The following general rules may be adopted as guides in all our investigations respecting chemical synthesis:—
"1st. When only one combination of two bodies can be obtained, it must be presumed to be a binary one, unless some cause appear to the contrary.
"2nd. When two combinations are observed, they must be presumed to be a binary and a ternary.
"3rd. When three combinations are obtained, we may expect one to be binary and the other two ternary.
"4th. When four combinations are observed, we should expect one binary, two ternary, and one quaternary," etc.