of definite Proportions in Chemical Affinity. 115 
_ Dr. Thomson, in his System of Chemistry, gave an outline of 
Mr. Dalton’s theory , before the latter had himseli made it public, 
and in his paper upon the oxalic acid in the Philosophical Trans- 
actions for 1807 has shown a very great coincidence in the com- 
position of this acid, by caleulation, with the result of actual 
analysis. In the 4th section of this paper he states that “ it has 
been ascertained by decisive experiment, that elementary bodies 
always enter into combinations in determinate proportions, which 
may be represented by.numbers;”’ and he finds an atom of oxy- 
gen to be G, if an atom of hy drogen be estimated as 1. The 
Doctor has taken this estimate by supposing one atom of oxygen 
to unite to one of hydrogen, to form waters estimating the 
weight of hydrogen to be 144 "and oxygen to be 852, to form 
100 parts of water: hence ‘an atom of hydrogen being to one of 
oxygen in the same ratio, it gives when reduced 1 to 6: 48, and 
the Doctor rejected the decimals. Mr. Dalton takes his estimate 
from the proportion given by Humboldt and Gay-Lussac. It 
appears from their experiments tobe 87:4 of oxygen to 12°6 of 
hydrogen; and the number representing oxygen will be 6° 96, which 
approximates very nearly to 7, the number ‘chosen by Mr. Dalton, 
Sir H. Davy estimates the specific gravity of hydrogen to 
oxygen as | to 15; and as it takes 2 measures of hy drogen to | 
of oxygen in the formation of water, the ratio of the hydrogen 
to oxygen is as 2 to 15*; “ and it may be regarded as composed 
of 2 proportions of hy drogen and 1 of oxygen, and the number 
representing hydrogen ‘will be 1, and that representing oxygen 15.” 
This value is not taken by considering the number of ating in 
the gases, a doctrine truly hypothetical, but from the proportions 
in which they unite, and their specific gravity. In considering 
the doctrine of proportion, as Sir H. Davy has observed, we cer- 
tainly have no reason to suppose bedies composed of atoms that 
are indivisible uniting one to one, &c. In the present state of 
cour knowledge this hypothesis of atoms cannot with propriety be 
‘admitted, as we have no evidence upon which we can be certain 
of the weight or number of atoms in any compound. “ Our 
numerical expressions,” as the author last quoted justly says, 
af ought ofily to relate to the result of our experiments.” 
It is not a little perplexing to observe Mr. Dalton,in his work, 
found his whole calculation upon the combination of oxygen and 
hydrogen as uniting atom to atom to form water; and in his 
chapter upon the combination of these gases, to find him de- 
clare that © after all it must be allowed to be possible that waler 
may be a ternary compound,” ‘This conclusion appears in a great 
measure to be drawn from his experiments upon muriatic: and 
* Elements of Chemical Philosophy, p. 112 
H 2 fluoric 
