270 
THEORY OF GALVANISM. 
hydrogen. But when the positive wire is of a non-oxidable 
metal, oxygen gas is disengaged ; and in the production of this 
gas the electric fluid might be expected to act, instead of being 
employed in carrying hydrogen to the negative wire. 
The same class of phenomena has been explained by Mr* 
Davy’s theory Davy on a different theory. According to bis view, bodies 
tion by the walc h are capable of entering into chemical union, are 
pile* Original invariably in opposite electrical states, oxygen for example, is 
disposition of negative and hydrogen positive. From the known laws of 
bodies to electrical attraction and repulsion, it will follow that oxygen 
trie plates will be attracted by positive and repelled by negative surfaces, 
pass^o^tTe t0 an< ^ contrary process will happen with respect to hydrogen. 
opposite poles Jt is easy then to conceive that these opposite attractions may 
of the pi!e,&c. . . _ . . , 
produce the decomposition or water. lo explain the locomo- 
tion of its elements, we may imagine a chain of particles of water, 
extending from the point P to the point N, fig. 2, and con- 
sisting each of an atom of oxygen united to an atom of hydrogen. 
In fig. 2, the combination is represented as undisturbed, and 
the chain is consisting of six atoms of water. But when the 
attractive force of the point P for oxygen, and N for hydrogen, 
begin to act, an atom of oxygen and another of hydrogen are 
removed, as shewn by fig. 3, and new combinations happen 
between the remaining atoms ; the second of oxygen uniting 
with the first of hydrogen, and so on. The terminating atoms 
being supposed to be removed, a new change will followsimilar 
to the first, and thus the process will continue to be carried on, 
not only when the chain of particles is a short one, but when 
it extends to a very considerable length. 
This doctrine The theory of Mr. Davy, which I have thus attempted to 
t h /gen end i za* M ustrate > derives probability from its being founded on a general 
tion of observ- property of bodies (their different electrical energies) which 
appears to be established experimentally, as far at least as experi- 
ment can be applied to so delicate a subject. Ir has the advan- 
tage also of explaining a number of facts, chiefly arising out of 
his own researches, which scarcely admit of being brought 
under any former generalization. Thus the invisible transfer- 
ence of an element to a considerable distance, even through 
fluids 
