I i i a ae i ies 
Voltaic Electricity on Alcohol, §c. 351 
ties aiding the action, as is known to happen in other cases. 
But on heating the acid strongly in the oxyhydrogen flame, and 
then transferring it in its pure and transparent state to the table 
blowpipe flame, I could get no action on the galvanometer or ~ 
sparks with thirty-six pairs of four-inch plates, whilst it remained 
clear and transparent, although I observed both, when it became 
dark coloured from carbonaceous matter derived from the flame. 
The most probable explanation, therefore, seems to be, although 
the case is not free from ambiguity, that the carbonaceous matter 
merely gives a certain degree of conducting power to the mass, 
and that this is the source of the action on the galvanometer and 
of the sparks. I have thought it worth while to detail the re- 
sults, because they may at least serve as a caution to others in 
making similar experiments; if they are insufficient to alter our 
ideas of the non-action of the electric current on this substance. 
In the whole circumstances, however, I cannot help having 
some doubts as to the universality of the law of the necessary 
correspondence between voltaic decomposition and equality of 
atomic constitution, although I am by no means prepared to re- 
ject it altogether, and, on the contrary, think that, to a great 
extent, it may be well founded. It certainly appears very pro- 
bable, that, when the constituent elements of substances are 
nearly balanced in number, and particularly when that number 
is small, they will be peculiarly in that state of electric opposi- 
tion of character which will make them susceptible of voltaic 
agency ; and, on the other hand, where the number of atoms is 
considerable and unequally balanced, the mode of combination 
may partake more of that character of union of which organic 
nature appears to present numerous instances in its ternary and 
quaternary combinations, and in which the electric nature of the 
combined elements is probably less directly opposed, and will 
therefore be less adapted for voltaic action, and may very pro- 
bably even afford many cases where it will be altogether exclud- 
ed. Feebleness of affinity in the constituent elements may, how- 
ever, sometimes supply the place of simple atomic constitution, 
