256

DECOMPOSITION OF WATER

be placed in such a manner across a furnace, that the end towhich the retort is luted be a little elevated. Then light a fire inthe furnace, and, when the gun-barrel has become red-hof, applya lamp under the retort containing the water. The steam of thisfluid will thus pass through the red-hot iron barrel, and becomedecomposed. Its oxigen unites with the iron, and its hidrogenbecomes liberated in the form of gas, and may be collected inthe pneumatic trough.

The same experiment may be repeated, substituting a coatedglass or earthen tube for a gun-barrel, and weighing the ironwire accurately before and after the experiment. The wire willbe found to have gained weight, in proportion to the quantityof oxigen which formed the water. If this weight be added tothat of the hidrogen gas it will make up exactly the weight ofthe w'ater that has disappeared.

From experiments of this kind, conducted with the utmostaccuracy, as well as from synthetic investigations, it appears thatwater is composed of 85 parts of oxigen, and 15 of hidrogenby weight very nearly. But as hidrogen gas is at least eleventimes lighter than common air, the proportion of gases, by vo-lume, required to form water, is about two of hidrogen to oneof oxigen.

Remark.. .Water may be decomposed in a similar apparatus,by distilling it over red-hot charcoal instead of iron wire, butin this case the oxigen of the water combines with part of thecharcoal and forms carbonic acid gas, while its hidrogen unitesto another portion of the charcoal, and produces carbonated hi-drogen gas. (See vol. I. p. 245.)

DECOMPOSITION OF WATER BY ELECTRICITYAND GALVANISM.

Water has also been decomposed by the agency of electricityand galvanism. If uninterrupted electrical discharges be madeto pass through a quantity of water, part of it is decomposed int<?