( 441 ) 
XIX.—On a Mode of Taking the Density of Vapour of Volatile Liquids, at Tem- 
peratures below the Boiling Point. By Dr Lyon Puayratr, C.B., F.R.S., and 
J. A. WANKLYN, F.R.S.E. 
(Read 7th January 1861.) 
The interest awakened by Gay-Lussac’s great discovery of the simplicity in 
the relation of the volumes of gases has greatly increased in recent times, when 
chemists have discovered that, in a large number of instances at least, the formula 
of a body, as deduced physically from its vapour density, exactly coincides with 
that deducible from chemical considerations of its reactions, and from the nature 
of the products arising in consequence of them. 
The processes at present used for determining the vapour densities of bodies, 
are those of Gay-Lussac and Dumas. 
Although the operations by which both these processes are carried out are 
well known to chemists, it will save considerable repetition in description if 
we give a short account of them at the outset, and will enable the modifications 
in practice which we make of both processes to be more readily appreciated. 
Gay-Lussac’s process consists in measuring the volume of vapour produced 
by a known weight of liquid under circumstances which enable us to compare 
the former with an equal volume of air at normal temperatures and pressure. 
A graduated tube is filled with mercury, and inverted in a mercurial trough. A 
known weight of the volatile body is introduced into the tube, and the whole 
apparatus is transferred to a bath, which is heated 20° to 40° C. above the boiling 
point of the enclosed volatile liquid. By this arrangement, readings of the 
volume occupied by the vapour can conveniently be made, at different temper- 
atures. The pressure to which the vapour is subject is obtained by adding the 
barometric pressure to that of the column of water or oil contained in the bath, 
and deducting from the sum of these pressures the pressure of the column of 
mercury contained in the graduated tube. Some minute but necessary correc- 
tions require to be applied; the barometric column, and that in the graduated 
tube, must be reduced to 0° C., and an allowance be made for the expansion of 
the graduated tube. 
Dumas’ method for taking vapour densities is the converse of Guy-Lussac’s, and 
consists in determining the weight of a given volume of vapour at a known 
temperature and pressure. 
It is practised as follows:—A flask with the neck contracted and drawn to a 
point is weighed in dry air, charged with a portion of volatile liquid, heated to a 
suitable temperature, so as to convert the liquid wholly into vapour; and then, 
VOL. XXII. PART III. ) 5X 
