TAKING THE DENSITY OF VAPOUR OF VOLATILE LIQUIDS. 449 
hydrogen than when taken alone. The circumstances, moreover, under which 
we have taken the specific gravities in mixture with hydrogen, are just those 
which are admitted to raise the specific gravities of gases. They are lower 
temperatures than the temperatures of the observations with which we make 
comparisons, and (in two of the experiments) slightly higher pressures. 
It follows, therefore, that hydrogen hinders the contraction of alcohol and 
ether vapours; in other words, it diminishes the expansion-coefficient of these 
vapours. 
The differences observable between the two sets of specific gravities are small, 
but they are constant—just, in fact, what might be expected. 
Before leaving this division of the subject it may be useful to state the general 
behaviour of gases, as presented to us in our researches. We are led to the con- 
clusion, that at very high temperatures all vapours and gases expand nearly, if not 
quite uniformly; but at lower temperatures, that there are differences in the rate of 
expansion of different vapours. The admixture with permanent gases, and more 
especially with hydrogen, effects a slight alteration in the expansion-coefficient of 
the vapour, rendering the vapour more truly gaseous. 
To pursue this division of the enquiry, and to estimate these small differences, 
will require a more delicate apparatus than the one we have been employing. 
We pass on to what in fact is the more immediate subject of the paper—viz., how 
to render available for research our knowledge that vapours may be made into 
gases, at whatever temperature they are evolved. 
The method of research which we have hitherto described, involves the heat- 
ing of the liquid to be examined above its boiling point; this is requisite in order 
to ensure its total conversion into vapour. 
By the plan about to be described, the vapour is never heated higher than the 
temperature at which its density is to be taken. The principle is that of Dumas’ 
method—viz., the weighing of a given volume of vapour; the mode of carrying it 
out is, however, so different, that a minute description of the apparatus will be 
necessary. 
Those who have occupied themselves with Dumas’ mode of taking vapour den- 
sities, will be well aware, that one of the difficulties of using it for temper- 
atures near the boiling point of a liquid, is due to the circumstance that it is 
impossible to know when all the liquid has assumed the state of vapour. 
Our modification of the method consists in placing the liquid to be vaporised 
outside of that part of the apparatus which, when hermetically sealed, will con- 
tain the vapour mixed with gas. The apparatus consists of a pair of large bulbs, 
connected together by means of a neck, and terminating on either side in narrow 
glass tubes, which are bent and blown into small bulbs, as represented in the 
drawing. The bulbs C and C are large, and together capable of holding from 200 
VOL. XXII. PART II. 5 Zz 
