GASEOUS AND LIQUID STATES OE MATTER. 
579 
it corresponds, according to the experiments of Begnault, to an apparent error of a little 
more than one-fourth of an atmosphere at a pressure of twenty atmospheres, and ac- 
cording to those of Natterer, to an approximate error of one atmosphere when the 
pressure attains 107 atmospheres. These data are manifestly insufficient, and I have 
therefore not attempted to deduce the true pressure from the observed change of volume 
in the air-tube. It will be easy to apply hereafter the corrections for true pressure when 
they are ascertained, and for the purposes of this paper they are not required. The 
general form of the curves representing the changes of volume in carbonic acid will 
hardly undergo any sensible change from the irregularities in the air-tube ; nor will any 
of the general conclusions at which I have arrived be affected by them. It must, how- 
ever, always be understood that, when the pressures are occasionally spoken of, as indi- 
cated by the apparent contraction of the air in the air-gauge, the approximate pressures 
only are meant. 
To obtain the capacity in cubic centimetres from the weight in grammes of the mer- 
cury which filled any part of a glass tube, the following formula was used. 
C=W 
1 +0*000154 . t 
13'596 
1-00012, 
where C is the capacity in cubic centims., W the weight of the mercury which filled the 
tube at the temperature t, 0-000154 the coefficient of apparent expansion of mercury 
in glass, 13-596 the density of mercury at 0°, and 1-00012 the density of water at 4°. 
The volume of the gas V, at 0° and 760 millims. of pressure, was deduced from the 
double observations, as follows, 
Y=C 1 
' 1 +at’ 760 
where C is the capacity of the tube (figure, page 576) from a to d, or from a to e, t the 
temperature, a the coefficient of expansion of the gas for heat (0-00366 in the case of air, 
0-0037 in that of carbonic acid), h the height of the barometer reduced to 0° and to the 
latitude of 45°, d the difference of the mercurial columns in the U-tuhe similarly reduced. 
Having thus ascertained the volumes of the air and of the carbonic acid before com- 
pression, at 0° and 760 millims., it was easy to calculate their volumes, under the same 
pressure of 760 millims., at the temperatures at which the measurements were made 
when the gases were compressed, and thence to deduce the values of the fractions repre- 
senting the diminution of volume. But the fractions thus obtained would not give 
results directly comparable for air and carbonic acid. Although the capillary glass 
tubes in the apparatus (fig. 6) communicated with the same reservoir, the pressure on 
the contained gases was not quite equal, in consequence of the mercurial columns, 
which confined the air and carbonic acid, being of different heights. The column always 
stood higher in the carbonic-acid-tube than in the air-tube, so that the pressure in the 
latter was a little greater than in the former. The difference in the lengths of the mer- 
curial columns rarely exceeded 200 millims., or about one-fourth of an atmosphere. This 
