ELASTIC FORCES OF AQUEOUS VAPOUR. 575 
pletely freed from air. A first series of experiments is made 
with the level of the mercury at m; this level is then lowered 
to m', by allowing more mercury to flow out; the space occu- 
pied by the vapour is now double what it was in the first expe- 
riments ; if the least quantity of air was contained in the arm 
a 6, the tension of vapour found by the second experiments 
would no longer be the same as that found by the first. 
If it is required to determine by the same proceeding the 
tensions of a slightly volatile liquid at low temperatures, the 
tube fy is placed in communication with a small air-pump, by 
means of which the air enclosed in the arm Jc is previously 
rarefied, and then the tube fg is sealed with the lamp. The 
arrangement of the handle of the stop-cock 7, penetrated by 
three holes, fig. 5 61s, renders these different manipulations 
very easy. 
When experiments are to be made at higher pressures than 
those of the atmosphere, the tube fg is sealed at the moment 
when the manometer contains no mercury. Mercury is then 
poured into the tube & /, the air enclosed in the tube Ai is thus 
forced into less and less space, and increases in elastic force. 
The method which I have just described is exceedingly con- 
venient for very volatile liquids, because it allows of the deter- 
mination of their elastic forces between very extended limits. 
I did not employ it for water, because I consider it, on account 
of the number of measurements it requires, much less exact for 
weak tensions than the methods I have previously described, in 
which I endeavoured to diminish the number of measurements 
as much as possible *. 
The menisci in the bent tube adc, and in the manometer 
* This method is similar to that recently employed by M. Magnus for 
determining the elastic force of aqueous vapour between 0° and 100°, in a 
memoir which has just appeared (Poggendorff’s Annalen, xi. 226). [A trans- 
lation of the memoir here referred to will be found at page 218 of the present 
volume of the Scientific Memoirs.] It differs in the mode of obtaining the 
constant temperatures at which the observations ought to be made. M. Mag- 
nus employs for that purpose an apparatus composed of several concentric cases 
of sheet-iron, which are heated from below by means of spirit-lamps. I had 
occasion in a note inserted in Annales de Chimie et de Physique, tome vi. 
p- 370, to offer some objections to that method. M. Magnus thinks to have 
refuted my objections, by describing an experiment in which two air-ther- 
mometers, placed in such a manner that one instrument completely surrounded 
the other, rose exactly in the same ratio when the temperature was raised 
oy to 250°. I do not think that that experiment removes the difficulty ; 
cannot conceive, in fact, how two air-thermometers could show any sensible 
difference when placed under the circumstances described by M. Magnus. 
