ELASTIC FORCES OF AQUEOUS VAPOUR. 571 
stead a bell-glass of smaller dimensions, containing about 20 
litres of fluid. The barometric tubes are then simply attached 
to their board, and the globe A is placed at a greater distance 
from the tubes. The globe is first surrounded with melting 
ice to determine the tension of the vapour at 0°; the ice is then 
removed, and into the bell-glass a concentrated solution of chlo- 
ride of calcium is poured, the temperature of which is constantly 
lowered by dissolving ice in it. This mixture ought to be con- 
stantly stirred, and we obtain easily, not a positively steady 
temperature, but one at least which oscillates within very nar- 
row limits, for as long a time as is required, by adding small 
quantities of broken ice as often as the temperature begins to 
rise. For obtaining the greatest degrees of cold, crystallized 
chloride of calcium was used, in alternate layers with snow. 
This mixture was constantly stirred until it became fluid; the 
temperature was thus gradually lowered until there remained 
but a very small quantity of ice swimming on the top of the 
liquid. The observation was made at the moment of minimum 
temperature, the liquid being rapidly stirred; this minimum 
could moreover be retained for some time, by adding small quan- 
tities of snow. 
For elevating the temperature of the bath, small quantities of 
a hot saturated solution of chloride of calcium are poured into 
the bell-glass, until a sufficient elevation of temperature is pro- 
duced. A hot solution of chloride of calcium is used in prefer- 
ence to pure water, that the frigorific power of the liquid com- 
posing the bath may not be diminished, and that we may again 
be enabled to lower the temperature by the addition of pow- 
dered ice. 
It is evident that the temperature will be rendered stationary 
so much the more easily as it approaches that of the surround- 
ing air, and it is in this case that the variations in the elastic 
force of aqueous vapour with the temperature are the most sen- 
sible. Below —20° a change of some tenths of a degree in the 
temperature produced no sensible variations in the elastic force 
of aqueous vapour. The experiments made in this manner are 
noted in Table III., Series F, G, H. 
_ The mode of procedure, modified in the manner I have just 
described, may serve for inferior temperatures to those of the 
surrounding air; but I have satisfied myself by direct experi- 
ment that it likewise affords accurate results for temperatures 
