56 Mr Adie on i,he Symplesoine(er. 
The bulb A at the upper end of the tube is drawn to a slen- 
der thread, and is at first left open. In order to introduce the 
gas and oil, I fill the bulb and tube with quicksilver : Then, 
holding the tube horizontal, a communication is formed between 
a gasometer, containing the gas to be used, and the slender pipe 
at the end of the bulb A, by means of a fiexible tube. As th& 
tube is brought to a vertical position, the quicksilver flows out 
till it descends in the tube to the level of the top of the cistern, 
and the gas enters to supply its place. The slender pipe is 
then to be sealed hermetically close to the bulb A, by a touch 
of the flame of a blowpipe. 
The tube A B C is now to be inverted, and the mercury 
poured out of the cistern C, allowing the column which occupies 
the tube to run towards the bulb, to prevent the escape of the 
gas. The tube being again turned into a vertical position, 
the portion of quicksilver which remains is removed, by pour- 
ing some of the oil over it, and heating the gas until, by its 
expansion, it forces the column of quicksilver which is left at 
the lower end of the tube, into the cistern ; then, holding the 
tube nearly horizontal, the oil will enter as the gas cools, and 
the remaining quicksilver may be poured out of the cistern C. 
The inclosed gas which has thus been introduced, changes 
its bulk, or occupies more or less space, according to the pres- 
sure of the atmosphere upon the surface of the oil in the cis- 
tern C. The scale m n for measuring the change in the bulk 
of the gas occasioned by a change of pressure, is formed experi- 
mentally, by placing the instrument in an air-tight glass-case, 
along with an accurate barometer and thermometer. 
The glass-case is furnished with a condensing and exhausting 
syringe, by which any density may be given to the inclosed 
gas, so as to support a column of quicksilver in the barometer of 
^8, 29} SO, or any other required number of inches. The height 
of the oil in the tube of the Sympiesometer corresponding to 
these points being marked on its scale, and the spaces between 
being divided into an hundred parts, these parts correspond 
with hundredths of an inch, on the scale of the mercurial baro- 
meter. 
As the bulk of the gas is altered by any change that takes 
place in the temperature of the atmosphere, it is necessary 
