120 Proceedings of the Royal Society of Edinburgh. [Sess. 
II. Method of Experiment. 
Measurements were carried out at 15° C. and at pressures sometimes 
reaching 100 atmospheres. The gases mainly used were nitrogen, 
hydrogen, and oxygen ; a few special tests were carried out with firedamp 
and carbon dioxide. The nitrogen and oxygen employed were the com- 
mercial gases supplied by the British Oxygen Company; they usually 
contained about 2 per cent, of impurity. Hydrogen and, on some occa- 
sions, C0 2 were made in the laboratory. The firedamp was obtained in 
compressed form from South Wales, and contained about 98 per cent. CH 4 . 
Tests were carried out with (a) activated cocoanut charcoal, ( b ) activated 
birch charcoal, (c) German impregnated charcoal, (d) common wood charcoal, 
and ( e ) activated colloidal silica made in the laboratory from the hydrogel 
by the method described in a paper recently given before the Royal Society, 
London.* In addition, a few special tests, dealt with in the next section, 
were made upon anthracite. 
The method consisted in filling a small steel cylinder with the substance 
to be tested, the material being well shaken down within. Before loading 
the cylinder the charcoal or silica was dried in thin layers in a gas furnace ; 
it was inserted hot. The cylinder was fitted with a gas-cock screwed and 
soldered into place. After it had cooled it was charged with dry gas 
to the desired pressure, and a sufficient time allowed to elapse to allow the 
heat of adsorption to dissipate and the pressure to stabilise. The cylinder 
was then provided with a throttle-valve and pressure-gauge and placed in 
a water-bath. The volume contained by it was determined by allowing 
the gas to flow out very slowly through a meter. After the discharge of 
1 litre, or, in some cases, 5 litres of gas, the throttle was closed, and a 
pressure-reading taken after stability had again been attained. The opera- 
tion was repeated until no more gas was discharged. The pressure-gauges 
were tested from time to time against a large gauge which had been 
calibrated at the National Physical Laboratory. The meter was tested 
against displacement. 
III. General Results. 
The experimental method described ascertained the volume, at 15° C., 
discharged between any given pressure and atmospheric pressure. When 
constructing the graphs it was necessary to add to that volume the amount 
of gas retained by the adsorbent at atmospheric pressure and 15° C. This 
* H. Briggs, “The Adsorption of Gas by Charcoal, Silica, and other Substances,” 
Proc. Roy. Soc., A, 1921 (in course of publication). 
