THE COMBINING VOLUMES OF HYDROGEN AND OXYGEN. 
407 
was lowered, since the mixture now approximated to electrolytic gas, and there would 
have been some risk of shattering the vessel. 
The small residue of wet hydrogen was compressed into the neck of the explosion 
vessel and sparked for a few moments. In order to collect any gas bubbles caught 
between the mercury and the glass, the reservoir, j3, was lowered until most of the 
mercury had left the cylinder. A paper cone, mounted on a split cork, was fitted 
round the neck of the explosion vessel, and charged with a mixture of acetone and 
solid carbon dioxide. The water rapidly condensed as a film of ice on the walls of the 
neck, and, at the very low pressure in the vessel, the quantity of gas retained by the 
ice must have been negligibly small. When every trace of water had disappeared, 
mercury was allowed to rise to the bottom of the neck, and the dry hydrogen was 
sparked again for several minutes to eliminate any oxygen that might have escaped 
combination at the final explosion. Since any water vapour formed would at once be 
removed as ice, there was little chance of the reaction 2H 2 + 0 2 = 2H 2 0 being 
reversed. 
Measurement of the Hydrogen Residue. 
The tubing of the reservoir, (3, was now clipped, and the tap, y, turned to admit the 
dry hydrogen into the little pump, S. The mercury thread in the capillary was shot 
into the trap, e, and the gas passed to the pump through the tube, £, containing 
phosphoric oxide which had been distilled in oxygen, and the spiral, >?, which was 
cooled in liquid air. Any impurities condensed in this spiral could be pumped out and 
examined after all the hydrogen had been removed. The delivery tube of the pump, 
made of very fine-bore capillary was turned up and sealed as an inserted join into the 
small collecting vessel, 6. 
This vessel carried a 3-way tap, k, communicating on the one hand with the air, and 
on the other with the pipette by means of a capillary T-piece below the dead-space. 
The little pump was worked “ tight,” and any gas bubbles adhering to the turned-up 
end of the delivery tube were liberated by lowering the reservoir, X. The pipette was 
now filled with hydrogen from the train, and a setting was made in the manner already 
described. As soon as the points were set the tap, C, was closed, and the hydrogen 
in 6 was driven over into the pipette by raising the reservoir, X, and opening the tap, k. 
The mercury in the dead-space was depressed, and the original pressure was recovered 
by withdrawing mercury from the volume-adjuster until both points were set again. 
From the weight of mercury taken out, the volume of the hydrogen residue could be 
calculated. 
At first sight, this method of measuring a few cubic centimetres of gas may appear 
needlessly cumbrous, but in actual practice we found it both convenient and accurate. 
It had this great advantage, that no new pressure standard, with its possibly attendant 
errors, was introduced ; the hydrogen residue was measured under precisely the same 
conditions as the gases before synthesis. 
