﻿Volume, and Temperature of Rarefied Gases. 309 



if it be partially exhausted, owing to the external pressure of 

 the atmosphere not being balanced by equal pressure in the 

 interior. In Mendeleeff's experiments, for example, in which 

 glass vessels were used, and in which the volume of the gas 

 was ascertained by weighing the mercury run out of the vessel, 

 it appears not unlikely that at low pressures the vessel con- 

 tained less gas than at high pressures. The values of the 

 products of pressure and volume, pv, might really remain 

 constant with decrease of pressure ; but as he read pressure 

 directly, and assumed a volume which was probably too 

 great, he registered an increase. We are informed, how- 

 ever, by Professor Mendeleeff, that he introduced a correction 

 for this possible deviation. 



All the researches described aim at a direct measurement 

 of pressure. Now there is a limit to the possibility of such 

 measurement, and when the pressure is low, the readings 

 necessarily involve considerable relative error, the extent of 

 which is difficult, if not impossible, to gauge. 



We have therefore adopted a different plan, which admits 

 of the observation of the behaviour of gas at pressures much 

 lower than most of those previously employed. We have 

 met with difficulties, subsequently to be alluded to, which 

 may vitiate the results for some gases ; but for hydrogen the 

 results are fairly trustworthy. 



To describe the method in a few lines : — Two McLeod 

 gauges are exhausted in communication with each other : one 

 is heated to a known high temperature ; and at the same in- 

 stant, both are closed ; after cooling the pressure is read in 

 both. The readings afford data from which the expansion of 

 the gas may be calculated. 



Description of the Apparatus. 



The apparatus used for these experiments consisted essen- 

 tially of two McLeod gauges placed side by side on a stand. 

 These gauges were connected on the right to apparatus for 

 exhausting them, and on the left to apparatus for admitting 

 into them, when exhausted; measured quantities of the gas 

 under examination. 



The apparatus for exhausting shown on the right-hand side 

 of the diagram consisted of a water-pump A worked by high- 

 pressure water, and capable of giving a vacuum of 11 millim. 

 of mercury. 



The exhaustion of the apparatus was commenced by this 

 pump and completed by the mercury-pump B. This mercury- 

 pump discharged the gas taken out at each stroke into a 



