Oi} GASEOUS EXPLOSIONS. QAO 
higher than those of Regnault. Mr. Swann’s results have now been 
published, and he has supplied the Committee with a complete copy of 
his paper. ‘The method employed by him had previously been used by 
Professor Callendar for determining the specific heat of superheated 
steam. A current of the gas is passed over an electrical heater, the 
energy supplied to which can be accurately measured, and the temperature 
of the gas is measured before and after passing the heater by means of 
platinum thermometers. The rise of temperature, which amounted to 
about 5° C., can be measured correct to 0°001° C., and an at least equal 
degree of accuracy is obtainable in the measurement of energy supplied. 
A correction of the order of 10 per cent. of the total supply of heat has, 
however, to be applied for the loss of heat from the gas as it passes to 
the thermometer. It is assumed that with a given inflow and outflow 
temperature of the gas this loss of heat is independent of the rate of 
flow, and its amount is determined by experiments at different rates of 
flow. 
It will be remembered that a correction, amounting to about 5 per 
cent., was applied in Regnault’s experiments for the flow of heat by 
conduction along the substance of the pipe which connected his heater 
with his calorimeter. Regnault assumed that with a given temperature 
difference this correction was independent of the rate of flow of gas. It 
was pointed out by Professor Callendar that this assumption could not be 
justified, and that it would lead to too low a value of the volumetric heat. 
The corrections involved in the methods employed by Swann have been 
fully discussed by the author, and also by Callendar. They appear to 
admit of determination with an order of accuracy approaching 1 in 1000 
in the result. On this account, and because of the close agreement of 
Swann’s results with those obtained by Joly, the Committee consider that 
there is now little doubt that Regnault’s figures were too low, and that the 
volumetric heat of air at 100° C. may be taken as being within 1 per cent. 
of 5°0 calories per gm. molecule, or 19°8 foot Ib. per cubic foot. 
They have come to this conclusion with the less difficulty because 
to a great extent they have Regnault’s own authority for it. He appears 
to have been fully aware of the uncertainty introduced into his results 
by the heat-flow along the connecting pipe; he discusses it in the same 
way as Callendar and Swann, and arrives at the same conclusion, that is, 
that it would be in such a direction as to make his results too low. The 
passage of his original paper in which he deals with this matter is of 
very considerable interest, both historically and in connection with the 
work of the Committee, and they have therefore reproduced it in full in 
Appendix A. 
For the volumetric heat of CO. the Committee also feel justified in 
adopting Swann’s values as correct to within 1 per cent. They are as 
follows :— 
At 20° 4. At 100° @. 
Specific heat at constant pressure : 5 0-202 0:221 
' Volumetric heat :— : 
Cals. per gm. molecule! . ci : : 6:93 776 
Foot Ib. per cubic foot Bs, igs ee De 30°7 
* Taken as 44 grammes, 
