] 42 M, V. Regnault on the Expansion of Gases. 



These results are not in discordance with those observed by 

 Messrs. Joule and Thomson under analogous conditions. But 

 I explain them otherwise, and I draw from my experiments very 

 different conclusions from those which the English physicists 

 have deduced from theirs. 



Thomson and Joule's experiments on the passage of air 

 through an orifice in a thin plate gave far more considerable di- 

 minutions of temperature than those I obtained under analogous 

 circumstances. This may in a measure arise from the fact that 

 the velocity of outflow was greater in their experiments, but 

 especially, I think, from the fact that in their experiments the 

 air is directly compressed by the pump in the long tube of uni- 

 form section, where it circulates rapidly as far as the small orifice 

 through which it diffuses into the atmosphere. But 1 think in 

 this case the temperature is far from being uniform in the various 

 parts of the tube. 



The air arrives greatly heated by the working of the pump : 

 I admit that it loses this great excess of heat during its long pas- 

 sage through the double worm ; but the layers of air near the 

 orifice furnish directly the work which drives the gas outwards, 

 losing thus a certain quantity of heat which they have not time 

 to take from the fresh layers which arrive to replace them. In 

 short, I think we cannot assume for a natural gas in motion the 

 principles which are propounded theoretically for the ideal gas, 

 even when they would be almost true for the natural gas in a state 

 of rest. As to the vis viva which the molecules assume in ex- 

 panding, inasmuch as its creation only takes place at the passage 

 of the orifice, it certainly only takes place at the expense of the 

 heat of the emerging gas, the temperature of which should thus 

 be much lowered. 



In previous memoirs I have frequently dwelt on the slowness 

 with which air enclosed in a metal reservoir, surrounded by a 

 large mass of water, reacquires the temperature of this water 

 when it has been cooled by the expansion it experiences when a 

 portion escapes. When this reservoir is cylindrical, and its 

 diameter is only 20 centims., from ten to fifteen minutes elapse 

 before the open air-manometer which communicates with this 

 reservoir resumes the stationary condition. 



Messrs. Joule and Thomson speak of a considerable disengage- 

 ment of heat which they have observed when air strikes against 

 an obstacle opposed to its emergence, especially when it grazes 

 the bulb of a thermometer which is kept in the conical space of 

 a gutta-percha funnel, thus greatly diminishing the section which 

 remains free for the passage of the gas, &c. I have never suc- 

 ceeded in observing such a phenomenon; and the fact appears 

 to me contradictory to the experiments I have made with the 



