Regnaulfs Researches upon the Dilatation of Gases. 67 



that would enable me to measure directly the increase of the 

 volume of the gas, (it being apparently under the same pressure 

 at 0° and 100° Cent.) This method is evidently the only one 

 that we can employ for gases that do not follow the law of Mar- 

 riotte under slight differences of pressure." 



The instrument used was founded on the principle of M. Pouil- 

 let, pyrometre a air, and the results afforded by this series of ex- 

 periments are — 



For hydrogen, 



a 



it 



a 



U 



a 



a 



atmospheric air, 

 oxide of carbon, 

 carbonic acid, . 

 protoxide of nitrogen, 

 cyanogen, 

 sulphurous acid, 



0.36613 

 0.36706 

 0.36688 

 0.37099 

 0.37195 



0.38767 

 0.39028 



* The atmospheric air has afforded a number a little higher 

 than the mean of the preceding experiments ; but the difference 

 is insensible, and besides it may be caused by the air not follow- 

 ing rigorously the law of Marriotte. 



" The oxide of carbon has given the same number as in the 

 former experiments. 



" The coefficients for carbonic acid and protoxide of nitrogen 

 are much higher than those determined by the former experi- 

 ments, and this is no doubt owing to the fact that these gases do 

 not follow the law of Marriotte, and that their volumes at 100° 

 Cent., under the greater pressure to which it finds itself subjected 

 at this temperature in the first experiments, are smaller than they 



ought to be according to this law." 



As for the difference in the coefficients of cyanogen and sul- 

 phurous acid, Regnault has ascertained that it was owing to an 

 error in the first experiments, the gases not being perfectly dry. 



The dilatation of gases under different pressures, calculated 

 after the change of the elastic force. — "Philosophers admit gene- 

 rally, that the dilatation of gases is constant between the same 

 limit of temperature, whatever be the pressure to which they 

 are subjected ; consequently, that it is entirely independent of 

 the original density of the gas. But it is difficult to cite the 

 experiments upon which this law is founded. Many observers 

 having obtained the same value for the coefficient of the dilata- 

 tion of air under different barometrical pressures, have concluded 



