of H. Y. Regnault. 



275 



Series. 



No. of Experi- 

 ments in each, 

 series. 



Mean value 

 1 -t- 100a. 



Difference between 

 greatest and least 

 values in eacii series. 



1 



■1 A 



14 



"I oacofy 



i.obblo 



f\r\ ^ /\(\ 



2 



18 



1.3G633 



.00118 



3 



12 



1.36679 



.00135 



4 



6 



1,36650 



.00130 



5 



4 



1.36706 



.00025 



The magnitude of the fifth determination as compared 

 with the other four, is due not to errors in the experi- 

 ments, but to the fact that the first four indicate the 

 coefficient of expansion of air when at constant volume, 

 and the last indicates the coefficient of expansion of air 

 under constant pressure. 



In the second part of the same memoir he determines 

 the coefficients of expansion of some other gases, and dis- 

 proves the opinion, previously entertained, that all gases 

 expand equally under the influence of heat. In the third 

 part he proves by more than thirty experiments with vari- 

 ous apparatus that the coefficient of expansion increases 

 with the pressure. He also points out that for gases in a 

 state of extreme dilatation the coefficient of expansion is 

 the same in all, so that Gay-Lussac's law holds good if we 

 understand it as applied only to gases in this limiting 

 state. 



The second memoir is on the determination of the 

 density of gases. The values here given are considered 

 by Roscoe (Elementary Chemistry, Appendix, p. 443), to 

 be the most reliable of all of Regnault's determinations. 

 Those for oxyyen and nitrogen are confirmed in a re- 

 markable manner by the experiments of Stas on the com- 

 bining weights of these elements. Regnault's researches 

 give for oxygen a combining weight of 15.960, which is 

 exactly the value obtained by Stas, and for nitrogen, 

 14.025, which is .017 in excess of Stas' determination. 



The third memoir discusses experiments on the weight 



