140 PROCEEDINGS OP THE AMERICAN ACADEMY. 



Summary. 



Improvements in the surroundings, the stability, and the manipulation 

 of the apparatus previously described for the determination of the co- 

 efficient of expansion of gases by heat under constant pressure are set 

 forth. The possibility of measuring the pressure of a gas to an accuracy 

 of one thousandth of a millimeter of mercury is shown, and some of the 

 difficulties of the determination of volume and temperature to a corre- 

 sponding degree of accuracy are pointed out. 



Measurements of the coefficient of expansion of hydrogen (0.0036609) 

 and of carbon dioxide (0.0037282) under constant pressure as far as 

 five significant figures are recorded, and their probable reliability is dis- 

 cussed. The importance of knowledge of the amount of adsorption of 

 gases by glass is mentioned in this connection. The results must not be 

 considered as final measurements, but their consistency shows that the 

 apparatus is capable of attaining a high degree of precision. 



