RICHARDS AND MARK. — THERMAL EXPANSION OF HYDROGEN. 139 



Carbon Dioxide. 



The final value, it will be noticed, is in accord with that obtained in 

 the preliminary work, 0.003727, having been increased by about the 

 same small amount as has the figure for hydrogen. Except for the pos- 

 sibility of some error dependent on the nature of the gas itself, such as 

 adsorption, the relative rates of expansion of hydrogen and carbon diox- 

 ide seem to be well established. 



As has been suggested in the last paragraph, the value obtained from 

 these experiments, as well as all those previously found by others, may 

 contain a constant error of uncertain magnitude caused by the different 

 amount of adsorption of the gas by the glass at the two different tempera- 

 tures employed. In the case of hydrogen and glass this correction is 

 probably small. A similar error would, of course, enter equally into the 

 determinations of increased tension at constant volume, but the ratio of 

 the results from the two methods might nevertheless be changed, since 

 the constant volume measurements were made between 0° C and 100° C. 

 Possibly the platinum-iridium bulb used by Chappuis may have exerted 

 a slightly adsorbing action upon his hydrogen, and, in the case of carbon 

 dioxide, the adsorption is probably of some importance. It is hoped that 

 determinations of the actual amount of this correction will soon be made 

 in this laboratory. 



The American Academy of Acts and Sciences has given valuable pe- 

 cuniary assistance to this investigation through a grant from the Rumford 

 Fund, for which it is a pleasure to acknowledge our indebtedness. 



