8 THE COMPRESSIBILITY OF LITHIUM, SODIUM, 



Although it is thus evident that Davy had a distinct conception of the 

 significance of changing atomic volume, it is equally evident that he was 

 no more cognizant of the relation of compressibility to that problem than 

 were others who have more recently touched upon the subject. This con- 

 nection is now worthy of brief comment. 



Upon consideration, it is evident that the proposed interpretation in- 

 volves the assumption that chemical affinity in its action exerts pressure; 

 this internal pressure is supposed to cause a diminution of the volume of 

 the substance. Here the main difference between the present series of 

 papers and the independent work of others may be especially emphasized. 

 It is now contended that if chemical affinity exerts pressure, this pressure 

 must have more effect in its action upon the volume of a compressible sub- 

 stance than upon that of a relatively incompressible one. To have any im- 

 portant meaning, the volume-change must be considered in relation to the 

 compressibilities of the substances concerned. 



If, then, compressibility is not merely an isolated physical fact, but 

 rather a phenomenon of chemical significance immediately connected with 

 the most important chemical properties of material, it becomes a highly 

 interesting problem to determine the compressibility of as many simple sub- 

 stances as possible. The fact that the results are not always easy of inter- 

 pretation in the more complex cases does not prevent an approximate 

 analysis of the figures in the simpler cases ; and even this approximate 

 interpretation holds out the promise of a resulting gain in our knowledge 

 concerning the mechanism of chemical affinity. 



It was with these facts in mind that three years ago we began the study 

 of compressibility, in the first place devising a new method for its deter- 

 mination, because none of the already existing methods were adequately 

 convenient or precise. This method has already been described in detail 

 in Publication 7 of the Carnegie Institution of Washington.* The essen- 

 tial feature of the method is the comparison of the unknown compressibility 

 with the compressibility of mercury, by first compressing mercury in a 

 given apparatus, measuring both pressure and change of volume, and then 

 displacing most of the mercury by the substance to be studied, and again 

 noting the relationship of pressure to volume. Obviously in such a method 

 as this, the compressibility of the apparatus itself is eliminated, occurring 

 in both series of measurements. The relation of volume to pressure was 

 easily determined by causing the mercury meniscus to make contact with 

 a very fine platinum point in a tube of narrow diameter ; and the decrease 

 in volume was determined by adding weighed globules of mercury and 

 noting the pressures needed to cause similar contact. Several forms of 



*This paper by Richards and Stull appears also in more abbreviated form in Journ. 

 Am. Chem. Soc., 26, p. 399 (1904), and Zcit. phys. Chem., 49, p. 1 (1904). 



