J8 



THE COMPRESSIBILITY OF LITHIUM, SODIUM, 



enough to affect the use to which the preliminary values for the compressi- 

 bilities have already been put.* 



Although these early experiments are thus treated with but few words, 

 the experience gained in them was valuable, and contributed considerably 

 to the success of the later work. Moreover, they furnished the first data 

 concerning the compressibilities of the most compressible series of metals, 

 and the first evidence which showed that the usually accepted density of 

 lithium as found by Bunsen is considerably too great. This matter will 

 be discussed in a subsequent paper. 



FINAL DETERMINATIONS OF THE COMPRESSIBILITY OF ALL THE 



ALKALI METALS.t 



For this purpose the jacket indicated in fig. 2 (V) was used, a single 

 one serving for all determinations. In the first place, it was filled with 

 pure mercury, and then with mercury and paraffin oil, in each case having 

 the total compressibilities of the contents measured by first adding and 

 then removing small weighed amounts of mercury in the way already 

 described. In every case, as has been said, 3 mg. of rubber lubricant was 

 put on the stopper. The following table contains the figures, which are 

 given in order to make the somewhat unusual operations more obvious. 



Observations during compression of mercury and oil. 



The left-hand half of the table gives data for computing the difference 

 between the compressibility of mercury and glass which may be said 

 incidentally to agree with previous results although this fact is only 

 useful in the present research as showing that no accidental error was 

 made in obtaining the figures. 



*Richards, Proc. Amer. Acad., 39, 584 (1904) ; Zeit. phys. Chem., 49, 18 (1894). 

 fThe experimental part of this work was performed by W. N. Stull. 



