POTASSIUM, RUBIDIUM, AND C/ESIUM. 



23 



although of course this was done on a much larger scale. The curve was 

 satisfactory in its slight curvature and its return-point ; hence the experi- 

 ment was deemed successful. From the curve it appeared that W = 1.148 

 grams. It had previously been found that W = 0.072 gram and that 

 Wk = 3.168 grams, as given at the foot of the third and second columns of 

 the table on page 19. Then 



[ 



1.148 0.072 (3.168 0.072)^ 1 0.972 



/ . o o I 



400 X 13.57 X 6.947 



+0.00000371=0-0000151 



In the cases of rubidium and caesium a slight correction was needed for 

 the glass vessel containing the metal. This correction has already been 

 discussed in the previous paper dealing with the compressibility of bromine 

 and need not be further detailed here. It is enough to say that the small 

 vessels in experiments 7, 8, and 9 weighed, respectively, 2.65, 1.21, and 

 2.88 grams. 



Thus the elements form a series of increasing compressibility with in- 

 creasing atomic weight. Lithium is not 3 times and caesium is over 16 

 times as compressible as mercury. 



The relations of these compressibilities to one another and to other con- 

 stants of the elements are of interest. In parallel columns below are given 

 the compressibilities, the atomic weights, the atomic volumes, and finally 

 an approximate constant obtained by dividing compressibilities by atomic 

 volumes. 



Comparison of fundamental constants. 



Thus all these properties increase together, the compressibilities increas- 

 ing much less rapidly than the atomic weights, but almost at the same rate 

 as the atomic volumes, especially with the three lightest elements. 



A further matter of interest may be discovered by the careful study of 

 the curves drawn from the figures given, namely, the different magnitude 

 of the second differential coefficient in the five cases.* In the case of lith- 

 ium ?nd sodium the compression curve, when corrected for the mercury 

 and oil present, is almost a straight line, while in the case of potassium, the 



*Richards and Stull, Carnegie Institution of Washington Publication No. 7, p. 44. 



