340 
. DR. E. H. GRIFFITHS AND MR. EZER GRIFFITHS ON THE 
With the exception of sodium, the smoothed curves have been drawn through the 
experimental results of Dorsey and of Ayres in the range 0° C. to —180° C., and 
f* 
the curves extrapolated in order to deduce the value of aev a dt at various tempe- 
Jo 
ratures. 
For sodium the value 0'000224, determined by us in some preliminary experiments 
by a new method, has been utilized. 
This method, which will be described in a later paper, did not indicate any marked 
change in the coefficient over the range 0° C. to 95° C. 
For low temperature a coefficient of decrease of the same order as that of 
cadmium was assumed, a procedure which is justified by the fact that the 
coefficient of expansion is proportional to the atomic heat, as shown by Gruneisen, 
and at low temperature the atomic heat curves of sodium and cadmium are 
similar. 
Nernst and Lindemann* adopt a different procedure, basing their method on 
Gruneisen’s observation that the coefficient of expansion is proportional to the atomic 
heat; they obtain the approximate relation 
C„-C, = (yTA. .......... (6) 
where A is a constant characteristic of the substance which can be deduced from 
measurements of the compressibility and coefficient of expansion made at one tempe¬ 
rature. 
In the cases of Al, Ag, Pb, and Gu the values so obtained are compared with the values 
calculated by means of equation (5), which are utilized in all subsequent comparisons 
of theory and experiment in this paper. 
It is shown in Part II. of this paper that no one of the theoretical formulae hitherto 
published is capable of representing the atomic heat-temperature curves over the 
entire range. Since data based on the experimental observations are necessary to 
test any theory, we summarise below the values, at convenient intervals of tempera¬ 
ture, of the atomic heat read from curves which we believe represent the most 
probable paths through the observations of Nernst at liquid air and hydrogen 
temperatures; of Dewar at 50 C. abs. and through our own numbers! from 
100° C. abs. to 400° C. abs. 
* ‘ La Theorie du Rayonnement et les Quanta,’ p. 265, 1912. 
t Vide Tables VI. to XIY. and fig. 4. 
