SPECIFIC AND ATOMIC HEAT 



67 



heats, especially for elements with small atomic volumes, as 

 appears from the following table : 



To a certain extent then the atomic heat is a periodic 

 function of the atomic weight, like the atomic volume. 

 Only the deviations in the first period, lithium to sodium, 

 are very large, with 1-76 as minimum for carbon ; in the 

 second period, sodium to potassium, much smaller, with 4-63 

 as minimum for silicon, whilst after potassium the values 

 no longer falling below 6, but still with minima for the 

 small atomic weights. 



It has often been attempted, but with little success, to 

 refer the variations from Dulong and Petit's law to 

 secondary causes. Spring pointed out the possibility that 

 the work of expansion might affect the values differently, 

 and measured the specific heat of some alloys (d'Arcet's 

 and others) which show a minimum of density, in the 

 neighbourhood of the minimum, but without arriving at 

 any clearly marked regularities l . Weber found that the 

 abnormally low atomic heats of carbon, boron, and silicon ap- 

 proach 6 on rise of temperature. Thus the atomic heats are : 



Graphite at 1000 

 Boron 233 



Silicon 232 



5-58 

 4-03 

 5-7 



This is in accordance with the observation that small 

 atomic heats are associated with small atomic volumes. 



1 See also Kicharz, Wied. Ann. 48- 712. 

 E 2, 



