130 CHEMICAL DISCOVERY AND INVENTION 



maxima and minima in each case corresponding to the same 

 atomic weight. 



In the course of experiments on the specific heats of the 

 elements at low temperatures between the boiling-points of 

 liquid nitrogen and hydrogen, Sir James Dewar has observed 

 that the mean atomic heats (at 50 Abs.) of the elements are a 

 periodic function of the atomic weights. The atomic heat is to 

 be understood as the product of multiplying the specific heat by 

 the atomic weight, and at ordinary temperatures from the freezing 

 to the boiling-point of water, the value of this product is almost 

 uniformly 6-2 to 64 and periodicity has not been noticed. 



At the temperature of only 50 absolute or 223 below centi- 

 grade zero the specific heats are very small, that of carbon in the 

 form of diamond being only 0-0028 or ^ 7 - of the specific heat 

 of water. When these numbers are multiplied by the atomic 

 weights the product is still of small numerical value. The 

 figures for the atomic heats range from 0-03 for carbon to 6-82 

 for caesium. Doubtless the values in some cases will be slightly 

 corrected by future experiment, but in the meantime the 

 recognition of the periodic relation, that is, the rise and fall at 

 regular intervals, is an interesting observation which brings the 

 outstanding property of specific heat into the same category as 

 the rest of the physical properties of the elements. 



The diagram published in connection with Dewar's paper in 

 the Proceedings of the Royal Society, vol. 89 (1913), p. 169, shows 

 that the rise and fall of the curve follows very nearly the same 

 course as the curve of atomic volumes originally pointed out by 

 Lothar Meyer, and incorporated into the comprehensive diagram 

 already given (page 129). 



The importance of the periodic scheme, as arranged by 

 Mendeleeff, deserves a little further notice in view of the influence 

 it has had on the progress of theoretical chemistry. First of all 

 it should be noticed that several of the elements now known, 

 and amounting at the present time to eighty-three, were not 

 known in 1869-70 when the scheme was published. The table 

 of elements first arranged by MendeleefE in 1869 contained only 

 sixty-three then recognised elements, the atomic weights being 

 in many cases uncertain. But inasmuch as intervals too wide 

 to be accounted for by mere experimental inexactitude occurred 

 in several parts of the table, vacant spaces were left which 

 suggested that substances existed somewhere in nature of 



