242 PROFESSOR W. A. TILDEN ON THE SPECIFIC HEATS OF METALS, 



The same Platinum rolled thin. 



W. trj. w. t'i. t\. S. 



23-4692 -1158 -1152 17-30 100-12 -03177 



REGNAULT'S mean value for the specific heat of platinum by the method of mixture 

 is '03243. 



If we now compare gold with platinum 



S.H. of Gold : S.H. of Platinum :: 1 : - = 1-0369. 



In this case the atomic weights are known with a greater degree of certainty than 

 those of cobalt and nickel. The value for the atomic weight of gold may be regarded 

 as very accurately determined, while the atomic weight of platinum is still uncertain. 



Assuming Au = 19574 (H = 1), the atomic weight of platinum is ; - = 188'8, 



' 



from which it is obvious that the law of DULONG and PETIT cannot be applied to such 

 results in an absolute sense, for the atomic weight of platinum is approximately 

 193-41 (CLARKE). 



Copper and Iron. The copper was prepared by electrolysis from a solution of 

 specially purified copper sulphate. After being washed and dried, the metal was 

 fused in a covered ROSE crucible, supplied with a stream of hydrogen. The buttons 

 on cooling gave off gas sufficient to produce a large cavity in each, but on afterwards 

 heating to redness in a vacuum they yielded practically nothing. The metal was 

 bright within and without. 



Series IV. 



Specific Heat of Copper. 

 Pure and after fusion. S.G. 8 "522 at 20/20. 



Specific heat. 

 09248 

 09241 

 09205 

 09234 



Arithmetical mean . . '09232 



Assuming the value G3'12 preferred by CLAKKE for the atomic weight of copper, 

 and the specific heats determined by REGNAULT and by KOPP, we obtain the following 

 product of their combination : 



63-12 X -09515 (REGNAULT). . = 6-00 

 63-12 X '0925 (Korp) . . . . = 5-81 

 63-12 X '0923 (W. A. T.) . . = 5'82 



