430 Dr. J. A. Harker on the Specific 



plishing the feat. The increase in the scintillating power 

 produced by crushing the crystals may be due to the 

 exposure of more of the molecular impurities. If any 

 difficulty is felt regarding the action of a single electron, we 

 can assume as well, that the continued action of the rays 

 results in an unstable chemical change effected upon the 

 impurity, the flash being due to the toppling over of the 

 molecular complex into its original condition. 



The fact that the flash lasts as long as ^^ of a second 

 seems to point to some such action, for the wave train emitted 

 even in this brief length of time is measured in miles, and a 

 simple vibration started by the impact of an electron could 

 scarcely be supposed to persist for so long a time. 



LI. 2he Sjiecific Heat of Iron at High Temperatures. B;i 

 J. A. Hakker, JJ.Sc, Assistant at the National Physical 

 Laboratory, Teddington*. (From the National Physical 

 Laboratory.) 



A LTHOUGH a knowledge of the specific heat of iron at 

 -^- high temperatures would seem to be of vital importance 

 to the metallurgical industry, yet it would appear that no 

 determination of this physical constant has been made since 

 our high-temperature thermometric scale has been rendered 

 definite and precise by the investigations of the last fifteen 

 years. 



Probably on account of its difficulty, the work done on the 

 subject of specific heats at high temperatures is comparatively 

 meagre. Apart from observations dating from more than 

 forty years ago, a careful search in the literature of the 

 subject has revealed the work of only one observer, who 

 carried his observations up to really high temperatures — 

 Pionchon, a pupil of Berthelot, who worked in Paris in 1886. 

 His experiments were made by the usual mixing calori- 

 meter method, but were carried out with exceptional skill 

 and accuracy. His specimens were heated in a specially 

 constructed gas-furnace, those such as iron, which were of a 

 readily oxidizable character, being protected by enclosure in 

 a thin platinum envelope. His temperatures were measured 

 by the Violle method of calorimetric estimation, using a 

 small ingot of platinum heated alongside the specimen. 

 Employing the values previously determined by Violle for 

 the specific heat of platinum, Pionchon found for the nielting- 



* Communicated by the Physical Society : read May 26, 1905. 



