358 



Dr. T. S. Hnmpidge. 



[June 21, 



XIII. "Reply to a Note by Professor J. E. Reynolds on the 

 Atomic Weight of Glucinum or Beryllium." By T. S. 

 Humpidge, Ph.D., B.Sc. Communicated by Dr. Frankland, 

 F.R.S. Received June 7, 1883. 



In the above-mentioned note of Professor Reynolds* the author 

 criticises the results detailed in a paper which I recently had the 

 honour to contribute to the Society ,f and draws an inference from 

 the specific heats of different specimens of the metal which I cannot 

 admit to be founded on facts. 



Professor Reynolds remarks that all the results obtained by Nilson 

 and myself tend in one direction viz., to a considerable, though 

 irregular, rise in the specific heat as the impurities diminish. 



If, however, we compare the three determinations from which this 

 inference is drawn — 



Percentage of Specific heat at 



glucinum. 45° — 50°. 



Nikon 87-09 0-4084 



Humpidge 93'97 0'4453 



Nilson 94-41 0*4246 



we find that, together with a general rise in the specific heat, there 

 is also a fall between that of the second and third sample which is 

 proportionally greater that the general rise. In other words, the 

 irregularity nearly counterbalances the regularity. On the other 

 hand, as I have already stated, my result is probably slightly too high, 

 owing to the heat produced on the absorption of the turpentine by the 

 porous metal. In ISTilson's experiments the metal was enclosed in a 

 platinum capsule. 



But even admitting this general rise in the specific heat with 

 diminished impurities, it hardly appears that the specific heat of the 

 pure fused metal could be as much as 50 per cent, greater than in the 

 crystalline state. It rather appears that glucinum is either an excep- 

 tion to Dulong and Petit's law of atomic heats or to Mendeleeif's 

 periodic law. I hope shortly to contribute some further evidence to 

 the solution of this exceedingly interesting question, and am now 

 making preparations for a revision of the specific heats of the solid 

 elements in the pure state, and at different temperatures. 



Next with regard to the purity of the metal as prepared by my 

 process. The 7 decigrammes which were used for the determination 

 of the specific heat was the first sample prepared, and included all I 

 had then extracted. It ought not, therefore, to be compared with 



* Bead May 24, 1883, " Proc. Roy. Soc," vol. 35, p. 248. 

 f Read April 12, 1883, lb., p. 137. 



