:286 Mr. G. H. Livens on the 



The annealing temperature of silica was not determined, as 

 the small temperature expansion coefficient and good heat 

 conductivity rendered it somewhat difficult to produce ob- 

 servable strain by heating in the small pieces of transparent 

 silica which were available. 



Two plates of silica glass 0*75 in. X 0*5 in. x 0*1 in. were put 

 into optical contact with the ends of a silica ring 1*0 in. diam. 

 - andO'5 in. long, in order to make a small cell suitable for 

 investigating the ultra-violet absorption spectra of liquids. 

 This was held in a steel clamp, with a few thicknesses of 

 paper inserted between the steel and the silica, and heated to 

 1030° C. for 1 hour. 



While cooling, contraction of the clamp fractured the cell, 

 but portions ot' it, examined when cold, were found to have 

 joined. On forcing them apart some of the material of the 

 plate was torn away by that of the ring, and vice versa ; from 

 the nature of the union we judge that a rather higher tem- 

 perature, probably in the neighbourhood of 1100° C, will be 

 required to give results equal to those attained with glass. 

 The planeness of the silica plates was found to be unchanged. 



In conclusion, we wish to thank Mr. F. Twyman, of 

 Messrs. Adam Hilger, Ltd., for continued help and many 

 suggestions. 



Research Laboratories, 

 Adam Hilger, Ltd. 



XXVII. On the Energy in the Electromagnetic Field. 



To the Editors of the Philosophical Magazine. 



The University, Sheffield, 

 DEAR SlRS,— February 2nd, 1916. 



IN your issue of last month * Mr. E, A. Biedenn <nn 

 discusses some consequences of the assumption that 

 the magnetic or kinetic energy density in the electro- 

 magnetic field is H 2 /87r, and as the outcome of this discussion 

 he proposes a modified form for this expression, viz. : 



~ {H 2 + (divA) 2 }, 



H being the magnetic force vector and A the magnetic 

 vector potential. 



As this new form of expression for the density would 

 carry with it all the mathematical and physical uncertainties 

 involved in any theory interpreted in terms of the scalar and 



* P. 146. 



