126 Prof. W. M. Thornton on the 



pure, has absorption*. When ellipsoids or! such materials 

 are suspended in a steady field of force there is a true 

 instantaneous polarization, and a slow increase ichich proceeds 

 in every case to a constant final value independent of the 

 intensity of the external field ~\ . We may therefore write 



i=L^^ » 



for the extra displacement observed, though independent of 

 v, can be regarded as caused by an apparent increase of v 

 when h is constant. 

 In this case 



so that 



hav lev 2 



Jcav kv 2 , 



~ 4:7TIV $7TW ^ ' 



If, as in paraffin, w is constant, we should have £ = Au + Bi' 2 . 



Such a relation was given by Steinmetz for air J, and the 

 writer was informed by Dr. Kapp in 1910 that he had found 

 it to fit the materials quoted by Baur§ better than the 

 formula t = av* used by him. Baur's statement is, "Every 

 dielectric, whatever its thickness be, requires a certain voltage 

 to break it down, and this is proportional to the two-thirds 

 power of its thickness " || . This is certainly not true for 

 paraffin, for the differences between the observed values and 

 those calculated by Baur vary from + 35 to — 12 per cent., 

 while those in Table I. from a quadratic formula vary from 

 + 1*7 to —1*8 per cent. For porcelain his differences range 

 from +10 to —24, and for air from +19 to —35 per cent. 



The following tables show how the best experimental values 

 of the breakdown voltage are represented by t = Av + Bv' 2 , 

 where A and B are determined by drawing a tangent to 

 each curve through the origin. 



* Phil. Mag. loc. cit. p. 398. t Loc. cit. § 4. 



\ C. P. Steinmetz, " Notes on the Descriptive Strength of Dielectrics," 

 Trans. Amer. I. E. E. vol. x. p. 64 (1893). 



§ C. Baur, ' Das Elektrische Kabel,' 1910, p. 46. 



!| Baur, " On the Electric Strength of Insulating Materials," ' The 

 Electrician,' Sept. 6, 1901, p. 759. 



