Allen — The Electrical Resistance of Porcelain. 



297 



Sample D was also tested for voltages up to 230 at a temperature 213° C, 

 and the results obtained were : — 



The investigations of Curtis and others show that the insulation 

 resistance of some materials depends upon the applied voltage. His results 

 are tabulated 1 for different substances for a range of 50 to 500 volts, and 

 only one-third of the number given show variation of resistance with the 

 applied voltage. 



Another test was tried on sample B to determine whether, by using a 

 given area of contact-surface in one case and then diminishing it to one-half, 

 the resistance of the sample would be doubled. The results obtained were as 

 follows ; the resistance being in megohms : — 



The resistance is therefore practically doubled in the Latter case, and this 

 suggests an absence of creepage over the top of the walls of the sample. 



The samples, excepting E, were heated by means of a sandbath, and those 

 in the form of a dish were loaded with mercury to a suitable height, and 

 floated on mercury contained in a larger vessel. Sample E was placed in 

 another tube, which was immersed in oil heated electrically by a spiral of 

 manganin wire wound round this tube. With care a steady temperature 

 could be obtained during a set of observations. 



It must be noted that, on account of dielectric absorption in some of the 

 samples, the galvanometer sometimes took several minutes before indicating 

 a steady value. This was accompanied by a directional effect ; the resistance 

 being greater in one direction through the walls of the sample than in the 

 other. 



1 Bulletin of the Bureau of Standards, 1915, vol. ii, no. 3, p. 359. Table I. 



SCIENT. PROC. R.D.S., VOL. XV., NO. XXVII. 3 C 



