i riq^i 



Electrical Conductivity of Mica in Intense Fields. 125 



This close agreement over so large a rang*' is remarkable, 

 and suggests that there may be some definite physical basis 

 for the formula. However, it may be argued that the 

 formula can only be an approximation, as, if it were a 

 general one applicable to all fields, we would expect it only 

 to contain odd powers of X, whereas it, or its equivalent 

 C = a\e bX , contains both odd and even powers. The formula 

 C = aX(e bX + e~ bX ), which only contains odd powers, is in- 

 distinguishable from the preceding one over the range 

 covered, as the second term is negligible. The first of these 

 formulae gives a value 5'3 x 10 14 ohm cms. for the specific 

 resistance of mica in weak fields, in which Ohm's law 

 would be approximately obeyed, while the second gives- 

 2'65 x 10 14 ohm cms.: in this case the agreement with Ohm's 

 law would be much closer, and would extend over a much 

 greater range, as will be seen from a comparison of the 

 curves in fig. 3. As the value given by Kaye and Laby for 

 the specific resistance of mica is 9 x 10 15 ohm cm., it seemed 

 to be desirable to make some determinations of the con- 

 duction current in weaker fields, using an electrometer to 

 measure the current. 



Accordingly some experiments were carried out in this 

 way, the current being measured by the rate of rise of 

 potential of a standard ^ microfarad condenser, the potential 

 being measured by means of a Dolezalek electrometer whose 

 sensitivity was about 100 scale-divisions per volt. The 



values of ^ or the specific conductivity are plotted against 



X in fig. 3. Here, as before, C is in microamperes per sq.. 

 cm., and X is in mega volts per cm. The upper curve has 

 for its equation C = aX(V x -f-£ -5X ), and the second curve 

 the equation C = aXe bX , where a and b are the values ob- 

 tained from the previous results, so that either of these 

 curves will fit the values obtained with large values of X.. 

 It will be seen that they both fail to represent the current 

 correctly at small fields, though giving results of the right 

 order of magnitude. The simpler formula differs less from; 

 the observed value than the other. The measurement of 

 these small currents is rendered very uncertain by soakage 

 and polarization effects, w r hich make it hard to know how 

 much of the observed current is to be ascribed to conduction. 

 The reading marked ® was obtained after the voltage had 

 been on for an hour ; it lies slightly above the value previously 

 obtained at this pressure, indicating a small rise of the 



