Insulating Properties of Indiarubber. 509 



will be seen that both sets of readings at 51 volts give values 

 of the resistance which are higher than those obtained when 

 charging at 197 volts. On July 6th, an experiment was made 

 in which the pressure was first increased by steps to a maxi- 

 mum and then diminished in the same steps. The values of 

 the resistance so obtained are given in Table II. It will be 

 seen that on dropping the pressure from 197 to 121 volts the 

 current increases as electrification at the lower voltage pro- 

 ceeds, and after 60 minutes gives a nearly constant resistance 

 of 1536 megohms, which is considerably higher than the 

 value which was obtained when the pressure was increasing, 

 viz., 1329 megohms. On the pressure again falling to 

 51 volts, the current for the first few minutes opposes the 

 charging battery. The resistance calculated from the current 

 after 72 minutes' electrification was 2016 megohms, while 

 the value obtained when first charomo- at 51 volts was 

 1316 megohms* 



9. The cable was again placed in the tank on November 

 27th, 1900, and was kept in until the end of the experiments. 

 On December 14th, an experiment was made to determine 

 the curve of charging current. The pressure was kept on 

 for 24 hours, and at the end of that time readings of the dis- 

 charge current were taken. The results are given in figs. 3 

 and 7 . Jt will be seen that the observed values of the charging 

 current lie very approximately on the broken line curve which 

 is plotted from the equation 



c-l-60xl0-- 8 =ll-7xl0- 8 *- 181 . 



This equation has been determined by assuming a probable 

 value for the constant to he subtracted from the current 

 readings. \\' the assumed value he too great, the curve log 

 (c — a constant) plotted to log "/" is concave to the origin, 

 and vice versa ; the correct value of the constant makes the 

 curve become a straight line. If logarithms of the discharge 

 current be plotted to logarithms of the time which has 

 elapsed since the cable was short-circuited, the points lie on 

 the straight line 



log c = — 8'06O— "343 log /. 



The discharge current may therefore be represented by the 

 equation 



c=87'lxl0- 8 i- 348 . 



On March 9th two experiments were made. In the first, the 

 cable was charged at 104 volts for 20 minutes, and readings 

 of the discharge current taken ; in the second the pressure 

 was 206 volts, the time of charge being the same and the 

 temperature practically constant for the two experiments. 



