40G On the Disruptive Discharge of Electricity. 



the results of my own measurement, and have also entered 

 values given or deduced from data given by other experi- 

 menters. 



Column 2 gives the relative difference of potential (or of 

 electrostatic force) required, compared with that for air, to 

 pass a *5-centim. spark between the parallel disks, the sub- 

 stance being at the atmospheric pressure. It is necessary to 

 make the comparison at a standard distance and pressure; for, 

 as has been shown, the curves for the different dielectrics, when 

 the distance or pressure is varied, are not similar to one ano- 

 ther. The gases were compared directly for the length of 

 spark mentioned ; and the ratio was taken from the mean of 

 eight or so readings. As the liquids could not be compared 

 directly for so long a spark, the comparison was made at 

 •4 centim. and the readings reduced to *5 centim. by means 

 of the appropriate laws. The paraffin examined is of low 

 melting-point. We liquefied it and poured it into the vessel 

 containing the disks, took several "3-centim. sparks through 

 it in the liquid condition, and, after allowing it to solidify for 

 twenty- tour hours, took one spark through it to determine its 

 specific resistance in the solid state. The latter is double the 

 former. 



The numbers in column 3 are the squares of those in 

 column 2 ; they are proportional to the dielectric strengths, 

 provided K, the specific inductive capacity, is constant. In 

 calculating the values of the dielectric strength for oil of 

 turpentine and paraffin, I have taken K, = 2'2 and 2 respec- 

 tively, from the table in Gordon's ' Electricity ' *. 



I have calculated the values entered under Faraday from 

 his equivalent mean intervals!, taking those for the posi- 

 tive discharge. It will be observed that the numbers are 

 uniformly less than those of column 2, and that they agree 

 better with those of column 3. This is because the electrodes 

 used were two balls, in which case the difference of potential 

 is proportional to the square root of the distance. 



Those entered under Wiedemann and Eiihlmann I have 

 calculated from the formulas | which they give for the dis- 

 charge through different gases at different pressures when 

 the electrodes are two equal balls, the negative one being un- 

 insulated. The comparison is made at 80 millims. pressure. 

 The order is the same as in column 3 and column 5. Those 

 entered under De La Rue and Miiller§ and Eontgen || are the 

 values given by these physicists. It will be observed that 



* Vol. ii. p. 134. f ' Experimental Researches/ vol. i. § 1388. 



I Pogg. Ann. cxlv. p. 364 . 



§ Phil. Trans, vol. clxix. p. 95. || Phil. Mag. Dec. 1878, p. 446. 



