Measurement of Dielectric Constants. 173 



into the calculation. It might have been expected that these 

 secondary effects, conduction, &c. would be entirely eliminated 

 by the rapidly alternating charges of a Hertz oscillation ; it 

 appears, however, that even with such extremely short times 

 of charge the secondary effects still increase the value of the 

 dielectric constant, and indeed in a greater degree than before ; 

 provided, of course, that we assume the truth of Maxwell's 

 hypothesis. 



I have investigated two solid bodies, ebonite and glass, and 

 two liquids, petroleum oil and water, by means of three dif- 

 ferent methods. The times of charge were 0*5, 5 x 10~ 4 , and 

 3 x 10~ 8 seconds respectively. 



Experiments with Hertz Oscillations. 



In a previously published paper on the study of phenomena 

 of electric resonance *, I have already given a full investi- 

 gation of all the conditions necessary in measurements by this 

 method. In the present experiments exactly the same 

 apparatus was used. In two parallel wires (fig. 1) st and s't 1 , 

 31 centim. apart, and each of length 1122 centim. (represented 

 in the figure by broken lines), electric oscillations are excited 



Fig. 1. 



U 



~r& 



in a known manner by the oscillatory charging of condensers 

 AB and A / B / by means of a spark F. From the ends of the 

 two parallel (leading) wires t and t' two flexible copper wires, 

 each of length 69*7 centim., lead to two circular condenser- 

 plates, each of radius R=9*68 centim.: these latter can be 

 moved parallel to each other by the aid of a micrometer- 

 screw. On bridging over the leading wires at certain places 

 with a metal slider, the vacuum-tube gg' begins to glow. 

 Every alteration of the capacity at the end of the wires dis- 

 places the ventral segment of the oscillation, and the slider 

 must therefore be brought to a new position in order to make 

 the tube glow once more. 



I next measured very exactly the wave-length of a few 

 oscillations with air alone between the condenser-plates. After 

 this, the condenser-disks were screwed apart, and the sub- 

 stance to be investigated placed between them. By a few 



* Lecher, Wied. Ann. xci. p. 850, 1890. 



