427 
enced by the nature of the metals between which the interrup- 
tion spark occurs. The opinion of such a man is of great 
authority; yet it is not easy to see how this can produce such 
an effect; and the following observations appear to show that 
other circumstances must also be concerned in it :— 
«1, I tried the effect of frictional electricity from a 
machine of two 18-inch plates in powerful action. The lower 
electrode was an inch ball, screwed into the opening of the 
pump-plate (which is of glass): the upper was @ point, 7 
snches distant. The receiver was filled with dry hydrogen. 
On exhausting to 1:0, the light appeared filling the receiver, 
and did not pass in a central stream till 009. At 0°06 a 
few faint bands were seen near the ball, which, with its stem, 
had a faint envelope surrounded by a brighter one, but no 
difference of colour. The light was greenish, producing no 
fluorescence, and very much fainter than that produced by 
Ruhmkorff. Nothing was gained by including a jar of one- 
third foot coating discharging at 010. With the air vacuum 
0'-15, the light was violet, much fluorescence, and the bands 
less distinct.* 
«2, An ‘electric egg,’ 8'°5 high, and 6i diameter, was 
filled with hydrogen.} The balls of its wires, 01-25 diameter, 
were set 6! apart, and it was exhausted to 0:17 (the pump 
not acting well then). The Ruhmkorff was excited by four 
Groves giving in air a spark 0:52. When the current 
ee tt linn wal OP te Seba ted tee a tet ee oe 
* Within the last few days I had an opportunity of reading the last two 
years of Poggendorff’s “ Annalen.” In the 7th Number of this year I find an 
important observation of Herr Van Willigen. He saw dark bands in the 
discharge of the Leyden jar through a vacuum of 0i-12 (containing vapour of 
oil of bergamot), when a wet string is included in the circuit. Hence, he, 
and Poggendorft subsequently, infer that they require a certain retardation 
of the discharge. It seems rather to confirm Mr. Grove’s view of a certain 
undulation in the current being necessary. 
+ The gases were always dried by being slowly passed through a capil- 
lary tube immersed Gi in sulphuric acid. 
