F. Dellmann on Atmospheric Electricity. 405 



be taken. In order to avoid the loss of electricity, inseparable 

 Irom the transport of the sphere, an insulated wire was led from 

 Its earner to the measuring instrument, in such a manner that it 

 could be conveniently drawn up with the rod. With this con- 

 trivance the apparatus was a mean between a moveable and a 

 fixed one, but for the reasons given in the introduction it could 

 not be used. With the last arrangement, of course the elec- 

 tricity Itself of the atmosphere is obtained, whereas the elec- 

 tricity opposite to that of the aii- was before carried down on the 

 sphere. For, as Peltier senior everywhere justly asserts, the 

 sphere becomes charged by induction, and not, as the youno-er 

 Peltier imagines, by conduction,— by the passage of electrickv 

 trom the earth through the observer to the sphere The 

 electricity similar to that of the air is conducted away by the 

 discharging wire h, as is proved from the fact that an insulated 

 electrometer becomes charged with the same when itsconductincr 

 wires are connected with the wire h. Although the atmospheric 

 electricity is stronger above than below, and hence acts strongest 

 on the elevated sphere, it still manifests an action, even on the 

 conducting wire which leads from the sphere to the uieasurin<' 

 instrument ; for when this wire is well insulated and drawn up 

 into the air without the sphere, a deflection is visible in the elec- 

 trometer. Hence with this apparatus, as with every fixed one 

 a mixed action is obtained,— not the pure action of 'the air sur' 

 rounding the sphere. On the other hand, the apparatus em 

 ployed by me has but the one defect, inseparable from every 

 moveable apparatus, that a loss of electricity is occasioned bv 

 the transport of the sphere. This would be no defect if the loss 

 were always proportional to the quantity of electricity • but as I 

 have already shown*, it depends upon the shell-lac, and upon the 

 relative moisture. This loss being then unavoidable, we must 

 seek to familiarize ourselves with its conditions, in order to avoid 

 it as much as possible, and render it either sufliciently small to 

 be neglected, or else capable of being estimated. For this pur- 

 pose the following measurements were employed. 



In the first place, it was found that under the most unfavour- 

 able circumstances 25 seconds were required for bringino- down 

 the sphere and charging the measuring instrument; under ordi 

 nary circumstances, however, a httle more than half this time 

 was sufficient. It must here be noticed that the electricity 

 becomes free only when the sphere is lowered, and that it is 

 only when free that loss commences. In the following measure- 

 ments the loss per cent, was always calculated for 25 seconds 

 The sphere was charged with an arbitrary quantity of electricity^ 

 • Poggendoilt's Amirilen, vol. Ixxxvi. p. 5-11. 



