June 7, 1883] 



NATURE 



129 



with sulphuric acid, the outer surface of the glass tube, 

 ad, will be kept dry, and almost completely insulated. 

 The distance between the jar and the bell should be as 

 great as possible, in order that the hoar-frost may not 

 form a bridge across the intervening space. 



From this apparatus a telegraph wire is led on poles 



Fig. 



provided with insulators to a convenient chamber of 

 observation. The conducting wire may, when the above- 

 described kind of insulators is used, be an ordinary iron 

 wire 2 mm. in diameter. The poles should be at most 

 40 metres apart. 



The galvanometer should be constructed with a great 



« U f 



Fig. a. 



number (about 10,000) coils, and be provided with a pair 

 of astatic needles, near which the mirror is affixed. In 

 order to avoid too great oscillations, the needles should be 

 hung side by side on fine threads of cocoon silk, the 

 distance between the ends of the threads may be regu- 

 lated according to circumstances. The readings should 



= k 



be made with a telescope and scale. It is besides clear 

 that the conductive resistance of the galvanometer 

 should be exactly measured, and that the readings of the 

 instrument should be verified, as, for instance, with an 

 inductor whose action on the galvanometer has been 

 ascertained in absolute measure. For this purpose an 

 ordinary Daniell's element may serve, and may in fact 

 be the best, as a similar element should also be used for 

 the actual measurements. If an iron wire is used it 

 must of course be replaced with one of copper near the 

 chamber of observation. The earth conductor of the 

 current is a zinc disk about 4 square decimetres in area. 

 The theory of the apparatus is this : — The entire quantity 

 of electricity which is suffused in a certain part of the 

 atmosphere which is situated above a certain horizontal 

 plane, as that formed by the points in the " utstromnings " 

 apparatus, produces in each one of these points an electro- 

 motive force. And if the potential of these quantities ot 

 electricity on all the points be denominated as V a , and 

 the potential of the aggregate electricity on the zinc disk 

 as V : , the electromotive force E will be — 



E = Va - Vz, 

 and the strength of the current i 



R 



where R denotes the whole conductive resistance, and 

 k a constant dependent on the construction of the gal- 

 vanometer, &c. 



Generally, V z is assumed = o j but this is, in the present 

 case, not correct ; we therefore put — 



• /£ 



—**■ 



If a constant element is introduced into the current, we 

 have, if the electromotive force is denominated e, and the 

 internal resistance of the element is not taken into account — 



when i\ means the intensity of the current which is created 

 by the galvanic element. 



If the positive pole is turned first against the " utstrom- 

 nings " apparatus, we obtain — 



and if the negative pole is turned against it — 

 i+t\ = V, 



i.e. if 8 and 8" means the deflexion of the galvanometer in 

 each case. 



We obtain therefore in i a measure of - , and in i, 



R l 



a measure of — . If the deflexions are always reduced 

 R 



to the same value for . which is easily done as e is 



constant, we obtain measures capable of being compared 

 with E or Va — V z . 



In the deflexions observed when a constant galvanic 

 element is introduced into the circuit, one obtains, when 

 the element is turned in the first instance with the positive 

 pole against the "utstromnings" apparatus, and in the 

 second against the earth-plate, a relative measure of the 

 potential due to induction in the air on the particular 

 occasion. From this it will appear that the observations 

 should always be effected in the following manner : — 



1. With the constant element in the current — 



(a) With the positive pole against the apparatus. 

 (i) With the positive pole against the disk in the 

 earth. 



2. Without the element in the current — 



(a) First deflexion. 

 (6) Constant deflexion. 



