Intelligence and Miscellaneous Articles, 71 



that the electrification remains adherent to the strata of air which 

 were directly in contact with the flame during the discharge. The 

 electrified gases ascend by virtue of their raised temperature, then 

 move and disseminate themselves after the manner of smoke until 

 they are uniformly distributed in the atmosphere of the room : the 

 deflection of the galvanometer is then near its maximum. As to 

 the disappearance of the electricity, it takes place either by ex- 

 change with the outer air, or by the receiving flame itself conti- 

 nually neutralizing the surrounding electricity, or by contact of 

 the air with the walls of the room. 



The loss will be diminished when the movements of gas due to 

 the presence of the flames are suppressed. This is shown by expe- 

 riment when the discharge is effected by a sharp point and the 

 receiving flame is replaced by running water : the needle of the 

 electrometer is again deflected immediately the discharge com- 

 mences ; but after this it remains some time stationary, attains 

 the maximum deflection a little later, and the return to zero takes 

 place more slowly. At the end of an hour the loss w T as only two 

 thirds ; it would doubtless be much slower in air absolutely still. 

 We have moreover obtained an identical proof of it by studying the 

 air enclosed in a room communicating with the outside only by the 

 usual leakages of the doors and windows : electricity was almost 

 always found there, of the same sign as that of the outer air ; for 

 opening a window was sufficient to augment greatly the indications 

 of the instrument. 



The electrification produced by a Leyden jar is always very 

 slight; but much more energetic effects can easily be obtained. 

 On the electricity furnished by a Holtz machine being discharged 

 by a flame during one minute, the air w ; as so much electrified that 

 the potential about the middle of the room at the instant of the 

 maximum exceeded 2000 volts ; from this the mean density of the 

 electricity in the air can be deduced, supposing its distribution 

 uniform. 



From these experiments it follows that, for the examination of 

 the lower strata of the atmosphere, it is sufficient to determine the 

 potential in a room of some metres dimensions, the walls of w 7 hich 

 should be formed by wire netting wdth large meshes, in communi- 

 cation with the ground, in order to eliminate the action of external 

 electric masses. The exchanges of gas with the atmosphere, how- 

 ever gentle the wind may be, will be sufficient to compensate the 

 loss produced by the wails and the collecting-apparatus (flame or 

 running water), and to communicate to the electrometer a potential 

 constantly proportional to the electrification belonging to the sur- 

 rounding air. That potential will be quite different (most fre- 

 quently of contrary sign) from that obtained by the usual methods. 

 If electricity plays an important part in natural phenomena, it is to 

 be presumed that the proper electrification of the air is peculiarly 

 effective : Sir W. Thomson's suggestion therefore merits the atten- 

 tion of all observers. — Comptes Rendus de V Academic des Sciences, 

 Nov. 13, 1882, t. cxv. pp. 917-919. 



