June 4, 1903] 



NATURE 



i03 



For the study of the variations of the electric field 

 at a given place a large mass of material is furnished 

 by the electrograph curves obtained at various observ- 

 atories. There is a well-marked annual variation in 

 the intensity of the electric field ; the maximum occurs 

 in winter and the minimum in summer, the midwinter 

 values being five or six times as high as those of mid- 

 summer. The daily variation is less regular, and its 

 character depends on the place of observation and on 

 the season of the year. Three types are recognisable 

 according to Exner. Most commonly there are 

 maxima about 8 a.m. and 8 p.m., with night and noon 

 minima between them. There may, secondly, as on 

 the Eiffel Tower ^ and in winter at many low level 

 stations also, be a minimum in the early morning 

 hours, and a flattened maximum over the day hours. 

 Finally, as in Ceylon and on the Indian Ocean, there 

 may be no daily variation. 



A great advance was made in 1899 by Elster and 

 Geitel. They proved, in agreement with previous ex- 

 periments of Linss, that an electrified body exposed 

 in the open air loses its charge comparatively rapidly 

 by leakage through the air ; the leakage is more rapid 

 the clearer and more free from dust the air may be. 

 They showed that the phenomena were entirely in 

 agreement with the supposition that the atmosphere 

 contains positively and negatively charged ions free 

 to move under the action of the electric field. An 

 interesting account of the application of our know- 

 ledge of gaseous ions to the explanation of many of the 

 phenomena of atmospheric electricity has been given 

 by Geitel. - 



Charged conductors exposed in the open air are 

 found to lose i or 2 per cent, of their charge or more 

 per minute; the leakage from negatively charged 

 bodies is often somewhat greater than that from posi- 

 tively charged bodies ; this difference is especially 

 great on mountain peaks, where a negative charge 

 may be neutralised many times as fast as a positive 

 one, indicating an excess of positive ions. Ebert ' 

 found in balloon ascents an increased rate of neutralisa- 

 tion in the upper atmosphere as on mountain peaks, 

 but without any marked difference between positive 

 and negative leaks. Many observers, especially in 

 Germany, have lately been carrying out measurements 

 of this " Elektricitatszerstreuung. " 



There have, however, been very few absolute 

 measurements from which the number of ions present 

 per c.c. in the open air could be determined. Measure- 

 ments of this kind have been made by Ebert and by 

 Rutherford and Allen. The latter observers found 

 {Phil. Mag., December, 1902) for the number of Ions 

 per c.c. of air drawn in from outside their laboratory 

 values which on certain occasions were as low as 30 

 per c.c, the charge carried by each ion being 

 about 3x10"*' electrostatic units, according to recent 

 determinations by J. J. Thomson {Finl. Mas;., March) 

 and by H. A. Wilson {Phil. Mag., April). Rutherford 

 and Allen also showed that the velocity of the ions of 

 the free atmosphere under a given strength of field 

 was approximately the same as that of the ions pro- 

 duced by Rontgen and Becquerel rays, being about 1.4 

 cm. per second for a potential gradient of a volt per 

 cm. ; we are probably therefore justified in assuming 

 an identity in other properties also. With the above 

 values for the number of ions and their velocity, the 

 charge on the ground should be neutralised at the rate 

 of about a half per cent, per minute. 



In connection with the question of the origin of the 

 ions in the atmosphere, some remarkable phenomena 



1 ChativeaUj CR., vol. cxvii. p. 1069(1893). 



2 *' Ueber die Anwendung der Lehre von den Gasionen auf die Erschein- 

 ungen der aimospharischen Elektricitat." (Braunschweig, 1901.) 



^ " Terrestrial Magnetism," vol. vi. p. 97 (1901). 



NO. 1753, VOL. 68] 



have to be considered. Even in dust-free air in a 

 closed vessel in the dark there is a continuous produc- 

 tion of ions, generally at rates not differing greatly 

 from 40 per c.c. per second, if we interpret the measure- 

 ments in the light of the most recent determinations 

 of the ionic charge. It has, however, been shown by 

 McLennan and Burton,' and by Strutt (Nature, 

 February 19), that the greater part of the effect 

 is due to the walls of the vessel, that ordinary 

 substances in varying degrees resemble radium 

 in being radio-active and producing radio-active 

 emanations, the effects, however, being of incomparably 

 smaller intensity. The two first-named experi- 

 menters also found that a part of the ionisation 

 is due to an extremely penetrating radiation from 

 sources outside the vessel. Rutherford and Cooke 

 (Nature, April 2) have obtained a similar result. 

 Elster and Geitel found that negatively charged bodies 

 exposed in the open air become temporarily radio- 

 active, just as they do when exposed to the emanations 

 from radium or thorium. ' Vessels in which freshly 

 fallen rain or snow have been evaporated to dryness 

 show a similar temporary radio-activity.^ The atmo- 

 sphere apparently contains an emanation like that from 

 radium. Air pumped out of the ground shows these 

 effects to an abnormally marked degree, as Elster and 

 Geitel have proved. The surface of the ground, and 

 to a still greater extent the exposed portions of trees, 

 must, it will be observed, under normal fine weather 

 conditions become radio-active in virtue of their nega- 

 tive charge, and produce, therefore, an abnormal 

 amount of ionisation in the air near them. 



It is probable, in the light of Lenard's experiments, 

 that sunlight ionises the air which it traverses, 

 especially in the upper atmosphere, while it is still 

 strong in ultra-violet rays. 



The conductivity of the air increases in a sense the 

 difficulty of the problem of the origin of the earth's 

 electric field. For it would seem that the electric field 

 in fine weather regions should rapidly diminish, and in 

 a few hours disappear; t}ipj:/> rpn st be som e process- 

 by w hich the electric field is continuallv bein^ re- 

 genecatetU i^eavmg aside, however, the consideration 

 of the origin of the electric field, we may atternpt to 

 explain its variations as due to the variations in the 

 conditions determining its rate of destruction. What- 

 ever increases the conductivity of the air will diminish 

 the electric field, and vice versd. Examples of the 

 application of this principle will be found in the paper 

 by Geitel already mentioned. To take only one, the 

 increase in the electric field accompanying fogs (a 

 phenomenon well shown in the Kew electrograph 

 curves) may be explained as due to the entangfing of 

 the ions by the fog particles ; the leakage of electricity 

 under such conditions has been found by Elster and 

 Geitel to be very slight. 



In regions enjoying fine weather, if we assume the 

 existence of a flow of electricity in the direction of 

 the electric field, there will be a downward earth-air 

 current ; there must then be a compensating current 

 accompanying precipitation, negative electricity being 

 brought down in the rain, and the positive charge 

 being left behind in the atmosphere and carried by 

 upper air currents to other regions. There is, as we 

 shall see later, reason to believe that an excess of 

 negative electricity is brought down to the earth's sur- 

 face by rain. It is, however, doubtful whether we can 

 explain in this way the existence of the normal electric 

 field at a distance from regions where rain is falling; 

 for the positively charged upper air currents would 

 continually be losing their charges, and we should 

 expect a rapid falling off in the intensity of the field 



1 In a paper read before the American Physical Society, December, 1902. 



2 C. T. R. Wilson. Camb. Phil. Proc, vol. xi. p. 428; vol. xii. pp. 17 

 and 85 ; M'Lennan, P/ii7. Mag., April. 



