1895.] on Atmospheric Electricity. 465 



be dangerous, equalisation may take place through the air gap to earth 

 by means of a small spark. So far, the air gap answers its purpose, 

 but as soon as a spark passes through the gap, it destroys the insulating 

 power of the air, and the main current consequently takes a short-cut 

 through the gap. At Pontresina, in the Engadine, lightning conductors 

 put up in this way are so sensitive that a flash of lightning several 

 miles away causes a small spark by induction, and instantaneously 

 puts out every electric lamp in the town. 



If we accept the view that an electric discharge destroys the 

 insulating power of the gas, it follows that the outer regions of 

 the atmosphere must conduct, for we have ample reason to suppose 

 that electric currents are passing continuously through those regions. 

 The aurora borealis in the arctic regions is, according to Norden- 

 skiold's observations, a permanent phenomenon, and the diurnal 

 changes of terrestrial magnetism show that in our latitudes electric 

 currents traverse the air above us. However small a conductivity we 

 may assign to the atmosphere, the earth could not remain electrified 

 inside such a shell of partially conducting gases. Lord Kelvin drew 

 the same conclusion in the Royal Institution lecture, on the assump- 

 tion that gases at much reduced pressures cease to insulate. We 

 may leave it an open question whether the normal electric stress 

 could in itself cause a discharge in the outer regions ; but we cannot 

 deny that under existing conditions these regions do not insulate, and 

 Lord Kelvin's argument still holds good. 



But the question of the ending of the lines of force — in other 

 words, the location of the positive charge corresponding to the 

 negative electrification of the surface of the earth — can only be solved 

 by balloon or kite experiment, and we may briefly mention the more 

 important results which have so far been obtained. 



Observations made up to heights of about 1000 feet seem to 

 indicate a strengthening of the electric field — i. e. the fall of potential 

 per metre is greater at a height of, say, 200 metres than on the 

 surface of the earth. The observations of Dr. Leonhard Weber * 

 bring out this point clearly. In one case the fall of potential at a 

 height of 350 metres was found to be six times that at the earth's 

 level. This increase is in itself not surprising, if we remember that 

 every particle of dust raised from the ground must itself be nega- 

 tively electrified, and probably the observed increase in the electric 

 force is sufficiently accounted for by the presence of electrified dust. 



Observations made at greater heights in balloons, on the other 

 hand, seem clearly to indicate that this increase soon ceases, and that 

 a diminution already takes place at moderate heights. Thus the 

 observations of Dr. O. Baschin f gave for the fall of potential in volts 

 per metre the numbers 49, 28, 13 at heights of 760, 2400, 2800 metres 

 respectively, and at a height of 3000 metres no measurable fall at all 



* 'Elektrotechmsehe Zeitscbrift,' April 1888. 

 t ' Meteorologische Zeitschrift,' September 1894. 



