82 ME. GEOEGE C. SIMPSON ON THE 



for the month of June being "39. On September 19 the temperature rose to 

 16-4 C., after having been below 5 for the previous few days; the air again was very 

 hazy and sultry and the ionization went down to '24, the mean for the month 

 being '44. On the contrary, a clear day in the winter would be accompanied by 

 comparatively high value of the ionization : February 22 ionization '40, mean for 

 month, -21. Much to my regret I cannot support this conclusion by actual figures, as 

 Karasjok was so enclosed by low hills that it was impossible to obtain even a rough 

 arbitrary scale of the clearness of the air by the visibility of distant objects. But 

 there can be no doubt that the maximum of the transparency of the atmosphere 

 corresponded with the maximum of the ionization. 



Potential Gradient.The yearly course of the potential gradient in Karasjok 

 conforms to the general rule for the northern hemisphere formulated by HANN* in 

 the following words : " The maximum of the potential gradient occurs in December, 

 January or February ; it falls rapidly in the spring ; remains nearly at the same level 

 during the summer and then rapidly rises again in October and November." 



The fact that the potential gradient runs so exactly opposite to the dissipation 

 makes it appear as though there were a constant charge of negative electricity being 

 continually given to the surface of the earth during the whole year, and that the 

 amount at any moment on the surface itself (measured, of course, by the potential 

 gradient) is determined by the rate at which the charge is being dissipated. How 

 this charge is supplied to the earth still remains, in spite of many theories, one of the 

 unsolved problems of atmospheric electricity. 



Two types of daily variation of the potential gradient are known, t The first is a 

 double period, having a minimum between 3 and 5 A.M. and a second about midday, 

 the corresponding maxima falling at about 8 A.M. and 8 P.M. Good examples ot 

 this are Batavia and Paris. The other type consists of a single maximum and 

 minimum, the former falling in the evening and the latter between 3 and 5 A.M. To 

 this type belong the records made at high altitudes and at some places during the 

 winter. 



The daily course of the potential gradient for the whole year at Karasjok belongs 

 to the latter class, there being only one maximum and one minimum. Taking the 

 four seasons each by itself, we see that the winter and spring curves are of the same 

 type, while that for the summer shows a slight tendency to form a minimum at 

 midday, and the autumn curve has a distinct double period. As stated above, the 

 morning and evening maxima of the autumn curve were considerably strengthened by 

 the mists which formed over the river. The nearest place to Karasjok at which 

 measurements have been made of the potential gradient is SodankylaJ in Finland, 

 and the curves for the two places are in surprising agreement. 



* ' Lehrbuch der Meteorologie,' p. 715. 



t HANN, 'Lehrbuch der Meteorologie,' p. 716. 



} ' Expedition polaire, 1882-83.' 



