316 



ATMOSPHERIC ELECTRICITY. 



used in the determination. Thus during the early morning from .5 a.m. to 10 a.m. the 

 potential is high and the observations are numerous. Daring the period of minimum potential 

 from 1 P.M. to 7 p.m. there are relatively few observations. If the numbers are examined 

 in more detail, it will be seen that the paralleHsm is very close. It will also be seen that 

 the same relationship is repeated in each of the four seasons, the period of high potential 

 coincides with the period of numerous observations and vice versa. 



That there are more observations in the early morning and fewer in the early afternoon 

 is not chance, it is due to the fact that the air is more often calm in the early morning 

 than in the afternoon. This is clearly brought out in the curves on figure 33, page 102, 

 which show the frequency with which calms (0—1 miles per hour) occurred during the same 

 period as that for which we have the potential gradient observations. The similarity between 

 the curves showing the daily variation of calms and the daily variation of potential gradient 

 (figures 33 and 92) is most remarkable and points to some close relationship between the 

 two phenomena. 



Reasons for such a relationship do not appear difficult to find. 



It is a well-known fact that, generally speaking, when the electrical conductivity of the 

 air is low the potential gradient is high, also that a high humidity of the air causes a 

 low3ring of the conductivity. Thus one would expect p3riods of high humidity to coincide 

 with high potential gradient. Unfortunately no humidity measurements were made at Cape 

 Evans, but during calm weather one might expect the humidity to depend largely on the 

 temperature. 



Now during the greater part of the year the temperature is lowest in the early morning 

 and highest in the afternoon, we might therefore conclude with a certain amount of justifica- 

 tion that the humidity is highest in the early morning and lowest in the early afternoon. 

 Thus we see that the high potential might be accounted for in this way during the greater 

 part of the year ; but what about the remaining part ? We have already shown (page 67) 

 that during the three winter months, May, June, and July, the temperature variation is very 

 small, and what there is of it makes the afternoons warmer than the night and early morn- 

 ing. It must, however, be pointed out that all hours were used in this determination. Would 

 the effect be the same if only the calm hours had been considered ? The fact that there 

 are more calms during these months in the early morning than in the afternoon would lead 

 us to expect that the afternoons would be the warmer ; for we have repeatedly shown that 

 the more the wind the higher the temperature. 



To test whether or not this is so the temperature of every hour during May, June, and 

 July, for which the potential gradient has been used, has been tabulated and the daily 

 variation of the temperature of these hours determined. The following table contains the 

 results and shows that the temperature is practically constant throughout the twenty-four 

 hours and that there is no temperature variation at all similar to that of the potential 

 gradient variation : — 



Table 159. 

 Mean temperature of the ho'irs from which the potential gradient variation has been determined 



in May, June and July. °F. 



