Variation of Atmospheric Electrical Quantities. 293 



mean, and is more or less distinctly apparent in the indi- 

 vidual days except August 31. Thus it seems to indicate a 

 general phenomenon. It was often noticed at Eskdalemuir 

 during this month that an early morning ground-mist was 

 dissipated between 8 and 10 a.m. and was followed by bright 

 sunshine with cloudless sky lasting a short time but be- 

 coming overclouded towards noon. Although at 10 A.M. 

 the mist had quite cleared on the slopes near the Obser- 

 vatory, it was often visible in the valleys. It is possible 

 that the dissipation of the valley-mist to higher levels might 

 cause a decrease in the mobility of the ions, and so in the 

 conductivity. 



In order to study the eclipse effect more closely, the 

 results for August 20 and 22 may be considered. The 

 general form of the variation for both days was similar 

 except that on August 22 the increase in the conductivity 

 and the electric charges from 7 to 10 a.m. was very marked, 

 and also that instead of a gradual fall from 4 to 8 p.m. there 

 was an increase, this being especially marked in conduc- 

 tivity. The mean for the two days shows a deeper depres- 

 sion in conductivity between 10 a.m. and 1 p.m. than is 

 obtained on August 21. This reduces the probability that 

 any eclipse factors produced the diminution of electrical 

 quantities on the day of eclipse between these hours. 

 But when we consider the retarded maximum in the con- 

 ductivity and both charges, there is some justification for 

 considering it to be the result of an indirect solar effect. 

 Some observers have considered it sufficient in eclipse work 

 only to investigate the electrical variation during the actual 

 phases, but a lag in producing any electrical effect would 

 make later hours important. Wiechert's reasoning * with 

 respect to solar radiation and the maintenance of atmo- 

 spheric ionization, indicate the probability of a lag in any 

 eclipse effect which may be due to the suspension of the 

 ionizing influence of ultra-violet light. Although most 

 eclipse observers in the past do not appear to have noticed 

 any striking electrical changes, it is to be hoped that a 

 longer series of observations will be possible in the future at 

 stations situated near the line of totality r with sufficient 

 observations on preceding and following days to obtain the 

 normal form of the diurnal variation. 



The few results obtained by G. Dobson f and other ob- 

 servers { show generally a diminution of conductivity and 



* " Les Recherches sur l'Electricite Atmospheric) ue," Arch, des Sc, 

 Phys. et Nat. 1912, p. 385. 



t Q. J. Roy. Met. Soc. 1913, p. 221. 



% Ludeling & Nippoldt, Abhand. d. Kon. Preuss. Met. Inst. 1908. 



