158 



DR. C. CHREE: ATMOSPHERIC ELECTRICITY POTENTIAL GRADIENT 



and ordinary days were relatively few, the corresponding means were 294, 278 and 

 212. Thus at Eskdalemuir -and I think the same would prove true at Kew the 

 mean potential gradient tends to fall when we include days of negative potential and 

 of large irregular oscillations. It is thus very probable that the difference between 

 the mean values of potential gradient at Kew and Edinburgh in 1912 respectively 

 300 and 167 is due in some measure to the difference in the type of days utilised. 



Diurnal variation data were obtained by Mr. DOBSON from the three species of days 

 for individual months, and for the three seasons winter, equinox, and summer, and 

 the results are shown graphically in his paper. 



The curves for individual months and for the winter season are very irregular, but 

 those for equinox and summer are less so. The O tt equinoctial curve has larger 

 ordinates than the ordinary day curve at every hour of the 24, and except for an hour 

 or two the same is true of the summer curves. The difference in the type of the 

 diurnal variations derived from the two series of days at these seasons does not seem 

 to be large, but confirmation would be desirable from a period of years. 



Edinburgh has a much dryer climate than Eskdalemuir, and the difference between 

 all and rainless days there may well be less. In any case, in view of the comparison 

 instituted by Messrs. CAUSE and SHEARER with the older Kew data, it seemed worth 

 while to compare their results with those given by the selected days of the same year 

 at Kew, investigating at the same time how far that year was fairly representative of 

 normal conditions. 



The mean potential gradient at Kew in 1912 differed by only 4 volts from the 

 mean for the 15 years and so was unquestionably normal. 



Expressing monthly means at Edinburgh and Kew in terms of their respective 

 mean annual values, we obtain the annual variations recorded in Table XI. 



TABLE XL Mean Monthly Potentials as Fractions of Mean for Year. 



If, following Messrs. CARSE and SHEARER, we divide the year into a 6-months 

 summer, April to September, and a 6-months winter, we find for the ratio of the mean 

 winter potential to the mean summer potential 1'08 : 1 at Edinburgh and 1'46 : 1 at 

 Kew in 1912, as compared with 1'59 : 1 at Kew in the 15-year average. 



