fiB MR. GEORGE C. SIMPSON ON THE 



The yearly course of the potential gradient is shown in Curve I., fig. 1, on which 

 each point gives the mean potential gradient for a week, the means for the months (as 

 in Table I.) being shown by points enclosed within circles. It will at once be seen how 

 irregular the potential gradient is during the winter when taken for such short time 

 intervals as a week ; on the contrary the monthly means fall very nearly on a regular 

 curve. It must be remembered that, as stated above, these values during the winter 

 are obtained from both fine and disturbed days. If only fine days had been used not 

 only would the curve have been more regular, but also the mean potential gradient 

 would have been greater. The trend of the curve may be summed up as a regular 

 rise in the potential gradient from. October to the middle of February, followed 

 by a more rapid fall until the end of May, after which the potential gradient remains 

 nearly constant during the summer months. 



Dissipation. The mean values of the dissipation for each month are shown in the 

 following table. In order to find the effect of the seasons, and whether the total 

 absence of the sun for nearly three months during the winter and its presence for an 

 equal length of time during the summer influences the electrical conditions of the 

 atmosphere, the observations have been grouped into periods of three months, the 

 winter three months containing the period of no sun and the summer three months 

 that of permanent sun. 



TABLE II. Dissipation. 



On Curve II. these values of the dissipation (a ) have been plotted, also the 

 weekly values. If no observations were made for a week a gap has been left in the 

 curve. From the curve it will be seen that the yearly course of the dissipation is 

 strikingly similar to that of the potential gradient when inverted, the one falling and 



