118 



rainfall observations would agree so closely with those of 

 barometric pressure and temperature. 



Instead of comparing the differences between the amounts 

 of rainfall it would perhaps be more correct to compare 

 their ratios, but the results would be substantially the same. 

 Thus dividing the entire series of 11 years into 3 groups, 

 the first including the four years 1858-61, one of which was 

 a year of maximum frequency of solar spots ; the second the 

 four years 1862-65 ; and the third the three years 1866-68, 

 one of which was a year of minimum frequency, we have 

 the following amounts and their ratios : — 



Sum of Sum of 



Rainfall under Rainfall under 



S.E. & S. winds. S.W, &W. winds. Ratio. 



Inches. Inches. 



4 years 1858-61 40-24 62-13 0-64 



4 years 1862-65 49-10 40-05 1-22 



3 years 1866-68 56-06 26-09 2*14 



Here we have a small ratio in years of maximum solar 

 activity, and a large ratio in years of minimum, and a ratio 

 of intermediate value for the intervening years. 



It will I think be admitted that the results of this inves- 

 tigation support very strongly the hypothesis which led me 

 to undertake it. They show also strikingly that the future 

 progress of meteorology must depend to a much greater 

 extent than has been generally supposed upon the know- 

 ledge we may obtain of the nature and extent of the 

 changes which are constantly taking place on the surface of 

 the sun; and therefore, in the interests of meteorological 

 science, it is evidently very desirable that observations of 

 solar phenomena should be greatly multiplied by the estab- 

 lishment, in various parts of the world, of observatories 

 specially devoted to this object, so that a continuous daily 

 or even hourly record may be obtained of the state of the 

 solar disc and its appendages, and the results discussed in 

 connection with those of observations of meteorological 

 phenomena, 



