125 



The mean ratio is 801, and the ratios for the years of 

 maximum solar spot frequency are all above this mean, 

 while those for minimum years are all below it, with only 

 one unimportant exception. 



In order now to eliminate as far as possible the effects of 

 accidental disturbing causes we may take the means of the 

 ratios of every three successive years, and in this way we 

 obtain the corrected ratios in the fifth column of the above 

 table. For convenience of comparison I have added in the 

 sixth column the number of groups of solar spots observed 

 in each year by Schwabe, and a glance at the two sets of 

 numbers will show the remarkably close agreement which 

 exists between them in the times of their maxima and 

 minima, which seems to me fully to justify the conclusion 

 that both classes of phenomena are intimately connected, 

 either as cause and effect, or as effects of the same cause. 



Excluding the amounts of rain which fell during calms 

 the corrected ratios become : — 



1855 2-77 1860 6-42 



1856 2-15 1861 4-37 



1857 3-32 1862 4-04 



1858 5-40 1863 2-80 



1859 6-31 



It will be observed that the course of these numbers is 

 almost identical with that of the numbers obtained when 

 the amounts of rain which fell during calms are combined 

 with those which fell under north-east, east, and north-west 

 winds. 



The close agreement which has thus been shown to exist 

 at St. Petersburg between the times of maximum and 

 minimum frequency of solar spots, and those of the varia- 

 tions in the distribution of rainfall under different winds, 

 gives increased value to the results derived from the Oxford 

 observations, and affords additional support to the hypo- 

 thesis I ventured to advance in a former paper — that 



