256 



General Strachey on the Correspondence of 

 Table VII. 



[May 31, 



Rainfall near periods of maximum sun-s 



jpots. 



Uainfall near periods of minimum sun-spots. 



Year. 



Two 

 years 

 before. 



One 

 year 

 before. 



Maxi- 

 mum. 



One 

 year 

 after. 



Two 

 years 

 after. 



Year. 



Two 

 years 

 before. 



One 

 year 

 before. 



Mini- 

 mum. 



One 

 year 

 after. 



Two 

 years 

 after. 



1817 



in. 

 - 4-4 



in. 

 +2-6 



in. 

 + 1-7 



in. 

 -15 



in. 

 +3*3 



1823 



in. 

 4-6-6 



in. 

 +0-1 



in. 

 -0-5 



in. 

 +8-1 



in. 

 -2-6 



1 SQH 



_l_ Q.Fi 

 + oO 



— A o 



-rO A 



iQ.fi 



+4-8 



1 S^LL 







1 -1 

 — 11 





+ o 1 



1837 . 



+ 5-7 



+61 



-5-8 



-33 



+2-2 



1844 



-4-9 



-2-4 



-30 



-6-2 



-1-4 



1849 .. 



-121 



+5-3 



-10 



-5-2 



-3-3 



1856 



-5-9 



-1-1 



-3-0 



-3-5 



-71 



1860 



- 7-1 



4-1-0 



+7-1 



-4-5 



+ 1-6 



1867 .. 



+38 



+5-8 



+3-5 



+0-2 



-0-9 



Mean... 



- 2-9 



+2-5 



+ 1-6 



-1-0 



+1-7 



Mean ... 



4-0-9 



413 



-0-8 



4-4-3 



-1-2 



Mean 

 difference 

 from mean 



1 



1 6-0 



2-5 



4-0 



42 



2-1 



Mean 

 difference 

 from mean 



I 50 

 J 



2-9 



1-9 



4-8 



3-0 



Mean of mean differences 3" 



8 inches. 



Mean of mean differences 35 inches. 



In this case the mean deviation o£ the original observations from the 

 arithmetical mean of the whole series (24-9 inches) is 4-1 inches. This 

 result therefore is quite analogous to that obtained from the Indian 

 observations. 



I have not attempted to make detailed calculations in other cases, but, 

 so far as I can judge, the evidence of the alleged periodicity will be gene- 

 rally found to fail when it is tested by comparison with the individual 

 observations on which it has been made to rest. 



It will serve to illustrate the argument on which this paper is based 

 if we consider what would be the consequence of applying it to a case in 

 which a well-ascertained periodicity exists, as that of the diurnal baro- 

 metric oscillation. The following Table gives an example, taken at 

 random from an old Madras register, the intervals being made two-hourly, 

 so as to reduce the number of calculations. The entries are the differ- 

 ences of the observed barometric heights from the mean of the whole in 

 thousandths of an inch. 



