74 Mr. G. M. Whipple. Results of an [Jan. 15, 





1800 

 + lln 



1801 

 + lln 



1802 

 + lln 



1803 

 + lln. 



1804 



+ lltt 



1805 

 + lln 



1806 

 + lln. 



1807 

 + II71 



1808 

 + ll?i 



1809 

 + 11» 



1810 

 + lln 



Padua 



-1-3 

 -5-0 



-0-3 

 + 1-5 



-1-7 

 +0-2 



+ 1-1 

 -1-9 



+ 4-2 

 -2-5 



+ 4-2 



o-o 



-4-9 

 + 3*0 



+ 3 -4 

 + 4-7 



-2-8 

 -5'6 



-2-8 

 + 2-6 



+ 1-7 

 +3-3 



These show that the years of greatest rainfall at Padua are repre- 

 sented by the formula [1804 or 5 + llw], and of least by [1806 + llw], 

 whilst for Milan the maximum occurs at [1807 + 11%] , and the minimum 

 at [1808+11»]. 



Fig. 2. 



Curves of Rainfall Variation in Cycles of 5 to 13 Years. 



Further inspection of the tables will show numerous other instances 

 of incongruity to be found in every one of the cycles, leading forcibly 

 to the conclusion that either no short term of exactly five, six, seven, 

 eight, nine, ten, eleven, twelve, or thirteen years exists in the annual 

 amount of rainfall at any of the stations whose observations have 

 been here discussed, or that the effect of abnormal falls is so great that 

 it cannot be eliminated by upwards of a hundred years' observations. 



In any case, I think it may now be stated with certainty that all 

 predictions as to rainy or dry years, based upon existing materials, 

 must in future be considered as utterly valueless. 



My thanks are due to Mr. R. H. Scott, Director of the Meteoro- 

 logical Office, for several valuable suggestions, and to Padre F. Denza, 

 Director of the Moncalieri Observatory, for manuscript copies of the 

 Italian observations. 



