112 M. Flaugergues on the Quantity of Rain, and [AuC« 



Days. 



C January 345 



Winter 941 < February 281 



I March." 315 



C April 363 



Spring 1001 < May 335 



(.June 303 



1942 



("July 229 



Summer 726< August 198 



L September 299 



{October 419 

 November 433 

 December 401 



1979 



We see by these tables that the number of rainy days in the 

 year at Viviers increases sensibly. That the order of the months 

 relative to the number of rainy days is this : November, Octo- 

 ber, December, April, January, May, March, June, September, 

 February, July, August ; that the sum of the rainy days of 

 winter and spring is nearly equal to that of the rainy days of 

 summer and autumn ; and that the number of rainy days during 

 autumn and winter is to the number of rainy days during sum- 

 mer and spring as nine to seven. 



I have traced from the numbers in the preceding table (follow- 

 ing the same method as for the curve of the mean monthly 

 quantity of rain), a curve, which may be called the curve of the 

 frequency of rain, the ordinates of which are proportional to the 

 number of rainy days in each month. This curve, F G H I K 

 (fig. F), has two minima and two maxima. The first minimum 

 corresponds to the month of February, the first maximum to the 

 month of April ; the second minimum to the month of August, 

 and the second maximum to the month of December. 



I have traced likewise the same curve for all the days of the 

 year, taking ordinates proportional to the number of times that 

 it has rained on the same day of the year during the 40 years' 

 observations. This curve presents singular irregularities: very 

 dry days placed in the middle of a group of wet days ; and veiy 

 wet days in the midst of a series of dry days, &c. 



During the course of these 40 years, the days of the year on 

 which it has rained the oftenest are Oct. 31, on which it rained 

 23 times, and Nov. 4, on which it rained 21 times ; and the days 

 on which it rained the least are July 7 and 13, on which it 

 rained only twice. 



If we divide the total quantity of rain water which has fallen 



