Variation of the Barometer at Puris. 75 



It now remains for us to determine the value of the period 

 from 9 h p. m. to 3 h a. m., and that of 3 h a. m. to 9 h p. m. ; but 

 the observations are not very numerous. 



The observations fitted for this purpose are 495 in num- 

 ber, from 1815 to 1826 inclusive. The following are the re- 

 sults : — 



The period from 4 h a. m. to 9 h a. m. is thus rendered evi- 

 dent ; but that from 9 h p. m. to 4 h a. m. is not established by 

 the observations. M. Bouvard considers it probable that this 

 uncertainty may arise from the maximum in the evening and 

 the minimum in the morning not taking place at 9 h , and from 

 the hours at which the observations were made. 



Of the four daily barometrical periods the best established 

 is therefore that of from 9 h a. m to 3 h p. m. Its value, we 

 know, is not the same for all climates, and it diminishes as the 

 latitude increases. 



M. Bouvard then gives the table of the daily variation for 

 different latitudes from Humboldt, which we have already pub- 

 lished in No. viii. p. 300. He substitutes in it his own re- 

 sult for Paris, viz. m .76, in place of 0.70, as given by Arago, 

 and the result of M. Gambard for Marseilles. 



In order to give our readers a complete view of the results, 

 we have drawn up the following table, including the results ob- 

 tained by our correspondent A for Rome, and the results given 

 by Mr Daniel in his Meteorological Essays. 



