751 



nearer than at the Havannah, in the system 

 of American climates. 



The diminution in the extent of the horary 

 variations, in advancing from the equator * to- 

 wards the pole, was remarked by M. Ramond-f- 

 as soon as he began to compare the results of 

 his observations at Clermon-Ferrand with 

 those which I had collected in the torrid zone. 

 xc The extent of the variations," says this 

 able naturalist, " is half less in France than 

 between the tropics. The maximum of the 

 variations in our climates is in spring: the 

 ascent of the day in Europe is nearly equal to 

 the preceding lowering, while in the tropics 



* At Senegal (lat. 15° 53') a well-informed traveller M. 

 de Beaufort, found recently, by means of observations that 

 comprehend two months and a half, the extent of the 

 horary oscillations to be 2.7mm. He gives for 7 h in the 

 morning 0.7629 m (th. 21o) ; for noon 0.7654 (th. 25°); for 4 h 

 in the afternoon, 0.7663 m (th. 23°) ; and for 8 h in the even- 

 ing, 0.7667 m (th. 19°). Reduced to the temperature of zero^ 

 the observations of noon, and at 4 h in the afternoon give, 

 0.7619 m , and 0.7631 and not, as is asserted in a letter ad- 

 dressed to M. Jomard (January 25th 1824), 0.7631 m , and 

 0.7658 m . {Bulletin de la Soc. de Geographie, p. 14, 59) . 

 Hertha, 1825, n. 3, p. 143. These observations little ac- 

 cord with what has been found in every other part of the 

 earth, where the barometer has every where been seen lower 

 at 4 h in the afternoon, than at 8 h in the morning, and at noon, 



t Mem. de Vlnstitut, 1808, p. 107, and 1812, p. 46. 



