71 



A great mass of water follows with extreme 

 slowness the changes of temperature observed in 

 the atmosphere, and the maximum of the mean 

 temperatures of each month does not occur at 

 the same time in the ocean and in the air. The 

 increment of the heat of the seas necessarily 

 undergoes a delay ; and as the temperature of 

 the air begins to diminish, before that of the 

 water has reached it's maximum, it follows, that 

 the extent of the thermometrical variations is 

 smaller at the surface of the sea than in the at- 

 mosphere. We are still very far from knowing 

 the laws of these phenomena, which have a great 

 influence in the economy of nature. 



Mr. Kirwan admits, that between the eigh- 

 teenth degree of north and the eighteenth of 

 south latitude, the mean temperatures of the 

 months differ only five centesimal degrees, and 

 this estimation is somewhat too low ; for we 

 know by observations, carefully calculated, that 

 at Pondicherry, at Manilla, and in several other 

 places between the tropics, the mean heats of 

 the months of January and August differ eight 

 or ten degrees. Now the variations of the air 

 are at least a third less in the basin of the seas 

 than on the continent ; and the ocean under- 

 goes a part only of the changes of tempera- 

 ture of the atmosphere that surrounds it. 

 Hence it results, that, if the equinoctial oceans 

 did not communicate with the seas of the tern- 



