Hot mid Thermal Springs. S95 



continually rise, and thus impart to the surface a temperature 

 higher than it possesses over shallows. The phenomenon is, 

 therefoi-e, in no way dependent on the latitude, but only on the 

 depth of the sea. 



The higher temperature of the surface of the sea in deep 

 places seems also to prove, that only a part of the warmed water 

 which rises from the depths is used in the evaporation of the 

 water at the surface. Now, as the air over the sea partakes of 

 the heat arising from this source, but to a very limited extent, 

 we must conclude, as in the case of lakes, that the mean tem- 

 perature of the surface of the sea is higher than ihat of the sur- 

 rounding atmosphere. And this is shewn to be the case by the 

 observations made four times a-day (at 6 a. m., and 6 p. m., at 

 mid- day, and at midnight) by Peron. He found the mean 

 temperature of the water of the sea at its surface, at a distance 

 from land, between latitudes 49° N. and 45° S., always higher 

 than that of the air immediately in contact with it. Accord- 

 ing to Humboldt's observations, the ocean is also on an average 

 rather warmer than the air with which it is in contact. The 

 maximum temperature of the seas amounts to between 82°.4 and 

 84°. 1, and the mean temperature of the atmosphere near the 

 equator is only 78o.9 to 80°.5.* 



The same was observed by J. Davy,-|- during his voyage to 

 Ceylon, between latitudes 48° N. and 35° S. He made obser- 

 vations on the temperature of the sea and of the air with the 

 help of two assistants, from the iSth Februa.y to the 12th 

 August 1816, every two hours, even during the night, and 

 from them he calculated the mean for each day. Taking the 

 mean of all these observations, which comprises 150 days, we 

 find, that the sea was 1°.795 warmer than the air. Of these 

 150 days there were only 17° on which the atmosphere gave a 

 mean temperature superior to that of the sea, on five days they 

 were equal, and on 128 days the mean temperature of the air 

 was lower than that of the sea. It may, indeed, with great 

 probability be assumed, that the observations on those seventeen 



* Reisc, i. 353. Sec especially the data in the table, p. 352. 

 t Philosophical Transactions, 1817, ii. p. 27''>. 



