1 48 Prof. BischofF o?i the Temperature of 



increases a certain number of degrees, is required, it may be 

 measured either in a direction perpendicular to the horizon or to 

 the surface of the earth at that place. In an horizontal plain 

 these two directions become identical, and therefoi-e it is un- 

 necessary to make any distinction. This is not the case in 

 mountains. A simple consideration shews, however, that the 

 direction in which the increase of temperature in mountains must 

 be measured is perpendicular to the slope of the hill, and not 

 perpendicular to the horizon. If we suppose, for example, that 

 a mountain forms a perpendicular precipice BC, Fig. 7, it is 

 clear that the increase of temperature, caused by heat in the in- 

 terior of the earth, cannot proceed in a vertical direction ; for BC 

 is a part of the earth's surface, and the surface is under the im- 

 mediate influence of the temperature of the air. From B to C, 

 therefore, there can be only such an increase of temperature as 

 takes place in the same direction in the atmosphere *. 



That which is so evident in a mountain having a perpendicu- 

 lar side must also take place in mountains the slope of which is 

 more or less inclined to the horizon. The increase of tempera- 

 ture in mountains can, therefore, only be measured in a direction 

 perpendicular to their surface. So that in a conical mountain 

 ABD, PI. II, Fig. 8, we can only measure the increase of tem- 

 perature from the exterior to any point F in the interior of the 

 mountain, in the direction EF or GF. But since the increase of 

 temperature towards the centre of the earth has hitherto always 

 been measured in a direction perpendicular to the horizon, it re- 

 mains to be determined in what ratio are the lines EF and BF. 



Let the mean temperature at B = 1% at F = t° + 1°, at 

 F,=it°-\-x°; and let the angle B A C = a; let the temperature 

 be supposed to increase ]°, in every n feet, in a direction per- 

 pendicular to the slope of the mountain ; and suppose the tem- 



• If the increase of temperature in mountains proceeded in a direction per- 

 pendicular to the horzon, and at the rate of 2°.25 in every 1 15 feet, we should 

 have an increase of temperature of 191°.25 from the summit to the foot of the 

 celebrated tremendous precipice of Canivalleda, which has an almost per- 

 pendicular height of 9800 feet ; so that at the foot there would be a tem- 

 perature of 248°, as Von Humboldt found 5C°.75 at the top at a time when 

 the actual temperature in the Guayra, on the sea coast, was only 81°;50. See 

 Gilbert's Annal., vol xxi\-. p. 25. 



