DEPENDING ON THE HEIGHT ABOVE THE EARTH'S SURFACE. 453 



above the Earth's surface. The fall is, consequently, 1" of Fahrenheit 

 for an elevation of 186 feet. This, according to the theory, would be 

 the case if the temperature were a function of the density only ; and 

 for every additional height of 186 feet there would be an equal fall 

 of temperature. It appears, however, from the discussion of a large 

 number of thermometrical observations at different elevations, contained 

 in a Memoir by Mr Atkinson, in Vol. II. of the Memoirs of the Royal 

 Astronomical Society, that the decrement of temperature is 1° of Fah- 

 renheit at an elevation of 250 feet. Also, that the decrements are not 

 exactly equal for equal increments of height, but rather for increments 

 of height which increase in a slow arithmetical progression. The ex- 

 planation of the difference in these two respects between the facts of 

 experience and the above results of the theory, is to be sought for in 



d9 

 the term involving -j^. By neglecting this term we, in fact, supposed 



the temperature to be given when the density is given, and consequently 

 neglected the tendency which contiguous portions of air of different 

 densities have to assume the same temperature. When a limited por- 

 tion of air is suddenly rarefied, its temperature falls, in the first instance, 

 but in a short time it assumes the temperature of the surrounding 

 bodies, its density remaining the same, and the time required to pro- 

 duce this effect is greater as the portion of rarefied air is larger. 

 A similar cause must be in operation in the atmosphere, tending to 

 produce a nearer approximation to equality of temperature in the upper 

 and lower regions, than would exist if the temperature were a function 

 of the density merely : although the temperature of the different parts 

 can never be entirely equalized, on account of the radiation from the 

 cold parts above into the still colder spaces beyond the limits of the 

 atmosphere. As the effect of the above-mentioned cause is to increase 

 the temperature of the colder and diminish that of the warmer portions 



dQ 

 of the atmosphere, so far as this action is concerned, -j- will be a positive 



quantity. 



Again, in consequence of the unequal distribution of the Sun's 

 lieat, different columns of the air are differently heated. Motion con- 



3m2 



