TEANSACTIONS OF SECTION C. 345 



rock, •with little or no warning-, disintegrates and discharges itself in efflorescence 

 in the air, producing an effect as destructive to the city built there as in the 

 former case, with quite as picturesque an effect, though from an exactly opposite 

 cause. So much was being done now in ascertaining the component parts of stone 

 for the purpose of hardening, as in the recent experiments on the Houses of 

 Parliament, Cleopatra's Needle, and other well-known works, that it occurred to 

 him that an analysis of these two rocks of similar component parts, but with 

 varying conditions, would be well worth the attention of the chemist and the 

 practical constructor. 



The paper was illustrated with views of Hierapolis, &c. 



4. Report on the Miocene Plants of the North of Ireland. 

 See Reports, p. 162. 



MONDAY, AUGUST 25, 1879. 



The following Report and Papers were read : 



1. Sixth Report on the Conductivities of certain Rocks. 

 See Reports, p. 58. 



2. On some Broad Features of Underground Temperature. 

 By Professor J. D. Everett, F.R.S. 



r ' The temperature at the surface of the ground is not sensibly influenced by the 

 flow of heat upwards from below, but is determined by astronomical and atmo- 

 spheric conditions. 



The rate of increase in travelling downwards from the surface may conveniently 

 be called the temperature-gradient, and averages about one degree Fahrenheit for 

 fifty or sixty feet. This is about five times as steep as the temperature-gradient in 

 the air. 



If we draw isothermal surfaces for mean annual temperature in the ground, 

 their form beneath mountains and valleys will be flatter than that of the surface 

 above them. This is true even of the uppermost, and the flattening increases as 

 we pass to lower ones, until at a considerable depth they become sensibly horizontal 

 planes. 



The temperature-gradient is consequently steepest beneath gorges, and least 

 steep beneath ridges. 



In a place where the surface of the ground and the isothermal surfaces beneath 

 it are horizontal, the flow of heat will be vertical, and the same quantity of heat 

 will flow across all sections which lie in the same vertical. In this case the flow 

 ^across a horizontal area of unit size will be equal to the product of the temperature- 

 gradient by the conductivity, if we employ the latter term in an extended sense,_ so 

 as to make it include convection by the percolation of water, as well as conduction 

 proper. It follows that, in comparing different strata lying in the same vertical, 

 the gradient will vary in the inverse ratio of the conductivity. It seems probable 

 that the same law of inverse proportion between gradient and conductivity holds 

 approximately even when the strata compared are not in the same vertical, but are 

 widely distant. 



As regards the modes of observation which have been employed for the deter- 

 mination of gradients, shafts full of water, and wells of large diameter, afford so 



