Observations of Underground Temperature. 133 



•06744, which, from his five years' data alone, we found (§ 16) 



for the corresponding constant with reference to the sand at the 



Experimental Garden and the sandstone of Craigleith Quarry. 



From them, calculating as above (§ 36), we find 260-5 and 690- 7 



k 

 as the values of- for the terrestrial substances of these localities 

 c 



respectively, — results of which the meaning is illustrated by the 



statements of §§ 36 and 37. 



39. To deduce the conductivities of the strata in terms of uni- 

 form thermal units, Professor Forbes had the " specific heats " 

 of the substances determined experimentally by M. Kegnault. 

 The results, multiplied by the specific gravities, gave for the 

 thermal capacities of portions of the three substances, in terms 

 of that of an equal bulk of water, the values -5283, -3006, and 

 •4623 respectively. Now these must be the values of c if the 

 thermal unit in which, k is measured is the thermal capacity of 



k 

 a French cubic foot of water. Multiplying the values of - found 



above by these values of c, we find for k the following values : — 



Trap-rock of Calton Sand of Experimental Sandstone of 

 Hill. Gardens. Craigleith. 



124-2 78-31 3193 



The values found by Professor Forbes were — ■ 



111-2 82-6 298-3 



Although many comparisons have been made between the con- 

 ducting powers of different substances, scarcely any data as to 

 thermal conductivity in absolute measure have been hitherto 

 published, except these of Professor Forbes, and probably none 

 approaching to their accuracy. The slightly different numbers 

 to which we have been led by the preceding investigation are no 

 doubt still more accurate. 



40. To reduce these results to any other scale of linear mea- 

 surement, we must clearly alter them in the inverse ratio of the 

 square of the absolute lengths chosen for the units*. The 



* Because the absolute amount of heat flowing through the plate across 

 equal areas will be inversely as the thickness of the plate, and the effect 

 of equal quantities of heat in raising the temperature of equal areas of the 

 water will be inversely as the depth of the water. The same thing may be 

 perhaps more easily seen by referring to the elementary definition of ther- 

 mal conductivity (footnote "to § 11 above). The absolute quantity of heat 

 conducted across unit area of a plate of unit thickness, with its two sides 

 maintained at temperatures differing by always the same amount, will be 

 directly as the areas, and inversely as the thickness, and therefore simply 

 as the absolute length chosen for unity. But the thermal unit in which 

 these quantities are measured, being the capacity of a unit bulk of water, is 



