292 SIXTH REPORT — 1836. 



temperature of the interior of the earth, from the surface to the 

 greatest depths yet reached by human enterprise. 



(b.) The plan of experiment pi'oposed for general adoption, 

 and specially required to be followed by those who undertake 

 to use the instruments furnished by the Association, is intended 

 to reduce the effect of known and equalize those of unknown 

 sources of fallacy. 



For this end certain precautions must be observed, suited to 

 experiments in air, water, androck respectively; for none of these 

 are wholly free from the influence of sources of serious error. 



The temperature of the air in the gallery of a mine varies ac- 

 cording to the place of the observation as compared with the 

 entrance and exit of the current ; according to the rate of this 

 current as it passes through a confined, open, or complicated 

 passage ; according to the place of the thermometer in the sec- 

 tion of the air passage ; according to lights, respiration, and 

 other local conditions. 



On all these accounts the experiments in air are the least ac- 

 curate indications of subterranean temperature : however care- 

 fully made there is in the result always too much of the effect of 

 local influences which cannot be estimated. (They are however 

 extremely valuable in combination with those hereafter noticed.) 



The water of a mine offers a less exceptionable subject of ex- 

 periment. If it be a small continuous subterranean spring, 

 discovered at a known depth, without any sign of efflux under 

 violent pressure, its temperature carefully taken will be found 

 to be nearly constant. The composition and specific gravity of 

 the water may be of importance in the combination of the re- 

 sults. It should therefore be correctly stated. But water merely 

 lying in the galleries of a mine, or collected from the sides of 

 the shafts, is never to be referred to as a standard of subterra- 

 nean temperature. 



It is however in the solid rock that the best observations, and 

 those most suited to the purpose of philosophical reasoning, are 

 to be obtained. The principal sources of fallacy in this class of 

 experiments arise from the unequal and varying influences of 

 the air-currents, moisture, &c., on the surface of the rock ; local 

 chemical actions and electric currents may also be noticed as 

 affecting the precision of the result, and if known should be re- 

 corded. The only experimental caution, however, available in 

 this case, is to sink the thermometer to a sufficient depth from 

 the surface, in a hole very little larger than itself, and to record 

 the observations after moderate intervals of time. 



(c) The instruments furnished have been compared with one 

 known standard, and all that is required of the observer is to re- 



