But if these are the mean temperatures of the surface, the natural temperatures of the caves themselves should be still higher, on account of the allowance to be made for increase of temperature with descent into the interior of the earth.  This element will very materially affect our calculations in such a case as the lower part of the ice in the Glaciere of the Pre de S. Livres, and the strange suggestive beginning of a new ice-cave 190 feet below the surface, on the Montagne de l’Eau, near Annecy.  In any open pit or cave, the ordinary atmospheric influences find such easy access, that the temperature cannot be expected to follow the law observed when perforations of small bore are made in the earth, as in the case of the preliminary boring before commencing to dig a well;[216] but the two glacieres mentioned above are so completely protected in their lowest parts, that they may be treated as if they were isolated from external influence of all ordinary kinds; and it may fairly be said that the mean temperature there ought to be considerably higher than at the surface.

It is not very likely that the results of the above calculations are strictly in accordance with what a careful series of observations on the spot might show.  The distance between Geneva and the Glacieres of S. Georges and S. Livres is sufficiently small to make it probable that the reality is not very far different from the calculated temperature; but the other two caves are comparatively so far off, that the temperature and elevation of Geneva are not very safe data to build upon.

FOOTNOTES: 

[Footnote 211:  Bischof, Physical Researches, 189.]

[Footnote 212:  Philosophical Magazine, v. 446 (1834).]

[Footnote 213:  Annules de Chimie et de Physique, liii. 2-10.  See also Bischof, 136.]

[Footnote 214:  The English edition of Bischof affords here a proof of the danger of frequent changes from one scale to another.  Bischof in the first instance rendered Boussingault into degrees Reaumur, and this was in turn reduced to degrees Fahrenheit; the result being that the authorised English edition of his book gives 2 deg..25 F. for 127.5 feet, which does not come within 10 feet of Boussingault’s statement.]

[Footnote 215:  M. Thury calculates a decrease of 1 deg.  C. for every 174 metres between Geneva and S. Bernard, which is less than the decrease given in the text.  He arrives at this conclusion by correcting the mean temperature of Geneva from 8 deg..9 C., the observed mean of eighteen years, to 9 deg..9 C., in consequence of supposed local causes, which unduly depress the temperature of Geneva.  With the mean 8 deg..9 C. a result nearly in accordance with that of the text is obtained.]

[Footnote 216:  Professor Phillips found, in the course of his investigations in the Monk Wearmouth mines, some hundreds of yards below the sea, that when a new face of rock was exposed, its temperature was considerably higher than that of the gallery or shaft in which it lay.  In some cases the difference amounted to 9 and 10 degrees.  The rock soon cooled down to an agreement with the surrounding temperature.]