16 REPORT — 1871. 



at the highest part of this vertical, that is close under the crest, than at a 

 lower level in the same vertical. It would be absurd to pretend to fix the 

 amount of the correction with accuracy ; but it seems not unreasonable to 

 estimate that, in the present case, the numer of isothermals cut through by a 

 vertical line descending from the crest of the ridge to the tunnel itself is 

 about seven-eighths of the number which wovdd be cut through in sinking 

 through an equal distance in level ground, other circumstances being the same. 

 Instead of 1° in 93 feet, we should thus have 1° in ^ of 93, that is, in 81 feet. 



This is a slow rate of increase, and is about the same as Mr. Fairbaim 

 found at Dukenfield. The rocks penetrated by the tunnel consist of highly 

 metamorphosed material, and are described as belonging to the Jurassic 

 series. No fossils have been found in them. For two-thirds of the length 

 of the tunnel, beginning from the Italian end, they are remarkably uniform, 

 and it is in this part that the observations have been taken. The following 

 account of them has been given by Prof. Ansted (Pop. Sci. Review, Oct. 

 1870, p. 351) : — " The rocks on which the observations have been made are 

 absolutely the same, geologically and otherwise, from the eatrance to the 

 tunnel, on the Italian side, for a distance of nearly 10,000 j'^ards. They 

 are not faidted to any extent, though highly inclined, contorted, and sub- 

 jected to slight slips and slides. They contain little water and no mineral 

 veins. They consist, to a very large extent indeed, of silica, either as 

 quartz or in the form of silicates, chiefly of alumina, and the small quantity 

 of lime they contain is a crystalline carbonate." 



This uniformity of material is very favourable to conduction, and the high 

 inclination of the strata (in which respect these rocks resemble those at 

 Dukenfield) also appears to promote either conduction proper or aqueous con- 

 vection, which resembles conduction in its effects. As regards Mons. Gior- 

 dano's estimate of the mean air- temperature at the crest, it is obtained in 

 the following way : — The hill of San Theodule is 430 metres higher, and 

 the city of Turin is 2650 metres lower than the crest ; the temperature of 

 the former has been determined by one year's observations to be — 5°'l C, 

 and that of the latter is 12°-5 C. If a decrease of 1° C. for every 174 metres 

 of elevation be assumed (1° F. for 317 feet), we obtain, either bj' com- 

 parison with San Theodule or with Turin, the same determination — 2°-6 

 for the air-temperature at the crest of the ridge over the tunnel. 



This mode of estimating the temperature appears very fair, though of 

 course subject to much uncertainty ; and there is another element of uncer- 

 tainty in the diff'erence which may exist between the air-temperature and 

 the rock-temperature at the summit. 



These two elements of uncertainty would be eliminated if a boring of 

 from 50 to 100 feet were sunk at the summit, and observations of tempera- 

 ture taken in it. The uncertain correction for convexity would still remain 

 to be applied. It would therefore be desirable also to sink a boring, of about 

 the same depth, in the plateau which extends for about a quarter of the 

 length of the tunnel, beginning near the Italian end, its height above the 

 tunnel being about a third of a mile. 



In November last, when very little information had reached this country 

 respecting the temperature-observations in the tunnel, an urgent appeal was 

 addressed, jointly by your Committee and by the Geographical Society (of 

 which Prof. Ansted is Foreign Secretary), to M. Sismonda, requesting him 

 to use his influence with the Italian authorities to -secure a series of accu- 

 rate observations of the temperature in the sides of the tunnel, before time had 

 been allowed for this temperature to undergo sensible change from its original 



