162 Prof. J. Prestwich. Underground Temperatures, [Feb. 12, 



recent geological times, and the rate of increase of temperature in 

 descending beneath the surface, contend for a crust of minimum 

 thickness as alone compatible with these phenomena. 



The question of underground temperature, which is a subject 

 equally affecting the argument on both sides, had engaged the 

 author's attention in connexion with an inquiry respecting volcanic 

 action, and he was induced to tabulate the results to see how far the 

 usually received rates of increase were affected by various interfering 

 causes — not that most of them had not received due attention, but it 

 was a question whether sufficient allowance had been made for them. 



Although Gensanne's first experiments were made in 1740, and 

 others were subsequently made by Daubuisson, Saussure, and Cordier, 

 in coal and other mines, it was not until the construction of deep 

 artesian wells commenced in the second quarter of this century, and 

 Walferdin introduced his overflow thermometer, aud precautions 

 were taken against pressure, that the more reliable observations were 

 made and admirably discussed by Arago. The Coal Commission of 

 1866 collected a mass of important evidence bearing on the question, 

 and in 1867 a Committee of the British Association was appointed 

 to collect further information. Under the able superintendence of 

 Professor Everett, a series of valuable experiments with improved 

 instruments has been made, and full particulars published in the 

 Annual Reports of 1868—1883. 



But notwithstanding the precautions taken, and the accuracy of the 

 experiments, they present very wide differences in the thermometric 

 gradient, ranging from under 30 to above 120 feet per degree Fahr. 

 Consequently different writers have adopted different mean values. 

 On the Continent one of 30 metres per degree C. has been commonly 

 adopted, while in this country some writers have taken a mean of 

 50 feet per degree, and others of 60 feet or more. The object which 

 the author has in view is to see whether it is not possible to eliminate 

 ihe more doubtful instances, and to bring the probable true normal 

 gradient within narrower limits. In so doing he confines himself 

 solely to the geological side of the inquiry. 



In a general list, Table I, he gives all the recorded observations in 

 the order of date. The list embraces observations at 530 stations in 

 24-8 localities. The most reliable of these he classifies under three 

 heads, in Tables II, III, and IV. 



1. Coal mines. 



2. Mines other than coal. 



3. Artesian wells and bore-holes. 



To which tunnels are added in a supplement. 



The author then proceeds to point out that the gradients given in 

 many of the earlier observations were wrong in consequence of 

 neglecting the height of the surface, and from the exact mean annual 



