Dr Martens on the Glaciers of Spitsbergen. 293 



summer to the depth of a metre, and the surface is warmed 

 only by the direct rays of the sun. It is evident that this in- 

 fluence cannot act on ground covered with glaciers. The 

 atmospheric heat cannot penetrate across masses of ice of such 

 a thickness, and the ground on which they rest is necessarily 

 frozen. It was our intention to assure ourselves directly of 

 this; but circumstances, independent of our wishes, prevented.* 

 In his excellent work on the temperature of the earth,i 

 Professor Bischof has fully demonstrated that an ordinary 

 glacier cannot melt under the influence of central heat when 

 the mean temperature of the earth which it covers is at zero. 

 Now, he shews that,} among the Alps, it is at 2002 metres 

 above the sea that this medium is found. This point being 

 still 468 metres below the line Avhich separates the lower and 



* M. Elie de Beaumont has inserted in the Journal of the Institut, of 

 15th June 183G, a physico-mathematical note on the relation qui cxiste entre 

 I'eijaisseur que les glaces perpitueUes peuvent aequirir dans un lieu donni, et I'ac- 

 croissement de temperature qu'on observe dans les lieux profonds. If we apply to 

 the glaciers of Spitzbergen the consequences which result from these for- 

 mula?, we will find that the face in contact with the ground ought to be in 

 a, state of melting, and that consequently the increase of the entire mass iu 

 thickness will be necessarily limited, if they acquire such a power that their 

 base is comprised in a zone of terrestrial temperature above zero. In fact, 

 the mean temperature of the air in Spitzbergen is very probably 7° C. ; but 

 temperatures above zero not being able to traverse cither the ice or the 

 snow which covers the ground during the whole year, we will take no ac- 

 count of them, in this general mean, from the monthly means of July and 

 August which are above zero. The moan transmissible by the snow and ice 

 is thus found lowered to nearly 9 metres C. If we admit, Avith M. E. de 

 Beaumont, that the increase of temperature is 1° for 30 metres, in propor- 

 tion as we descend vertically into the earth ; if we suppose, moreover, that 

 the conducting power of the ice for heat is equal to that of the earth, it 

 must follow that a glacier must be 270 metres in thickness before it attain 

 the isogeofherme of zero. If it pass this limit, it melts. Among the glaciers 

 hitherto observed in Spitzbergen, the most dense, that of Horn Sound, is 121 

 metres of vertical height at the margin of the sea. It docs not, therefore, 

 reach the melting point, since its base is on a level witli the isogeothcrmc of 

 — 5° C. But' if it were possible to visit the interior of the island, wo would 

 probably there find masses of ice whose height would approach to 250 metres. 



t Die W'aermclehrc des Innern unsera Erdkoerpers, 1837; and Professor 

 Bischof in Jameson's Philosophical Journal, vol. xxiii. p. 310'. 



J Ibid. chap, xviii. 



