292 Pr Martens on the Glaciers of Spitsbergen. 



Biselx,* discussed and admitted by Gilbert, + and lastly re- 

 produced, in very recent times, by M. Agassiz. J The weight 

 of the ice, the dilatation of the water freezing in its crevices, 

 and their enlargement as a necessary consequence, appear to 

 me to be the three causes which operate most powerfully in 

 producing the progression of the Spitzbergen glaciers. In fact, 

 the sinking of which Saussure speaks could not be regarded 

 as one of the means of accounting for it, for these glaciers 

 never melt on that surface which is in contact with the soil. 

 The following are my proofs : — At Magdalena Bay there 

 were, as I have already stated, two small glaciers which ter- 

 minated on the top of a talus and did not descend to the sea. 

 I have examined them closely ; not the smallest streamlet of 

 water escapes from their base. With regard to those which 

 advance into the sea, it is impossible to assure one's self directly 

 whether they give rise to streamlets at their lower part, since 

 the latter would be lost in the sea below the ice. The rivu- 

 lets which run laterally between the glacier and the moun- 

 tain owe their existence to the melting of the snows. The 

 experiments which we made on the temperature of the earth, 

 appear to me equally to demonstrate that the glaciers of 

 Spitzbergen do not melt under the influence of terrestrial 

 heat. At Bell Sound, the earth was frozen very hard at least 

 to the depth of a metre. We were obliged to put warm wa- 

 ter into the hole in order to continue the bore. On the 3d 

 and 4th of August 1838, a thermometer sunk !n .43, and 

 observed for thirty-three hours, afforded a mean of -f 2°. 86 C. ; 

 the maximum was 4- 8°.4 ; the minimum + 0°.2. At Mag- 

 dalena Bay another thermometer sunk m .35, and, observed 

 from hour to hour for ten days, gave a mean of -f- 1°.55 ; the 

 maximum being + 2°.40 ; the minimum + 0°.40. The diur- 

 nal variations followed those of the atmospheric temperature. 

 Thus, in Spitzbergen, the ground is frozen in the middle of 



* Ibid. t. 64, p. 183, 1820 ; and Biblioth. Univ. de Geneve, t. xi. and xii. 

 (one. sir.) 



t Annalen der Physik, t. 64. p. 183. Sec the notes. 



% Bulletin de la Socit'te' de Geologie do Fiance, t. ix. p. 443 ; and 

 Biblioth. Univ. de Geneve, May 1840, vol. xxvii. p. 134-5 ; and Jameson *a 

 Journal, vol. xxvii. p. 383. 



