Ancient Extent of the Glaciers of Chamonix. 61 



characterise it. Mr J. D. Forbes studied them on the Mer 

 de Glace of Chamonix ; but it is on the glaciers of the Aar 

 that the observations have been prosecuted with most care 

 and perseverance. From 1842, MM. Agassiz and Desor, as- 

 sisted by MM. Wyld, Otz, and Dollfus-Ausset, have been un- 

 ceasingly occupied with this question ; they have ascertained 

 that, in its medium part, this glacier advances 71 metres a 

 year. Toward the lower extremity, the rapidity of the pro- 

 gression decreases till it does not exceed 39 metres : it accele- 

 rates a little, on the contrary, towards the top, where the 

 glacier annually traverses a space of 75 metres.* 



The inclination of the slope on which the glacier descends 

 does not appear to have any influence on the rapidity of its 

 progress, but it is singularly modified by the walls of the hol- 

 low in which it moves. The friction of the ice against these 

 walls considerably retards the advance of the lateral parts of 

 the glacier. Besides this, if a promontory jut out towards the 

 middle of the valley, the glacier, arrested on one of its sides, 

 turns round the obstacle with extreme slowness, or rather this 

 side remains behind, while the central portion and the oppo- 

 site edge continue to advance with their relative quickness. 



* The following is a short account of the method in which the advance of a 

 glacier was measured. On the two banks, two rocks were chosen opposite each 

 other ; each of these I'ocks was marked by a white cross painted upon the stone. 

 A series of stakes were then fixed in the ice in a line between those two points, 

 80 as to form a straight line perpendicular to the axis of the glacier. After 

 some days, an observer placed himself before one of the crosses, and directed 

 a telescope, bearing a level, towards the one opposite. The glacier having 

 advanced, and the stakes along with it, the latter were no longer in their 

 original line. Then a guide placed on the glacier and carrying a pole sur- 

 mounted by a very visible object, placed the pole in the direction of the former 

 line. This direction was indicated to him by the signal of the observer, whose 

 eye was applied to the tfelescopo. The latter caused the pole to be carried back 

 till it was exactly at the point formerly occupied by the line of stakes. This 

 done, the guide measured the distance on the ice from the foot of the pole to 

 that of the stakes. This interval was exactly the length traversed by the glacier 

 between the observations. This year, the process has been modified by MM. 

 Uollfus, Otz, and myself, in such a manner as to permit us to follow the daily 

 progress of the glacier of the Aar with such an exactness, that the error of ob- 

 servation cannot exceed two millimetres, or about a line. 



