ftewing are those which have been collected > When, going from 

 sthe inferior extremity of a glacier of the second kind, we rise 

 gradually to those of the first kind, even to the highest peaks. 



ol ^hOften,"' savs Mr Hugi, *' blocks of ice detach themselves 

 ^ora the extremity, or even from more elevated parts of a gla- 

 cier of the second, and rest upon the soil. These blocks, ex- 



'^|)06ed to the rays of the sun, and at an elevated temperature, 

 do not melt, as is the case with ice in general ; but if they are 

 not of very large dimensions, they first break into many pieces. 



? I have often examined these detached blocks, particularly on 

 the glacier of Alctch, in the Lake of Morile, which has been al- 

 ready mentioned, lo 



^aMi^/This little lake being completely empty, the glacier which 

 formed one of its sides broke through all its thickness, and filled 

 the bottom of the lake with its debris ; some of the blocks had a 

 diameter of 40 feet, but the greater number were from 4 to 12 

 feet. An examination of these fragments is particularly neces- 

 sary to enable us to know the I lOonstitJAtion pf ikhe^ io^j; aB|d the 

 stratification of the glacier. • pffh to uh/'j^ odi nocfif ooiTn-Hji, 



-msf*i/The mass is formed of crystals imbedded in each other, in 

 such a manner as to be moveable not only in the broken blocks 

 of which we speak, but upon the borders of the glaciers them- 

 selves; above all, where protuberances and ridges are found. Not- 

 withstanding this mobility, these crystals do not separate from 

 each other ; a certain force is even necessary to detach one of them 

 from the mass, and they are seldom detached without breaking. 

 These crystals, of which the larger have a diameter of two 

 inches, and the smaller of one inch, are articulated into each 

 other in all positions and directions, and each of them aids to 

 enclose its neighbour in the mass. But if only one of them be 

 detached, it is easy to detach all the rest successively witli the 

 fingers, and thus destroy a whole block. A mass is often de- 

 composed of itself into a heap of crystals, when some of its 

 crystals are separated. It is scarcely possible to assign to these 

 crystals a determinate form. They are rather oblong than cu- 

 bical ; and they have very often from one side, and rarely from 

 two, a strong jutting out of articulation, with surfaces and 

 angles imperfec^y, .defined* i i,Th«r.,.s«r£ai»i.i« iw»Mgfe,A94tfur- 



