Charpentier's Essay on Glacien. 109 



that the underside of glaciers, on the contrary, is always 

 frozen, even during the heats of summer. 



After some brief observations on the ice of northern re- 

 gions, which he divides into floating and terrestrial ice, the 

 former proceeding from true glaciers produced by the conge- 

 lation of the neve, and the others from ordinary congealed 

 water, M. de Charpentier arrives at the second part of his 

 work, which treats of the erratic formation. 



Although it may perhaps be difficult to give a good defini- 

 tion of this formation, without confounding it with the dilu- 

 vial or alluvial soils, it is so well known to geologists under 

 the name of erratic blocks, that there is no need of charac- 

 terising it by a description. The size of the blocks, which 

 in no degree diminish in proportion as we recede from the 

 places where they originate, as takes place in regard to the 

 diluvial formation, the rough surface and sharp angles of 

 many of these blocks, the almost invariable want of selection 

 according to size and stratification, the great height to which 

 these accumulations rise, — such are the prominent features of 

 the erratic formation. The rocks of which these blocks are 

 composed in Western Switzerland, are almost all found in situ 

 in the valley of the Rhone and its lateral valleys. The hard- 

 est rocks furnish the lai'gest blocks ; M. de Charpentier men- 

 tions many blocks of granite from 40,000 to 100,000 cubic 

 feet, and, as an exception to the general rule, a block of lime- 

 stone, the largest known, situate near Bex, which contains 

 161,000 cubic feet. These blocks must have travelled a great 

 distance, some of them 25 and even 60 leagues, from the place 

 of their origin. 



According to M. de Charpentier, the deposits of erratic 

 blocks appear under three diiferent forms ; sometimes scat- 

 teredy in blocks dispersed here and there, or in insulated 

 blocks, and more or less covered with earth or diluvium. 

 This is the most common form. Sometimes they are accu- 

 mulatedy or collected in mounds or small hills, presenting the 

 same modifications and the same structure as moraines. 

 Finally, at other times they are stratified in short thick beds, 

 and readily distinguished from the diluvial formation by the 

 great number of angular blocks they contain. 



One curious fact is the assemblage of blocks of tlie same 



