1845.] MODIFICATIONS OF DE SAUSSURE's THEORY. 101 



those who have attempted to illustrate De Saussure's theory by 

 experiment, but who, like him, neglected the form and undula- 

 tions of the bed in which it rests. 



4 MODIFICATIONS OF DE SAUSSURE'S THEORY. 



De Saussure and his immediate followers appear to have 

 considered the crevasses which occur transversely in most glaciers 

 as the result of the inequalities of the beds down which they 

 are constrained to move ; but other writers have imagined that 

 the part which these crevasses perform in the phenomena of 

 glacier motion is fundamental, and essential to the existence of 

 the movement at all. Some writers have remarked that the 

 fall of ice blocks over the precipice wnich often occurs near 

 the lower end of glaciers, leaving the superior portions unsup- 

 ported, allows them to advance to fill the position formerly 

 occupied by the portion of the now fallen ice. But in this 

 case it would appear that cause and effect are in some degree 

 confounded. The ice about to be projected over the cliff must 

 either advance towards its fall by its own gravity, or by the 

 pressure of the parts behind. If its own gravity suffices, the 

 same cause will urge the ice behind it to move similarly, whether 

 the block in question fall or not ; and if it be the pressure from 

 behind which shoves it on, then still more is the pressure of 

 the entire glacier the cause of motion of the entire glacier, 

 irrespective of the precipitation of its more advanced part. 



Thus, M. Martins' theory of the progression of glaciers is, 

 that the weight of the parts causes them to separate by fissures 

 into wedge-shaped masses, without their sliding along the bottom ; 

 that the fissures become filled with frozen snow, and that thus 

 the glacier is perpetuated and extended year by year. " Cette 

 progression," he says, " n'est done ni un glissement ni un 

 affaissement difficiles a comprendre, puisque la glace doit adherer 

 au sol, mais un demembrement successif."* Besides other 

 objections, it is now universally admitted that the glacier-proper 

 does not grow by the consolidation of snow in its fissures. 



* Martins sur les Glaciers de Spitzberg et de la Suisse. Bibl. Univ. Juillet 1840. 



