xv.] FORBES' GLACIER DISCOTERIES. 503 



'To a person accustomed to the rigour of reasonings 

 about mechanical problems, the very first data for a 

 solution were evidently wanting namely, the amount of 

 motion of a glacier in its different parts at different 

 times. A few measures had indeed been made from 

 time to time by MM. Hugi and Agassiz, of the advance 

 of a great block on the glacier of the Aar from one year 

 to another, but with such contradictory results as corre- 

 sponded to the rudeness of the methods employed ; for 

 in some years the motion appeared to be three times as 

 great as in others. I then pointed out to M. Agassiz, 

 how, by the use of fixed telescopes, the minutest motions 

 of the glacier might be determined, a suggestion which 

 :us, 1 believe, since put in practice. It seems very 

 ilar that ingenious men, with every facility for esta- 

 blisliing facts for themselves, should have relied on con- 

 clusions vaguely gathered from uncertain data, or the 

 hazard. -d assertions of the peasantry about matters in 

 which they take not the slightest interest. The supposed 

 immobility of the glaciers in winter, the supposed 

 tei velocity of the sides than the centre of the 

 ice, were amongst the assumptions traditionally handed 

 down, upon no sufficient authority; and I believe that I 

 may safely affirm that not one observation of the rate of 

 motion of a glacier, either on the average or at any par- 

 ticular season of the year, existed when I commenced 

 my experiments in 1842. Far from being ready to 

 admit, as my sanguine companions wished me to do in 

 1841, that the theory of glaciers was complete, and the 

 cause of their motion certain, after patiently hearing 

 all that they had to say, and reserving my opinion, I 

 drew the conclusion that no theory which I had then 

 d of could account for the few facts admitted on 

 all hands, and that the very structure and motions of 



i nained still to be deduced from observation.' 

 , (Travels, pp. 126129) : 



4 I theory !< inn-, tin- daricr moves 



;nl without sensibly incorporating new matter into 

 substance continually fed by tli< supplies {'mm 



