Ancient Moraines in Argyleshire. 201 



resistance in front will diminish, and the glacier may puslAhe 

 upper part of the moraine before it, while it glides over the 

 lower ; or it may override the whole, and the ultimate effect 

 will then be to raise the glacier to a higher level. It is per- 

 haps in this way that the insignificant size of the terminal 

 moraines of some large glaciers (such as that of the Unter Aar) 

 is to be accounted for. They may be, as it were, engulfed, in 

 consequence of the glacier riding over and covering them. 



It is not difficult to find a probable cause for the glacier of 

 Glenmessan stopping at AB, and piling up its terminal mo- 

 raines there. The lateral valley of Corusk w, though short is 

 steep in the sides, and would have its separate glacier, whose 

 course would be in the direction fc, h, at right angles to the 

 motion of Glenmessan glacier. When the latter, therefore, ar- 

 rived at the line g, A, it probably encountered a rampart of ice 

 flanked by piles of gravel and clay, issuing from the gorge at k, 

 and stretching right across its path. Its southward march would 

 thus be stopped, and stopped perhaps so long, that after the 

 rampart of ice had disappeared, it was not able either to propel 

 its own massive moraines A, B, or to override them. If the 

 work of M. Collomb, Preuves de V existence oVAnciens Gla- 

 ciers dans les Vallees des Vosges, is consulted, it will be seen 

 from his two maps (pp. 12 and 180), that the ancient moraines 

 are generally found at the junction of lateral valleys with 

 the principal one. Sometimes the glacier in the lateral valley 

 had arrested the glacier in the principal, and sometimes the 

 latter had arrested the former. In another point the ancient 

 moraines in the Vosges illustrate those we have been describ- 

 ing; they are generally " multiple," consisting not of one 

 mound or ridge, but of several, ranged, as he expresses it, par 

 echelon, or one behind another. Occasionally there are two, 

 like A, B — often there are three, and of various forms and mag- 

 nitudes. At Kirchberg, for instance, there is a double one 

 slightly curved, 400 metres long, and one of the mounds is 

 10 metres ( = 32J- feet) high. At Hussern there is one 3 5 

 metres (49 feet) high, at Sondernach one 50 metres (164 feet) 

 high ; at Wessenberg there is a triple moraine, the highest 

 point of which is 35 metres (115 feet) above the river, which 

 has cut a breach through the middle of the three ridges. 



