Let us now proceed across the valley of Chamouni to the Montanvert; where, climbing the adjacent heights to an elevation of six or eight hundred feet above the hotel, we command a view of the Mer de Glace, from Trelaporte almost to the commencement of the Glacier des Bois. It was from this position that Professor Forbes first observed the bands. Fifteen, sixteen, and seventeen years later I observed them from the same position. The number of bands which Professor Forbes counted from this position was eighteen, with which my observations agree. The entire series of bands which I observed, with the exception of one or two, must have been the _successors_ of those observed by Professor Forbes; and my finding the same number after an interval of so many years proves that the bands must be due to some regularly recurrent cause. Fig. 36 represents the bands as seen from the heights adjacent to the Montanvert. [Illustration: DIRT-BANDS OF THE MER DE GLACE, AS SEEN FROM LES CHARMOZ. Fig. 36. _To face p. 368._] I would here direct attention to an analogy between a glacier and a river, which may be observed from the heights above the Montanvert, but to which no reference, as far as I know, has hitherto been made. When a river meets the buttress of a bridge, the water rises against it, and, on sweeping round it, forms an elevated ridge, between which and the pier a depression occurs which varies in depth with the force of the current. This effect is shown by the Mer de Glace on an exaggerated scale. Sweeping round Trelaporte, the ice pushes itself beyond the promontory in an elevated ridge, from which it drops by a gradual slope to the adjacent wall of the valley, thus forming a depression typified by that already alluded to. A similar effect is observed at the opposite side of the glacier on turning round the Echelets; and both combine to form a kind of skew surface. A careful inspection of the frontispiece will detect this peculiarity in the shape of the glacier. [Sidenote: FROM THE CLEFT-STATION.] From neither of the stations referred to do we obtain any clue to the origin of the dirt-bands. A stiff but pleasant climb will place us in that singular cleft in the cliffy mountain-ridge which is seen to the right of the frontispiece; and from it we easily attain the high platform of rock immediately to the left of it. We stand here high above the promontory of Trelaporte, and occupy the finest station from which the Mer de Glace and its tributaries can be viewed. From this station we trace the dirt-bands over most of the ice that we have already scanned, and have the further advantage of being able to follow them to their very source.