reduce the fragments to less size, and even to fine powder. Looking at the general conditions of glacier movements, the kind of ground these masses of ice pass over, and to the introduction of fragments from the sides, and even through the crevasses to the bottom, we

should expect that the grooving and scratching would be considerable on the bottom and sides, mingled with an extensive smoothing of surface, as if, in the application of a huge polishing apparatus, acting, as a whole, with minor deviations, in one direction, harder grains were strewed about, so that scratching as well as polishing was effected.* This scratching and smoothing by glaciers has been chiefly observed, with reference to their geological value, in modern times, though the rounded and polished surfaces frequently seen have been long known by the name of Roches Moutonnées, that assigned them by De Saussure.

From the general grinding of glaciers on their beds, the friction

of the fragments on cach other, and the decomposition of many kinds of rock in regions where the alternations of frost and thaw are so common, particularly in the warmer parts of the year, much finely-comminuted mineral matter could scarcely fail to be exposed to the action of any running water, finding its way amid the glacier, and along its sides and bottom.* The streams and rivers derived from glaciers have commonly a marked character, as above noticed, from the quantity of fine mineral matter in mechanical suspension. These sometimes fall into lakes, and leave the fine sedimentary matter behind them, as is the case with many amid and on the skirts of the Alps, while some have a considerable course, as, for example, the Durance (bearing the glacier waters of Monte Viso), and many tributaries of the Po, fed by glacier streams from the southern side of Mont Blanc, and other Italian portions of the high Alps.

Independently of the mass of fragments which may be borne forward by a glacier, when it is on the increase outwards, from a fitting combination of conditions, it ploughs up the ground before it, thrusting forward the loose substances, no matter how accumulated, and with them, should they come in its course, fields, woods, and houses. We remember seeing the Glacier des Bois thus crushing and forcing all before it during its advance in 1819.† These accumulations, to which the transported blocks and minor fragments of rock are being added, as the ice melts, which once supported and carried them onwards,‡ are known as terminal moraines, and by their position a glacier is inferred to

There is much finely-comminuted mineral matter distributed over some parts of many Alpine glaciers. It is sometimes so fine as to enter the interstices of the more porous ice, thus distinguishing the latter from the more compact bands. These "dirt bands," as Professor Forbes terms them, were of much service to him in his examination of the structure of glaciers. Alluding to the discoloration from this finely-comminuted detritus, the Professor observes, "The cause of the discoloration was the next point, and my examination satisfied me, that it was not, properly speaking, a diversion of the moraine, but that the particles of earth and sand, or disintegrated rock, which the winds and avalanches and water-runs spread over the entire breadth of the ice, found a lodgement in those portions of the glacier where the ice was most porous, and that consequently the dirt bands' were merely indices of a peculiarly porous veined structure traversing the mass of the glacier in these directions."-Travels, &c., p. 163. Upon careful examination these "dirt bands" were found to be quite superficial.

In 1820 it attained its greatest known modern advance into the valley of Chamonix.

Respecting the blocks and fragments of rocks thus carried outwards, M. Rendu has remarked that some of them can be occasionally traced to the very commencement of a glacier. -Théorie des Glaciers de la Savoie.

be, for the time, either retreating, advancing, or stationary.* That glaciers advance and decrease is well known, and this to considerable distances, so that many a terminal moraine left at one time, may be again forced forward at another, part of it so caught in the advance of the ice as to be employed in grooving and scratching the solid rocks beneath, then bared and passed over by the glacier. Enormous blocks are often left by glaciers in their retreat; indeed, under such circumstances, they would not only leave the terminal moraines, marking their extension for the time, and during periods of increase, but also their whole load of blocks and fragments, up to the new limits of the decreased glaciers.

Supposing a glacier to advance and retreat from causes which, though variable on the minor scale, are constant for considerable intervals of time, there would be no small amount of blocks and fragments of rock, too considerable to be borne onwards by river action, left either perched on various parts of the mountain sides, or distributed over the valleys, within the range of increase and decrease of these masses of ice in glacier regions. This great and constant general action, continued through long time, would scarcely otherwise than very considerably modify the state of the area from that original condition, when the glaciers were first formed, even supposing no alteration in the relative level, ast respects the sea, of the mountain masses amid which they occur. Avalanches aid in the general descent of fragments of rocks, carrying many, with their snows, to lower levels, sometimes falling on glaciers, sometimes into deep valleys, where the fragments are merely exposed to the ordinary action of rivers.

Taking the general causes and movements of glaciers in the Alps for his guides, the observer is enabled to infer how far glaciers would be found in other regions. M. Elie de Beaumont has pointed out, that from the little variation of climatal con

* Professor Forbes, after quoting M. Venetz (vol. i. of the Transactions of the Swiss Nat. Hist. Society), as pointing out "that passes the most inaccessible, traversed now, perhaps, but once in twenty years, were frequently passed on foot, sometimes on horseback, between the eleventh and fifteenth centuries," considers the evidence important, as "showing that a very notable enlargement of these boundaries (glacier boundaries), was consistent with the limits of atmospheric temperature, which we know that the European climate has not materially overpassed within historic times."-Travels, &c., pp. 43, 44.

† Professor Forbes mentions one of green slate, pushed forward by the glacier of Swartzberg, valley of Saas, and now left at a distance of about half a mile from the glacier by its retreat, estimated by M. Venetz to contain 244,000 cubic feet. This mass, if about 14 cubic feet be taken to the ton, would weigh no less than 17,428 tons. Remarques sur deux points de la Théorie des Glaciers, Annales des Sciences Géologiques, 1842. He observes that glaciers being due to annual and not merely to diurnal conditions, there could be only perpetual snows, and not glaciers, under the equator, where the variations of temperature are only diurnal.

ditions in tropical regions, glaciers would not be expected among the mountains there situated, and sufficiently high to be clothed with perpetual snow. Where the alternations of frost and thaw, snow and rain, would be insufficient to produce the needful amount of névé, assuming this to be the storehouse whence the glaciers are supplied, these would not be found. Looking, therefore, at the different known regions of the world, their varied relief, as regards the distribution of high and low land, the different amount of water supply from the atmosphere, either in the shape of snow, hail, or rain; changes of temperature during various times of the year, and their amount; prevalent or periodical winds -one set dry, the other bringing abundant moisture, and proximity or distance from the sea- the observer finds no want of modifying conditions for the presence or absence, and geological importance of glaciers. At one time glaciers were somewhat doubted among the great range of the Himalaya, but several are now known. The height of the lowest part of the Pinder glacier is estimated at about 11,300 feet above the sea, and that of the Kuplinee glacier at 12,000, which, the height of the perpetual snow line near them being considered at about 15,000 feet, would give a glacier descent of 3,700 feet for the former, and 3,000 feet for the latter.* The lowest part of the glacier of the Ganges is 12,914 feet above the sea, according to Captain Hodgson.

* Captain Strachey, Bengal Engineers, Jameson's Edinburgh New Phil. Journal, vol. xliv., p. 119, and Journal of the Asiatic Society of Bengal, No. viii., p. 794.

CHAPTER XIII.

ARCTIC GLACIERS REACHING THE SEA.-NORTHERN ICEBERGS AND THEIR EFFECTS.-ANTARCTIC GLACIERS AND GREAT ICE BARRIER.-GEOLOGICAL EFFECTS OF ANTARCTIC ICEBERGS.-GLACIERS OF SOUTH GEORGIA. -GLACIERS OF SOUTH AMERICA.-TRANSPORT OF DETRITUS BY RIVER ICE. GEOLOGICAL EFFECTS OF COAST ICE.-EFFECTS OF GROUNDED ICEBERGS.-GENERAL GEOLOGICAL EFFECTS OF ICE.

PROCEEDING from the temperate parts of the world, where lands rise sufficiently high into the atmosphere to obtain a constant covering of snow, and the fitting conditions permit glaciers to descend amid the adjacent valleys at lower levels,* to the Arctic and Antarctic regions, we find glaciers not only covering various portions of land, but jutting into the sea, the line of perpetual snow having descended towards its level. If the observer will in imagination, and by reference to the view of part of it previously given (fig. 84), fill up the valley of Chamonix with sea to the height of about 4000 feet above the village of Chamonix (3,425 feet above the sea), and, therefore, so that the perpetual snow line descended (in round numbers) to within about 1,000 feet from the sea level,† it will readily be seen that numerous glaciers would jut into the sea, resting upon and grating along the rocks forming their bases and sides, until the emersion in the water became such that they floated at their extremities, the transport of fallen fragments being continued in the manner that it now is, until the glacier reached the sea. Here the conditions for their further transport would be modified. Instead of terminal moraines, the blocks would be thrown into deep water, and those which now fall off the lateral moraines would be distributed at greater or less

* In the Pyrenees, the conditions for the production of glaciers would appear to be such, that, where they occur, they are almost always found on the northern slopes of the mountains.

Taking 8,500 feet above the sea as the snow line for the Alps, the altitude inferred by Professor Forbes.