channel, and filling the bottom of the valley. As it slackens its pace, it deposits its burden, first the coarsest material, and then the finer; the finest of all being carried into the sea to form deltas. The matter deposited by the overflow of rivers forms alluvial meadows. Rivers that flow into lakes there deposit the matter held in suspension, and flow out pure and transparent. Lakes are thus gradually silted up, and in the end become alluvial plains.

264. Action of Water on Limestone.-This action is of a chemical nature. Limestone (carbonate of lime) is hardly, if at all, soluble in pure water; but when the water contains carbonic acid, it dissolves the limestone (or chalk) freely. Now, rain brings down with it carbonic acid from the atmosphere, and in sinking through the superficial soil, it imbibes still more from decaying organic matter. In afterwards trickling through the cracks of limestone rocks, it eats them away and forms cavities. Hence the caverns, sometimes miles long, with which many limestone regions are honeycombed.

265. Hard Water.-Water issuing from chalk or other limestone strata is thus always charged with dissolved carbonate of lime, and is known as hard water. One grain of the carbonate in a gallon is counted 1° of hardness. Up to 6° water may be reckoned soft; 8° is moderately hard; at 16° the hardness is excessive. Springs from chalk strata. commonly range from 16° to 18° of hardness, but some springs are four or five times as hard.

The quantity of limestone thus carried into the sea in solution is truly vast. It has been calculated that the Rhine carries into the German Ocean annually as much carbonate of lime as would form a bed of limestone a foot thick and four square miles in area. Notwith

standing this, the relative amount present in sea-water is surprisingly small (see par. 170).

MELTING-SNOW, GLACIERS, ICEBERGS.

266. The Snow-line.-A temporary covering of snow, when melted, has much the same denuding effect as rain. Every one is familiar with the idea of lofty mountain summits remaining white throughout the year, even in warm countries. There is everywhere a limit of elevation above which the heat of summer is insufficient to melt all the snow that falls. This limit is the snow-line. Its average height in the tropics is eighteen miles, and it generally declines towards the polar regions. The height of the snow-line is not regulated by latitude alone, but is modified by local peculiarities. On the north side of the Himalayas, for example, it is 19,000 feet; on the south side, 15,000 feet. The explanation of this anomaly is to be found in the principle stated in par. 229, and in the greater condensation of vapour on the south side, and greater dryness of the air, and consequent evaporation, on the north side.

But although summer does not melt all the snow on high mountains, it melts a great part of it. The powerful rays of the summer sun send torrents of water from snowy mountains into the valleys below; and hence rivers that have their head sources in such mountains are full and turbid in summer, and comparatively shrunken up in winter. This is the case with the Rhine, the Rhone, the Indus, and others.

267. Glaciers.—It is natural to ask why, in the course of years, the snow on lofty mountains does not accumulate to an indefinite depth. Now, snow summits rid themselves of their growing burden in two ways. some places it is hurled down in avalanches to a lower

In

level, where it melts; in other places it moves down as rivers of ice-glaciers. The glaciers of Switzerland are the best known, and their origin may be thus described: The upper end of an alpine valley has generally a cupshaped expansion. In this the snow accumulates by sliding down from the slopes. Such an accumulation is called a névée. The depth is often hundreds of feet, and the pressure of the mass on the lower strata condenses the snow into ice, and pushes it down the slope. That pressure turns snow into ice you can verify for yourselves, if you will squeeze and knead a snowball long enough. The ice thus formed at the bottom of the névée behaves as if it were a viscous fluid like honey. It flows down the valley by gravity, obeying the windings, contractions, and expansions like a river. When it comes to a precipice, it tumbles over in fragments—a cataract of ice; but the pieces unite at the bottom, and continue their course as before.

268. How Ice becomes plastic.-How is it that a substance, any piece of which is brittle like glass, can behave in the mass as if it were plastic? This anomaly is explained by the fact that ice, under pressure, melts at a temperature below 32°. Put a piece of ice at 32° under a piston, and subject it to a pressure of several atmospheres. A portion of it melts; but in becoming water it requires latent heat, and this it takes from the remaining ice and from its own substance, cooling both below 32°-say to 30°. No sooner, however, does the water thus cooled down escape from under the pressure, than it turns to ice again. Now, within the substance of a piece of ice under pressure, the stress must bear more on some points than on others. At such a point a small quantity of the ice melts, and the water escapes into the minute cracks which the crushing force produces around

the spot; but the pressure being now removed, it immediately congeals and solders up the cracks again. The ice is thus able to yield and change its shape without losing its continuity.

269. Glaciers move without slope-Icebergs.-It does not require a valley, or even a slope, to make a glacier. If constant additions were made to a snow-field even on a plain, the pressure and partial melting would convert the bottom layers into ice, and make it spread out in all directions. The whole of Greenland is enveloped in one continuous glacier or ice-cap, which is in constant motion towards the coasts. Opposite the mouths of valleys the glacier, thousands of feet thick, protrudes for miles into the sea, until the buoyancy of the water breaks off a mass which floats away as an iceberg.

The South Polar land, within the parallel of 70° S., is enveloped in a similar covering, presenting everywhere ice-cliffs of the nearly uniform height of about 200 feet. The ice-sheet must extend considerably beyond the land, and dip into the water to a depth of about 200 fathoms. Masses are being constantly torn off from this margin, and form the tabular icebergs of the southern seas. Being usually about 200 feet above water, they must have a depth in all of at least 1400 feet.

270. The Depth of the Ice-cap limited.—The ice sheet would seem to be limited to this thickness by its own weight. This weight not only makes it spread out by a glacier movement, and waste itself at the margin, but it causes melting below. The soil on which the ice rests, being protected from external cold, may be assumed to be about the same temperature as the surrounding seawater, or from 32° to 28°. The weight of the mass will thus be able to cause liquefaction in the lower strata, the water working out channels for itself between the ice

and the land. It is well known that great streams of muddy water issue into the frozen sea from below the great glaciers of Greenland, winter as well as summer.

271. Moraines.-A valley glacier, like those of the Alps, carries on its surface, stones and other debris that fall from the sides of the valley. When the glacier has extended so low that it melts as fast as it advances, the rubbish is deposited, forming a moraine. Long lines of it are also formed at the margins, when the glacier contracts in breadth in summer. Great quantities of sand and stones are imbedded in the substance of the ice; and thus a glacier, as it creeps along, acts like a huge rasp or polishing machine, wearing away the rocks, and leaving them rounded and covered with stric or scratches. The water arising from the melting of the surface in summer may be seen running in considerable streams, wearing out channels on the surface of the ice. When such a stream comes to one of those crevasses or cracks which are found here and there in glaciers, it plunges to the bottom, where it forms channels, and flows out at the foot, a turbid river, charged with the mud formed by the grinding action of the glacier.

272. Ancient Glaciers.-Noting the effect of existing glaciers on the ground over which they slide, and observing like features elsewhere, we infer with certainty that there were glaciers at one time where none have been seen in the memory of man. Geologists have collected unquestionable proofs that much of the lands in the north temperate zone were for ages covered with ice, as Greenland is now. The strata known as the till or boulder-clay is believed to be the work of such widespread glacier action. The Greenland glaciers that protrude into the sea are at the present day forming similar deposits-submarine moraines.