the deposit will probably be unstratified, like the terminal moraine of a glacier; but whenever the materials are under the influence of a current of water as they fall, they will be sorted and arranged according to their relative weight and size, and therefore more or less perfectly stratified.

In a former chapter it was stated that some ice islands have been known to drift from Baffin's Bay to the Azores, and from the South Pole to the immediate neighborhood of the Cape of Good Hope, so that the area over which the effects of moving ice may be experienced, comprehends a large portion of the globe.

We learn from Von Buch that the most southern point on the continent of Europe at which a glacier comes down to the sea is in Norway in lat. 67° N.* But Mr. Darwin has shown, that they extend to the sea, in South America, in latitudes more than 20° nearer the equator than in Europe; as, for example, in Chili, where, in the Gulf of Penas, lat. 46° 40′ S., or the latitude of central France; and in Sir George Eyre's Sound, in the latitude of Paris, they give origin to icebergs, which were seen in 1834 carrying angular pieces of granite, and stranding them in fiords, where the shores were composed of clay-slate. A large proportion, however, of the ice-islands seen floating both in the northern and southern hemispheres, are probably not generated by glaciers, but rather by the accumulation of coast ice. When the sea freezes at the base of a lofty precipice, the sheet of ice is prevented from adhering to the land by the rise and fall of the tide. Nevertheless, it often continues on the shore at the foot of the cliff, and receives accessions of drift snow blown from the land. Under the weight of this snow the ice sinks slowly if the water be deep, and the snow is gradually converted into ice by partial liquefaction and re-congelation. In this manner, islands of ice of great thickness and many leagues in length, originate, and are eventually blown out to sea by off-shore winds. In their interior are inclosed many fragments of stone which had fallen upon them from overhanging cliffs during their formation. Such floating icebergs are commonly flat-topped, but their lower portions are liable to melt in latitudes. where the ocean at a moderate depth is usually warmer than the surface water and the air. Hence their centre of gravity changes continually, and they turn over and assume very irregular shapes.

In a voyage of discovery made in the antarctic regions in 1839, a dark colored angular mass of rock was seen imbedded in an iceberg, drifting along in mid-ocean in lat. 61° S. That part of the rock which was visible was about 12 feet in height, and from 5 to 6 in width, but the dark color of the surrounding ice indicated that much more of the stone was concealed. A sketch made by Mr. Macnab, when the vessel was within a quarter of a mile of it, is now published. This iceberg, one of many observed at sea on the same day, was between 250 and 300 feet high, and was no less than 1400 miles from any certainly known land. It is exceedingly improbable, says Mr. Darwin, in his notice of this phenom

* Travels in Norway.

Journ. of Roy. Geograph. Soc. vol. ix.

+ Darwin's Journal, p. 288.

p.

526.

enon, that any land will hereafter be discovered within 100 miles of the spot, and it must be remembered that the erratic was still firmly fixed in the ice, and may have sailed for many a league farther before it dropped to the bottom.*

Captain Sir James Ross, in his antarctic voyage in 1841, 42, and 43, saw multitudes of icebergs transporting stones and rocks of various sizes, with frozen mud, in high southern latitudes. His companion, Dr. J. Hooker, informs me that he came to the conclusion that most of the southern icebergs have stones in them, although they are usually concealed from view by the quantity of snow which falls upon them.

In the account given by Messrs. Dease and Simpson, of their recent arctic discoveries, we learn that in lat. 71° N., long. 156° W., they found "a long low spit, named Point Barrow, composed of gravel and coarse sand, in some parts more than a quarter of a mile broad, which the pressure of the ice had forced up into numerous mounds, that, viewed from a distance, assumed the appearance of huge boulder rocks."t

This fact is important, as showing how masses of drift ice, when stranding on submarine banks, may exert a lateral pressure capable of bending and dislocating any yielding strata of gravel, sand, or mud. The banks on which icebergs occasionally run aground between Baffin's Bay and Newfoundland, are many hundred feet under water, and the force with which they are struck will depend not so much on the velocity as the momentum of the floating ice-islands. The same berg is often carried away by a change of wind, and then driven back again upon the same bank, or it is made to rise and fall by the waves of the ocean, so that it may alternately strike the bottom with its whole weight, and then be lifted up again until it has deranged the superficial beds over a wide area. In this manner the geologist may account, perhaps, for the circumstance that in Scandinavia, Scotland, and other countries where erratics are met with, the beds of sand, loam, and gravel are often vertical, bent, and contorted into the most complicated folds, while the underlying strata, although composed of equally pliant materials, are horizontal. But some of these curvatures of loose strata may also have been due to repeated alternations of layers of gravel and sand, ice and snow, the melting of the latter having caused the intercalated beds of indestructible matter to assume their present anomalous position.

There can be little doubt that icebergs must often break off the peaks and projecting points of submarine mountains, and must grate upon and polish their surface, furrowing or scratching them in precisely the same way as we have seen that glaciers act on the solid rocks over which they are propelled.

Journ. of Roy. Geograph. Soc. vol. ix. p. 529

Ibid. vol. viii. p. 221.

In my Travels in N. America, pp. 19, 23, &c., and Second Visit to the U. S., vol. i ch. 2, also in my Manual of Geology, a more full account of the action of floating ice and coast-ice, and its bearing on geology, will be found.

To conclude: it appears that large stones, mud, and gravel are carried down by the ice of rivers, estuaries, and glaciers, into the sea, where the tides and currents of the ocean, aided by the wind, cause them to drift for hundreds of miles from the place of their origin. Although it will belong more properly to the seventh and eighth chapters to treat of the transportation of solid matter by the movements of the ocean, I shall add here what I have farther to say on this subject in connection with ice.

The saline matter which sea-water holds in solution, prevents its congelation, except where the most intense cold prevails. But the drifting of the snow from the land often renders the surface-water brackish near the coast, so that a sheet of ice is readily formed there, and by this means a large quantity of gravel is frequently conveyed from place to place, and heavy boulders also, when the coast-ice s packed into dense masses. Both the large and small stones thus conveyed usually travel in one direction like shingle-beaches, and this was observed to take place on the coast of Labrador and Gulf of St. Lawrence, between the latitudes 50° and 60° N., by Capt. Bayfield, during his late survey. The line of coast alluded to is strewed over for a distance of 700 miles with ice-borne boulders, often 6 feet in diameter, which are for the most part on their way from north to south, or in the direction of the prevailing current. Some points on this coast have been observed to be occasionally deserted, and then again at another season thickly bestrewed with erratics.

The accompanying drawing (fig. 19), for which I am indebted to

Boulders, chiefly of granite, stranded by ice on the coast of Labrador, between lat. 500 and 600 N.

Lieut. Bowen, R. N., represents the ordinary appearance of the Labrador coast, between the latitudes of 50° and 60° N. Countless blocks, chiefly granitic, and of various sizes, are seen lying between high and low-water

mark. Capt. Bayfield saw similar masses carried by ice through the Straits of Belle Isle, between Newfoundland and the American continent, which he conceives may have travelled in the course of years from Baffin's Bay, a distance which may be compared in our hemisphere to the drifting of erratics from Lapland and Iceland as far south as Germany, France, and England.

It may be asked in what manner have these blocks been originally detached? We may answer that some have fallen from precipitous cliffs, others have been lifted up from the bottom of the sea, adhering by their tops to the ice, while others have been brought down by rivers and glaciers.

The erratics of North America are sometimes angular, but most of them have been rounded either by friction or decomposition. The granite of Canada, as before remarked (p. 221), has a tendency to concentric exfoliation, and scales off in spheroidal coats when exposed to the spray of the sea during severe frosts. The range of the termometer in that country usually exceeds, in the course of the year, 100°, and sometimes 120° F.; and, to prevent the granite used in the buildings of Quebec from peeling off in winter, it is necessary to oil and paint the squared stones.

In parts of the Baltic, such as the Gulf of Bothnia, where the quantity of salt in the water amounts in general to one fourth only of that in the ocean, the entire surface freezes over in winter to the depth of 5 or 6 feet. Stones are thus frozen in, and afterwards lifted up about 3 feet perpendicularly on the melting of the snow in summer, and then carried by floating ice-islands to great distances. Professor Von Baer states, in a communication on this subject to the Academy of St. Petersburg, that a block of granite, weighing a million of pounds, was carried by ice during the winter of 1837-8 from Finland to the island of Hockland, and two other huge blocks were transported about the years 1806 and 1814 by packed ice on the south coast of Finland, according to the testimony of the pilots and inhabitants, one block having travelled about a quarter of a mile, and lying about 18 feet above the level of the sea.*

More recently Dr. Forchhammer has shown that in the Sound, the Great Belt, and other places near the entrance of the Baltic, ground-ice forms plentifully at the bottom and then rises to the surface, charged with sand and gravel, stones and sea-weed. Sheets of ice, also, with included boulders, are driven up on the coast during storms, and "packed" to a height of 50 feet. To the motion of such masses, but still more to that of the ground-ice, the Danish professor attributes the striation of rocky surfaces, forming the shores and bed of the sea, and he relates a striking fact to prove that large quantities of rocky fragments are annually carried by ice out of the Baltic. "In the year 1807," he says, "at the time of the bombardment of the Danish fleet, an English sloop-of-war, riding at anchor in the roads at Copenhagen, blew up. In

* Jam. Ed. New Phil. Journ. No. xlviii. p. 439

1844, or thirty-seven years afterwards, one of our divers, known to be a trustworthy man, went down to save whatever might yet remain in the shipwrecked vessel. He found the space between decks entire, but covered with blocks from 6 to 8 cubic feet in size, and some of them heaped one upon the other. He also affirmed, that all the sunk ships which he had visited in the Sound, were in like manner strewed over with blocks."

Dr. Forchhammer also informs us, that during an intense frost in February, 1844, the Sound was suddenly frozen over, and sheets of ice, driven by a storm, were heaped up at the bottom of the Bay of Täarbeijk, threatening to destroy a fishing-village on the shore. The whole was soon frozen together into one mass, and forced up on the beach, forming a mound more than 16 feet high, which threw down the walls of several buildings. "When I visited the spot next day, I saw ridges of ice, sand, and pebbles, not only on the shore, but extending far out into the bottom of the sea, showing how greatly its bed had been changed, and how easily, where it is composed of rock, it may be furrowed and streaked by stones firmly fixed in the moving ice."*

CHAPTER XVI.

PHENOMENA OF SPRINGS.

Origin of Springs-Artesian wells-Borings at Paris-Distinct causes by which mineral and thermal waters may be raised to the surface-Their connection with volcanic agency-Calcareous springs-Travertin of the Elsa-Baths of San Vignone and of San Filippo, near Radicofani-Spheroidal structure in travertin -Lake of the Solfatara, near Rome-Travertin at Cascade of Tivoli-Gypseous, siliceous, and ferruginous springs-Brine springs-Carbonated springs-Disintegration of granite in Auvergne-Petroleum springs-Pitch lake of Trinidad.

Origin of springs.-THE action of running water on the surface of the land having been considered, we may next turn our attention to what may be termed "the subterranean drainage," or the phenomena of springs. Every one is familiar with the fact, that certain porous soils, such as loose sand and gravel, absorb water with rapidity, and that the ground composed of them soon dries up after heavy showers. If a well be sunk in such soils, we often penetrate to considerable depths before we meet with water; but this is usually found on our approaching the lower parts of the formation, where it rests on some impervious bed; for here the water, unable to make its way downwards in a direct line, accumulates as in a reservoir, and is ready to ooze out into any opening which may be made, in the same manner as we see the salt water flow into, and fill, any hollow which we dig in the sands of the shore at low tide.

*Bulletin de la Soc. Géol. de France, 1847, tom. iv. pp. 1182, 1188.