to every hundred miles. Inequalities of three to four hundred fathoms, sufficient to endanger the safety of a cable, might easily exist between any two such points, and subsequent survey has proved this to be often the case. No systematic soundings were made for the Lisbon-Madeira cable, but from those taken it was inferred that the average depth was 2,000 fathoms. During the laying a bank with only 100 fathoms was crossed, and the cable was suspended in a festoon and broken.

With this experience before them, the Silvertown Company determined to make a careful survey for their Cadiz-Tenerife cable. Mr. J. Y. Buchanan, F.R.S., who had been on the scientific staff of the Challenger, was on board the Dacia, and 552 soundings were taken by that ship alone. Two important banks were discovered. The first of these was a coral patch about six miles long and three and a half miles broad. Its shallowest part showed a depth of 435 fathoms. At one end there was a precipitous wall, 285 fathoms in height. Whilst sounding on this ledge the sounder struck ground at 550 fathoms, tumbled over, and struck again at 620 fathoms, and, continuing to fall, eventually found a resting-place at 835 fathoms.

The second bank at its most shallow point was only forty-nine fathoms below the surface, and also had a perpendicular wall. As this was discovered at night-time, a buoy was put over in 175 fathoms, and the ship lay by, in order to continue the work by daylight. On attempting the following morning to raise the mushroom anchor, to which the buoy was moored, the wire mooring-rope parted at seventyfive fathoms from the bottom, and was found to have been almost chafed through at that point, thus proving the existence of a roughedged wall, at least an equal distance from the ground. In spite, however, of these numerous soundings, a fresh bank in the direct path of the cable was discovered, during the laying, by a pioneer ship, a little way ahead. It was at night, and a rocket was fired without delay. The engineer in charge of the laying ship, seeing the signal, and noting, by the dynamometer, the decreasing strain on the cable, although too late to avoid the bank, put the ship's engines full speed astern, and paid out a sufficient amount of slack cable to prevent a repetition of the accident which occurred in the Madeira-Lisbon line.

As this experience proves the impossibility, even with numerous soundings, of discovering every inequality which might prove dangerous to a cable, it becomes a question, in the case of a very long line such as the proposed Pacific, what proportion of the whole cost should be devoted to a preliminary survey. In the neighbourhood of land, careful soundings always repay the time devoted to them; but in wide stretches of mid-ocean, where the bottom is likely to be more uniform, and where work on such an elaborate scale would be a matter of years rather than months, a much greater distance between the soundings becomes imperative. The total length of the Pacific cable

1

route is about 7,000 nautical miles, or three times the length of an average Atlantic cable. A zigzag course, which would give three direct lines of soundings, with thirty miles between the soundings in each particular line (or, altogether, one sounding for every ten miles), would be quite sufficient for practical purposes.

An expedition which was undertaken by the Silvertown Company, to survey a cable route down the west coast of Africa, revealed some interesting facts with regard to the ocean bed in that quarter. A spot called on the Admiralty charts The Bottomless Pit,' lying off Little Bassam on the Ivory Coast, was explored, and a depth of 425 fathoms in close proximity to forty-seven and sixty-seven fathoms on either side of it was found. This formed, perhaps, at one time the mouth of the river Akba. Along the eastward end of the Guinea coast the descent from the hundred-fathom line is well marked, but on nearing the area influenced by the outflow of the Niger and the Congo the slope becomes abnormally gentle. Nine-tenths of the rivers of Africa empty themselves into the Gulf of Guinea, and the sand and mud brought down by them have changed the steep descent to a slowly shelving bank. The water, too, of these rivers, which drain a district remarkable for its heavy rainfall, causes the Gulf to be less salt than any other portion of ocean water in the world. While sounding off the Congo, a submarine cañon or gully two miles broad and 242 feet deep was discovered. This cañon was formed, not like a land cañon by the wearing away of the river bed, but by the heaping up of the mud brought down by the stream into banks on either side. The current was so strong here, that during soundings the ship's engines had to be kept half-speed ahead, and with a hemp line it would have been impossible to get reliable results.

The information gathered from various scientific and telegraphic expeditions goes to prove that the normal depth of the Atlantic Ocean is about 2,500 fathoms, or nearly three miles. In some parts, however, it is almost twice as deep, for off Porto Rico in the West Indies, the Blake, belonging to the United States Navy, found a depth of 4,561 fathoms, or nearly five and a quarter miles. One of the deepest cables in the Atlantic is that of the South American Company, which in one part, between Senegal and the Island of Fernando Noronha, lies in 2,830 fathoms, or a little more than three miles. When laying a line in this depth it is calculated that, with the ship steaming at eight knots an hour, the length of cable from the stern of the ship to the spot where it touches the ground is over twenty-five miles, and that it takes a particular point in the cable more than two hours and a half to reach the bottom from the time that it first enters the water. The deepest sounding yet recorded was taken early in the present year by H.M.S. Penguin in the South Eastern Pacific, about 550 miles to the north-west of New Zealand. This gave 5,155 fathoms, or a depth of nearly six miles.

As a result of these numerous surveys, much valuable information has been obtained with regard to the configuration of ocean beds. Contrary to the opinion formerly held, the bottom of the sea does not present so many striking irregularities as the surface of the earth. Except for islands of volcanic origin and some coral patches, the bed of the Atlantic is an undulating plain of fairly uniform flatness, and may be better compared to a tray with a sharply ascending rim than to a basin. The slope of the land, as a general rule, is continued out into the sea until it reaches a depth of about 100 fathoms, and then increases rapidly to 1,500 and 2,000 fathoms, reaching finally the normal depth of 2,500 fathoms. The area between the 100-fathom line and the shore-usually known as the continental platform-is really submerged land, and if the sea level were suddenly lowered to that extent, England would be connected by dry land to Denmark, Holland, Belgium, France, Ireland, Orkney, and Shetland. Nearly the whole of the North Sea, with the exception of some of the Norwegian fiords, would be laid bare, while the coast of Ireland would be extended 100 miles to the westward. On the other hand, the raising of the sea level to the extent of 100 fathoms would put a large portion of Europe under water, as, indeed, has been several times the case with that continent. Not only in the Atlantic, but in the Pacific, on the west coast of North America especially, the continental platform rises abruptly from the margins of the real oceanic depressed areas, and this phenomenon is one of the strongest arguments in favour of the theory of the permanence of the great ocean beds.

When the Atlantic first came to be sounded in a scientific manner in the course of the Challenger expedition, the result, after the great depths previously reported, was generally felt to be disappointing. Sir C. Wyville Thomson,2 indeed, who was chief of the scientific staff, subsequently described that ocean, with its average of 2,000 fathoms, as a 'thin shell of water.' When, however, it is remembered that over large areas the depth is at least 2,500 fathoms, or 15,000 feet-the height of Mont Blanc-and that in one place a sounding gave 4,561 fathoms, or 27,366 feet-only 2,000 feet less than Mount Everest, the highest point in the world-his expression appears decidedly misleading. The sea level may, in fact, be taken as the relief equator of the globe, almost equidistant from the highest land elevation and the lowest depths of the sea. But while the average height of the land is only 1,000 feet, the average depth of the water is 13,000 feet. Hence an enormous disproportion exists between the mass of land above sea level and the volume of water beneath it. Taking the area of the sea in comparison to the land as 2 to 1, and multiplying by thirteen, the number of times by

2 Sir Wyville Thomson, who died in 1882, was succeeded in his work by Dr. John Murray, F.R.S., who is responsible for all the volumes containing the reports of this expedition.

which it exceeds it in depth, we find that the total volume of ocean water is thirty-six times the volume of the land above sea level.

But although the ocean bed has some depths almost equal to the highest mountains, it is, as a whole, much more uniform than the land. In the Atlantic, for instance, as already stated, only a few volcanic islands break the regularity of the level plateau at the bottom. Near to the land the sea bottom, as a rule, reproduces the leading features of the coast, and on these depends the distance of the 100-fathom line from the shore. Thus the low east coast of England is subtended by the shallow sandbanks of the North Sea, while the precipitous mountains of Norway find their counterpart in the great depth of its fiords.

Ocean deposits may be arranged according to the depth at which they are found in the following manner:

(1) Shore deposits.

(2) Pteropod ooze.

(3) Globigerina ooze. (4) Grey ooze.

(5) Red clay.

(6) Radiolarian and Diatom ooze.

The area of shore deposits is, as Professor Geikie has pointed out, the marginal belt of sea floor skirting the land. The sand and mud brought down by rivers sink to the bottom long before they reach the real ocean depths. The Gulf of Guinea is an exceptional case, and the matter brought down in suspension by its numerous rivers can be traced to a distance of nearly 200 miles out to seaward, and to a depth of 1,600 fathoms. Indeed, off the Congo River shore mud has been found 600 miles from its mouth, and in a depth of 3,000 fathoms. But, as a rule, shore deposits rarely extend beyond the 100-fathom line, and their discovery at greater distances has usually been shown to be due to exceptional agencies. Thus some sand brought up from deep soundings off the north-west coast of Africa was proved to have been carried out to sea by the Harmattan, a powerful and extremely dry wind blowing from the desert of Sahara, and bearing with it a fine dust in such large quantities as to throw a plentiful deposit on the decks of vessels 200 or 300 miles from land.

Pteropod and globigerina oozes are formed from the remains of the shells of the small marine organisms which bear those names. The former is not found at greater depths than 1,500 fathoms, and the latter than 2,500 fathoms, owing to the amount of free carbonic acid gas in the water increasing with the depth, and dissolving their delicate shells of carbonate of lime. These oozes are the best for the purposes of submarine telegraphy. They are so yielding that the cable becomes embedded in them, and their presence is a guarantee against strong under-currents, the scouring effects of which have been traced to a depth of 1,000 fathoms. Shore deposits, on the other

hand, are often very injurious to the sheathing wires of a cable, owing to the iodine contained in seaweed and decaying vegetable matter, which is known to corrode iron rapidly.

Grey ooze is intermediary between globigerina ooze and red clay, and is evidently a mixture of the two. Red clay itself is formed by the decomposition of pumice-stone, and from minerals containing felspar. The action of the waves washes pumice-stone off volcanic rocks, and, being lighter than water, it floats for a long time on the surface of the sea. Towing nets invariably inclose large quantities when lowered in mid-ocean. Some may also be derived from submarine volcanic disturbances; but whether this is the case or no, it is certain that the red clay deposit formed by it accumulates very slowly. This is proved by the frequent presence of meteoric iron in this deposit. I know of no recent discovery in physical geography,' says Professor Geikie, ' more calculated to impress deeply the imagination than the testimony of this meteoric iron from the most distant abysses of the ocean. To be told that mud gathers on the floor of these abysses at an extremely slow rate conveys but a vague notion of the tardiness of the process. But to be told that it gathers so slowly that the very star dust from outer space forms an appreciable part of it, brings home to us, as nothing else could do, the idea of undisturbed and excessively slow accumulation.'

In the red clay are found nodules of almost pure peroxide of manganese, collected round some hard centre like a shark's tooth or a whale's earbone. Curiously enough, no other parts of the structure of large marine animals are recovered from the bottom; but the two just mentioned occur in great abundance, no fewer than 600 shark's teeth and 100 earbones of whales having been brought up on one occasion in a single haul of the dredge. More striking still is the fact that no fossil remains, no portion of a ship, nor any article of human manufacture, has ever been retrieved from the depths of the ocean. A satisfactory explanation of this problem has yet to be forthcoming.

Red clay is deposited in 2,500 to 3,000 fathoms, and beyond that depth Radiolarian and Diatom ooze is found. This ooze is composed of the skeletons of the Radiolaria, or star-shaped organisms, and of the cases of the vegetable Diatoms. As these are of siliceous formation, they are impervious to the action of carbonic acid gas, which dissolves the shells of the Pteropods and Globigerinæ. Diatoms are classed as vegetables owing to their structure, and their mode of reproduction, which is by self-division of the cell. Although individually so small as to be quite invisible to the naked eye, they often occur in such large masses as to give the sea a deep red hue. Mr. J. Y. Buchanan, in the course of his voyage with the Buccaneer on the west coast of Africa, passed through a number of Diatom banks, one of which was 200 miles long and forty to fifty fathoms deep.

6