in the method of taking deep-sea soundings. The ordinary plan is to carry the lead-line (a strong line or small rope of fine tanned Manilla yarn) from the stern along the ship's side to the bows, and there drop the lead into the sea. As it sinks the rope runs out off the drum on which it is coiled, and when the lead strikes bottom the running ceases. The introduction of fine steel pianoforte-wire for the rope, by Sir William Thomson, is a great improvement upon this clumsy method. The wire sinks quickly through the water, and is pulled in again with a very great saving of time and labour. But the most ingenious of all contrivances for finding the depth of the sea is Siemen's Bathometer, a very recent invention. The bathometer simply stands in the captain's cabin like a barometer, and indicates the depth of the sea over which the ship is passing, just as a barometer indicates the height of the atmosphere above. The action of this ingenious contrivance depends on the attraction of the earth on a column of mercury. This attraction is proportional to the earth's density, and the relative distance of its crust from the mercury column. Earth being denser than water, exercises a greater downward attraction on the mercury. If then there are say a hundred fathoms of water just under the mercury instead of a hundred fathoms of earth or rock, there will be less downward attraction on it. Taking advantage of this law, the mercury column is adjusted so as to indicate the power of the attraction and give the depth of water it corresponds to.

Arrived at the place from which the cable is to be laid, the first thing done is the laying of the shore-end from the ship at her anchorage to the cable-hut on the beach. The cable-hut is generally a small erection of galvanised iron or stone and lime to contain the end of the cable and a few instruments. Cables are never if possible landed in harbours, or where there is danger from anchors. A suitable retired cove is generally selected, not far distant from the town where the telegraph office is, and a short land-line connects the end of the cable to the office. In the cable-hut the land-line and cable meet and are connected together.

The distance from the ship to the cable-hut is accurately measured by the ship's boats or steam-launch, so as to fix the amount of shore-end necessary to reach the shore. This is then coiled in a flat barge or raft, and payed out by hand as the launch tugs the raft ashore. When the water becomes too shallow for the raft to float, the men jump into the water and drag the heavy end ashore. A trench has been cut in the beach up to the cable-hut, and into this the end is laid. In a few moments a test from the cable-hut to the ship announces that the shore-end is successfully fixed.

Everything is now ready for the ship to begin paying out. The anchor has been got up; the paying-out gear is all in working order; the men are all at their appointed places. The cable is being held fast at the ship's stern, and the running out by its own weight is prevented. But directly all are aboard again, the word is given, the screw revolves, the cable is let go at the stern, and the real work of paying out begins.

The cable passes from the tank to the stern of the ship, and from thence to the bottom of the sea. The weight of that part which hangs in the water between the ship and the bottom pulls it out of the ship as the latter moves along. If the ship were stationary, still the cable would run out, but then it would simply coil or kink itself up on the bottom. The object is, however, to lay it evenly along the bottom, neither too tight nor too slack, so as at once to economise cable, and to allow of its being easily hooked and hauled up again from the bottom without breaking from over-tightness. The speed of the ship has therefore to be adjusted to the rate at which the cable runs out. It should be a little under the rate of the paying out, so that there is a slight excess of cable for the distance travelled over. Now the rate at which the cable runs out from the ship is greater, the greater the depth; therefore the speed of the ship must be varied as the depth changes. The rate at which the cable runs out, however, is not entirely dependent on the depth. It can be controlled on board by mechanism. The cable can be held back against the force pulling it overboard. But there is a limit to the extent to which it may be held back, and the tension on it must not be so great as to overstrain it. It is necessary, therefore, to know what tension there is on the cable at any time. To achieve this, two apparatus are used: the friction brake (for holding the cable back) and the dynamometer (for indicating its tension).

The cable is made to run cleanly out of the tank by being allowed to escape through a funnel-shaped iron framework called a 'crinoline.' This prevents it from lashing about or flying off as the ship rolls. It then passes over pulleys to the paying-out drum, round which it is passed several times. The paying-out drum is controlled by an Appold's friction brake, which is simply a belt or strap of iron with blocks of wood studded to it clasping the periphery of the drum and restraining its revolution by the friction of the wooden blocks. The tighter the belt is made to clasp the drum, the greater the friction on the drum, and the greater the force required to make it revolve. After passing several times round the drum, so as to get a good hold of it, the cable passes through the dynamometer to the sheave or grooved pulley projecting from the stern, and from thence it passes to the water. The dynamometer is simply a 'jockey pulley' riding on the cable; that is to say the cable is made to support a pulley of a certain weight; and according as the tension on the cable, due to the weight of cable in the water, is greater or less, so will the weight of this jockey pulley supported by the cable cause the cable to bend less or more. Although this jockey pulley is the essential part of the dynamometer, there are three pulleys altogether, two fixed pulleys at the same level, with the riding pulley between. The cable passes over both the fixed pulleys and under the riding pulley. The weight of the latter bends the cable into a V shape; and as explained above, the depth of this V is greater as the tension on the cable is less. In short the tension of the cable can be told from the depth of the V, which is therefore graduated into a scale of tensions.

By regulating the friction on the brake the cable can be held back, under restrictions of tension indicated by the dynamometer, and the speed of the ship adjusted to give the proper percentage of slack. Sixteen per cent. for a depth of two thousand fathoms is a usual allowance. The slack should vary with the depth, because of the possibility of having to hook the cable and raise it up from the bottom to the surface to repair it. The revolutions of the drum, the tension on the cable, the number of turns of the ship's screw, are constantly observed night and day as the laying goes on.

In the electrical testing-room the same watchful activity prevails. A continuous test is kept applied to the cable, to see that its insulating power keeps steady; in other words, to detect any 'fault' that may occur in the cable which is being laid. This is done by charging the cable throughout, from the end on board the ship to the end left behind in the cable-hut, with a current of electricity, and observing on a galvanometer (an instrument described in a recent paper) the amount of electricity which leaks through the gutta-percha from the copper wire inside to the sea-water outside. To shew that the copper wire too keeps continuous from ship to shore, a pulse of electricity is sent along the cable at stated intervals, usually every five minutes. This is either sent from the ship to the cable-hut, or from the cable-hut to the ship by the electricians left on duty there. The resistance too of the