that the crew had to depend upon the air contained in the submarine. Also, they had to stop the engine, for without air it cannot work: nor can it work without giving off fumes, which, if admitted to the ship, would soon suffocate the crew. Just before closing up, therefore, the engine is stopped and electric motors take up the task of driving the ship.

Now suppose that, while running submerged, the commander espies, through his periscope, an unsuspecting enemy. He tries forthwith to get as close as he can. Having noted the direction of the vessel and which way she is going and as far as possible her speed, he submerges more deeply, in all probability, lest the white streak which represents the wake caused by his periscope should reveal his presence. For possibly she is one of those terrible destroyers in fair fight with which he has but a poor chance. His only safety lying in complete invisibility, he therefore submerges entirely, trusting to his calculations to lead him in the desired direction. Thus he attempts and, if he have good luck, he succeeds in getting reasonably near to his foe.

Then he must try so to man[oe]uvre that his bow shall at the right moment be pointing towards the quarry, for his torpedo tubes are in the bow and they are fixed, or nearly so at all events, so that he can only fire them in a direction nearly, if not precisely, in the direction of the centre line of his ship.

Nay, he must do even more than that. It will not do to fire the torpedo directly at the ship, for a torpedo is comparatively slow. Suppose it is capable of

forty miles an hour, and the other ship is a mile away: the torpedo will take ninety seconds to reach it. And in that time it may have travelled a mile or so itself. So the submarine man has to allow for that.

Occasionally, therefore, he comes up a little for a moment in the hope of getting a sight of the enemy while not revealing his own presence. Or perhaps he may decide to risk being seen and caught, trusting to the chance of getting his own blow in first. He needs to be a most resourceful man, with clear and keen judgment and supreme self-confidence, or he can never grapple with such a task.

Supposing, then, that he succeeds in getting undetected into a favourable position, as he thinks; at the critical moment the other ship may change its course, and the whole scheme goes awry. Perhaps he then tries to follow, but that is bad, for the end of a ship is not nearly so good a target as the side and the part hit is not so vulnerable. The first torpedo may, however, so disable the vessel as to give him chance to get into position for a second and better shot.

Anyway, when he thinks he has got his best chance he lets off a torpedo, immediately diving to be safe out of harm's way for a while. Then he rises to see the result of his work. If successful he would be sure to hear the sound, for water is an excellent sound-conductor and a submarine is like a gigantic telephone ear-piece.

It must be a nerve-racking job at the best of times, for the submarine is a very vulnerable craft. A member of the crew of a German submarine captured

during the war is reported to have said that out of ten submarines attacked, nine were sunk. That may or may not be true, but it is certain that a very little damage, which would hardly affect an ordinary craft, is enough to sink a submarine. That is because, in order to be able to sink at will, the reserve of buoyancy has to be very low. An ordinary surface ship has at least as much of its bulk above water as below: hence it can take on board a weight of water almost equal to, if not exceeding its own weight before it sinks. At the best a submarine has not more than 30 per cent of excess and so it sinks if water amounting to only 30 per cent of its weight gets into it. In other words, the reserve in one case is at least 100 per cent: in the other at most 30 per cent.