into the bottom of the ship to the extent of perhaps 5,000 tons. This extra weight with the machinery and certain heavy items in the equipment should be quite sufficient to balance the armour and guns.

(2). The Production of Suitable Propelling Machinery.-Before discussing this feature in detail it may be explained here that almost all submarines are propelled by oil engines on the surface and by electric motors when submerged. The electricity for under-water propulsion is stored in a battery of accumulators, which is charged by the motors when run as dynamos or generators by the oil engines. This operation is carried out on the surface while the vessel is cruising at reduced speed, part of the power being used for propulsion while the remainder is absorbed in generating electricity.

Although steam turbines have been to some extent adopted in the French Navy for the propulsion of submarines, the system is unlikely to continue, as the weight of oil fuel used amounts to nearly three times that needed for oil engines.

Already marine oil engines capable of developing 750 B.H.P. per cylinder have been built to order, and will shortly be in service, while experimental engines have given over 2,500 B.H.P. per cylinder on numerous tests. There is no reason to suppose the latter figure to be a limit, although much experimental work will have to be done before larger engines prove reliable.

Judging by the performances of existing battleships, our submarine would require engines of 20,000 horse power to drive. her at twenty-one knots on the surface.

The engines needed would therefore each have to develop 6,600 H.P., assuming three sets to be used in conjunction with three propeller shafts; and allowing six cylinders to each engine, we should have 1,100 H.P. per cylinder against a possible 2,500 H.P. In all probability high speed engines such as are at present used in submarines will be adopted; as this plan makes for lightness, both as regards the engines and the electric motors for propulsion under-water.

A cruising speed of ten knots when submerged should be sufficient for vessels which will mostly operate on the surface. More especially is this the case when, as we shall see later, the electrical equipment is capable of giving much higher speeds for short periods.

Owing to the absence of wave-making when a vessel is submerged the power needed for propulsion is only about half that required on the surface, and this holds good for submarines, in spite of the increased displacement and the resistance caused by the conning tower and other excrescences. In view of this

feature 1,800 H.P. or 600 H.P. on each shaft should easily give a speed of ten knots under water. The capacity of the storage battery would be based on this power for, say, six hours, but the actual motors are always made much larger than the output of the battery on this basis.

When it is remembered that motors of 600 H.P., run as dynamos, would take at least nine hours to charge the battery, one good reason for increasing their power is apparent. Furthermore, the extra power allows of high speeds under water for short intervals-an advantage which may at times be of vital importance.

In the battleship being considered the electric motors should each have a capacity of at least 2,000 H.P., which would allow of charging the battery in rather less than three hours. This power also would give a speed of about fifteen knots below the surface for considerable periods, while a still further increase of power would be available for a few minutes when manoeuvring. We are now in a position to consider the weight of the machinery and the space occupied.

Based on existing submarine engines, the weight should not exceed one ton to every 50 H.P., which gives us 400 tons for 20,000 H.P. The weight of the motors with their accessories may be taken at 150 tons, while the battery will account for another 500 tons. These figures give us a total weight of 1,050 tons, to which must be added, say, 150 tons for auxiliaries, making 1,200 tons altogether. According to figures given by Sir Philip Watts some ten years ago for a battleship of about 16,000 tons displacement, a little over 1,700 tons were allowed for propelling machinery. We are considering a submarine battleship displacing 18,000 tons which on the same basis would have machinery weighing 1,890 tons, so that our estimate of 1,200 tons is well below the allowance for existing surface craft.

Again, the use of oil engines instead of steam machinery with oil fuel reduces the weight of oil needed for a given radius of action by from 33 to 50 per cent.

As regards the space occupied, the oil engines and motors would take up more room than steam turbines of the same power, but there is an enormous saving by substituting a battery for the steam boilers. The battery in question would occupy about 30,000 cubic feet of space, whereas boilers of 20,000 H.P. require 180,000 cubic feet up to the level of the main deck where the depth is about equal to that of the submarine battleship under consideration.

So far as the propelling machinery is concerned, the battle cruiser offers a much more difficult problem than the battleship,

owing to the high speed needed both on the surface and when submerged. The functions of battle cruisers are, however, likely to be considerably curtailed in future submarine navies. At present the type is primarily intended for chasing and destroying enemy cruisers, but when it becomes practicable for the latter to avoid destruction by diving out of sight the main use of the battle cruiser has been eliminated. On the other hand, the destruction of enemy commerce raiders (function 4) has to be considered. The only chance of doing this will be to catch them in shallow water where they cannot dive low enough to be safe from ramming, probably when coming out of some neutral port. This operation will mean gun-power, armour, and speed to ensure success; in other words, exactly the features to be found in the battle cruiser. Unless the attacking vessel possessed heavy guns, armour, and speed the attacked ship might either show fight successfully or get away. Whether submarine battle cruisers will be retained for the sole purpose of chasing elusive enemy commerce raiders remains to be proved, but even, if they are, oil engines of sufficient power are within sight. In other words, what is practically an oil turbine may be available at any moment with almost unlimited power and of far less weight than the type of oil engine at present in use.

Even without this development, experimental Diesel engines, as already stated, have given 2,500 H.P. per cylinder, so that with three shafts having eight cylinders to each, 60,000 H.P. is available, or more than enough power for the submarine battle cruiser under consideration. The speed under water must, of course, be greatly increased, and this will mean a much heavier battery; at the same time, lighter armour would be used, as in existing battle cruisers, leaving more flotation available for the propelling machinery.

Light cruisers with surface speeds up to thirty knots should be quite practicable, in fact, designs have already been made of vessels approaching this type in speed, armament, and displacement.

The submarine destroyer is a much more difficult proposition, owing to the high speed necessary, which involves an enormous amount of power. Our L. class destroyers are very similar in size to the F. class of submarines, but there is a vast difference in speed. The actual speed of our largest destroyers is not given in the Navy List, but we know that their turbines develop 24,500 H.P. or 4,500 H.P. more than even the oil engines we have proposed for the future submarine battleship.

Both the L. class destroyers and the F. class submarines (on the surface) weigh about 950 tons each, and assuming for the

moment that the hulls are equally heavy, each type is capable of carrying the same weight of propelling machinery.

We may also assume that a given power wil! produce the same speed in either case.

In the destroyer the machinery consists of steam turbines and boilers, while the submarine is provided with oil engines, electric motors, and a battery. Now if the machinery in the submarine can be kept within the weight of the turbines and boilers in the destroyer equal speed should be possible.

The present type of oil ́engine is much heavier than a steam turbine of equal power, but the oil Turbine, if successfully developed, should not weigh more than its steam competitor. We have therefore to balance the electric motors and battery in the submarine against the boilers in the destroyer.

These boilers weigh about 300 tons, hence the motors and battery must be kept within this weight. If twelve knots be taken as the cruising speed under water the power required will be about 900 H.P., and the weight of the battery-based upon our previous estimate for the submarine battleship-should not exceed 200 tons. As explained in connection with the battleship, the power of the motors must be far greater than actually required to give the cruising speed, so that the battery can be charged within a reasonable time. In this case the motors should develop 3,000 H.P., and their weight, with accessories, might reach 75 tons. These weights give us a total of 275 tons against the 300 tons for the boilers. It must be remembered also that oil engines or oil turbines consume far less oil than steam turbines driven by oil-fired boilers, hence quite fifty tons less fuel would be needed for a given radius of action.

These considerations show that the submarine destroyer is within the range of practical politics provided that the oil turbine materialises.

We have, however, assumed that the hull of the submarine is no heavier than the hull of the destroyer, but owing to the pressure of the water when submerged, the submarine's skin plating will have to be thicker and heavier, while the framework supporting must also be strengthened, but the 75 tons we have in hand should be sufficient for this purpose.

Of the auxiliary craft needed for every navy, the mine layers will certainly be of the submarine type, and it is probable that a certain number of submarine oil tank vessels will be employed for the purpose of supplying commerce raiders in distant waters W. O. HORSNAILL.

MR. JOSEPH CONRAD AND "VICTORY."1

EVERY new book by Mr. Conrad suggests a new estimate of his work, for every one of his books is a new experiment. He has the rare capacity (a capacity quite different from the self-consciousness of his art) of putting his brain into compartments and of pervading each work with an entirely distinct atmosphere. But with the appearance of Victory the need of a new estimate is more urgent than ever. This is partly fortuitous. It is just twenty years since he published his first novel (Almayer's Folly), and after twenty years, with nineteen books behind him and the assured future of a great reputation before, it is most proper that we should try to focus his achievement afresh. But apart from that, Victory calls for a special recognition. Things have been attempted here which have probably never been attempted before. It is not only a new experiment but a new experiment of an astonishing kind. One could date it, of course, by its technique, by its curious surprises in method so quiet as to pass unnoticed, by its air of wordless melancholy, but in doing so one would not lessen the difficulty of criticism. Far from it. It has often been remarked that between the earlier and the later work of Conrad there is a gulf. True--if one adds that the gulf is not unbridged. And that is what makes broad criticism easy. But how are we to compare novels of the same period so that through the similarity of their technique we can see the deep lines of their divergence? Take, for instance, the case of Chance and Victory. Wherein do these two latest of his novels differ so profoundly? The answer to this question should help us in all criticism of Conrad, because, being a scrupulous artist, he never does anything without a reason, and because his reasons spring not alone from a definite but from a discoverable attitude towards his art.

The main difference between Chance and Victory would appear to be this-that the value of Chance is almost more in the branches than in the trunk, whereas the value of Victory is in the trunk itself. In short, Chance is indirect, Victory is direct. And unlike the directness of such earlier works as The Outcast of the Islands and The Secret Agent, the whole attitude of Victory is that of an austere simplicity of execution. In The Outcast of the Islands the external exuberance, the tropical weight, of Conrad's early manner distracts the attention, in The Secret Agent there is irony between the reader and the work. But in Victory all such things are eliminated-it is like looking through a clear pane of (1) Victory, by Joseph Conrad (Methuen).