The general structure of the Titanic is shown by the midship section, page 83, and the side elevation, page 129. For about 550 feet amidships she contained 8 steel decks, the boat deck, promenade deck, bridge deck, shelter deck, saloon deck, upper deck, middle deck, and lower deck. The highest steel deck that extended continuously throughout the full length of the ship was the shelter deck. For 550 feet amidships the sideplating of the ship was carried up one deck higher to the bridge deck. The moulded or plated depth of the ship to the shelter deck was 64 feet 3 inches and to the bridge deck 73 feet 3 inches. This great depth of over 73 feet, in conjunction with specially heavy steel decks on the bridge and shelter decks, and the doubling of the plating at the bilges, (where the bottom rounds up into the side,) conjoined with the deep and heavy double bottom, served to give the Titanic the necessary strength to resist the bending stresses to which her long hull was subjected, when steaming across the heavy seas of the Atlantic. The doubling of the plating on the bridge and shelter decks served the same purpose as the cellular steel construction which, as mentioned in the previous chapter, was adopted for the upper deck of the Great Eastern.

The dimensions of the frames and plating of the hull were determined by the builder's long experience in the construction of large vessels. The cellular double bottom, which extended the full width of the ship, was of unusual depth and strength. Throughout the ship, its depth was 5 feet 3 inches; but in the reciprocating engine-room, it was increased to 6 feet 3 inches. The keel consisted of a single thickness of plating, 1½ inches thick, and a heavy, flat bar, 3 inches in thickness and 19½ inches wide. Generally speaking, the shell plates were 6 feet wide, 30 feet long, and 2½ to 3 tons in weight. The largest of these plates was 36 feet long and weighed 4¼ tons.

Amidships, the framing, which consisted of channel sections 10 inches in depth, was spaced 3 feet apart. Throughout the boiler-room spaces, additional frames, 2½ feet deep, were fitted 9 feet apart, and in the engine- and turbine-rooms, similar deep frames were fitted on every second frame, 6 feet apart. These heavy web-frames extended up to the middle deck, a few feet above the water-line, and added greatly to the strength and stiffness of the hull.

Had the inside plating of the double bottom been carried up the sides and riveted on the inner flanges of these frames, as shown in the sketch on page 107, it would have served the purpose of an inner skin; and when the outer skin of her forward boiler-rooms was ruptured by the iceberg, it would have served to prevent the inflow of water to these two large compartments. Mr. Ismay, the President of the International Mercantile Marine Company, in his testimony at the Senate Investigation, stated that among the improvements, which would be made in the Gigantic, now under construction for the company, would be the addition of an inner skin. Doubtless he had in mind the construction above suggested.

The 10-inch channel frames extended from the double bottom to the bridge deck, and some of these bars were 66 feet in length and weighed nearly 1 ton apiece. The frames were tied together along the full length of each deck by the deck beams of channel section, which, throughout the middle portion of the ship, were 10 inches deep and weighed as high as 1¼ tons apiece. The transverse stiffness of the framing was assured by stout bracket knees, riveted to the frames and deck beams at each point of connection, and by the 15 watertight bulkheads, which were riveted strongly to the bottom and sides of the ship, and also by 11 non-watertight bulkheads, which formed the inner walls of the coal bunkers on each side of the main bulkheads.

The bridge, shelter, saloon, and upper decks were supported and stiffened by four lines of heavy longitudinal girders, worked in between the beams, which were themselves carried by solid round pillars placed at every third deck beam. In the boiler-rooms, below the middle deck, the load of the superincumbent decks was carried down to the double bottom by means of heavy round pillars.

Such was the construction of the Titanic; and it will be agreed that, so far as the strength and integrity of the hull were concerned, it was admirably adapted to meet the heavy stresses which are involved in driving so great and heavy a ship through the tempestuous weather of the North Atlantic.

The first sight of such a gigantic vessel as the Titanic produces an impression of solidity and invulnerability, which is not altogether justified by the facts. For, to tell the truth, the modern steamship is a curious compound of strength and fragility. Her strength, as must be evident from the foregoing description of the framing of the Titanic, is enormous, and ample for safety. Her fragility and vulnerability lie in the fact that her framework is overlaid with a relatively thin skin of plating, an inch or so in thickness, which, while amply strong to resist the inward pressure of the water, the impact of the seas, and the tensile and compressive stresses due to the motion of the ship in a seaway, etc., is readily fractured by the blow of a collision.

In a previous chapter it was shown that when the Titanic is being driven at a speed of 21 knots, she represents an energy of over 1,000,000 foot-tons. If this enormous energy is arrested, or sought to be arrested, by some rigid obstruction, whether another ship, a rock, or an iceberg, the delicate outside skin will be torn like a sheet of paper.

It was shown in Chapter IV that protection against flooding of a ship through damage below the water-line is obtained by subdividing the hull into separate watertight compartments, and that, roughly speaking, the degree of protection is proportionate to the extent to which this subdivision is carried. Applying this to the Titanic, we find that she was divided by 15 transverse bulkheads into 16 separate compartments. But, in this connection it must be noted that these bulkheads did not extend through the whole height of the ship to the shelter deck, as they did in the case of the Great Eastern, and therefore it cannot be said that the whole of the interior space of the hull received the benefit of subdivision. As a matter of fact, only about two-thirds of the total cubical space contained below the shelter deck was protected by subdivision. Water, finding its way into the ship above the level of the decks to which the bulkheads were carried, was free to flow the whole length of her from stem to stern. Furthermore, the value of the subdivision below the bulkhead deck depends largely upon the degree to which this deck is made watertight. If the deck is pierced by hatchways, stairways, and other openings, which are not provided with watertight casings and hatch covers, the integrity of the deck is destroyed, and the bulkhead subdivision below loses its value.

It was largely this most serious defect—the existence of many unprotected openings in the bulkhead deck of the Titanic—that caused her to go down so soon after the collision.

Referring now to the side elevation of the Titanic on page 129, it will be noted that the only bulkhead which was carried up to the shelter deck was the first, or collision bulkhead. The second bulkhead extended to the saloon deck, and on the after side of this and immediately against it was a spiral stairway for the accommodation of the crew, which led from their quarters down to the floor of the ship. Here the stairway terminated in a fireman's passage, which led aft through the third and fourth bulkheads, and gave access through a watertight door to the foremost boiler-room. The seven bulkheads, from No. 3 to No. 9, extended only to the upper deck, which, at load draft, was only about 10 feet above the water-line. Bulkhead No. 10 was carried up one deck higher to the saloon deck, as were also bulkheads 11, 12, 13, and 14. Bulkhead No. 15 terminated at the upper deck.

Now, it will be asked: what was the factor in the calculations which determined the height of these bulkheads? The answer is to be found in the Board of Trade stipulations, to which reference was made in Chapter IV, page 62. These stipulations establish an imaginary safety line, below which a ship may not sink without danger of foundering. The safety line represents the depth to which a ship will sink when any two adjoining compartments are opened to the sea and therefore flooded. If the two forward compartments are flooded, for instance, the bow may sink with safety, until the water is only three one-hundredths of the depth of the ship, at the side, from the bulkhead deck. If two central compartments are flooded, the ship is supposed to settle with safety until the bulkhead deck at that point is only three one-hundredths of the depth of the side, at that place, above the water.

The raising of the height of the bulkheads, by one deck, at the engine-room, is due to the operation of this rule; for here the two adjoining compartments, those containing the reciprocating engines and the turbine, are the largest in the ship, and their flooding would sink the ship proportionately lower in the water.

Now it takes but a glance at the diagrams on page 66 to show that the application of the Board of Trade rule brought the bulkhead line of the Titanic down to a lower level than that of any of the other notable ships shown in comparison with her. It was the low bulkheads, acting in connection with the non-watertight construction of the bulkhead deck, that was largely answerable for the loss of this otherwise very fine ship.

Another grave defect in the Titanic was the great size of the individual compartments, coupled with the fact that the only protection against their being flooded was the one-inch plating of the outside skin. If this plating were ruptured or the rivets started along the seams, there was nothing to prevent the flooding of the whole compartment and the entry, at least throughout the middle portion of the ship, of from 4,000 to 6,000 tons of water—this last being the approximate capacity of the huge compartment which contained the two reciprocating engines. Now, if safety lies in minute subdivision, it is evident that in this ship safety was sacrificed to some other considerations. The motive for the plan adopted was the desire to place the coal-bunkers in the most convenient position with regard to the boilers. By reference to the hold plan of the Titanic, page 129, it will be seen that her 29 boilers were arranged transversely to the ship. With the exception of the five in the aftermost compartment, they were "double-ended," with the furnaces facing fore and aft. To facilitate shovelling the coal into the furnaces, the coal-bunkers were placed one on each side of each transverse watertight bulkhead. The coal supply was thus placed immediately back of the firemen, and the work of getting the coal from the bunkers to the furnaces was greatly facilitated. Now, while this was an admirable arrangement for convenience of firing, it was the worst possible plan as far as the safety of the Titanic was concerned; since any damage to the hull admitted water across the whole width of the ship. The alternative plan, which should be made compulsory on all large ocean-going passenger steamers, is the one adopted for the Mauretania, Kaiser Wilhelm II, Imperator, and a few other first-class ships, in which the coal-bunkers are placed at the sides of the ship, where they serve to prevent the flooding of the main boiler-room compartments. It is probable that any one of the ships named would have survived even the terrific collision which sank the Titanic.

The objection has been raised against longitudinal coal-bunkers, that they are not so conveniently placed for the firemen. A large force of "coal passers" has to be employed in wheeling the coal from the bunkers to the front of the furnaces. This, of course, entails an increased expense of operation.

The use of transverse coal-bunkers must be regarded as one among many instances, in which the safety of passenger ships is sacrificed to considerations of economy and convenience of operation.

CHAPTER VII
HOW THE GREAT SHIP WENT DOWN

The Titanic, fresh from the builder's hands, sailed from Southampton, Wednesday, April 10, 1912. She reached Cherbourg on the afternoon of the same day, and Queenstown, Ireland, at noon on Thursday. After embarking the mails and passengers, she left for New York, having on board 1,324 passengers and a ship's complement of officers and crew of 899 persons. The passenger list showed that there were 329 first-class, 285 second-class, and 710 third-class passengers.

The weather throughout the voyage was clear and the sea calm. At noon on the third day out, a wireless message was received from the Baltic, dated Sunday, April 14, which read: "Greek steamship Athinai reports passing icebergs and large quantity of field ice to-day in latitude 41.51 north, longitude 49.52 west." At about 7 P.M. a second warning was received by the Titanic, this time from the Californian, which reported ice about 19 miles to the northward of the track on which the Titanic was steaming. The message read: "Latitude 42.3 north, longitude 49.9 west. Three large bergs five miles to southward of us." Later there was a third message: "Amerika passed two large icebergs in 41.27 north, 50.8 west on the 14th of April." A fourth message, sent by the Californian, reached the ship about an hour before the accident occurred, or about 10.40 o'clock, which said: "We are stopped and surrounded by ice."

These wireless warnings prove that the captain of the Titanic knew there was ice to the north, to the south, and immediately ahead of the southerly steamship route on which he was steaming. The evidence shows that Captain Smith remarked to the officer doing duty on the bridge, "If it is in a slight degree hazy we shall have to go very slowly." The officer of the watch instructed the lookouts to "keep a sharp lookout for ice." The night was starlit and the weather exceptionally clear.

After leaving Queenstown the speed of the Titanic had been gradually increased. The run for the first day was 464 miles, for the second 519 miles, and for the third day, ending at noon Sunday, it was 546 miles. Testimony given before the Court of Inquiry under Lord Mersey, showed that the Chief Engineer had arranged to drive the vessel at full speed for a few hours either on Monday or Tuesday. Twenty-one of the twenty-nine boilers were in use until Sunday night, when three more were "lighted." It is evident that the engines were being gradually speeded up to their maximum revolutions. Both on the bridge and in the engine-room there was a manifest reluctance to allow anything to interfere with the full-speed run of the following day. This is the only possible explanation of the amazing fact that, in spite of successive warnings that a large icefield with bergs of great size was drifting right across the course of the Titanic, fire was put under additional boilers and the speed of the ship increased.

It was shown in a previous chapter on "The Dangers of the Sea," that one of the greatest risks of high-speed travel across the North Atlantic is a certain spirit of sangfroid which is liable to be begotten of constant familiarity with danger and a continual run of good luck. If familiarity ever bred contempt, surely it must have done so among the captain and officers of the Titanic on that fatal night. One looks in vain for evidence that the situation was regarded as highly critical and calling for the most careful navigation;—calling, surely, for something more than the mere keeping of a good lookout—an imperative duty at all times, whether by day or night. Yet the fate of that ship and her precious freight of human life hung upon the mere chance of sighting an obstruction in time to avoid collision by a quick turn of the helm. The question of hitting or missing was one not of minutes but of seconds. A ship like this, nigh upon a thousand feet in length, makes a wide sweep in turning, even with the helm hard over. At 21 knots the Titanic covered over a third of a mile in a minute's time. Even with her engines reversed she would have surged ahead for a half mile or so before coming to a stop. Should she strike an obstruction at full speed, the blow delivered would equal that of the combined broadsides of two modern dreadnoughts.

And so the majestic ship swept swiftly to her doom—a concrete expression of man's age-long struggle to subdue the resistless forces of nature—a pathetic picture both of his power and his impotence. As she sped on under the dim light of the stars, not a soul on board dreamed to what a death-grapple she was coming with the relentless powers of the sea. Latest product of the shipbuilder's art, she was about to brush elbows with another giant of the sea, launched by nature from the frozen shipyards of the north, and she was to reel from the contact stricken to the death like the fragile thing she was!

At 11.46 P.M. the sharp warning came from the lookout: "Iceberg right ahead." Instantly the engines were reversed and the helm was put hard a-starboard. A few seconds earlier and she might have cleared. As it was, she struck an underwater, projecting shelf of the iceberg, and ripped open 200 feet of her plating, from forward of the collision bulkhead to a few feet aft of the bulkhead separating boiler-rooms numbers 5 and 6. It was a death wound! How deeply the iceberg cut into the fabric of the ship will never be known. Probably the first incision was deep and wide, the damage, as the shelf of ice was ground down by contact with the framing and plating of the ship becoming less in area as successive compartments were ruptured.

Whatever may have been the depth of the injury, it is certain from the evidence that the six forward compartments were opened to the sea. Immediately after the collision the whistling of air, as it issued from the escape pipe of the forepeak tank, indicated that the tank was being filled by an inrush of water. The three following compartments, in which were located the baggage-room and mail-room, were quickly flooded. Leading fireman Barrett, who was in the forward boiler-room, felt the shock of the collision. Immediately afterwards he saw the outer skin of the ship ripped open about two feet above the floor, and a large volume of water came rushing into the ship. He was quick enough to jump through the open door in the bulkhead separating boiler-rooms 6 and 5, before it was released from the bridge. The damage just abaft of this bulkhead admitted water to the forward coal-bunker of room No. 5, which held for a while, but being of non-watertight and rather light construction, must have soon given way; for the same witness testified to a sudden rush of water coming across the floor-plates between the boilers.

In spite of the frightful extent of the damage, the Titanic, because of the great height to which her plated structure extended above the water-line, and the consequent large amount of reserve buoyancy which she possessed, would probably have remained afloat a great many hours longer than she did, had the deck to which her bulkheads extended been thoroughly watertight. As it was, this deck (upper deck E) was pierced by hatchways and stairways which, as the bow settled deeper and deeper, permitted the water to flow up over the deck and pass aft over the tops of the after bulkheads and so-called watertight compartments. See page 129.

Now, it so happened that for the full length of the boiler-rooms there had been constructed on upper deck E what was known as the "working-crew alleyway." On the inboard side of this passage six non-watertight doors opened on to as many iron ladders leading down to the boiler-rooms. Not only were these doors non-watertight, but they consisted of a mere open frame or grating, this construction having been adopted, doubtless, for purposes of ventilation. Unfortunately, although there was a watertight door at the after end of this alleyway, there was none at its forward end. The water which boiled up from the forward flooded compartments, as it flowed aft, poured successively through the open grating of the alleyway doors, flooding the compartments below, one after the other.

It does not take a technically instructed mind to understand from this that the safety elements of the construction of the Titanic were as faulty above the water-line as they were below it. The absence of an inner skin and the presence of these many openings in her bulkhead deck combined to sink this huge ship, whose reserve buoyancy must have amounted to at least 80,000 tons, in the brief space of two and one-half hours.

Not until the designer, Mr. Andrews, had made known to the captain that the ship was doomed was the order given to man the lifeboats. The lifeboats, forsooth! Twenty of them in all with a maximum accommodation, if every one were loaded to its full capacity, of something over one thousand, for a ship's company that numbered 2,223 in all. Just here, in this very fatal discrepancy, is to be found proof of the widespread belief that a great ship like the Titanic was practically unsinkable, and therefore in times of dire stress such as this, was well able to act as its own lifeboat until rescuing ships, summoned by wireless, should come to her aid.

The manner of the stricken ship's final plunge to the bottom may be readily gathered from the stories told by the survivors. As compartment after compartment was filled by overflow from the decks above, her bow sank deeper and her stern lifted high in the air, until the ship, buoyed up by her after compartments, swung almost vertically in the water like a gigantic spar buoy. In this unaccustomed position, her engines and boilers, standing out from the floor like brackets from a wall, tore loose from their foundations and crashed down into the forward part of the ship. Probably it was the muffled roar of this falling machinery that caused some of the survivors to imagine that they witnessed the bursting of boilers and the breaking apart of the hull. As a matter of fact, the shell of the Titanic went to the bottom practically intact. One by one the after compartments gave way, until the ship, weighted at her forward end with the wreckage of engine- and boiler-rooms, sank, straight as an arrow, to bury herself deep in the ooze of the Atlantic bottom two miles below. There, for aught we know, with several hundred feet of her hull rising sheer above the ocean floor, she may now be standing, a sublime memorial shaft to the fifteen hundred souls who perished in this unspeakable tragedy!

CHAPTER VIII
WARSHIP PROTECTION AGAINST RAM, MINE, AND TORPEDO

The most perfect example of protection by subdivision of the hull into separate compartments is to be found in the warship. It is safe to say that there is no feature of the design to which more careful thought is given by the naval constructor than this. Loss of stability in a naval engagement means the end of the fight so far as the damaged ship is concerned. Nay, even a partial loss of stability, causing the ship to take a heavy list, may throw a ship's batteries entirely out of action, the guns on the high side being so greatly elevated and those on the low side so much depressed, that neither can be effectively trained upon the enemy. Furthermore, deep submergence following the entrance of large quantities of water, will cut down the ship's speed; with the result, either that she must fall out of line or the speed of the whole fleet must be reduced.

In the battle of the Sea of Japan it was the bursting of heavy 12-inch shells at or just below the water-line of the leading ship of the Russian line that sent her to the bottom before she had received any serious damage to her main batteries. Later in the fight, several other Russian battleships capsized from the same cause, assisted by the weight of extra supplies of coal which the Russians had stowed on the upper decks above the water-line.

In the matter of subdivision as a protection against sinking, there is this important difference between the merchant ship and the warship, that, whereas the merchant ship is sunk through accident, the warship is sunk by deliberate intention. The amount of damage done to the former ship will be great or small according to the accidental conditions of the time; but the damage to the warship is the result of a deliberately planned attack, and is wrought by powerful agencies, designed to execute the maximum amount of destruction with every blow delivered.

A large proportion of the time and money which have been expended in the development of the instruments of naval warfare has been devoted to the design and construction of weapons, whose object is to sink the enemy by destroying the integrity of the submerged portion of the hull. Chief among these weapons are the ram, the torpedo, and the mine. There can be no question that the damage inflicted by the ram of a warship would be far greater, other things being equal, than that inflicted by the bow of a merchant ship. The ram is built especially for its purpose. Not only is it an exceedingly stiff and strong construction; but it is so framed and tied into the bow of the warship, that it will tear open a long, gaping wound in the hull of the enemy before it is broken off or twisted out of place. The bow of the merchant vessel is a relatively frail structure, and many a ship that has been rammed has owed its salvation to the fact that immediately upon contact, the bow of the ramming ship is crumpled up or bent aside, and the depth of penetration into the vessel that is rammed is greatly limited. Furthermore, because of its underwater projection, the ram develops the whole force of the blow beneath the water-line, where the injury will be most fatal. Even more potent than the ram is the torpedo, which of late years has been developed to a point of efficiency in range, speed, and destructive power which has rendered it perhaps the most dreaded of all the weapons of naval warfare. The modern torpedo carries in its head a charge of over 200 pounds of guncotton and has a range of 10,000 yards. Ordinarily, it is set to run at a depth of 10 to 12 feet below the water; and should it get home against the side of a ship, it will strike her well below the armour belt and upon the relatively thin plating of the hull.

Most destructive of all weapons for underwater attack, however, is the mine, which sent to the bottom many a good ship during the Russo-Japanese war. The more deadly effects of the mine, as compared with the torpedo, are due to its heavy charge of high explosive, which sometimes reaches as high as 500 pounds. Contact, even with a mine, is not necessarily fatal; indeed the notable instances in which warships have gone to the bottom immediately upon striking a mine have been due to the fact that the mine exploded immediately under, or in close proximity to the ship's magazines, which, being set off by the shock, tore the ship apart and caused her to go down within a few minutes' time. This was what happened to our own battleship Maine in Havana harbour, and to the Russian battleship Petropavlovsk and the Japanese battleship Hatsuse at Port Arthur.

Enough has been said to prove that when the naval architect undertakes to build a hull that will be proof against the blow, not merely of one but of several of these terrific weapons, he has set himself a task that may well try his ingenuity to the utmost. Protection by heavy armour is out of the question. The weight would be prohibitive and, indeed, all the side armour that he can put upon the ship is needed at the water-line and above it, as a protection against the armour-piercing, high-explosive shells of the enemy.

Heavy armour, then, being out of the question, he has to fall back upon the one method of defense left at his disposal,—minute subdivision into watertight compartments. Associated with this is the placing at the water-line of a heavy steel deck, known as the protective deck, which extends over the whole length and breadth of the hull and is made thoroughly watertight.

The double-skin construction, which was used to such good effect in the Great Eastern, is found in every large warship; and in a battleship of the first class, the two skins are spaced widely apart, a spacing of three or more feet being not unusual. The double-hull construction, with its exceedingly strong framing, is carried up to about water-line level, where it is covered in by the protective deck above referred to. Below the protective deck the interior is subdivided into a number of small compartments by transverse bulkheads, which extend from the inner bottom to the protective deck, and from side to side of the ship. The transverse compartments thus formed are made as small as possible, the largest being those which contain the boilers and engines. Forward and aft of the boiler- and engine-room compartments the transverse bulkheads are spaced much closer together, the uses to which these portions of the ship are put admitting of more minute subdivision.

By the courtesy of Naval Constructor R. H. M. Robinson, U.S.N., we reproduce on page 143 from his work "Naval Construction" a hold plan and an inboard profile of a typical battleship,—the Connecticut,—which give a clear impression of the completeness with which the interior is bulkheaded. Although the ship shown is less than one-half as long as the Titanic, she has 27 transverse bulkheads as against the 15 on the larger ship; and all but nine of these are carried clear across the ship from side to side.

Equally complete is the system of longitudinal bulkheads. Most important of these is a central bulkhead, placed on the line of the keel, and running from stem to stern. On each side of this and extending the full length of the machinery spaces, is another bulkhead, which forms the inner wall of the coal-bunkers. Forward and aft of the machinery spaces are other longitudinal bulkheads, which form the fore-and-aft walls of the handling-rooms and ammunition-rooms.

To appreciate the completeness of the subdivision, we must look at the inboard profile and note that the spaces forward and aft of the engine- and boiler-rooms are further subdivided, in horizontal planes, by several steel, watertight decks or "flats," as they are called. Including the compartments enclosed between the walls of the double hull, the whole interior of the battleship Connecticut, below the protective deck, is divided up into as many as 500 separate and perfectly watertight compartments.

Moreover, in some of the latest battleships of the dreadnought type the practice has been followed of permitting no doors of any description to be cut through the bulkheads below the water-line. Access from one compartment to another can be had only by way of the decks above. Furthermore, all the openings through the protective deck are provided with strong watertight hatches or, as in the case of the openings for the smoke stacks, ammunition-hoists, and ventilators, they are enclosed by watertight steel casings, extending to the upper decks, far above the water-line.

In the later warships, further protection is afforded by constructing the first deck above the protective deck of heavy steel plating and making it thoroughly watertight, every opening in this deck, such as those for stairways, being provided with watertight steel hatches. This deck, also, is thoroughly subdivided by bulkheads and provided with watertight doors.

It sounds like a truism to say that a watertight bulkhead must be watertight; yet it is a fact that only in the navy are the proper precautions taken to test the bulkheads and make sure that they will not leak when they are subjected to heavy water pressure. Before a ship is accepted by the government, every compartment is tested by filling it with water and placing it under the maximum pressure to which it would be subjected if the ship were deeply submerged. If any leaks are observed in the bulkheads, decks, etc., they are carefully caulked up, and the test is repeated until the bulkhead is absolutely tight.

Now, here is a practice which should be made compulsory in the construction of all passenger-carrying steamships. Only by filling a compartment with water is it possible to determine whether that compartment is watertight. To send an important ship to sea without testing her bulkheads is an invitation to disaster. The amount of water that may find its way through a newly-constructed bulkhead is something astonishing; for although the leakage along any particular joint or seam of the plating may be relatively small, the aggregate amount will be surprisingly large.

Let us now pass on to consider the actual efficiency of the watertight subdivision as thus so carefully worked out in the modern warship. Thanks to the Russo-Japanese war, which afforded a supreme test of the underwater protection of ships, the value of the present methods of construction has been proved to an absolute demonstration.

The following facts, which, were given to the writer by Captain (now Admiral) von Essen of the Russian Navy, at the close of the Russo-Japanese war, and were published in the "Scientific American," serve to show what great powers of resistance are conferred on a warship by the system of subdivision above described. The story of the repeated damage inflicted and the method of extemporised repairs adopted, is so full of interest that it is given in full:

"Immediately after the disaster of the night of February 8th," when the Japanese, in a surprise attack, torpedoed several of the Russian ships, "the cruiser Pallada was floated into drydock, and the battleships Czarevitch and Retvizan were taken into the inner harbour, and repairs executed by means of caissons of timber, built around the gaping holes which had been blown into their hulls by torpedoes. The repairs to the Pallada were completed early in April, and about the 20th of June the Czarevitch and Retvizan were also in condition to take the sea. On the 13th of April, during the sortie in which the Petropavlovsk was sunk with Admiral Makaroff on board, the battleship Pobieda, in returning to the harbour, struck a contact mine, and was heavily damaged. Similar repairs were executed, and this ship was able to take her station in the line in the great sortie of August 10.

"On June 23 Captain von Essen's ship, the Sevastopol, was sent outside the harbour to drive off several Japanese cruisers that were shelling the line of fortifications to the east of Port Arthur. This she accomplished; but in returning she struck a Japanese mine, which blew in about 400 square feet on the starboard side, abaft the foremast, at a depth of about 7 feet below the water-line. The rent was from 7 to 10 feet in depth and 35 to 40 feet in length. The frames, ten in all, were bent inward, or torn entirely apart, and the plating was blown bodily into the ship. She was taken into the inner harbour, where the injured portion of the hull was enclosed by a timber caisson in the manner shown in the engravings on page 155. The caisson—a rectangular, three-sided chamber—was built of 9-in. by 9-in. timbers, tongued and grooved and carefully dovetailed. The floor of the caisson abutted against the bilge keel. The outer wall, which was at a distance of about 10 feet from the hull, had a total depth of about 34 feet, the total length of the caisson being about 75 feet. Knee-bracing of heavy timbers was worked in between the floor and the walls, and the construction was stiffened by heavy, diagonal bolts, which passed through from floor to outside wall, as shown in the drawing. Watertight contact between the edge of the caisson and the hull of the ship was secured by the use of hemp packing covered with canvas. The whole of the outside of the caisson was covered with canvas, and upon this was laid a heavy coating of hot tar. The caisson was then floated into position and drawn up snugly against the side of the ship by means of cables, some of which passed underneath the ship and were drawn tight on the port side, while others were attached to the top edge of the caisson and led across to steam winches on deck. After the water had been pumped out, the hydraulic pressure served to hold the caisson snugly against the hull. The damaged plating and broken frames were then cut away; new frames were built into the ship, the plating was riveted on, and the vessel was restored to first-class condition without entering drydock.