the bottom is now carefully covered with cement or asphalte up above the rivet heads. Those portions of the vessel at the extreme ends, just above the keel, where the space is too narrow to admit of their being cleaned or painted, may be advantageously filled up solid in the same manner." The Liverpool Underwriters require that "fresh Portland cement shall be laid on so as to cover the frames and rivets heads. The cement is to be raised in the centre to the level of the limber holes, and to be taken up to the upper part of the bilge." All the lower compartments of H.M. iron ship Warrior are filled in with brick-work to prevent the accumulation of bilge water.

579. Where some remedy is not applied, it is recommended that a small quantity of water from the sea, should be let into the ship daily and pumped out again, the strength of the acid will thus be kept down, and the ship remain sweet and clean. The insides of iron ships, are also injured by salt, saltpetre, &c.

580. Mr. PEACOCK, who has made the subject his especial study, observed that on the inside of old ships, where copper pipes have lain in the bilge, or the urine of cattle, or brine from the boilers had lodged under the angle-iron frames, and where the rivet heads and plates had not been properly coated, or on the outside where metal valve seatings occurred, great deterioration took place. Injury from salt-water is increased when brass is near. It is stated that a penny piece accidentally dropped into the bilge of an iron ship and left there for a year, had nearly corroded through, within one-sixteenth of the outer surface, so that had it not been discovered, the steamer might have eventually foundered at sea. The iron steamer Prince of Wales, had a round hole, of four inches diameter, eaten through the bottom by her copper pump; but for the discovery she would have foundered. The constant dropping of the sounding rod is likely to injure the plate immediately below, if not protected; it has made a hole through the bottom of a steamer.

581. To preserve the interior from oxidation, Mr. GRANTHAM, says "good paint becomes a perfect enamel, and while carefully renewed periodically, no sensible decay is perceptible. But to do this requires great care and constant watchfulness. Good white lead is perhaps the best application, but red lead is preferred, though, probably, without any satisfactory reason. Care should be taken that paint is applied under the frames before they are rivetted to the plates, and that no wood should be allowed to come into contact with the iron without a similar thick coat of paint being previously applied. It has been customary, in some cases, to apply a coat of boiled oil to the plates and frames while building, to prevent corrosion when they

are necessarily exposed to the atmosphere." This is, doubtless, a wise step; but the oil should be kept from touching those parts which are to form the joints. For some information applicable to the preceding, see dunnage, iron, and metals.

582. In one case where a master had some copper sheathing to carry, he placed it alongside and touching the iron framework and plates of his vessel. Considerable damage was done to the iron by galvanic action, and much worse consequences would have ensued had there been any salt-water also in contact. An iron ship has been much injured by laying some months alongside a copper bottom ship, that had her copper bottom in contact with the iron ship.

583. With ordinary precautions it would be nearly impossible for a fire to take place, or to gain head, in the hold of an iron ship, provided the hatches were properly secured; for the bulkheads make each division perfectly air-tight, and effectually stop out the atmospheric air, without which fire will not burn,-thus confining the injury, when it does occur, to the compartment in which it originates. 584. The Liverpool Underwriters' Association call the attention of commanders and officers of iron ships to the following remarks:— Newly-launched iron ships, while fitting out, should be kept, if possible, with the head in the opposite direction to that in which they were built, or so near to it as circumstances will permit.

Compass deviations observed in port should be tested at sea as soon as opportunity occurs, especially in new iron ships. The vibration of the machinery in iron steamers may effect the magnetism of the ship, and cause a small alteration in the deviation of the compass.

Compass deviations usually change in amount very gradually as the ship changes her geographical position. The deviations of a compass placed near vertical iron, like a steering compass, generally change more, on change of geographical position, than those of an elevated or standard compass. This change may not show itself while the ship is upon certain courses, but must be guarded against when the course is altered. When an iron ship has been long on one course, and then is put on a new course, she is likely to err in the direction of the old course; thus a ship, after being for some time on a westerly course, and then changing to north or south, will go to the west of her new course.

Besides the ordinary deviation of the compass, there is a deviation caused by the heeling of iron ships, which may increase or decrease the deviation observed when the ship is upright.

There appears to be no deviation from heeling when the ship's head by compass is east or west, but it increases as the ship's head

is moved from these points, and is greatest when the ship's head by compass is near north or south. Cases have been observed in which the deviation resulting from heeling has amounted to as much as two degrees for each degree of heel of the ship-that is, without altering the real direction of the ship's head the apparent alteration in direction has amounted to 40 degrees, by heeling the ship from 10 degrees to starboard to 10 degrees to port.

In north latitude, in ships built head to the northward, with their compasses in the usual position, the deviation from heeling is much larger than in ships built with head to the southward. In north latitude, the north end of the compass needle is drawn to the high or weather side of the ship, as she heels over; the effect being when this deviation is not allowed for, that an iron ship, with a list on northerly courses goes to windward of her apparent course, and on southerly courses goes to leeward of her apparent course. The deviation which arises from the heeling of the ship will vary with the dip of the magnetic needle. In high south latitudes, where the dip is south, the north end of the needle has been observed to deviate towards the low side of the ship. A small deviation towards the low side of the ship has also been observed in north latitude, in some ships which were built in a southerly direction.

It is desirable, therefore, that all iron ships which are liable to heel over should be swung, at least once, with a list to port and with a list to starboard, as well as upright, so as to enable the navigator to estimate what allowance he must make when the ship heels at sea.

A properly trained compass adjuster, with the requisite instruments, can ascertain very nearly what the heeling error will amount to, and can approximately correct it with a vertical magnet, without actually heeling the ship; but until compass adjusters shall be invariably submitted to an examination to test their competency, the only safe plan is to heel and swing the ship so as to ascertain the actual heeling deviation.

The compasses of those iron ships which change their latitude very much cannot be properly compensated by fixed magnets only, but should be partly corrected by vertical iron. The record of careful observations made in high southern latitudes for ascertaining the deviation of the compass when ship's head by compass is east or west, will greatly assist the compass adjuster in perfecting the magnetic compensation of the compasses whose deviations are so observed.

The caps and pivots of the compass cards should be frequently examined at sea, and the blunt pivots and the cracked or otherwise injured caps should be replaced by new ones. Compass errors arising

from mechanical causes of this kind are not unfrequent, and are often wrongly attributed to changes in the magnetism of the ship.

The Admiralty variation chart of the world, constructed by StaffCommander F. J. EVANS, R.N., will be found of great service in ascertaining how much of the total error observed by amplitude or azimuth in the compasses of iron ships is due to variation, and how much to deviation.

The azimuth tables for latitudes between 30 degrees and 60 degrees inclusive, by Staff-Commander J. BURDWOOD, R.N., published by the Admiralty, give the sun's true bearing for nearly every degree of azimuth through the day, so that, by inspection only, when time at ship's place is known, the true direction of the ship's head may be at once ascertained.

The Admiralty Manual," for ascertaining and applying the deviations of the compass, containing charts of magnetic variation, horizontal intensity, and dip; Towson's "Practical Information on the Deviation of the Compass, for the use of masters and mates of iron ships; "the Reports of the Liverpool Compass Committee, and other works connected with the magnetism of iron ships, published by the Admiralty, and Board of Trade, may also be consulted with advantage. W. W. RUNDELL, Secretary.

Underwriters' Rooms, Liverpool, 17th Feb., 1870.

In the P. and O. Steamers and other well conducted ships, the error of the compass is ascertained when possible at least twice a day (morning and evening,) by azimuth and amplitude.

585. Iron Ships-Bulkheads. Mr. COURT, Secretary to the Liverpool Underwriters' Association, issued 1st January, 1859, a Report on bulkheads. In reference to section 9, it may be said that compensation in strength should always be made for the metal taken out of the plates where the bulkheads are rivetted to the ship's sides. Vessels have been known to break in two (especially when unequally laden in different compartments), at the line of bulkhead, owing to the rigidity of the hull at this part, and to the fact that the rivets run in a direct line from keel to gunwale, post-office stamp fashion. It might be better for the angle-iron of the bulkheads to be made with a wider flange, by which space would be obtained for spreading the rivets, and thereby avoiding the direct line. Longitudinal bulkheads give strength, and are of great assistance for stowing a shifting cargo, or where there are different goods liable to injure each other. A midship partition carried from the foremost bulkhead through to the aftermost one would, in combination with the three intermediate

Y

bulkheads, render the ship longitudinally much stronger, and safer to withstand any casualty that may occur; see next page. Bulkheads frequently rust away at the point where they connect with the floor plates, so that when they are wanted they are found useless. Every officer in a ship that is fitted with iron bulkheads should make himself acquainted with all the doors and valves in them, it should not be left to the engineer or carpenter individually.

Mr. COURT'S REPORT ON BULKHEADS IN IRON SHIPS.

SECTIONS 1, 2, and 3, are introductory.

SEC. 4. To determine the size of the compartments we have the following data, viz:-merchant vessels generally load to about seven-tenths of their extreme capacity; that is to say, if the extreme outside dimensions, representing the buoyant power of a vessel, be ten, the weight of ship and cargo is generally about seventenths of this quantity. Vessels laden much beyond this are considered deep. We have, therefore, in general, about three-tenths of the extreme buoyancy of the vessel as a margin, and if this three-tenths be destroyed by leakage arising from accident, or otherwise, the vessel will sink in still water; but in a seaway, the motion of a vessel from the summit of a wave to its base generates a momentum downwards, and the consequence would be that the vessel would sink in a seaway some time before the margin of three-tenths was destroyed. From this it would appear that if a vessel be divided into four equal compartments by three watertight bulkheads, and one of these compartments be filled with water, such vessel would be just safe in a seaway, because nine and a half-tenths of her buoyancy only would be destroyed,-seven-tenths by the weight of the cargo, and two and a half tenths by the destruction of the injured compartment.

5. Practically, it is difficult accurately to divide a vessel into four equal compartments, and it would also be inconvenient, because the forward and after bulkheads would be thrown much too far from the ends of the vessel.

6. The practice latterly has been to put a bulkhead at each end, so as to shut off the space devoted to ship's use from the rest of the vessel. The united contents of these portions vary a little above or below one-tenth of the vessel's extreme capacity in cargo-carrying ships. Now, assuming this one-tenth as the amount required to resist or overcome the descending momentum of the vessel in a seaway (referred to above), and dividing the space between the end bulkheads into four equal, or nearly equal, compartments, by three additional bulkheads, we have provision for nearly every case of emergency.

7. These remarks are independent of the character of the cargo; if of iron and one compartment be filled with water, the vessel will lose the buoyancy due to that compartment, less one-seventh of the weight of iron in it, as the specifie gravity of iron is one-seventh less in water than in air. Thus, if a vessel of 1,600 tons register be laden with iron down to 20 feet, and having a dry side of 6 feet out amidships, ship and cargo weighing about 3,500 tons, such vessel would, if one of these equal compartments were stove in, increase her draught to 24 feet from such cause. With salt or coal the increased draught would be 21 or 22 feet in all; with East India produce the draught would gradually increase