episode of the Crimean War that the development of the sea-going turret

ship was directly due. In the Sea of Azov, in the spring of 1855, Commander Cowper Coles, of H.M. steamer _Stromboli_, constructed in a single night, of barrels, spars and boards, a raft capable of bearing heavy artillery, which he named the _Lady Nancy_; by means of which he brought within range and destroyed by shell fire the Russian stores at Taganrog. The naval operations of this war had drawn general attention to the special problems in connection with the navigation of shallow waters by vessels with a heavy armament, and Commander Coles’ exploit immediately excited official interest. Models of armed rafts were submitted by him for Admiralty inspection, and shortly afterwards he was himself ordered home to give advice upon the requirements of this form of construction: in connection with which the necessity for armour protection for the gun or guns was a point early insisted on by him. In that same year he sketched a design for a belted shallow-draught vessel for the attack of stationary forts which he equipped with guns of the heaviest pattern, each working in a fixed hemispherical shield. From the fixed shield to a revolving turret was a small step. In a short time Commander Coles made himself the enthusiastic exponent of armour-protected guns, mounted in cupolas or turrets on or near the centre-line of a ship so as to give a command over nearly the whole sweep of the horizon. By such a system, he argued, a vessel could be endowed with a concentrated offensive power on any bearing unapproachable by broadside armament, however designed; all guns were effective on almost any bearing without diverting the ship, their force required no evolution to elicit, existing as it did when the ship was at anchor, in dry dock or on a constant course. The height of the turrets gave them a plunging fire, an effect particularly useful now that ships’ sides were armoured and their decks alone remained penetrable. His advocacy of the turret system, aided by the technical assistance of Mr. Brunel, made a deep impression on a large section of the public and gained the interest of the Prince Consort. He did not profess the technical knowledge of a shipbuilder or designer; but in his insistence on the advantages to be derived from the method of mounting guns on the centre-line he wielded arguments of great natural force, and enlisted in his favour the professional sympathies of eminent builders and naval men. In 1860 he produced before the newly founded Institution of Naval Architects a plan of a sea-going ship carrying nine turrets, seven on the centre-line and two off-set so as to allow ahead fire from three turrets. In the following year he wrote to the Admiralty undertaking to prove that a vessel could be built on his principle of armament 100 feet shorter than the _Warrior_ and in all military respects her superior: “I will guarantee to disable and capture her in an hour; she shall draw four foot less water, require only half the crew, and cost the country for building at least £100,000 less. I am ready to stand or fall on these assertions.” Such a pronouncement could not be lightly passed over. Moreover, coast-defence vessels embodying the turret system--light-draught vessels characterized by small tonnage, small cost and indifferent sea-going qualities, in combination with massive protection and a large offensive armament--were already being built by the private firms of this country for various foreign powers. In ’61, for instance, Denmark had ordered the _Rolf Krake_, a turret gunboat carrying a 4½-inch belt and four 68-pounder guns, a pair in each of two armoured turrets; which three years later proved her value in action against a nominally superior force. Prussia had ordered her first ironclad, a turret ship. Holland, Italy, Brazil, Russia--all were known to be purchasing coast-defence vessels of the turret type. And two sea-going turret ships which had been ordered by the American Confederates, and which were building in this country--the _Wyvern_ and _Scorpion_--had been seized and purchased by our government. In these circumstances the Admiralty, though there was a preponderance of official opinion against the idea, resolved to countenance the turret system and give it a trial. The _Royal Sovereign_ was cut down from a three-decker of 120 guns, armoured with a 5½-inch belt and a 1-inch deck, and equipped with four turrets carrying a total of five 12½-ton guns--two in the foremost and one in the remaining turrets. At the same time the _Prince Albert_, also a four-turret ship, was laid down by the firm of Samuda to an Admiralty order. These ships were a distinct success so far as the armament was concerned. They were certainly not ocean-going ships. There were many faults and undesirable features to be found in them. But the disposition of the armament was found satisfactory, and the captain of the _Royal Sovereign_ reported most favourably of his ship, describing her as the most formidable man-of-war; “her handiness, speed, weight of broadside, and the small target she offers, increase tenfold her powers of assault and retreat.” Time, and the progress of artillery, were on the side of Captain Cowper Coles. He saw, and the Admiralty advisers felt, that although it was possible to work existing guns on the broadside, yet increase in the size and weight of guns would sooner or later necessitate the mounting of them on accurately balanced turntables secured by central pivots on the centre-line. Only by such a method could the largest gun be worked and the full weight of metal be poured, as required, on either broadside. In fact the turret, the original object of which was purely defensive, was now regarded from a quite different point of view: as a convenient device by which guns of the highest calibre could be carried and worked. Was complicated machinery objected to? The common winch, the rack and pinion, were in constant use on every railway turntable, nor had the American turrets ever failed in action or caused a loss of confidence in their reliability. Reliance upon a central pivot was disliked? Yet the pivot was already in use for holding the broadside guns of our ironclads--a mere bolt 4 inches in diameter and itself exposed to gunfire.[169] The Admiralty constructors were insistent on the practical difficulties which lay in the way of designing a satisfactory sea-going turret ship. The advantages which had been claimed for turrets were obvious, said Sir Edward Reed; the larger and heavier the individual gun, the greater the gain of mounting it in a turret. But enthusiastic advocates of this method lost sight of the fact that turrets were incompatible with masts and sails, and with the forecastle and high freeboard necessary for good sea-going qualities. At that time, 1865, it was possible to protect and work eight of the largest guns, mounted on the broadside, with as little expenditure of weight as would be required to mount four of the guns two in a turret on the centre-line; while in the latter case they could only fire in two different directions at the same time, whereas in the former they could fire in eight. In order to allow both sides in the controversy to come to grips with the practical difficulties, a committee was formed at the Admiralty in May, ’65, and Captain Coles was asked to produce a turret-ship design by the aid of a draughtsman and with the drawings of the _Pallas_ for guidance. His design, a vessel showing two 600-pounders each mounted in a centre-line cupola, was not considered suitable. So the Board resolved to build a ship to Sir Edward Reed’s design--a fully rigged and masted, high-freeboard ship, with an armour belt and protected bow and stern batteries, and with two centre-line turrets amidships mounted over a central battery, each carrying two 25-ton 600-pounder guns. This was the _Monarch_. She was the first truly ocean-going turret ship, and her performances at sea in ’69 in company with central-battery ships like the _Bellerophon_ and _Hercules_ proved her to be a valuable and efficient unit; by this experiment it was demonstrated, said Mr. Brassey, “that it was practicable to design a thoroughly seaworthy turret ship, although for sea-going purposes a central battery presents great advantages over the turret system.” In the meantime Captain Coles had protested vigorously against the design of the _Monarch_ as representative of his system. The plan was not his; the turrets were mounted so high that there was a large area to protect and the ship, unlike the low-freeboard ships of his own design, presented a large target. But his chief objection was, that the presence of a forecastle and an armoured bow battery annihilated the whole advantage of turret guns by preventing ahead fire from them. After protracted negotiations he obtained Admiralty permission to have a ship built to satisfy his own views and independently of criticism from Admiralty officials. In ’69 the _Captain_, built by Messrs. Laird to his drawings, was launched at Birkenhead. The _Captain_, although generally similar to the _Monarch_ (the growth of artillery limited the number of the turrets to two), differed from her in one important respect: her designed freeboard was only 8 feet as compared with 14; and, by some error in calculation, this dimension proved to be only 6 feet when the vessel was in sea-going trim. This low freeboard, in conjunction with her large sail-area, produced a condition of instability at large angles of heel which led to disaster and sealed the doom of the fully rigged turret ship. Even in the _Captain_ ahead fire was not found possible. In the original plans she had the low freeboard favoured by her designer; but in the later plans poops and forecastles were added to give the necessary sea-going qualities, and ahead fire was thereby sacrificed. Complete mastage was given her: iron masts in the form of tripods to avoid the use of shrouds and to give as clear an arc of fire as possible. The rigging was all stopped short at, and worked from, a narrow flying deck which was built above the turrets. This flying deck provided a working space for the crew, who in a moderately rough sea would not be able to make use of the low upper deck. [Illustration: THE _MONARCH_ From a photograph by Symonds, Portsmouth] On the night of September 6th, 1870, the _Captain_ capsized in a heavy sea off C. Finisterre. In St. Paul’s Cathedral the memorial brass, erected in commemoration of this disaster, records that the _Captain_ was built in deference to public opinion expressed in parliament and through other channels, and in opposition to the views and opinions of the Controller and his department; and that the evidence all tended to show that they generally disapproved of her construction. § The difficulty of combining the turret system with a full rig of masts and sails had for a long time been recognized. Some eighteen months before the loss of the _Captain_, the Admiralty, in the presence of the increasing efficiency of steam machinery, had decided to construct a mastless sea-going turret ship. American experience greatly influenced this decision. In America, where the principle of machinery for propulsion and for working the guns had been accepted with a greater readiness than in Europe, the line of development had been more direct. From the original _Monitor_ a whole series of derivatives had been produced, and from coast-defence vessels of a single turret advance had been made to ocean-going mastless turret ships of low freeboard, carrying the largest smooth-bore guns. These ocean monitors, lacking though they did some features which were considered indispensable in British warships, yet exerted an undoubted influence upon our own construction. Weakly designed in many respects, with small fuel capacity, and unsteady as gun platforms, they were regarded by some writers as the true progenitors of the class of warship which now superseded the masted vessels of the ’sixties. The problem of the naval architect henceforth was greatly simplified. Masts and sails, which had in the past proved such an embarrassment, were now frankly abandoned, with the result that a thousand difficulties which had beset the designer of the turret ship were swept away. No longer had the stability curve to conform to the conflicting requirements of the sailing vessel and the gun platform. The large weight gained by dispensing with masts and sails could be embodied as an addition to the armament or to the fuel carried. The single screw, which in the case of a ship intended to use sails had been almost a necessity, could be replaced by twin screws of greater power; and the change would remove the liability of complete disablement, and give a number of constructive advantages which it is unnecessary to enumerate. Indeed, it may be said conversely, that the adoption of twin screws so improved the reliability of the propelling machinery as to make practicable the abandonment of masts and sails. In April, 1869, the _Devastation_ was commenced. Designed by Sir Edward Reed, she “forestalled, rather than profited by, the dreadful lesson of the _Captain_ and by her success gave proof of the judgment and initiative of the Board and their adviser.” Sir Edward Reed had recognized, more fully than his critics, the conflicting elements inherent in the rigged turret ship. And it is significant that, just at a time when the assured success of the _Monarch_ must have been a gratification to her designer, he should record: “My clear and strong conviction at the moment of writing these lines [March 31st, 1869] is that no satisfactorily designed turret ship with rigging has yet been built, or even laid down.” The _Devastation_ design was a development of those of some previous mastless turret ships, the _Cerberus_, the _Hotspur_, and the _Glatton_ class, which had embodied Sir Edward Reed’s ideas as to the requirements of coast-service vessels. At first given four 25-ton guns, the _Devastation_ was ultimately armed with four M.L. guns each weighing 35 tons and carried in turrets on the centre-line, one at each end of a central breastwork, 150 feet in length, built round the funnels. This central breastwork, raised above the upper deck and armoured along its sides with 10-inch steel, supported the two turrets and enabled the guns to be carried at a desirable height above the water-line. The upper deck itself was low. The sides, up to its level, were protected by a complete belt of armour 8 inches in thickness. The abolition of masts and rigging had a striking effect on the design. Compared with the _Monarch_, of nearly the same tonnage, she carried heavier guns, double the weight of armour, double the amount of fuel, and required little more than half the crew to work her. The loss of the _Captain_, confirming the doubts which experts had expressed as to the seaworthiness of rigged turret ships, caused an alarm for the safety of all turret ships, built and building. In the public mind, in consequence of the reported shortcomings of the American monitors and the known deficiencies of our coast-defence vessels, the belief was growing that the turret system was inherently unsafe. It was believed, also, that mastless ships, having no spread of sail to steady their motion, would be liable to excessive and dangerous rolling. To allay the uneasiness as to the safety of the _Devastation_ and her type a Committee on Designs was formed. The Committee, composed of some of the most eminent of naval architects and officers, made a report in the spring of ’71 which, though it met with considerable opposition from one school, nevertheless “formed the groundwork upon which the English Admiralty determined to construct their policy for the future.” The Committee pronounced altogether against fully rigged ships for the line of battle; it was impossible, in their opinion, to combine in the same vessel great offensive and defensive power and a full spread of canvas. They considered the _Devastation_ class as the most suitable type of armoured ship for future service, and found them to have sufficient stability for safety and to be in almost all respects a satisfactory design of warship. As regards the _Devastation_ herself they recommended some minor alterations, the effect of which was to improve the stability of the ship and to give greater accommodation for the crew. The main alteration consisted in the carrying up of the ship’s sides amidships to the level of the central breastwork, and in continuing the breastwork deck outward to the sides, to form unarmoured side superstructures. Besides the _Devastation_, two others of the type were laid down shortly afterwards, the _Thunderer_ and the _Dreadnought_. The three ships differed from each other slightly in dimensions, but embodied the same characteristic features. Of chief interest is the transition of the unarmoured side superstructures, in the _Devastation_, to an armoured central battery of the same width as the ship, in the _Dreadnought_. The influence of Sir Edward Reed, who had now given place to Mr. Nathaniel Barnaby as Chief Constructor at the Admiralty, was apparent in this evolution. In ’73 he stated publicly his objections to the carrying up of the _Devastation’s_ sides, and pictured a shell entering the unarmoured superstructure and blowing up all the light iron structure in front of the guns. The result was seen in the _Dreadnought_, in which the breastwork was made a continuation of the ship’s side and armoured. More freeboard was also given to the forecastle and the after deck than was found in the _Devastation_ and _Thunderer_, with the desire to make the vessel drier and more comfortable; and, owing to the height at which the turrets were carried, this was found possible without restricting the arcs of fire of the guns. The movement from the monitor type toward the modern battleship in respect of freeboard is clearly traced in these three ships of the _Devastation_ class. Low freeboard, in spite of its effect in rendering inconspicuous the ship in which it was embodied, was gradually being abandoned. High freeboard was foreshadowed for future ships. The loss of the _Captain_ had led to a serious study, by naval architects and mathematicians, of the stability of warships at large angles of rolling, and the advantages of high freeboard were by this time widely appreciated. High freeboard not only made a ship more habitable; by the form of stability curve it gave it allowed a vessel’s beam to be reduced with safety, and thereby contributed to a steadier and more easily propelled ship than would have been obtained without it. In other respects these three ships show the lines along which progress was being made. In the turrets of the _Devastation_ the twin 35-ton guns had been loaded and worked by hand; but in the forward turret of the _Thunderer_ the new hydraulic system of Messrs. Armstrong was applied with success to two 38-ton 12-inch guns; and this system was adopted for both turrets of the _Dreadnought_. The guns were loaded externally, the turrets being revolved by steam, after firing, till the guns were on the requisite bearing; they were then depressed by hydraulic power, and the 700-pound projectiles were rammed into their muzzles by a telescopic hydraulic rammer. In 1879 an accident occurred in the _Thunderer_ which helped, it is said, to hasten the return to breech-loading guns. Simultaneous firing was being carried out; one of the guns missed fire without anyone either inside or outside the turret being aware of it. The guns were loaded again, and, on being discharged, one of them burst. Such double-loading, it was clearly seen, would not have obtained with breech-loading guns. The _Devastation_ had twin screws driven by independent engines, but these were non-compound engines of the trunk type working with a maximum steam pressure of 30 lbs. per square inch. In the _Dreadnought_ an advance had been made to compound the three-cylinder vertical engines, working with 60 lbs. per square inch in engine-rooms divided by a longitudinal watertight bulkhead. § The evolution of the battleship was being forced along at a hot pace by the evolution of artillery. No sooner had the mastless turret ship received the sanction of the Committee on Designs as the standard type for warfare of the immediate future, than a sudden increase in the power of guns necessitated the consideration of new principles and brought into being a new type. So far, defence had managed to compete fairly successfully with offence; the naval architect, by devoting as much as 25 per cent of the total of a ship’s weight to protective armour, had been able to keep level with the artillerist. But it was clear that he could not follow much further, by the existing methods. Armour could not be thickened indefinitely. Penetrable armour was no better than none; worse, in fact, since it was a superfluity, and in a ship a superfluity was doubly wasteful, implying a loss of strength in some other direction. Armour might have to go altogether? It seemed that, after all, the predictions of Sir Howard Douglas might well come true; that, just as gunpowder had forced the foot soldier, after burdening him with an ever-increasing weight, to dispense altogether with body-armour, so rifled artillery would render ship armour increasingly ineffectual and, eventually, an altogether useless encumbrance. The advance in artillery took place in connection with Italian construction. In 1872 Italy laid down the _Duilio_, and a year later the _Dandolo_, two mastless turret ships of a novel class, engined by Penn and Maudsley, and equipped with two diagonally placed turrets each designed to carry two 60-ton Armstrong guns; guns which were afterwards changed to 100-ton guns of 17¾ inches bore. In the same ships the Italians introduced a solution of the armour difficulty. They abandoned vertical armour altogether, except for a very thick belt over the central portion of each vessel which was to protect the vital machinery and the gun turrets. The reply to these was the _Inflexible_, laid down in ’74. We have already seen how, in the last of the _Devastation_ class, the central armoured breastwork was widened to the full beam of the ship. It had been proposed by Mr. Barnaby to take advantage of this arrangement to off-set the two turrets of the _Dreadnought_ at a distance each side of the centre line of the ship, so as to allow a powerful ahead fire. Although not then approved, this suggestion was embodied in the _Inflexible_ as her most distinctive feature. In this, however, she was forestalled by the Italians. Her two turrets, each weighing 750 tons, were carried diagonally on a central armoured citadel plated with compound armour of a maximum thickness of 24 inches. Forward and aft of this citadel the unarmoured ends were built flush with it, and along the centre line was built, the whole length of the ship, a narrow superstructure. This superstructure did not contribute anything to her stability; nor was such contribution needed in view of the comparatively high freeboard. But it rendered unnecessary a flying deck such as had been fitted in the _Devastation_ class, and provided accommodation for the crew, without restricting to any appreciable degree the arcs of fire of the big guns. The _Inflexible_ was of over 11,000 tons displacement, the heaviest and most powerful warship that had ever been built. She was 320 feet in length and 75 feet broad at the water-line; this unprecedented beam being required, in spite of the high freeboard, on account of the height at which the turrets were carried. Nevertheless, so improved was her propulsive efficiency as compared with that of former ships, so great the gain resulting from Mr. Froude’s historic researches on ship form and the action of propellers, that a speed of 15 knots was obtained at a relatively small expense in horse-power. The idea of sails was not yet altogether dead. In deference to a strong naval opinion she was originally designed to carry two pole masts, with sails for steadying her motion in a seaway and as a standby in the event of her propelling machinery being disabled. But this scheme was modified owing to the possibility of falling masts and rigging interfering with the working of guns and screw in action. It was decided that she should be brig-rigged for peace service; and that, on an anticipation of war, she should be docked to allow the cruising masts to be removed and replaced by two short iron masts without yards for signalling and for carrying crows’ nests. But it was in the bold abandonment of armour for the ends of the ship and its concentration on the sides of the citadel that the _Inflexible_ design was most freely criticized. Armour, except in the form of an under-water protective deck, was not used at all forward and aft of the citadel. The ends of the ship were left unprotected, but subdivided; the compartments near the water-line formed watertight tanks filled with coals, stores, or--next to the side of the ship--cork. This criticism was directed from two directions. To many naval men the attempt to beat the gun by adding to the thickness of the armour was a game no longer worth the candle. The point of view, moreover, that the defensive power of a ship was accurately represented by the defensive power of an armour patch upon its side was condemned as altogether too partial and theoretical. The same fallacy was abroad in respect of guns. “Men were apt to think and speak as if the mounting of a single excessively heavy gun in a ship would make her exceptionally powerful, no matter what number of powerful, but still less powerful, guns were displaced to make room for it. The targets and guns at Shoeburyness were held to be real measures of the defensive and offensive powers of ships.”[170] On the other hand, experience was at this time bringing to light the inefficiency of heavy naval artillery. In ’71 a paper by Captain Colomb attracted attention, in which he analysed the effective gun power of the _Monarch_, and showed, by the light of experiments carried out by her against a rock off Vigo in company with _Captain_ and _Hercules_, that “in six minutes from the opening of her fire on the sister ship at 1000 yards, she will have fired twelve shot, of which one will have hit and another may have glanced, and it remains an even chance whether the single hit will have penetrated the enemy’s armour.” In the following summer Mr. Barnaby was himself impressed with the difficulty which the _Hotspur_ experienced in hitting the turret of the _Glatton_ at a range of 200 yards in the smooth water of Portland Harbour: an experiment which, while confirming confidence in the reliability of a turret and its power to withstand shock, led him to question whether we were wise to put so much weight into the protection of turrets, and whether it might not be a better plan to stint armour on guns in order to add to their number and power. From another direction the criticism was more directly effective. In ’75 Sir Edward Reed, now a private member of parliament, made a pronouncement on his return from a visit to Italy in the following words: “The Italian ships _Duilio_ and _Dandolo_ are exposed, in my opinion, beyond all doubt or question, to speedy destruction. I fear I can only express my apprehension that the Italians are pursuing a totally wrong course, and one which is likely to result in disaster.” The Italian Minister of Marine indignantly refuted the assertion, based as it must have been (he said) on incomplete information; and the construction of the _Duilio_ and the _Dandolo_ proceeded. But the remarks of the ex-Chief Constructor applied with equal force to the _Inflexible_; and in the following session he stated as much in the House of Commons. It was possible, he insisted, that in an action the cork and stores which filled the unarmoured ends of the _Inflexible_ might be shot away, and the ends riddled and water-logged; and that in such an event the citadel, though intact, would not have sufficient stability to save the ship from capsizing. The reply of the Admiralty was to the effect that Sir Edward Reed had assumed an extreme case, and that such a complete destruction as he had envisaged was, even if possible, never likely to occur in a naval action. The effect of both statements was to cause widespread anxiety in the public mind, and a lamentable loss of confidence in the projected warship. A decision was therefore made to appoint another Committee, of unquestioned eminence and freedom from bias, to investigate and report on the _Inflexible_ design. In due course the Committee reported. They confirmed in a long statement the Admiralty point of view that the complete penetration and water-logging of the unarmoured ends of the ship, and the blowing out of the whole of the stores and the cork by the action of shell fire, was a very highly improbable contingency; they found that the ship, if reduced to the extremest limit of instability likely to occur, viz. with her ends completely riddled and water-logged, but with the stores and cork remaining and adding buoyancy, would still possess a sufficient reserve both of buoyancy and of stability; and, balancing the vulnerability of the citadel with its 24-inch armour and the destructibility of the unarmoured ends, they came to the conclusion that the unarmoured ends were as well able as the armoured citadel to bear the part assigned to them in encountering the risks of naval warfare, and that therefore a just balance had been maintained in the design, so that out of a given set of conditions a good result had been obtained. Except that a recommendation was made that the system of cork chambers should be extended, no structural alteration from the existing design was proposed. The _Inflexible_ was followed by its smaller derivatives, the _Ajax_ and _Agamemnon_, _Colossus_ and _Edinburgh_, and by the _Conqueror_, an improved _Rupert_, with a single turret. Movement was in the direction of smaller displacements and less armour; construction was influenced at this time more by Italian than by French practice. § All through this transitional decade, 1870-80, experience and various new developments were imperceptibly causing a gradual change of opinion as to what constituted the best type of battleship. At no period, perhaps, was the warship more obviously a compromise, at no time were the limitations of size and weight more keenly felt. So many considerations interacted with one another, so conflicting were the claims made of the naval architect, that it appeared indeed almost impossible to embody them in a satisfactory design. (And yet nothing is more remarkable than the unanimity with which designers, given certain conditions, arrived at the same final result: the _Duilio_ and the _Inflexible_ are a case in point.) Whatever the design might be, it was open to powerful criticism. And the chief part of this criticism was directed, as we have seen, against the use and disposition of the armour. In ’73 Mr. Barnaby had questioned the wisdom of expending a large weight in the protection of turrets. Three years later Commander Noel, in a Prize Essay, was advocating unarmoured batteries, with a view to multiplying the number of battery guns, utilizing for offence the weight thus saved. In ’73 Mr. Barnaby had argued that the stinting of armour on the hull in order to thicken it on the battery would drive the enemy to multiply his light and medium machine-guns. Within a few years warships were bristling with Gatling and Gardner, Nordenfelt and Hotchkiss guns, which by their presence gave a new value to armour, however thin. Mr. Froude, too, in his experiments in connection with the _Inflexible_, brought into prominence the advantage which thin armour on a ship’s ends conferred on her stability. The idea of substituting cellular construction for armour was proving attractive. While the French continued to favour the complete water-line belt, the Italians went to the limit in the _Italia_ and _Lepanto_, in which the water-line was left entirely unprotected by side armour. Such armour as was carried was embodied in the form of a protective deck, a feature found above water and in conjunction with a side belt in our _Devastation_ class, and under water and without side armour in the _Inflexible_ and smaller contemporary ships. The protective deck, which covered the vitals of a ship and deflected shot and shell from its surface, was a device which found increasing favour with naval architects. It was advocated by the Committee on Designs in ’71 as possessing important advantages over a similar weight of side armour. If placed at some distance below water it formed the roof of a submerged hull structure which was immune from damage by gun-fire, the sides of this hull being protected sufficiently by sea-water. If, as was subsequently done, the protective deck were placed at a small distance above water, and if the sides of it were bent down so as to meet the ship’s sides at a distance below water beyond which a shot was unlikely to penetrate, the deck offered other advantages: the vital machinery, though now partly above water, was still protected, the sloping parts of the deck being able to deflect shots which would have penetrated a much thicker vertical plate; moreover, if the ship’s sides were riddled in action, the protective deck still preserved a large portion of the water-line area intact, and thereby secured her lateral stability. The ram was still in favour, but opinion was slowly changing as to the necessity for bow-fire. “It is my impression,” wrote Commander Noel in ’76, “that too great a value was attached by some of the authorities, two or three years ago, to bow-fire; and that the manœuvring of a fleet in action will be more for the purpose of using the ram effectually, and the guns in broadsides on passing the enemy.” The firing of the heavy guns in the approach to ram was considered undesirable, owing to the obscuring of the scene by smoke. In short, bow-fire was not of primary importance, and the disposition of armament which sought to obtain a concentration of bow-fire at the expense of broadside fire was based on a false principle. Commander Noel advocated a broadside ship, of moderate tonnage, with an unarmoured battery of moderate-size guns, with an armour belt round her water-line of 10-inch armour tapering to 5 inches forward and aft, and backed by wood and coal. Watertight subdivisions he proposed as a defence against the ram and the torpedo. As the decade progressed the navy and naval affairs were less and less a subject of public interest. The design of warships continued to be discussed by a small circle, but the Board, alive to the transitional nature of the citadel ships, and under the influence of a national movement for retrenchment and economy, had almost ceased to build. In the three years ’76, ’77, and ’78 England laid down only two armoured battleships, while France laid down a dozen. In ’78 four foreign ships building in this country were hastily purchased on a Vote of Credit. But by 1880 the French armoured navy was once more equal in strength to that of England. The gun, by its rapid evolution, was blocking design. The long debates over sails and steam had been settled; it was now the achievement of powerful breech-loading guns of large and small calibre which threw all existing ideas of warship design into the melting-pot. It became known that the French at last possessed efficient breech-loading guns; and artillerists showed that, in spite of the inconvenience of long-barrelled guns in ships, long barrels and slow-burning powder were necessary if greater powers were to be developed, and that our short-barrelled muzzle-loaders were already becoming obsolete. In the summer of ’79 public interest was aroused by the arrival at Spithead of some Chinese gunboats built by the firm of Armstrong. These gunboats each carried two 12-ton breech-loading guns mounted on centre pivots, one forward and one aft: guns so powerful and efficient compared with any mounted in the Royal Navy, that the possibilities of the diminutive craft were instantly appreciated. The contest between B.L. and M.L. was approaching a climax. The 100-ton M.L. gun was undergoing proof at Woolwich. In August a committee of naval officers visited Germany to witness and report upon the trials of Krupp’s new breech-loaders, and these trials, and those of Armstrong in this country, confirmed the formidable character of the new ordnance. Armour was also improving its power; compound armour (of combined steel and iron) was found to possess unexpected powers of resistance to penetration. The torpedo, moreover, in its growing efficiency was now beginning to have an effect, not only on the details of ship design, but on the whole nature of naval warfare. The influence of the torpedo in its various forms had been appreciated in the early days of the decade.[171] The catastrophic but, happily, fictitious Battle of Dorking, fought in the pages of _Blackwood’s Magazine_ in 1871, had been preceded by a naval action in which all but one of our fine ironclads had been sunk by torpedoes in attempting to ram the French fleet. The moral was obvious. From that time onwards the potential effect of the torpedo was seen to be very great. The ram seemed at last to have found a check. And it appeared that, in combating the ram, the torpedo had once more given the primacy to the fast-improving gun. Broadside actions of the old type, carried on at high range and speed, were predicted.[172] In 1880 a new type of battleship was evolved of sufficient permanence to form the basis of whole classes of future ships. An intimate account of the genesis of the _Collingwood_ design is given us by the biographer of Sir Cooper Key, to illustrate the manner in which that prescient administrator succeeded in forecasting the trend of future construction. In ’66, he says, Captain Key had put on paper a résumé of his ideas on warship design which was clearly several years in advance of current opinion. Briefly, he had maintained that the specifications for our first-class battleships of the future should be drawn to cover the following features so far as possible:--moderate speed, small length and great handiness; perfect protection for vital parts and a complete water-line belt, rather than protection for personnel and above-water structure; a main-deck armament of broadside guns of medium calibre amidships, and of lighter calibre towards the ends, in combination with an upper-deck armament of four large guns in two unarmoured barbettes, one mounted before the foremast and one abaft the mizzen-mast; no sails. But for some years no approach was made to this ideal ship of Captain Key’s; the ideas it embodied were antagonistic to those held by the great majority of his brother officers. “In 1878 there had been laid down by the French, at Toulon, a ship called the _Caiman_. She was 278 feet long, and had a speed of 14½ knots. She carried a single 42-cm. breech-loading rifled gun at the bow, and another at the stern, each mounted _en barbette_, and she further carried on each broadside, between the barbettes, two 10-cm. guns, besides machine-guns. She was heavily armoured by a water-line belt 19½ inches thick amidships, and tapering in thickness towards bow and stern. The middle part of the ship, between the barbettes, was further protected by a steel deck 2·8 inches thick. Evidently, there was in this ship some approach to that general ideal which had been in Sir Cooper Key’s mind in 1866--not, however, more than this. She gave a sort of hint to the constructors at the Admiralty, and, before Sir Cooper Key joined the Board, a new design, based indeed on the _Caiman’s_ hint, but yet differing widely from her, and, by as much as she differed, approaching more nearly to Sir Cooper Key’s ideal, was in process of completion there. The ship was the _Collingwood_.” The _Collingwood_ was of 9150 tons displacement, 325 feet in length, 68 feet in breadth, and 15·7 knots speed. There was in her, for the first time in the navy, that particular disposition of guns which Captain Key had recommended in ’66: two guns at bow, two at stern, on turntables, and a strong broadside armament between them. In the end the adoption of a breech-loading system led to a larger barbette and a smaller battery armament: to 43-ton guns at bow and stern and only 6-inch guns on the broadsides; and in this way the final design differed more than did the original from the ’66 ideal. “The bow and stern guns were protected by barbette and other armour, but Key had required that some protection should be given to the turntables and the machinery for working them. Hydraulics had greatly increased the quantity and importance of this machinery, and as by its means the crews of the guns were very much diminished, we can imagine the admiral concurring in the change as a natural development of his principle. So we can understand him as now definitely concurring in the abandonment of sail power for first-class battleships.” In ’78 he had flown his flag in the _Thunderer_ at sea, and he had then experienced the reliability of the gun machinery and the difficulties attendant on the manœuvring of a modern fleet under sail. Both in armament and in disposition of armour the _Collingwood_ was a great but a natural advance on the citadel ships of the _Inflexible_ type. The symmetrical placing of the big gun turntables, one forward and one aft, proclaimed the advent of new tactical ideas--the recognition of the battleship as a unit which must take its place in the line with others, and the rejection of “end-on” methods of fighting which involved a concentration of bow-fire. The provision of the powerful secondary armament was a tribute to the growing efficiency of French torpedo craft, while at the same time serving, offensively, to force an enemy to protect himself against it: to spread his armour over as large a surface as possible in the attempt to preserve his stability in a protracted action. The concentration of armour on the fixed barbettes and on a partial belt over the central portion of the ship was in accordance with the _Inflexible_ arrangement. But, in consequence of the strictures which had been passed on that vessel and on the exposure of her large unprotected ends, the _Collingwood_ was given a longer belt, though not so thick. Fifty-four per cent of her length was covered with 18-inch compound armour, as compared with 42 per cent, and 24-inch armour, in the former ship. Although this longer belt appeared to confer greater longitudinal stability on the ship, its narrowness was such that it was of doubtful efficacy, as Sir Edward Reed was not slow to point out. So narrow was this belt, so big still remained the unarmoured ends, that the slight sinkage caused by their filling would bring the whole of the armour belt, he said, under water. Thus all the advantage arising from a longer citadel was more than destroyed by this lowering of the armour, and, so great was the consequent danger of the vessel capsizing, that he hesitated to regard the _Collingwood_ as an armoured ship. The _Collingwood_ was laid down in July, 1880. But what was there to show that her design would be in any degree permanent? Was it safe to consider it sufficiently satisfactory to form the master-pattern for a number of new ships, urgently required? For a short time there was uncertainty. “The French type, where there were isolated armoured barbette towers generally containing single heavy guns placed at the ends and sides of the ships upon the upper deck, with broadside batteries of lighter guns, entirely unprotected by armour, upon the deck below, did not commend itself to the English naval mind, yet, in the sort of despair which possessed us, the new Board turned somewhat towards the French system. The _Warspite_ and _Impérieuse_ were laid down in 1881, and were again a new departure in British design.... It was intended to adhere to sail power in these new types, and it was only after they were approaching completion that the utter incongruity of the proposal was realized, and sail power was given up in the last of the armoured ships to which it was attempted to apply it.” But the Admiralty still wished, without alarming the public, to regain as soon as possible a safe balance of armoured construction over that of France. “There was no design before the Board which was more likely to perpetuate itself than that of the unlaunched _Collingwood_. Suppose a bold policy were adopted? Suppose it were assumed that the time had come when diversities of type were to cease, would it be made less likely by the frank abandonment of sail power?” The bold step was taken. Four more ships to the _Collingwood_ design were laid down in ’82, the five being thereafter spoken of as the “Admiral” class. “At the time, little note was taken of this very great step in advance. Even at this day it is scarcely remembered that this is the step which made possible, and led up to, our present great battle fleet, and that never before had so many as five first-class ironclads of a definite type been on the stocks together.... In the Admiral class there was the definite parting with sail power, the rejection of the tactical ideas brought to a climax in the _Inflexible_, and, above all, the definite adoption of the long-barrelled breech-loading rifled gun. Without question, we must say that we owe the Admiral class, and all that has followed, in great part to the enterprising and yet well-balanced mind that then governed the naval part of the Council at Whitehall.” § At this point in the evolution of the ironclad it is convenient to bring our survey to an end. The _Collingwood_ marks the final return (with one or two notorious exceptions) to the truly broadside ship, the ship with armament symmetrically disposed fore and aft, intended to fight with others in the line. From the Admiral class onwards the modern battleship evolved for years along a continuous and clearly defined curve of progression. It only remains to close this brief and necessarily superficial historical sketch with a few remarks upon the classification of warships. In tracing the types of ironclads which superseded each other in direct succession, no mention has been made of other than those which formed in their time the chief units of naval force. Other war-vessels there were, of course, subsidiary to the main fighting force, whose value and functions we now briefly indicate. So long as sails remained the sole motive power, warships retained the same classification as they had received in the seventeenth century. “Up to the time of the Dutch Wars,” says Admiral Colomb, “ships were both ‘royal’ and of private contribution; of all sorts and sizes and ‘rates.’ Fighting was therefore promiscuous. Fleets sailed in the form of half-moons, or all heaped together and, except for the struggle to get the weather gage, there were no tactics. Actions were general.” Then, in order to protect their fleets from the fire ship, the Dutch first introduced the Line of Battle: “in which formation it was easy for a fleet to leeward to open out so as to let a fire ship drift harmlessly through.” And so the efficacy of the fire ship was destroyed. “But now, with a Line, each ship had a definite place which she could not quit. Hence the diversities in sizes began to be eliminated. The weakest ships, which might find themselves opposite the strongest, were dropped for ships ‘fit to lie in the line,’ i.e. for what were afterwards called ‘line-of-battle ships.’ These ships would be individually as powerful as possible, only subject to the objection of putting too many eggs in one basket. Uniformity would thus be attained. The fleet of line ships, however, required look-outs or scouts, which could keep the seas and attend, yet out-sail, the fleet. Hence the heavy frigate. Lastly, there was the much lighter attendant on commerce (either by way of attack or defence), the light cruiser.” Although this differentiation of types was based ostensibly upon displacement or tonnage, in reality it was formed on a more scientific basis. Admiral Sir George Elliot demonstrated, in 1867, that the real basis was not a rule of size, but a _law of safety_, similar to that which operates in the natural world; a law so important that it should under no circumstances be disregarded. He showed that sailing ships conformed to this law. He showed that the reduction of a vessel’s size, for instance, endowed her with smaller draught and an increased speed; that the dispensing with one quality automatically gave another in compensation; and that thus the weakly armed vessel always possessed the means, if not to fight, to escape from capture.[173] With the coming of steam and armour, all this was changed. Size had now no inherent disability; on the contrary, the larger the ship the greater the horse-power which could be carried in her, the greater her probable speed and sea endurance. The small ship had no advantages. The old classification had clearly broken down. The first ironclads, the _Warrior_ and her successors, although of frigate form, belonged to no particular class; they were of a special type intended to cope with the most powerful ships afloat or projected; and subsequent ships were designed with the same end in view. These ships being faster as well as more powerful than those of a smaller size, there was no object in attempting to build others of a frigate class for the purpose of outsailing them. As material developed, and as the warship became more and more obviously a compromise between conflicting qualities, differentiation of types was once more seen to be necessary. Attempts were made to classify on the bases of displacement, material, defensive and motive power, service, system of armament. In the end British construction divided itself into two categories: armoured and unarmoured vessels. And each of these categories was subdivided into classes of ships analogous to those of the old sailing ships. But, during the transitional period 1860 to 1880, when armour and iron ships, steam engines, rifled guns, and fish torpedoes, were all in their infancy and subject to the most rapid development, no such classification was recognized. The circumstances of the Crimean War, with the adoption of armour and the sudden and enormous growth in the unit of artillery force which took place soon afterwards, led to the first differentiation of ironclads, into ocean-going and coast-defence vessels. We have already noted this fact. We have seen how, especially to the lesser Powers, the turreted monitor appeared to offer an economical and effective form of naval force; and we have noted how, in America, the evolution proceeded in the opposite direction, viz. from coast-defence monitor to ocean-going turret ship. This differentiation prevailed for many years. It prevailed even in the British navy, in spite of its being in full opposition to the offensive principle on which that navy had always based its policy. Later, although convinced that in any war involving this country and its colonies the chief combats must be fought in European waters, naval opinion saw the necessity for a type of ship designed primarily for the defence and attack of commerce: a speedy, lightly armed and protected type capable of overhauling and injuring a weaker, or of escaping from a more powerful enemy. The American War of ’62, in which no general sea action was fought, gave the impulse to the construction of the type which eventually became known as the _cruiser_. Vessels were built in ’63 expressly to overtake Confederate vessels and drive from the seas the Southern mercantile marine. These vessels were to annihilate the enemy’s commerce without being drawn themselves to take part in an engagement, unless in very favourable circumstances. Several such ships were built. The first, the _Idaho_, was a complete failure; the next attempt was little more successful; and those subsequently constructed, the _Wampanoag_ class, the finest ships of the type which existed at the close of the war, which were designed for 17 knots and to carry sixteen 10- or 11-inch smooth-bore cast-iron guns on the broadside and a revolving 60-pounder rifle in the bows, suffered from miscalculations in design and from the weakness peculiar to long and heavily weighted timber-built ships. “These pioneers of the type,” says Brassey, “were followed, both in England and in France, by vessels believed by the builders of their respective countries to be better adapted for the work for which they were designed.” At first England and France had built and appropriated small ironclads to this secondary service; in France the _Belliqueuse_, in England the _Pallas_, were designed to this end. But in ’66 the first ship of the cruiser type was built for the British navy: the _Inconstant_, of Sir Edward Reed’s design, an iron-built, fine-lined vessel with a speed of 16 knots and a large coal capacity. She was followed by the corvettes _Active_ and _Volage_, and then, in ’73, by the _Shah_ and _Raleigh_. Experience with the early cruisers showed the advantages of large displacement. “The greater number of the American corvettes had now been launched. A trial of one of them showed that the high hopes which had been entertained of their performance were fallacious. It now appeared no longer necessary that the English corvettes should possess such extraordinary power and speed, qualities which necessarily required very large displacements. The Admiralty, however, still believing in the wisdom of the policy which they had previously adopted, decided to follow a totally different course from that which all other navies had been compelled by financial considerations to follow. So far from diminishing the size of their ships, increased displacement was given to the new designs.”[174] Full sail power was still required, for the high-power steam engine used by the cruiser for fighting purposes was most uneconomical. The _Raleigh_, for instance, burned her six hundred tons of coal in less than 36 hours, at full speed. But after the _Raleigh_ came a slight reaction. With a view to economy a smaller type of vessel was designed, the smallest possible vessel which could be contrived which would possess a covered-in gun deck in combination with other features considered essential in a frigate class; the result was the _Boadicea_ or the _Bacchante_ class. In the late ’seventies size again increased, and the _Iris_ and _Mercury_, unsheathed vessels of steel, with coal-protection for their water-line and extended watertight subdivision of the hull, were laid down. From the unarmoured, unprotected cruiser was in time evolved, by the competition of units, the armoured cruiser. Russia led the way. Her _General-Admiral_, the first belted cruiser, was built to compete with the _Raleigh_ and _Boadicea_. Then England designed the _Shannon_, partially belted and with protective deck and coal protection, to outmatch her. Eventually the cleavage came, and the cruisers were themselves divided into two or more classes, in accordance with their duties, size and fitness for the line of battle. * * * * * Of the development of torpedo craft this is not the place to write; although the torpedo was fast growing in efficiency and importance, it had not, before 1880, become the centre and cause of a special craft and a special system for its employment in action. But after that date the creation of torpedo flotillas began to exercise a marked and continuous effect upon the evolution of the ironclad. The fish-torpedo, improving at a phenomenal rate in the first years of its development, and at first esteemed as of defensive value and as a counter to the ram, became, after 1880, an offensive weapon of the first importance. The ram, already suspected of being placed too high in popular estimation, suffered a decline; the danger of its use in action was emphasized by naval officers, whose opinion alone was decisive: its use, as an eminent tactician explained, reduced the chances of battle to a mere toss-up, since there was “only half a ship’s length between ramming and being rammed.” The gun developed in power, in range, and accuracy; but not (up to the end of the century) at so great a rate as its rival, the torpedo. The steam engine affected all weapons by its continuous development. It depressed the ram, enhanced the importance of the gun, and endowed the torpedo with a large accession of potential value in placing it, in its special fast sea-going craft, within reach of the battleship; moreover, it enabled the cruiser to regain its old supremacy of speed over the line-of-battle unit. Armour, quick-firing guns, secondary armament, watertight construction, net defence, all influenced the development of the various types. But it was the torpedo, borne into action by the high-speed steam engine, which had the greatest effect on naval types in the last two decades of the century, and which at one time bid fair to cause a constructional revolution as great as that of 1860. The torpedo, according to a school of French enthusiasts, had destroyed the ironclad battleship and dealt a heavy blow at English sea power by paving the way for an inexpensive navy designed for a _guerre de course_. The ironclad was dead, they cried, and might as well be placed in the Louvre museum along with the old three-deckers! In Italy and Germany, too, the logic of facts seemed to point to a vast depreciation in the power of existing navies: the fate of the expensive ironclad seemed assured, in the presence of small, fast, sea going torpedo-boats. Still, it was noticed, England laid down battleships. True; this was quite in keeping with her machiavellian policy. Had she not resisted--“not blindly, but with a profound clairvoyance”--all the inventions of the century? Had she not successfully baulked the development of Fulton’s mines, steam navigation, the shell gun, and the ironclad itself? And, now that steam had made the blockade impossible and the torpedo had attacked the ironclad effectually, making sea-supremacy an empty term, could not the British Empire be destroyed by taking the choice of weapons out of England’s hands? The prospect was alluring. Yet the ironclad survived the menace and remained the standard unit of naval power. Expensive, designed with several aims and essentially complex,--a compromise, like man himself,--it could not be replaced by a number of small, cheap, uni-functional vessels, each constructed for one sole and special purpose, without loss of efficiency and concentration of power. Nor could it be supplanted by a type which, like the sea-going torpedo-boat, could only count on an ascendancy over it in certain moments of its own choosing--for example, at night-time or in a fog. To every novel species of attack the ironclad proved superior, calling to its aid the appropriate defensive measures. FOOTNOTES [1] Sir Harry Nicolas: _History of the Royal Navy_. [2] The greatest authoritative works on ancient and medieval shipping, it should be mentioned, are the _Archéologie Navale_ and the _Glossaire Nautique_ of M. Jal, published in 1840 and 1848 respectively. [3] Corbett: _Drake and the Tudor Navy_. [4] Corbett. [5] Oppenheim. [6] Corbett. [7] Navy Records Soc.: Edited by Sir John Laughton. [8] Cases were known where ships, unfit for sea, completed their voyage in safety, to fall to pieces immediately on being taken into dock and deprived of that continual support which they derived from the water when afloat (_Charnock_). [9] Chief-constructor D. W. Taylor, U.S.N. [10] Creuze: _Shipbuilding_. [11] Manwayring. [12] Navy Records Soc.: 1918. _Edited by_ W. G. Perrin, Esq., O.B.E. [13] Captain John Smith’s _Sea Man’s Grammar_ also appeared in the early part of this century. [14] Sir J. Knowles, F.R.S. [15] Willett: _Memoirs on Naval Architecture_. [16] It has been suggested that the restricted draught given to the Dutch ships, owing to the shallowness of their coast waters, had the result of necessitating a generous breadth, and therefore made them generally stiffer than vessels of English construction. [17] Derrick in his Memoirs refers to this ship us having been built of burnt instead of kilned timber, and as having special arrangements for circulating air in all its parts. [18] Charnock. [19] Colomb: _Sea Warfare_. [20] Creuze: _Papers on Naval Architecture_. [21] Even the scientific Sir William Petty cast a veil of mystery over his processes. “I only affirm,” he writes, “that the perfection of sailing lies in my principle, _finde it out who can!_” (See Pepys’ Diary for 31st July, 1663.) [22] Creuze: _Shipbuilding, Encycl. Brit._, 7th Edition, 1841. It should be mentioned that the work of Dr. Colin McLaurin, of Edinburgh, in giving a mathematical solution for the angles at which a ship’s sails should be set, had received considerable attention on the Continent. [23] See a paper by Mr. Johns, R.C.N.C., in _Trans. I.N.A._ 1910. [24] Willett: _Memoirs on Naval Architecture_. [25] At the beginning of the eighteenth century the English first rates carried 100 guns. The second rate comprised two classes: (1) a three-decker of 90; (2) a two-decker of 80. Ships of these rates were few in number and very expensive. The bulk of our fleets consisted of third rates: two-deckers of 70 guns in war and 62 in peace time and on foreign stations (_Charnock_). [26] Sir C. Knowles: _Observations on Shipbuilding_. [27] _Letters of Sir Byam Martin_: N.R. Soc. [28] Sir C. Knowles: _Observations on Shipbuilding_. [29] In 1784 Thomas Gordon published a treatise entitled _Principles of Naval Architecture_, drawing attention to the work of the French scientists and advocating increased length and breadth, finer lines, and a more systematic disposition of materials, for improving the strength and seaworthiness of our royal ships. No notice was taken of his communications to Lord Sandwich, but there is no evidence that his predicted fate overtook him: “to be traduced as an innovator theorist, and visionary projector, as has been the fate of most authors of useful discoveries in modern times, particularly in Britain.” “The bigotry of old practice,” recorded Mr. Willett in 1793, “opposes everything that looks like innovation.” [30] Fincham says their armament was established as, thirty 32-pounders on the lower deck, thirty 24-pounders on the middle deck, thirty-two 18-pounders on the upper deck, and on the quarter-deck and forecastle eighteen 12-pounders. [31] James: _Naval History_. [32] _Letters of Sir Byam Martin_: N.R. Soc. [33] Sharp: _Memoirs of Rear-Admiral Sir W. Symonds_. [34] Hannay: _Ships and Men_. This formula was known before, for Bushnell mentions it in his _Compleat Shipwright_ of 1678. [35] Sharp: _Memoirs of Admiral Sir W. Symonds_. [36] E. J. Reed: _On the Modifications to Ships of the Royal Navy_. [37] _Ibid._ [38] Lieut.-Col. H. W. L. Hime: _The Origin of Artillery_. [39] In the _Histoire d’Artillerie_ of MM. Reinaud and Favé long excerpts from Bacon are examined, from which it appears that he suggested the use of gunpowder in military operations. Gibbon says: “That extraordinary man, Friar Bacon, reveals two of the ingredients, saltpetre and sulphur, and conceals the third in a sentence of mysterious gibberish, as if he dreaded the consequences of his own discovery.” [40] Lieut. H. Brackenbury, R.A.: _Ancient Cannon in Europe_. Vol. IV and V of Proc. R.A.I. [41] Schmidt: _Armes à feu portatives_. [42] Sir Harry Nicolas, in his _History of the Royal Navy_, attributes the documents to the reign of Edward III: an error of more than seventy years. The mistake is exposed by a writer in Vol. XXVI of _The English Historical Review_, in an article on “Firearms in England in the Fourteenth Century.” The writer also gives the English records relating to the use of firearms at Cressy. [43] Brackenbury. [44] The secrecy of the early writers of Italy on gunnery and kindred subjects has been remarked on by Maurice Cockle in his _Bibliography of Military Books_. He attributes it to two motives: fear that the Infidel (the Turk) might profit by the knowledge otherwise gained, and a desire to keep the secrets of the craft in the hands of their countrymen, whose knowledge and assistance the foreigner would then be forced to purchase. [45] _The Great Cannon of Muhammad II_: Brig.-Gen. J. H. Lefroy, R.A., F.R.S. Vol. VI of Proc. R.A.I. [46] Ascribing the deliverance of Constantinople from the Saracens in the two sieges of A.D. 668 and 716 to the novelty, the terrors, and the real efficacy of Greek fire, Gibbon says: “The important secret of compounding and directing this artificial flame was imparted by Callinicus, a native of Heliopolis in Syria, who deserted from the service of the caliph to that of the emperor. The skill of a chemist and engineer was equivalent to the succour of fleets and armies.” For the story of the manner in which its mystery was guarded at Constantinople, of its theft by the Infidel, and of the use he made of it against the Christian chivalry at the crusades, see Chapter LII of _The Decline and Fall of the Roman Empire_. [47] Grose: _Military Antiquities_. [48] Hayley’s MSS.: quoted by M. A. Lower. [49] Oppenheim. [50] Oppenheim. [51] Corned powder was graded in France in the year 1540 into three sizes by means of sieves which varied with the types of guns for which they were intended (see Hime: _Origin of Artillery_). By the end of the century the manufacture had evidently improved in this country. “Some do make excellent good corn powder, so fine, that the corns thereof are like thime seed,” wrote Thos. Smith in his _Art of Gunnery_, A.D. 1600. [52] Oppenheim. [53] Bourne: _The Art of Shooting in Great Ordnance_, 1587. [54] Sir J. K. Laughton: _Armada Papers, N.R.S._ [55] Smith demolished, to his own satisfaction, a theory current that some molecular movement of the metal took place at the moment of gunfire. “I asked the opinion of a soldier, who for a trespass committed was enjoined to ride the canon, who confidently affirmed, he could perceive no quivering of the metal of the piece, but that the air which issued out of the mouth and touch-hole of the piece did somewhat astonish and shake him.” [56] The advantages of large calibres had been appreciated in the previous century. Sir Richard Hawkins, in his _Observations_, printed in 1593, compares the armament of his own ships with that of his Spanish opponents, and says: “Although their artillery were larger, weightier, and many more than ours, and in truth did pierce with greater violence; yet ours being of greater bore, and carrying a weightier and greater shot, was of more importance and of better effect for sinking and spoiling.” [57] Oppenheim. [58] A significant view of the attitude of these professionals toward any innovation in gunnery material is afforded by the entry of Mr. Pepys in his diary for the 17th April, 1669. [59] An anonymous writer in the _Pall Mall Gazette_. [60] Le Sieur Malthus, gentil-homme Anglois, Commissaire Général des Feux et Artifices de l’Artillerie de France, Capitaine General des Sappes et Mines d’icelle & Ingeniéur és Armées du Roy, published his _Pratique de la Guerre_ in 1668. This notable but almost-forgotten artillerist introduced the use of mortars and bombs into France, in