HotFreeBooks.com
Scientific American Supplement, No. 586, March 26, 1887
Author: Various
1  2  3     Next Part
Home - Random Browse

SCIENTIFIC AMERICAN SUPPLEMENT NO. 586



NEW YORK, MARCH 26, 1887

Scientific American Supplement. Vol. XXIII, No. 586.

Scientific American established 1845

Scientific American Supplement, $5 a year.

Scientific American and Supplement, $7 a year.

* * * * *

TABLE OF CONTENTS.

I. BIOGRAPHY.—George W. Whistler, C.E.—By Professor G.L. VOSE.—Full biography of the eminent railroad engineer.

II. CHEMISTRY.—A Newly Discovered Substance in Urine.—A substance possessing greater reducing power than grape sugar found in diabetic urine.

On Electro Dissolution and its Use as Regards Analysis.—By H. N. WARREN, research analyst.—Interesting decomposition of cast iron with production of boron and silicon; experiments with other metals.

III. ELECTRICITY.—No Electricity from the Condensation of Vapor.—Note on Herr S. Kalischer's conclusions.

On Nickel Plating.—By THOMAS T.P. BRUCE WARREN.—Notes on this industry, and suggested improvement for procuring a bright coat.

The Electro-Magnetic Telephone Transmitter.—New theory of the telephone's action.

IV. ENGINEERING.—Fuel and Smoke.—By Prof. OLIVER LODGE.—The second and concluding one of these important lectures.

Gas Engine for Use on Railroads.—The application of six horse power Koerting gas engine to a dummy locomotive.—1 illustration.

New Gas Holder at Erdberg.—The largest gas holder out of England.—3 illustrations.

Tar for Firing Retorts.—Simple arrangement adapted for use in ordinary gas retort benches; results attained.—1 illustration.

The Anti-Friction Conveyer.—An improvement on the screw of Archimedes; an apparatus of wonderful simplicity and efficacy in the moving of grain.—2 illustrations.

The Retiro Viaduct.—Combined iron and stone viaduct over the river Retiro, Brazil.—5 illustrations.

Western North Carolina Location over the Blue Ridge.—Interesting instance of railroad topography.—1 illustration.

V. METALLURGY.—Chilled Cast Iron.—The various uses of this product; adaptability of American iron for its application.

VI. MISCELLANEOUS.—Coal in the Argentine Republic.—Note.

History of the World's Postal Service.—Conclusion of this interesting article.—The service in Germany, China. Russia, and elsewhere.—10 illustrations.

Snow Hall—The new science and natural history building of the University of Kansas.

VII. NAVAL ENGINEERING.—Improvement in Laying Out Frames of Vessels.—The Frame Placer.—By GUSTAVE SONNENBURG.—Ingenious apparatus for use in ship yards.—1 illustration.

Sea-going Torpedo Boats.—The inutility of small torpedo boats at sea.—The construction of larger ones discussed.

VIII. ORDNANCE.—Firing Trial of the 1101/2 Ton B.L. Elswick Gun. Full dimensions of this piece and it projectiles.—Results of proof firing.—9 illustrations.

IX. PHOTOGRAPHY.—Experiments in Toning Gelatino-Chloride Paper.—Trials of ten different gold toning baths, formulas, and results.

Printing Lantern Pictures by Artificial Light on Bromide Plates from Various Sizes.—By A. PUMPHREY.—The processor producing smaller or larger transparencies from negatives.—1 illustration.

X. PHYSICS.—A New Mercury Pump.—Simple air pump for high vacua.—1 illustration.

The Laws of the Absorption of Light in Crystals.—By H. BECQUEREL.

Varying Cylindrical Lens.—By TEMPEST ANDERSON, M.D., B. Sc.—Combination of two conoidal lenses.—Range of power obtained.

XI. PHYSIOLOGY.—Elimination of Poisons.—Treatment of poison cases by establishment of a strong diuresis. The Filtration and the Secretion Theories.—Experiments on the action of and secretions of the kidneys.

XII. TECHNOLOGY.—Furnace for Decomposing Chloride of Magnesium.—Furnace with rotating chamber for use by alkali manufacturers.—1 illustration.

Notes on Garment Dyeing.—The production of blacks on silk and wool.—Formulas for mordants.

Studies in Pyrotechny.—II. Methods of Illumination.—Continuation of this valuable treatise.—9 illustrations.

The "Sensim" Preparing Box.—New machine for treatment of fiber.—An improvement on the ordinary gill box.—3 illustrations.

* * * * *



THE RETIRO VIADUCT.

We give engravings of the viaduct over the river Retiro, Brazil, our illustrations being reproduced by permission from the Proceedings of the Institution of Civil Engineers. In a "selected paper" contributed to the volume of these proceedings just published, Mr. Jorge Rademaker Grunewald, Memb. Inst. C.E., describes the work as follows:



This viaduct was constructed in the year 1875, according to designs furnished by the author, for the purpose of passing the Dom Pedro Segundo State Railway over the valley which forms the bed of the river Retiro, a small confluent on the left bank of the river Parahybuna. It is 265 kilometers (165 miles) from Rio de Janeiro, and about 10 kilometers (6.4 miles) from the city of Juiz de Fora, in the province of Minas Geraes, Brazil. It has a curve of 382 meters (1,253 ft.) radius, and a gradient of 1 in 83.3. Its total length is 109 meters (357 ft. 7 in.); width between handrails, 4 meters (13 ft.); and greatest height above the bed of the river, 20 meters (65 ft. 7 in.).

The viaduct is composed of seven semicircular arches, each end arch being built of ashlar masonry, and of 6 meters (19 ft. 8 in.) diameter; five intermediate arches, 15 meters (49 ft. 2 in.) in diameter, are of iron. The four central piers are of iron erected on pillars of ashlar masonry. The metallic part of this viaduct is 80 meters (262 ft. 6 in.) long, and is constructed in the following manner: The arches, and the longitudinal girders which they support, are made of two Barlow rails riveted together, with an iron plate 1/2 inch thick placed between them. The spandrels are formed of uprights and diagonals, the former being made of four angle-irons, and the latter of one angle-iron. Each pair of arches, longitudinal girders and uprights, is transversely 3 meters (9 ft. 10 in.) from center to center, and is connected by cross and diagonal bracing. On the top of the longitudinal girders are fixed cross pieces of single Barlow rails, upon which again are fastened two longitudinals of wood 12 in. square in section, and which in their turn carry the rails of the permanent way.

The gauge of the Dom Pedro Segundo Railway is 1.60 meters, or 5 ft. 3 in. nearly, between the rails. At each end of the transverse Barlow rails is fixed the customary simple iron handrail, carried by light cast-iron standards. The iron piers are each formed of four columns, and the columns consist of two Barlow rails, with a slotted iron plate 1/2 inch thick let in between the rails, and the whole being riveted together connects each pair of side columns.

The details show the system of cross and diagonal bracing. The columns are each supported by four buttresses formed of plates and angle-irons. These buttresses, fastened with bolts 8 ft. 3 in. long, let into the masonry pillars, secure the stability of the viaduct against lateral strains, due mostly to the centrifugal force caused by the passage of the trains.

The Barlow rails, which constitute the peculiarity of the structure, are from those taken up from the permanent way when the Vignoles pattern of rail was adopted on this railway. The whole of the foundations were built without difficulty. The principal parts of the iron work were calculated to resist the strains resulting from a weight of 4 tons 8 cwt. per lineal meter traveling over the viaduct at a velocity of 60 kilometers, or about 37 miles, per hour.

In spite of its fragile appearance this viaduct has, up to the present time, served in a most satisfactory manner the purpose for which it was built.—Engineering.

* * * * *



SEA-GOING TORPEDO BOATS.

All investigations of the sea-going qualities of torpedo boats show that while the basin experiments are highly satisfactory, those made at sea prove with equal force the unreliability of these craft when they leave the coast. At the beginning of the Milford Haven operations, the boisterous weather necessitated the postponing of operations, on account of the unfitness of the torpedo boat crews to continue work after the twelve hours of serious fatigue they had already undergone. In the French evolutions, the difficulties of the passage from Bastia to Ajaccio, although not remarkably severe, so unfitted fifteen of the twenty boats that they could take no part in the final attack. In two nights we find recorded collisions which disable boats Nos. 52, 61, 63, and 72, and required their return to port for repairs.

Of the twenty-two torpedo boats leaving Toulon a few days before, but six arrived near the enemy, although their commanders displayed admirable energy. One had run aground, and was full of water; another had been sunk by collision; another's engine was seriously injured; and as for the rest, they could not follow.

Of the boats under the command of Admiral Brown de Colstoun, but five remained for service, for the sixth received an accident to her machinery which prevented her taking part in the attack.

During the operations off the Balearic Isles, only one of six boats attacked, and none was able to follow the armorclads, all meeting with circumstances quite unexpected and embarrassing.

With the weather as it existed May 13, the armorclads had the torpedo fleet completely at their mercy, for even if they had not been destroyed by the excellent practice of the Hotchkiss gunners, they would have been of no use, as they could not with safety discharge their torpedoes. In fact, the search lights discovered distinctly that one of the boats, which burned her Coston's signal to announce victory, did not have her torpedo tube open, on account of the heavy sea.

Furthermore, their positions were frequently easily discovered by the immense volume of smoke and flame ejected while going at great speed. This applies as well by night as by day. It was also reported that after the four days' running the speed of the boats was reduced to twelve knots.

With such evidence before us, the seaworthiness of boats of the Nos. 63 and 64 type may be seriously questioned. Weyl emphasizes the facts that "practice has shown that boats of No. 61 type cannot make headway in a heavy sea, and that it is then often impossible to open their torpedo tubes. On this account they are greatly inferior to ships of moderate tonnage, which can certainly make some progress, fire their torpedoes, and use their artillery in weather when a torpedo boat will be utterly helpless. The torpedo boat abandoned to itself has a very limited field of action."

Du Pin de Saint Andre admits the success of the torpedo boat for harbor and coast work, but wisely concludes that this can prove nothing as to what they may or may not be able to do at sea.

In an article which appeared in the Revue des Deux Mondes in June last, he presented able reasons why the torpedo boats of to-day's type, being destitute of most, if not all, of the requisites of sea-going craft, cannot go to sea, take care of themselves, and remain there prepared to attack an enemy wherever he may be found. Invisibility to an enemy may facilitate attack, but it has to be dearly paid for in diminished safety. Further, the life that must be led in such vessels in time of war would very quickly unfit men for their hazardous duties.

He points out that the effect of such a life upon the bodies and minds of the officers and crew would be most disastrous. The want of exercise alone would be sufficient to unfit them for the demands that service would make upon them. He has intelligently depicted the consequences of such a life, and his reasoning has been indorsed by the reports of French officers who have had experience in the boats in question.

No weapon, no matter how ingenious, is of utility in warfare unless it can be relied upon, and no vessel that is not tenantable can be expected to render any service at sea.

From the evidence before us, we must conclude that the type of torpedo boat under discussion is capable of making sea passages, provided it can communicate frequently with its supply stations and secure the bodily rest so necessary to its crew. But even in a moderate sea it is useless for attack, and in the majority of cases will not be able even to open its impulse tubes. Should it succeed in doing this, the rolling and yawing will render its aim very uncertain.

An experiment conducted against the Richelieu in October last, at Toulon, before Admiral O'Neil, the director-general of the torpedo service, has added its testimony to the uncertainty of the Whitehead torpedo. The Richelieu had been fitted with Bullivant nets, and the trial was made to learn what protection they would afford.

The weather was fair, the sea moderate, and the conditions generally favorable to the torpedo; but the Whitehead missed its mark, although the Richelieu's speed was only three knots. Running at full speed, the torpedo boat, even in this moderate sea, deemed it prudent to keep the launching tube closed, and selected a range of 250 yards for opening it and firing. Just at the moment of discharge a little sea came on board, the boat yawed, the torpedo aim was changed more than 30 deg., and it passed astern without touching its object.

While the Milford Haven operations have taught some valuable lessons, they were conducted under but few of the conditions that are most likely to occur in actual warfare; and had the defense been carried on with an organization and command equal to that of the attack, the Navy's triumph would, perhaps, not have been so easily secured, and the results might have been very different.

May not the apparent deficiencies of the defense have been due to the fact that soldiers instead of sailors are given the control of the harbor and coast defense? Is this right? Ought they not to be organized on a naval basis? This is no new suggestion, but its importance needs emphasis.

These operations, however, convinced at least one deeply interested spectator, Lord Brassey, to the extent of calling attention "to the urgent necessity for the construction of a class of torpedo vessels capable of keeping the sea in company with an armored fleet."

There is no one in Great Britain who takes a greater interest in the progress of the British Navy than Lord Brassey, and we take pleasure in quoting from his letter of August 23 last to the Times, in which he expressed the following opinion: "The torpedo boats ordered last year from Messrs. Thornycroft and Yarrow are excellent in their class. But their dimensions are not sufficient for sea-going vessels. We must accept a tonnage of not less than 300 tons in order to secure thorough seaworthiness and sufficient coal endurance.

"A beginning has been made in the construction of vessels of the type required. To multiply them with no stinting hand is the paramount question of the day in the department of construction. The boats attached to the Channel fleet at Milford Haven will be most valuable for harbor defense, and for that purpose they are greatly needed. Torpedo boat catchers are not less essential to the efficiency of a fleet. The gunboats attached to the Channel fleet were built for service in the rivers of China. They should be reserved for the work for which they were designed.

"We require for the fleet more fast gunboats of the Curlew and Landrail type. I trust that the next estimates for the Navy will contain an ample provision for building gun vessels of high speed."

As torpedoes must be carried, the next point to which we would call the attention of our readers is the very rapid progress that has been made in the boats designed to carry automatic torpedoes.

A very few years ago the names of Thornycroft and Yarrow were almost alone as builders of a special type of vessel to carry them. To-day, in addition, we have Schichau, White, Herreshoff, Creusot, Thomson, and others, forming a competitive body of high speed torpedo-boat builders who are daily making new and rapid development—almost too rapid, in fact, for the military student to follow.

As new types are designed, additional speed gained, or increased seaworthiness attained, public descriptions quickly follow, and we have ourselves recorded the various advances made so fully that it will be unnecessary to enter into details here.

As late as October, 1885, an able writer said: "The two most celebrated builders of torpedo boats in the world are Thornycroft and Yarrow, in England. Each is capable of producing a first class torpedo boat, from 100 ft. to 130 ft. long, and with 10 ft. to 14 ft. beam, that will steam at the rate of from 18 knots to 22 knots per hour for 370 knots, or at the rate of 10 knots per hour for 3000 miles. A second class torpedo boat is from 40 ft. to 60 ft. long, and with 6 ft. or 8 ft. beam.

The use of these boats is gradually being abandoned in Europe except for use from sea-going ships; but in Europe the harbors are very small, and it has been found that practically every torpedo boat for coast defense must be able to go to sea. The tendency is, therefore, to confinement to the first class boats."

In a paper on "Naval Torpedo Warfare," prepared in January, 1886, for a special committee of the American Senate, by Lieutenant Jaques of the American Navy, we find the following reference to the progress in torpedo boat construction: "The development in torpedo boats has been phenomenal, the last year alone showing an advance from a length of 120 ft. and a speed of 19 knots, which were considered remarkable qualities in a first class boat, to a length of 140 ft. and a speed of 23 knots loaded (carrying 15 tons), and 25 knots light, together with the introduction of novel features of importance.

"Although Messrs. Yarrow and Thornycroft have brought the second class boats to a very high standard in Europe, I believe they will soon be abandoned there even for sea-going ships (very few are now laid down), and that the great development will be in overcoming the disadvantages of delicacy and weakness by increasing their size, giving them greater maneuvering power and safety by the introduction of two engines and twin screws, and steel plate and coal protection against rapid firing ammunition. Yarrow and Co. have already laid down some boats of this character that give promise of developing a speed of from 23 to 25 knots."

In the Russian boat recently built at Glasgow, progress in this direction is also seen in the 148 ft. length, 17 ft. beam, the maneuvering powers and safety element of the twin screws. But while the boat is fitted for the 19 ft. torpedo, a weapon of increased range and heavier explosive charge, it suffers from the impossibility of broadside fire and the disadvantages that Gallwey has named: "The great length of this torpedo, however, makes it a very unhandy weapon for a boat, besides which its extra weight limits the number which can be carried."

While perhaps Messrs. Thomson have been the first to show the performance of a twin screw torpedo boat in England, the one completed in June last by Yarrow for the Japanese government recalls the intelligence that Japan has exercised in the selection of types.

Commencing as far back as nine years ago, the Japanese were probably the first to introduce sea-going boats, and they have been the first power to initiate the armor type, one of which was shipped last summer to be put together in Japan. As before stated, it was built by Messrs. Yarrow and Co., was 166 ft. long, 19 ft. beam, with twin screws, 1 in. steel armor, double engines, with bow and broadside torpedo guns, the latter so arranged as to greatly increase their efficiency.

While the advances are not restricted to the English builders, a glance at the points to which Thornycroft and Yarrow have brought their improvements up to the present time will indicate that their achievements are not only equal to but greater than those of any other builders.

The former has boats under construction 148 ft. long, 15 ft. beam, to make 420 revolutions with 130 lb. of steam, the guaranteed speed being 23 knots on a continuous run of two hours' duration, with a load of 15 tons. They will have triple-expansion or compound direct-acting surface-condensing engines and twin screws, Thornycroft's patent tubular boilers, double rudders, electric search lights, three masts and sails.

While the armaments of the various boats differ, Thornycroft is prepared to fit the launching tubes with either air or powder impulse, to mount the tubes forward or on deck, and also the fittings for machine and rapid firing guns.

Yarrow and Co. have contracted for boats varying in length from 117 ft. to 166 ft., with fittings and armament as may be required. They have obtained excellent results in their last English boat of the Admiralty type. They are, in fact, prepared to guarantee a speed of 23 knots in a length of 125 ft. and 25 knots in a length of 140 ft., carrying in both causes a mean load corresponding to fuel and armament of 10 tons.

And so the progress goes on, but it will not stop here; it has already incited a marked development in ship construction, and the endeavors to withstand torpedo attack have improved the defense against gun fire also.

In quoting a German opinion on the development of the Russian torpedo fleet, Charmes refers to the type which will, no doubt, be most successful upon the sea, namely, the torpedo cruisers, and it is to this type, more than for any other, that we may expect torpedo boats to be adapted. Already, writers have dropped the phrase "torpedo boats" for "torpedo vessels."—Engineering.

* * * * *



FIRING TRIAL OF THE 1101/2 TON B.L. ELSWICK GUN.

The firing trial of the first new 1101/2 ton breech loading gun approved for H.M.'s ships Benbow, Renown, and Sanspareil was commenced recently at the Woolwich proof butts, under the direction of Colonel Maitland, the superintendent of the Royal Gun Factories. We give herewith a section showing the construction of this gun (vide Fig. 8). It very nearly corresponds to the section given of it when designed in 1884, in a paper read by Colonel Maitland at the United Service Institution, of which we gave a long account in the Engineer of June 27, 1884.

The following figures are authoritative: Length over all, 524 in.; length of bore, 487.5 in. (30 calibers). The breech engages in the breech piece, leaving the A tube with its full strength for tangential strain (vide Fig.). The A tube is in a single piece instead of two lengths, as in the case of the Italia guns. It is supplied to Elswick from Whitworth's works, one of the few in England where such a tube could be made. There are four layers of metal hoops over the breech. Copper and bronze are used to give longitudinal strength. The obturation is a modification of the De Bange system, proposed by Vavasseur.



The maximum firing charge is 900 lb. of cocoa powder. The projectile weighs 1,800 lb. The estimated muzzle velocity is 2,216 ft. per second. The capacity of the chamber is 28,610 cubic inches, and that of the bore 112,595 cubic inches. The estimated total energy is 61,200 ft. tons. It will be a few days probably before the full powers of the gun are tested, but the above are confidently expected to be attained, judging from the results with the 100 ton guns supplied to Italy. On January 7 last we gave those of the new Krupp 119 ton gun. It had fired a projectile with a velocity of almost 1,900 ft. with a charge of less than 864.67 lb., with moderate pressure. The estimated maximum for this gun was a velocity of 2,017 ft. with a projectile weighing 1,632 lb., giving a total energy of 46,061 ft. tons, or 13,000 ft. tons less than the Elswick gun, comparing the estimated results.

The proof of the Elswick gun is mounted on a carriage turned out by the Royal Carriage Department, under Colonel Close. This carriage is made on bogies so as to run on rails passing easily round curves of 50 ft. radius. The gun is fired on an inclined length of rails, the recoil presses of the carriage first receiving the shock and reducing the recoil. The carriage is made to lift into the government barge, so as to go easily to Shoeburyness or elsewhere. It can be altered so as to provide for turning, and it allows the piece to be fired at angles of elevation up to 24 deg. The cheeks of the carriage are made to open and close, so as to take the 12 in. gun and larger pieces. The steel castings for it are supplied from the Stanners Close Steel Works.



The first round was fired at about noon. The charge was only 598 lb., consisting of four charges of 112 lb. and one of 130 lb. of Waltham Abbey brown prism No. 1 powder. The proof shot weighs, like the service projectile, 1,800 lb. Thus fired, the gun recoiled nearly 4 ft. on the press, and the carriage ran back on the rails about 50 ft. The projectile had a velocity of 1,685 ft. per second, and entered about 52 ft. into the butt. We cannot yet give the pressure, but unquestionably it was a low one. The charges as the firing continues will be increased in successive rounds up to the full 900 lb. charge.

Figs. 1 and 2 show the mounting of the 1101/2 ton gun in the barbette towers of the Benbow. The gun is held down on the bed by steel bands and recoils in its bed on the slide (vide Fig. 2). The latter is hinged or pivoted in front and is elevated by elevating ram, shown in Fig. 2. When the slide is fully down, the gun is in the loading position. The ammunition lift brings up the projectile and charge, which latter is subdivided, like those employed in the German guns, in succession to the breech, the hydraulic rammer forcing them home.



The simplicity of the arrangement is apparent. The recoil always acts parallel to the slide. This is much better than allowing its direction to be affected by elevation, and the distributed hold of the steel bands is preferable to the single attachment at trunnions. Theoretically, the recoil is not so perfectly met as in some of the earlier Elswick designs, in which the presses were brought opposite to the trunnions, so that they acted symmetrically on each side of the center of resistance. The barbette tower is covered by a steel plate, shown in Fig. 1, fitting close to the gun slide, so that the only opening is that behind the breech when the gun is in the forward position, and this is closed as it recoils.

The only man of the detachment even partly exposed is the number one, while laying the gun, and in that position he is nearly covered by the gun and fittings. Common shell, shrapnel shell, and steel armor-piercing projectiles, have been approved for the 1101/2 ton gun. The common shell is shown in Fig. 3. Like the common shell for all the larger natures of new type guns, it is made of steel. It has been found necessary to support the core used in casting these projectiles at both ends. Consequently, there is a screw plug at the base as well as at the apex. The hole at the base is used as a filling hole for the insertion of the bursting charge, which consists of 179 lb. of powder, the total weight of the filled shell being 1,800 lb.



The apex has a screw plug of larger diameter than that of the fuse. This is shown in Fig. 4. The fuse is a direct action one. The needle, B, is held in the center of a copper disk, C C, and is safe against explosion until it is actually brought into contact with an object, when it is forced down, igniting a patch of cap composition and the magazine at A, and so firing the bursting charge of the shell below. E E E are each priming charges of seven grains of pistol powder, made up in shalloon bags to insure the ignition of the bursting charge, which is in a bag of serge and shalloon beneath.

The use of this fuse involves the curious question of the physical conditions now existing in the discharge of our projectiles by slow burning powder. The forward movement of the shell is now so gradual that the inertia of a pellet is only sufficient to shear a wire of one-tenth the strength of that which might formerly have been sheared by a similar pellet in an old type gun with quick burning powder. Consequently, in many cases, it is found better not to depend on a suspending wire thus sheared, but to adopt direct action. The fuse in question would, we believe, act even on graze, at any angle over 10 deg.. Probably at less angles than 10 deg. it would not explode against water, which would be an advantage in firing at ships.

Shells so gently put in motion, and having no windage, might be made, it might naturally be supposed, singularly thin, and the adoption of steel in place of iron calls for some explanation. The reason is that it has been found that common shells break up against masonry, instead of penetrating it, when fired from these large high velocity guns.

The shrapnel shell is shown at Fig. 5. Like the common shell, it is made of steel, and is of the general form of the pattern of General Boxer, with wooden head, central tube, and bursting charge in the base. It contains 2,300 four ounce sand shots and an 8 lb. bursting charge. It weighs 1,800 lb. The fuse is time and percussion. It is shown in Figs. 6 and 6A. It closely resembles the original Armstrong time and percussion pattern.



The action is as follows: The ignition pellet, A, which is ordinarily held by a safety pin, is, after the withdrawal of the latter, only held by a fine, suspending wire, which is sheared by the inertia of the pellet on discharge, a needle lighting a percussion patch of composition and the composition ring, B B, which burns round at a given rate until it reaches the communication passage, C, when it flashes through the percussion pellet, E, and ignites the magazine, D, and so ignites the primer shown in Fig. 6, flashes down the central tube of the shell, and explodes the bursting charge in the base, Fig. 5. The length of time during which the fuse burns depends on how far the composition ring is turned round, and what length it consequently has to burn before it reaches the communication passage, C. If the fuse should be set too long, or from any other cause the shell strikes before the fuse fires the charge, the percussion action fires the shell on graze by the following arrangement: The heavy metal piece containing the magazine, D, constitutes a striker, which is held in place by a plain ball, G, near the axis of the fuse and by a safety pellet, H. On first movement in the gun, this latter by inertia shears a suspending wire and leaves the ball free to escape above it, which it does by centrifugal force, leaving the magazine striker, D, free to fire itself by momentum on the needle shown above it, on impact. There is a second safety arrangement, not shown in the figure, consisting of a cross pin, held by a weak spiral spring, which is compressed by centrifugal force during flight, leaving the magazine pellet free to act, as above described, on impact.

The armor-piercing projectile is shown in Fig. 7. It is to be made of forged steel, and supplied by Elswick. In appearance it very closely resembles those fired from the 100 ton gun at Spezia, but if it is made on the Firmini system, it will differ from it in the composition of its metal, inasmuch as it will contain a large proportion of chromium, probably from 1 to 2 per cent., whereas an analysis of Krupp's shell gives none. In fact, as Krupp's agent at Spezia predicted, the analysis is less instructive than we could wish.—The Engineer.

* * * * *



GAS ENGINE FOR USE ON RAILROADS.

The industrial world has reason to feel considerable interest in any economical method of traction on railways, owing to the influence which cost of transportation has upon the price of produce. We give a description of the gas engine invented by Mr. Emmanuel Stevens. Many experiments have been made both at Berlin and Liege during the past few years. They all failed, owing to the impossibility the builders encountered in securing sufficient speed.

The Stevens engine does not present this defect, as will be seen. It has the appearance of an ordinary street car entirely inclosed, showing none of the machinery from without. On the interior is a Koerting gas motor of six horse power, which is a sufficiently well known type not to require a description. In the experiment which we saw, the motor was supplied with a mixture of gas and air, obtained by the evaporation of naphtha. On the shaft of the motor are fixed two pulleys of different sizes, which give the engine two rates of speed, one of three miles and the other of 81/2 miles an hour. Between these two pulleys is a friction socket, by which either rate of speed may be secured.

The power is transmitted from one of the pulleys by a rubber belt to an intermediate shaft, which carries a toothed wheel that transmits the power to the axle by means of an endless chain. On this axle are three conical gear wheels, two of which are furnished with hooked teeth, and the third with wooden projections and fixed permanently in place. This arrangement enables the engine to be moved forward or backward according as it is thrown in right or left gear. When the conical pinions are thrown out of gear, the motive force is no longer applied to the axle, and by the aid of the brakes the engine may be instantly stopped. The movement of the pinions is effected by two sets of wheels on each of the platforms of the engine, and near the door for the conductor. By turning one of the wheels to the right or left on either platform, the conductor imparts either the less or the greater speed to the engine. In case he has caused the engine to move forward by turning the second wheel, he will not have to touch it again until the end of the trip. The brake, which is also operated from the two platforms, is applied to all four wheels at the same time. From this arrangement it is seen that the movement is continuous. Nevertheless, the conductor has access to the regulator by a small chain connected with the outside by a wheel near at hand, but the action is sufficiently regular not to require much attention to this feature.



The gas is produced by the Wilford apparatus, which regularly furnishes the requisite quantity necessary for an explosion, which is produced by a particular kind of light placed near the piston. The vapor is produced by passing hot water from the envelope of the cylinder of the motor through the Wilford apparatus. The water is cooled again in a reservoir (system Koerting) placed in direct communication with the cylinder. Any permanent heating is therefore impossible.

The noise of the explosions is prevented by a device invented by Mr. Stevens himself. It consists of a drum covered with asbestos or any other material which absorbs noise.

According to the inventor, the saving over the use of horses for traction is considerable. This system is soon to be tried practically at Antwerp in Belgium, and then it will be possible to arrive at the actual cost of traction.—Industrie Moderne, Brussels.

* * * * *



WESTERN NORTH CAROLINA LOCATION OVER THE BLUE RIDGE.



The interesting piece of railroad location illustrated in this issue is on the mountain section of the Western North Carolina Railroad. This section crosses the Blue Ridge Mountains 18 miles east of Asheville, at a point known as Swannanoa Gap, 2,660 feet above tide water. The part of the road shown on the accompanying cut is 10 miles in length and has an elevation of 1,190 feet; to overcome the actual distance by the old State pike was somewhat over 3 miles. The maximum curvature as first located was 10 deg., but for economy of time as well as money this was exceeded in a few instances as the work progressed, but is now being by degrees reduced. The maximum grades on tangents are 116 feet per mile; on curves the grade is equated one-tenth to a degree. The masonry is of the most substantial kind, granite viaducts and arch culverts. The numbers and lengths of tunnels as indicated by letters on cut are as follows:

Ft. in all of these.

A. Point Tunnel. 216 ft. long.[1] B. Jarrett's " 125 " " C. Lick Log " 562 " " D. McElroy " 89 " " E. High Ridge " 415 " " F. Burgin " 202 " " G. Swannanoa " 1,800 " "

[Footnote 1: For the sake of economy of space, our cut omits the Point and Swannanoa tunnels (the latter is the summit tunnel), but covers all of the location which is of interest to engineers, the remainder at the Swannanoa end being almost "on tangent" to and through the summit.]

The work was done by the State of North Carolina with convict labor, under the direction of Mr. Jas. A. Wilson, as president and chief engineer, but was sold by the State to the Richmond & Danville system.—Railroad Gazette.

* * * * *



NEW GASHOLDER AT ERDBERG.

The new gasholder which has been erected by Messrs. C. and W. Walker for the Imperial Continental Gas Company at Erdberg, near Vienna, has been graphically described by Herr E.R. Leonhardt in a paper which he read before the Austrian Society of Engineers. The enormous dimensions and elegant construction of the holder—being the largest out of England—as well as the work of putting up the new gasholder, are of special interest to English engineers, as Erdberg contains the largest and best appointed works in Austria. The dimensions of the holder are—inner lift, 195 feet diameter, 40 feet deep; middle lift, 1971/2 feet diameter, 40 feet deep; outer lift, 200 feet diameter, 40 feet deep. The diameter over all is about 230 feet. The impression produced upon the members of the Austrian Society by their visit to Erdberg was altogether most favorable; and not only did the inspection of the large gasholder justify every expectation, but the visitors were convinced that all the buildings were in excellent condition and well adapted for their purpose, that the machinery was of the latest and most approved type, and that the management was in experienced hands.

THE NEW GASHOLDER

is contained in a building consisting of a circular wall covered with a wrought iron roof. The holder itself is telescopic, and is capable of holding 31/2 million cubic feet of gas. The accompanying illustrations (Figs. 1 and 3) are a sectional elevation of the holder and its house and a sectional plan of the roof and holder crown. Having a capacity of close upon 3,200,000 Austrian cubic feet, this gasholder is the largest of its kind on the Continent, and is surpassed in size by only a few in England and America. By way of comparison, Hamburg possesses a holder of 50,000 cubic meters (1,765,000 cubic feet) capacity; and there is one in Berlin which is expected to hold 75,000 cubic meters (2,647,500 cubic feet) of gas.

GASHOLDER HOUSE.

The gasholder house at Erdberg is perfectly circular, and has an internal diameter of 63.410 meters. It is constructed, in three stories, with forty piers projecting on the outside, and with four rows of windows between the piers—one in each of the top and bottom stories, and two rows in the middle. These windows have a height of 1.40 meters in the lowest circle, where the wall is 1.40 meters thick, and of 2.90 meters in the two top stories, where it is respectively 1.11 meters and 0.90 meter thick. The top edge of the wall is 35.35 meters above the base of the building, and 44.39 meters from the bottom of the tank; the piers rising 1.60 meters beyond the top of the wall. The highest point of the lantern on the roof will thus be 48.95 meters above the ground.

GASHOLDER TANK.

The tank in which the gasholder floats has an internal diameter of 61.57 meters, and therefore a superficial area of 3,000 square meters; and since the coping is 12.31 meters above the floor, it follows that the tank is capable of holding 35,500 cubic meters (7,800,000 gallons) of water. The bottom consists of brickwork 1.10 meters thick, rendered with Portland cement, and resting on a layer of concrete 1 meter thick. The walls are likewise of brick and cement, of a thickness of 3.30 meters up to the ground level, and 2.40 meters thick to the height of 3.44 meters above the surface. Altogether, 2,988,680 kilos. of cement and 5,570,000 bricks were used in its construction. In fact, from the bottom of tank to top of roof, it reaches as high as the monument at London Bridge.



The construction of the tank offered many and serious difficulties. The bottom of the tank is fully 3 meters below the level of the Danube Canal, which passes close by, and it was not until twelve large pulsometer pumps were set up, and worked continually night and day, that it was possible to reach the necessary depth to allow of the commencement of the foundations of the boundary wall.

ROOF OF HOUSE.

The wrought iron cupola-shaped roof of the gasholder house was designed by Herr W. Brenner, and consists of 40 radiating rafters, each weighing about 25 cwt., and joined together by 8 polygonal circles of angle iron (90x90x10 mm.). The highest middle circle is uncovered, and carries a round lantern (Fig. 1). These radiating rafters consist of flat iron bars 7 mm. thick, and of a height which diminishes gradually, from one interval to another on the inside, from 252 to 188 mm. At the outside ends (varying from 80x80x9 mm. in the lowest to 60x60x7 mm. in the last polygon but one) these rafters are strengthened, at least as far as the five lowest ones are concerned, by flat irons tightly riveted on. At their respective places of support, the ends of all the spars are screwed on by means of a washer 250 mm. high and 31 mm. thick, and surmounted by a gutter supported by angle irons. From every junction between the radial rafters and the polygonal circle, diagonal bars are made to run to the center of the corresponding interval, where they meet, and are there firmly held together by means of a tongue ring. The roof is 64.520 meters wide and 14.628 meters high; and its total weight is 103.300 kilos. for the ironwork—representing a weight of 31.6 kilos. per square meter of surface. It is proposed to employ for its covering wooden purlins and tin plates. The whole construction has a light, pleasing, and yet thoroughly solid appearance.

RAISING THE ROOF.

Herr Brenner, the engineer of the Erdberg Works, gives a description of how the roof of a house, 54.6 meters wide, for a gasholder in Berlin, was raised to a height of 22 meters. In that instance the iron structure was put together at the bottom of the tank, leaving the rafter ends and the mural ring. The hoisting itself was effected by means of levers—one to each rafter—connected with the ironwork below by means of iron chains. At the top there were apertures at distances of about 26 mm. from each other, and through these the hoisting was proceeded with. With every lift, the iron structure was raised a distance of 26 mm.



Herr Brenner had considerable hesitation in raising in the same way the structure at Erdberg, which was much larger and heavier than that in Berlin. The simultaneous elevation to 48 meters above the level, proposed to be effected at forty different points, did not appear to him to offer sufficient security. He therefore proposed to put the roof together on the ground, and to raise it simultaneously with the building of the wall; stating that this mode would be perfectly safe, and would not involve any additional cost. The suggestion was adopted, and it was found to possess, in addition, the important advantage that the structure could be made to rest on the masonry at any moment; whereas this had been impossible in the case at the Berlin Gasworks.



HOISTING.

At a given signal from the foreman, two operatives, stationed at each of the forty lifting points, with crowbars inserted in the holes provided for the purpose, give the screws a simultaneous turn in the same direction. The bars are then inserted in another hole higher up. The hoisting screws are connected with the structure of the roof, and rise therewith. All that is requisite for the hoisting from the next cross beam is to give a forward turn to the screws. When the workmen had become accustomed to their task, the hoisting to a distance of 1 meter occupied only about half to three-quarters of an hour. At the outset, and merely by way of a trial, the roof was lifted to a height of fully 2 meters, and left for some time suspended in the air. The eighty men engaged in the operation carry on the work with great regularity and steadiness, obeying the signal of the foreman as soon as it was given.

THE GASHOLDER.

The holder, which was supplied by the well-known firm of Messrs. C. and W. Walker, of Finsbury Circus, London, and Donnington, Salop, was in an outer courtyard. It is a three-lift telescopic one; the lowest lift being 200 feet, the middle lift 197 ft. 6 in., and the top lift 195 ft. in diameter. The height of each lift is 40 feet. The several lifts are raised in the usual way; and they all work in a circle of 24 vertical U-shaped channel irons, fixed in the wall of the house by means of 13 supports placed at equal distances from the base to the summit (as shown in Fig. 2). When the gasholder is perfectly empty, the three lifts are inclosed, one in the other, and rest with their lower edges upon the bottom of the tank. In this case the roof of the top lift rests upon a wooden framework. Fixed in the floor of the tank are 144 posts, 9 inches thick at the bottom and 6 inches thick at the top, to support the crown of the holder in such a way that the tops are fixed in a kind of socket, each of them being provided with four horizontal bars, which decrease in thickness from 305 by 100 mm. to 150 by 50 mm., and represent 16 parallel polygons, which in their turn are fastened diagonally by means of iron rails 63 by 100 mm. thick, arranged crosswise. The top of this framework is perfectly contiguous with the inside of the crown of the gasholder. The crown itself is made up of iron plates, the outer rows having a thickness of 11 mm., decreasing to 5 mm. toward the middle, and to 3 mm. at the top. The plates used for the side sheets of the holder are: For the top and bottom rows, 6.4 mm.; and for the other plates, 2.6 mm.

* * * * *

A new bleaching compound has been discovered, consisting of three parts by measure of mustard-seed oil, four of melted paraffin, three of caustic soda 20 deg. Baume, well mixed to form a soapy compound. Of this one part of weight and two of pure tallow soap are mixed, and of this mixture one ounce for each gallon of water is used for the bleaching bath, and one ounce caustic soda 20 deg. Baume for each gallon is added, when the bath is heated in a close vessel, the goods entered, and boiled till sufficiently bleached.

* * * * *



GEORGE W. WHISTLER, C.E.

[Footnote: A paper by Prof. G.L. Vose, Member of the Boston Society of Civil Engineers. Read September 15, 1886.]

By Prof. G.L. VOSE.

Few persons, even among those best acquainted with our modern railroad system, are aware of the early struggles of the men to whose foresight, energy, and skill the new mode of transportation owes its introduction into this country. The railroad problem in the United States was quite a different one from that in Europe. Had we simply copied the railways of England, we should have ruined the system at the outset, for this country. In England, where the railroad had its origin, money was plenty, the land was densely populated, and the demand for rapid and cheap transportation already existed. A great many short lines connecting the great centers of industry were required, and for the construction of such in the most substantial manner the money was easily obtained. In America, on the contrary, a land of enormous extent, almost entirely undeveloped, but of great possibilities, lines of hundreds and even thousands of miles in extent were to be made, to connect cities as yet unborn, and accommodate a future traffic of which no one could possibly foresee the amount. Money was scarce, and in many districts the natural obstacles to be overcome were infinitely greater than any which had presented themselves to European engineers.

By the sound practical sense and the unconquerable will of George Stephenson, the numerous inventions which together make up the locomotive engine had been collected into a machine which, in combination with the improved roadway, was to revolutionize the transportation of the world. The railroad, as a machine, was invented. It remained to apply the new invention in such a manner as to make it a success, and not a failure. To do this in a new country like America required infinite skill, unbounded energy, the most careful study of local conditions, and the exercise of well matured, sound business judgment. To see how well the great invention has been applied in the United States, we have only to look at the network of iron roads which now reaches from the Great Lakes to the Gulf of Mexico, and from the Atlantic to the Pacific.

With all the experience we have had, it is not an easy problem, even at the present time, to determine how much money we are authorized to spend upon the construction of a given railroad. To secure the utmost benefit at the least outlay, regarding both the first cost of building the road and the perpetual cost of operating it, is the railroad problem which is perhaps less understood at the present day than any other. It was an equally important problem fifty years ago, and certainly not less difficult at that time. It was the fathers of the railroad system in the United States who first perceived the importance of this problem, and who, adapting themselves to the new conditions presented in this country, undertook to solve it. Among the pioneers in this branch of engineering no one has done more to establish correct methods, nor has left behind a more enviable or more enduring fame, than Major George W. Whistler.

The Whistler family is of English origin, and is found toward the end of the 15th century in Oxfordshire, at Goring and Whitchurch, on the Thames. One branch of the family settled in Sussex, at Hastings and Battle, being connected by marriage with the Websters of Battle Abbey, in which neighborhood some of the family still live. Another branch lived in Essex, from which came Dr. Daniel Whistler, President of the College of Physicians in London in the time of Charles the Second. From the Oxfordshire branch came Ralph, son of Hugh Whistler, of Goring, who went to Ireland, and there founded the Irish branch of the family, being the original tenant of a large tract of country in Ulster, under one of the guilds or public companies of the city of London. From this branch of the family came Major John Whistler, father of the distinguished engineer, and the first representative of the family in America. It is stated that in some youthful freak he ran away and enlisted in the British Army. It is certain that he came to this country during the Revolutionary War, under General Burgoyne, and remained with his command until its surrender at Saratoga, when he was taken prisoner of war. Upon his return to England he was honorably discharged, and, soon after, forming an attachment for a daughter of Sir Edward Bishop, a friend of his father, he eloped with her, and came to this country, settling at Hagerstown, in Maryland. He soon after entered the army of the United States, and served in the ranks, being severely wounded in the disastrous campaign against the Indians under Major-General St. Clair in the year 1791. He was afterward commissioned as lieutenant, rose to the rank of captain, and later had the brevet of major. At the reduction of the army in 1815, having already two sons in the service, he was not retained; but in recognition of his honorable record, he was appointed Military Storekeeper at Newport, Kentucky, from which post he was afterward transferred to Jefferson Barracks, where he lived to a good old age.

Major John Whistler had a large family of sons and daughters, among whom we may note particularly William, who became a colonel in the United States Army, and who died at Newport, Ky., in 1863; John, a lieutenant in the army, who died of wounds received in the battle of Maguago, near Detroit, in 1812; and George Washington, the subject of our sketch. Major John Whistler was not only a good soldier, and highly esteemed for his military services, but was also a man of refined tastes and well educated, being an uncommonly good linguist and especially noted as a fine musician. In his family he is stated to have united firmness with tenderness, and to have impressed upon his children the importance of a faithful and thorough performance of duty in whatever position they should be placed.

George Washington Whistler, the youngest son of Major John Whistler, was born on the 19th of May, in the year 1800, at Fort Wayne, in the present State of Indiana, but then part of the Northwest Territory, his father being at the time in command of that post. Of the boyhood of Whistler we have no record, except that he followed his parents from one military station to another, receiving his early education for the most part at Newport, Ky., from which place, on July 31, 1814, he was appointed a cadet to the United States Military Academy, being then fourteen years of age. The course of the student at West Point was a very satisfactory one. Owing to a change in the arrangement of classes after his entrance, he had the advantage of a longer term than had been given to those who preceded him, remaining five years under instruction. His record during his student life was good throughout. In a class of thirty members he stood No. 1 in drawing, No. 4 in descriptive geometry, No. 5 in drill, No. 11 in philosophy and in engineering, No. 12 in mathematics, and No. 10 in general merit. He was remarkable, says one who knew him at this time, for his frank and open manner and for his pleasant and cheerful disposition. A good story is told of the young cadet which shows his ability, even at this time, to make the best of circumstances apparently untoward, and to turn to his advantage his surroundings, whatever they might be. Having been for some slight breach of discipline required to bestride a gun in the campus for a short time, he saw, to his dismay, coming down the walk the beautiful daughter of Dr. Foster Swift, a young lady who, visiting West Point, had taken the hearts of the cadets by storm, and who, little as he may at the time have dreamed it, was destined to become his future wife. Pulling out his handkerchief, he bent over his gun, and appeared absorbed in cleaning the most inaccessible parts of it with such vigor as to be entirely unaware that any one was passing; nor did the young lady dream that a case of discipline had been before her until in after years, when, on a visit to West Point, an explanation was made to her by her husband.

It was at this time of his life that the refinement and taste for which Major Whistler was ever after noted began to show itself. An accomplished scientific musician and performer, he gained a reputation in this direction beyond that of a mere amateur, and scarcely below that of the professionals of the day. His sobriquet of "Pipes," which his skill upon the flute at this time gave him, adhered to him through life among his intimates in the army. His skill with the pencil, too, was something phenomenal, and would, had not more serious duties prevented, have made him as noted an artist as he was an engineer. Fortunately for the world this talent descended to one of his sons, and in his hands has had full development. These tastes in Major Whistler appeared to be less the results of study than the spontaneous outgrowth of a refined and delicate organization, and so far constitutional with him that they seemed to tinge his entire character. They continued to be developed till past the meridian of life, and amid all the pressure of graver duties furnished a most delightful relaxation.

Upon completing his course at the Military Academy he was graduated, July 1, 1819, and appointed second lieutenant in the corps of artillery. From this date until 1821 he served part of the time on topographical duty, and part of the time he was in garrison at Fort Columbus. From November 2, 1821, to April 30, 1822, he was assistant professor at the Military Academy, a position for which his attainments in descriptive geometry and his skill in drawing especially fitted him. This employment, however, was not altogether to his taste. He was too much of an artist to wish to confine himself to the mechanical methods needed in the training of engineering students. In 1822, although belonging to the artillery, he was detailed on topographical duty under Major (afterward Colonel) Abert, and was connected with the commission employed in tracing the international boundary between Lake Superior and the Lake of the Woods. This work continued four years, from 1822 to 1826, and subsequent duties in the cabinet of the commission employed nearly two years more.

The field service of this engagement was anything but light work, much of it being performed in the depth of winter with a temperature fifty degrees below zero. The principal food of the party was tallow and some other substance, which was warmed over a fire on stopping at night. The snow was then removed to a sufficient depth for a bed, and the party wrapped one another up in their buffalo robes, until the last man's turn came, when he had to wrap himself up the best he could. In the morning, after warming their food and eating, the remainder was allowed to harden in the pan, after which it was carried on the backs of men to the next stopping place. The work was all done upon snow-shoes, and occasionally a man became so blinded by the glare of the sun upon the snow that he had to be led by a rope.

Upon the 1st of June, 1821, Whistler was made second lieutenant in the First Artillery, in the reorganized army; on the 16th of August, 1821, he was transferred to the Second Artillery, and on the 16th of August, 1829, he was made first lieutenant. Although belonging to the artillery, he was assigned to topographical duty almost continually until December 31, 1833, when he resigned his position in the army. A large part of his time during this period was spent in making surveys, plans, and estimates for public works, not merely those needed by the national government, but others which were undertaken by chartered companies in different parts of the United States. There were at that time very few educated engineers in the country, besides the graduates of the Military Academy; and the army engineers were thus frequently applied for, and for several years government granted their services.

Prominent among the early works of internal improvement was the Baltimore & Ohio Railroad, and the managers of this undertaking had been successful in obtaining the services of several officers who were then eminent, or who afterward became so. The names of Dr. Howard, who, though not a military man, was attached to the Corps of Engineers, of Lieut.-Col. Long, and of Capt. William Gibbs McNeill appear in the proceedings of the company as "Chiefs of Brigade," and those of Fessenden, Gwynne, and Trimble among the assistants.

In October, 1828, this company made a special request for the services of Lieutenant Whistler. The directors had resolved on sending a deputation to England to examine the railroads of that country, and Jonathan Knight, William Gibbs McNeill, and George W. Whistler were selected for this duty. They were also accompanied by Ross Winans, whose fame and fortune, together with those of his sons, became so widely known afterward in connection with the great Russian railway. Lieutenant Whistler, says one who knew him well, was chosen for this service on account of his remarkable thoroughness in all the details of his profession, as well as for his superior qualifications in other respects. The party left this country in November, 1828, and returned in May, 1829.

In the course of the following year the organization of the Baltimore and Ohio Railroad, a part of which had already been constructed under the immediate personal supervision of Lieutenant Whistler, assumed a more permanent form, and allowed the military engineers to be transferred to other undertakings of a similar character. Accordingly, in June, 1830, Captain McNeill and Lieutenant Whistler were sent to the Baltimore and Susquehanna Railroad, for which they made the preliminary surveys and a definite location, and upon which they remained until about twenty miles were completed, when a lack of funds caused a temporary suspension of the work. In the latter part of 1831 Whistler went to New Jersey to aid in the construction of the Paterson and Hudson River Railroad (now a part of the Erie Railway). Upon this work he remained until 1833, at which time he moved to Connecticut to take charge of the location of the railroad from Providence to Stonington, a line which had been proposed as an extension of that already in process of construction from Boston to Providence.

In this year, December 31, 1833, Lieut. Whistler resigned his commission in the army, and this not so much from choice as from a sense of duty. Hitherto his work as an engineer appears to have been more an employment than a vocation. He carried on his undertakings diligently, as it was his nature to do, but without much anxiety or enthusiasm; and he was satisfied in meeting difficulties as they came up, with a sufficient solution. Henceforward he handled his profession from a love of it. He labored that his resources against the difficulties of matter and space should be overabundant, and if he had before been content with the sure-footed facts of observation, he now added the luminous aid of study. How luminous and how sure these combined became, his later works show best.

In 1834 Mr. Whistler accepted the position of engineer to the proprietors of locks and canals at Lowell. This position gave him among other things the direction of the machine shops, which had been made principally for the construction of locomotive engines. The Boston and Lowell Railroad, which at this time was in process of construction, had imported a locomotive from the works of George and Robert Stephenson, at Newcastle, and this engine was to be reproduced, not only for the use of the Lowell road, but for other railways as well, and to this work Major Whistler gave a large part of his time from 1834 to 1837. The making of these engines illustrated those features in his character which then and ever after were of the utmost value to those he served. It showed the self-denial with which he excluded any novelties of his own, the caution with which he admitted those of others, and the judgment which he exercised in selecting and combining the most meritorious of existing arrangements. The preference which he showed for what was simple and had been tried did not arise from a want of originality, as he had abundant occasion to show during the whole of his engineering life. He was, indeed, uncommonly fertile in expedients, as all who knew him testify, and the greater the demand upon his originality, the higher did he rise to meet the occasion. The time spent in Lowell was not only to the great advantage of the company, but it increased also his own stores of mechanical knowledge, and in a direction, too, which in later years was of especial value to him.

In 1837 the condition of the Stonington Railroad became such as to demand the continual presence and attention of the engineer. Mr. Whistler therefore moved to Stonington, a place to which he became much attached, and to which he seems during all of his wanderings to have looked with a view of making it finally his home. While engaged upon the above road he was consulted in regard to many other undertakings in different parts of the country, and prominent among these was the Western Railroad of Massachusetts.

This great work, remarkable for the boldness of its engineering, was to run from Worcester through Springfield and Pittsfield to Albany. To surmount the high lands dividing the waters of the Connecticut from those of the Hudson called for engineering cautious and skillful as well as heroic. The line from Worcester to Springfield, though apparently much less formidable, and to one who now rides over the road showing no very marked features, demanded hardly less study, as many as twelve several routes having been examined between Worcester and Brookfield. To undertake the solution of a problem of so much importance required the best of engineering talent, and we find associated on this work the names of three men who in the early railroad enterprises of this country stood deservedly in the front rank: George W. Whistler, William Gibbs McNeill, and William H. Swift. McNeill had graduated from the Military Academy in 1817, and rose to the rank of major in the Topographical Engineers. Like Whistler, he had been detailed to take charge of the design and construction of many works of internal improvement not under the direction of the general government. These two engineers exercised an influence throughout the country for many years much greater than that of any others. Indeed, there were very few works of importance undertaken at that time in connection with which their names do not appear. This alliance was further cemented by the marriage between Whistler and McNeill's sister. Capt. William H. Swift had also graduated from the Military Academy, and had already shown marked ability as an engineer. Such were the men who undertook the location and construction of the railroad which was to surmount the high lands between the Connecticut and the Hudson, and to connect Boston with the Great West.

The early reports of these engineers to the directors of the Western Railroad show an exceedingly thorough appreciation of the complex problem presented to them, and a much better understanding of the principles involved in establishing the route than seems to have been shown in many far more recent works. In these early reports made in 1836 and 1837, we find elaborate discussions as to the power of the locomotive engine, and a recognition of the fact that in comparing different lines we must regard the plan as well as the profile, "as the resistance from curves on a level road may even exceed that produced by gravity on an incline;" and in one place we find the ascents "equated at 18 feet, the slope which requires double the power needed on a level road," resulting in a "virtual increase." We find also a very clear expression of the fact that an increased expenditure in the power needed to operate the completed road may overbalance a considerable saving in first cost. To bear this principle in mind, and at the same time to work in accordance with the directors' ideas of economy, in a country where the railroad was regarded very largely as an experiment, was by no means an easy task. The temptation to make the first cost low at the expense of the quality of the road in running up the valley of Westfield River was very great, and the directors were at one time very strongly urged to make an exceedingly narrow and crooked road west of Springfield; but Major Whistler so convinced the President, Thomas B. Wales, of the folly of such a course, that the latter declared, with a most emphatic prefix, that he would have nothing to do with such a two-penny cow-path, and thus prevented its adoption.

Mr. Whistler had many investigations to make concerning the plans and policy of railroad companies at a time when almost everything connected with them was comparatively new and untried. When he commenced, there was no passenger railroad in the country, and but very few miles of quarry and mining track. If at that time an ascent of more than 1 in 200 was required, it was thought necessary to have inclined planes and stationary power. It was supposed that by frequent relays it would be possible to obtain for passenger cars a speed of eight or nine miles an hour. Almost nothing was known of the best form for rails, of the construction of the track, or of the details for cars or engines. In all of these things Major Whistler's highly gifted and well balanced mind enabled him to judge wisely for his employers, and to practice for them the truest economy.

Major Whistler's employment upon the Western Railroad began while he was still engaged upon the Stonington line. In connection with his friend McNeill he acted as consulting engineer for the Western road from 1836 to 1840. From 1840 to 1842 he was its chief engineer, with his headquarters at Springfield. The steep grades west of the Connecticut presented not only a difficult problem in location and construction, but in locomotive engineering as well. At the present day we can order any equipment which may best meet the requirement upon any railroad, and the order will be promptly met by any one of our great manufactories. But in the early days of the Western Railroad it was far otherwise, and the locomotive which should successfully and economically operate the hitherto unheard of grade of over 80 feet to the mile was yet to be seen. The Messrs. Winans, of Baltimore, had built some nondescript machines, which had received the name of "crabs," and had tried to make them work upon the Western road. But after many attempts they were given up as unfit for such service.

These "crabs" were eight wheeled engines, weighing about 20 tons, with a vertical boiler. The wheels were 31/2 feet in diameter, but the engine worked on to an intermediate shaft, which was connected with the driving axle in such a way as to get the effect of a five foot wheel. These engines did not impress Major Whistler at all favorably. And it is related that one Sunday the watchman in charge of the building in which some of them were kept, hearing some one among the engines, went in quietly and overheard Major Whistler, apparently conversing with the "crab," and saying: "No; you miserable, top-heavy, lop-sided abortion of a grasshopper, you'll never do to haul the trains over this road." His experience in Lowell was here of great value to him, and he had become convinced that the engine of George Stephenson was in the main the coming machine, and needed but to be properly proportioned and of sufficient size to meet every demand.

With Major Whistler's work upon the Western Railroad his engineering service in this country concluded, and that by an occurrence which marked him as the foremost railroad engineer of his time. Patient, indefatigable, cautious, remarkable for exhaustless resource, admirable judgment, and the highest engineering skill, he had begun with the beginning of the railroad system, and had risen to the chief control of one of the greatest works in the world, the Western Railroad of Massachusetts. Not only had he shown the most far-sighted wisdom in fixing the general features of this undertaking, but no man surpassed him, if, indeed, any one equaled him, in an exact and thorough knowledge of technical details. To combine the various elements in such a manner as to produce the greatest commercial success, and to make the railroad in the widest sense of the word a public improvement, never forgetting the amount of money at his disposal, was the problem he had undertaken to solve. He had proved himself a great master in his profession, and had shown how well fitted he was to grapple with every difficulty. He was equally a man of science and a man of business. And to all this he added the most delicate sense of honor and the most spotless integrity. He was in the prime of manhood, and was prepared to enter upon the great work of his life.

It was not long after the introduction of the railroad that intelligent persons saw very plainly that the new mode of transportation was not to be confined to the working of an already established traffic, in densely populated regions, but that it would be of equal service in awakening the energies of undeveloped countries, in bringing the vast interior regions of the continents into communication with the seaboard, in opening markets to lands which before were beyond the reach of commerce. And it was seen, too, that in event of war, a new and invaluable element had been introduced, viz., the power of transportation to an extent never before imagined.

Especially were these advantages foreseen in the vast empire of Russia, and an attempt was very early made to induce private capitalists to undertake the construction of the lines contemplated in that country. The Emperor, besides guaranteeing to the shareholders a minimum profit of four per cent., proposed to give them, gratuitously, all the lands of the state through which the lines should pass, and to place at their disposal, also gratuitously, the timber and raw materials necessary for the way and works which might be found upon the ground. It was further proposed, to permit the importation of rails and of the rolling stock free of duty. Russian proprietors also came forward, and not only agreed to grant such portions of their land as the railroads might pass through, gratuitously, but further to dispossess themselves temporarily of their serfs, and surrender them to the use of the companies, on the sole condition that they should be properly supported while thus employed.

With regard to the great line, however, which was to unite the two capitals, St. Petersburg and Moscow, it was decreed that this should be made exclusively at the expense of the state, in order to retain in the hands of the government and in the general interest of the people a line of communication so important to the industry and the internal commerce of the country. The local proprietors agreed to surrender to the government, gratuitously, the lands necessary for this line.

It was very early understood that the railroad problem in Russia was much more analogous to that in the United States than to that in England. The Emperor, therefore, in 1839, sent the Chevalier De Gerstner to the United States to obtain information concerning the railroads of this country. It was this person who obtained from the Emperor the concession for the short railway from St. Petersburg to Zarskoe Selo, which had been opened in 1837, and who had also made a careful reconnoissance in 1835 for a line from St. Petersburg to Moscow, and had very strongly urged its construction on the American plan. The more De Gerstner examined our roads, the more impressed he was with the fitness of what he termed the American system of building and operating railroads to the needs of the empire of Russia. In one of his letters in explaining the causes of the cheap construction of American railroads, after noting the fact that labor as well as material is much dearer in America than in Europe, he refers to the use of steep grades (93 feet to the mile) and sharp curves (600 feet radius), upon which the American equipment works easily, to the use of labor saving machinery, particularly to a steam excavating machine upon the railroad between Worcester and Springfield, and to the American system of wooden bridge building, and says: "The superstructure of the railroads in America is made conformable to the expected traffic, and costs therefore more or less accordingly;" and he concludes, "considering the whole, it appears that the cheapness of the American railroads has its foundation in the practical sense which predominates in their construction." Again, under the causes of the cheap management of the American roads, he notes the less expensive administration service, the low rate of speed, the use of the eight wheeled cars and the four-wheeled truck under the engines, and concludes: "In my opinion it would be of great advantage for every railroad company in Europe to procure at least one such train" (as those used in America). "Those companies, however, whose works are yet under construction I can advise with the fullest conviction to procure all their locomotive engines and tenders from America, and to construct their cars after the American model."

Notwithstanding this report, the suggestions of De Gerstner were not at once accepted. The magnitude of the enterprise would not admit of taking a false step. Further evidence was needed, and accordingly it was decided to send a committee of engineer officers to various countries in Europe, and to the United States, to select such a system for the road and its equipment as would be best adapted to Russia. These officers, Colonels Melnikoff and Krofft, not only reported in the most decided manner in favor of the American methods, but also stated that of all persons with whom they had communicated, no one had given them such full and satisfactory information upon all points, or had so impressed them as possessing extraordinary ability, as Major George W. Whistler. This led to his receiving an invitation from the Emperor to go to Russia and become consulting engineer for the great road which was to connect the imperial city upon the Baltic with the ancient capital of the Czars.

When we consider the magnitude of the engineering works with which the older countries abound, we can but regard with a feeling of pride the fact that an American should have been selected for so high a trust by a European government possessing every opportunity and means for securing the highest professional talent which the world could offer. Nor should it be forgotten that the selection of our countryman did not arise from any necessity which the Russian Government felt for obtaining professional aid from abroad, growing out of a lack of the requisite material at home. On the contrary, the engineers of the Russian service are perhaps the most accomplished body of men to be found in any country. Selected in their youth, irrespective of any artificial advantages of birth or position, but for having a genius for such work, and trained to a degree of excellence in all of the sciences unsurpassed in any country, they stand deservedly in the front rank. Such was the body of men with whom Major Whistler was called to co-operate, and whose professional duties, if not directed specially by him, were to be controlled by his judgment.

Accepting the position offered to him in so flattering a manner, he sailed for St. Petersburg about mid-summer in 1842, being accompanied on his voyage by Major Bouttattz, of the Russian Engineer Corps, who had been sent to this country by the Emperor as an escort. Arriving in St. Petersburg, and having learned the general character of the proposed work, he traveled partly by horse and partly on foot over the entire route, and made his preliminary report, which was at once accepted.

The plan contemplated the construction of a double track railroad 420 miles long, perfect in all its parts, and equipped to its utmost necessity. The estimates amounted to nearly forty millions of dollars, and the time for its construction was reckoned at seven years. The line selected for the road had no reference to intermediate points, and was the shortest attainable, due regard being paid to the cost of construction. It is nearly straight, and passes over so level a country as to encounter no obstacle requiring a grade exceeding 20 feet to the mile, and for most of the distance it is level. The right of way taken was 400 feet in width throughout the entire length. The roadbed was raised from six to ten feet above the ordinary level of the country, and was 30 feet wide on top.

One of the most important questions to settle at the outset in regard to this great work was the width of the gauge. At that time the opinion in England as well as in the United States among engineers was setting very strongly in favor of a gauge wider than 4 feet 81/2 inches, and the Russian engineers were decidedly in favor of such increased width. Major Whistler, however, in an elaborate report to the Count Kleinmichel argued very strongly in favor of the ordinary gauge. To this a commission of the most distinguished engineers in Russia replied, urging in the most forcible manner the adoption of a gauge of six feet. Major Whistler rejoined in a report which is one of the finest models of an engineering argument ever written, and in which we have perhaps the best view of the quality of his mind. In this document no point is omitted, each part of the question is handled with the most consummate skill, the bearing of the several parts upon the whole is shown in the clearest possible manner, and in a style which could only come from one who from his own knowledge was thoroughly familiar with all the details, not only of the railroad, but of the locomotive as well.

In this report the history of the ordinary gauge is given, with the origin of the standard of 4 feet 81/2 inches; the questions of strength, stability, and capacity of cars, of the dimensions, proportions, and power of engines, the speed of trains, resistances to motion, weight and strength of rails, the cost of the roadway, and the removal of snow are carefully considered. The various claims of the advocates for a wider gauge are fairly and critically examined, and while the errors of his opponents are laid bare in the most unsparing manner, the whole is done in a spirit so entirely unprejudiced, and with so evident a desire for the simple truth, as to carry conviction to any fair minded person. The dry way, too, in which he suggests that conclusions based upon actual results from existing railways are of more value than deductions from supposed conditions upon imaginary roads, is exceedingly entertaining. The result was the adoption of the gauge recommended by him, namely, five feet. Those who remember the "Battle of the Gauges," and who know how much expense and trouble the wide gauge has since caused, will appreciate the stand taken thus early by Major Whistler; and this was but one among many cases which might be mentioned to show how comprehensive and far-reaching was his mind.

The roadbed of the St. Petersburg and Moscow Railway was made 30 feet wide on top, for a double track of 5 foot gauge, with a gravel ballasting two feet deep. The bridges were of wood, of the Howe pattern, no spans being over 200 feet in length. The stations at each end, and the station and engine houses along the line, were on a plan uniform throughout, and of the most ample accommodation. Fuel and water stations were placed at suitable points, and engine houses were provided 50 miles apart, built of the most substantial masonry, circular in form, 180 feet in diameter, surmounted by a dome, and having stalls for 22 engines each. Repair shops were attached to every engine house, furnished with every tool or implement that the wants of the road could suggest.

The equipment of rolling stock and fixed machinery for the shops was furnished by the American firm of Winans, Harrison & Eastwick, who from previous acquaintance were known by Major Whistler to be skillful, energetic, and reliable. Much diplomacy was needed to procure the large money advances for this part of the work, the whole Winans contract amounting to nearly five millions of dollars; but the assurance of Major Whistler was a sufficient guarantee against disappointment or failure.

In 1843 the plans for the work were all complete, and in 1844 the various operations along the line were well under way, and proceeding according to the well arranged programme. In 1845 the work had progressed so far that the construction of the rolling stock was commenced. The locomotives were of two classes, freight and passenger. The engines of each class were made throughout from the same patterns, so that any part of one engine would fit the same position on any other. The passenger engines had two pairs of driving wheels, coupled, 6 feet in diameter, and a four wheeled truck similar to the modern American locomotive. The general dimensions were: Waist of boiler, 47 inches, 186 two inch tubes 101/2 feet long; cylinders, 16 x 22 inches. The freight engines had the same capacity of boiler and the same number and length of tubes, three pairs of driving wheels, coupled, 41/2 feet in diameter, a truck and cylinders 18 x 22 inches, and all uniform throughout in workmanship and finish. The passenger cars were 56 feet long and 91/2 feet wide, the first class carrying 33 passengers, the second class 54, and the third class 80. They all had eight truck wheels under each, and elliptic steel springs. The freight cars were all 30 feet long and 91/2 feet wide, made in a uniform manner, with eight truck wheels under each. The imperial saloon carriages were 80 feet long and 91/2 feet wide, having double trucks, or sixteen wheels under each. They were divided into five compartments and fitted with every convenience.

Early in 1847 the Emperor Nicholas visited the mechanical works at Alexandroffsky, where the rolling stock was being made by the Messrs. Winans, in the shops prepared by them and supplied by Russian labor. Everything here was on the grandest scale, and the work was conducted under the most perfect system. Upon this occasion the Emperor was so much gratified at what had already been accomplished that he conferred upon Major Whistler the decoration of the Order of St. Anne. He had previously been pressed to wear the Russian uniform, which he promptly declined to do; but there was no escape from the decoration without giving offense. He is said, however, to have generally contrived to hide it beneath his coat in such a manner that few ever saw it.

Technically, Major Whistler was consulting engineer, Colonel Melnikoff being constructing engineer for the northern half of the road, and Colonel Krofft for the southern half; but as a matter of fact, by far the larger part of planning the construction in detail of both railway and equipment fell upon Major Whistler. There was also a permanent commission having general charge of the construction of the road, of which the president was General Destrem, one of the four French engineers whom Napoleon, at the request of the Emperor Alexander, sent to Russia for the service of that country.

The year 1848 was a very trying one to Major Whistler. He had already on several occasions overtasked his strength, and had been obliged to rest. This year the Asiatic cholera made its appearance. He sent his family abroad, but remained himself alone in his house. He would on no account at this time leave his post, nor omit his periodical inspections along the line of the road, where the epidemic was raging. In November he had an attack of cholera, and while he recovered from it, he was left very weak. Still, he remained upon the work through the winter, though suffering much from a complication of diseases. As spring advanced he became much worse, and upon the 7th of April, 1849, he passed quietly away, the immediate cause of his death being a trouble with the heart.

Funeral services were held in the Anglican (Episcopal) Church in St. Petersburg. His body was soon afterward carried to Boston and deposited beneath St. Paul's Church; but the final interment took place at Stonington. The kindness and attention of the Emperor and of all with whom Major Whistler had been associated knew no bounds. Everything was done to comfort and aid his wife, and when she left St. Petersburg the Emperor sent her in his private barge to the mouth of the Baltic. "It was not only," says one who knew him weil, "through his skill, ability, and experience as an engineer that Major Whistler was particularly qualified for and eminently successful in the important task he performed so well in Russia. His military training and bearing, his polished manner, good humor, sense of honor, knowledge of a language (French) in which he could converse with officers of the government, his resolution in adhering to what he thought was right, and in meeting difficulties only to surmount them, with other admirable personal qualities, made him soon, and during his whole residence in Russia, much liked and trusted by all persons by whom he was known, from the Emperor down to the peasant. Such is the reputation he left behind him, and which is given to him in Russia to this day."

In 1849 the firm of Winans, Harrison and Eastwick had already furnished the road with 162 locomotives, 72 passenger and 2,580 freight cars. They had also arranged to instruct a suitable number of Russian mechanics to take charge of the machinery when completed. The road was finished its entire length in 1850, being opened for passenger and freight traffic on the 25th of September of that year, in two divisions, experimentally, and finally opened for through business on November 1, 1851. In all of its construction and equipment it was essentially American of the best kind, everything being made under a carefully devised system, by which the greatest economy in maintenance and in management should be possible. The use of standard patterns, uniformity in design and duplication of parts was applied, not only to the rolling stock, but to the railroad as well, wherever it was possible. Indeed, the whole undertaking in all its parts bore the impress of one master mind.

On the death of Major Whistler the government with jealous care prevented any changes whatever being made in his plans, including those which had not been carried out as well as those already in process of execution. An American engineer, Major T.S. Brown, was invited to Russia to succeed Major Whistler as consulting engineer. The services of the Messrs. Winans also were so satisfactory to the government that a new contract was afterward made, upon the completion of the road, for the maintenance and the future construction of rolling stock.

While the great railroad was the principal work of Major Whistler in Russia, he was also consulted in regard to all the important engineering works of the period. The fortifications at Cronstadt, the Naval Arsenal and docks at the same place, the plans for improving the Dwina at Archangel, the great iron roof of the Riding House at St. Petersburg, and the iron bridge over the Neva all received his attention. The government was accustomed to rely upon his judgment in all cases requiring the exercise of the highest combination of science and practical skill; and here, with a happy tact peculiarly his own, he secured the warm friendship of men whose professional acts he found himself called upon in the exercise of his high trust in many cases to condemn. The Russians are proverbially jealous of strangers, and no higher evidence of their appreciation of the sterling honesty of Major Whistler, and of his sound, discriminating judgment, could be afforded than the fact that all his recommendations on the great questions of internal improvement, opposed as many of them were to the principles which had previously obtained, and which were sanctioned by usage, were yet carried out by the government to the smallest details.

While in Russia Major Whistler was sometimes placed in positions most trying to him. It is said that some of the corps of native engineers, many of whom were nobles, while compelled to look up to him officially, were inclined to look down upon him socially, and exercised their supposed privileges in this respect so as to annoy him exceedingly, for he had not known in his own country what it was to be the social inferior of any one. The Emperor, hearing of this annoyance, determined to stop it; so, taking advantage of a day when he knew the engineer corps would visit a celebrated gallery of art, he entered it while they were there, and without at first noticing any one else, looked around for Major Whistler, and seeing him, went directly toward him, took his arm, and walked slowly with him entirely around the gallery. After this the conduct of the nobles was all that could be desired.

Major Whistler's salary while in Russia was $12,000 a year; a sum no more than necessary for living in a style befitting his position. He had abundant opportunity for making money, but this his nice sense of honor forbade. It is even stated that he would never allow any invention to be used on the road that could by any possibility be of any profit to himself or to any of his friends. He was continually besieged by American inventors, but in vain. The honor of the profession he regarded as a sacred trust. He served the Emperor with the fidelity that characterized all his actions. His unswerving devotion to his duty was fully appreciated, and it is said that no American in Russia, except John Quincy Adams, was ever held in so high estimation.

Major Whistler married for his first wife Mary, daughter of Dr. Foster Swift of the U.S. Army, and Deborah, daughter of Capt. Thomas Delano of Nantucket. By her he had three children: Deborah, his only daughter, who married Seymour Haden of London, a surgeon, but later and better known for his skill in etching; George William, who became an engineer and railway manager, and who went to Russia, and finally died at Brighton, in England, Dec. 24, 1869; Joseph Swift, born at New London, Aug. 12, 1825, and who died at Stonington, Jan. 1, 1840. His first wife died Dec. 9, 1827, at the early age of 23 years, and is buried in Greenwood Cemetery, in the shade of the monument erected to the memory of her husband by the loving hands of his professional brethren. For his second wife he married Anna Matilda, daughter of Dr. Charles Donald McNeill of Wilmington, N.C., and sister of his friend and associate, William Gibbs McNeill. By her he had five sons: James Abbot McNeill, the noted artist, and William Gibbs McNeill, a well known physician, both now living in London; Kirk Boott, born in Stonington, July 16, 1838, and who died at Springfield, July 10, 1842; Charles Donald, born in Springfield, Aug. 27, 1841, and who died in Russia, Sept. 24, 1843; and John Bouttattz, who was born and who died at St. Petersburg, having lived but little more than a year. His second wife, who outlived him, returned to America, and remained here during the education of her children, after which she moved to England. She died Jan. 31, 1881, at the age of 76 years, and was buried at Hastings.

At a meeting held in the office of the Panama Railroad Company in New York, August 27, 1849, for the purpose of suggesting measures expressive of their respect for the memory of Major Whistler, Wm. H. Sidell being chairman and A.W. Craven secretary, it was resolved that a monument in Greenwood Cemetery would be a suitable mode of expressing the feelings of the profession in this respect, and that an association be formed to collect funds and take all necessary steps to carry out the work. At this meeting Capt. William H. Swift was appointed president, Major T.S. Brown treasurer, and A.W. Craven secretary, and Messrs. Horatio Allen, W.C. Young, J.W. Adams, and A.W. Craven were appointed a committee to procure designs and estimates, and to select a suitable piece of ground. The design was made by Mr. Adams, and the ground was given by Mr. Kirkwood. The monument is a beautiful structure of red standstone, about 15 feet high, and stands in "Twilight Dell." Upon the several faces are the following inscriptions:

Upon the Front.

IN MEMORY OF GEORGE WASHINGTON WHISTLER, CIVIL ENGINEER, BORN AT FORT WAYNE, INDIANA, MAY, 1800, DIED AT ST. PETERSBURG, RUSSIA, APRIL, 1849.

Upon the Right Side.

EDUCATED AT THE U.S. MILITARY ACADEMY. HE RETIRED FROM THE ARMY IN 1833 AND BECAME ASSOCIATED WITH WILLIAM GIBBS M'NEILL. THEY WERE IN THEIR TIME ACKNOWLEDGED TO BE AT THE HEAD OF THEIR PROFESSION IN THIS COUNTRY.

Upon the Back.

HE WAS DISTINGUISHED FOR THEORETICAL AND PRACTICAL ABILITY, COUPLED WITH SOUND JUDGMENT AND GREAT INTEGRITY. IN 1842 HE WAS INVITED TO RUSSIA BY THE EMPEROR NICHOLAS, AND DIED THERE WHILE CONSTRUCTING THE ST. PETERSBURG & MOSCOW RAILROAD.

Upon the Left Side.

THIS CENOTAPH IS A MONUMENT OF THE ESTEEM AND AFFECTION OF HIS FRIENDS AND COMPANIONS.

While the monument thus raised to the memory of the great engineer stands in that most delightful of the cities of the dead, his worn-out body rests in the quaint old town of Stonington. It was here that his several children had been buried, and he had frequently expressed a desire that when he should die he might be placed by their side. A deputation of engineers who had been in their early years associated with him attended the simple service which was held over his grave, and all felt as they turned away that they had bid farewell to such a man as the world has not often seen.

In person Major Whistler was of medium size and well made. His face showed the finest type of manly beauty, combined with a delicacy almost feminine. In private life he was greatly prized for his natural qualities of heart and mind, his regard for the feelings of others, and his unvarying kindness, especially toward his inferiors and his young assistants. His duties and his travels in this and in other countries brought him in contact with men of every rank; and it is safe to say that the more competent those who knew him were to judge, the more highly was he valued by them. A close observer, with a keen sense of humor and unfailing tact, fond of personal anecdote, and with a mind stored with recollections from association with every grade of society, he was a most engaging companion. The charm of his manner was not conventional, nor due to intercourse with refined society, but came from a sense of delicacy and a refinement of feeling which was innate, and which showed itself in him under all circumstances. He was in the widest and best sense of the word a gentleman; and he was a gentleman outwardly because he was a gentleman at heart.

As an engineer, Whistler's works speak for him. He was eminently a practical man, remarkable for steadiness of judgment and for sound business sense. Whatever he did was so well done that he was naturally followed as a model by those who were seeking a high standard. Others may have excelled in extraordinary boldness or in some remarkable specialty, but in all that rounds out the perfect engineer, whether natural characteristics, professional training, or the well digested results of long and valuable experience, we look in vain for his superior, and those who knew him best will hesitate to acknowledge his equal.—Journal of the Association of Engineering Societies.

1  2  3     Next Part
Home - Random Browse