User:Marino108LFS/sandbox/Lorem Ipsum

February 10th, 1852.JAMES MEADOWS RENDEL, President, in the Chair.No. 866. " The Construction and Duration of the PermanentWay of Railways iu Europe, and the modifications most suit-able to Egypt, India, &c." By William Bridges Adams. 1In examining the " Permanent Way " of various kinds, used up tothe present period, it is necessary first to settle the exact meaningof the term. In its simple sense, it would imply a way of greatdurability ; but that it does not satisfactorily fulfil this condition,may be assumed from the numerous changes and attempts at im-provement it is constantly undergoing.It must therefore be understood, that the term " permanent way "means some kind of way left by the contractor, when he hasremoved his temporary way, and which may be of greater, or less,durability, or permanence, according to irs structure, but whichmust, in most cases, be governed by the quality of the material thatcan be most easily procured. Where timber is plentiful, it islargely used for the substructure ; if it is scarce, stone is substituted ;and if the capital is restricted, the iron is pinched too closely, as onthe early railways, in the United States. Whilst the ballast alsovaries according to locality, from broken stone, and burnt clay, tosand and gravel.The principal requirements of " permanent way " are ; that it bewell drained, and especially in contiguity to the substructure. Thatthe weight and damaging power of the engines and rolling stock,should be considered as the datum for calculation. That thestrength, hardness, and tenacity of the rails, and the immobility ofthe substructure, should be adapted for the hardest work to whichthe railway is to be subjected. That the substructure should havean amount of bearing surface, proportioned to the load to be borne,and the nature of the soil, or ballast ; and a sufficiently firm hold1 The discussion on this paper extended over a portion of three evenings,but an abstract of the whole is given consecutively.b 2Digitized by'4PERMANENT WAV.in the ground, to prevent looseness, or lateral movement, from theside lurches of the engines, or trains. That the rails should possessso much vertical and lateral stiffness, either in themselves, or bytheir fastenings, as to prevent all deflection, and have sufficienthardness of surface not to laminate, or to disintegrate, beneath therolling loads; and have sufficient breadth, or tread-surface, todiminish the effect of the crushing power of the wheels. Theyshould be as smooth as possible on the running surface, to preventconcussion ; and be laid at the proper angle, and the curves re-gularly bent, so as to insure the accurate tread of the wheels:whilst the joints should be so made, that the rails may, practically,become continuous bars, yet with freedom to expand and contractwithout being too loose. And with all this, there should beinterposed, between the rails and the solid ground, some medium,sufficiently elastic to absorb the effect of the blows of the wheels,without being crushed, or forced down into the ballast, and yetstiff enough to keep the upper surface of the rails in a uniformplane.If these conditions are fulfilled the railw ay will be a " permanentway " in the full sense of the term.The ballast may be considered the foundation of the railway ;and the depth of it must vary with the structure. If sleepers areused in the ordinary manner, they should be sunk into the ballast,and yet have a sufficient quantity below, to prevent them from gra-dually sinking. Therefore, the thicker and more efficient thesleeper, the greater must be the depth of the ballast ; and this makesthe choice of the sleeper an important consideration ; which again,is regulated by the kind of rail intended to be used.The earliest, cheapest, or rather lowest priced and worst form,of wrought-iron rails, is the flat tire-bar rail, spiked down on alongitudinal balk. This was adopted on most of the Americanrailways at the outset, but is now abandoned, on account of theenormous wear and tear, and the great danger to passengers, as theends curled up, and occasionally killed a passenger, through thebottom of the carriage.A variation of this kind of way, with the bars edgewise, has beenfor many years in use in quarries and other places where only a slowmotion of the trams is required.The next in order is the single T fish-bellied rail, ranging from28 lbs. to 35 lbs. per yard. It was originally patented by Mr.Birkinshaw, in the year 1820, and was first employed to any consider-able extent, by Mr. G. Stephenson, on the Liverpool and ManchesterRailway (1829-80) ; but is now no longer manufactured. This railDigitized by GooglePERMANENT WAY.516 a remarkable instance of correct reasoning from unproved data,for though all chair-rails are practically girders, the chairs being thesupporting piers, yet in a rail 15 feet long, when the chairs yielded)that reduced portion of the rail which was supported by them,would also yield and bend ; so that instead of making the rail asuccession of fish-bellies, it ought rather to have been one singlecurved line at the bottom, as in the case of the cast-iron fish-belliedrail.The next is the single T parallel rail used with a cast-iron chair :this form, which was first extensively used by Mr. Vignoles, aboutthe year 1833, is a valuable rail, under some special applications,hereafter to be noticed.The next is the double X parallel rail, generally used with chairs :this is a convenient, as well as a very strong mechanical form, andis well adapted for traffic. This double table-rail was firstpractically adopted on the Grand Junction Railway, by Mr. Locke,M.P., V.P., under whose directions the precise form was carriedout by Mr. C. Manby (now Secretary of the Institution of CivilEngineers), to whom the order was given for manufacturing thefirst quantity, at the Ebbw Vale Ironworks, South Wales, early inthe year 1835, when a very elaborate series of investigations wasmade by Mr. Locke, on the various forms of rails, and chairs, thematerial for, and the distances between the sleepers, &c. The rail was4£ inches deep, with a flat bearing-surface of 2£ inches wide onthe top and bottom tables, which were precisely similar, and theweight was about 62 lbs. per yard. The chairs were of cast-iron,and were originally intended to have in each a metal filling-inpiece, with a round oak trenail, or plug, between it and the faceof the jaw of the chair, in order to force it against the web ofthe rail : this plan was, however, after a very short trial, aban-doned, and solid plugs of oak were substituted for the iron filling-in piece and oak trenail. These oak plugs, or keys, were com-pressed into iron moulds, under hydraulic presses, by Mr. Beattie,at the Crown-street works, Liverpool, where, under Mr. Locke'sdirections, the same process had been in operation for two yearspreviously, and the compressed plugs were first tried on the Liver-pool and Manchester Railway.The foot-rail, a single (reversed) J. section, with a broad foot, orbase, to stand on longitudinal timbers, or on cross sleepers, to whichit was bolte.il down, through holes in the foot, was generallyknown as the * Stepiienson and Vignoles rail.' It is now commonlyused by contractor for a temporary way, with transverse sleepers,to which it is attached by dogs, or spikes, and it was, at one period,extensively used in the United States for permanent way, butDigitized by Google6PERMANENT WAY.being sometimes insufficiently fastened down by spikes, it not un-frequently caused accidents.Then comes the well-known bridge-rail, first used by Mr. Bruneion the Great Western Railway in 1838, and subsequently, on manyother lines.The rail manufactured by Sir John Guest, for Mr. Stevens, C.E.(U. S. America), is generically a foot-rail, but as it is 7 inches inheight, with a foot only 4| inches in width, and the vertical web only4 inch in thickness, it is obvious that it must be supported laterally.Probably this is accomplished by bolting it between two longitu-dinal timber balks, into which the foot is grooved ; but the Authorhas not been able to gain information respecting its real mode ofapplication.Split rails, or rails rolled in two pieces, both in the form of thehalf of a double I, and also as a foot-rail, with a vertical joint be-tween them, and put together so as to " break joint," are extensivelyused in America, being manufactured under a patent in England.The object of this form of rail is, in fact, to give a longitudinalscarf through the whole length, so as to prevent bad joints.Barlow's " saddle-back rail " is the latest great innovation ; itis used without sleepers ; or, in the words of the patentee, " is atrough rail, bearing on the ballast, and riveted to bearers."(Fig. 1.)Fig. 1.Mr. W. H. Barlow's Saddle-back Kail.Two general systems, known as the transverse and the longi-tudinal methods, have been adopted, for securing rails upon timbersleepers, and sometimes the two have been combined. But thedesirability of adopting a material chemically durable, originallyinduced the use of stone blocks ; and as these have been almostaltogether abandoned, it will be well to inquire, why this has takenplace, and if for sufficient reasons.The blocks were usually about 2 feet square, by 1 foot in depth,containing 4 cubic feet of stone. They were of various qualitiesaccording to the localities, and were placed, sometimes rectangularlyin line, with intervals between them, and at others diagonally,nearly touching at the corners, and forming a continuous foundation,DigitizPERMANENT WAY.7or substructure ; in some instances they were laid in ballast, and inothers, only in the natural soil.To insure solidity, which was the object then aimed at, the cast-iron chairs were fitted accurately to the stones, and fastened to themby a pair of iron spikes, each driven into an oak trenail, bedded about6 inches deep, in a hole bored in the stone block. It was soon found,however, that the chairs worked loose and cut into the stones ; and toremedy this, a layer of patent felt, an eighth of an inch in thickness,was placed between the chair and the stone, but the felt beingcrushed by the heavy weights, the chairs became looser than beforeThe most probable solution of the defect is this : — the stone blockspartially subsided, irregularly deflecting the rails, which were helddown by the spikes, until the elasticity of the rail, in seeking to re-gain its form, gradually drew the fastenings, and then an incessanthammering ensued between the chair and the stone, on the passageof trains, gradually cutting into the surface of the stone a recess ofthe form of the chair. To pack the stone blocks, or to restore themto their proper position, at so great a depth, involved a costly ope-ration. Moreover, the noise in travelling was a nuisance to thepassengers, and the hardness wa< found to be very destructive to therolling stock, and therefore the u*e of stone blocks was discontinued,and in many cases they were removed from existing lines, and werereplaced by transverse timber-sleepers, previou>»ly only used in em-bankments of new construction, being in fact merely " contractors'way," but of larger scantling and materials.Before the question was finally disposed of, an experiment wasmade on a portion of the Leeds and Manchester line, by levellingthe solid rock, in which there could be no subsidence— and fasten-ing the chairs to it, exactly as they had previously been attached tothe stone blocks. The result it is said, was, to produce a road sohard and rigid, that it was impossible to continue the use of it, andthis was considered to have definitely settled the question. If,however, instead of discarding the stone blocks, they had been usedas foundations for supporting the timber sleepers, to which the railscould have been attached, it is probable that a much more reallypermanent way would havt resulted.The mechanical change that was effected, by the substitution oftimber for stone, was comparative freedom from noise, which wasundoubtedly an indication of a deflecting rail, rising and fallingwith the chairs on the blocks.The same rail, transferred to the transverse sleepers, would alsorise and fall, but without noise. And that they do deflect is provedby the oozing of watery mud from beneath the sleepers, in wetweather ; by the quantify of dust, in dry weather, and by the necessityDigitized by Google8PERMANENT WAY.for continually packing the sleepers, which latter forms a large itemof the cost of " maintenance of way."Supposing the rails to be integrally strong enough to prevent de-flection, when supported at their two ends, the seat of the rail inthe chairs requires to be of sufficient area to prevent that indentationboth of rail and chair, which may be occasionally seen to the depthof a quarter, or even three-eighths of an inch.A simple method of avoiding the use of cast-iron chairs, and atthe same time of firmly attaching the rails to the timber sleepers,would be to fasten, by transverse pins to both sides of the verticalweb, two short bracket blocks of oak, resting upon the timbersleepers, to which they could be spiked, and supporting the top webof the rail ; the bottom web resting upon and being grooved into thesleepers. The inner top of the bracket blocks would not touch theflanches of the wheels. The rails would thus be lower, the leveragewould be less, the thickness of ballast would be reduced, the chairsentirely done away with, and considerable economy be effected.(Fig. 2.)Fig. 2.Hail without Chairs.Again, it is essential, that the chair should have such a bearingsurface on the timber-sleeper, and be so secured to it, that the side-lurch of the engine, and the springing of the sleeper across thewidth of the road, cannot disturb it.The last consideration is, the bearing surface of the sleepers them-selves on the ballast. There are generally under each parallelpair of rails, of 15 feet in length, six sleepers of about 9 feet inlength, by 10 inches in width, and 5 inches in thickness.The rail thus becomes a girder-bridge supported on piers, andwhen the rail deflects, the sleepers are forced into the ballast indetail.To overcome this depression of the sleepers, either their numbermust be extended, if needful, till they touch each other, like thecorduroy, or plank roads of America, or el.«e the depth and strengthof the rails must be increased, so as to prevent all deflection.Digitized by GooglePERMANENT WAY.9In the longitudinal system of timber-sleepers, as used on theGreat Western and other lines, with bridge-rails, having a con-tinuous bearing on the timber, the bearing surface of the rails onthe timber is about 6 inches in width, and that of the sleepers onthe ballast is about 13 inches. The depth of the rail being about3£ inches, and the depth of the timber 6 inches.The quantity of timber used in the transverse system, withfour intermediate sleepers to each length of rail, is rather lessthan . 16 cubic feet, and on the longitudinal system it is rathermore than 16 cubic feet; or with about the same quantity oftimber, the bearing surface on the transverse plan will be about12 inches in width; and on the longitudinal plan, about 13inches, giving an advantage of 1 inch in width to each rail, ortwo inches to the pair of rails, in the longitudinal plan : but asin the latter plan, the sleepers are 1 inch deeper over tiie wholesurface, the effective quantity of timber is nearly 3 cubic feetmore. In comparing the expense of the two systems, the costof the fastening of the rails may be assumed to be equal inboth cases, and as the quantity of timber has been shown to beabout equal, and the quality ought to be the same, the cost shouldbe alike. For a long period the transverse timber system was considered the cheapest, but the reason was, that on many lines, thesleepers were very small, and of inferior quality, while in the caseof the longitudinal system, it was essential always to have goodtimber. In addition to the increase of bearing surface in the lon-gitudinal system, tiie broad base of the bridge-rail makes it muchstiffer laterally, than the double X rail? and its smaller depthotfera less opportunity of accident, from sudden lurches of theengine. But, on the other hand, the bri<ige-rail being only 3&inches in depth, is liable to deflect under loads, by which the doubleX rail, 5 inches in depth, would not be aflected.Tho'igh the longitudinal timber tends to stiffen the bridge-railvertically, the real effect depends entirely on the fastenings. Therail has two lateral flanches about If inch wide by £ inch in thick-ness, through which holes are pierced at intervals of 3 fe<-t.Iu the earlier examples these rails were fastened down by screws ;but as the rails and the timber deflected, a sliding movement tookplace between them, similar to that between the plates of a carriage-spring ; the screws were loosened, drawn out, or broken, and thetimber was crushed in detail. To remedy this, in some cases, thethe timber was cross-boarded, wiih manifest advantage, and in mostcases bolts with nuts passing through therail-flanchesand the timber,were substituted for the screw-bolts ; subsequently, plates were in-troduced under the rail-ends, and also below the timber, and theDigitized by Google10PERMANENT WAY.bolts passed through all. But the author is not aware of any fasten-ing yet being found quite effectual in preventing deflection, or theworking movement between the rails, and the timbers, or the loosen-ing of the bolt-heads.On the old Croydon line, where the longitudinal system was used,with very light bridge-rails, it was not uncommon to find railsbroken into two, or more lengths, by constant deflection, and theywere repaired by placing a plate beneath the fractured ends.With regard to the cost of repairs, the mere packing of thesleepers, under equal weights, should be less in the longitudinalsystem, because there is more bearing surface ; but to set againstthat, the wood is more crushed than in the transverse system withchairs, and the access to the bolts is very troublesome, while thefastening of the double I rails in the chairs, by wooden keys, is verysimple, though in the joint chair, it is insecure, and the rising of theends is a common cause of accident. The author, therefore, in theyear 1848, suggested a new variation, combining the use of thedouble I rail with th* longitudinal sleeper, so as to avoid deflectionand to obtain the continuous bearing. Dividing the naif balk oftimber, about 7 inches by 14 inches, into two parts, the double Xrail was half-grooved into each, and bolts being passed through,below the rail, it was held as in a continuous v* ooden vice, embeddedto the upper lip, the flanche of the engine-wheels being cleared by ashallow groove. Four lengths of this were laid down on theEastern Counties' line, where it appeared to answer well, ai d is stillin good order.With respect to the durability of timber-sleepers there appears tobe some difference of opinion, arising, probably, from the variousdescriptions of timber employed. There have be*n two piincipalsystems employed, for rendering timber artificially durable. Theone was by saturation with certain chemical soluiions, or mineralsalts, which act chiefly on the albumen, produc ng a kind oftanning effect, but by which the mechanical strength of the timberis stated to be impaired ; moreover the salt being soluble in water,is apt to be washed out again. The other system, now verygenerally employed, is by * creosoting,' or injecting coal-oil, underpressure, by which both mechanical toughness and chemical dura-bility are produced.Engineers have rectmlv shown a dispositon to resort to othermaterials, and it has been consid red, whether it would not bepossible to employ stone, which is chemically and mechanically themost durable material, a».d would afford a perfect foundation, pro-vided the rails could be properly attached to the blocks.If the thickness of these blocks was reduced from 12 inches toDigitized by GooglePERMANENT WAY116 inches, the superficies remaining the fame, and they were laidbeneath each line of rails, nearly touching each other, and so asto support longitudinal sleepers, 8 inches by 6 inches deep, withbridge- rails, bolted down between every joint of the stones, throughrails and timber to a small block below, a continuous bearing wouldbe formed of chemically-durable sleepers, equal to 60 feet of surfacebeneath two rails, whilst a sufficient amount of elasticity would begiven. At the same time it is probable, that no other ballast wouldbe required, as the extended bearing surface would prevent dis-placement, and a true permanent way, when once settled and undis-turbed by the rolling loads, would thus be formed.Of course, the question of the cost of stone must depend chieflyon locality. In some districts it might be desirable to use cast-iron, or slate-flags, or the slag of the iron furnaces might be runinto moulds. On this plan any kind of rail might be used ; forin-tance, a deep foot-rail, with a half-foot angle-plate, boltedlaterally to its break-joint, to save timber, or a girder-rail betweentimbers, or probably the double I rail bedded between two timbers,as before described.The objections to the use of timber-sleepers, rendered very pro-minent by the inferior wood used in transverse structures, inducedMr. Reynolds, in 1836, to propose inverted troughs of cast-iron,lined with wood. 1Subsequently, in 1846, Mr. Greaves, of Manchester, introduceda new metal sleeper, possessing several useful properties. (Figs. 3,4, and 5.) This sleeper is a semi-spheroidal bowl of cast-iron,Fig. 3.Greaves Mir I ace- parked Sleeper*, with Douglas' fish-joint Chairs.1 Vide Trans. Inst. C.E., vol. ii., 1838, page 73.Digitized by Google12PERMANENT WAY.Fig. 5.Greaves surface-packed Sleepers.having the chair, of any form, cast upon its apex, and is connectedfirmly with the opposite slteper, at intervals, by a transverse bar,which preserves tne gauge and retains the travelling table of therail at the proper angle. The joint sleeper has Douglas* fish chairapplied to it ; this is a double-headed chair, with loose fishingplates of cast, or wrought iron, attached to the rails by split ironkeys.There are several advantages in this system ; the form of thesleeper is t-trong, — it holds well in the ground ; — the chair is notliable to be detuched, — the whole bearing surface is directly beneaththe load, — the ballast is always kept dry and elastic,— and there is .an extremely ingenious contrivance for packing it through twoholes, with a pointed rammer, from the surface, so that the sleeperand rail can be forced upwards, without disturbing the general bedof ballast, or they may be lowered by taking out a portion from theinterior. The danger of admitting water, by opening the ground, isthus obviated, and in certain situations this is of great importance.The sleepers are in use on several lines in Great Britain, on theContinent, and in the Colonies, and Mr. Robert Stephenson, M.P.,V.P., h.js adopted them for the Egyptian Railway.The next kind of cast-iron sleeper was that introduced by Mr.Peter W. Barlow, M. Inst. C. E., in the year 1849, on the SouthEastern Railway. (Figs. 6 and 7.) Each sleeper is in two pieces,and consists of a plate 3 feet long by 7 inches wide, with two half-chair heads cast upon it, except that for the joint-chair, which hasDigitized by GooglePERMANENT WAY.1.1Fig. 6.Mr. P. W. Barlow's Cut-iron Sleepers.Fig. 7.three half-chair heads on it ; these are made to grip the lower halfof the rail tightly, by means of screw-bolts, passing through thechair-heads below the rails, and holding them as in a vice. Thecasting of these plates so as to fit the rail perfectly, and to preserveall the metal in a state of rest, is a work of some nicety.Another cast-iron sleeper was brought forward by Mr. WilliamH. Barlow, M. Inst. C. E. : it was very similar to the former, exceptthat it was cast in one piece, the rails being fixed in the chairs bykeys, in the usual manner, or by resting the rail on a wooden block,keyed in below it. 11 These sleepers are in use, and are much approved, on the Londonderryand Enniskillen and Londonderry and Coleraine Railways ; and since thepaper was read Mr. W. H. Barlow has adopted, with excellent results, on theMidland Railway, a form of cast-iron sleeper, combining the best propertiesof Mr. P. W. Barlow's sleeper, with the suspended fish-joint. — Sec, Inst. C. E.(Figs. 8 and 9.)Fig. 8.Mr. W. H. Harlow's Cast iron fish a>int Sleeper, as used on the Midland Hailway.Fig. 9.Digitized by Google14PERMANENT WAY.In both these plans, it is evident, that the rail must be sufficientlystiff in itself, to prevent deflection, as the sleepers are not con-tinuous. Mechanically considered, the latter plan appears the best,as the wood-keys can yield slightly and equalize the deflection ofthe rail, whereas, any bending of the rail, in the other plan, wouldbe apt to break the sleepers.The cast-iron sleeper introduced by Mr. Samuel, M. Inst. C. E.,is a wedge-shaped trough, deep enough to take the total depth ofthe rail, with two inclined side wings, at the upper part, to bearon the ballast. The rail is grooved into two pieces of timber,which clip it at the bottom and the sides, and being forced into thetrough, the wood is compressed. The troughs are, however, notcontinuous bearings, and therefore the rail must be strong enoughnot to deflect.This is, mechanically, the best arrangement of the cast-ironsleepers hitherto examined ; but as it is evident, that the retentionof the rail in the timber cushion depends on the tenacity of the cast-iron trough, it must be made very strong, to prevent the wedgingaction from splitting it across the bottom. It is, however, a defec-tive arrangement to make the bearing wings pitch downwards, as itinduces unsteadiness ; they should rather have been reversed, if notmade horizontal.The cast-iron sleeper, introduced by Mr. Hoby, like the sleeper ofMr. Samuel, is a trough, though of a different shape, the internalsection being that of a prolonged chair, into which folding wedgesare driven laterally, to secure the rail in its place. The side-bear-ing wings are pitched downwards, and the sleepers are not con-tinuous. The rail fits in the bottom of the iron trough, and alsoagainst one side, and it will require very nice moulding and rollingtosecure an accurate fit, but the elastic wood wedges will reduce thechance of fracture.Another variety is an imported French sleeper, consisting of an,ordinary chair, cast solid on a plate of 1 8 inches square, and merelyresting on the surface of the ballast.Another variety of sleeper is that of Major Cochrane. It is afiat cast-iron plate, with a pair of ribs upon it, forming a channelabout two inches square, filled with timber, on which a flat-bottomed rail is held down by wedges, driven into projectingchair- heads, in the usual manner : it is evident, that if the raildeflects, the timber must be crushed, as it presents a very smallbearing surface, and a rail with sufficient depth would possess avery unsteady base, unless the sleepers were cast in the form of atrough.Digitized by GooglePERMANENT WAY.15Cast-iron may be used, efficiently, for railway sleepers, but themasses must be larger than have hitherto been applied, and theform must be continuous ; or if detached, the sleepers should onlyserve as piers, or points of support, in which case the rail must benon-deflecting.Having examined the different kinds of sleepers, with referenceto the vertical and lateral support they afford to the rails, the nextquestion is that of the various modes of attaching the rails to thesleepers, in order to prevent the jolting and unsteadiness usuallyexperienced on ordinary railways.About the year 1832, or 1833, a reward of £100 sterling wasoffered by a railway company, for the best mode of securing rails tochairs. Many plans were sent in, but none were integially adopted.It was said, that an amalgamation was effected between three of thecompetitors, and the reward was divided.The fastening so rewarded was not successful, and it was soonfound, that the wooden key was the only arrangement that did notrapidly get loose. Many of these, however, were too small ; therails were loosened laterally, and a creeping motion was induced bythe action of the breaks on the wheels, not unfrequently causingaccidents.The mode of fastening the chairs to the sleepers, by means of ironspikes was insufficient. Rust got in between ; the two irons werethen rapidly worn away, and all became loose. This gave rise tothe improved chairs of Messrs. Ransomes and May, hardened inchills to prevent the rails from crushing them, but not preventingthem from crushing the rails, and with large-sized compressed tre-nails of oak, to attach them to the sleepers. (Figs. 10, J 1, 12, 13,14, 15, and 16.)'Fig. 10.Ransomes and May 8 Chair* and compressed Wedges and Trenails.Elevation of Chair, showing the inclination of the Rail.1 Vide Minutes of Proceedings Inst. C. E., 1842, vol. ii. page 72.*Digitized by Google16 PERMANENT WAY.Fig. 11. Fig. 13.JOINT CHAIR. JNTEKMKDIATE CHAIR.Rails, Ch.iw, and Sleeper*, with the « cramp gauge' fixed.Digitized by Google1'EltMANENT WAY. 17Fig. 16.A simple mode of preparing oaken trenails has since beenproposed. The ordinary round trenail requires to have the holebored very accurately in the sleeper, to prevent splitting. Thenew mode is, to cast in the chair a trenail hole of an oblongform, say 1$ inch by 2 inches in the direction of the grain of thesleeper, the upper side of the hole being as much larger than thelower, as is required to form the head of the trenail. A piece ofstraight-grained oak, not over dry, is driven into this hole, and iscompressed, and the sap forced out in the act of driving. A hole1± inch in diameter being previously bored in the sleeper, the tre-nail is driven into it, and cannot split it, because it is of less widththan the hole across the grain. But it drives very forcibly iu thedirection of the grain, precisely in the mode of a chisel-shaped brad,which forms its own hole without splitting, and it offers the greatestamount of resistance in the direction of the lurch of the engine.If chairs continue to be used, it is probable that this method willbe found advantageous, as the square trenails can be prepared withgreater facility than the cylindrical ones.The wood keys should also have ample bearing surface, so as notto be crushed with the side-lurch of the engine.Apart from the question of securing the rails to the sleepers, isthat of the best mode of securing the ends of the rails to each other,so as to approximate to a continuous bar.One of the earliest attempts to obviate the evils arising frominsufficient connexion at the rail-joints, was on the BlackwallRailway.In that case the rail-ends were connected together by a ecarf-joiut, about 6 inches in length, with the poiuts dovetailed, and thewhole wedged into the chair. As these rails were only used withcDigitized by Google18PERMANENT WAY.Fig. 17.rope- traction, and were not tested by locomotive work, for whichthey were not intended, no deduction can be drawn from them ; butthey were weakened by the scarf, and were expensive in manu-facture.On examining the joints of a railway out of order, it will bofound, that the forward end of each rail, in the direction of the tr^ffi' ,stands, apparently, slightly higherthan the hinder end of the rail infront of it (Fig. 17). When theweight comes on it, so as to press itdown to the chair, it becomes thelowest, as the under part has beenhammered in the chair, and the railis reduced in depth When oncerails are thus hammered, it becomesSunk sleeper. impracticable to secure them firmlyin the chairs, except by cutting off their ends.The joints of the rails are evidently the weakest parts of the lineof way ; therefore the great object is how to strengthen them ; andthis cannot be done by taking away metal from the already tooweak part. Supporting the joint by extra sleepers appears a simpleremedy, but the sleepers are not practically supports, as they onlybear on loose ballast, and not on any continuous surface. It becomesnecessary, then, to join the rails firmly together. The operation of* fishing,' by the addition of pieces of metal to each side, then pre-sents itself ; but to secure these * fishes,' becomes the next question.The simplest mode is to take the rail-joint wholly out of the chair,and to place a sleeper, with a chair on it, on each side of the joint,thus suspending the joint between the two chairs. The * fishes'being then driven in on each side of the rails, the rails and chairsare connected together. This was practised by the Author in 1847.The damaged rails were thus supported by the chairs in theundamaged part, and they were made practically continuous, whilepermitting expansion and contraction, by the sliding of the fishes.Hut putting the fishes into the chairs involved the necessity for freshcastings, and it was found a simpler operation to punch holes in therails, and connect the fishes together by four bolts passing throughall, the holes in the rails being larger than the holes in the fishes,to permit of expansion and contraction. The first fishes tried wereof cast-iron, and they auswered very well ; but subsequently wrought-iron was substituted, the edges of the fishes bearing against the topand bottom of the web of the rail, leaving the middle untouched, sothat the bolts were held as it were in spring tension endways, andDigitized by GooglePERMANENT WAY.19therefore did not work loose. So perfect was this arrangement, thatold rails, which had become useless, when battered at the ends, sothat one end projected nearly a quarter of an inch above the other,when once firmly connected by the fishes, gradually rolled, under theworking of the trains, to a perfectly level surface.This plan did not, however, get into general use, and the Authorhaving many other subjects to occupy him, the matter lay in abey-ance, but some experienced engineers, convinced of the efficiency ofthe system, became possessed of a portion of his interest in it, and havesince been enabled to extend its adoption, with good practical effect,entirely confirming the soundness of the original views relative to it.The application of the * fishing- plates ' to ordinary rails is verysimply and rapidly effected, and the results are very satisfactory(Figs. 18 and 19).The next proposition for improving joints was by ProfessorGordon. It was, to make a horizontal lap, by cutting away 3 inchesfrom the top of one rail, and 3 inches from the bottom of its neigh-bour, to half the depth of each, and then lapping them togetherFig. 18.PUhing-pktes applied to ordinary Raila.C 2Digitized by Google20 PERMANENT WAY.Fig. 19.Enlarged Section of Rail and nsbingplatea.and securing them in the chair. It was expected that the hinderrail in the line of progress, thus lapping over the forward one,would keep both down together. But whether from bad iron, orfrom the diminished strength of the parts, it was found that theends broke off, and the plan was abandoned.At length, in the year 1849, the authorities on the EasternCounties Railway resolved to try the experiment of the ' fish-joint/ Some half-dozen rails were jointed with cast-iron fishes,and they were found to answer.Many persons went to examine them, yet though all appearedsatisfied, the railway companies and the engineers exercised a wisediscretion in waiting till experiment hail become experience. Pro-gress under such circumstances is slow, and, moreover, as thetrial seemed to throw a new light on the question of good joints,many new experiments were made.Mr. J. Fowler, M. Inst. C. E., produced a chair-joint, (Fig. 20,)formed of three pairs of jaws cast together upon a sole-plate, the twoextremities resting upon two sleepers below : it is said to answerwell on the East Lincolnshire line, and has been in use theresince 1849. Various other plans also were tried, or suggested.Fig. 20.Fowler's Joint Chair.Digitized by GooglePERMANENT WAY.21In applying tlie fish-joint to the repair of rails on lines, whenfunds had grown scarce, an objection arose to the cost of an extrasleeper and chair, in addition to the fishes. To meet this objection,Mr. Samuel contrived a modification of the 'fish/ (Figs. 21, 22,and 23.)Pig. 21.SAMUEL'S FISH CHAIR.Side View.Fig. 22.Plan.A chair was cast with only one jaw, made to ht against one sideof the rails, a wrought-iron fish being placed against the other side,and all bolted together through the rails and fishes. This requiredonly one sleeper, and the value of the modification of course resolvesitself into a question of bearing surface on the ballast. If oneDigitized by Google22PERMANENT WAY.sleeper be sufficient there is no use for two, but it would be anunwise economy to save sleepers and to expend more than theamount in maintenance of way.In the original specification of the fishing plan, it was foreseen,that many evasions might be resorted to, and various modes ofaccomplishing the same object were suggested and shown. Double-headed chairs were specified, and also the mode of casting the chair-heads to fill the channels of the rails. A plan of splicing rails wasalso provided for. It was to roll the rail double, with a verticaljoint from top to bottom, and to fix them together * break-joint,'t. e., with the joints of one pair of half rails, abutting on the solidportion of another half rail, so that the rail would practically becontinuous, the rails being riveted, or bolted, or fished togetherlaterally. It will of course require more iron to construct rails onthis plan, than in the solid form; but, though hitherto unused inEngland, the system has been rather extensively used in the UnitedStates and in the colonies, and is frequently preferred to otherarrangements, the rail being very quiet, the abseuce of soliditypreventing any ringing noise.An unwise economy has generally been practised, in restrictingthe quantities of Railway material, or fixed plant, though there hasbeen no such economy in rolling stock. At the outset of railways,iron was the most costly commodity, and all methods were sought tolessen its consumption. This idea was continued as a kind of super-stition, long after i>ou became cheaper, so that at the outset ofthe Author's experiments, he found, that to increase the weightof the permanent way was considered a kind of misdemeanour, tobe avoided by any means, and a second chair to a joint was held tohe a great extravagance, yet almost immediately on the productionof the fish-joint, iron seemed to increase in quantity, and its cost tolessen in many ways. The quantity of cast-iron sleepers used onthe South Eastern and the Midland Railways, appeared to haveoverthrown the superstition, and timber was suddenly pronouncedto be wholly unfitted to be classed amongst the materials of per-manent, way.Occasional breakages of cast-iron sleepers, owing to the desire tolessen the first cost of material, and thus unduly reducing its propor-tions, caused attention to be turned to the capabilities of wrought-iron.The saddle-back rail of Mr. W. H. Barlow, previously alluded to,(page 6, Fig. 1,) was the first result. This rail is usually 13 inchesin width, and 5f inches deep, and weighs 126 lbs. to the yard. 1 It1 The weight now generally used is 90 lbs. to 94 lbs. per yard.— Sec. Inst. C. E.Digitized by GooglePERMANENT WAY.23is stated, that « it will bear a pressure of 40 tons without spreadingout in width, and of 27 tons between bearings, 4 feet 4 inchesapart, without its elasticity being injured ; the stiffness of the railvirtually increases its bearing surface, for if a weight be placed overthe centre sleeper of a length, and the rail be flexible, the whole ofthat weight will be borne by the sleeper, whereas, if the rail berigid, the weight will be equally distributed over the adjoiningsleepers."If this rail be as rigid as it is described to be, with a proper ex-tent of bearing surface, and is sufficiently heavy to remain quies-cent under the rolling loads, there ought to be no cost of " mainte-nance of way," but only the expense of renewal from time to time*But upon examination it appears, that though the rail is 13 incheswide, this is not all true bearing surface. If the trough, orhollow, were filled with adhesive ballast, that could not leave it,the whole width would be bearing surface, but it appears to lie onthe ballast as a saddle lies on a horse's back. It is stated, that con-siderable noise is experienced in running over this rail. If so, thiswould indicate a hollowness. In dry sandy ballast, however wellthe hollow of the rail might be packed, it must always be fallingaway from the upper part of the trough. Then again, the sidesurface to resist displacement, by the lurch of the engine, is but 4inches, in a wedge form, tapering down to a sharp edge, like theform of a plouglishare. This want of hold must render the processof packing very difficult.It is stated, that " in this wrought-iron rail, there is a sensationof hardness, which appears to result from the rails ha\i ng been laiddown a short time, and not having been worn to sufficient evennesson the upper surface." There can be no doubt, that the mere rough-ness, from the rolls, on the surface of rails, is sufficient to producepositive blows at high speeds, and were it practicable to producea smooth surface on them, there would be a great advantage in it,not merely by the even face, but by hardening. But assuming thesaddle-back rail to be a rigid girder, another consideration ensues.It is stated, by experimenters, that the limit of weight whichcan be borne by the best wrought-iron, before it begins to becrushed, is 1 1 tons to the square inch. Rail-iron has been stated tobe only equivalent to 8 tons per square inch. But the impact of adriving-wheel with 7 tons weight on it, is only a point, and underfavourable circumstances a line, unless the rail deflects and partlylaps round the wheels. If this be correct, it follows, that a railought to laminate and roll out under heavy engines, as has actuallyoccurred in many instances. Though it must be stated, on theDigitized by Google24PERMANENT WAY.other side, that there are many rails which have been down foryears under heavy engines, and are still not injured, probably escap-ing- lamination by elastic deflection.If it be resolved to continue the use of inflexible girder rails, oneof two methods must be resorted to. The first is, to make the railof steel, or to steel the surfaces, and the other is, to increasetiie width. As the weight on the engine is greater than that on thetrain carriages, or waggons, it might be desirable to construct therails with two surfaces, the double surface for the engines, and thesingle surface for the train, with a channel rolled between the twofor the holding down bolts, which will be presently alluded to.Reasoning on the absolute necessity, in regard to economy ofmaintenance of way, of obtaining an inflexible rail, that will notcrush either the ballast, or the sleepers in detail, the questionchanges, from Mr. Barlow's rail to his mode of connecting hisrail-ends together, for if the connexion does not produce equalstrength with the other portions of the rail, mischievous resultsought to ensue, unless the middle of the rail be three times thebreaking, or deflecting strength, and the joint only twice, and thent here is a waste of one-third of the strength of the rail.The mode used is, to place below the two rail-ends a saddle ofwrought-iron, about 2 feet in length fitting the trough. To thissaddle the two rail-ends are riveted firmly at the outer edges, with-out any provision for expansion, or contraction. The upper por-tions of the rails abut closely together, and thus the lower part is intension, and the upper part in compression, and the two togetherform a beam, the strength of which must depend on the power ofthe rivets to resist being sheared off.Supposing the upper, or abutting portion to be removed, thelower portion with the rivets would be apt to bend under a load,consequently the strength of the joint must depend on the abuttingends and the strength of the rivets. If, therefore, there be anyexpansion, the rails must elongate through the whole length, andslide over the surface, or they will buckle upwards, or sideways.But it is said there is no expansion ; and if so, it is possible thatthe contact of the earth carries off the heat as it is generated, justas water carries off the heat applied to the bottom of a boiler.It would appear clear, that when the upper portion of this rail isworn by the traffic, a very large quantity of iron will have to beremoved, and renewed. This of course is a matter for commercialconsideration on one side, and of mechanical convenience on theother.Assuming that, for certain purposes, the solid rail might beDigitized by GooglePERMANENT WAY.25desirable, it remains to be proved that the saddle-back is the bestform, either for strength, or for packing. Lateral and vertical stiff-ness, with large bearing surface vertically, and security against lateralshifting, are the objects sought for ; and with these points in view,the T form section, with the rail, or rib, on the upper surface, appearsbe well adapted for general use. In this form there is a bearingsurface horizontally of 14 inches in width, the total depth is 9 inches,and the depth below the horizontal surface is 6£ inches, giving greatvertical stiffness, and serving as a keel to prevent lateral movementin the ground, the rail being used without any sleepers. To formthe joints, two cheek-plates, about 2 feet in length, might be boltedto the vertical and horizontal portion, with allowance for expansionand contraction in the holes through the rails, which are madelarger than in the cheek -plates. This rail would lie very firmly inthe ballast, and the packing from the surface would be very easy,the vertical web forming a solid back to ram the ballast against.If it were required to key more firmly in the ground, a timber mightbe bolted to the lower edge. It might be objected that there wouldbe a difficulty in rolling this rail, but practice decreases those ob-jections daily ; in fact, there is, practically, no mechanical difficultyto large demands. It might be objected to this rail that it mightbe noisy, but probably it would be less so than a trough-form rail.If a quiet rail be required, without vertical deflection, it may beproduced by rolling an upper table with a deep thin vertical web,and bolting it laterally between two timbers. To secure adhesionbetween the timbers, the sides may be rolled to the form of theplanking of a clincher-built ship, multiplying in that mode horizontalbearing surface on the timber, or a T~f° rinea ' rail may be used.A webbed girder-rail may be used with transverse sleepers, with-out chairs, by adding to a foot-rail, a deep web, to be inserted tothe depth of 3 inches in the sleepers, so that the feet rest horizon-tally on the t-urface of the timber and are fastened down by dog-head spikes. At the joints the rails might be attached together byplates fitting into a longititdidal groove rolled in the web (fig. 24).This rail, whose upper table would be only 2£ inches above thesurface of the sleepers, would possess great vertical and horizontalstiffness, and would be a cheap and serviceable form of railway.Solid rails, such as have been described, might be difficult to bendfor curved lines, on account of their width ; but by making a girderrail of several pieces, the difficulty may be overcome. A flat plate14 inches wide mav be bolted between a bridge, or foot-rail, at thetop, on a T-girder below, the bolt-holes in the flat plate being dis-posed in the curved line required, and the rail and T"gprd«' bent toDigitized by Google20PERMANENT WAY.Fig. 24.Webbed-foot Rail.it. To make the joints, cheek-plates may be bolted laterally to thegirder, and the rail break-joint with it. The bolts require to bestrong, to make the girders firm.Another variety of girder-rail is formed by a bridge, or foot-rail,bolted on two wide and deep angle-plates which break -joint witheach other and with the rail, and thus no cheek-plate is requiredat the joints. This rail would be quiet, the division of parts pre-venting ringing. Many other varieties of form have been devisedfor particular localities.The question of adhesion must be viewed from two differentaspects. In driving-wheels, the maximum of adhesion, up to thepropelling power, is required. In all other wheels adhesionshould be got rid of altogether. The Author is not aware, that anyexperiments have been made, to ascertain whether the conical formof the periphery of the driving-wheels assists, or retards propulsion ;but it is clear, that if a conical-formed tire presses over the wholewidth of the rail, with equal intensity, there must be a universalgrinding action, which, even supposing it to be beneficial in the caseof the driving-wheels, must be mischievous in all other cases. There-fore it is clear, that steel rails and cylindrical peripheries aremechanically desirable, in one point of view, for all bearing-wheels,unless other circumstances intervene, which render the friction alesser evil in comparison with something else. That a considerablefriction does take place, involving the rubbing of tires on rails,may be judged by the fact, that the rails in use, present a burnishedappearance, and that after the passage of a train particles of metalresembling black-lead dust are found on the rail surface.Experiments to determine this question of friction and adhesionon firm inflexible rails of iron and steel, and with conical andcylindrical wheels, are much wanted, and would prove valuabledata for railway companies, involving as they do the question ofdiminution of cost of haulage and of maintenance of way.Digitized by GooglePERMANENT WAY.r 27There is another subject, in railway maintenance, which appearsto have been overlooked. Provision has been made for drainage,but not for drainage in immediate contact with the rail. It isknown by experience, that the vibration of passing trains commonlyopens a thinner, or thicker air-space, between the vertical edges ofthe sleepers and the ballast, when the weather is dry, and that intothese portions of the way rain finds its way as to a drain. Thesleeper being absolutely loose from all lateral adhesion, springs upand down beneath the trains, and packing is required. The oozingof the fine washed sand from the sides of the sleepers points this outto the way-men. Moreover, even in continuous rainy weather, theballast being an absorbent material, or sponge, and the sleeper non-absorbent, the greater mass of water will be found at the point ofcontact, precisely as the water of gravel-beds is intercepted by aclay stratum.Impressed with this defect, the Author was led to consider thepracticability of getting the drains close to the sleeper, and the mostobvious mode was, to make the sleeper itself the drain, by construct-ing it of perforated iron, or supposing iron rails to be used withoutsleepers, to make the rail itself the sleeper. The plan proposed toaccomplish this is to roll a rail with a deep thin vertical web ; twoangle-pieces of cast, or of sheet iron, bent to the requisite form arebolted to the rail above and below, breaking-joint ; these angle-plates are perforated, and the water will find its way through theperforations, precisely as it does in ordinary clay drain-pipes, andmay be led away to the exterior at convenient intervals. Theform of this rail will give perfect rigidity ; and if perfect rigiditybe compatible with durability of the running surface — a point notyet satisfactorily made out — and if the bearing surface be in excessof the load to disturb it, it ought not to require any maintenanceof way. In using the term rigidity, it must not be understoodas applying to a torsive strain, such as an engine flanch wouldtend to produce, as it would have a slight elastic yielding in thatlirection. The packing of this kind of rail would be exposed tothe same difficulties as ordinary longitudinal, or transverse sleepers,from the depth of bearing below the surface ; but this might beobviated by reversing the angle, and placing the flat above, as in thetriangular sleepers of the Dover line. Or the two sides might becurvilinear, for the purpose of inducing more stability, and prevent-ing rocking on the ballast.A copious use of water would do away with the annoyance ofdust, and be economical for maintenance of way.Supposing a rigid rail to be impracticable, on account of lamina-Digitized by Google28rEHM A NKNT WAV.tion, it will become necessary to devise a mode of producing yield-ing, or vibration of the upper surface, without in any way disturb-ing the portion which rests on the ballast. For this purpose, theAuthor would propose to use a rail with a deep thin vertical web,bolted so as to break- joint, between two angle-plates at the upperportion, as before described, but with a space between the bottomof the rail and the bottom bearer. If the rail deflects, it must, onaccount of its great depth and fastenings, deflect only on a con-siderable length, and that without disturbing the bottom. If thepressure be excessive, it will take a bearing on the bottom and willin fact be a spring-rail. The bottom may be of cast-iron, but inall cases it will be sufficiently solid to prevent its yielding in thecentral part.The girder-rails, or sleepers, thus described, are in common withthe trough-rail subject to a defect. They are mostly of wrought-ironand therefore liable to oxalate. If this be a serious objection, itmay be remedied by a coating of zinc, or other metal, or by bitu-miuizing, or covering with mineral paint, or coal-tar.How far such a principle of rail might be available as a pipe forthe transit of water, for other purposes than watering the road, is amatter for consideration.In the American Union the railways run through the streets oftowns. It has hitherto been found difficult to construct a line ofthis kind, so as to bear the rough usage of carts and waggons thatrun over the rails in every direction. Some of these lines are con-structed of heavy cast-iron, and cost as much as 24,000 dollars permile, equivalent to nearly £5,000. It would appear probable, thatsolid girder-rails of wrought-iron, such as have been before de-scribed, would be found to answer the purpose.With regard to the applications of " permanent way " on the Con-tinent, two patterns appear to have been set. One, the Americanpattern of the foot -rail and longitudinal balk, fastened by doijs, whichprevails very commonly in Germany ; the other, the English doubleX rail and chair with transverse sleepers. In France, the Englishsystem is the general type. But the American system has scarcelybeen u*ed in England, save by contractors, for temporary work.It has, however, been stated, that longitudinal timber sleepers onstone foundations, have been found to form the most really per-manent way of any used in the United States, where " maintenance,"on account of the hijih price of labour, is a very costly item.In speaking both of the longitudinal and transverse sleepers, thecombined systems, partially introduced, have not been alluded to.In truth, the rail being rendered non -deflecting, the other consider-Digitized by GoogleTERMANENT WAY.29ation is, how can it best be secured to a sufficient mass of bearingsurface on the ballast, to secure it against sinking ? With a non-deflecting rail, it is a matter of no consequence, whether the bearingsurface be equally distributed, or be placed upon sleepers, like piers,save that, the farther the bearings are apart, the stronger andheavier must be the rail and the sleeper, or pier foundations.In discussing the durability of way, the question of constructingthe levels and the inclines, forms an important item.Supposing a line to be required to be worked by fast, or expresstrains, not stopping, and by a number of other trains, stopping atvery frequent stations, the plan is occasionally resorted to, of makingsidings from the main-line, at the stations, in which the stoppingtrains may pull up, while the express trains may run by unimpeded.Iti starting the trains from the stations, great surplus power isrequired, and in arriving, great damage is done by the friction ofthe stopped wheels on the rails. But if it were practicable so toarrange it, that the main line might be on the level, while thesidings, or stopping stations, were on an elevation, less engine-powerwould suffice to start and to work the trains, and no break-powerwould be required to stop them. The artificial levels might easilybe arranged, by constructing the stations on brick arches alongsidethe main line, and letting the siding road run over the arches, whichwould form summit levels, the offices being below. And bya proper disposition of rails on elastic timber, the vibration mightbe sufficiently diminished, bearing in mind also that no rapid move-ment would exist on the immediate summit.The question of permanent way in Egypt, India, &c, involvessome new consideratiotis. Railways, in themselves evidences of ahigh state of civilization, are unquestionably, calculated to becomethe most effective implements for bringing semi-barbarous countriesinto a state of civilization. But as railways, in such countries,must pass through wild districts, thinly peopled, where fences andpolice cannot be maintained, and where, in many cases, the wholeline must, in fact, be a level crossing where most people are pil-ferers, it becomes necessary to revert to such means as may placethe railway in such a state, that it will watch itself in considerablelength, without needing fences, or police, by having nothing prac-ticable to pilfer.In Egypt, up the valley of the Nile, there can be no need of theseprecautions, lie Mould be a bold thief who would venture on ap-propriating any railway material belonging to Abbas Pasha. Butthere are other circumstances to contend against. The dry heat ofthe atmosphere is fatal to timber, and the soil along which theDigitized by Google30PERMANENT WAY.Egyptian line will be carried, will vary from the extreme moistureof irrigated land to parched dust. Therefore, the deeper the foun-dations of a discontinuous sleeper road can be placed, the betterchance there is of their remaining firm. What the effect of Mr.Barlow's saddle-back rail, or of girder-rails, would be, in such aclimate, with an intense sun by day and with condensation bynight, would require some careful experiment to determine. It isprobable that ample provision for expansion and contraction wouldbe required. In this case, a hollow girder-rail, with a free currentof air through it, would be the best adapted for maintaining anequal temperature. Any of these kinds of rails, without timber,and with few parts, firmly fixed together, would offer little tempta-tion to a marauder.In hot climates, well peopled, when once introduced, railways willmultiply as a means of health. They afford the most efficient meansof obtaining the coolness, essential to keep down fever, and howeverstagnant the atmosphere may be, a locomotive at speed can alwaysproduce on a train of passengers the effect of a cooling breeze.Such will eventually be the result of railways in the East. Asthe sea-breeze is called the * Doctor * in the West Indian Islands, sowill the railways of the East acquire similar names. The wealthywill make moving houses on the eastern railways, for the sake of anartificial climate wherein their jaded faculties may be restored.In the flat parts of India, there are two evils to be guardedagainst ; the flooding of the ground at seasons, which might float upa line, in which much buoyant timber was used, and the other isthe white ant. It is doubtful, however, if the white ant wouldnot find creosote disagree with his taste, quite as much as the teredonaval is finds it impede his tunnelling operations. It is apprehended,however, that all timber, in arid climates, would require to be wellcovered with ballast, to prevent injury from the hot cokes, fromthe ash-pan.If broad flat stones could be used without ballast, and at a smallcost, they would form a most valuable foundation ; but stonescannot be used without timber, and fastenings, and therefore itseems probable, that continuous wrought rails of the compoundgirder form, may be best adapted for such circumstances; and ifthey be made wide enough to take an efficient bearing, on ordinaryground, as plank roads are made in America, they may be found thecheapest structure, and at the same time, a structure that manufac-turers could turn out complete from their establishments, and whichwould require no skilled labour to apply.It must be received as an axiom, that even where wood isDigitized by GooglePERMANENT WAY. 31plentiful, it is not a cheap material. It requires much skilledlabour to apply it— unless the form reduces it to machine work —and if the wood be hard it is very costly, and if soft, it has littledurability, save by creosoting.Girder-rails of sufficient size will probably be largely used in ourAustralian colonies, where, as in America, they cannot afford to paypolicemen, and they require a structure, not easily damaged by care-lessness, or wilfulness; a structure. also, be it said, that affords pecu-liar facilities for moving from one district to another, if by changeof circumstances a more eligible route were found, for be it remem-bered, that in such districts the bare i way,' and the * siding/ constitutealmost the whole property of the railway.The cost of the various kinds of structure examined, has not beenhere entered into, because it must vary with the locality, and thequality of the material ; and the structure which is cheapest in oneplace may be dearest in another. It is also well understood, thatcost depends chiefly on quantity, in a material practically withoutlimit, such as iron.An apology is perhaps due for the prolixity of this paper, reca-pitulating many things familiar to engineers ;• but when reflectingthat the permanence of way must depend precisely on the details ofits structure, and the care that is taken to guard against deterio-rating causes, and which may sometimes make all the differencebetween a dividend and a loss, those who examine deeply will notconsider the prolixity a waste of time.In some of the views which he has taken, or rather in the elabo-ration of principles which must have occurred to all who have hadtime to think, the Author would disclaim all captious criticism.His object has been to arrive at mechanical truth, and to indicatein what direction further experiments should be made, in further-ance of railway progress, which is another result of the in>tinct thathas prompted the great Anglo-Saxon race to push steadily forward,in this great work of civilization. For in truth the road is the greatcivilizer of humanity. People only quarrel with those they donot know. Give them access to each other and the quarrels cease.It is a marvellous instinct that prompts this law-loving nation and itsoffshoots to the work of road-making, and if he who makes twoears of wheat grow, where only one grew before, be worthy of anovation, what shall be said to those, amongst the noble array ofengineers, who will make two lines of railways at the cost of one ?The absence of roads is isolation, and if by some strange chanceEngland should be deprived of her roads, and be prevented fromreconstructing them, another generation might behold the inhabit-Digitized by Google321' Kit MA N KM WAV.ants reduced again to the conditiuti in which they were found byJulius Caesar.From this state, roads have delivered us, under the disputationof Providence, giving latent faculties to man. The most |>erfeetroad yet known is the raihoad ; and no higher service cau be done tohumanity than making railroads still more perfect, thus removingthe chief obstruction to man's universal interchange of ideas, andthe general spread of ihe benefits of civilization.The paper is illustrated by a series of diagrams from which thewoodcuts have been compiled. 1Mr. W. B. Adams described the different sections of rails, ofwhich diagrams were exhibited, and directed attention to a girder-rail of simple construction, hollow, so as to preserve as nearly aspossible a uniform temperature, under the extreme thermometricvariations between day and night, or even to serve as an occasionalconduit, and also as a conduit for water, which latter, it was con-tended, might be advantageously employed in preventing the cloudsof dust from which the passengers now suffered inconvenience, andat the same time causing more permanent contact between theballast and the supports of the rails. This, he submitted, would bean efficacious, simple, and cheap mode of constructing the per-manent way.Mr. P. W. Barlow objected, that the question of the cost of thepermanent way had been almost entirely lost sight of, in the paper ;now this was so important a consideration, in all constructions, thatit should have been fully considered, before entering on the discus-sion of the various expensive innovations alluded to by the Author.'With respect to the system of cast-iron supports, for which hewas responsible, it was necessary to state, that it had been proposed,before it was considered practicable to roll rails, either of suchweight, or of such an extended section, as those introduced by hisbrother, Mr. W. H. Barlow, and now known as the 4 saddle-backrail as soon as it was found, that such rails could be manufactured,he had advised their being tried, and he was fully satisfied with theresult. Previous to that time, however, being under the necessityof determining the construction of about sixty miles of line, he hadnot hesitated to employ the cast-iron supports, which had, under thecircumstances, fully answered his expectations. It must be admitted,however, that the saddle-back rail was, under ordinary circum-1 Several of the blocks, for the cuts, have been lent by members.— Sec.Inst. C. E.Digitized by GooglePERMANENT WAY.33stances, preferable ; a single line of way could be laid with thoserails, weighing 90 lbs. per yard (at the then low price of iron), ata cost of £900, or £1000, per mile. The portions of line so con-structed, on the Midland Railway, were very successful ; there wasan even motion in the carriages, less noise at the joints, and evidentlyless wear and tear in the rolling stock, in spite of the rigidity of therails. He apprehended, that the expense of the ' maintenance ofway ' would be considerably reduced.Mr. W. B. Adams submitted, that the question of cost need notnecessarily enter into the discussion, especially as the prices ofmaterials were constantly varying, and the quantities being given,each engineer could make his own calculation as to first cost.Mr. Brcxel said, he was so fully impressed with the necessityfor attempting to reduce the expense of maintenance of way, thathe had not hesitated to try the * saddle-back ' rail, and as far as thelimited experience of only twelve months enabled him to form anopinion, he was well satisfied with the result of the experiment, foralthough it had not yet induced the happy position, assumed byMr. Adams, of there being * no cost of maintenance of way/ therewas every indication of considerable economy in the ordinary ex-pense of repairs. The piece of railway, where the experiment *astried, was about seventy-five miles from London, and exposed to allthe traffic of the Great Western Railway, under somewhat dis-advantageous circumstances ; yet it appeared to stand well, and, atpresent, the expense of maintenance of way was quite as low asthat for the system of bridge-rail, with longitudinal timber-sleepers.It was well known, that Mr. Brunei was a decided advocate of thelatter system of construction ; he believed a line thus made to be atleast as easily kept in order as with transverse sleepers ; the jointswere as readily made and were more solid, than those of any othersystem, and the cost of the original construction was not greater.The saddle-back rail appeared to possess most of these advantages,and at the same time to offer economy in the first cost ; the joints,in particular, were rendered very even and secure, by fasteningthe ends of the rails upon a wrought-iron saddle-piece, with four,or six rivets, which united them perfectly.He was so satisfied with the general result, that he had also usedthese rail* on a short length of twelve miles, on a railway in Corn-wall, and was about to adopt them for a considerable length on theSouth Wales Railway.He must observe, that in the description, given by Mr. Adams,of the effect of traffic on bridge-rails and longitudinal bearings,results were mentioned which, during all his exj>erience of theDyDigitized by Google34PERMANENT WAY.system, had escaped his notice. The actual results of his first trialof the system were these;— the rails were too light, and beingonly attached to the timbers by a large sort of ordinary wood-screw,traversing the flanch of the rail and passing nearly through thetimber below, the deflection, or * creep ' of the rail loosened thescrews, and after being tightened down a few times, the holesbecame galled and the screws no longer held firm. They werethen abandoned and were replaced by screw-bolts with the nutson the under side of the sleepers ; these in turn were reversed, thenuts being screwed on from above, and thus were accessible forconstant examination ; the corners of the heads were turned up,like fangs, to bite into the timber and prevent the bolt from turning.When the bridge-rail was well bolted down, with small joint-plates, introduced beneath the junction of the ends, no system couldgive a smoother way for the carriages, or generally succeed better ;it was, however, more expensive, in first construction, than Barlow'ssaddle-back rail.It was generally considered necessary, in laying rails, to providefor their contraction and expansion. Now, however strange theassertion might appear, Mr. Brunei did not hesitate to state, thatsuch provision was unnecessary, under ordinary circumstances,in this climate. This would become evident, when it w as consideredthat 10° or 12° of heat only represented an amount of expansionequal to 1 ton per square inch of pressure ; now if the rails wereheld very firmly by their fastenings, there would be sufficientresistance to counteract any effect of contraction, or expansion,between the points of attachment, as a change of temperature ofeven 50° would only render it necessary to provide for a tension, orcompression equal to 5 tons per square inch in the whole, or 2i tonsalternately, of extension and compression.A fact had been mentioned to him, by Mr. Hawkshaw, relativeto a cast-iron parapet, which completely confirmed this view, and itwas found, that in riveting together the ends of the bridge rails, thesame effect was produced. When this was first mentioned, everybodydoubted it, and to satisfy himself, Mr. Brunei caused a length of 100feet of rails to be welded together, and another length of a quarterof a mile to be riveted together, with accurate gauges fixed at eachend ; the result was conclusive, that there was not any accumulatedmotion, either at the extremities, or in any part of the two lengths.This and other subsequent trials convinced him, that it was onlynecessary to fix the rails securely upon steady sleepers, to provideagainst every tendency to contract, or expand, arising from thevariation of temperature in this climate, although in a very hotDigitized by GooglePERMANENT WAY.35summer he had found the surface of the rail heated to as much as130°.With respect to the various modifications of the forms of rails,proposed by the Author of the paper, in the diagrams exhibited, andwith every desire to remark courteously on the papers kindly pre-pared for the meetings, he must say, that he could not find anysuggestions calculated to advance the practice of laying permanentway. They might be ingenious, but they were not useful, and mostof the forms had been already tried and condemned.As to the split-rail, stated to have been tried in America,there was no experience of its use in this country, to enable anopinion to be hazarded upon its utility. The arguments in favourof the girder double rail were based upon assumed, but notdemonstrated, defects in the saddle- back rail, and it did not appearthat any experiments had been made, to prove either the positiveexistence of the defects, or of the merits of the proposed plan forobviating them. It was presumed, that the saddle-back rail didnot take a solid bearing on the ballast, in consequence of thedifficulty of packing it ; now it was well ascertained, that ballast ofalmost any kind became, in time, so perfectly rammed into thecavity, that a nearly solid mass was formed, whereon the spreadingflauches took a solid bearing, at a sufficient depth to receive thenecessary lateral support, without that hindrance to perfect drainagewhich would be likely to occur from an extreme depth of web,unless there was such a thickness of ballast as would be very ex-pensive in practice. On the other hand, the shallower proposedforms had the disadvantage of taking their bearing so near thesurface, that the packing would be affected by any slight changes ofweather.With good materials, and judgment and care in applying them, apractically perfect permanent way could be made, either with stoneblocks, or transverse sleepers and chairs, or with longitudinaltimber-bearings and bridge-rails, and any of these systems werepreferable to the proposed modifications; but Mr. Brunei hoped,that with Barlow's saddle-back rail a permanent way as good, ornearly as good as any of these, might be constructed, and at a lesscost.Mr. Ashcboft said, from his experience of the system of'fishing* the ends of the rails, on the Eastern Counties Railway,where it was already tried upon a length of twenty miles, he was ofopinion, that there would be an economy of fifty per cent, in thecost of maintenance of way ; and he thought, that by the applicationof the * fishes,' to retain the ends in the same plane, and to renderd 2Digitized by Google36PER M ANENT WAY.them steady under the wlieels, any line, with transverse sleepers andchairs, could be cheaply put into, and be easily retained in, goodworking order.Mr. Bird said, he had been informed, by eminent German en-gineers, that * seite laschen,' which were nearly identical with the* fishing-plates,' had been used for several years in Germany. Onthe railway between Leipzic and Dresden, which was nearly thefirst line executed in ihat country, having been constructed in theyear 1838, a portion was laid with single T rails and 4 seitelaschen,' holes being punched in the ends of the rails for thesefastenings. They had subsequently been employed, throughout theentire length of the Lubeck Railway, and were tried on many otherlines ; indeed the * seite laschen ' were included in most of the spe-cifications ; their form was such as to allow them exactly to fill thecavity on each side the rail, immediately beneath the top table, andthey were held together by transverse bolts and nuts. On therailway near Olmutz, in addition to the * seite laschen,' the twolines of rails were connected transversely by long Iwlts, to preservethe gauge accurately. In Rhenish Prussia the ends of the railswere connected by * seite laschen,' of such a form as to adaptthe upper part to the vertical web of the rail, and with a horizontalweb to be attached to the sleeper. He was not aware, that in anypart of Germany fishing- plates, between the supports, had beenadopted.Mr. Samuel was not awar e of the * fishing-plates ' having beenadopted in Germany, otherwise than as a means of keeping togetherthe ends of the rails in the joint-chair, of which they merelyextended the hold on each side. He had never seen them used, inGermany, with the suspended joint, in the manner he had employedthem on the Eastern Counties and other lines.He gathered from the paper, that Mr. Adams claimed the inven-tion of the fished suspended joint ; the claim was perfectly valid,and Mr. Samuel had become first connected with it, by introducingit practically on the Eastern Counties Railway. He subsequentlyused a * fished-joint chair,* which lie considered a valuable modi-fication of Messrs. Adams' and Richardson's * fish-joint ;' it wascapable of easy application to the permanent way of a line of theordinary transverse sleeper construction, and had the merit ofavoiding interference with the usual position of the sleepers andrails, which became inevitable, when the ordinary * fishing-plates'were applied to a permanent way already laid down. The result ofthe experience of several miles, laid on this system, on the EasternCounties Line, had been very satisfactory.Digitized by GooglePKRMANENT WAV.37Mr. Brunei, had always understood, that it was to Mr. Samuelthe railway world was indebted, for the valuable iutroduction of the* fishing-plates ' with the suspended joint, and all who had noticedthe even surface of the rails and the smoothness of the travelling, onthe portions of the lines so treated, must admit the great advantagesof the system.Mr. P. W. Barlow would correct an impression which appearedto obtain, as to the extent of fracture of the cast-iron bearing-plateson his system ; now the fact was, that the amount of breakage didnot exceed that of the cast-iron chairs, on railways of the ordinaryconstruction. He was convinced, that under all circumstances,the cast-iron bearing-plates could be economically employed, andthis opinion was confirmed by the offer of Mr. Taylor, the con-tractor, to maintain and renew, for twenty years, that portion ofthe South Eastern line, laid with cast-iron bearing-plates, for £100per mile per annum. It must be borne in mind, that the portionof line alluded to, was that on the Greenwich and the main line,where, from the confluence of so many trains, the traffic was nearlythree times as great as the average of that of the other portions ofthe railway ; and also, that the rails had been more than ordinarilydegraded, from the employment of bad ballast.In a Report he had recently made, on * Permanent Way/ 1 he badgiven the details and the results of some extended experiments,having for object, ' to ascertain whether the weights of iron requiredto obtain the necessary bearing surface, and, at the same time, toinsure against fracture, would make the cost greater than the valueof the increased durability of the material :* and without troublingthe meeting by a recapitulation of the particulars of the Report, towhich he would beg to refer them, he would confidently state, thathe cost was even less than that of an ordinary wooden road. *The experience he had, in the maintenance of way, convinced him,that after the selection of a good quality of ballast, the cost of main-tenance of a line depended on the extent of bearing surface, whichshould be so obtained as not to interfere with the drainage. It wouldappear, that in the commencement of the railway system, the en-gineers had admitted this fact, as with five stone blocks, each 2 feetsquare, at intervals of 3 feet from centre to centre, under each railof 15 feet in length, there was an extent of bearing surface of 40feet, under each pair of rails on a single line. On some of themore recent railways, with transverse timber-sleepers, the extent of1 Vide Report on Permanent Way, by Peter W. Barlow, addressed to theDirectors of the South Eastern Railway. 4to. London, 1849.Digitized by Google38 PERMANENT WAY.bearing surface did not exceed 24 feet, part of which was useless,as the plate-layers only packed about 18 inches on each side of thechair. In the estimate of cost of his system of cast-iron bearing-plates, he assumed 28 feet of surface, and he believed that wouldbe found sufficient for heavy traffic, as from the proximity of thebase to the surface of the ballast, complete drainage could beobtained.He understood it to have been stated, that the actual cost ofmaintaining the transverse timber-sleeper permanent way of theSouth Eastern Railway, had not exceeded £30 per mile for sixmonths. Pie must be permitted to observe, that it was very pos-sible to reduce present expenditure, at the risk of deterioration ofthe road, and that half a year was a very limited period to judgefrom.He had tried two small pieces of line laid with Greaves' iron-sleepers ; on one piece, no breakage occurred, but on a part of therailway, where the heavy engines passed over at full speed, severalof the castings had been broken ; he thought they had been madetoo light to bear the severe test of heavy engines, with hard gravelballast upon brick arches. They were generally reported to standwell, requiring less repair than the ordinary sleepers. The cost ofGreaves' system, if equal strength of metal was used in the sleepers,was necessarily greater than Mr. Barlow's cast-iron bearing-plates,with equal bearing surface.The quality of timber employed for the sleepers, and the modeof preserving them from decay, were important questions, in themaintenance of permanent way. He believed the most economicalsystem was, to use only the soundest timber that could be obtained ;and after trying the processes of Kyan, Payne, and Burnett, he mustadmit that the only one which had not completely failed was Bethell'sprocess, which was saturation with creosote.Mr. W. H. Barlow explained, that the introduction of thesaddle-back rails, on the Midland Railway, arose from the greatexpense incurred in renewing the timber-sleepers. In the firstinstance, he proposed a rail weighing 127 lbs. per yard, but in spiteof the exertions of the manufacturers, for which they were entitledto great credit, the difficulty of manipulating such, masses of ironwas so great, that, as a temporary expedient, a length of about ninemiles was laid with cast-iron bearing-plates, beneath rails weighing80 lbs. per yard ; the weight of the bearing-plates was about300 tons per mile, and although the road was laid under circum-stances of considerable difficulty, it had completely succeeded, andhe would not desire to have a inore perfect length of line; theDigitized by Googlel'KKMANENT WAV.39maintenance of a portion of this length had recently been let to acontractor, for £42 per mile per annum ; this demonstrated theeconomy of the system. The rails were 20 feet long, with tworigid cross-ties, formed of bars so deep and stiff as to maintain thebevil of the rails, which was a very important feature in all perma-nent way.As soon as the manufacturers had completed their arrangements,a length of way was laid with saddle-back rails, weighing 127 lbs.per yard, resting on cast-iron joint-plates, info which the rails werekeyed ; this formed a very durable road, requiring very little main-tenance. Subsequently, finding there was an excess of strength,the weight of the rail had been reduced to 100 lbs., and was nowbeing still further reduced to 90 lbs. per yard, the joints beingmade by riveting the meeting ends upon an internal wrought-ironsaddle-piece, somewhat similar to that used by Mr. Brunei andMr. Hawkshaw, for bridge-rails on longitudinal timbers.The traffic of the Midland Railway was, from local circumstances,peculiarly heavy, and like all narrow-gauge roads, laboured underthe disadvantage of the want of base so requisite for heavy trafficand high speeds ; the cross-ties had therefore been extended on eachside beyond the rails, and with considerable advantage.The cost of a line, with saddle-back rails weighing 90 lbs. per yard,when those rails were costing £6 per ton, was £2,100 per mile forthe two lines of way.Mr. W. B. Adams had assumed, in his paper, that the permanentway was a constant source of expense, arising chiefly from thedeflection of the rails, and the sinking of the sleepers, under thepressure of the passing loads. These effects, he argued, could onlybe obviated, by giving to the permanent way a certain amount ofrigidity, upon an elastic absorbent, which would be best obtained byusing deep stiff rails like girders, with lateral strength, and plentyof bearing surface on the ballast, so as to avoid the undue depres-sion of the substratum. For these reasons, although, he believed,not strictly in accordance with the regulations of the Institution,he had ventured to suggest some hitherto untried forms of railsand methods of supporting them, which he felt convinced wouldattain the end which was admitted to be very desirable.Captain W. S. Moorsom, in reference to Mr. Brunei's observation,was of opinion, that only under peculiar circumstances, could theusual precautions, against the effects of expansion and contraction,be dispensed with. The experience of Mr. Brunei and of Mr.Hawkshaw was certainly conclusive (as far as it went), but on theother hand, there were many instances of lengths of line being forcedDigitized by Google40PERMANENT WAY.out of gauge, or lifted up, on sudden changes of temperature, owing,it was supposed, to the rails being too firmly fixed in the chairs.He had seen a length of nearly half a mile so lifted, on a railwayunder his charge ; he therefore again recommended caution in dis-regarding the effects of changes of temperature.The general questions of the first cost, and of the subsequentexpense of maintenance of the permanent way, were of vital import-ance in railway economy, but he thought more details should begiven, to enable a correct appreciation of the value of the varioussystems to be arrived at. The cost per mile of Mr. Barlow's saddle-back rail appeared low, but probably the amount stated only includedthe rails and the labour of laying them, without the ballast. Thisshould be distinctly explained.The data essential for making a comparison between the first cost,and the subsequent maintenance of different systems of permanentway, should comprehend —1. The weights, and quantities of the respective materials.2. The prices of the same — in place, and the amount per mile.3. The character of the ballast.4. The description of engines used on the road.5. The number of trains — per annum.6. The speeds.7. The length of way subject to this traffic.8. The time during which the expense of maintenance had beenexperienced.9. The cost of maintenance per mile, per annum, including allmaterials and ballast.10. The cost of maintenance, per train, per mile run, per annum.As an example of his meaning, he would give a sketch ofthe particulars of the construction of the Waterford and KilkennyRailway in 1846, when iron was at the high price of £12 perton.The sleepers were continuous longitudinal bearings of half balks,cut to an isosceles triangular section, with stiff cross-ties formingstruts also. The double I rails weighed 56 lbs. per yard. Thechairs, cast by Messrs. Ransomes and May, of their ordinary pattern,but with a parabolic section on the length of the bed, weighedrespectively, for the joints, 27 lbs. each, and for the middle bear-ings, 14 lbs. each; the saddle-pieces weighed 7 lbs. each. Thelengths of bearing varied from 30 inches, in the middle, to 15 inchesat each end of the rails, before reaching the joints.The first cost of this permanent way, if the rails cost £7 per ton,Digitized by GooglePERMANENT WAY. 41the chairs and saddles £5 per ton, and the timber for the sleepers1*. 6d. per cubic foot, would be per mile of single line of way-Rails .......£ 616Chairs and saddles ..... 143Pins, trenails and keys (Ransomes' and May'scompressed) ...... 87Timber-sleepers, ties and dowels . . 453Cost of laying 88Total cost per mile . . . £1,387The railway, about thirteen miles in length, with a gauge of5 feet 3 inches, had now been regularly travelled over, for betweenthree and four years, by eight, or ten trains daily on each line,making about three thousand trains per annum. The weight of thelocomotives used was about 22 tons, and the average speed of thetrains was twenty -four miles per hour.The cost of maintenance of way, including the renewal of allmaterials and ballast supplied, since the line had been out of thecontractor's hands, had been, for the first year, £53 12*. per mile;for the second and third years, £35 12*. per mile ; and for the firsthalf of the current year, at the rate of £31 per mile per annum, orat the rate of 2-51 pence per mile run, per annum.He believed this cost of maintenance was less than that of manylines of equal traffic.Mr. R. J. Hood thought it was almost impossible to draw anaccurate deduction, as to the comparative cost of maintenance ofway, from any published reports, as it was rarely stated what wasincluded under that head ; for instance, on one line it was strictlylimited to keeping up and renewing the lines of way, the ballastingand the ditches ; whilst on another line the keeping up the fenceswould be included, and perhaps on account of the vicinity of theline to populous districts, the fencing was of an expensive kind. Heagreed with Captain Moorsom, in the necessity for specifying every-thing very clearly.As to the modifications of forms, proposed in the paper, theyappeared to be chiefly untried attempts, to remedy some of theacknowledged evils of existing systems, but he could not agree incommending them.Cast-iron sleepers had been laid down to some extent on theBrighton Railway, but Mr. Hood dreaded the expense of repair andof their renewal, or, what was more probable, the substitution ofDigitized by Google42PERMANENT WAY.some other system. There were also, on the same railway, somebridge-rails laid on transverse sleepers; it was not a satisfactorysystem, and was frequently out of repair.The 'fished* suspended joints were decidedly the best, even withonly four, instead of five sleepers ; he used 1 fishing-plates ' 18 incheslong", with four transverse bolts $ inch diameter, and found thesurface of the rails so even, that the rolling stock was less injuredthan before their adoption. lie was certain that system woulddiminish the cost of maintenance of way.Mr. Hawksiiaw corroborated the statement, as to the apparentabsence of expansion and contraction, in lengths of iron securelytied down. About ten years since he erected at the station of therailway, at Salford, a parapet 750 feet long, composed of cast-ironplates, each 9 feet long, and 1 inch thick, firmly riveted together,and held down tightly ; it retained a perfectly straight line, underall variations of temperature, without any special provision beingmade for contraction, or expansion.He commenced using the systems of fishing rails, and of rivetingthem firmly together in long lengths, as far back as in 1846: oneend of the joint-plate being riveted to the rail in the woiks, andthe other on the line, when in place ; no ill effects had ever beenexperienced, on his lines, from changes of temperature.When rails were driven hard up endwise against each other inthe joint-chairs, without there being fishing-plates to keep themdown, and in the same plane, there was when under expansion a ten-dency in the rails to rise. The best feature of the fishing-plates wastheir power of holding the ends of the rails firmly down, and ofkeeping the travelling surface even.The question of the relative cost of maintenance of way could notbe properly discussed, without a minute detail of all the dates andcircumstances. For instance, a line under his charge was constructedfor some miles across a coal district, where, in consequence of theunder-ground workings, the surface constantly subsided, and, ofcourse, a large quantity of ballast was required to restore the level,independent of the constant labour in packing and keeping up theline for the traffic. The mechanical questions of good forms andproportions, of good quality of materials and of ballast, and ofcare in construction, were now generally well understood, and thebest systems were acknowledged to be the cheapest in the end. Hemust repeat, that the cost of the maintenance of way was so dependenton a variety of circumstances, that it was extremely difficult to forma comparison between any two lined.Digitized by GooglePERMANENT WAY.43After very extensive experience of almost all ordinary forms ofrails and systems of construction, he gave the preference to a strongbridge-rail, firmly attached to longitudinal bearers of sound balk;this he, at present, considered as the best system of construction.He did not think that the fibre of the longitudinal timber-balk wasmore crushed under the bridge-rail, than that of the transversesleepers under the chairs ; and he believed the cost of maintenanceof way to be somewhat less with longitudinal, than with transversesleepers. 1He had used Greaves' cast-iron sleepers successfully, and he wasof opinion, that in certain situations, particularly where the sandwas used for packing or the gravel ballast was small, the systemwould answer well ; on some foreign railways they would be founduseful ; particularly in the East, where only such timber couldbe used as was known to resist the ravages of the white ant,and there it was sometimes expensive and difficult to be obtained.Mr. C. May directed attention to the joint-chairs, slightly alludedto in the paper ; cast in 1849, by Messrs. Hansomes and May, atIpswich, from the design of Mr. John Fowler, M. Inst. C.E.The object of the extended chair was, to prevent the injuriouseffect on the rolling stock, and permanent way, arising from theends of the rails being permitted to move, as in the joint-chairsof the ordinary construction. This evil had been felt by allengineers, and there were few who had not attempted a remedy.Mr. Fowler had tried an experiment on one mile of line, undervery considerable traffic, and, after an experience of two years, heexpressed himself satisfied, that there was less action at the ends ofthe rails, which were held down more firmly, the injurious effects onthe rolling stock were diminished, and the cost of maintenance ofway had been reduced.This joint-chair (Fig. 20, page 261) was 2 feet 6 inches long,and had three jaws on each side, so as to hold the ends firmly for15 inches of each rail, giving that length of solid bearing.Experience had shown, that the body of the chair might beshortened about 9 inches, without diminishing its utility, as itslength was determined by the power of properly packing the twosleepers on which it rested. Only three chairs had been broken,in two years, and those all near the same place, showing thata slight addition of metal to that part would be beneficial.It was not contended, by Mr. Fowler, that this was the best1 In Mr. Hawkshaw'8 Report on the Rolling Stock and Permanent Way ofthe Lancashire and Yorkshire Railway, for Manchester, 1850, these points arediscussed with great minuteness.— Sec. Inst. C. E.Digitized by Google44PERMANENT WAY.method of securing the end of the rails ; but as it had proved suc-cessful, aud it was not the object of a patent, he had requestedMr. May to lay the facts before the meeting.Mr. Wilson corroborated the statement relative to Mr. Fowler'sjoint-chairs; they had been used, in 1849, on the Lincoln branch ofthe Manchester, Sheffield, and Lincolnshire Railway, where therewas considerable traffic, with heavy engines, and at high speeds ;they were subjected to severe trials, by running engines, at consider-able speeds, over obstructions 4 inch in height placed on the rails,over the joint-chair, and the shocks were borne without any injuryto the joint. The bearing was originally 2 feet 6 inches long, butit was found, that the base of the chair might safely be reduced to1 foot 9 inches, which had been done, without diminishing thesteadiness of the chairs. The part of the line, on which theywere used, was excellent, and the cost of maintenance of way wasless than in other parts.Mr. Allport confirmed the evidence, as to the length, on whichthese joint-chairs were used, being the best part of the whole line,although it was subject to the action of the heavy engines and trainsof the Great Northern Railway, passiug over it at considerablespeed.Captain Simmons, R. E., being fully impressed with the im-portance of the subject, had listened with much attention to thediscussion of a question which, however, he thought was of toogreat magnitude to be fully canvassed at a meeting of the Institu-tion; he would, therefore, confine his observations to a very fewpoints'.He had made some experiments on the permanent way describedby Captain Moorsom, as at first he entertained some misgivings asto the solidity of the line, on account of the apparent weakness ofthe rails, the distance between the main chairs, and the absence ofpower in the intermediate saddle-pieces to afford lateral support tothe rail ; the experiment might, however, be considered generallysatisfactory.In considering the original construction of the railway, and thecost of maintenance of way, he thought that the amount of re-sistance to trains should be imported into the calculation. It hadbeen stated in evidence, that at a speed of 60 miles per hour therewas a resistance equal to about 60 lbs. per ton, on a narrow-gaugeline, whilst, at the same speed, the resistance was reduced to18 lbs. per ton, on a broad-gauge line. Now as there was not anyvery great difference in the form, arrangement, or weight of thelocomotive engines on the two lines, he had been at a loss toaccount for the stated difference of resistance.Digitized by GooglePERM AN ENT WAY.45He would now venture to suggest, whether the difference mightnot, to some extent, proceed from the different systems of formingthe line ; the longitudinal bearing affording that regular and con-tinuous vertical and lateral support to the rail, which it onlyreceived at intervals from the cast-iron chairs on transversesleepers ; whilst, with the latter system, it was very difficult to pre-vent a certain amount of rocking motion in the chairs, which mustproduce considerable resistance ; and at the joints, unless the railswere very heavy, or were * fished,' the resistance was very muchincreased.The system introduced by Mr. Fowler appeared to be a decidedimprovement; the joint-chairs were very steady, and the increasedlength, of 30 feet, given to the rails, was a very advantageousinnovation, as it got rid of half the usual number of joints ; pro-vided, however, that the extra cost of such heavy lengths of rail didnot militate against the introduction of the system.Captain Simmons had examined the roads laid with Mr. W. H.Barlow's saddle-back rails ; the line appeared to be good, and theconstruction solid ; the principal objection against it seemed to bethe anticipated heavy expense of renewal, when the upper table ofthe rails should be worn down, or the surface be laminated, orcracked. In conversation on the subject, with Mr. Morries Stirling,he had been informed, that it was practicable, at a slight expense, toimport a very durable surface to the rails, by the introduction of amixture of metal during the process of rolling ; that the compoundmetal, welded perfectly, and was as tenacious and much harder thanthe ordinary rails. If this process could be adopted, at a reason-able cost, it would prove a most advantageous adjunct in theuse of Mr. Barlow's saddle-back rail. .The cast-iron bearer-plates, introduced by Mr. P. W. Barlow,had also received his attention ; they appeared to have answeredwell, where they had been originally well packed and had beencarefully attended to ; but they were liable to fracture, if the sub-stratum was not constantly kept well rammed beneath them : inshort, like almost all other systems, if it was well attended to, itwas durable, but, if not, the cast-iron way as soon got out of orderas any other road.Mr. B run el was anxious not to be misunderstood in what he hadstated, relative to the contraction and expansion of rails ; he did notmean to imply, that iron rails were exempt from the ordinary effectsdue to changes of temperature ; but that when rails were effectivelyheld down to longitudinal, or other firm bearings, the effect of theamount of contraction, or expansion, occurring under the limitedXDigitized by Google46PEKMANENT WAYrange of alterations of climate in Great Britain, would be resisted,and thus would not produce any of the effects of buckling upwards,or curving outwards, which would occur on lengths of rail ineffi-ciently attached to the sleepers. He had tried lengths of severalmiles of rails firmly riveted together, and securely attached tolongitudinal sleepers, without finding any ill effects' from the con-traction, or expansion.He desired also to be understood, as to the opinion he expressedupon the numerous modifications of forms of rails exhibited in thediagrams, many, if not all of which, were made the subject ofpatents, although they had not been tried in any way, and most pro-bably no person would ever attempt to try them ; the only effect,therefore, to be anticipated from this accumulating of numerousfanciful forms in one patent was, that when a really good form wasdevised, it would be found to bear so close a resemblance to someone of these imagined sections, that either the use of the positiveimprovement would be prohibited, or a fertile field would be openedfor litigation, on a case of infringement. For instance, some of thenew forms shown, though in themselves impracticable, or objection-able, would have sufficed to prevent the use of Barlow's saddle-back rail, which though, perhaps, not positively new in form, mustbe admitted to be a decided improvement upon the rails of thatdescription in ordinary use, and to be a good disposition of material ;there was a good bearing, sufficient depth to give stiffness, and abreadth of base to insure firmness. Now the employment of thissaddle-back rail, which was rather remarkable for its judicious pro-portions, and the good disposition of the metal, than for its novelty,might (as lie before said) have been precluded by some such fancifulsections as those which had been theoretically advocated, although,he must say of them, that among them there was little that was new,and that little was not good. It was notorious, that engineers fre-quently found their practice restricted, by the claims of some tlieo-retical patentee, whose obsolete invention never would have beenheard of, but for the adaptation, in practice, of some, perhaps theonly useful, portion of an invention, originally applied to somewidely-different purpose. He feared that he should incur the ireof the inventive world but in tSese discussions he held it to beincumbent upon the Members; to give their unfettered opinions,however heretical they might be considered.Reference had been made to the principle of having a false topto rails, to be removed when it became abraded, or laminated by thetraffic. Now though this sounded well theoretically, it was gene-rally considered to be useless in practice. Throwing aside theDigitized by GooglePERMANENT WAY. 47danger and other disadvantages, liable to result from the use of arail composed of so many parts, as would be necessary for the re-moval and replacing of a false top, it would suffice to consider themonetary part of the question. If a rail weighing 90 lbs. per yardrequired to be changed, it was true that another rail of the sameweight must be substituted, but the old metal could be re-rolled andconverted into new rails at an expense of about two pounds per ton,which would probably be less than the cost of replacing the falsetops in more complicated rails. It had been admitted, that the ave-rage duration of rails was ten years, and many engineers contended fora much longer duration ; now a very simple calculation would suffice*to demonstrate, that after ten years' use, it would be cheaper, andcertainly much more economical in practice, to take up the entirerail, thau to take to pieces and refix a complicated subrail, iu orderto replace the upper table. As a matter of figures, it must beevident, that at the present price of iron and with the actual facili-ties for rolling rails, no positive advantage could result from renew-ing the upper table of any form of compound rail yet proposed.Mr. Brunei trusted that the remarks he had made would not beconsidered invidious : but he felt strongly the prejudicial effect ofthe present system of accumulating in one patent almost every pos-sible modification of form, or process, and thus obstructing, if notfrustrating, the attempts of others to introduce improvements whichwould become the unrestricted property of the public.Mr. W. B. Adams, in answer to previous speakers, considered,that the amount of expansion and contraction must depend verymuch on locality and circumstances, as the iron might be laidwhere the heat would be carried off with such rapidity as not toaffect the metal. It might, no doubt, be practicable to overpowera given force of expansion and contraction, by a greater forceholding down the rails, but it then became a question whether itwas desirable to do so, and whether the iron might not be injured,by the unnatural restriction to which it was subjected.He could not comprehend, how the ballast remained firmlypressed upwards into the apex of the interior of the saddle-backrail, when the vibration occasioned by the travelling was considered,and he thought the deep cross-ties, necessary for preserving theinward level of the upper table, indicated a certain tendency to un-steadiness in that form, although its base was so widely spread.With respect to the use of the fishing-pieces, he thought commonsense must point out the advantage of giving to the joints, whichwere the weakest parts, such an amount of support, or stiffness, aswould render them as steady as the intermediate parts, and in noDigitized by Google48PERMANENT WAY.manner could this be done so simply, or so effectually as by apply-ing what might be termed ' splints/ to unite the two ends andimpart to them as much strength as was possessed by the body ofeach rail, even if as much weight of metal was abstracted fromthe entire length as would supply the fishing-pieces ; by this meansthe cost of fishing was much reduced. Experience had shown, thatby the application of the system to old lines, the expense of relayingwas avoided, and the increased comfort to the travellers, and thediminished wear and tear of the rolling stock were generallyadmitted.The experiment tried by Mr. Hartley, had proved, that a certainelasticity of substratum was essential for a good permanent way.He must submit, that the opinions, adverse to the forms of railsproposed in the paper, had been supported by assertion rather thanby argument, or by the deductions of experience. If the variousforms and arrangements were as defective as had been stated, theycould not, by the simple fact of being included in a patent, becomesufficiently important to interfere with the inventions, or improve-ments of engineers. The cast-iron subway, the saddle-back rail,the chilled chairs and wooden trenails, innumerable turn-tables andswitches, and almost all the various parts composing a railway werealready patented, so that, at all events, if there was any error inaccumulating several forms in one specification, the sin had beencommitted, in emulation of very numerous, if not very good, pre-cedents. The discussion of the question, whether patents should bepermitted to be secured at all, belonged rather to another subject,and should not be imported into the present proceedings.The moveable top for the rail, which had been so strongly ob-jected to, had been proposed merely for certain localities, where thedifficulties and cost of transport of rails of heavy weight would beconsiderable, when it should become necessary to replace the originalrail. There were other forms, more analogous to the saddle-backrail, with which the latter might have been more correctly com-pared. There was a marked mechanical distinction between thesaddle-back rail and the solid girder rail. In the specification ofthe former, it was described as " a trough rail bearing on the ballastand riveted to bearers,*' and the intention was. by widening a bridgerail to obtain a broader base. In the latter, the object w as to obtaina T girder of such a depth as would prevent undue deflection, andwith sufficient breadth to give ample bearing surface on the ballast,whilst the tread of the rail should be so slightly elevated, that thereshould be little, or no tendency to lateral yielding, or rockinjr.In fact the saddle-back rail might be said to represent the shallowDigitized by GooglePERMANENT WAY.49bridge rail of the longitudinal system, whilst the deep girder railexemplified the T section of the transverse sleeper system.A comparison of the actual proportions of the two forms wouldshow (Fig. 25), that the girder rail possessed more bearing surfaceFig. 25.Adama' Girder Rail. Burlaw*! Saddle-back Hail.on the ballast, combined with a deep keel to prevent oscillation,and there was no greater difficulty in packing one form thanthe other, or rather, if there was any difference, it was in favourof the girder rail. As to the cost of manufacturing the deepgirder rail, the same objections had been raised to the saddle-backform, but they had been overcome by the energy and skill of theiron manufacturers, and if there was a positive advantage in anygiven form of rails, means of producing them at moderate costwould soon be discovered. A crack, or flaw in the horizontal topof a " saddle-back " rail, would seriously injure it, but such a crack,in a " girder- rail" would not impair its efficiency. The matter re-solved itself into the question of which form, with a given weight ofiron, offered the best bearing surface on the ballast, and possessed the 'greatest number of collateral advantages. Experience alone mustdecide ; meanwhile Mr. Adams felt bound to endeavour to clear him-self from the imputation of having proposed forms without ideas.The object of the paper was to direct attention to the T girderrail and its varieties ; — to suggest the resumption of stone blocks,combined with either transverse, or longitudinal timber-sleepers ;— the substitution of durable materials for the perishable ballastnow in use ; — giving to the rails a sufficient amount of elas-ticity to avoid the jarring which now acted so prejudicially on therolling stock, and detracted from the comfort of the passengers ;— and whilst providing for the drainage of the ballast, to afford themeans of employing water, in dry seasons, for consolidating thepacking, and laying the dust during the passage of the trains. Intreating these various topics, the mention of the proposed methodsDigitized by Google50PERMANENT WAY.for meeting certain difficulties, or for obtaining certain ends, wasunavoidable, and experience had shown, that in bringing forward anynew proposition, unless the first claim included every apparent modeof attaining the object, an unscrupulous improver, seizing upon theidea which had cost the inventor much thought, would, by a slightmodification, obtain the credit and reap all the advantage of theinvention.When in 1847 Mr. Adams specified the (i fish-joint," for thedouble-table rail supported on chairs,, he thought that a good andvalid title to an original idea had been recorded ; but a host of fol-lowers in the same track immediately obtained patents for improve-ments, on the method of securing the joint of the double-table rail,used with chairs. This experience determined him, in future, tocombine as many of these, so called, improved forms, or optionalmodes, as his imagination could supply, in order to avoid those dis-putes, which the protection of a successful invention was nearlycertain to induce.It was well known, that upon all lines of railway, there were useda multitude of patented inventions, and therefore, inventors mightbe permitted to hope, that even Mr. Brunei's antipathy to patentedinventions would eventually be modified, and that the deep girderrail might be admitted for trial on the Great Western, or someother line of railway.Mr. Erring ton said more reference should have been made tothe use of stone blocks for railways, as, although they had beengenerally discarded in Great Britain, he thought there were manylocalities where stone might be used with advantage. It would bewell to inquire into the causes of the block road being so completelyabandoned. He had recently examined the Lancashire and PrestonRailway, which had been open for twelve years, and in that periodthe blocks had only been replaced, to the extent of oue-fiftieth ofthe whole.The timber sleepers on the line had been replaced to the extentof one-half, though they were of excellent quality, and had beensupplied at a time when there was no difficulty in getting larchtimber. One-fortieth of the quantity of rails had been replaced,and it would now be requisite to replace one-twentieth more. Thechief injury to the rails appeared to have arisen from the splittingat the joints, which clearly pointed out, that " fishing," or someother system, than had been hitherto pursued, was required. Whether,ultimately, there would not be a great deal of wear and tear, evenwith that new system, appeared doubtful, as the nuts, bolts andfishing-pieces would probably have to be replaced.Digitized by GooglePERMANENT WAY.51With reference to the cost of maintenance of way, it was desirableto define what was meant by the term. In a document, publishedunder the authority of the London and North Western RailwayCompany, it was stated, that the cost of the road, over an average offour hundred miles, had been hitherto £200 per mile per annum.Now it was impossible, that the small sums (£70 to £100) men-tioned for maintenance of way, could comprise the same items asthose in the document alluded to ; he had known cases where, onone mile of railroad, there had been deposited 2,000 cubic yards ofballast, in one year. The upholding, referred to in the Londonand North- Western Railway Report, included the maintenanceof every great work on the line — bridges, tunnels, slips, drains,and ballast, in feet, every tiling an engineer could desire to havedone, in order to preserve a perfect plane for the rolling stock, andthat condition was generally rigidly enforced. In some cases it wasnot usual, in maintaining a road, to replace a key, or spike, or rail,the moment it was wanted, and to include it in the maintenance, butit was reserved for a future time, when the cost was included in thatof more special restoration out of the capital. He thought, on thepresent occasion, although the object was, perhaps, rather to discussthe mechanical merit of certain forms of construction, yet in sopurely commercial a matter as a railway, the vital question of costshould not be lost sight of. It was yet to be proved, whether thesaddle-back rail could be maintained at anything like the expensethat had been stated ; but he could not believe, that in the pricenamed, anything approaching to a general maintenance of way hadbeen included. It was therefore desirable to obtain further informa-tion on that point.Mr. P. W. Barlow stated, that the cost of maintenance given byhis brother at £42 per mile, included the upholding of the fences,ballasting, and the ordinary repairs of the road ; any extraordinaryexpenses, such as the giving way of tunnels, or other large works, orlarge slips, were not included. Everything was comprehended thatwould be comprised in a contract for maintenance. Mr. Wythes,the Contractor for maintaining the way with cast-iron bearings, hadinformed him, that when the road was consolidated, one man per milewould be sufficient to keep it in repair, and in that he included theordinary cost of fences and of maintaining the road.Mr. Bidder said much valuable time had been wasted in thediscussion of the speculative portions of the paper, which hethought was not quite of the character pointed out in the list ofsubjects circulated among the members of the Institution. No moreimportant subject could be treated of, than the construction andDigitized by Google52 PERMANENT WAY.maintenance of permanent way ; a paper on such a subject ought toenter fully into detail, and the first thing to be clearly ascertainedwas, what was meant by the term " maintenance of way/' which wasstated to have cost all sorts of sums, varying between £42 and£200 per mile. Descriptions of the mode of laying transverse, orlongitudinal sleepers were not required, but a paper on this importantsubject, should have traced the results of all the experiments upon,and trials of, the various systems which had been tried. Now thepaper under consideration chiefly contained observations on a greatmany forms of rails, but did not attempt to supply information onthe methods of doing the best work in the most economical manner,whether in the construction of a new line, or in repairing an oldone.Allusion had been made to the rail first laid on the Black wallRailway ; it had bevelled joints, which were described as beingweak ; now in fact the joint was not weakened, as it was made bytwisting the web of the rail and cutting off a portion of the pro-tuberant end. The rail was certainly rendered somewhat more ex-pensive, by the introduction of a process, which he would recommendon every railway, namely, by causing the iron to undergo additionalhammering in the course of the manufacture ; that had only costthirty shillings per ton, and he believed the extremely durability ofthe old rails was, to some extent, owing to the extra-manufacturingprocess to which the iron was submitted. If instead of introducinglarge and heavy rails, a more perfect mode of manufacture hadbeen sought for, he thought they would have had more economicalrailways.The great question (although he admitted they were scarcely in acondition to settle it) was, which was the best system of constructionto be adopted, at the present time, for a new line of railway ; theywould have to choose between the longitudinal system (which was,perhaps, brought to the greatest perfection by Mr. Brunei), — thetransverse system, — and the saddle-back rail. There remained thento be considered, the prices of wood and of iron, the relative durabilityof all the materials employed, and their value when renewed. Ifthe price of w rough t-iron was assumed to be £5 10*., and cast-iron£4 per ton, and sleepers 4*. each, it was practicable to use heavyrails, and cast-iron bearings, if otherwise deemed advantageous ; butif iron rose to £12 per ton, it would be necessary to use more timber.He had been in the habit of assuming the durability of a goodrail to be fifteen years ; that an ordinary unprepared sleeper wouldlast six years, and a creosoted sleeper fifteen years : then arose thequestion of the value of the old materials, and, at the presentDigitized by GooglePERMANENT WAY.prices, he thought those of the longitudinal system were of the leastvalue, next came the transverse system, and the most valuable werethose of the saddle-back rail system. Therefore, in England, heshould be inclined to adopt the latter ; at the same time, in Norway,it had been found more expedient to adopt the longitudinal system,because sleepers cost there only about sixpence each, and for thatlocality, which offered several peculiarities, it was considered thesafest kind of road ; and he begged to acknowledge the handsomemanner in which Mr. Brunei had supplied him with all the informa-tion he required.Then came the question of the best mode of repairing an existingroad ; and, from observation and careful inquiry, he had arrived atthe conclusion, that the fish-joint was the most effectual system.He was of opinion, that the simple packing of the road could bedone, as had been stated, for £40 per mile ; but on the question ofmaintenance of way, generally, the Institution required to be fur-uished with a considerable amount of data, before the subject couldbe satisfactorily discussed; and he trusted some Member wouldsupply a comprehensive paper on the subject, in the course of thenext Session.Mr. G. H. Phipps said, the statement, of there being, practically,no prejudicial effect from riveting together long lengths of rails,had excited considerable surprise ; but, from his own observation,he was prepared to admit the fact of there being no perceptibleeffect of contraction, or expansion, if the rails were firmly fasteneddowu upon longitudinal bearings ; he thought this was as demon-strable by theory as it appeared to be positive in practice.Mr. Hawkshaw regretted there had not been more precision inthe terms of the discussion. It was just as practicable to define theduration of a rail, as that of a grindstone ; it must be evident,that it was dependent on certain considerations ; a rail might lastany term, between five years and twenty years. The fact was, thelife of a rail was in proportion to the number, velocity, and weightof the trains running over the line. He had some rails, weighing42 lbs. per yard, on the Preston and Wyre Railway, which hadbeen laid twelve, or thirteen years, and now appeared capable ofserving for ten years more, whilst, on other lines, he had renewedheavier rails, which had only been laid from three to nine years.When, therefore, the duration of permanent way was mentioned, thenature and amount of the traffic should be specified, or otherwisethe conclusions arrived at would be erroneous. The cost of main-tenance of permanent way had been referred to, without any state-ment of the weight and number of the trains, of the expense of theDigitized by Google54PERMANENT WAYballast, of the cost of the iron, or of the precise character of theworks.It was important, also, to consider which system of permanentway offered the least resistance to the passage of trains. He wasaware of the difficulty of treating the subject ; but it was evident,that a railway might be cheaply laid, and yet, by increased deflection,might offer great resistance, and if it thus augmented the cost ofpower, there was not much gained by the innovation. A few yearsago he had been obliged to take up a piece of railway, which offeredless resistance to trains than any line he had yet seen ; it was laidby Mr. Hartley, who had attempted to make a line like a lathebed ; the construction was very rigid, the raita were rubbed byhand, into blocks of stone, until a perfectly smooth surface wasobtained, and while the line was in working order, it offered onlysmall resistance ; but when, partly from its inherent rigidity, it oncegot out of order, there was no remedy but to take it up and to putdown what must be considered to be an inferior road, but which wasbetter adapted to the nature of the traffic.Captain Simmons, R. E., said, that even if, in addition to the grosscost of maintenance of way, the separate expense of each kind ofmaterial was given, with all the details that had been mentioned,and all necessary allowances were made, it would still be nearlyimpossible to arrive at an accurate comparison of the cost of anytwo lines, on account of the difference in the rolling stock. Theweight on the driving- wheels of the engines varied with every class,and the hammering effect of certain kinds of locomotives, must havea very serious effect on the permanent way.Mr. Brassey, in answer to the President, said his experience didnot enable him to accord with the statement, that a line could bemaintained in effective working condition for £30, or £40 per mile.He admitted that his practice in the maintenance of railways,though extensive, had beeu confined to lines laid on transversesleepers, generally with rails of 75 lbs. per yard, chairs of 25 lbs.each, and with iron spikes, and he must state, that he had not beenable to find any road which had been more economically maintained,yet he could not have it well done for less than £80 or £90 permile, and he believed he had not done worse than other contractors,nor had he many idle men in his employ. If he had to construct arailway and to keep it in repair for twenty years, he should userails of 75 lbs. per yard, on transverse timber sleepers, with 25 lbs.chairs, fastened down with iron spikes, as being less liable to rotthan wooden trenails.Mr. Bethell, through the Secretary, begged to correct what heDigitized by GooglePERMANENT WAY.55contended was an erroneous statement relative to the inflammabilityof creosoted timber. During its very extensive use, for upwardsof fifteen years, there was only one instance of a structure sufferingfrom fire ; and that one was a bridge in Wales, which had been seton fire by a workman driving a red-hot bolt into a creosoted beam.When Mr. Bethell's premises, at Battersea, were burned down in1847, the stacks of unprepared timber caught fire long before thatwhich had been creosoted. He conceived the reason for this, to be,that unprepared timber, when heated, gave off a vapour which caughtfire at a lower temperature than that at which the vapour fromcreosoted timber would inflame, because creosote oil required ahigh temperature to volatilize it. Therefore he contended, thatcreosoted timber would not catch fire so soon as unprepared wood ;but he admitted that when once it was on fire it burned for a longertime.There was not any instance on record of any of the stacks ofcreosoted sleepers catching fire, although they were generallyplaced close beside the railways, and were constantly exposed tothe felling upon them of red-hot cinders and sparks from the loco-motive engines.With respect to the duration of prepared sleepers, it was now anaccepted fact, that if carefully creosoted, they would endure for atleast twenty years ; and as regarded the price, the general estimateof 5$. per intermediate sleeper, and of lOj. per joint sleeper, hadproved to be rather over than under the actual contract prices.Mr. R. Stephenson, V.P., said the subject under discussion hadoccupied his attention for many years, and he had some time agoarrived at the conclusion, that the maintenance of permanent waywas, as nearly as possible, an insoluble problem, unless by firstcarefully analyzing every element by which it was influenced ; andeven then, as those elements were inconstant, and as no two casescould be found exactly parallel, it was evident, that no correctcomparison of cost could be arrived at. He quite agreed, that ifthe traffic was small and the speed slow, stone blocks sufficed for arailway for the conveyance of heavy minerals, at a speed of aboutten, or twelve miles per hour, indeed he thought that a stone-blockroad offered less resistance than a timber-sleeper line.With these slow trains resistance was a very important object,because by its diminution, an engine of a given power, was enabledto drag after it ten tons, or twenty tons, in addition to the ordinaryload ; but when it was necessary to convert the power of the engineinto velocity, the case was entirely changed, and then that kind ofpermanent way, best adapted to resist repeated shocks, with the leastDigitized by Google56PERMANENT WAY.injury to the rolling stock, became necessary, and under these cir-cumstances the stone blocks made the worst road.Even if it was admitted, for the sake of argument, that there wasa little more resistance with one system than with the other, itbecame of little importance, as at high velocities the questionof atmospheric resistance came into consideration, and that wasentirely apart from the present question. He could not supposethe remarks had intended to imply, that any ordinary differencein the construction of a line could cause any material difference inthe resistance of trains. At the time of the discussion on thatsubject, a wide difference of opinion was entertained by men ofexperience, but that had nothing to do with the present question.Therefore the distinction he drew was, that on a mineral road, withmoderate speeds, stone blocks could be safely applied ; but with thehigh speeds required by passenger traffic, they were quite out of thequestion ; he conceived that his position was proved by the resultof the experiment tried by Mr. Hartley, who in constructing a veryrigid unyielding line, as nearly as possible resembling a lathe bed,had only produced a perfect block road, which was soon proved tobe unfitted for traffic at a great velocity. The cost of maintenanceof way must depend, in a great degree, upon the nature of the strata,and the character of the country, through which the line passed ; onecutting would keep the men continually employed in repairs, whilstin another cutting, the grass would grow even beneath the rails; theformer being in a wet clay, with sand, and the latter in a dry, firm,sandy soil, without any water near it. The simple introduction ofa little water, which would be found to vary in all localities, madeall the difference ; so that if the costs of maintenance of each railwaywere separately analyzed, it would be found, that instead of beingable to compare one line with another of the same length, it wouldscarcely be possible to compare two miles of the same line. It wastherefore a problem, which, at present, could not be solved. Hedid not wish to be understood as decrying the value of a record ofactual facts ; on the contrary, it would be highly useful, as if all thecircumstances and considerations were clearly stated, an approximatejudgment might be formed. But it required great care in ascer-taining whether the cases, to which the facts were applied, involvedthe same facts to the same extent. The question was one of greatimportance to the railway world, and he hoped the Institutionwould continue to receive, from resident engineers on railways, theresults of the cost of maintaining permanent way, and that thepapers would clearly define what was comprehended under the headof maintenance.Digitized by GooglePERMANENT WAY. 57It was imperative to arrest the impression, which was entertainedby some persons, that it was desirable to obtain a perfectly rigidroad ; Mr. Stephenson's experience had led to exactly the oppositeconclusion. It was quite true, that a rigid road might, theoreti-cally speaking, be considered the best, but in practice it was theworst, for although from its rigidity it might offer the least resist-ance, it increased the wear and tear prodigiously. His attentionhad been called, some years ago, to the fact of the rails in theWatford Heath cutting, on the London and North Western Railway,being split and crushed, to such an extent, that although thecutting was through gravel, the rails were removed two, or threetimes. On a personal examination of the spot, he soon perceivedthat the destruction of the iron, arose entirely from the gravelhaving been extremely well consolidated, and the stone blocks sofirmly bedded and packed, that the line was rigidly unyielding, andconsequently the iron rails were literally crushed under the fasttraffic.The chief causes of deterioration of the permanent way, were themileage, the speed, and the weight of the trains passing over theline ; though the weight per se of the engines and its distributionupon the wheels, must also be considered. Wear and tear was, then,entirely a question of certain weights passing along the rails, atgiven speed ; in a given time ; the destruction of the surface beingaccelerated nearly as the speed, and the general dilapidation beingaugmented in an increasing ratio, as the number and the weight ofthe trains were increased beyond a given point.Mr. Stephenson trusted, that the question of ' permanent way 'was only opened by the present paper, as it was one of vitalimportance and could not be too frequently brought before theInstitution of Civil Engineers.

Attribution

This article incorporates text from this source, which is in the public domain: Weick, Fred Ernest (1854). The Construction and Duration of the Permanent Way of Railways in Europe, and the Modification Most Suitable to Egypt, India etc. London, Printed by W. Clowes and sons.