THE TRAIN NOW DEPARTING
Home Up Maps and Plan Main Index Site Search
 


 

NOTES AND EXTRACTS

ON THE HISTORY OF THE

LONDON & BIRMINGHAM RAILWAY


CHAPTER 3.

PROBLEMS WITH THE TRACK

 
THE CAST-IRON PLATE RAIL


The development of the railway locomotive languished for some years following Trevithick’s withdrawal from the field.  This was not wholly due to the absence of his inventive genius, nor to the lack of a commercial driver for change, for the inflationary pressures of the Napoleonic War had greatly increased the cost of horse fodder and with it that of horse-powered traction.  The principal reason, as Trevithick had discovered at Pen-y-Darren and later in London, was the absence of rails capable of withstanding a locomotive’s weight and hammer blow. [1]  More than a another decade was to pass before rails became available that could reliably withstand the stresses of locomotive traction.

This chapter reviews the main changes that took place in rail design and technology leading up to the world’s first passenger-carrying, steam-worked railway, the Stockton and Darlington.


Iron plate rails mounted on stone blocks ― the Derby Canal Railway.


The development of the wagonway progressed with the coming of the Industrial Revolution and the widespread use of iron.  First, wooden rails were reinforced with iron plating to reduce wear and to further reduce friction.  It was then a natural progression to the all-iron L-shaped rail, [2] which was not attacked by rot and could better withstand the weight of the wagons:


“Notwithstanding the imperfection both of the railway and the impelling force, as compared with the improved apparatus of the present day, the advantage was so considerable, that a single horse could draw three tons of coals from the pit to the river.  There was, however, a drawback on the advantages, from the expense of repairing the wear and tear of the decayed wood, which was, indeed, under some circumstances, so great as to render the use of rail-ways, made of this material, a very doubtful benefit.  At length iron was introduced, and found to succeed remarkably well.  But in the first instance the railways were not made wholly of iron.  Flat bars of this metal were fastened on the top of the existing wooden rails, and this was considered a great improvement.  A greater still, however, which soon succeeded, was making the rails wholly of iron, cast in short bars, united at their extremities, and resting on square blocks of stone, instead of logs of wood, arranged at short intervals along each side of the road.”

The British Magazine, Vol. I., p. 121, 1830.


The introduction of iron reinforced wooden rails, followed by the replacement of wooden rails with iron plates, increased wear on the wagons’ wooden wheels.  To combat this, iron tyres were introduced, followed by iron wheels, both of which further reduced a railway wagon’s rolling resistance:


“The next improvement in order of time appears to have been the use of cast-iron as a substitute for the wooden rails, and these were tried on a small scale at the Colebrookdale iron works in Shropshire about 1767, at the suggestion of Mr. Reynolds, one of the partners in that concern.  About this same time cast-iron wheels, turned in a lathe, and made with great truth and accuracy, began to be used, and then it was that the great advantage of these roads became apparent, for the advantage to be gained by a rail-road depends in a great measure on the perfection of the workmanship bestowed upon it, to make it truly smooth and level, and on making the carriages that run upon it as free from friction and inequalities of motion as possible.”

Elements of Civil Engineering, John Millington (1839).


It is doubtful that iron wagon wheels were first used at Colebrookdale, as Millington claims.  Like much of our railway history their first use was probably in the North East of England. [3]

Although there were variations in design, by the end of the 18th Century a typical iron wagonway consisted of L-shaped plate rails, which today we would describe as ‘angle irons’.  A wagon’s flat-rimmed wheels ran along the rail’s flat surface (about 3½ inches wide), while the uprights (about 4 inches high) kept the wheels aligned with the track.  This contrasts with a modern railway, where the opposite applies ― flanges on the wheels guide them along a flangeless track.


Plateway mounted on stone blocks ― the Derby Canal Railway.


Plate rails were made of cast-iron, generally in lengths of three to six feet, drilled to receive spikes [4] and supported on either transverse timber sleepers or stone blocks.  Track-beds were probably unballasted at this time, for  a civil engineering manual of the period speaks of the need to support their ends where timber sleepers were in use:


“. . . . a transverse timber sleeper may be let into the ground and a large heavy stone, or mass of rubble-work in mortar, may be sunk below each of its ends to give it a firmer bearing and prevent its sinking deeper.”

Elements of Civil Engineering, John Millington (1839).


Where the lines were laid on timber they were spiked directly into the sleepers, whereas when laid on stone blocks, holes had first to be bored into the blocks and plugged with wood, and the lines then spiked into the plugs.

An advantage of using stone blocks was that the track-bed between the lines was left clear for the horses’ hooves, but to achieve this with timber the sleepers had to be let into the ground, or covered with gravel, where they were more prone to rot.  But stone blocks had their disadvantages; they were more prone than wood to vibration, which shook the rails loose, while the wooden plugs gradually saturated and swelled, splitting the stone — such was the construction of the Pen-y-Darren Tramroad [5] on which the first steam locomotive-hauled train was to travel in 1804.

Besides colliery owners building wagonways, canal companies also made considerable use of them during the canal era (little realising they would evolve to supersede the canals), principally as feeders in situations where they provided a more economic alternative to a branch canal, or where the terrain was unsuitable for a waterway.  It was in the construction of such feeders that the civil engineer Benjamin Outram’s name is now associated.


The Derby Canal Railway.


One example of Outram’s work was the ‘Derby Canal Railway’ (aka the ‘Little Eaton Gangway’), a typical wagonway of the period.  Now abandoned, the Derby Canal was opened throughout in 1796 to form a link between the Trent & Mersey and the Erewash canals in Derbyshire.  As a more economic alternative to a branch canal, Outram built a wagonway to link the Denby collieries with the canal at Little Eaton wharf — Priestly described his plateway thus:


“From the northern end of the main line [of the canal] at Eaton, a railway proceeds by Horsley and Kilboum, to Smithy House, which is nearly four miles and three quarters in length.  From Smithy House there is a branch one mile and three quarters in length, to the collieries at Henmoor, situated one mile and a half east of the town of Belper; another one mile and a half in length, by the potteries, to the extensive coal works near Denby Hall; with a collateral branch out of the last mentioned branch, three quarters of a mile in length, to other collieries north of Salterswood.”

Historical Account of the Navigable Rivers, Canals etc., Joseph Priestley (1831).


The Derby Canal Railway used cast-iron L-shaped plate rails, approximately three feet long, which were spiked into stone sleepers.  Its horse-drawn wagons had detachable bodies, each of a capacity of about 1¾ tons, which were loaded into barges by crane at Little Eaton Wharf, an early form of containerisation.  The wagonway’s principal cargo was coal, but it also carried stone, pottery and other goods.  It remained in use in its original horse-drawn form until closure in 1908.


――――♦――――

 

 
THE EDGE RAIL.


William Jessop was one of the great civil engineers of his era, his name being linked to a wide range of civil engineering projects that include canals, docks and wagonways.  He is also credited with the first use of the cast-iron edge rail, the predecessor of the modern railway line. [6]

Around 1794 ― the date is uncertain ― Jessop engineered a wagonway to carry coal from Nanpantan, on the Charnwood Forest Canal, to Loughborough.  This was to have been laid with plate rails of the Outram pattern, but, so the story goes, the trustees of a turnpike road objected to the obstruction posed by the rails’ raised flanges at a point where the track crossed the carriageway.  However, a further problem with plate rails was that the debris thrown up by the horses’ hooves lay on the rail’s flat surface and obstructed the free running of the wagon wheels.  Jessop’s use of the edge rail solved both problems; where the rails crossed a road, their upper edges could be recessed so as to lie level with the road surface, thereby posing no obstruction to traffic, and elsewhere their raised running surfaces remained substantially free of stones and dirt.


“Edge rails succeeded plate-rails, having been first used in 1785; the inconvenience arising from the dust laying on the latter probably led to their introduction originally, although the many other advantages possessed by them might hot have been contemplated at the time, as the form of edge-rails is certainly very superior, combining the least expenditure of material with the greatest possible strength and the friction upon them is less than upon tram-rails.”

A Glossary of Civil Engineering, S. C. Brees (1844).


The earliest iron edge rail consisted of a cast-iron bar, laid on edge and generally three feet long (due to cast-iron being brittle, it could only be laid in short lengths).  The underside of the rail was elliptical to provide greater strength between the supports, which gave rise to the description ‘fish bellied’.  The rails were mounted in a succession of iron chairs, each being spiked to either a timber or a stone block sleeper.  Vehicles that ran upon edge rails required flanged wheels, which kept them aligned with the track.

The iron edge rail was not adopted immediately and co-existed with the plate-rail for some years:


“It was of course a great saving of material to cast a flange on the tire of the wheels only, instead of along the whole length of the line; but plate-rails with the flange upon them continued in use notwithstanding, in virtue of the force of prejudice.  In 1797, Mr. Jessops edge-rails were laid down on the Lawson Main Colliery Railway, near Newcastle; but in 1800, Mr. Wm. Outram laid plate-rails, with a flange, on the Little Eaton Railway, in Derbyshire.  In 1801 the Wandsworth and Croydon Railway was laid with these flange rails, and in 1803 the Croydon and Merstham Railway.  It was not till about 1812 or 1815 that edge rails got the mastery.”

The Railway Register, Hyde Clark (1847).


In 1794, Jessop entered into partnership with Benjamin Outram and others in an ironworks, ‘Benjamin Outram and Company’, [7] that Outram had set up some years previously at Butterley in Derbyshire, and the business began manufacturing both types of rail.


――――♦――――

 

 
LOSH AND STEPHENSON’S PATENT


In 1816, George Stephenson contributed to the development of the iron edge rail by devising an improved method for joining and fixing the rails in place.  At the time, railways were poorly laid, the outcome being an uneven track on which locomotives and wagons were subjected to excessive wear and tear from the considerable jolts they experienced in passing over protruding rail joints, sometimes being derailed.  Stephenson replaced the existing butt joints between each rail with half-lap joints, which extended the rails over each other for a short distance at their ends.  He also redesigned the supporting chair, so that the joint between the rails rested upon the apex of a curve in place of a flat surface.  To reduce the number of rail joints, the chairs were moved from 3 feet to 3 feet 9 inches or 4 feet apart.


Fish-bellied rail, showing (top) curvature in the supporting chair and (bottom) a half-lap joint.


The effect of these changes, while reducing the number of rail joints, helped to maintain an even line.  Were a sleeper to tilt from the horizontal, the rail would remain tangential to the curved base of the chair in which it was seated, while the half-lap joint with the adjacent rail kept it locked in place.

Stephenson, together with William Losh, part owner of Losh, Wilson & Bell, who at the time employed Stephenson for two days a week at their Walker Ironworks, applied for a patent to cover this invention.  Registered on 30th September 1816, the patent ― an extract of which is at Appendix I. ― included the steam suspension and the application of malleable iron to rail vehicle wheels referred to in Chapter 2.


――――♦――――

 

 
THE MALLEABLE IRON RAIL


An important step in the development of the iron rail took place at the Tindale Fell Railway (aka Lord Carlisle’s Railway), an early mineral line that served the extensive collieries and lime works at Tindale Fell in Cumbria.  The line had been laid partly with cast-iron and partly with malleable iron [8] rails.  Over a period of years its operators noticed that while the malleable iron rails remained in good condition, those of cast-iron became worn and liable to fracture, an outcome that ran contrary to rceived wisdom on the subject.  Because malleable iron was vulnerable to rust and was also more expensive than cast-iron to manufacture, it had been considered unsuitable for use in railway lines.  However, when put to the test it was discovered that the impact and friction of traffic rolling along a malleable iron line work-hardened the surface, [9] which caused it to resist rust, while malleable iron’s greater manufacturing cost was more than offset by the much smaller gauge of bar necessary to provide the equivalent strength of cast-iron.

The experience at Tindale Fell came to the attention of the civil engineer Robert Stevenson (1772–1850), who referred to it in a report: [10]


“Before the period alluded to, the rails in use had been almost invariably made of cast-iron or timber; but my father, in his notes, says, ’I have no hesitation in giving a decided preference to malleable iron, formed into bars from twelve to twenty feet in length, with flat sides and parallel edges, or in the simple state in which they come from the rolling mills of the manufacturer.’  He also recommends that they should be fixed into guides or chairs of iron, supported on props placed at distances in no case exceeding three feet, and that they should be connected with a clamp-joint, so as to preserve the whole strength of the material.  It is not a little singular that this description, given about forty years ago [1818], may, to use engineering phraseology, be not inaptly called a ‘specification of the permanent way’ of our best railways at the present day.”

Biographical Sketch of the Late Robert Stevenson: Civil Engineer, Alan Stevenson (pub. 1861).


So wrote Stevenson Jnr. of his father, known principally for his construction of the Bell Rock and other northern lighthouses, although he was also involved in a range of civil engineering work.  George Stephenson obtained a copy of Stevenson’s report, which he passed to Michael Longridge, part owner of the Bedlington Ironworks, [11] who in turn passed it to John Birkinshaw, the works’ principal agent; and to Birkinshaw must go most of the credit for establishing the Bedlington company’s reputation:


“The Bedlington Iron and Engine Works will be remembered in industrial history more for the contribution made to the development of the early railways than for any other single reason.  It is no coincidence that the companys peak of production and fame was paralleled by the excitement of railroad and locomotive development in this country and abroad.”

The North Eastern Railway, William Weaver Tomlinson (1915).


Birkinshaw contacted Lord Carlisle’s agent for information on the use of malleable iron rails, and was told that:


“Our rails are one and a half inches square, and stand upon stones about ten inches square, and are placed at one yard distance from centre hole to centre hole.  Our railway carries four tons weight, and has never cost us any thing yet, as to expense of the malleable iron, except creasing [track maintenance].  The iron I cannot see the least alteration with, although it has now been laid eight years.  The cast-iron is a daily expense; it is breaking every day.”


In 1820, Birkinshaw registered a patent ― and extract of which is at Appendix II. ― for rolling rails of malleable iron:


“These rails are generally rolled into lengths of fifteen feet, subdivided into bearing lengths of three feet each; eighteen feet lengths were recommended by the patentee, but experience has shewn that the former are the most practicable.  The joinings of the ends of these rails, were at first square at the ends, similar to the old cast-iron rails; but they are now formed with a half lap . . . . and thus they now possess all the properties of the improved cast-iron rails.”

A Practical Treatise on Rail-roads, Nicholas Wood (1836).


This was a vital breakthrough in railway engineering, for this new type of rail could withstand the stresses caused by locomotive movement.


“KlLLINGWORTH COLLIERY,
June 28, 1821.

Robert Stevenson, Esq.

Sir, ― With this you will receive 3 copies of a specification of a patent malleable-iron rail invented by John Birkinshaw of Bedlington, near Morpeth.  The hints were got from your Report on Railways, which you were so kind as to send me by favour of Mr Cookson some time ago.  Your reference to Tindale-fell Railway led the inventor to make some experiments on malleable-iron bars, the result of which convinced him of the superiority of the malleable over the cast-iron ― so much so, that he took out a patent.  Those rails are so much liked in this neighbourhood, that I think in a short time they will do away the cast-iron railways.  They make a fine line for our engines, as there are so few joints compared with the other.  I have lately started a new locomotive engine, with some improvements on the others which you saw: it has far surpassed my expectations.  I am confident a railway on which my engines can work is far superior to a canal.  On a long and favourable railway I would stent my engines to travel 60 miles per day with from 40 to 60 tons of goods.  They would work nearly fourfold cheaper than horses where coals are not very costly.

I merely make these observations, as I know you have been at more trouble than any man I know of in searching into the utility of railways; and I return you my sincere thanks for your favour by Mr Cookson.  If you should be in this neighbourhood, I hope you would not pass Killingworth Colliery, as I should be extremely glad if you could spend a day or two with me, ― I am, sir, yours most respectfully,

(Signed) G. STEPHENSON


Biographical Sketch of the Late Robert Stevenson: Civil Engineer
, Alan Stevenson (pub. 1861).


――――♦――――

 

 
RAILS FOR THE STOCKTON AND DARLINGTON


Prior to Stephenson being appointed Engineer to the Stockton and Darlington Railway Company, the directors reached two important decisions; that the line was to be a railway laid with edge rails, as opposed to a tramway laid with plate rails, and that Stephenson was to undertake a further survey of the line.  Both decisions were possibly assisted by a letter (dated 22nd June 1821) to one of the Committee members from that great promoter of railways, William James:


“This railway pioneer [James], who, in the capacity of engineer of the Stratford and Moreton Railway, had visited most of the railroads in the kingdom, described the edge-rail as ‘infinitely preferable’ to the plate-rail, and eulogised, in no measured terms, the North-country engineers, ranking Stephenson next to Watt in point of mechanical ability.”

The North Eastern Railway, William Weaver Tomlinson (1915).


Following completion of the survey toward the end of 1821, Stephenson was appointed engineer to the Company.  Despite his connection with William Losh and the Walker Ironworks, he declined to recommend the use of Losh’s patent cast-iron rails, opting instead for Birkinshaw’s patent malleable iron fish-bellied rails of 28 lbs. per yard.  In Stephenson’s opinion:


“The great object in the construction of a railroad is that the materials shall be such as to allow the greatest quantity of work to be done at the least possible expenditure; and that the materials also be of the most durable nature.  In my opinion Birkenshaws patent wrought-iron rail possesses these advantages in a higher degree than any other.  It is evident that such rails can at present be made cheaper than those that are cast, as the former require to be only half the weight of the latter, to afford the same security to the carriages passing over them, while the price of the one material is by no means double that of the other.  Wrought-iron rails, of the same expense, admit of a greater variety in the performance of the work, and employment of the power upon them, as the speed of the carriages may be increased to a very high velocity without any risk of breaking the rails; their toughness rendering them less liable to fracture from an impulsive force, or a sudden jerk.  To have the same advantages in this respect, the cast-iron rails would require to be of enormous weight, increasing of course the original cost.”

A Practical Treatise on the Construction and Formation of Railways, James Day (1848).


But malleable iron had yet to win the day, for after consulting a number of eminent engineers the Committee concluded unanimously that two-thirds of the railway should be laid with malleable iron and the remainder with cast-iron, the chairs in both cases to be of cast-iron. [12]  By the 8th July 1823, the Belington Ironworks had delivered 900 tons of malleable iron rails, [13] while the Neath Abbey Company supplied 243 tons of cast-iron rails, chairs, and crossing plates. [14]

The sleepers posed a similar problem to the rails ― uncertainty.  At the outset there was a considerable discussion as to whether stone blocks or wooden sleepers were more suitable for the permanent way.  Eventually the Committee decided to try both, the stone being sourced from local quarries at Brusselton and timber shipped to the Tees from Portsmouth, where it had been recovered from scrapped wooden-wall warships.  In time it was realised that stone sleepers were too unyielding for the weight of the new locomotives, causing damage to the iron rails, and both they and the wooden blocks were replaced by more compliant transverse wooden sleepers.  The redundant stone blocks served out their days as edgings to the platforms of stations, and in the seawall and slipway at Saltburn-by-the-Sea (a partly successful speculative development carried out by Edward Pease’s youngest son, Henry) where they can still be seen, the fastening holes for the iron chairs being plainly visible.


CHAPTER 4

――――♦――――

 

 
APPENDIX I.

LOSH AND STEPHENSON’S PATENT

from:
Patents for Inventions.
Abridgements of Specifications Relating to Railways,

(1868).

A.D. 1816, September 30. ― No. 4067.


LOSH, William, and STEPHENSON, George. ― “Improvements in the construction of railways and tramways.”*

“The invention relates to edge round-top’d fish-backed, plate tramway, and barrow-way plate rails.  In the construction of our edge railways our objects are, to fix both the ends of rails, or separate pieces of which the ways are formed, unmoveable in or upon the chairs or props by which they are supported, and to place them in such a manner that the end of any rail shall not project above or fall below the corresponding end of that with which it is in contact, or with which it is joined; also to form the joinings of the rails with the pedestals or props which support them in such a manner that if these props should vary from their perpendicular position in the line of the way the joinings of the rails with each other would remain as before such variation, and so that the rails shall bear upon the props as firmly as before.  The formation of the rails or plates, of which a plate railway consists, being different from the rails of which the edge railways are composed, we are obliged to adopt a different manner of joining them, both with each other and with the props and sleepers on which they rest; but in the joining these rails or plates upon their chairs and sleepers we fix them down unmoveable, and in such a manner that the end of one rail or plate does not project above or fall below the end of the adjoining plate, so as to present an obstacle or cause a shock to the wheels of the carriages which pass over them; and we also form the joinings of these rails or plates in such a manner as to prevent the possibility of the nails which are employed in fixing them in their chairs from starting out of their places from the vibration of the plates, or from other causes.”

The patentees also describe improvements in the construction of railway wheels and locomotive engines.

___________


* “A grant unto William Losh, of the town and county of Newcastle-upon-Tyne, iron founder, and George Stephenson, of Killingworth, in the county of Northumberland, engineer, for their invented new method or new methods of facilitating the conveyance of carriages, and all manner of goods and materials along railways and tram-ways, by certain inventions and improvements in the construction of the machine, carriages, carriage wheels, railways and tram-ways employed for that purpose ― 30th September, 1816, Patent Record Office, No 4067.”


Engineers and Mechanics Encyclopedia 1839.


Because the base of the chair is not flat, but forms an arch, should the sleeper tilt, as shown in the diagram, the rail remains tangential to the arch. It is further prevented from protruding by being fixed to its neighbour by a half-lap joint.


――――♦――――


 

APPENDIX II.

JOHN BIRKINSHAW’S PATENT

from
The Engineer’s and Mechanic’s Encyclopedia
Luke Hebert (1836).

 

Specification of the Patent granted to JOHN BIRKINSHAW, of Bedlington Iron-Works in the County of Durham; for an Improvement in the Construction of malleable Iron Rails, to be used in Rail Roads, whereby the Cost is reduced, and the Expense of Repairs of broken Rails saved. Dated October 23, 1820.

“My invention consists in the adaptation of wrought or malleable iron bars or rails of a peculiar form, instead of cast-iron rails, as heretofore.  From the brittle nature of cast-iron, it has been found, by experience, necessary to make the bars of a railroad sufficiently strong to bear at least six times the weight intended to be carried along the road, by which the original cost of a railroad was considerably augmented; or if light rails were used, the necessity of frequently repairing entailed a heavy expense upon the proprietors.

To obviate these objections, I have invented a bar to be made of wrought, or malleable iron, the original cost of which will be less than the ordinary cast-iron rails or bars, and, at the same time, will be found to require little (if any) reparation in the course of many years.  The rails or bars which I have invented are formed as prisms, though their sides need not of necessity be flat. Figs. 1 and 2 show sections of the bar thus formed; the upper surface upon which the wheel of the carriage is to run is slightly convex, in order to reduce the friction; and the under part rests in the supporting-blocks, chairs, rests, standards, or pedestals, which are mounted upon the sleepers.  The wedge-form is proposed, because the strengths of the rail is always in proportion to the square of its breadth and depth.  Hence this form possesses all the strength of a cube equal to its square, with only half the quantity of metal, and, consequently, half the cost.  Sufficient strength, however, may be still retained, and the weight of metal further reduced, by forming the bars with concave sides, as shown in section, by Figs. 3 and 4.  The mode of making iron bars of a great variety of forms, we have already generally explained in our account of the iron manufacture.”


――――♦――――


 

FOOTNOTES

1.

‘Hammer blow’ refers to the vertical forces transferred to the track by a locomotive’s driving wheels.  Although most is due to the unbalanced reciprocating motion, which in slow-moving early locomotives would not have been significant, the piston thrusts also contribute to it.  The result is that rails are subjected to an intense and regular pounding, which can cause damage.

2.

There is some dispute over the invention of cast-iron rails.  They are generally credited to John Curr, being first used underground in mines at Sheffield c. 1776, only later being used overground.  Another claim is that cast-iron rails were used at Coalbrookdale c.1768.

3.

The cast-iron waggon wheel was really a north-country invention.  As early as may, 1731, Elias Thornhill of Sunderland, whitesmith, obtained a grant of a patent for ’his new invention of making the rim or edge of coal waggon wheels with iron or steel and with iron ribs or tabbs and iron bolts, rivets, and screws for the fastening the same.’ (Archaeologia Aeliana, vol. xxiv., p. 226)

The North Eastern Railway, William Weaver Tomlinson (1915).

4.

Commenting on the Dowlais-Merthyr railway, William Taitt of Dowlais wrote on 17th March 1791:


“We are now making Rails for our own Waggon way which weigh 44 li or 45 li [pounds] per yard. The Rails are 6 feet long, 3 pin holes in them, mitred at the ends, 3 Inches broad Bottom, 2½ In. top & near 2 In. thick . . . .”


The rails were spiked to transverse wooden sleepers.

5.

The Pen-y-Darren tramroad was a single track plateway with a gauge of 4 feet 4 inches over the flanges of the L-shaped cast-iron plate rails, or 4 feet 2 inches between the inside of the flanges.  The plates were 3 feet long, weighed 56 pounds each, and were spiked to rough stone blocks about 18 inches square.

6.

Most writers on this subject credit Jessop with the first use of the iron edge rail, but flanged wheels on wooden rails were in use in the North East of England well before:


“The flanged cast-iron wheel had been used in the North of England for half a century before Jessop introduced the iron edge rail, and was not only mentioned by Bishop Pococke in 1760, but described and illustrated in 1765 by Monsieur Jars, who actually gives the depth of the flange, viz., from an inch to an inch and a half (Voyages Métallurgiques, vol i., p.202). . . . The flanged waggon wheels had merely to be transferred from the wooden to the iron rails when the latter were laid down.”

The North Eastern Railway, William Weaver Tomlinson (1915).

7.

Following Outram’s death in 1805, ‘Benjamin Outram and Company’ was rename the ‘Butterley Company’, and continued trading (as an engineering company) until 2009.

8.

Malleable iron is a form of cast-iron that is easier to work with than pure iron.  It is made by melting scrap steel and pig iron and then carefully controlling the cooling of the mixture over many hours.  This results in an iron that is very tough but not brittle.

9.

In metallurgy: work hardening is the increase in hardness of a metal induced, deliberately or accidentally, by hammering, rolling, drawing, or other physical processes.

10.

One point, however, deserves particular notice here, as likely to be attended with the most important advantage to the railway system, which is the application of malleable iron instead of cast-iron rails.  Three miles and a half of this description of railway have been in use for about eight years on Lord Carlisle’s works, at Tindal Fell, in Cumberland, where there are also two miles of cast-iron rail; but the malleable iron road is found to answer better in every respect.  Experiments with malleable iron rails have also been made at Mr. Taylor’s Works, at Ayr, and Sir John Hope’s at Pinkie; and, upon the whole, this method, as in the case of the Tindal Fell Railway, is not only considerably cheaper in the first cost than the cast-iron railway, but is also much less liable to accident.  In the use of malleable iron bars, the joints of the railway are conveniently obtained about twelve feet apart, and three pedestals are generally between each pair of joints.”

Report of a Proposed Railway from the Coal-field of Mid-Lothian to the City of Edinburgh p.26 (1818).

11.

Soon to enter partnership with the Stephensons in the Newcastle locomotive manufacturing firm of Robert Stephenson & Company.

12.

George Stephenson to William James, postmarked 20th December, 1821:


"With respect to the Stockton and Darlington Railway Company advertising for cast-iron rails, it was merely to please a few of the subscribers who have been brought over by some of the cast-iron founders, but they have only advertised for one-third to be cast-iron."

The Two James’s and the Two Stephensons ( 1861) p. 48.

13.

“The specifications for the malleable iron rails prescribed that they should be fifty-six pounds per double yard; that the breadth of the top of the rail should be two and one-fourth inches, and the depth at the end two inches; that the depth at the middle should be three and one-fourth inches; that the depth at the top flange should be three-quarters of an inch; that the thickness of the web at the top should be three-quarters of an inch; that the thickness of the web at the bottom should be half an inch; that the edge should be rounded and the surface flat; that the rails should be perfectly straight, and fit to the chairs accurately; and that a sample rail and chair, or patterns thereof, should be furnished to the company.”

A History of the Stockton and Darlington Railway, J. S. Jeans (1875).

14.

“The weight and dimensions prescribed for the cast-iron rails and accessory chairs were as follow:― ‘The length of each rail to be 4 feet, cast from good pig-iron; the weight per double yard to be 115 lbs.; the weight of the chairs to be 10 lbs. each, or 15 lbs. per double yard; the breadth at the top of the rail to be 2¼ inches; the depth at the end to be 4 inches; the depth at the middle to be 6 inches; the depth of the top flange to be 1 inch; the thickness of the web at the top to be five-eighths of an inch.’”

A History of the Stockton and Darlington Railway, J. S. Jeans (1875).

 



[Home] [Up] [Maps and Plan] [Main Index] [Site Search]

Correspondence should be addressed to.... Webmaster@Gerald-Massey.org.uk