Tramway Engineering

Railways in the Canal Age, from contemporary sources

The books of Nicholas Wood (1795-1865), a colliery viewer (manager or engineer), and Thomas Tredgold (1788-1829), an engineering writer, were eagerly read by those wanting to introduce the new transport mode of the steam railway, especially in the United States, even before the success of the Liverpool and Manchester Railway. These treatises represent the acme of tramway engineering in all its wonderful variety and newness, and envisage its use as a system of general land communication, rather than as simply an adjunct to mines and collieries. Their effect was evident in the planning of the enterprises of the 1830's. Every significant American early railway project, including the Camden and Amboy, Delaware and Hudson, Baltimore and Ohio, the South Carolina and the State of Pennsylvania, sent representatives to England to find out the latest information just as these books were published. In fact, the debt to them was even acknowledged in a few cases. When I discuss what these men had to say, I will put complementary material that is not from their hands in square brackets.

The canal age tramway of horse-drawn trains on cast-iron plate rails that was the immediate ancestor of the steam railway really goes back no farther than about 1790, and itself was descended from the wooden railways that had served coal mines and quarries for nearly 200 years, since the end of the 16th century, which evolved in their turn from German mine railways. Lewis (Reference 4) gives an authoritative account of these developments. Wooden railways survived into the 19th century, and were very well-developed on Tyneside. Single waggons, not trains, were characteristic of wooden railways.

Tredgold considered the advantage of railways was that they provided a smooth surface for wheels, and a good track for horses, at the same time. One horse could do the work of eight on a common road. Railways were a worthy successor to the 48,500-mile network of Roman roads in Europe. In Florence, marble wheel-tracks had been used long ago, and Dr Thomas Young (of hieroglyph decoding and wave theory of light fame) had proposed roads paved with iron some years before. On tramways, wheels rotated on their axles, as in road vehicles. Heavy loads could be supported on six or eight wheels, with the loads equalized on pairs of wheels. These were not bogie vehicles, but the use of equalization should be noticed. Later inventors have sometimes claimed this idea, falsely, as their own.

Railways would have the advantage over canals in speed, because the resistance to movement was independent of speed, while it increased as the square of the speed on a canal. It happens that this is not quite true. Since the speed of propagation of a wave in a canal is finite and not large, a canal boat soon overtakes its bow wave and the resistance actually decreases as the boat rides up on it. Nevertheless, attaining this speed requires considerable power, and the resulting disturbance destroys the canal banks, so that attempts to take advantage of it came to naught. On railways, the resistance does increase with speed, lack of rigidity leading to an increase proportional to speed, and wind resistance to its square. Tredgold's rail-roads were far from attaining these speeds, however.

Railways came to Tyneside about 1680, he said, replacing the earlier carriage of coals by panniers on horses. [They flourished about this date, but the introduction was much earlier.] These were wooden railways, but the Surrey Iron Railway, built under the authority of the first railway Acts for a common carrier in 1800 and 1804 was, as its name advertised, of iron. The first segment completed was from Wandsworth to Croydon, 9-1/2 miles. The urban location explains the need for parliamentary authority. In 1805 it was extended to the chalk quarries at Merstham, 8-1/2 miles further. The line was double, the maximum gradient 1 in 120, and a new kind of Z-shaped cast-iron plate rail was used. The engineer of these lines was William Jessop, the leading Civil Engineer of the age, and an early proponent of steam power and railways. Tredgold, a firm advocate of edge rail, says the 'effect was short of edge rail,' meaning that train resistance was greater. The motive power was horses, and the traffic general.

Tredgold uses the term tram-road for plateways, where the wagons had plain wheels, but makes it clear that trams are the vehicles, and the rails are called flat or plate rails. [It appears that tram or dram (in Wales) was a term for an underground coal cart. The word may be related to the Swedish meaning of tram--log or beam--and so may have referred to the construction of the vehicle, in which the corves rested on two beams. The first known use of the word tram is in a will of 1555 (where the reference to any kind of railway is far-fetched), but it did not become general until the end of the 18th century, the earlier terms for railways being quite different and local. The normal Tyneside term was waggonway, and the waggon was always a railway vehicle. Railway and railroad originated in Shropshire. Gangway is confined to Derbyshire and Nottinghamshire.] Tredgold's term rail-road refers to the edgeway, where the wheels are flanged, as distinguished from the L-plate tramway. He enunciates the advantages of this kind of way, principally the reduction of resistance to motion and the heavier loads that can be carried. He then refers to a third kind of railway, Palmer's single-rail, where the carriages are in two parts like panniers, one on each side of the rail, which is suspended at 3' height. The advantage is the elimination of level crossings, when the rail is raised to a height of 10'. [The two examples of Palmer's monorail actually built were materials-handling facities operated by manpower.]

Tredgold's enthusiasm for the rail-road is clear, and he mentions how it spread from the Tyne to Cumberland, then to Yorkshire, and even to Derbyshire. [It actually spread to Shropshire, and radiated from there and from Tyneside. From Derbyshire and beyond into South Wales, the plateway was dominant. Cumberland had the first railway in Britain, a German mining railway in Tudor times.] L-rails, he pointed out, were rather weak for the amount of iron they contained. Gritty plain common-road wheels on plateways soon wore through the chilled surface of the cast iron leading to accelerated wear, and effect that was absent with edge rail, that shed grit. The shortness of cast-iron rails was a great disadvantage that could be overcome with rolled wrought-iron rails, as had been shown by John Birkenshaw of Bedlington Iron Works, Durham. [The first cast-iron edge rails of around 1808 had a flat base, so they resembled the later Stevens and Vignoles rails of a later date. By the use of chairs, much iron could be saved, so T-shaped cast-iron rails were afterwards preferred.] Birkenshaw's rail had a mushroom head, very appropriate for the soft wrought iron. Tredgold suggested correctly that the rail would be stronger with more metal on the side opposite the running head.

In Nicholas Wood we find another strong supporter of the edge rail. In his history of land transport, he states that pack animals were still the predominant mode as late as 1750 in all of Scotland, and parts of England. Even in 1825, pack trains were still found in the Welsh mountains and the Scottish highlands. The modern canal age began with the Sankey Brook improvement in 1755, he states, but railways came to Newcastle sometime in 1602-49. The gauge was four feet [Tyneside gauges were four feet or a bit more, and Shropshire gauges were less than four feet. Four feet was typical, however.], and the vehicles were called waggons. The common-road vehicles they replaced were called wains. The wooden wheels turned on wrought iron axles. The newer edgeways used cast-iron wheels with conical treads 4" wide and ledges (flanges) 1" deep. The axle holes were square, so the axles turned in [brass, originally iron on Tyneside] journals, and the treads were case-hardened [chilled] to resist wear.

Wood gives the date of October 1820 for Birkenshaw's malleable iron rail, and lengths of 12 to 15 feet. [Some say 18 feet. The length was a great advantage over cast iron rails, which could be no longer than 4 feet.] Birkenshaw suggested welding the rails end to end, which would have given a very early continuous welded rail (CWR). R. Stevenson of Edinburgh used malleable iron bars 12' to 20' long as rails, with flat sides and parallel edges, probably on top of wooden rails. J. Hawkes invented a composite rail in 1817, but no details are given. [A cast-iron head was cast on a malleable iron bar. It proved impossible to manufacture reliable rails by this method.] Cast iron had the advantage of a harder surface, but once this wore through, wear was rapid.

[Cast iron for railways came from Coalbrookdale, centre of the Shropshire kind of railways. Iron wheels were introduced around 1729, and Richard Reynolds of Coalbrookdale produced iron rails in 1767. These were flat bars 5 to 7 feet long, 4 inches wide, and 1-1/4 inches thick, with three ears for holes for pinning them down to the wooden rails, which thus became longitudinal sleepers. Wooden rails wore out quickly, especially when iron wheels were used, and had high rolling friction. Iron rails greatly reduced the effort required to move a waggon, and gave much longer life. Where the timber was too flexible, the cast iron fractured. Wrought-iron bars were used in such cases, but were expensive until puddled iron was available after 1784. The flanged plate rails used without timber support, resting on stone blocks, became popular after 1800.]

Railways could be worked by horses, men, gravity, stationary engines, or locomotive engines. [Even sails were suggested, but never seriously used.] The best form of traction was not generally agreed upon at this time. [The most advanced opinion favoured locomotives for level sections, stationary engines for inclines, where the load was not primarily downhill. Trains of waggons were not used until the iron railway made them practical. The word convoy does not refer to a train, but to the brake-handle of a waggon. Gang was used for a train, and gave the word gangway.] Wood states that Blenkinsop used the cog in 1811, Chapman, at Hetton Colliery, a chain in 1812, Brunton, at Butterley, legs in 1813, and finally Blackett, at Wylam Colliery, established the practicality of adhesion. [Of course, Richard Trevithick had shown that friction on a plain rail was practical as early as 1803, and deserves at least as much credit as Blackett, or Wylam's Hedley, who introduced an 8-wheeled locomotive into regular service on that plateway.] Wood carried out experiments to measure train resistance, the first of their kind.

These books are especially valuable in presenting a thorough view of the technology, and the opinions of its users, before the Liverpool and Manchester introduced the modern fast, heavy steam railway. The 1826 Reports of the American engineer, William Strickland, are also very informative. Railways had been a common adjunct of canals and heavy construction projects since the 1770's. William Jessop had pioneered the use of stationary steam engines for drainage and water supply, as well as recommending railways where they could be useful. An understanding of this technology is necessary for the appreciation of the rapid engineering progress that occurred in the 1830's, since railway engineering was not without ample precedents in canal engineering.

In fact, the comprehensive article on Canals in Rees' Cyclopaedia (1820) [perhaps written by John Farey] demonstrates that railways were considered as a part of canal engineering, and were already widely used, both as relatively level roads, and as inclined planes. [There are no page numbers in the Cyclopaedia, making references difficult.] The same is evident in William Chapman's Observations on the Various Systems of Canal Navigation (1797). Rees associates the origin of railways with the coal trade of Tyneside. Coals had long come from the mines to Tyneside in carts or wains, or in panniers on horseback, where it was transferred into Tyne keels. The keels carried the coal below Tyne Bridge, where it was loaded into coastal traders for delivery to eastern England and London. The rapidly increasing trade towards the end of the 17th century led to the introduction of wooden railways, around 1680, made of beechwood logs smoothed and rounded on the tops pegged to transverse log sleepers, embedded in earth. The large waggons, pulled singly by one horse on the descending gradient, had wheels with hollow treads and flanges 2" deep on the insides, and a capacity of about 3 tons. Railways were introduced slowly, says Rees, their next application being at Whitehaven in 1738. [The Shropshire railways were earlier.] Elevated staithes for gravity loading of boats was soon introduced, with wagon turntables on the elevated tracks for access to different parts of the structure. Railways also ran into foundry yards, where the waggons could be loaded and unloaded by cranes.

Attempts to protect the wooden rails by cast-iron plates were abandoned due to breakage of the plates. In 1768, says Rees, Richard Lovell Edgeworth (1744-1817) won the Gold Medal of the Society of Arts for his design of 3 smaller waggons in a train, or gang, to replace one large waggon, which made cast-iron plates practical. [Edgeworth's memoirs, as quoted in the Dictionary of National Biography, says the prize was for a 'perambulator' or land-measuring machine instead. Whatever the prize was for, Edgeworth did propose trains of waggons to spread the load. See Lewis, op. cit., p. 304.] In 1788, he suggested rollers for waggon axles to reduce friction. He projected drawing waggons by chains along the track, moved by stationary engines. Edgeworth also proposed running stagecoaches on railways in 1802, when he suggested four tracks, with fast and slow lines in each direction. Dr Anderson proposed a double-track railway from London to Bath, as well as goods containers that could be forwarded by road at the ends of the railways. In 1794, Samuel Homfray obtained a parliamentary Act for the Cardiff-Merthyr railway, on which Trevithick's locomotive ran on 21 February 1804, drawing 10 tons of iron and some 70 people at nearly 5 mph for 9 miles. The Shropshire canals sent forth railway branches from about 1797, and Benjamin Outram began his career of tramway construction at that time. William Chapman proposed flat-bottomed boats with one or two tracks that could carry up to 8 waggons, saving transloading. [There was a ferry across the Avon for waggons of the stone-carrying tramway at Bath in 1739.] This account of origins is incomplete, and contains a few errors, but is of some value because of Rees' date, and the use of him as a reference by later writers.

Rees gives two examples of the use of tramway plates. The first was on the Caldon Branch of the Trent and Mersey canal, 4 miles to a lime-works, and in service 'long before 1794.' Here, simple L-shaped plates with triangular noses and notches to keep them in line, and two lugs on the inner side for wooden pegs to hold them in place. These plates were laid on longitudinal wooden rails for support, but carried plain, not flanged, wheels. They tended to break at the lug holes, and the ends of the flange were often knocked off. They were 3 ft long, 1-3/4" thick, had two holes 18 in apart, and weighed 42 pounds. The second example was of the Surrey Iron Railway, where the plates had no lugs, simply countersunk square notches at the ends, where they were supported on stone sleepers. An octagonal plug was driven into a hole in the sleeper and drilled for a flat-pointed, square-headed iron spike. A small drain hole was drilled clear through the sleeper below the wooden plug so that water would not collect. If such water froze, the sleeper might be cracked. There is no mention of the iron edge rail in Rees. [The L-shaped tramway plate was introduced by John Curr of Sheffield Park Colliery in about 1787, for underground use in guiding the wheeled trams that carried corves of coal. (Lewis, p. 317)]

The method of constructing a tramway, dramway, or gangway is as follows. In locating the line, the amount of rise and fall is more significant than for a canal, since it is more difficult to move in one direction than in the other on a railway with a gradient, unlike the equal facility on a level canal. Large changes of level can be made by inclined plane, and the location of these must be carefully chosen so the plane can be even and straight. The remaining steps of preparing the foundation is much the same for railways and canals. A width of 4 yards is necessary for a single line, 7 yards for a double. A level bed of gravel is first prepared. Stone sleepers 8-12 in thick and weighing 150-200 pounds, otherwise of any shape, are prepared flat on the bottom, and a flat area is also prepared to receive the tramplate on the top. A hole 1-1/2" diameter and 6" deep is drilled in the flat area.

Longitudinal and transverse gauges are prepared, with pins that will fit snugly in the holes in the tops of the sleepers. The pins of the longitudinal gauge are spaced the length of the tramplates, and those of the transverse gauge are spaced so that the distance between the flanges of the tramplates will be the required amount, usually 4 feet. The word 'gauge' for railway track comes from these tools. The sleepers are then placed one by one using the gauges, and gravel is placed around them and firmly packed. A mason's level is applied to the tops of the gauges to ensure that the track is level across. If it is not, gravel is added to or removed from beneath the sleeper until it is. Wooden plugs are hammered, not too tightly, into the sleepers, and drilled to suit the spikes that will be used to fasten the plates. The plates are then laid, and spiked into place. Finally, the pathways for the horses between the rails, and for the attendants, to one side, are made, taking care that the track is properly drained. This produced a road that was excellent for horses and waggons, moving at a walking pace, providing much cheaper and regular transport than the common road, and a serious competitor of the canal for light traffic. It only proved inadquate for heavy locomotives and fast trains at a later time.

The independent tramways for which statistics are given in Rees are the following: Cardiff and Merthyr, Carmarthenshire, Sirhowy, Surrey Iron and Swansea and Oystermouth, in addition to ten canals with railway branches. There is no mention in Rees, or elsewhere, of the existence of tramways anywhere outside of Great Britain. We understand by a tramway to mean a specially constructed hard road for vehicles expressly made for it that are guided by the road so that the amount of bearing surface is reduced to a minimum. This permits the operation of trains or gangs of vehicles, but we have seen that this may well have been a rather late invention. The employment of metal on metal contact to reduce friction and minimise wear is another later, but essential, development. The very earliest references to tramways may merely refer to hard roads for common wains or carts. Similarly, grooves in Roman streets may have guided wagon wheels, but are not tramways. Wagonways in mines are a special case, and may, in fact, have provided the original idea for the railway, when they were extended short distances from a hillside mine adit to navigable water.

Biographical Notes

Thomas Tredgold (1788-1829) was born near Durham, and apprenticed to a cabinetmaker. While working as a joiner, he taught himself architecture and engineering. In 1820, he published Elementary Principles of Carpentry, the first English work on the strength of timber and rational timber design. This was followed by books on the strength of cast iron, extending the investigations of Thomas Young, heating of buildings, railways (the work discussed here), and the steam engine. Tredgold knew nothing of analytic elasticity theory, and his mathematics were not advanced, so his works can be difficult to understand. He was the father of the engineering discipline Strength of Materials.

Nicholas Wood, M.I.C.E, F.R.S. (1795-1865) was born in County Durham, the son of a tenant farmer. His remarkable abilities were recognised by Sir Thomas Liddell, the owner of Killingworth Colliery, to which he was sent in 1811 to learn the occupation of colliery viewer, or manager. There he became a friend of George Stephenson's, and Robert Stephenson was apprenticed to him from 1819 to 1821. He was Stephenson's colleague in railways and the safety lamp, and a judge at the Rainhill Trials. In 1844 he removed to Hetton Hall, as a partner in the Hetton Coal Company. His book on railways was very well-received, and went through several editions, the last in 1838.

Abraham Rees, D.D., F.R.S. (1743-1825) was born at Llanbrynmair, and became a celebrated nonconformist (Presbyterian) minister, residing in London, and the last to wear a wig while officiating. He was elected a Fellow of the Royal Society, and was a member of the Linnean and American Societies. He revised the cyclopaedia of Ephraim Chambers, which had appeared in 1728 as two folio volumes, the new edition appearing in 1778, and then began its great expansion, which was mainly his own labour. The first installment of the New Cyclopaedia appeared on 2 January 1802, which was complete in 45 quarto volumes (the last six of plates) in 1820.

William Chapman (1749-1832) was born at Whitby, and well-educated. He was a friend of Watt and Boulton, engineer of the Kildare Canal in Ireland, and, with John Rennie, engineer of London and Hull docks, among other activities in the same fields. In 1797, he published Observations on the Various Systems of Canal Navigation. In 1820, he tested the chain-traction locomotive at Hetton Colliery.

Richard Lovell Edgeworth (1744-1817) was born at Bath into a wealthy landowning family whose estate was at Edgeworthstown, Ireland. He attended Trinity College, Dublin and Oxford University, where his fruitless career was marked by dissipation and carousing. He was an MP in the last Irish Parliament. He devised a plan, not executed, of telegraphing race results from Newmarket (for some kind of scam?), and, more seriously, proposed a Tellograph (this is the correct spelling, from Latin tellus, not Greek tele) for telecommunication in Ireland at the time of the French invasion threat in 1797. He invented a land-measuring machine (the Perambulator) for which he won a prize, as well as a turnip cutter and a 1-wheel chaise. In 1817, he wrote an Essay on the construction of roads and railways.

Benjamin Outram (1764-1805) was born at Alfreton, Derbyshire, and named after Benjamin Franklin, a friend of his father's. His second son, Sir James Outram, was Major General in the Indian Army, more famous than his father to the general public. Benjamin founded Butterley Foundry in partnership with William Jessop and others, and the foundry supplied much tramway iron. He was a canal engineer who specialised in tramways, for which he is most famous. His last work was the Surrey Iron Railway. 'Tram' is not short for Outram, of course.


  1. Nicholas Wood, A Practical Treatise on Rail-Roads and Interior Communication in General....(London: 1825)
  2. Thomas Tredgold, A Practical Treatise on Rail-Roads and Carriages shewing the principles of estimating their strength, proportions, expense and annual Produce....(London: 1825)
  3. Abraham Rees, The Cyclopaedia or Universal Dictionary... (London: 1820), Volume 6, Article Canals. The plates are in Vol. 2 of the plates.
  4. M. J. T. Lewis, Early Wooden Railways (London: Routledge and Kegan Paul, 1970) is a modern survey of British and Continental wooden railways, prior to the iron plateways, with a glossary and sources.
  5. There have been excellent recent accounts of some tramways, such as the Little Eaton Gangway, the Cheltenham and Gloucester Tramroad, and the South Wales tramways, in short books and magazine articles.

Return to Railway Index

Composed by J. B. Calvert
Created 13 February 2000
Last revised 13 February 2000