The Railways

  Far from detecting the frailty of his bridge, Stephenson had inspected it routinely on the very day of the collapse and ordered that an extra layer of ballast be spread across its timber decks as a precaution against fire. The ballast was in place by afternoon, adding an extra nine-ton load on each girder. Had Stephenson been young and with everything still to prove, his engineering career would probably have ended there; as it was, he was lucky to escape a manslaughter charge. Fellow engineers lined up in support of his convenient theory that the disaster had arisen from a derailment.

  The Dee affair showed that the immediate future of iron bridge spans lay with wrought iron. Stephenson used the material in two of his most famous bridges, further along the Chester & Holyhead Railway: at Conwy and the Britannia Bridge that crosses the Menai Strait to Anglesey. Both were tubular box-girder bridges, most unlike anything seen on Britain’s railways before. The trains ran not on top of each rigid linear box, like the road traffic on a modern example such as the 1990s Skye Bridge, but through the dark tunnel of the interior. For the traveller, the effect at Conwy is as startling as a solar eclipse: instead of the grand views anticipated, the bridge offers only a plunge into 400ft of noise and darkness. (The approach is peculiar enough as it is: trains run right under the ramparts of Edward I’s fortified town, the castellations of which are extravagantly imitated on the bridge portals.)

  Crossing the Britannia Bridge used to involve a similar journey through darkened iron tubes, raised in this case over 100ft above the water – another case of demands imposed by the need for free navigation below – and extending together for more than 1,500ft. Stephenson was not entirely confident that these flat spans would be self-sufficient, so he carried up the stone piers of his bridge sufficiently high to allow the addition of suspension chains. A consultant engineer then persuaded him that chains were superfluous, and so the bridge never received them. So things stood until 23 May 1970, when some boys set off down the tubes, ostensibly on a quest for birds’ eggs. To light their way, they used burning newspapers as torches – not a prudent technique in a 120-year-old structure with a timber roof sealed with tarred hessian. The boys escaped the resulting inferno, in which the white-hot tubes split along their original joins, sagging by twenty-nine inches at the centre. A total loss, Stephenson’s spans were reconstructed in steel with arched supports, the Admiralty no longer insisting by this date on clearance sufficient for square-rigged men-o’-war.

  Soon after its reopening, a road deck was added along the top of the railway at the Britannia Bridge. This recreated in reversed form the arrangement of another of Stephenson’s innovative railway crossings, the High Level Bridge of 1845–9 across the Tyne to Newcastle. Here the railway deck is supported on shallow wrought-iron arches, from which the horizontal road deck is suspended on vertical hangers. It is an efficient as well as an elegant solution – two bridges for the price of one – but the formula was never repeated elsewhere in Britain.

  Other rare birds included a type of bridge invented by Brunel which depended on the suspension principle, as used at Chepstow (which has been rebuilt) and most splendidly at the Royal Albert Bridge of 1853–9 at Saltash. The suspension bridge had been brought to something like perfection by Thomas Telford as recently as the 1820s, in the graceful form of the Menai Bridge that still shadows Stephenson’s mutilated Britannia Bridge between Caernarvonshire and Anglesey. But the concentrated loads imposed by trains were generally considered to rule out bridges of the springy, flexible suspension type. The Stockton & Darlington built one over the Tees, it is true, but this deflected alarmingly under the weight of coal trains – by as much as twelve inches, Brunel reckoned, when he visited in 1831 – and it was soon replaced by a girder bridge. When Brunel came back to the suspension principle for railways, he reconfigured it by using big iron tubes for rigidity and strength. At Saltash, where high clearances were again demanded by the Admiralty, the iron tubes are of such broad span – two of 455ft, each longer than most cathedrals – that only one pier is needed in mid river. The tubes are arched elliptically between the piers, with vertical hangers supporting the bridge decks, together forming closed suspension spans. The bridge was a success technically, financially and aesthetically, but no one ever built another like it. Like Stephenson’s box-girder structures, it belongs to a period of exceptional experiment and diversity, after which things settled down on a more even course.

  The main line of development lay with the various types of wroughtiron trussed girder. Some of these were of closely latticed form, others had larger members in criss-cross or zig-zag patterns. Trains ran sometimes on a deck on top of the trusses, sometimes between them, making a flickering blur of the view of river, estuary, road or valley below when travelling at speed. Another common type used the bow-string principle: the upper member of the truss describes an arch, with rigid ties to the lower beam. Others had solid sides of riveted plates.

  Flat or flat-bottomed trusses were especially useful in urban settings, where they allowed railways to cross streets with full-height clearances for pedestrians on both sides. This represented an advance on the early type used at Bermondsey, with its three arches for the roadway and pavements. Which is not to say that these dank undersides made for a congenial environment for anyone except the rough sleeper and the feral pigeon. When Morrissey sings ‘Under the iron bridge we kissed’, the picture is instantly there: the underbelly of the inner-city railway, with its own discouraging smell and its temperature usually a few degrees below that of the streets around. Morrissey’s Manchester, with its multitudinous railways snaking towards the city centre across flat terrain, is especially blighted (by contrast, its subterranean approaches make inner Liverpool unusually railway-free). It is strange to read of the wonder such bridges inspired when they were new, whether they took the railway under the road or over it. The 1850 edition of Black’s Picturesque Tourist and Road and Railway Guide Book catches this sentiment, commending the first mile of the Bolton & Preston Railway with its nine ‘much admired’ bridges of iron beams. As with skew bridges, a technical achievement that was striking when new has become banal by familiarity and repetition.

  Quirks and novelties multiplied in the construction of iron bridges. The crossing of navigable waterways was often the cause, especially in lowlying country. Swing bridges, including the straightforward balanced type pioneered by G. P. Bidder, were far from rare. There were also joint road–rail swing bridges, lifting bridges, rolling lift bridges, telescopic bridges and railway drawbridges. Special arrangements were required each time to allow signalling and telegraphs to be broken and reconnected. At the second Arun Bridge at Ford in Sussex, which carried water and gas pipes as well, railway traffic had to stop for up to forty minutes when the telescopic section was winched open to let a boat pass.

  With or without an opening span, a combination of straight trussed girders and bowstring trusses was a good way to take a long bridge (strictly, a viaduct) across a wide expanse of water. That way, the wider bowstring trusses could accommodate the deeper flows of the central section, at once making space for larger vessels and reducing the number of costly piles that had to be sunk far below the surface. The grandest example is the second Tay Bridge of 1881–7, Britain’s longest railway bridge, designed by the partnership of William Henry Barlow and his son Crawford. For seventy-one of its eighty-four spans, the passenger has a clear view towards Dundee on the north shore and Fife on the south side. These outer sections have trussed girders spanning beneath the carriage. Then the thirteen navigation spans are reached and the prospect is suddenly interrupted by the rapid uprights and diagonals of the bowstrings, before the straight spans resume for the final stretch of the elevated two-mile journey.

  The first Tay Bridge, photographed in the aftermath of its collapse in December 1879

  That is not all. Just upstream of the crossing are the stubs of the first Tay Bridge, designed by Sir Thomas Bouch, which failed catastrophically one stormy Sunday night in December 1879, wh
en not much more than a year old. The collapse took out the navigation spans, which were of the same length as those of the present bridge. The trussed-girder spans match too, for the very good reason that the North British Railway reused what could be salvaged from the old bridge, this time set on broader and sturdier piers. The sight of the indecently close and prominent stumps, which the North British chose not to remove, inspires uneasy thoughts even today, but these are as nothing compared to the thrill of horror when the collapse occurred. For the spans took with them an entire train with all its passengers and crew: a loss as complete as if a ship had foundered in mid ocean. Nobody on the land even saw it happen, except for one or two who observed the puzzling descent of some lights out on the darkened water – either the fire of the locomotive, or lamps still burning in the falling carriages. Even the number of fatalities was uncertain. Three hundred were rumoured at first, but the final estimate was given as seventy-five, calculated on the basis of the sale of fifty-nine tickets for the journey plus an allowance for any children and railwaymen, who travelled ticket-free. (This total has since been questioned, most recently in 2011 following a local campaign for a permanent memorial to the victims, inscribed with every proven name.) As a catastrophic failure of transport technology at its most ambitious and up-to-date, the episode had something of the impact of the Titanic disaster thirty-three years later.

  Widely admired when new for its daring and structural economy, Bouch’s design proved inadequate in terms of resistance to the winter gales that roared through the firth. On top of that, it was carelessly built. It seems heartless to add to these lethal faults that it was also rather ugly – its successor too. The same goes for many more of the major iron bridges built for the Victorian railways. The most spectacular of these have vanished in their turn, not after structural failure, but because the lines that they served have closed.

  Bouch was responsible for two of the biggest examples, at Belah and Deepdale, on the line sponsored by the Stockton & Darlington, which was pushed across the north Pennines in 1857–61. The route’s lifeblood was the rich iron ores of the Furness division, required by the ironworks at Middlesbrough, and the coke that flowed in the other direction to fuel the furnaces at Barrow and Millom. This was wild, harsh country: Deepdale Viaduct had a wind gauge which rang a warning bell in extreme weather, so that an extra speed restriction could be imposed. Belah Viaduct was the tallest ever built in England, at 196ft. Beautiful, however, it was not; in photographs, the viaduct has a spindly and temporary look, like something run up from scaffolding. The same applies to the Crumlin Viaduct of 1853–7 across the Ebbw Valley in Monmouthshire, which at 200ft held the British record for height. Its design was sophisticated, as the first large-scale implementation of the Warren triangular girder principle: stresses were distributed in such a way that every member was subject only to tension or compression, rather than sideways forces that might cause bending or shearing. What the viaduct looked like in its final days between closure and demolition is captured in the comedy thriller Arabesque (1965), in which Gregory Peck and Sophia Loren are pursued along the triangulated spans by rifle shots from a helicopter containing an armed and dangerous Alan Badel.

  The last entry on the roll-call of great railway bridges of Britain introduced a new material, steel, and a new structural principle, that of the balanced cantilever. This is the Forth Bridge, built in 1883–90 in succession to the bridge planned by Bouch, and probably still the most famous railway bridge in the world. Its co-designers were the veteran English engineer John Fowler, born in 1817, and a younger man, Benjamin Baker. Both were knighted on its completion. For good measure, the contractor William Arrol, the resourceful son of a Scottish blacksmith, was knighted too. It was as if the shameful memory of the knighthood conferred on Sir Thomas Bouch a few months before his disgrace could now be purged. Work had actually started on Bouch’s own crossing of the Forth, which aimed to revive the long-abandoned suspension type. The fall of the Tay Bridge killed this immediately, but Bouch’s workshops were kept, and enlarged for use by his successors. Not everyone admired the mighty new bridge that resulted; William Morris, opposed on Ruskinian principle to building in iron or steel, called it ‘that supreme specimen of ugliness’. Baker’s response was that his bridge had a visual honesty that was meant to be graceful and comprehensible. He must have been cheered to receive a letter from the eminent architect Alfred Waterhouse, praising the absence of ornament that made the bridge ‘a style unto itself’, with ‘a beauty of its own’.

  Waterhouse’s sentiments have a modernist ring to them. However, the decline in railway building meant that opportunities to forge a new aesthetic for twentieth-century railway bridges were limited. When large-scale bridge building resumed after 1945, the beneficiaries were the roads, not the railways. Nor was Britain any longer one of the chief sites of innovation. New technologies such as welding and pre-stressed concrete were adopted in time, but often as an exercise in catching up. The emphasis also shifted away from new lines and towards the modernisation of old ones. For the post-war engineer of railway bridges, the hardest challenge was often how to construct and install a crossing on an existing line with the minimum delay to traffic. Increasingly, such bridges were prefabricated in whole or part, so that they could be slid into place sideways and as fast as possible. Sometimes this was done to replace a bridge that had become worn out, or to increase clearances for overhead electrification. But many of the biggest post-war railway bridges – at least until the Channel Tunnel line came along – were needed in order to carry lines under or over the new trunk roads and motorways. One of the railway marvels of 1869 was the new Runcorn Bridge on the London & North Western, a spectacular lattice-girder design, which allowed a more direct route across the Mersey from London and the south. Exactly a hundred years later, Railway World magazine proudly reported that British Rail’s largest single-span concrete bridge to date had just been inched into place to take the Runcorn-to-Frodsham line over the new M6, a few miles away from the Runcorn Bridge proper. As an engineering exercise this was notable, but this time there was no advantage to the railway – quite the reverse, as the new road route favoured the competition.

  Trackbed, verges, cuttings, embankments, buildings: all these components of the network required cumulatively a vast amount of land, much of it taken straight out of fruitful agricultural production. The average area needed to make a single route-mile has been calculated at some eleven acres. Where lines clustered together, the impact could be drastic. By 1911, a section drawn across the Taff Valley at a certain point south of Pontypridd transected railways running at six different levels. Other acquisitions of terrain resulted from the common practice by which an extra curve was inserted where two lines branched, to make a triangular junction. The company usually had to buy up the land enclosed, which often became inaccessible. There is an extreme example of this sort of useless purchase at Norwood Junction, in the outer suburbs of South London. Half a dozen lines split and branch and rejoin here, carving up into triangles and segments an area of Surrey otherwise big enough to host a giant housing estate. All around this unofficial nature reserve are the inter-war streets that filled up the district after the railways arrived. By 1873, the landholdings of the companies of England and Wales alone already amounted to 109,762 acres, comfortably more than the entire county of Rutland. A high proportion of this was green, open land.

  Presented with the conundrum of what to do with these elongated territories, some railways went in for a bit of farming themselves. Posterity has remembered the agricultural use of railway land chiefly in terms of wartime emergency, when the acreage of land in Britain used for food production rose by an astonishing 80 per cent. In particular, the Dig For Victory campaign encouraged temporary allotments on railway land within easy reach of willing urbanites. Self-seeded, their crops sometimes outlived the war: in the 1970s the nature writer Richard Mabey found ‘bizarre horticultural relics’ on the Metropolitan’s embankments at Neasden and
Brent: ‘sprawling loganberry bushes, forests of perpetual spinach and one vast and still-sprouting asparagus crown over a yard wide’. That these useful and rewarding plots had to be given up after the end of hostilities says a great deal about the tight control Britain’s railways have always exercised over their land.

  There were earlier attempts to farm alongside the rails, on a scale beyond that of urban allotments. The most straightforward use of the land was for fodder crops (for obvious reasons, grazing livestock was out of the question). Cutting of the lineside of the Liverpool & Manchester for hay was noted as early as 1837, but the practice seems never to have become general. Lineside fires caused by sparks probably had something to do with this. Even so, there were exceptions. At Eckington in Worcestershire the landowner’s settlement with the Bristol & Gloucester Railway allowed him the use of both sides of the embankment, which therefore had lineside fencing along the top of the slope rather than at the bottom. Financial stringency explains the hay-cropping on the Garstang & Knott End Railway in Lancashire, a badly run mid-Victorian branch line that had over-reached itself by needlessly buying enough land to put down double tracks. The railway’s staff cultivated the unused strip and local farmers were invited to buy a share of the resulting haystacks. The much grander Great Central Railway achieved a return from its newly made embankment at Eydon in Northamptonshire from 1899: the slopes were sown with lucerne, which was mown by the platelayers and sold to farmers where it lay. The practice was abandoned after a few years, because too many weeds were creeping in. Hay-cropping beside the tracks came back during the Second World War – along the London, Midland & Scottish alone, 440 farmers took up the company’s offer in 1940 – but once again it failed to endure.