Speed Kills: Salisbury, 1906

About Forums Den of Writers Blogs Speed Kills: Salisbury, 1906

Viewing 8 posts - 1 through 8 (of 8 total)
  • Author
  • #3684

    One reason that railway accidents make such sensational news is because, in contrast to the daily carnage on the roads, they are so rare. It was once said that you are safer in a British railway train than almost anywhere else. And one reason for that is the responsibility and care with which the vast majority of British railwaymen have always carried out their duties. They wouldn’t do anything so reckless as racing, would they? With trains full of passengers?

    Well, not recently, but there have been times when they came pretty close to it….

    Until the Government-enforced Grouping of 1923, which combined all railways in Britain into four large groups (‘The Big Four’), there were over a hundred independent railways in Britain. Many were small local lines, but there were about twenty large main-line companies, and in some places competition between them was intense, sometimes to the point of absurdity. At one time, due to the rivalry between the South Eastern Railway and the London Chatham and Dover Railway, nearly every town of any importance in Kent had two railways and two stations, both companies nearly bankrupting themselves in the process. And in south-west Scotland the bitter feud between the Glasgow and South Western Railway (the Wee Sou’ West) and its much larger rival the Caledonian Railway (the Caley) was legendary.

    There were circumstances where cooperation rather than competition was obviously to the benefit of all concerned, and nowhere more so than on the services from London to Scotland. Queen Victoria’s love for Balmoral had resulted in Scotland becoming a fashionable place for the gentry and the wealthy to visit, and the Scotch (in those days the word was not reserved for whisky) expresses were the most prestigious in the country. On both the East Coast and West Coast routes the companies concerned (the Great Northern, the North Eastern and the North British on the east, and the London and North Western and the Caledonian on the west), cooperated with each other to run the trains, and even pooled their resources to build jointly-owned coaches for them. But there was rivalry between the routes. And, despite a widespread distrust of high speed on the part of the Victorian public and a well-known aversion to it on the part of Queen Victoria herself, one way of making a train service more attractive to the public, then as now, was to shorten the journey time. In other words, to make it faster.

    The first of what became known as the Races to the North, that of 1888, wasn’t really a race at all but more of a schedule-cutting war. The East Coast and West Coast companies shortened their timetables between London and Edinburgh alternately, tit for tat, but the trains still kept to them. That didn’t stop the press from sensationalising the competition and calling it a race, nor the enthusiasts (yes, there were those, even then, long before the word anorak had ever been heard of) from getting excited. Eventually common sense prevailed and an uneasy truce descended, until the rivalry flared up again in the summer of 1895.

    The race of 1895 was an altogether more exciting (or irresponsible, depending on how you look at it) affair. The trains concerned were the night expresses that left King’s Cross and Euston at 8pm, and the destination was Aberdeen, or more strictly Kinnaber Junction, near Montrose, where the two routes merged to run on the same tracks for the last 38 miles into Aberdeen. It started like the 1888 affair, with progressively tighter timetables, but by mid-August all restraint had been abandoned, and the timetable with it. Featherweight trains set out from King’s Cross and Euston to reach Aberdeen as soon as humanly possible, crowds came to see them off, and the press and enthusiasts went berserk. Never mind that passengers were being decanted at Aberdeen around five in the morning before anything was open: prestige was at stake.

    The fastest run of all was made by the West Coast companies on the night of 21-22 August, reaching Aberdeen in 8 hours 42 minutes at an average speed of 63.3mph for the 540 miles. That might not seem very fast in the twenty-first century, but the journey included three stops to change engines at Crewe, Carlisle and Perth, which meant running at speeds up to and possibly exceeding 80mph, and in 1895 it is doubtful if anything man-made had ever moved faster. The fastest part of the run was made on the London and North Western Railway between Crewe and Carlisle by 790 Hardwicke, a small locomotive even by the standards of the time, covering the 141 miles, including the stiff climb up Shap Fell in Cumbria, in 126 minutes. The record stood for many years, and the engine is preserved at York.

    That was the last word. Amateur observers and sensation-hungry pressmen may have felt disappointment as the excitement subsided, but those in the know breathed a collective sigh of relief. None of the locomotives involved had speedometers (very few British steam engines ever did), and so speed limits at curves and junctions were nominal, their observance down to the drivers’ judgement and experience. As the competition hotted up drivers had been pushing nearer and nearer to the limits of safety, and there is little doubt that at times the trains were a hairsbreadth from disaster. The men responsible for running the trains, both the footplate crews and their supervisors, knew well enough that they had been lucky to get away with it.

    The racing left a legacy of some very fast schedules by the standards of the day, but all that came to an end almost exactly one year later, when one of the very trains involved in the racing, the 8pm Scotch express from Euston, came to grief at Preston on the night of 15 August 1896.

    In 1896 Sir Richard Moon, the redoubtable chairman who had ruled the London and North Western Railway with a rod of iron, had been retired for some years, but the regime of strict economy he had imposed still lingered. The LNWR would have none of the policy of one man, one engine, one duty that was practised by many railways in Victorian times, at least for express trains. It was wasteful of both personnel and locomotive resources – it meant, for example, that all the time a driver was not at work his locomotive would be sitting in the shed doing nothing – and on the North Western waste was anathema. There was no reserving of the most experienced crews or the best engines for prestigious trains: the first available crew was allocated to the first available engine, and the combination allocated to the next train. So it came about that on 15 August the Scotch express, double-headed, was in the charge of two drivers neither of whom had ever driven that train before or had ever run through Preston Station without stopping. But what both men did know was that they had a tough job on their hands to keep time on the tight schedule.

    At the northern end of Preston Station there was a very sharp curve with a speed limit of only 10mph. Hardwicke had somehow managed to get through without incident the previous summer, but this train, according to witnesses, came through the station at about 45mph and didn’t slow down at all for the curve. What followed was a miracle of good luck. If either engine had turned over, as might have been expected, the carriages of the train would have piled up against the obstruction in a mangled heap of wreckage, but that didn’t happen. Although both engines derailed immediately both stayed upright, ploughing straight on across the tracks of the station yard. The carriages were not so fortunate, but even so only one person was killed.

    That didn’t stop the storm from breaking over the railway companies’ heads, and there was much shaking of heads and cries of ‘I told you so’. Bowing to public opinion, both routes decelerated their Scottish expresses, and the schedules remained static for nearly forty years, by which time they were among the slowest express schedules in the country.

    While peace descended on the lines to the north, rivalry was brewing in the south-west of England. Around the turn of the century some transatlantic liners had taken to calling at Plymouth, and since trains are faster than ships, this gave the opportunity for urgent mail and passengers in a hurry to reach London a day earlier by leaving the ship at Plymouth instead of at Southampton. Once again the prestige of serving the upper echelons of society was involved. The transatlantic liners were then the most prestigious and luxurious (and expensive) form of transport in the world – if you went first class, that is.

    In those days two railways connected Plymouth with London: the Great Western to Paddington and the London and South Western to Waterloo. Direct competition was avoided by an agreement that the LSWR should carry the passengers and the GWR the mails, but since the whole point of the exercise was to reach London quickly pride and prestige demanded that the boat trains be run very smartly, and a speed contest between the two railways developed nonetheless. The much-quoted words of George Jackson Churchward, the blunt-spoken locomotive superintendent who was revolutionising the GWR’s motive power at the time, suggest that, on that railway at least, this had official sanction. ‘Withhold any attempt at a maximum speed until I give you the word, then you can go and break your bloody necks.’

    Apparently he did give the word, for on 9 May 1904 the GWR ran an Ocean Mail special the 247 miles from Plymouth to London in 3 hours 47 minutes, averaging just over 65 mph. Not only did this include a stop to change engines at Bristol (at that time the GWR had no direct route to the West Country and all its trains went via Bristol, earning it the derisive nickname of the Great Way Round), but the line between Plymouth and Exeter had, and still has, a lot of curves and some of the steepest gradients of any British main line, making fast running impossible. So from Exeter onwards the train had to really get a move on, and that is exactly what it did. It was on this run, near Taunton, that the GWR locomotive City of Truro earned railway immortality by touching a speed that has been much disputed over the years but was certainly very close to 100mph.

    Although no necks were broken, bloody or otherwise, that maximum speed was kept quiet for fear of alarming the travelling public, for the Preston derailment was still fresh in many minds; but enough details of the journey were published to make it plain that a very fast run indeed had been achieved. This was a special demonstration run, and though there is no evidence to prove that it was a deliberate challenge to the rival LSWR, there is little doubt that it would have been seen as such in some quarters. There was no direct come-back from the LSWR, but the regular boat specials on both railways continued to run to fast schedules.

    What happened at the eastern end of Salisbury Station in the small hours of 1 July 1906 put a stop to all that. There were resemblances to the Preston accident – a night train, a sharp curve, a driver who’d never driven through the station non-stop before – but the results were very much worse.

    On 30 June 1906 the American Line steamer New York arrived at Plymouth. Forty-three of its passengers, mostly wealthy Americans, were landed by the London and South Western Railway’s tender and boarded the train waiting for them at Stonehouse Pool. The train was a short one, consisting, in order from the engine, of a luggage van, three first class coaches, and a kitchen car/brake van. It departed at 11.01pm.

    Although its schedule was brisk, no sanction had been given to anyone on the London and South Western to break their bloody necks. On the contrary, in June 1904, suspiciously soon after the GWR’s record run, the LSWR’s locomotive superintendent Dugald Drummond, a Scot with a fearsome reputation as a man not to be crossed, had issued the following curt notice:

    Devonport Boat Special.
    On 3rd May, I requested that this train should be run to scheduled timing.
    Any driver running it at higher than scheduled speed, will be taken off his engine.

    When the train arrived at Templecombe (a few miles east of Yeovil) where it stopped to change engines, it was a modest one minute early.

    The boat specials were not a permanent fixture in the timetable but were laid on to meet the transatlantic liners as and when they arrived at Plymouth, so it was not possible to assign regular crews to them and they were manned by whatever express drivers and firemen were available. Driver Robins, on the locomotive that backed onto the train at Templecombe, was an experienced express driver, but this was the first time he had driven a boat special.

    The locomotive that came off the train was a member of the famous T9 class (or ‘Greyhounds’, as they were nicknamed for their speediness) that was so successful that the last one was not withdrawn until the early nineteen-sixties (and is preserved). These engines followed the traditional Victorian practice of placing the boiler as low as possible between the tall driving wheels, but around the turn of the century the need for more power to cope with steadily increasing train loads was forcing designers to start using bigger boilers, too fat to fit between the wheels. Robins’ engine, a member of class L12, had a larger boiler than a T9’s, pitched several inches higher, with consequent effects on its centre of gravity. The L12s were less than two years old and there were only twenty of them. Driver Robins would have been far more accustomed to driving the older and more numerous T9s.

    Robins must have had Mr Drummond’s peremptory warning in mind when, as he was waiting to take over the train, he responded to a comment that the boat train was ‘running well tonight’ with ‘I shan’t get into Waterloo before time, else I shall have to go up to see the governor.’ This may be why he took it rather easy for the first few miles after leaving Templecombe at 1.27, so much so that when the train passed Dinton signal-box, 20 miles out, it was running four minutes late.

    But from then on Robins appears to have realised he was losing time, for the train’s speed increased markedly. Calculations made later, based on signal-box passing times, suggested that it must have approached Salisbury at 70mph or more. By the time it passed Salisbury West box between 1.56 and 1.57 it should have been slowing down, but if it did it was not by much. Signalman Mundy estimated its speed at 60 mph, about 25mph faster than usual for a boat special at this point. Robins was sounding the whistle and had shut off steam, but the brakes were not on.

    They should have been because, as at Preston, there was a sharp curve at the far end of the station, not as severe as the one at Preston but severe enough to be restricted to 30mph. For all other trains passing through Salisbury this was academic, as the passenger trains all stopped there and goods trains seldom if ever even reached that speed, but the boat specials ran through non-stop. It would seem that some of their drivers, no doubt wishing to get a run-up for the long gentle climb that begins just east of Salisbury, had been putting a rather liberal interpretation on that speed restriction, for in the previous couple of years notices had been issued imposing a minimum time of one minute between the Salisbury West and Salisbury East signal-boxes, and repeatedly reminding enginemen not to exceed 30mph through the station. This train, in charge of a man who had never driven a boat special before, was going considerably faster than that.

    At that time in the small hours there were no passengers waiting at Salisbury station, and the only witnesses as the train hurtled through the platforms were four railwaymen. It wasn’t easy for them to guess its speed accurately, as they were all on the westbound platform where the booking hall and station offices were, and their view was masked by the buildings on the island platform and by a milk train that started passing through the station from the east just as the boat train entered from the west. One of them didn’t notice anything unusual, but the other three all thought that it was going faster than normal, and the senior man present, Traffic Inspector Spicer, decided to report the driver.

    In the rearmost coach Guard Harrison was getting uneasy. The train was entering the station and the brakes still hadn’t come on. He could have applied them himself, but he didn’t, only giving them a light touch with the intention of alerting the driver. By the time he’d realised that the driver wasn’t going to react, and it would have to be down to him to do something, it was too late. He had only just closed his hand on the brake handle when a massive jolt threw him across his van.

    The enquiry estimated that the train must have been doing at least 60mph when it reached that 30mph curve, and the luck that had averted catastrophe at Preston was not with the LSWR at Salisbury.

    Careful examination of the track afterwards made it quite clear that the locomotive did not jump the the rails as it hit the curve. What it did do was to heel over until it crashed sidelong into the milk train going the opposite way on the next track. With a closing speed of perhaps 80mph, the result was a scene of appalling devastation. The first three vehicles of the boat train were totally destroyed, their wooden bodies ripped from the frames and smashed to pieces. The fourth fared little better, and the only vehicle remaining more or less intact was the kitchen car/brake van in the rear. Over half of the passengers on board, 24 out of 43, lost their lives, along with both Driver Robins and his fireman. The dead also included the guard of the milk train, several of whose vehicles were also destroyed, and the fireman of an engine that had been standing on a nearby siding minding its own business.

    The chairman of the enquiry, Major J W Pringle, concluded that Guard Harrison could have prevented the accident if only he’d made a decisive brake application. When asked why he hadn’t, Harrison said that he was concerned that a full brake application might divide the train. Why he should have believed that, when he would have been using exactly the same brake system as the driver would have, is hard to imagine. This does sound rather like an excuse.

    Major Pringle seems to have thought so. He regarded Harrison’s evidence as unreliable and considered that it was ‘a matter for regret that he showed himself so little fitted for responsibility’ by not applying the brakes. But it isn’t hard to understand Harrison’s indecision if you put yourself in his place. He would have assumed – indeed, wished to believe – that the driver knew what he was doing. He would have been expecting the brakes to come on at any moment. Quite likely he feared that Driver Robins wouldn’t thank him for interfering with the running of the train. And he had a quarter of a minute or less to make up his mind.

    With both the driver and the fireman killed, any explanation of why the train was going so fast can only be speculative. Nothing was found wrong with the brakes. Examination of the locomotive showed the controls set for drifting (running without steam) and the brake lever in the run (off) position. Witness statements that the driver gave the whistle a prolonged blast as the train approached the station proved that he was awake and alert. Driver Robins was familiar with the road, and other drivers testified at the enquiry that the curve was plainly visible from the cab, even at night, and gave their opinion that it should have been obvious to Robins that it was too sharp to be taken safely at the speed he was driving.

    It is possible that an older locomotive with a lower centre of gravity, such as a T9, might just have made it round that curve. Robins, running late and with a climb coming up, never having driven through Salisbury at speed before, may have decided to risk it and not made due allowance for the higher centre of gravity of the L12 locomotive he was driving. He may also have underestimated his speed. At least, no other explanation that makes sense has ever been put forward.

    There have, however, been some explanations that do not make sense. One canard, started by the press at the time, was that one or more of the passengers might have offered the driver money as an inducement for a fast run. There is no excuse for placing any credence in this silly tale, since the enquiry went to the trouble of thoroughly debunking it and the report is freely available on-line. Guard Harrison testified that no passengers got out of the train at Templecombe, where Driver Robins took it over, so there could have been no contact between passengers and driver until the train reached Waterloo. Other drivers who had driven the boat trains testified that they had never been given tips of any sort.

    Not quite so irresponsible is the suggestion that the crews may have been under pressure from above to make fast runs. It has been speculated that the LSWR may have been put on its mettle by the opening of the GWR’s new direct route to the West Country, cutting out the Great Way Round via Bristol and significantly shortening its journey times to Plymouth. This actually happened the very next day after the accident, but it would have been no secret beforehand.

    But it is highly unlikely to have had any bearing on the Salisbury accident. If anyone did put such pressure on it would have been in direct contravention of the unequivocal instructions of the head of department, Dugald Drummond. And though people have been known to disobey orders, the words Driver Robins spoke before taking the train over show that he had no intention of doing any such thing. It is surprising how so many accounts of the accident mention this theory, in view of the fact that both Drummond’s notice and Robins’ statement are quoted in the report of the enquiry.

    Although there have been several British railway accidents with a higher death toll, there have been few in which a train was so comprehensively wrecked or where a higher proportion of the people on it lost their lives. Public opinion was profoundly shocked, and any thoughts of speed contests between the Great Western and the London and South Western came to an abrupt end, though the GWR’s new shorter route gave it such an advantage that the competition would probably have subsided anyway. And the speed restriction at the eastern end of Salisbury Station was reduced to 15mph, where it remains to this day.

    • This topic was modified 3 years ago by RichardB.

    Another engaging story, Richard. I have no particular interest in railway history but your adept recounting of these incidents mean that I open your blogs with a sense of anticipation and I am never disappointed. You have a gift for giving us the right amount of background to bring the events to life. Thank you!

    John S Alty

    Another interesting and entertaining story from the world of railways – thank you Richard.


    Fascinating and absorbing stuff as ever, Richard. As well as the wonderful clarity and readability of the technical matters, I find the culture of the railway companies and the way it affected how they ran their services really eye-opening. We have it pretty bad these days, with the worst combination of competition and monopoly, but at least we don’t have companies taking ludicrous risks to win our business


    Well, to be fair, accidents through excessive speed aren’t all caused by the heat of competition. One curve at Morpeth was the scene of three (count them) derailments within thirty years, all in the BR period: 1969, 1884 and 1994. Fortunately only the first caused any fatalities, six of them. And at Eltham Well Hall in 1972 the driver was drunk.

    On the other hand, on one occasion in 1937 the London Midland and Scottish Railway, inheritors of the West Coast main line, gave official sanction to an episode of (as the railway writer W A Tuplin put it) ‘barely credible recklessness.’ And the motive was, yes, prestige and publicity.

    On the other side of the country the London and North Eastern Railway had introduced streamlined expresses and there had been several record-breaking runs, which its publicity department had not been slow to exploit. So when the LMS introduced its own streamliner it used the pre-service press demonstration run as an excuse for an attempt to beat the LNER’s maximum speed record. The attempt was made on a downhill stretch just south of Crewe, where the train was, incredibly, scheduled to enter the station via a series of points and crossovers limited to 20mph.

    What were they thinking? They were thinking about getting one-up on the LNER. They were not thinking about how long it might take to slow down from a speed never reached in normal service.

    114mph had been reached when, less than two miles from Crewe, the brakes were slammed hard on, and the train hurtled down the bank towards the station with flames and smoke coming off the brake blocks. It was still doing over 50mph when it hit the first crossover. Passengers were thrown about and there was a fusillade of breaking crockery from the dining car, but the train stayed on the rails and drew to a halt, more or less safely, in the station. No one was seriously hurt, but it was close, much too close, to being another Salisbury.

    • This reply was modified 3 years ago by RichardB.

    I didn’t know about that one – lucky it wasn’t catastrophic. I admit to a fondness for the LMS streamliners, and, contrary soul that I am, prefer them to the much more successful LNER ones. This is partly because I was given a ‘Coronation Scot’ for my train set in my childhood (including a carriage by my grandmother, which I think was the last Christmas present from her before she died), and partly because to me they look so wonderfully 1930s-futuristic. I’m led to understand that their rather high side area led to them being a inefficient when there was a strong crosswind, and the streamliner bodies were removed after a while.


    The usual reason given is for ease of maintenance in WWII. The streamlined casings were a diktat from the LMS publicity department (the LNER’s got them so we must have them too), and the Chief Mechanical Engineer referred to the batch of five un-streamlined ones he insisted on building ‘for comparison’ as ‘proper’ones.


    Richard, your narrative quality is spot on again. Brilliant read.

    Daeds, when I was knee-high to the proverbial grasshopper, we had a mix of streamliners and more conventional looking engines pulling the expresses that rolled into Newbury from London Victoria. I always hoped it would be a streamlined engine that would carry me off to exotic foreign parts such as Taunton and Watchet exactly because they looked so wonderfully futuristic, even though the first ones rolled into service in the 30s.

Viewing 8 posts - 1 through 8 (of 8 total)
  • You must be logged in to reply to this topic.