This page uses Cascading Style Sheets to present the content in the best possible manner. If you can see this message, then CSS (or JavaScript) is not enabled in your browser, and the page will not appear as the designer intended.

LMSS Masthead

Fostering Interest in Research & Modelling of the London, Midland & Scottish Railway


Part 1, Methods of Single Line Working

Peter Tatlow, President LMS Society.

Photo of Tyer's No. 6 tablet token instrument at Clachnaharry (HR)
Plate 1/1: Tyer's No. 6 tablet token instrument at Clachnaharry (HR). At the bottom right there is a drawer with a hole in it, which when drawn out has a slot for the tablet. The handle on the left-hand side is used to reseat the tablet once it has been replaced in the magazine. (Author's collection)

Upon its formation in 1923, the LMS inherited a number of single lines, which required special signalling provisions to prevent more than one train being admitted into a section at a time; most obviously those travelling in opposite directions unwittingly meeting in mid-section, with potentially catastrophic results. For example, as well as those on the other divisions, on the Northern Division alone there were to about 950 route miles of singe line.

Enginemen were under strict instructions not to enter a single line section without being in possession of a staff for the line ahead; or, in the case of the procedures to be described below, having had a sight of the staff and having been handed a ticket, or being in possession of an electric token for each section of single line. In all cases the speed at which hand exchange could take place was severely limited by the regulations, such as 10 mph.

Persons appointed to receive the staff or token from and deliver it to the driver at stations etc were limited to the station master, booking clerk and porter signalman. Not all crossing points were necessarily authorised to cross both passenger and freight trains. Some, such as The Mound on the main line (HR), Monikie (CR), Barnhill (CR), (GSWR) and Wigtown (PP&WJR) could only cross a passenger or freight with freight trains, when one of the freight trains had to enter a siding, otherwise barred to passenger trains.

The simplest method, permissible at the time of the grouping, was to work the line by 'only one engine in steam or two or more coupled together'. The usual authority to enter such a branch line was one unique staff identified for the section of single line concerned and often a length of wood about two feet long uniquely painted a distinctive colour to distinguish it readily from any neighbouring branches. Once issued, however, no other engine was allowed to enter the branch until the staff had been returned. With this method, therefore, apart from at a non-block post for an intermediate level crossing, no signalling was necessary along the branch other than to control the engine returning to the junction and the signal box from which the staff had been issued in the first place. Nonetheless, it was sometimes judged convenient to use block working to control the return of the train.

The next method was to use 'train staff and ticket', whereby there was still only one unique staff, or staff key, for each section of single line between crossing loops, junctions or double line of track. These were sometimes fitted with a key for ancillary purposes within the section. This method, however, had its limitations. The ideal was to operate a neat succession of trains in alternate directions, so as to restore the staff to its starting point each time. If, however, more than one train was due to travel in the same direction before one returning, then, the driver of the first train was shown the staff and given a ticket. Once his train reached the far end and the line was clear, the staff had to be shown to each subsequent driver and he be given a ticket authorising him to enter the section ahead, the staff being conveyed by the last train in that direction. The tickets were to be kept in a locked box only opened by key on the staff.

All this presumed that the actual sequence of trains was in accordance with the timetable or could be predicted reasonably far in advance. In the real world, however, a few trains running out of course meant that the 'staff & ticket' method of regulating single lines was unbending. Occasionally, an alternative means had to be found of returning the staff. Short of the unlikely event of a light engine being available, in the days before the motor car, this would necessitate someone carrying the staff by horse, bicycle or on foot, which inevitably involved considerable delay.

Sidings within the single line section, but too remote to be worked by the nearest signal box, could now be protected by incorporating a locking feature on the ground frame or providing a point lock only released by inserting the staff, key or token for the section with this becoming trapped in the lock until the points were reset to normal.

An alternative method was that of Pilot Guards, who should not be confused with Pilot Men. The latter were usually employed when, as a result of an accident or engineering work, single line working over double line was required. Pilot Guards, however, were different. He wore an arm-band lettered 'Pilot Guard followed by the title of the section of line over which he had control'. His possession of the staff, which he had to show to every driver, authorised all movements on the branch. For instance, on a colliery line, the sequence might begin with a number of trains of empty wagons being dispatched by him from the empty wagon sidings, after which he travelled to their destinations on the last train. Once the trains of loaded wagons were ready, he authorised their departure and returned on the final working.

The procedure was set out in LMS General Appendix to WTT for March 1937, pages 84 and 85, and specific locations at which it was authorised listed in the appropriate Sectional Appendix, although that for the Northern Division, makes no mention of Pilot Guards. In effect, Pilot Guard working was just staff and ticket working where a man was the staff and some companies appear have found this useful. In most cases, however, there seems to have been an additional reason, such as extra work to do, why using a man rather than a bit of wood made sense (i.e. the extra man also did something useful).

Greater flexibility in operation of trains was achieved by the introduction of the electric tablet instrument. In this method matching pairs of instrument were located one at each end of the single line section, each of which was capable of storing a number of tablets, or subsequently other types of token, uniquely identified for its particular section of the line. The two instruments were connected electrically and the system so designed that, although a token could be extracted from either end, only one could be taken out at a time.

Only once returned could another be drawn out, again from either instrument. Each section of line would be provided with a quantity of at least two dozen such tokens, which meant that a succession of trains could pass down the line in one direction, but, once clear of the section, one in the opposite direction could be sent as and when required. Nonetheless, on many sections the number of tokens in each instrument from time to time got out of balance, and provision was made in the Regulations for the signal lineman to be able to restore this.

For instance, mention is made of the following telegraph lineman instructions for the Maidens & Dunure Light Railway, which appeared in the GSWR Working Timetable as a general footnote, where there were more trains southbound than northbound:

A Telegraph Lineman was arranged, as in the 1913 Working Timetable, to travel with the daily goods train when required 'to redistribute the Tablets between Alloway Junction and Girvan, and the Train he travels with must stop at all Tablet Stations to allow him to put the Tablets into their places and rejoin the Train'.

This was to allow the lineman to transfer the tablets from one end of the section to the other, thereby overcoming the problem of an accumulation of tablets at one end, if the service was unbalanced, as it sometimes was.

And again, in the Highland Railway's Appendix to the working timetable for 1 May 1920:

TRANSFER OF TABLETS AT DAVIOT, MOY AND SLOCHD. The 6.30 a.m. Goods Train from Inverness to Aviemore to be delayed a minute or two at above Stations when necessary for Lineman to transfer Tablets.

The system invented by Edward Tyer was patented in March 1878, employing tokens in the form of flat metal 5, later 4, inch diameter by ½ inch thick discs known as tablets upon which would be inscribed the section of line to which it applied. Square tablets were also later used to distinguish them when the feature of instituting long sections was available. Once prototypes had been tested on the line between Lockerbie and Lochmaben and later between Newmains and Morningside, the system was put into regular use in 1880 with the opening of the line between Dalmally and Oban.

At first the tablets were made of brass, bound with a steel periphery, but later aluminium or even fibre were adopted to reduce cost and weight. To implement the system, an instrument would be installed in the signal box or station office at each end of the single line section. These were connected by means of either a single telegraph wire and earth return, or two wires. Within each instrument a number of tablets were placed. The design of the bench mounted instrument was such that these were secure within until released electrically by the combined action of the signalmen in both boxes. Thus, to obtain a tablet, the signalman, or a senior member of the station staff in the box or office seeking a tablet, after exchanging the call bell codes, requested the release of a tablet and if the line was clear the man at the other end would press a plunger on his instrument, which would send an electrical current to the other instrument thereby enabling the man there to withdraw a tablet. Once he had done so, no further tablets could be obtained from either end, until the first was returned, initially only to the instrument at the other end, but later at either end, thereby allowing shunting back into the section to be carried out. In due course a further safety feature was introduced at selected locations, whereby the starting signal was locked at danger until the tablet had been obtained.

To avoid confusion between tokens for adjacent sections, the tablets and later key tokens for adjacent sections were differently configured, by means of a variety of shapes to the perimeter and/or holes in the centre, such as round, triangular, square, diamond, and/or notches on the periphery, which the instrument could identify

Ten years later, as an alternative to Tyer's tablet instrument, FW Webb and A Thompson of the LNWR patented a floor mounted electric train staff instrument, using tokens resembling the familiar shape of the staff and initially some 22 inches long. The Railway Signal Co Ltd of Fazakerley undertook production of these and in 1906 they produced a miniature version of electric staff instrument to place on a bench. The key token instrument, patented by AT Blackall and CM Jacobs of the Great Western Railway, was introduced by Tyer & Co. Ltd. in 1912. This was more compact and mechanically simple, with few moving parts, and this soon became the preferred form of token instrument on Britain's railways. To facilitate the proper exchange between signalman and train crew by hand, a token was usually placed in a holder such as a stout leather pouch attached to a large hoop wrapped in leather. Of 10 to 18 inches in diameter, or 22½ by 14½ inches the hoop was large enough to be caught by an outstretched arm of either the footplate-man or station staff. Latterly the pouch had a hole through which the driver could look to verify that it contained the correct token without his having to remove it. A few lightly used branch lines normally operated by train staff working, but, in anticipation of seasonal excursion traffic, such as the Thurso, Aberfeldy (following alterations in LMS days), and Kirriemuir branches were also equipped with electric tablet apparatus, which could be switched in as required to cope with peaks in traffic, but were otherwise switched out. The actual procedure of passing and overtaking of trains on a single line has been described by the author in LMS Journal No 16 and The Dingwall and Skye Railway, published by Crécy in 2016. A nasty accident, in which fifteen people lost their lives, occurred on the Cambrian Railways at Abermule on 26 January 1921, when an unauthorised junior member of staff handed the token just given up by the driver to the station master without explanation, who, having recently returned from the goods yard, was unaware of the current situation. He then failed to check that the token was for the next section of the line as he supposed, and handed it back to the driver, with the result that the driver, again without checking the section to which the tablet applied, duly set off only to collide violently with the train in the opposite direction. Apart from the appalling slackness in working, it was realised the risk of misreading the token for the appropriate section could be reduced if the tablet could be seen without removing it from the pouch. Such a mistake might be further avoided if adjacent sections alternated between round tablet and key tokens, leading the LMS to institute a programme in 1928 of providing key token instruments in sections alternately with the tablet type on most of the Highland Section. On conventional double lines, it was not uncommon at slack periods in traffic to switch out an intermediate block post and double the length of the block section to the next block post along the line, thereby saving in manpower. Likewise, it was sometimes advantageous to do the same on single lines, but in this case another pair of token instruments at each end of the extended section was required. So, when the crossing was switched out, square tablet tokens were often used for the longer section. In the case of 'staff and ticket', when trains were double-headed, or were being assisted in the rear, a ticket for each engine would be issued. For tablets and key tokens, when exchanging tokens by hand, these had to be shown to both drivers and carried by the one in the rearmost engine. Likewise, when automatic tablet exchange apparatus was used the engine in the rear carried the token and the driver of the pilot engine of a double-headed train will have been assured that they were authorised to enter a single line section by seeing the driver of the train pick up the token. When given the token for a long section, however, a notice in the Appendix to the Working Timetable required that:

When a double-headed train enters a token section and the driver of the train engine has picked up a long section token he must give 1 crow and 1 whistle to indicate to the driver of the leading engine that he has received a long section token. The driver of the leading engine must acknowledge the whistle by repeating it.

Within three weeks of the opening of the final section of the Callander and Oban Railway from Dalmally to Oban on 1 July 1880, Tyer's tablet instruments were brought into use throughout the line. Tyer's instruments were installed on the S&DJR from 1886.

The Highland Railway had operated safely from its early days by the telegraph block system, whereby only one train, from either direction, was permitted to enter a section at time. The crossing points with trains travelling in opposite directions were primarily set out in the working timetable. Any additional train or alteration due to late running could only be authorised by the Traffic Superintendant's office in Inverness. These orders were transmitted by telegraph and had to be repeated back word for word. It took until 1897 before HR reluctantly decided to adopt Tyer's No 6 tablet instruments on its main lines and longer branch lines.

The Glasgow & South Western Railway introduced on all of their passenger carrying single lines Tyer's electric train tablet instruments in place of staff and ticket working between 1890 and 1896. One exception was the lightly used branch from Castle Douglas to Kirkcudbright until about 1930 when the LMS installed key token. From its opening in 1903 the Moniavie branch depended on Syke's single line 'lock and block' without tokens, something the Board of Trade was reluctantly persuaded into accepting. When in 1936 these instruments required renewal the LMS changed to key token to Dunscore and 'one engine in steam' from there to Moniaive using a train staff.

An unusual procedure was adopted on the Callander and Oban line to obviate the need for a late-night shift for a number of signalmen while allowing the Down night mail (and mixed) train to pass along the line in the small hours of the morning (in July 1934 dep Dunblane 12.2am, arr Oban 4.31am MX). To quote from the Appendix to the Working Timetable for March 1937:

The guard in charge of the train will exchange tokens at each token station and leave the token received from the driver in the place where the one to be handed to the driver is taken from.

According to both OS Nock and John Thomas, this meant that, as no other trains were scheduled, before going off duty every evening each signalman set the road for the down direction, in cooperation with the next signalman down the line withdrew a tablet and left it lying at an agreed place, perhaps in the signal box or locked in a waiting room. As the train stopped at each station, the guard or fireman, left the tablet for the previous section and a note that the train had arrived complete, collected the tablet for the new section and conveyed it to the next station, where he left it and repeated the procedure for the next section and so on for the rest of the journey. The first task of each signalman on coming duty in the morning was to replace the tablet in the appropriate instrument in preparation for normal single line working. This was fine in the days before vandals might have created mayhem, especially if the tablet were left in the catcher!

Again, to save calling out signalmen for the passage of a light engine at 5.00 am from Ayr to Dunure and its return working with a fish train at 6.30 am, special arrangements were made for the tablets for the sections to Alloway, Heads of Ayr and Dunure to be withdrawn the previous night and left on the instruments at their respective signal boxes. The signals were left at danger, and the guard acted as pilotman, the train stopping at each box to let him leave the tablet for the section the train had just left and pick up the one for the section ahead, repeating the process on the return journey. The guard also left a pilotman's ticket in each box indicating that the train had passed.

Photo of Loch Skerrow
Plate 1/2 At Loch Skerrow (PP&WJR) provision was made to operate the block sections in both directions either long to: Newton Stewart and Castle Douglas or short to: Gatehouse-of-Fleet and New Galloway. The signalman lounges in front of various instruments, which are from the left: two short section (Tyer's) key token instrument; a long section No. 6 instrument, using square brass tokens, and the switching instrument. There will have been another long section tablet instrument out of the picture. To the right of the tablet instrument is the end switch of the "long section relieving switches" for one of the sections. These switches were to allow reversion to short section working by surrendering the long section tablet at one of the intermediate switched out stations. (Norris Forrest, courtesy Transport Treasury)

Tablet Exchange Apparatus

It might be supposed that single lines were short branch lines and where operation was intermittent and train speeds low, but this was often far from the case. Although there were indeed many such branch lines, there were also long lengths of single line running a fairly intensive service over some distance in difficult terrain, where maintaining a reasonable speed and punctuality was highly desirable. Once token instruments were installed on single lines with significant traffic, the busy lines of any length benefited from the greater flexibility in operation of being better able to cope with peaks and disruption in flow. It remained, nonetheless, necessary for non-stop trains to slow down at crossing loops, whether crossing another train or not, to enable the exchange of tokens to be safely made by hand. Despite the railway companies trying to impose a sensible speed restriction, typically of 10 mph, a sense of bravado appears to have developed whereby the fireman was faced with the challenge of undertaking the exchange of the hoops containing the tablets at ever increasing speed and consequent risk of injury.

A partial solution was the introduction of manual token receiving and delivering apparatus which facilitated hand exchanging, as adopted by the Great Western Railway. A post beside the track with a curved arm projecting out of a leather covered cushioned pad fitted at the top was designed to receive a special hoop containing the token. Another adjacent post was fitted with a cast-iron socket into which the hoop was clipped, both posts being adequately lit to enable exchanges to take place at night. This may have benefited the station staff, but still did little for the footplate men.

Photo of Kyle of Lochalsh
Plate 1/3 4-6-0 Stanier Class 5 No. 44719 with the 6.20am mixed train from the Kyle of Lochalsh to Inverness runs into Garve station (HR) on 10 July 1958, as the fireman prepares to undertake a hand exchange of the single line token. (Author)

Automatic Tablet Exchange Apparatus

The introduction of automatic tablet/token exchange apparatus, therefore, had the double benefit of reducing the risk of injury to the footplate-man and/or signalman, where the train was travelling too fast, and enabled the train to maintain the speed other factors would determine. By February 1938 there were on the LMS Northern Division 538 miles of single line equipped the automatic tablet exchange apparatus enabling the tokens to be transferred without the train having to be slowed down for the purpose, nonetheless leaving 392 miles where exchanges were still necessarily carried out by hand.

Those lengths of line on the LMS which used an automatic form of tablet exchange apparatus included:

The Manson system: From 1899, the Highland main line from Perth to Inverness, via both the Direct and Dava routes; the lines further north to Wick and west from Dingwall to Kyle of Lochalsh; and east from Inverness to Keith. 163 exchange posts were installed along 340 route miles. In 1903 it was also adopted on 78 miles of the Callander & Oban line, which had 38 posts, while the Portpatrick & Wigtownshire Joint had 23 in 57 miles, until replaced by the Bryson system during World War 2, and across the North Channel the Northern Counties Committee of the LMS used the system on the Belfast to Londonderry line, together with the Portrush and Cookstown branches. The Belfast & County Down also used Manson's catcher.

The Whitaker system: Single line sections of the Somerset & Dorset Joint line from Bath to Broadstone; and from May 1906 on the Midland & Great Northern Joint in East Anglia with 92 posts in 106 miles. This system was also used by the Great Western (Norton Fitzwarren – Barnstaple) and by the Great Southern of Ireland, mainly on the singled lines of the former Midland Great Western, but latterly elsewhere.

The Bryson system: Parts of the Glasgow & South Western, including between Girvan and Challoch Jct from 1909; Dalrymple Jct to Dalmellington; Annbank to Cronberry; Belston to Hole House; and Dalrymple to Dalmellington; together with the Portpatrick & Wigtownshire Joint once it was changed from the Manson system during World War 2.

Each of these systems will be explored in subsequent instalments, during which it should be borne in mind that the weights of tablets, token and keys in a range of materials could vary considerably, viz:

Weights of Tablets, Key Tokens, Pouches and Rings etc

TabletAluminiumRound5½ to 8oz
BrassRound1lb 1½oz
Square1lb 3oz
Key tokenTyer'sShort section1lb
Rly Signal CoShort section6½oz
Tablet pouchFor MansonLeatherWithout key token adaptor1lb 6oz
With key token adaptor1lb 8oz
For BrysonLeather7½oz
15 in dia. ringMetal, steel?Circular2lb 3½oz

The act of accelerating or decelerating such items to or from up to 60 mph almost instantaneously and the resulting shock had a significant effect on the potential performance of the apparatuses to be described and called for robust construction.

It was not unknown for footplate-men to throw out the pouch and token instead of using the exchange apparatus, possibly resulting in damage to one or both. This and the occasional malfunction of the apparatus indicated that it was preferable for the ground post to be positioned away from areas accessible to the public, such as platforms, and staff cautioned. Likewise, footplate staff needed to act carefully to avoid hitting lineside obstructions when placing and recovering the token and its container before and after an exchange.

The possibility always existed of course that the exchange apparatus might malfunction, in which case the driver was required to stop, the HR instructions reading:

ENGINE CATCHERS MISSING TABLET In the event of Engine Catcher failing to pick up the Tablet, the Engine Driver must immediately make known to the Signalman or Station Staff by sounding his Engine Whistle as follows:- 1 Long and 2 Short Whistles After coming to a stop the Engine Driver will require to send his Fireman back for the missing Tablet, with which the Signalman, or Station Staff must meet the Fireman in order to reduce the delay to the Train as much as possible.

It was usual to position the ground apparatus for the Manson system towards the end of the loop, before the line became single, in the cess or left-hand side of the track. As the length of runs undertaken by such trains likely to carry out an exchange was of some measurable distance, engines were normally running forward, rather than in reverse and as a consequence the pick-up apparatus was ordinarily only on the left-hand side. Nonetheless, it was entirely feasible to mount the engine's equipment on the right-hand side facing either forwards or backwards. As will be explained, the former was adopted on the NCC. The latter appears only to have been done briefly on the Highland Railway's 0-6-4T Banking Tanks, returning light after having performed banking duties, not up to Drumochter, which was double line, but to Dava on the Old Line and Raven's Rock on the Skye Line.

On the other hand, it was usual with the Bryson system on the Glasgow & South Western Section to position the posts immediately after the line had become single and to equip engines with sockets for the exchange apparatus on both sides. All M&GN engines and all S&DJR 2-8-0s had the pick-up apparatus fitted both sides to allow for running in reverse.

In considering the three types of tablet exchange apparatus to be discussed in subsequent instalments a degree of partisanship appears to be exhibited by the proponents of the different systems as to their respective merits, especially the maximum speed at which exchanges could supposedly successfully be accomplished.

Photo of Clachnaharry swing bridge
Plate 1/4 After running across the Clachnaharry swing bridge over the Caledonian Canal, subject to an 8-mph speed restriction, the fireman gives up the token to the signalman standing outside his cabin as the train comes off the single line and moves onto a stretch of double line to Clunes on the line to the Further North. (LGRP, author's collection)

Edward Tyer

Edward Tyer, the developer and manufacturer of signalling and telegraph equipment, was born in Kennington, London on 6 February 1830. He started work in his uncle's city office, but before long decided to pursue a career in the field of electrical appliances, soon specialising in railway signalling and telegraph. After being Manager of the Railway Electrical Signals Co in England and of Winkworth & Cie in France, in 1862 he founded the firm of Tyer & Co, making specialist equipment for home and overseas. The patent for the first of his many inventions was taken out in 1852 and these continued to be granted until he was the age of 80. They included the instruments used in connection with block signalling and single line tablets, together with telegraphic communications. Aged 82 he died in Tunbridge Wells on Christmas Day 1912.


  1. Bryson W, Token exchange apparatus in Scotland (LMSR), Institution of Signal Engineers, Proceedings 1932/3, Part 2, pp 373-389 & insets 13-16.
  2. Electric token block system – book of instructions etc, LMS, 3 September 1934 ufn.
  3. Jenkins SC, The Portpatrick & Wigtownshire Joint Line, Part 2, LMS Journal No 20, pp67-80.
  4. Nock OS, The Caledonian Railway, Ian Allan, (1962), pp 118-9.
  5. Rolt LTC, Red for danger, David & Charles, 1982.
  6. Stirling D, The history & development of railway signalling in the British Isles, Vol 2, Part 3, Single lines, Friends of the National Railway Museum, 2002.
  7. Tatlow P, Achnasheen – busy times, LMS Journal No 16, pp 61-76.
  8. Thomas J, The Callander & Oban Railway, David & Charles, 1966, p147.
  9. Train staff or train staff & ticket block system – book of instructions etc, LMS, 3 September 1934 ufn.

Photo of hand exchange of single line token
Plate 1/5 One of the footplatemen exchanges single line tokens by hand with the signalman at Crieff in 1948 as LMS No. 14498 in reverse passes the signal box while a bystander looks on. (ER Morten [5070])

On to Part 2 >

Site contents Copyright © LMS Society, 2020