NER Dynamometer Car — The Rolling Laboratory That Witnessed History

Historical Background and Introduction

The story of the NER Dynamometer Car is, at its heart, a story about belief — the conviction that scientific measurement could replace guesswork in the development of steam locomotives. That conviction belonged to Wilson Worsdell, Chief Mechanical Engineer of the North Eastern Railway from 1890 until his retirement in May 1910. Worsdell had spent formative years at the Pennsylvania Railroad's Altoona Works during the 1870s, where American engineers were already pioneering instrumented locomotive testing. That transatlantic experience left a lasting imprint.

The immediate catalyst came in late 1903. Worsdell arranged to borrow the Great Western Railway's dynamometer car — itself only two years old, built at Swindon under G.J. Churchward in 1901 — to conduct trials with NER locomotive No. 532 on the East Coast Main Line. Worsdell and Churchward were reportedly friends, and the loan proved the concept thoroughly. Recognising the operational value of a permanent testing capability, Worsdell commissioned the NER's own vehicle. The car was completed and entered service with its first recorded run on 6 March 1906, hauled by NER R Class 4-4-0 No. 2109 on preliminary East Coast Main Line workings.

The question of builder is genuinely contested. The Science Museum Group catalogue attributes construction to Darlington Works, while detailed archival research within the LNER Encyclopedia community, based on NER works diagrams including Drawing 15246D, points to York Carriage Works as the more likely site of construction — the logical location for building a vehicle classified as a carriage. The car was assigned NER number 3591, painted in the company's Crimson Lake livery, and reportedly cost approximately £1,500.

The NER's timing placed it ahead of almost all its contemporaries. When the car emerged in 1906 it made the NER only the second British railway, after the GWR, to operate a purpose-built dynamometer car. The LNWR would not build one until 1908; the Lancashire and Yorkshire Railway completed theirs in 1912–1913. The Midland Railway and the Caledonian Railway never built comparable vehicles at all. Worsdell understood that his railway's "Big Engine Policy" — which had already produced Britain's first 4-6-0 passenger locomotives — demanded accurate data to validate its increasingly powerful designs.

His obituary made the point with admirable directness: the dynamometer car had been of the greatest service, marking the beginning of a new era in the creation of accurate data regarding tractive efforts. It was, in every sense, a rolling laboratory on wheels.

Design, Construction, and Technical Specifications

The dynamometer car is 15,500 mm (approximately 50 ft 10 in) long over buffers, a substantial vehicle by Edwardian standards, mounted on two bogies. Its most immediately distinctive external feature is a clerestory roof, with etched or decorative glass panels running the length of the raised central section — a design that admitted natural daylight to the instrument compartment below. NER-style duckets (observation lookout bays projecting from the body sides) were added sometime between 1936 and 1938, giving the car the characteristic profile it carried through the Mallard record run and the rest of its operational life.

Feature	Detail
Builder	NER (Darlington Works or York Carriage Works — see text)
Year built	1906
Original number	NER 3591
Length over buffers	approx. 15,500 mm (50 ft 10 in)
Bogie type (original)	NER pattern
Bogie type (from 1928)	Gresley design
Tare weight	[Data unavailable]
Maximum speed	Limited to test train speed; recorded 126 mph with Mallard
Seating	Staff working saloon; accommodation for 4–6 observers
Heating	Steam from locomotive
Lighting	Gas (original); electric [VERIFY: conversion date not confirmed]
Roof type	Clerestory with etched glass
Corridor connections	Added 1928 rebuild

The draw-bar pull measuring equipment was the car's functional core. The front coupler connected to a spring balance arrangement that measured the tractive effort exerted by the locomotive. This mechanical linkage transmitted readings via an upstand bracket to recording pens positioned over a large platen inside the vehicle. The car was not double-ended for this purpose: the draw-bar gear occupied the locomotive end only, meaning the vehicle had to be correctly oriented for each test.

Speed was recorded by a dedicated measuring wheel mounted between the bogies on the underframe, running directly on the rail surface. This small, independent wheel provided mechanical speed data entirely free from the wheel-slip or variation inherent in the locomotive's own driving wheels — a crucial distinction during high-speed trials. Both draw-bar pull and speed were transcribed by pens on continuously moving paper roll charts. During high-speed runs these rolls advanced at a rate of 24 inches per mile, with timing marks applied every half-second, allowing precise speed calculations at any point along the route.

The interior housed a small team of technical observers — typically four to six — with desks and tables arranged around the central recording apparatus. A drop-down bed suspended from the ceiling provided accommodation during long-distance overnight tests. A WC compartment occupied one end, window blinds reduced instrument glare, and protection grilles covered the end windows facing the locomotive. Communication with the footplate crew was maintained through speaking equipment, and a train plug connection at the engine end allowed electrical links to the locomotive where required.

The original instruments were entirely mechanical, driven by the vehicle's own motion through the recording wheel and draw-bar linkage. No external power source was needed for the core measuring functions. The body construction followed NER carriage practice: teak panelling on a composite underframe, with prominent rivet and bolt-head detail on the Indian Red solebars — the same constructional philosophy as standard NER passenger carriages of the period.

Technical Insight — The Recording Wheel: One detail that surprises many enthusiasts is that the dynamometer car's speed measurement was entirely independent of the locomotive being tested. The dedicated measuring wheel, running on the rail between the bogies, could not be affected by driving wheel slip — critically important when testing locomotives at their performance limits. During Mallard's record run, this wheel confirmed the 126 mph peak without any ambiguity about wheel slip distorting the reading.

Numbering, Liveries, and Configuration Variants

Unlike coaching stock with multiple diagram variants and lot numbers, the NER Dynamometer Car existed as a single vehicle throughout its life — but it wore several significantly different appearances across its 45-year career. Understanding these configurations is essential for the modeller choosing which era to represent.

NER configuration (1906–1924): Number 3591 in Crimson Lake livery. NER-pattern bogies, no corridor connections at either end, no duckets. The car in this condition is entirely unrepresented in ready-to-run model form.

LNER early configuration (June 1924–August 1928): Renumbered 3591Y and repainted in LNER imitation grained teak livery with lining. NER bogies retained, still no corridor connections. This brief transitional condition is also absent from the RTR market.

LNER mature configuration (August 1928–1946): The most familiar and most-modelled version. Following a works visit to York — the car entered on 3 April 1928 and emerged 25 August 1928 — it received Gresley-design bogies, corridor connections at both ends, and a new number: 23591. The clerestory remained, and duckets were fitted somewhere between 1936 and 1938. This is the condition in which the car made the Mallard record run and all the famous pre-war high-speed tests. Rapido Trains UK's 935001 models this condition in OO gauge.

Post-war LNER (from May 1946): Renumbered 902502 after returning from wartime storage. Livery remained LNER imitation teak but unlined. Rapido Trains UK's 935002 covers this condition.

British Railways (late 1948 onwards): Renumbered E902502 and reliveried in BR faux teak with Gill Sans lettering, unlined. This is the condition in which the vehicle worked the 1948 Locomotive Exchange Trials. Rapido 935003 covers this period.

A note for researchers: some Rapido Trains UK promotional material gives the post-1946 number as 905202, which is incorrect. The confirmed number from the Science Museum Group catalogue and official records is 902502.

Service History and Operating Companies

NER Service (1906–1922)

The car's NER career is the least well-documented in publicly available sources, with the primary evidence residing in the physical log book at the National Railway Museum. Testing concentrated on East Coast Main Line routes between York and Edinburgh, evaluating the performance of Worsdell's Atlantics and 4-6-0s. A photograph exists showing the car coupled to NER C7 Atlantic No. 706, though this particular pairing does not appear in the log book, suggesting the image may have been a publicity shot rather than a genuine test working. The car's wartime role during the First World War is not documented in available sources.

LNER Service (1923–1947)

The LNER inherited the dynamometer car at the 1923 Grouping and, significantly, never felt compelled to replace it. Its first LNER-branded test run came on 26 June 1924, hauled by locomotive No. 4419 on the Great Northern section. The 1928 rebuild — new bogies, corridor connections, renumbering to 23591 — was carried out under Herbert Nigel Gresley's authority and prepared the vehicle for an extraordinary decade of high-speed work.

On 5 March 1935, LNER A3 Pacific No. 2750 Papyrus — the direct A3 forerunner of the A4 streamliners — hauled 217 tons from King's Cross to Newcastle and back. The dynamometer car's records showed a peak of 108 mph between Grantham and Peterborough, with 100 mph or more sustained for 12.5 miles continuously. These figures gave Gresley the evidence he needed to propose the Silver Jubilee express service and press ahead with the streamlined A4 class.

A year later, on 27 August 1936, A4 No. 2512 Silver Fox reached 113 mph on Stoke Bank while hauling a fare-paying Silver Jubilee service with 270 tons gross — an unofficial record achieved during public service. The dynamometer car was present to capture the data, though the run ended unhappily when the locomotive suffered damage to its middle big end bearing.

Then came 3 July 1938. The official purpose was a brake trial for the newly developed Westinghouse quick-release brakes. The real purpose, known to those who arranged it, was a speed record attempt. Mallard — just four months old, and the first A4 fitted with the Kylchap double-blastpipe exhaust system — hauled the dynamometer car and six Coronation coaches (240 tons gross) north from Wood Green to Barkston Junction, then southward down Stoke Bank. The paper roll recorded speeds at successive mileposts from the summit: 87.5, 96.5, 104, 107, 111.5, 116, 119 mph, then in half-mile increments: 120, 122¼, 123½, 124½, 125, and finally a peak of 126 mph near milepost 90¼ between Little Bytham and Essendine. The observers immediately calculated a five-second average maximum of 125 mph. Gresley, characteristically cautious, declined to claim the higher figure publicly. Modern analysis by the Institution of Mechanical Engineers, and subsequent work by historian David Andrews examining the paper feed mechanism, has suggested the true maximum may have been closer to 124 mph — but the certified figure of 126 mph, recorded in that car on that afternoon, has never been officially revised or bettered.

The car saw no further test use until 30 May 1946, when it re-entered service after wartime storage.

Historical Insight — No Spare Car: One underappreciated aspect of the LNER's testing programme is that the company operated with a single dynamometer car throughout its entire existence. When the 1906 NER vehicle was in works or storage, the LNER had no fallback testing capability at all. The car's 1928 rebuild took nearly five months — during which no LNER dynamometer testing was possible. This financial constraint, driven by the LNER's perpetual undercapitalisation, makes the breadth of the car's testing record all the more remarkable.

British Railways Service (1948–1951)

Renumbered E902502 under the new administration, the car played a central role in the 1948 Locomotive Exchange Trials — the great post-nationalisation contest in which express passenger locomotives from all four pre-nationalisation railways were tested on each other's routes to assist BR in designing its new Standard classes. Three dynamometer cars were deployed across the network; the NER car handled East Coast Main Line workings, recording the performances of visiting locomotives including LMS Rebuilt Royal Scot No. 46162, Coronation Class No. 46236 City of Bradford, GWR King Class No. 6018 King Henry VI, and Southern Railway Merchant Navy Class No. 35017 Belgian Marine.

Subsequent BR-era tests included Bulleid's experimental Leader locomotive around Brighton and Eastleigh in 1949–1950, and Britannia Class No. 70006 on the Liverpool Street–Norwich route. The final recorded test run took place on 12 October 1951, from Stourton (Leeds) to Lancaster and return, behind Hunslet 500HP diesel locomotive No. E6 — a machine destined for export to Peru. A fitting last assignment for a vehicle that had measured everything from Edwardian express Atlantics to pioneering diesel-electric power. The car was formally withdrawn in December 1954.

Withdrawal, Preservation, and Surviving Examples

The NER Dynamometer Car is part of the National Collection, held by the Science Museum Group (object number 1975-7050), and is displayed in the Great Hall of the National Railway Museum, York. It stands alongside A4 No. 4468 Mallard — the locomotive whose record it certified — in one of the great juxtapositions of preserved railway history.

In the second half of 1963, York Carriage and Wagon Works undertook a cosmetic restoration to the post-1946 LNER condition (number 902502) for museum display purposes. The waist panels were regrained, LNER lettering transfers were applied, and the interior measuring equipment was cleaned and captioned. Darlington's Chief Mechanical Engineer's department restored the working instruments and prepared explanatory captions. The car was first displayed at the Museum of British Transport at Clapham before transferring to the NRM when it opened in York in 1975.

The vehicle is displayed in an essentially unaltered condition. Enthusiasts with an eye for detail have noted a few minor discrepancies from strict historical accuracy: the toilet compartment window uses clear glass rather than the obscured glazing that would have been fitted originally; the spacing of "Dynamometer Car" lettering on the body sides does not precisely match photographic evidence from the 1946–1951 period; and the speed recording wheel between the bogies is currently painted white, though its original operational colour is uncertain. These are minor points on a vehicle of outstanding historical completeness.

You can visit the NRM year-round (the Great Hall is free to enter), and staff can often provide access to additional information about the vehicle from the museum's archive holdings. The original paper roll from Mallard's 3 July 1938 record run also survives at the NRM and has been the subject of modern re-analysis — ask at the enquiries desk about access to this extraordinary document.

Other preserved British dynamometer cars worth seeking out include:

• GWR Dynamometer Car No. 790 (1901) — the car that preceded the NER vehicle and which Worsdell borrowed in 1903. Preserved at the South Devon Railway, Buckfastleigh, Devon.
• L&YR/LMS Dynamometer Car No. 45050 (1912) — restored to LMS Crimson Lake livery and displayed at the Midland Railway Centre, Butterley, Derbyshire.
• LMS Dynamometer Car No. 45049 (c.1948–49) — unrestored, also at Butterley.

Modelling Significance and Scale Replications

The NER Dynamometer Car holds a unique status in the model railway market: it is emphatically not a vehicle any prototype railway enthusiast would have expected to see produced in ready-to-run form. It ran in ones, it wore drab working liveries, it carried no passengers, and its external appearance — boxy, topped by a clerestory roof, with a peculiar underframe wheel visible between the bogies — resists easy glamour. And yet it has become one of the most celebrated modern RTR releases, precisely because of the extraordinary history it represents.

Rapido Trains UK holds a virtually uncontested position as the producer of this vehicle. The original OO gauge release was commissioned by Rails of Sheffield as a retailer exclusive, developed with access to the NRM's own archive drawings. With 230 separately fitted parts, working LED interior lighting activated by a magnetic wand, acid-etched clerestory glazing, and the recording wheel accurately modelled between the bogies, it set a new benchmark for coaching stock detailing at its launch.

OO Gauge (4mm/1:76 scale) — Rapido Trains UK (935xxx series)

Catalogue No.	Description	Livery Period	RRP
935001	No. 23591, LNER teak	1928–1938 (Mallard condition)	£139.95
935002	No. 902502, LNER teak	Post-1946	£139.95
935003	No. E902502, BR faux teak	Post-1948	£139.95
935004 (Rails excl.)	BR maroon	Unconfirmed prototype basis	£139.95

The 935001 release in Mallard record run condition is the most sought-after for display purposes, pairing naturally with Hornby's, Bachmann's, or Accurascale's A4 models. The BR faux teak version (935003) is the correct partner for 1948 Locomotive Exchange Trial scenes — pair it with your LMS or GWR express pacific.

The BR maroon livery (935004) deserves a brief cautionary note: no photographic or documentary evidence has been identified confirming the prototype ever carried this livery. It is widely considered a fictional but commercially attractive option. Purchase it for its modelling appeal rather than prototype accuracy.

Golden Age Models produced a premium brass OO model, manufactured by FC Models of Korea, in LNER teak liveries. Prices at issue were in the £250–350+ range. This is a secondary-market item now and represents a step up in detail — particularly in brass body construction — over the Rapido plastic moulding, though the Rapido model has generally been considered the superior everyday modelling option due to its finer moulded detail and running qualities.

N Gauge (2mm/1:148 scale) — Rapido Trains UK (955xxx series)

Catalogue No.	Description	Livery Period	RRP
955001	No. 23591, LNER teak	Mallard record run condition	£99.95
955002	No. E902502, BR faux teak	Post-1948	£99.95
955003	NRM "as preserved"	Museum condition (902502)	£99.95
955004 (Rails excl.)	RTC red/blue	Railway Technical Centre era	£99.95

The N gauge range includes a Locomotion Models exclusive in "as preserved" condition — a useful choice if your layout includes a museum or special display scene. The Railway Technical Centre red/blue livery (955004) represents a period beyond the vehicle's operational testing life and is a speculative modelling option.

Kit Options

D&S Models produced the vehicle as kit DS 800, an etched brass body with cast whitemetal fittings, available for OO, EM, and P4 gauges. The most recent production run was a special 2013 re-release timed for Mallard's 75th anniversary celebrations, priced at approximately £75. The kit is now out of production and commands a premium on the secondhand market. An RMweb build thread by user "TomsFineArt" documents a detailed DS 800 build and provides useful reference on fitting and finishing.

ModelU produces a set of five LNER dynamometer car engineers in OO scale — white metal or resin figures specifically designed and posed to populate the Rapido vehicle's interior. A small but thoughtful accompaniment for the serious detailer.

No O gauge or Gauge 1 ready-to-run products have been confirmed as available.

Unique Modelling Tips and Layout Integration

The dynamometer car is, first and last, a specialist vehicle — and that specialisation is both its modelling challenge and its modelling opportunity.

Choosing the right livery for your era

The era classification determines your correct variant. Era 3 (1923–1947) modellers should choose 935001 (No. 23591, LNER teak) for the pre-war high-speed period, or 935002 (No. 902502) for immediate post-war workings. Era 4 (1948–1994 first period) modellers should choose 935003 (No. E902502, BR faux teak) for the 1948 Exchanges period, pairing it appropriately with visiting locomotives from other regions. Era 2 (pre-grouping NER 1903–1922) is entirely unrepresented in RTR form — a genuine gap in the market for the dedicated NER layout builder.

Correct train formation

The dynamometer car always ran immediately behind the locomotive being tested. In front of the locomotive would be nothing — the engine led the test train. Behind the dynamometer car came the train load, which might be:

• A set of Gresley teak coaches (for LNER-era express tests — typically 6–8 vehicles)
• A brake van alone (for locomotive classification trials on secondary routes)
• A mixed rake of Gresley and Thompson vehicles (for post-war LNER tests)
• BR Mark 1 coaches in blood and custard (for early BR Exchange Trial workings)
• A saloon or inspection vehicle at the rear (for senior observers)

For an accurate Mallard record run, the formation behind the dynamometer car was six Coronation stock coaches (the articulated twin sets of the Coronation streamliner service), totalling around 240 tons gross. Bachmann produces Coronation twin-set coaches in OO, making a close approximation achievable without scratch-building.

Modelling Tip — Staging the Record Run: To model the Mallard record run scene in OO gauge you need: Hornby or Bachmann A4 No. 4468 Mallard in Garter Blue; Rapido 935001 (No. 23591, LNER teak); and three Bachmann Coronation articulated twin-set coaches. The whole consist fits comfortably on a minimum radius curve of 438mm (2nd radius), making this an achievable display formation even on a modest oval.

The unmodelled NER condition

For modellers of the pre-1923 NER period, the challenge is considerable. To represent the car in NER Crimson Lake livery (number 3591, NER bogies, no corridors), you would need to: repaint a Rapido body; remove the corridor connection mouldings and fill; replace the Gresley bogies with NER-pattern bogies (available from association suppliers or scratch-built); and apply NER-style lettering using transfers. NER Crimson Lake is a distinctive deep red approximately matching Humbrol 100 (Red Brown) over a white base. The effort is substantial, but it would produce a unique and historically significant model for any NER layout.

Weathering considerations

The dynamometer car led a hard working life and was not pampered. Photographs from the mid-1930s onwards show roof dirt, underframe grime, and moderate body soiling around the bogies and buffer beams. A light wash of Vallejo Smoke or equivalent around the clerestory panels, with AK Interactive Grime along the solebars and a dry-brush of light earth along the bogie frames, produces a convincingly working-condition appearance. The preserved museum condition (NRM display) is essentially clean and well-presented — appropriate for those pairing the model with a heritage exhibition scene.

Finally

The NER Dynamometer Car of 1906 is, in the fullest sense, a vehicle that changed history. Built to bring rigour and measurement to steam locomotive development on a progressive Edwardian railway, it outlasted its builder, its company, and the entire era of steam traction — and in so doing became the certified witness to the greatest speed achievement in railway history. When Mallard touched 126 mph on the afternoon of 3 July 1938, it was the paper rolls turning inside this car that made the record real. Without that car, that afternoon would be an anecdote. With it, the number is incontestable.

For the modeller, the vehicle offers something genuinely rare: a prototype that is both technically specific and historically compelling. You can run it as part of a high-speed LNER test formation in the 1930s, as a working vehicle in the sober BR faux teak of the 1948 Exchange Trials, or simply as a display piece paired with Mallard on a showcase shelf. Rapido Trains UK's OO and N gauge offerings are, by any measure, exceptional models of an exceptional vehicle.

Visit the National Railway Museum in York to see the real thing. Stand beside the car, and beside Mallard, and consider the afternoon in 1938 when four observers inside this vehicle watched the recording pens trace a line across a moving paper roll and knew, in real time, that they were witnessing something that would never happen again. It has not.

Frequently Asked Questions

What is a dynamometer car and what was the NER Dynamometer Car used for?

A dynamometer car is a specially equipped railway vehicle towed immediately behind a locomotive to measure its performance during service trials. The NER Dynamometer Car measured draw-bar pull (tractive effort) and speed via mechanical instruments recording continuously on paper rolls. It was used throughout its career to evaluate NER and LNER express locomotive classes, verify speed records, and assist British Railways with the 1948 Locomotive Exchange Trials.

Why did the LNER never build its own dynamometer car?

The LNER was the least financially secure of the four grouped companies from 1923, operating a geographically challenging network with significant track maintenance commitments. The inherited NER vehicle, upgraded with Gresley bogies and corridor connections in 1928, was judged adequate for the railway's testing needs. Building a replacement would have required capital the company consistently lacked. The 1906 vehicle's quality of construction effectively saved the LNER from a capital expenditure it could not easily afford.

Where can I see the NER Dynamometer Car today?

The car is on permanent display in the Great Hall of the National Railway Museum, York, alongside preserved A4 No. 4468 Mallard. NRM admission to the Great Hall is free, and the museum is open year-round. The original paper roll from Mallard's record run is also held at the NRM — enquire at the visitor services desk about archival access arrangements.

Which OO gauge model should I buy to represent the Mallard record run of 1938?

Choose Rapido Trains UK 935001, which represents the car as No. 23591 in LNER imitation teak livery in its 1928–1938 Mallard record condition. Pair it with a Hornby, Bachmann, or Accurascale LNER A4 No. 4468 Mallard in Garter Blue, and three Bachmann Coronation articulated twin-set coaches for an accurate six-coach test train of approximately 240 tons. The Rapido model includes working LED interior lighting and accurately represents the clerestory roof and recording wheel between the bogies.

Are there other scales available beyond OO gauge?

Yes. Rapido Trains UK has produced the car in N gauge (2mm/1:148) in the 955xxx catalogue series, covering LNER teak, BR faux teak, and NRM "as preserved" liveries at approximately £99.95 RRP. The N gauge range includes a Rails of Sheffield exclusive in Railway Technical Centre colours. No confirmed ready-to-run O gauge or Gauge 1 versions are currently available. D&S Models produced an etched brass kit (DS 800) for OO, EM, and P4 — now out of production and secondhand only.

Is the BR maroon livery version of the model based on a real prototype?

No confirmed documentary or photographic evidence exists showing the NER Dynamometer Car in BR maroon livery during its operational life. The maroon Rapido OO release (935004, a Rails of Sheffield exclusive) appears to be a commercially attractive but fictionalised version. Modellers wishing to maintain strict prototype accuracy should choose 935003 (No. E902502, BR faux teak) for the vehicle's actual BR appearance, as worn during the 1948 Exchange Trials and subsequent test workings through to withdrawal in December 1954.

How does the NER Dynamometer Car compare to the GWR car it predates?

The GWR Dynamometer Car of 1901 was the first purpose-built example on a British railway and remains preserved at the South Devon Railway, Buckfastleigh. The GWR car was shorter and mounted on four-wheeled axles rather than bogies, reflecting the more conservative construction standards of the 1890s. Worsdell specifically borrowed the GWR car in 1903 before building the NER vehicle, and the NER design improved on the GWR approach by adopting bogies for a smoother ride, a larger body, and more generous interior accommodation for test staff. The NER car's greater length and bogie mounting made it considerably better suited to sustained high-speed testing, as the LNER era would spectacularly demonstrate.

What happened to the paper roll from Mallard's record run?

The original paper roll recording the 3 July 1938 run is preserved at the National Railway Museum, York, and has been the subject of modern scholarly re-analysis. Research by the Institution of Mechanical Engineers and historian David Andrews examined the paper feed mechanism speed and suggested the true peak may have been closer to 124 mph rather than the certified 126 mph. The official record has never been revised. The roll remains one of the most significant documents in British railway history — a continuous trace of the seconds during which steam traction reached its absolute limit.