Whale, Bowen Cooke & Beames designs: other locomotives
Steamindex homepage

This section covers twentieth century development of LNWR locomotive design which in some ways represented a break with that of Webb, although the classic 0-8-0 type had been introduced under his supervision. Furthermore, he had introduced the 4-6-0 type before the end of the nineteenth century, whereas nothing larger than 4-4-0s were produced at Derby for passenger work. Crewe ceased to construct the 0-6-0 type after 1904. The Claughton class locomotives must have been amongst the kindest locomotives to the track ever produced (being perfectly balanced) and were extremely elegant-looking. The external Walschaerts valve gear was a marked advance upon the Swindon four-cylinder layout and pressaged Gresley's front-end layout by using rocking levers ahead of the cylinders to drive the valves for the inside cylinders. WW1 and the early death of Bowen Cooke, plus the engineering debacle which characterized the formation of the LMS hindered development of a locomotive which in some respects was potentially far in advance of anything produced for the West Coast mainline until well after the arrival of Stanier.

Whale designs: 4-6-0: Experiments 19in goods; Precursors  4-4-2T.
Bowen Cooke designs: 4-6-0  Claughtons Prince of Wales; 4-4-0 George the Fifth;  0-8-2T; 4-6-2T
Beames 0-8-4Ts ex-ROD (MM) class 0-8-0 G2A Detail: brakes/injectors


Summary of the Eight-Coupled Tender Engines

There is a major problem with the LNWR 0-8-0 and 2-8-0 classes. There is no question that Webb started the concept by coupling the "huge" boiler fitted to the Greater Britain type to an extended Coal Engine 0-6-0 chassis. The initial locomotive was a simple but Webb could not resist producing three- and four-cylindered compound variants. Thereafter, Whale extended the concept by placing a pony truck on the front which led to a 2-8-0 with strong similarities to some of the early 2-6-0s which were extended 0-6-0s. Then larger (Precursor/Experiment-type boilers were fitted) and this led to the Super Ds. Bowen Cooke introduced superheating and Belpaire boilers. Thereafter, activity continued under all the CMEs of the LMS and modifications were still taking place under British Railways. Thus it is safer to assemble all that is not solely Webb below and nowhere else. Both Talbot and Essery and Jenkinson attempted to simplify the complex stories.

Talbot summarises the essential details of the various classes of eight-coupled [what he termed] coal engines in the table below: this uses the 1911 class letter system but lists the classes in chronological order of introduction. Essery and Jenkinson also attempted to impose order upon the somewhat anrchic system..

Class Date introduced Details
2524 October 1892 Two-cylinder simple 0-8-0 with Greater Britain boiler.
A September 1893 Three-cylinder compound 0-8-0
B August 1901 Four-cylinder compound 0-8-0, with much larger boiler than A class.
E August 1904 B class converted with leading pony truck to 2-8-0.
C November 1904 A class converted to two-cylinder simple, with new cylinders but original boiler.
D March 1906 A class converted like C class but with larger boiler.
F May 1906 B class converted to 2-8-0 like E class but with larger boiler.
G November 1906 B class converted to simple with original inside cylinders, retaining 'piano front', and D class boiler.
C1 March 1909 A class converted like C class but with smaller cylinders.
G January 1910 G class built new without 'piano front'.November 1906
G1 January 1912 As G class but superheated; some converted from other classes, others built new.
G2 June 1921 Improved version of G1 class with higher boiler pressure.
G2a October 1935 G1' class converted like G2, with higher boiler pressure, stronger motion and increased brake power.

All the various saturated eight-coupled tender engines were eventually either converted to the G1 class or withdrawn, and another 170 G1s were built new. In a sense, therefore, the G1 class was derived from all the other classes (except the C1s). All 449 G1s were never in service at the same time, as before all conversions to G1 had been made in the 1930s, other G1s had already been converted to class G2a. With the sixty G2s, the number of superheated 0-8-0s of LNWR origin eventuaIJy totalJed 509.The rebuilding policy continued well into LMS days and some of the class were still in service virtually until the end of steam. Talbot attempted to illustrate this complicated story with diagrams which are not been reproduced herein.

No. 2524: 1892
In October 1892 Webb built the first eight-coupled coal engine, No. 2524. It was basically an enlarged 17in Coal Engine, with two inside cylinders, having valves and motion based on the 18in Goods but enlarged to 19½in. by 24in., and the Greater Britain boiler, complete with its combustion chamber part way along the barrel.  No. 2524 was converted to D class in December 1906.

No. 2524 rebuilt in 1903
Locomotive Mag., 1903, 9, 261
shows a line drawing elevation of No. 2524 as rebuilt with a new boiler and cylinders and remaining simple. The new cylinders were 18½ x 24in (that is smaller) and the boiler had a total heating surface of 1245.3 ft2 and a grate area of 20.5 ft2

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Plate 334: The official view of No. 2524 as built was taken on 3 November 1892. Metal brake blocks are used for the first time on a goods engine but otherwise it has the usual Webb features of the time such as steel bufferbeam and wooden-framed tender with wooden brake blocks. The ash chute from the combustion chamber can be seen beneath the boiler, just to the rear of the second splasher. There is one sandbox, alongside the smokebox, for the front of the engine, and another on the tender, just behind the footsteps, to assist with braking and for running in reverse; the latter would be needed especially when shunting a train in bad weather, or when reversing into a siding off a main line to allow a passenger train to pass. The blower valve is on the left-hand side.Figure 93: Official drawing of No. 2524. Plate 335: The number of the engine in this photograph has not been recorded but it seems certain to be No. 2524, since it has no coal rails on the tender and it appears to have no centre lamp socket on the bufferbeam; and no 'C' class were built until long after both these features became standard. The engine is well cleaned in lined black livery and is passing Coleham shed, Shrewsbury, with a northbound freight train.

A class: 1893
In September 1893 Webb introduced a three-cylinder compound 0-8-0, No. 50: a compound version of No. 2524, with two high-pressure cylinders outside, operated by Stephenson valve gear, and one low-pressure cylinder inside, with a slip eccentric. The boiler was very similar to that of No. 2524 but without a combustion chamber; instead the front tubeplate was recessed into the boiler for the same distance as a combustion chamber and the char hopper placed behind and to the right of the low-pressure cylinder. The blower valve was on the left-hand side of the smokebox but there was no linkage for the bypass valve, which on this engine was operated by a complicated linkage passing through the boiler. After comparative trials between No. 2524 and No. 50, the three-cylinder compound 0-8-0 was chosen as the standard heavy goods engine and 110 were built between 1894 and 1900. From No. 2528 the third pair of driving wheels were made flangeless, to assist curving and a bypass valve, operated as on a three-cylinder compound passenger engines, by pushing or pulling the handwheel connected to the boiler handrail. This valve allowed exhaust steam from the high-pressure cylinder to escape directly up the blastpipe on starting. The last 17in Coal Engine 0-6-0 was built in October 1892, after which no more 0-6-0s were built purely for freight work. The last Cauliflower was built in May 1902, being superseded for mixed-traffic duties first by the Bill Baileys and then by Whale's 19in Goods. Thus, although Webb is sometimes derided for his locomotive policies, both he and the LNWR had abandoned the 0-6-0 type some twenty years before certain much respected railways that built it widely had even come into existence!

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Plate 336: official view of No. 50 taken 26 September 1893. Except for features associated with the three-cylinder compound front end, there are no detail differences from No. 2524. The arrangement of sandboxes is the same, one at the front of the engine and one on the tender. The latter was important for the braking of heavy trains and for reversing into a siding off a main line, and was a feature of the Bill Baileys as well as the A and B class 0-8-0s. No. 50 was the first engine to have a 2,000 gallon tender, which must have been built for it, as it has cast H-section wheels. Plate 337: No. 2528 probably at Shrewsbury, before 1903. Plate 338: A class on completion in paint shop at Crewe Works. Plate 339: Another official photograph: No. 1867 on completion on 21 September 1899. Plate 340 No. 1817 in late 1890s..

Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002.
This is an excellent book which complememts the earlier Illustrated history. Chapter 1 covers what Talbot entitles the "compound era".

B class
In 1901 Webb produced a four-cylinder compound 0-8-0, later designated class B with two high-pressure cylinders outside and two low-pressure cylinders inside. Compared with the A class, the boiler was 5in. greater in diameter, and 200psi rather than 175psi. In general appearance and detail, it was very similar, with steel bufferbeam, metal brake blocks and wooden-framed tender. At first, it also had the same arrangement of sandboxes, one at the front of the engine and one on the front of the tender. This was changed soon after Whale took over in 1903; a second sandbox was fitted to the engine, on the driving splasher, being worked by rodding from the leading one, and the tender sandbox was abolished. All subsequent developments of the B class, the E, F, G, G1 and G2 classes, shared the same arrangement of sandboxes on the two leading splashers, while developments of the A class, the C, C1 and D classes, had only one larger one, on the leading splasher. Twenty six of the B class were converted to the E type 2-8-0).. .

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Plate 343: No. 1881 official on completion in August 1901. This later became the only B fitted to work vacuum-braked trains. The there was a boiler support or stay made of steel plate, which was later changed to a casting. Plate 344 (right): No. 2024 at Buxton about 1910. Page 208: Figure 96: drawing of No. 1881; Fig. 97: Weight diagram; Plate 345 No. 859 (driving wheels for 17in. Coal Engine and fully lined) at Wigan c1910; Plate 346: No. 1047 (plain black, rocking lever covers open) post-1916; Plate 347: No. 2272 at Bletchley in 1920; Plate 348: No. 1282 c1920..

London & North Western Ry. Locomotive Mag., 1903, 8. 116
Crewe Nos. 4310-4314 (running numbers: 1229-33) completed order for ten four cylinder type.

Hambleton, F.C. L&NW compounds: the "B" class mineral engines. Loco. Rly Carr. Wagon Rev., 1942, 48, 47.

Boiler explosion at Buxton

Hewison, C.H. Locomotive boiler explosions. 1983.
Boiler explosion at Buxton on 11 Novemeber 1921 at 00.45 to locomotive No. 134. This was due to faulty workmanship on the safety valves when the locomotive was services at a Glasgow engineering works and a failure by those responsible for routine maintenance to respond to reports by footplate crews on the odd behaviour of the safety valves on the locomotive.
Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002.
Pp 34-9 cover the boiler explosion in very considerable depth.

E class (2-8-0): 1904-
Whale appeared to dislike the front overhang on the Webb locomotives and arranged for pony trucks to be fitted to the B class.

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Page 211: Plate 349: No. 1886 Crewe official on 15 October 1904; Figure 98: Drawing of No. 2056; page 212: Figure 99 weight diagram; page 213: Plate 350 No. 18 at Willesden c1920; Plate 351: No. 1222 at Willesden, c1922; Plate 352 No. 1017 c1922.

F class: 1906-
Followed from the E class, but larger boilers (Precedent/Experiment type) were fitted. There were ten conversions from type B and two from E type.

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Talbot page 214: Plate 353: Official photograph of No. 1273 on 16 May 1906; Plate 354 oblique view of 1273; Figure 100: drawing of No. 2573: page 215: Figure 101 wight diagram; Plate 355: 'F' class No. 1036, at Manchester (London Road) about 1912. It is in full lined livery and has a Whale tender. The front buffers are the Whale type on wooden bases, and there is a sandbox under the cab. Its driving wheels, with only one balance weight, have come from a withdrawn 17in. Coal Engine.

Class C and Class C1
In November 1904 Whale began to convert A class three-cylinder compounds into two-cylinder simples, the first being No. 2541. New cylinders, 19½in. by 24in., and new motion were provided but otherwise the engines were little changed, the result being very similar to No. 2524 except for the latter's Greater Britain boiler. Fifteen conversions to C class were made, the last being in March 1906 when Whale decided that larger boilers were needed to feed 19½in. by 24in. cylinders. He therefore fitted the large boiler used on the F class to the C class chassis, thus producing the D class; 63 conversions to D1 class were made, including No. 2524, from March 1906 to March 1909. The fitting of large boilers to Ds left a quantity of small A class boilers unused, so the remaining A class were converted to simples as C class but with smaller cylinders, 18½in. by 24in. They were designated C1' class but outwardly, the C and C1 classes were indistinguishable from each other. Only five of the C class were converted to G1 class, the remainder (and all the C1 type) were eventually withdrawn..

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Plate 356: official view shows C1 class No. 2550 after conversion on 3 January 1910. It has Whale buffers at the front but retains its Webb tender. There is one sandbox at the front, as on No. 2524, and another at the rear of the engine, filled from inside the cab; there is no tender sandbox. As originally built, the C class did not have this arrangement, though they acquired it later. They and the D class originally had one sandbox at the front of the engine only. Plate 357: No. 2550 front view; No. 2550 cab view (tender not fitted); Plate 359 No. 1803 at Rugby c1920; Plate 360: No. 1814; Plate 361: C1 class No. 1859 in March 1911.

D class: 1906-

Using the larger boiler from the F class this class consisted of sixty conversions from the A type plus No. 2524 rebuilt into a form with an outward similarity to the G class: in fact their large sandboxes placed at the front of the locomotives gave them a slightly more modern appearance comparable with many of the 0-8-0s on other railways. Furthermore, once superheated they gave rise to the nickname Super Ds (Talbot caption to Plate 365).

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Plate 362: Official photograph of No. 1866 on 12 March 1906; page 219: Plate 363: No. 2526 at Tamworth c1910; Fig. 102: drawing of No. 2532; Plate 364: page 220: No. 2548 at Tamworth (Low Level) c1915; Plate 365 No. 2532; page 221: Figure 103 weight diagram; page 222: Plate 366: No. 1838 in July 1917; Plate 367 No. 1819 in 1920; Plate 368: No. 1880.

G class: 1906-

Following the successful conversion of  the A class three-cylinder compounds into D class two-cylinder simples, Whale turned his attention to the B class four-cylinder compounds to produce the G class simples. Instead of fitting new inside cylinders as on the D class, the outside cylinders were removed and the former low-pressure ones were retained. As these were lin diameter larger than on the Ds the boiler pressure was reduced to 160psi so that the tractive effort would be about the same. The distinctively shaped covers in front of the smokebox were also retained and led to the nickname 'piano front Gs', but had nothing to cover. No. 1274 was involved in haulage trials in 1910: see more recent work by Talbot.. Sixty of the class were built new following an order placed shortly before Whale retired. 26 of the class served with the ROD in WW1 (Plates 376/7). See also 0-8-2T.

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Plate 369: Non-official view of first of the class, No. 1900; Figure 104: Drawing of  piano front G1 class No. 2251; Plate 370: Piano front type No. 1229 at Longsight about 1920 (still retaining its E class leading crankpin): page 224: Plate 371: No. 1893 with piano front at Huddersfield c1920; Plate No. 372: official photograph of No. 2653 on completion in January 1910: page 225: Figure 105: weight diagram [Talbot notes errors in this official diagram as it combined features from both the new build and piano front types]; Plate 373: No. 1503 with capuchon; Plate 374: No. 2661; Plate 375: No. 1700 c1914 [with W.H. Whitworth on footplate]; page 227: Plate 376: ROD 1639 and 2014; Plate 377: ROD 2014 in France: Figure 106 drawing of No. 1660..

Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002.
Pp. 69-71: In early 1910 No. 1274 was used on haulage trials (to assess the value of high capacity wagons for the carriage of coal) between Nuneaton and Sudbury Reception Sidings. Not surprisingly, the 20 ton capacity wagons were far more economical to haul than those with lesser capacities. Pp. 71-2 give some information on the locomotives sent to France during WW1..

G1 class: 1912

Superheating having been proven on the George the Fifth class, the first G class, No. 2653, was fitted with a Schmidt superheater and reclassified G1 in January 1912. Whether it obtained new cylinders and a new boiler, as then being built for the G1s, is not known. Its original boiler, only two years old, could have been re-used on the next B to be rebuilt to G, or it could have been fitted with a superheater, as with the front tube plate recessed into the barrel 1ft. 9in., there was room to take a header from a 4-6-2 tank, and the original slide-valve cylinders could have been retained. In any case, Bowen Cooke must have been confident of its success, as the first new Gl class was completed at Crewe Works the following month. 170 were built new. In addition, the majority of classes A, B, C, D, E, F and G were eventually rebuilt to Gl also, totalling 279 rebuilds in all. The main external difference compared with the G class was the longer smoke-box, to accommodate the superheater header, though the barrel length of both boilers was the same at 14ft. l0in; internally, inside-admission piston valves replaced the G class slide valves, although the cylinders were the same size. The superheater damper was controlled by the lever on the smokebox, operated by the boiler handrail, while lower down,just above the running plate, is the snifting valve. Superheater dampers were soon found to be unnecessary and were removed. As originally built, the G1s had three-link couplings and steam brakes for the engine only, but the 1914 and subsequent batches were built with vacuum brakes and screw couplings, to enable them to work passenger trains if necessary. Earlier engines were then similarly equipped, but over a period of years, and when this was done, the engine brakes were converted from steam to vacuum, following the practice with passenger engines which had started in 1913. Possibly following WW1, steam heating was added, not so much for passenger trains, as for working banana specials from Garston Docks.

Bodman, Martin. Coals to Newcastle? the 'Jellicoe specials'. Backtrack, 2006, 20, 498-503.
G1 used to haul coal trains from Warrington or Chester (received from GWR) for onward transmission via Shap to Carlisle and thence either the CR or NBR to Grangemouth for shipment to the Orkneys to service the British Fleet during WW1: ironically although the G1 type appears to have dominated this traffic both of Bodman's illustrations are of compounds struggling along (including one at Bayston Hill and another on Shap..
Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Page 228: Plate 378: No. 1384 official photograph January 1912; Figure 107: drawing of No. 1384; page 229: No. 2246 in original condition; Figure 108 drawing of No. 2118 (B class conversion of April 1918); Figure 109 drawing of B class conversion No. 2057 of December 1921; page 230: Figure 110 cross-sectional diagrams; Plate 380: No. 1568 in original condition; page 231: Figure 111: general arrangement diagrams (side elevation and plan); page 232: Plate 381: No. 767 when nearly new; Plate 382: No. 1181 with vacuum brake and screw coupling; Plate 383: No. 1248 (rebuilt with Belpaire boiler in Mrach 1923): page 233: Plate 384: No. 1036 (F class rebuild by LMS); Plate 385 No.1217 (Belpaire boiler fitted post-1924); Plate 386: No. 2551 (converted from D class in May 1925).


G1 was tested against an ROD 2-8-0 on Sunday 18 January 1920 and 9 November 1919

Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002.
Pp. 83-8: the aim of the tests was to measure maximum effort on steep gradients. The tests were made between Mossend and Carlisle at the behest of the Caledonian Railway and rightly Talbot mentions the similar tests on Glenfarg bank made on behalf of the NBR at about the same time. No 2421 on 18 January 1920 managed a 16.5 ton maximum drawbar pull on stalling. The GCR 2-8-0 No. 2078 was tested on 9 November 1919. Very detailed results of these tests are presented: between Motherwell and Law Junction (vertical rise 213 ft) the 0-8-0 hauled 892 tons at 5.5 mile/h and achieved 367 dhp. The 2-8-0 hauled 884 tons at 4.5 mile/h and achieved 325 dhp. From Law Junction to Craigenhil (200 ft rise) the 0-8-0 achieved 7.4 mile/h and 266 dhp, and the 2-8-0 6.1 mile/h and 216 dhp. Talbot states that the 0-8-0 had the edgebeing 20% lighter and cheaper to build..

Strengthened motion

Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002.
Pp. 88-91 Diagram 23: the indirect Joy motion suffered from excessive wear and from 1920 direct motion was fitted to the Prince of Wales and G1 classes. The valve travel was slightly reduced and the gear was slightly less efficient. Pp 74-5 (Fig. 19): general arrangement diagram.

Exhaust beat

Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002.
Pp. 148-9 describe the extraordinary exhaust beat of the LNWR 0-8-0s: fortunately this was also recorded.

G2 class: 1921/2
Higher boiler pressure (175 psi): only LNWR type not to rebuilt into another type (stated Talbot), but reboilered with Belpaire boilers by LMS

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Page 234: Plate 387: No. 134 (built as a replacement for the B class compound No. 134 which exploded at Buxton); Plate 388 No 373 with Ross pop safety valves; Plate 389 No. 994 with Ramsbottom safety valves
Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002.
Pp. 91-5. Very low key description. Pp 98-101 describe comparative trials between G2 No. 574; 4F 0-6-0 No. 3866, and the latter assited by 3F 3756 on coal trains between Toton and Brent. All the locomotives performed adequately, although Talbot suspected that the G2 could have hauled higher tonnages..

2-8-0 proposed 19 in goods engine: 1907-
Talbot (p. 272) includes a drawing dated 16 April 1907 for a "fast" freight engine based on the 19 in goods 4-6-0: this would have shared the same driving wheel size, boiler and inside cylinders as the 4-6-0 and was intended for working over Shap and the Pennines. Same diagram also included in Atkins West Coast 4-6-0s at work..

0-8-0: post-grouping modifications

According to Essery & Jenkinson (below) the LNWR 0-8-0 classes (LMS 8900-9454 plus 8892-9 above) had a complex history well outside the scope of their survey. Essentially, the series had started with a Webb two-cylinder simple design of 1892. This was followed by several series of Webb three and four-cylinder compound variants and finally by the Whale and Bowen-Cooke developments (Classes 'C' and 'Cl') which reverted to two-cylinder simple propulsion and culminated in Bowen-Cooke's 'G2' Class of 1921.

Parallel with building new two-cylinder engines, Whale and Bowen-Cooke also pursued a policy of rebuilding the Webb compounds to the two-cylinder simple layout and this process went on right up to and after the grouping. Some of the Webb compounds were also converted to 2-8-0 type (see above). These compounds which were not rebuilt to simple form were scrapped during the 1920s, including the 2-8-0 variants but the rebuilds had a much longer life. The following summary should clarify the situation:

By the time of nationalisation, further LMS rebuilding had reduced the varieties to either 'G1' type (with 160 lb boilers) or 'G2'j'G2A' type (with 175 lb boilers). The 'G2As' were rebuilds of the 'G1' Class with 'G2' type boilers. In all, therefore, some of the LNWR 0-8-0s had undergone several stages of rebuilding during the 1920s and 1930s and many of them were, in most essentials, relatively new engines - hence their longevity over nominally younger engines. It is finally, perhaps, also worth recording that to generations of locomen, all ex-LNWR 0-8-0s of 'G1', 'G2' or 'G2A' type were known as 'Super Ds' in reference to their original derivation from Whale's Class 'D' engines.

Like the larger LNWR passenger engines, the LNWR 0-8-0s received cut down cab fittings to clear the LMS composite loading gauge and many of them were also fitted with Belpaire boilers, tender cabs and other modifications before finally being scrapped. Some 400 or so of the engines survived to receive BR numbers and many of them soldiered on for almost the whole of the BR steam period, the type finally becoming extinct (except for the preserved 9395) in 1964.

The rebuilding of the LNWR 0-8-0s was a massive and on-going activity and lasted well into the Stanier era. It presented a marked contrast to the normal 'scrap and build' philosophy of the LMS and was a much more significant activity than many of the more celebrated rebuilding programmes of other British railways at the same time. The LNER (very much wedded to rebuilding) never attempted anything on the scale of the LNWR 0-8-0 programme.

G1/G2/G2a :1924: Beames/Hughes: Belpaire boilers
Under the direction of Hughes all new boilers built for LNWR 0-8-0s were of the Belpaire type. Talbot (Eight-coupled... p. 95)  suggests that this change was probably instigated by Beames rather than by Hughes, and links the programme to the Buxton boiler explosion. Talbot (Eight-coupled...) notes that Belpaire type

Talbot noted that it was simple to switch between the two types of boiler, and the round-top type would sometimes be fitted after a Belpaire type was overhauled. T.M. Herbert (later Head of the LMS Research Department investigated the renewal of firebox stays at Springs Branch and reported on this in January 1930: the round-top type ran 644 miles per stay change whereas the Belpaire type ran only 244 miles per stay change: suggested modification were not implemented. See also Cook's Raising steam Table 7:...

Brake tests

Cox (Locomotive panorama 1 p. 24) refers to brake tests carried out during the period of LYR/LNWR amalgamation: A number of engine types were fitted with an economical idea by which a single brake cylinder under the cab applied both engine and tender brake blocks. How the L.N.W worked unfitted freight trains at all was something of a mystery. There is a record of some tests carried out on the L.Y.R. under equal conditions of load, running down a 1 in 100 gradient at a speed of 20 m.p.h. Horwich 0-8-0 No. 1369 stopped its train in 1,520 yards. From the same speed Crewe G2 class 0-8-0 No. 2182, also with vacuum brake on the engine, had accelerated to 22.5 m.p.h. after a distance of 3,860 yards under full brake application, and the train had to be brought to rest by the intervention of another locomotive inserted in the make up for just such an eventuality.

L.N.W. Gl class 0-8-0 No. 1585 with steam brake fared even worse. At the end of 3,055 yards under full braking, speed was 25 m.p.h., and there is a note on the test record that engine brakepower diminished when the driver blew the whistle, due no doubt to a common insufficient manifold supplying both!

After one or two run aways had occurred in service a widespread modification was introduced to fit separate brake cylinders on engine and tender and thus increase total brake power.

REBOILERING L.M.S.R. locomotives. Rly Mag., 1925, 56, 220. illus.

G2a: 1935
The date should be noted: Talbot (Eight-coupled...) notes that these were G1 0-8-0s equipped with new 175 psi boilers, strengthened motion and increased power. Furthermore, this quite extensive rebuilding did not appear to be minuted and Talbot assumes that the extensive programme of new boiler construction was the authority for this work. Figures 24 and 25 (pp 122 and 123) and 28 (page 147) show Crewe drawings modified to include LMS post-1935 modifications. Post-1923 activity at Crewe must have been similar to that at Darlington, Wolverhampton and Eastleigh where "minor improvements" (especially to boilers and valve gear) were gradually introduced without reference to the chief mechanical engineer's office.

Gregory, J.R. The L.N.W.R. eight-coupled goods engines. RCTS, 1950.
Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002.
Page 120 et seq describe the final evoluation of the LNWR 0-8-0 presumably under the not very watchful gaze of Stanier. The last G2A conversion did not take place until 1947 and the last G1 remained in service until 1955. The 8F 2-8-0s were not produced in quantity until WW2.

Retrospective and critical

The A.B.C. of British Railways locomotives. Part 3—Nos. 40000-59999. London, Ian Allan, 1949. 80 p. incl. 24 plates. illus., tables.
This edition contains notes on the boiler variations to be found within the class (see p. 4) at that time.
Bushell, George. LMS locoman: Wellingborough footplate memories. Truro: Bradford Barton, 1985?
This was quite a good experience to get acquainted with a 'Nor'-West Super D', as the Midland men always called these 0-8-0s. Despite their awkward cab layout, I grew to respect them for their efficiency and power when properly handled. To describe their footplate, one should start with the floor which was not level, but sloped upwards to the firebox and was formed of wooden pegs, about 1½in square, set grain-end upwards in a cast iron frame. As if this was not bad enough, the rear sand boxes were inside the cab, about 9in above floor level, and formed a radius from the cab panel by the handrail to the boiler back-plate where they were approximately 3ft apart. The tops of these boxes were decked with flat, hard wood except where the filler caps were recessed level. The firehole door was of cast iron, and opened inwards to form a smoke plate, being hinged at the top. The door was operated with a ratchet lever on the left-hand side, adjacent to the driver, through two little cranks and levers which projected downwards, taking up even more of the firehole area. With wear on the pins and the concave door warping downwards, it required a very tricky shovelling motion to place the coal accurately against the tube-plate under the brick arch. Since the grate was level and shallow, and the brick arch fairly low, everything was against the fireman, whose work with the shovel had to be really accurate to maintain a thin even fire all over the grate. The space between the fire bars was very narrow so, provided the fire didn't go into holes, very good results could be obtained with a dancing fire. On the tender end, projecting into the coal alley, was a cast-iron hand wheel each side, about 15in diameter and having a large knob on the rim. The wheel on the fireman's side worked the tender hand brake, and the other one operated the water scoop. The height of these wheels was such that many an unwary set of knuckles has been chipped on them.
Dunn, J.M. Reflections
A fair number of 0-8-0 and 2-8-0 (M.M.) tender engines were stationed at Abergavenny for working coal trains from the M.T. & A. line forward from Abergavenny Junction to the north but there were no passenger tender engines. Whenever there was a special excursion from the M.T. & A. line to the north, Crewe always arranged to send a suitable engine to work the train forward from Brecon Road Station. On one classic occasion a "Prince of Wales" was sent to work one of these trains and when it arrived at Abergavenny Shed not many hours before train time it was found that it required repairs which could not possibly be carried out in time.
The only engine available to take its place was a 0-8-0 superheater "GI" tender engine which was scarcely a suitable steed for a train having to run from Hereford to Shrewsbury in the timings of the Bristol Express. Still, there was nothing else and when the driver arrived and the situation was explained to him, he said he would take the goods engine and added that if the Locomotive Inspector would travel with him, he would do his utmost to keep time with the train as far as Shrewsbury where it was to be handed over to the Great Western. To cut a long story short, the goods engine did keep time and covered the 50¾ miles between Hereford and Shrewsbury in 65 minutes, the time allowed for the G.W.R.'s West to North (or Bristol) Express. The first 38 miles to Church Stretton is mainly against the collar and some fast running down the bank thence to Shrewsbury is called for, so that when the train came to a stand at the latter station, the cross-heads, quadrant-blocks, big-ends 'and one or two axles were all hot but the honour of the L. & N.W.R. was once more vindicated!
Essery, Bob. LMS Garratts. Steam Wld, 2009 (263), 28-39.
Annual mileage statistics are quoted for the G2A for 1950: 19,533 miles.
Essery, R.J. and David Jenkinson An illustrated History of LMS locomotives. Volume One: General review and locomotive liveries. 1981. page 92-3.
Gahan, J.W. The "Super Ds" — the last L.N.W.R. locomotives on B.R.. Rly Wld, 1965, 26, 47-51. 7 illus., 2 tables.
Errata p. 160.
The L.N.W.R. eight coupled goods engines. Rly Obsr, 1950, 20, 3-10; 50-4; 78-84; 102-5; 129-30. + 11 plates (incl. 1 folding). 27 illus., 2 diagrs., 13 tables.
Reeves, John. LMS locomotive operating costs 1933-1935. Part 1 — Freight tender engines. LMS Journal, (7),7-21.
Operating costs for G2 1933-5: average annual mileage: 23515. Repair costs: 3.95 pence/engine mile; coal issued per engine mile 73.55lbs.
Stokes, Ken. Both sides of the footplate. Truro: Bradford Barton, [1985?]
Had experience of type whilst working at Grimesthorpe: complained about poor cab, poor tender, badly designed firehole, a regulator which operated in a different direction to that on the Midland, and a vacuum brake handle which was badly located and got very hot.
Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002. 268pp. 288 plates. 36 diagrs.
This is an excellent book which complememts the earlier Illustrated history. Chapter 8 describes working with 'Super Ds' mainly during the early years of British Railways. Harold Walkley was a fitter working at Abergavenny Junction and he describes (pp. 183-6 entitled Double home to Crewe) a journey made on the footplate (with official pass) on M9338 to Crewe and back. Walkley considered the locomotive to be master of its load with excellent brakes and gave a reasoable ride on a moderately fast freight. Pp 186-7 entitled Overloaded by thirteen wagons from Salop was related by Fireman David Williams and showed how 49029 could cope with an 8F load when the locomative was called on to relace a failed 8F at Condover and worked the train forward to Pontypool Road. Pp 186-7 describe how the Super Ds used to work excursions from Pontllanfraith to Bournville for the benefit of shopkeepers visiting Cadbury's. Two locomotives worked the train as far as Abergavenny Junction and a single Super D worked forward to Bourneville via Hereford and Worcester and back hauling ten to twelve coaches. Pp 188-9 Night shift fitter, Brecon Road Motive Power Depot, 1947-1952 are obvious and reminiscent of J.M. Dunn. Pp. 190-1 cover "Valve and piston examination on a 'Super D'; pp 191-2: Changing a spring and page 192: The LMS automatic blowdown valve.
[End of class at Tredegar]. Rly Mag., 1959, 105, 621.
Illustration of 49064 about to leave Tredegar for Ebbw Junction on 8 August 1959. Locomotive fitted with snowplough and with former LNWR staff posed in front.
Treloar, Peter. A jolly good wheeze. Steam Wld, 2005 (219) 48-55.
Black & white photo-feature of LNWR 0-8-0s throughout  their long history: cites J.R. Gregory's L.N.W.R. eight-coupled goods engines (Ottley 6373). No. 1811 (as class C1?); Webb three-cylinder compound (Class A) No. 1879; Class B four-cylinder compounnd No. 1432; Class E 2-8-0 (compound) No. 1222; two class B compound 0-8-0s heading coal train through Harrow (nichname of Swammies); class F 2-8-0 No. 1273 (Crewe official photograph); G2 No, 253 climbing away from Oxenholme with freight (H. Gordon Tidey excellently reproduced); LMS No. 9454 (G2 with Belpaire boiler); ROD 1639 and 2014 at St. Omer (Class G); G2A passing 8F No. 48420 at Bletchley; G2A No. 49125 with tender cab at Bescot; and G2A No. 49064 on Brynmawr to Barry Island excursion on 28 July 1957.

George Whale designs

Reed (Crewe Locomotive Works) is very dismissive of Whale: "In 1903 Whale had come too late in life to the top position. He was no longer willing or able to extend himself all-out in activities which hitherto he had known only at second-hand, yet which needed just at that time a strong chief of unremitting energy and a forward overall outlook that could cope with the changing trends in transport, mechanical engineering and labour. At the works the close continuous control was eased, statistical and graphic checks were given up, though 'running' charts were elaborated to show performance in mileage, repairs, consumption and so on compared with the engines of other companies, all based on annual reports and Board of Trade returns.

Whale was unable to appreciate the features that led Webb to watch closely the investment charges, returns on investments, purchases of raw materials, and stocks in hand of materials and parts that permitted evaluation of the efficiency of the works, and the contribution made by the locomotive department to the prosperity of the whole company.

Interest in these things lapsed, and no continued progress in the efficacy of the works or of the whole department as a unit of the railway was made in the further life of the LNWR.

The new policy at Crewe, quite clearly, was the smooth and easy running of headquarters and various sub-departments, and complete simplicity allied to cheapness in first cost of all new locomotives, the latter taken to such a degree that the first large 4-4-2T had a rigid trailing axle to save the small extra expense of a radial axlebox. Whale felt all these things could be well arranged by his five principal indoor and outdoor assistants without constant check by him. He was rarely about the works, possibly because of the energy shown by Trevithick.

His infrequent appearances in the drawing office could have arisen from sheer lack of interest in design as such; Jackson, Sackfield and leading draughtsmen had a much easier time. He relaxed his hold on 'running', and left much to his two principal assistants. When he did inject himself into that side it was not always with the happiest results, and he made errors into which he would not have fallen during his previous 25 years. Webb's tight grip on every phase of the whole department was loosened, and gradually a laxity in higher echelons was discernible, though not in the ranks, for discipline below foreman level was maintained up to World War 1.

Whale's inexperience in the direction and management of a whole department, plus encouragement from Trevithick, Harrison and Turnbull to finish with Webb and all his compounds, led him to acquiesce in the scrapping of many engines under 16 years old, and of many others of 1860-87 vintage, at a rate faster than new ones of equal total traffic capacity were built. Withdrawal and construction were not planned in unison, and as a result the LNWR was markedly short of main-line motive power in 1904-05; at times in the summer of 1904 train operation was chaotic from this cause. This imbalance also led to greater piles of scrap in the steelworks yard, and possibly from this dates the great untidiness of the south and south-east sections of the Steelworks yard that was so prominent a feature in the last few years of the LNWR and early years of the LMSR.

Yet in his first five months as chief, before Trevithick's desire to eliminate the 3-cylinder compounds had taken effect, Whale it was who confirmed the order for the last 20 of the 30 4-cylinder compound .'Bill Bailey' 4-6-0s, and he did so when only six or seven of the first batch were at work with indifferent results, and after the order had been placed on Crewe for the first five of the large 2-cylinder simple Precursor 4-4-0s. There must have been serious regrets a few months later at this unfortunate decision, but instead of cancelling the order work was soft-pedalled on the 4-6-0s to such an extent that the last four or five did not leave the works until January/March 1905.

Talbot devotes just a single Chapter (12) to Whale and his designs: Precursor 4-4-0; Experiment 4-6-0; 19in goods (4-6-0); 4-4-2T (Precursor tanks) and the Renown class which were rebuilt from Webb four-cylinder compound 4-4-0s.


Experiment: 1905-
The arrangement adopted (Whyte notation, in order of size) creates anomalies. The Experiment class was intended for the Crewe to Carlisle section, but other than the smaller (6ft 6in), and extra, driving wheels was virtually identical to the Precursor type.

Brislee, F.J. Combustion processes in English locomotive fire-boxes. Proc. Instn Mech. Engrs., 1908, 74, 237-68.
Brislee analysed smokebox vacua on the Experiment and Precursor classes and advocated that a blower should be fitted to the 4-6-0s to suppl;ement the air supply to the grate. In Bowen Cooke's contribution to the discussion the performance of Precursor No. 276 Doric was compared with that of Experiment No. 1987 Glendower. The former ran 34,348 miles; with a coal consumption of 882.25 tons (average 57.53 tons/mile, whilst the latter ran 34,013 miles, consumed 793.44 tons of coal (average 52.25 tons/mile). (via Atkins)

Atkins, P. West coast 4-6-0s at work. 1981. Chap. 3
Expresses surprise that superheaters were not fitted; alsthough the boiler pressure may have been raised to 195 psi on some locomotives. The boilers were not interchangeable with those on the 19 inch goods. The initial trials with No. 66 Experiment between Crewe and Carlisle on 21 May 1905 with a 373 ton train were disappointing. The maximum ihp did not exceed 994 and time was lost. This was due to the long shallow firebox which restricted the free gas area. In 1910 Bowen Cooke noted that Precurors achieved 69,124 miles between general repairs as against 64,835 miles for the Experiments: Atkins observes that this corresponded to the difference in coupled wheel diameter. .

Nock, O.S. The L.N.W.R. Precursor family: the Precursors, Experiments, Georges, Princes of the London & North Western Railway. 1966.
The tests of the Experiments against Ivatt Atlantic Nos. 1449 and 1451 on the LNWR and GNR respectively are described in moderate detail. The tests against a NBR 4-4-2 and CR Cardean are only very lightly sketched whilst the behaviour of Worcestershire on the GWR was a sorry tale compared with the behavior of Churchward Polar Star on the LNWR was mainly in support of Churchward's "expensive locomotive policies".
Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Plate 421: No. 66 Experiment: official photograph dated 3 June 1905; Plate 422: No. 507 Sarmatian at Euston; Plate 423: No. 1455 Herefordshire at Crewe North shed in August 1910 during interchange trials with GWR Polar Star; Fig. 117: weight diagram; Plate 424: 353 Britannic at Carlisle in late 1900s; Fig. 118 drawing; Plate 425: No. 2027 Queen Empress in late 1900s (sandbox located within driving wheel splashers; Plate 486: No. 61 Atalanta at Manchester London Road in late 1900s; Plate 427: No. 2638 Byzantium at Willesden on 10 July 1921; Plate 428: No. 1709 Princess May at Willesden on 3 June 1922; Plate 429: 353 Britannic at Rugby in June 1919 fitted with bogie shields to protect leading axleboxes from water thrown up by pilot engine when picking up water from troughs; Plate 430: No. 1002 Warwickshire at Stalybridge in early LMS days (capuchon on chimney); Plate 431: No. 2650 Buffalo with Belpaire firebox and patch on smokebox in early LMS days; Plates 432/3: No. 1361 Prospero as rebuilt with four cylinders arranged on Dendy Marshall system and with superheater.

No. 1361 Prospero: Dendy Marshall four cylinder arrangement
Locomotive, 1915 15 Oct., p. 219
Dendy Marshall, C.F. The Dendy Marshall 4-cylinder system of locomotives. J.Instn Loco. Engrs, 1916, 6, 190-218. (Paper No. 42).

In 1915 No. 1361 was rebuilt as a 4-cylinder locomotive using priciples developed by C.F. Dendy Marshall: see Nock's The LNWR Prcursor family and C.F. Dendy Marshall's paper presented to the Instn of Locomotive Engineers. Illustration of locomotive c1926: LMS Journal, 2006 (15) p. 58.
Atkins, P.
West coast 4-6-0s at work.
includes this locomotive in his Chapter 3: the lower coal consumption and higher mileage between repairs is noted. A sectional arrangement diagram is also included.

Atkins, Philip. Cast in a unique mould. Backtrack, 2011, 25, 442-5.
Casting cylinders for unique (and unusual) locomotives. In this form it achieved a significantly longer mileage bewteen repairs and lower fuel consumption than the unmodified members of the class. It lasted until June 1933.. .

19in. goods: 1906-
Experiment type with smaller (5ft 2½in coupled wheels). At the end of 1906 the first of the '19in. Express Goods' or 'Experiment Goods' 4-6-0s appeared. This was simply an Experiment with 5ft. 2½in. driving wheels and the only differences from the Experiments were those resulting directly from the smaller wheels: the long continuous splasher without coupling-rod splashers and the larger cab windows. The 19in. Goods was in effect the replacement for the Cauliflowers which Webb had had in mind when he produced the Bill Baileys, namely a mixed-traffic 4-6-0 for faster goods trains and slower passenger trains. Minor details were the capuchon on the chimney and the new style of number plate, stating 'Crewe Works' and the engine building date beneath the number. These plates appeared originally on the first batch of 'Precursor Tanks', which were built immediately hefore the first' 19in. Goods'. All the' 19in. Goods' had 3in. wooden pads for the short-tapered buffers, except the last batch built in 1910 which had long-tapered buffers. Some retained their wooden pads until the 1920s. Both sandboxes were of the type that was near impossible to fill: the leading one inside the frames behind the brake hanger and leading part of the front driving wheel, and the other one in the same position in relation to the centre driving wheel. Eventually, it was realised that these sandboxes were quite impractical and they were replaced by one behind the footstep, as on the 'Jumbos', and another in the driving splasher, as eventually adopted on the Experiments.

Cook's Raising steam shows that the effect of fitting Belpaire boilers did not extend stay life: 121 miles per stay changed with Belpaire firebox as against 311 with round top.

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Page 259: Plate 434: official view of the first of the class, No. 285, taken on 3 January 1907. Plate 435: No. 2606 about 1910, with original sandboxes; page 260: Figure 119 weight diagram; Plate 436 No. 1640 in 1917: page 261: Plate 437: No. 1350; Figure 120: 4mm drawing (side, front elevations & plan); page 262: Plate 438: No. 1625 at Willesden c1917; Plate 439: No. 2270 c1915; Plate 440: arrival of express at Buxton headed by member of class in summer of 1923; page 263: Plate 441: No. 1434 at Willesden in early LMS period; Plate 442: No. 2607 (with Phoenix superheater fitted between 1912 and 1914) at Manchester London Road.

Jamieson, A. Reminiscences of the L.N.W.R. Trains Annual 1967. 74-81
Fireman at Patricroft: refered to the 19inch 4-6-0s as Whales' Mankillers.

Post-Grouping modifications
Atkins, P. West coast 4-6-0s at work. 1981. Chap. 3
Observes that no attempt was made to superheat these locomotives, although 75 Belpaire boilers were constructed for the class between 1925 and 1930. The boilers were not interchangeable with those on the Experiments as the firebox was deeper on the 19 inch goods. Bogie brakes were fitted during the same period.


Precursor: 1904-
Talbot (Illustrated history) greets the Precursor class with: "Some nine months after Mr Webb's retirement, the first of George Whale's designs appeared from Crewe Works, the 'Precursor' class. It was a large simple-expansion express-passenger 4-4-0 with two inside cylinders. Though typically North Western in general appearance, it seemed outwardly quite different from anything that had gone before but in fact could be regarded as a logical development from the 'Jumbos' and 'Cauliflowers'. Plate 408: official photograph of No. 513 Precursor as built on 16th March 1904. Plate 409 No. 513 in service at Crewe in c1905; Figure 115 weight diagram; Figure 116 drawings (s/f and r elevations). The class was gradually modified to incorporate some or all of the features introduced on Bowen Cooke's George the Fifth class including the wheels, larger cylinders, superheating and Belpaire boilers. Many of the original locomotives took the numbers and names of Webb three-cylinder compounds (which is an excellent a posteriori reason for dividing LNWR locomotive development into three files!). Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Talbot's Figures illustrate some of these variants: Fig 410: No. 2582 Rowland Hill in original condition at Euston; Fig. 411: No. 1312 Ionic at Crewe in original condition; Fig 412: No. 622 Euphrates with George the Fifth type driving wheels on Camden Shed c1915; Plate No. 413: No. 2166 Shooting Star rebuilt with George the Fifth superheated boiler at Manchester London Road in February 1913 and Plate 414: No. 515 Champion and Plate 415: No. 1419 Tamerlaine - three in same condition and location; Plate 416: 659 Dreadnought in 1917 as rebuilt with superheater damper, piping to cylinders for mechanical lubricator, original driving wheels and Claughton-type tender; Plate 417: rear view of 2062 Sunbeam soon after rebuilding in February 1913; Plate 418: 685 Cossack at Crewe on 28 August 1924 (original boiler); Plate 419 No. 2202 Vizier at Bletchley on 4 October 1924 in virtually original condition; Plate 420 LMS 5275 Tiger (former 1439) with Belpaire boiler but still with slide valves.

Atkins, Philip. Crewe at home and away. BackTrack, 1989, 3, 11-80.
Includes results from  the 1909 comparative tests against a GNR large Atlantic between King's Cross and Leeds and between Euston and Crewe; and with McIntosh Cardean type between Crewe and Carlisle and between Carlisle and Glasgow: illus shows 1405 City of Manchester at Glasgow Cnetral and 903 Cardean alongside Crewe Works; Ivatt 1449 on fourteen coaches south of Crewe
Atkins, P. West coast 4-6-0s at work. 1981. Chap. 3
Firstly it is observed that this was the heaviest 4-4-0 type in Britain at the time of construction. On initial tests between Crewe and Rugby hauling 375 tons on 27 March 1904 No. 513 Precursor achieved an average 1002 ihp and a maximum 1197 ihp at 52 mile/h. In 1910 Bowen Cooke noted that Precurors achieved 69,124 miles between general repairs as against 64,835 miles for the Experiments: Atkins observes that this corresponds to the difference in coupled wheel diameter.
K. Cantlie (pp. 738-42) Discussion on Webber, A.F. The proportions of locomotive boilers. J. Instn Loco. Engrs, 1937, 27, 738 et seq
Submitted a graph (Figure 9) which compared the boiler performance at speeds from 0 to 70 mile/h of the following locomotives: "his" 4-8-4 for the Chinese National Railways; the Lord Nelson type with new boiler; and same type with original boiler; the enlarged Claughton/Patriot boiler; the original Claughton boiler; George V; Precursor, Teutonic; Greater Britain and Jumbo (comparitive performance was in descending order as listed) .
The last of the L.N.W.R. passenger tender engines. Rly Mag., 1949, 95,  324-5.
Line up at Crewe of Precursor 25297 Sirocco, Claughton 6004 and Prince of Wales 25752 (very brief history and individual illus. of each locomotive.
Nock, O.S. The L.N.W.R. Precursor family: the Precursors, Experiments, Georges, Princes of the London & North Western Railway. 1966.
The tests of the Precurors against the Marsh/Field LBSCR I3 4-4-2Ts are described at the beginning of Chapter 6 on the George the Fifth class (that is on page 64) following the Chapter mainly on later locomotive exchanges. In 1909 Bowen Cooke in association with Marsh on the LBSCR tested the Precursor type against the superheated I3 4-4-2T on the through working bewteen Brighton and Rugby on the Sunny South Express. Nock computed the coal consumption on the tank engine at 27lb per mile and the water conumtion at 22 gallons per mile and this led to superheating on the George the Fifth class.

4-4-0 Renown: 1908-
By the end of 1907, all the Alfreds had been converted to Benbows but for some reason no Jubilee was so treated and they all continued to run as four-cylinder compounds until 1908, when Whale undertook a major rebuilding of No. 1918 Renown The outside cylinders were removed, the inside cylinders lined up to 18½in. diameter and an 'Alfred the Great' boiler fitted, while such Whale features as cab and buffers were applied. The result was in effect a small version of Whale's 'Precursor'. Even then, further conversions were carried out quite slowly. By 1914 only five Jubilees and two 'Benbows' had become 'Renowns' and it was not until after the war that conversions were made in any numbers. Even then, three 'Jubilees' and seven 'Benbows' were never converted.

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Page 269: Plate 452: No. 1918 Renown: official photograph seen here, taken on 18th June 1908; Plate 453: No. 1943 Queen Alexandra was rebuilt in September 1916 and so must have obtained the lined livery after wartime economies ceased and lining out was resumed in October 1921. Apart froni. minor details such as windshields on the sand pipes, it is in much the same condition as Renown on first conversion; page 270: Plate 454: No. 1946 Diadem in 1914; Plate 455: No. 1973 Hood in 1924 at Crewe South shed in LNWR livery; Plate 456: 1257 (formerly 1914) Invincible at Willesden on 3 May 1924; page 271: Figure 123 weight diagram.

2-6-4T (proposed): 1909

According to Talbot who reproduces an official outline drawing (p. 288 Figure 130) a 2-6-4T was proposed under Whale with 19in x 26in cylinders, but Bowen Cooke wanted a superheated locomotive and due to the extra weight on the front-end this was translated into the 4-6-2Ts actually built.


Precursor tanks: 1906-
Whale brought out the first of his '6 ft. four-coupled side Tanks'. They were commonly known as Precursor Tanks as they were a tank version of the Precursor class with smaller driving wheels, nominally 6ft. but actually 6ft. 3in., and were used on suburban passenger work in the London, Birmingham and Manchester areas. It has all the usual Whale features of the time: buffers, capuchon, smokebox lubricator, etc as well as being the first engines to have a new style of number plate stating 'Crewe Works' and the engine building date. Again, the leading sandbox is the type that was near-impossible to fill, inside and beneath the frames, behind the leading brake hanger. To make things easier, the filler opening was brought outside the frames to the right of the upper footstep. The rear sandbox was behind the trailing edge of the rear driving wheel. Both rear sandboxes were filled from a central point in the cab floor as on the 'Coal Tanks' and other tank classes. Two features are typical of Webb tank engines: the tank was covered by a smoothly finished side-sheet, with no rivet heads showing, and rising a few inches above the top of the tank itself to provide a place for stowing fire-irons, and toolboxes were provided just inside the bunker coal rails with doors giving access to them in the rear of the cab.

Page 264: Plate 443: official view of  No. 528, the first of the class, taken on 17 July 1906: Plate 444: one of the second batch, No. 762 at Buxton in original condition: page 265: Plate 445: rear view of No. 762 at Buxton c1910; Plate 446: No, 1536 at Carnforth in 1921; Plate 447: No. 803: page 266: Plate 448: No. 44 official photograph (but in full LNWR livery) at Camden on 22 February 1924 (ventilator in cab roof and improved sandboxes like Jumbos): page 267: Plate 449: No. 875 at Birmingham New Street c1910;. Plate 450: No. 874 (bogie with large wheels) at Birmingham New Street c1910; Plate 451: No. 1589 (with Jumbo style sandboxes and cab roof ventlator) in 1920: page 268: Figure 121 weight diagram for No. 528 showing incorrect size of carrying wheels; Figure 122: 4mm drawing (side/front and rear elevations).

Cox, E.S. Memorandum to S.J. Symes. LMS Journal, 2007 (20), 20-3.
This is a facsimile reproduction of a typewritten document entitled Standard 2-6-4 tank engines and "Precursor" tank engines on Watford and Tring residential services dated 9 November 1932 from the Chief Mechanical Engineer's Office at Euston. The document is indicative that typewriter maintenance was less than perfect, although its cleanliness cannot be faulted. The 2-6-4Ts consumed nearly 43 lbs of coal per mile as against 54 lbs for the LNWR locomotives: a 20% saving. Maintenance costs were lower for the newer engines by about 15%. Timekeeping was greatly improved with the new engines. Footplate observations indicated that the 2-6-4Ts were smooth running, achieved rapid acceleration, but were draughty when running bunker-first. The 4-4-2Ts were rough, sluggish and difficult to fire.

Steam railcars (railmotors)
These were of two types: six single car units of 1905-6 and a two coach unit of 1910 which should really be credited to Bowen Cooke. Most were withdrawn in the late 1920s, but one lasted on the Moffat branch until 1948. photographs of

Aves, Bill. The London & North Western Railway steam railmotors. Rly Bylines, 2005, 11 (1), 40-7.
The units were numbered in the carriage series, but there was an additional locomotive unit. Article includes excellent official photographs, including front view with doors open to show smokebox; a naked engine unit which clearly shows the smallness of the boiler and the coupling rod on one side, No. 1 is shown at Prestatyn and at Dyserth in 1905; No. 3 is shown at Bicester, and again at Bicester showing modifications to improve ventilation (and with folding steps lowered); the two car unit at Lees on 6 July 1910. Caption notes that trailer survived, but is incorrect to suggest that centre doors caused loading problems; interior of car No. 29988 (very similar to centre door trailers used on Delph Donkey KPJ); No. 2 at Oddington halt; No. 10697 at Moffat; and No. 29988 at St Rollox Works in April 1947.
Jenkinson, David and Lane, Barry C. British railcars, 1900 to 1950. Penryn: Pendragon, 1996.
Pages 16-17 show (in the main) different views: the two car unit is the same photograph as in Aves above, but displayed on a larger scale. There are two exterior shots of No. 3 at Wolverton on temporary bogies, an official photograph of No. 1 with steps unfolded; No. 4 at Bletchley on Bedford service; No. 29988 at Moffat on 13 September 1939.

LNWR steam rail motor (railcar) No. 2 at Wendlebury in 1905. Rly Arch., 2009 (24) 78 lower
LNWR steam rail motor (railcar) at Bicester in 1908. Rly Arch., 2009 (24) 79

Bowen Cooke designs

Talbot allocates Bowen Cooke a single Chapter (13): poor Beames does not even get that (being no more than an appendeage to the Chapter on the convoluted development of eight-coupled locomotives on the LNWR).

Nock, O.S. LNWR locomotives of C.J. Bowen Cooke. Truro: Bradford Barton, 1977. 112pp.
Ottley 12207
Yeadon, Willie B. A compendium of LNWR locomotives, 1912-1949. Oldham: Challenger 1995. 154pp.


Brian Reed's Crewe Locomotive Works notes that the first really big engine of the LNWR, the Claughton 4-cylinder simple 4-6-0 of 1913, marked the incapacity of Cooke, Jackson and Sackfield as designers able to move with the times into higher powers.

Vital lessons from the GWR Polar Star's running from Crewe North shed in August 1910 were unabsorbed, particularly in boiler, valves and valve motion, possibly because contrary to what is generally supposed it was the GWR that initiated the exchange. Despite the small 5ft 2in diameter of the boiler barrel and moderate working pressure of l75lb/sq in, the locomotive weight and axle loads were right up to the maxima acceptable to the chief engineer. Walschaerts motion was the only Crewe innovation in them except for the size.

Day by day performance was too variable for them to be called successful, and they were expensive to run and maintain. Cooke himself realised that there were major defects, and was known to say that when the post-war rush of work was over he would take them in hand.

Spliced main frames

An unusual feature of LNWR locomotive practice from 1903 was the spliced main frame, put first on the Precursors and used on all new 8-wheel, 10-wheel and l2-wheel Whale and Cooke engines thereafter.

The big 4-4-2T and 4-6-2T engines had frame splicing at both ends. This arrangement, though linked to the 'running' era, seems to have arisen through a Homer's nod on the part of Webb and Earl. UntiI 1903 most frame machining was done for years in the wheel shop and frames up to 321/2ft length were slotted there, but the frame slotters in the new main machine shop appear to have been' so designed and located that they could take only 28ft, and no subsequent replacement or move of these slotters was made to rectify this, possibly because the spliced frame proved so convenient to repair after collisions. The last 20 Webb compound 4-6-0s of 1903-05, however, had their frames slotted in the wheel shop.

Prince of Wales class: 1911-
Introduced by Bowen Cooke in 1911, as an evolutionary development of the post-Webb LNWR 4-6-0. The class had inside cylinders and appeared to be crude in comparison with the Saint class produced at Swindon. The class was fitted with Joy valve gear, the failure of which caused the Betley Road accident (through fatigue in the valve gear). It is strange to consider that the Stanier Class 5 was originally cited as an "improved Prince of Wales" (although all that was common was the black livery. And in case the reader wonders why an inside cylinder 4-6-0 design was still being built in the 1920s, it is worth remembering that the GER 1500 class was receiving additions at the same time (under the stewardship of Gresley)..

As Whale had produced a 4-6-0 version of the Precursors so Bowen-Cooke produced a 4-6-0 version of the Georges. It was officially described as the '6ft. six-coupled Prince of Wales' class, after the first to be built. It was almost exactly an elongated George, the only differences being the 3ft. 9in. bogie wheels instead of 3ft. 3in., and the snifting valve above the running plate at the base of the smokebox. The leading sandboxes were in the same place, under the running plate, and the others under the middle of the splashers. The tender was a new type; the shape of the frames being the same but the springs had straight tops and the top of the tender was solid with double beading instead of coal rails. This type of tender remained standard until the first production batch of Claughtons appeared, on which the top had the same curved coping but with a single row of beading only. The first batch of ten Princes were built without any preliminary testing of prototypes. They all had particularly fine names, but all were new and non traditional North Western names; probably taken from Royal Navy warships. Twenty were built by NBL in 1915. They enjoyed a wide route availability and continued to be built during, and following, WW1. Ninety were built by W. Beardmore at Dalmuir in 1921/2. These lacked names and were painted in plain black.

Page 281: Plate 475: official view of No. 819 Prince of Wales on 16 February 1912; Plate 476: No. 1452 Bonaventure at Euston about 1914: page 282: Figure 127: weight diagram: page 283: Figure 128: 4mm scale drawings(side elevation and plan): page 283: Plate 477: No. 2175 Loadstone NBL official photograph of 1915: page 284: Plate 478: No. 964 Bret Harte at Manchester London Road in 1916; Plate 479 No. 446 Pegasus (built NBL) at Stockport in 1917: page 285: Plate 480: No. 1089 Sydney Smith at Tamworth c19120; Plate 481: No. 242 (Beardmore-built) at Willesden on 2 June 1923; Plate 482: Beardmore No. 551 at Camden c1924; Plate 483 No. 88 Czar of Russia plain black official photograzph, Camden, 22 February 1924. Tishy type: No. 484 official photograph No. 964 Bret Harte on 2 May 1923; Plate 485: No. 2340 Prince of Wales (with Ross pop safety valves) on 4 June 1924; Plate 486: No. 964 Bret Harte with Belpaire boiler; page 288: Figure 129: weight diagram.Walschaerts gear type with Belpaire boiler; Plate 487: No. 497 with Belpaire firebox c1925: page 288: official photograph of Beardmore locomotive for Wembley Exhibition LMS No. 5845 at Crewe Works on 6 January 1925.

Accident at Cheadle Hulme on 28 April 1922

Ministry of Transport Accident Report by A. Mount. 3 August 1922.
Also reported on a similar accident to a 4-6-2T at Furness Vale on 5 May 1922. In both cases the Joy valve gear failed due partially to lack of maintenance and partially to inherent design faults. J.P.S. Main assisted Mount in his investigation

Accidents at  Cheadle Hulme and Furness Vale L. & N.W. Ry. broken connecting rods. Locomotive Mag., 1922, 28, 332-6.

Improvements to valve gear
The Betley Road accident was to show failings in the Joy valve gear, but probably before this, Beames had introduced the unusual arrangement of external Walschaerts valve gear driving inside cyclinders: four were modified in this way? and a fifth (fourth?) was built by Beardmore for the Wembley Empire Exhibition. The modified locomotives were known as Tishies.

An ALTERED valve gear on a London, Midland & Scottish Ry. locomotive. Loco. Rly Carr. Wagon Rev., 1923, 29, 157-8. illus., diagr.(s.el.). Bibliog.
4-6-0 locomotive for the L.M.S. Railway: British Empire Exhibition. Engineering, 1924, 117, 664-5. 2 illus., 5 diagrs. (incl.s.el.)
A new (Beardmore product) modified "Prince of Wales" locomotive was an exhibit at the Wembley Exhibition.
REBOILERING L.M.S.R. locomotives. Rly Mag., 1925, 56, 220. illus.
Belpaire boilers.
REBUILT "Prince of Wales" class locomotive, London, Midland and Scottish Railway. Rly Mag., 1923, 53, 50-3. illus.
The WALSCHAERT [sic] valve gear adapted to inside cylinder locomotives. Engineering, 1923, 115, 596. illus., 2 diagrs. (incl.s.el.)

Retrospective and critical
Allen, Cecil J. Half-a-century of train travel—No. 14. The "Georges" and "Princes" of the London & North Western. Rly Wld, 1963, 24, 136
Records of both fast running and heavy haulage during WW1.
Atkins, P. West coast 4-6-0s at work. 1981. Chap. 5. The 'Princes'.
Notes that the class was not as powerful as the George the Fifth 4-4-0s and suggests that Prince of Wales were built in preference because of their higher haulage capacity (greater adhsion) and higher route availability. Notes the very large order placed with William Beardmore in 1920 for 90 locomotives and 60 tenders. From 1915 the class was fitted with a modified form of Joy valve gear, and further modifications were wrought in 1918 (known as indirect motion: diagram p. 37): claimed reductions in repair costs both for the 65 Princes and 203 G1 class 0-8-0s are quoted. Atkins considers at length the failure of the valve gear at Cheadle Heath on 28 April 1922; at Furness Vale (on a 4-6-2T) on 5 May 1922, and most notably at Betley Road on 28 May 1923. The latter led to tests of the fractured connecting rod at the National Physical Laboratory and to the modified form of Walschaerts gear designed by Beames. Atkins notes a report on this modification in Rly Gaz. in, 1923 18 May. The valve gear was difficult to set and performance was poor: the adoption of marine-type big ends on the Joy-fitted engines largely cured the problem. Atkins refers to tests conducted bretween Leeds and Carlisle in December 1923 of No. 388 with indirect Joy valave gear when its drawbar efficiency almost reached that of a 4P compound. Cites article in Stephenson Loco. Soc. J., 1978, February in which E.A. Langridge refers to work on developing an outide cylindered Prince of Wales class with long-travel Walschaerts valve gear (this followed unsuccesful efforts to design a Prince of Wales with inside cylinders and Caprotti gear): Atkins suggests that it was this last design which led to the Class 5 being known initially as 'Rebuilt Prince of Wales class'
Beckerlegge, W. The L.N.W.R. "Prince of Wales" class 4-6-0 with Walschaerts gear. Rly Obsr, 1941, 13, 101. illus.
Brooks, Mike. Naming the first LMS Pacifics, Railway Wld., 1979, 40, 79.
Names which would have been given to Prince of Wales and Claughton classes if Derby dictat to remain nameless had not been enforced
Bulleid, H.A.V. Master builders of steam. 1963. p. 141.
On the Works side there was much to impress: Crewe boiler-making came as a surprise to Stanier, who had not fully appreciated the stalwart performances put up by the standard range of L.N.W.R. boilers, on such classes as the "Precursors" and "Experiments." These boilers were decidedly on the small side for the engine duties, and the L.N.W.R. footplate teams were expected to, and did, thrash them to keep time with heavy trains. They made a rather invigorating noise, and at night looked highly pyrotechnic. But these boilers did not require excessive maintenance and had long lives; and when it came to steam-raising capacity a " Precursor" class boiler topped 20,000 lb./hr. on test. Stanier obtained some of his boiler data at first hand, and was seen followmg foreman Tizard through the fire hole of a "Prince of Wales" class boiler in Crewe boiler shop: as a result he confirmed his opinion that the better circulation in a Great Western type of boiler was a great advantage. But incidentally he found Crewe boiler-making of Swindon quality but cheaper,
Cantlie, K. discussion on Skeat, W.O., The Great Eastern Railways "1500 Class" locomotives. Trans. Newcomen Soc., 1969, 42, 98.
The discussion had raised the question as to why the 1500 class used inside cylinders: Cantlie noted that "with an inside-cylinder engine, everything tended to stay fairly tight; with outside cylinders there was always something that needed to be tightened up at the end of a run" (presumably the remarks were based on LNWR experience).
Cox, E.S. A modern locomotive history: ten years' development on the L.M.S. — 1923-1932. J. Instn Loco. Engrs., 1946, 36, 100-41. Disc.: 141-70; 275-6. (Paper No. 457)
Fig. 12 (graph) shows coal consumption within period 1927-1938: a more or less constant 52lbs/mile
Cox, E.S. Memorandums to S.J. Symes from E. S. Cox – Prince of Wales class engines. LMS Journal, 2005, (10) 33-4.
13 May 1931: survey at major former LNWR sheds. Frame failures were common and attemt was made to cost this fault. Hot boxes were common on the leading coupled axle especially on the left hand side. Other problems included the dieblocks on the Joy valve gear, brakes, sanding, knocking, the smokebox door and piston valve rings.
Gregory, J.R. The "Tishy" Princes of the L.N.W.R.. Rly Obsr, 1948, 18, 40.
Groves, Norman. Locomotive masqueraders. Railway Wld., 1960, 21, 160.
Concerns locomotive exhibited at the Wembley Exhibition
Hall, G.L. Report on the failure of the right-hand connecting rod of an engine, which occurred on the 28th May at Betley Road, near Crewe. Ministry of Transport: Railway accidents.. . which occurred during the three months ending 30th June, 1923. London, HMSO, 1923. 15 p. + 4 plates. illus., 3 diagrs. (incl. s. el.), 5 tables.
The last of the L.N.W.R. passenger tender engines. Rly Mag., 1949, 95,  324-5.
Line up at Crewe of Precursor 25297 Sirocco, Claughton 6004 and Prince of Wales 25752 (very brief history and individual illus. of each locomotive.
Locomotive connecting rod failures. Rly Engr, 1923, 44, 447.
Editorial comment on the above accident.
McNaught, R.S. A "Princes" valedictory. Rly Mag., 1949, 95,146-8; 164-5.5 illus.
Nock, O.S. The L.N.W.R. Precursor family: the Precursors, Experiments, Georges, Princes of the London & North Western Railway. 1966.
Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Tuplin, W.A. The North Western "Princes". J. Stephenson Loco. Soc., 1961, 37, 292-7. 5 diagrs. (incl. 2 s. els.)
Tuplin, W.A. The North Western "Princes". Trains ill., 1953, 6, 212-19. 2 illus., 3 tables.
Tuplin, W.A. North Western steam. London, 1963.
On the Tuplin page there is an extensive extract from a lengthy description of the work involved in the out-and-back Hull job from Edge Hill shed in the early years of WW1.

Claughton class: 1913-
As will be seen the Claughton class generated a large, and mainly derogatory literture, and notes on its original entry into service are not yet available. Talbot (Illustrated history) is wonderfully neutral and his captions form the basis for the introduction. Although the Georges and Princes were capable of magnificent work, a more powerful engine was needed for the heaviest trains, especially north of Crewe, and early in 1913 Bowen Cooke produced the first Claughton class four-cylinder express-passenger 4-6-0 from Crewe Works, No 2222 Sir Gilbert Claughton. As first built, No. 2222 differed in a number of details from what later became usual. It had four windows in the front of the cab (later engines had only two), one set of safety valves (soon changed to two), the brake-valve outside on the left with the train-pipe connection round the firebox in front of the cab (later moved inside like all the others), standard lamp sockets (replaced by lamp irons immediately on No. 2222 but not on the first batch of nine until about 1915) and square lower corners on the front bufferbeam (cut away on the first ten after one fouled Bletchley platform later in 1913). The riveting round the smokebox was also changed on later engines as was the type of tender, and in the photograph reproduced by Talbot there is no cover over the reversing gear, though one was soon fitted.

The early performance was notweworthy:  The key reference is Bowen Cooke's own description of early trials which was reported at a meeting of the Institution of Civil Engineers. There were also reports in The Engineer in February 1914.On the first trial from Euston to Crewe 1617 ihp was achieved at Boxmoor and Nock (LNWR locomotives of C.J. Bowen Cooke) noted: "Taken all round, the run of Sunday, 2 November 1913, represnted a performance the like of which had never previosly been seen in Great Britain, and it is doubtful even if any French compound 'Pacifics' had greatly exceeded such a sustained output of power." No. 1159 Ralph Brocklebank with a load of 434 tons had run from Euston to Crewe at almost exactly 60 mile/h at an averagle 855 dbhp. 1155 ihp had been attained on Camden Bank; 1314 ihp approaching Atherstone at 70 mile/h and 1183 ihp at 72 mile/h passing Tamworth. On 4 November 1913 the 13.12 express from Crewe to Carlisle included the dynamometer car and a maximum of 1669 ihp was achieved between Carnform and Shap

Mean drawbar pull over whole trip 1.83 tons
Mean power at drawbar over whole trip 663
Average indicated horsepower from diagrams taken at 17 points 1387
Average drawbar horsepower at the 17 points where the indicator diagrams were taken 920

Nock's comments follow: In contrast to the run of 2 November, however, there were lengthy stretches over which the engine was being worked under very easy steam, as from Coppull down to Preston, and from Shap Summit down to Carlisle where for much of the distance the engine was running without steam. The train had passed Shap Summit 10 minutes early (!) and further hard running would have been pointless.

From a study of these tests, in the data published early in 1914, it is evident that the L.N.W.R. had a locomotive type of outstanding power capacity in the 'Claughton', and the records of indicated horsepower were, by a considerable margin the highest that had been attained by any British locomotive at that time. Nothing comparable had been published, and when I was researching for my book' 'Fifty Years of Western Express Running" and had the privilege of examining the dynamometer car records at Swindon I could find no records taken with 'Star' class 4-6-0s that came anywhere near the L.N.W.R. records. It is true that in all probability there was never an occasion to work a Great Western 4-6-0 to the same extent; and at that time there was no other British locomotive in the same range of maximum capacity. The other feature of the Euston-Crewe run of 2 November, which was repeated two days later, was the high ratio of equivalent drawbar, to indicated horsepower. On the eight sets of diagrams taken between Carnforth and Shap Summit, the average e.d.h.p. was 1190 against an i.h.p. of 1537—-a ratio of 77.6 per ent, between speeds of 37 and 69 m.p.h.

Nock then referred to the discussion at the Institution of Civil Engineers, in which Aspinall, Churchward, Gresley, Hughes, Raven and several engineers connected with Dominion and Colonial railways took part, and to which Dr. Schmidt, the designer of the superheater made a written contribution, and thought it surprising that not one of those participants, not even Churchward, mentioned Bowen Cooke's data (but it was rare to comment upon iterms brought up in the discussion if they were not included in the main paper). Those who consider this to be some statistical errors either on the part of Bowen Cooke and his team or to Nock's interpretation should consult Rutherford (below).

Following WW1, Nock (LNWR locomotives of C.J. Bowen Cooke page 85) noted that locomotive in the Patriot batch were fitted with improved lubrication to the cylinders and piston valves, Ross pop safety valves and continuous sloping grates and that Bowen Cooke had commented to Kenneth Cantlie (one of his pupils) that "he was going to take the Claughtons in hand".

Having read Nock's panageric the reader may be tempted to ponder on E.S. Cox 's statement on p.79 (v.1) of his Locomotive panorama: "Putting to sea in a sieve was nothing compared to the problem of how to put these engines to rights!"   Backtrack was the venue for a vigourous discussion on the merits, or otherwise, of the design.

Reed in his study of Crewe Works is damning: "The first really big engine of the LNWR, the Claughton 4-cylinder simple 4-6-0 of 1913, marked the incapacity of Cooke, Jackson and Sackfield as designers able to move with the times into higher powers. Vital lessons from the GWR Polar Star's running from Crewe North shed in August 1910 were unabsorbed, particularly in boiler, valves and valve motion, possibly because contrary to what is generally supposed it was the GWR that initiated the exchange. Despite the small 5ft 2in diameter of the boiler barrel and moderate working pressure of l75psi, the locomotive weight and axle loads were right up to the maxima acceptable to the chief engineer. Walschaerts motion was the only Crewe innovation in them except for the size. Day by day performance was too variable for them to be called successful, and they were expensive to run and maintain. Cooke himself realised that there were major defects, and was known to say that when the post-war rush of work was over he would take them in hand."

A.J. Powell in a somewhat obscure source (Steam Days) gave an excelent evaluation of the limitations of the design and how these could have been eradicated:. "The design of the 'Claughtons' exhibited some serious deficiencies, but also some redeeming features. Crewe had been forced by the Civil Engineer to use a smaller diameter boiler than intended, to keep axleloads down. A Belpaire firebox was used for the first time on the LNWR; this seems to have been an afterthought, since the inner firebox was shaped to suit a round-topped box and thus the full tube bank potential could not be realised. The grate area of 30.5 sq ft was ample, and the firebox was very deep, with the rear third of the grate level. As a result the ashpan was extremely shallow where it arched over the trailing coupled axle, limiting air flow to the grate. These factors led to erratic steaming. In addition the firebox sat very low over the trailing axleboxes and led to a very poor lubrication system for them, so that despite the adequate journal size (8¼in dia x 9¾in long) they were very prone to overheating.

The frames were relatively shallow (only 18in deep over the horn gaps) and gave considerable trouble with cracking; on at least one occasion a frame plate cracked in the works during the final crane lift after General Repair! The single, wide piston valve rings leaked badly when worn, a feature not confined to the 'Claughtons'! The coupled-wheel coil springs were weak and breakages frequent. The cab and heavy splashers were constantly working loose, leading to pipe fractures and vacuum brake failures. The very flat smokebox doors warped and admitted air, despite four 'dogs' round - the circumference. There were other lesser evils. On the credit side the four cylinders, cast in pairs giving a very rigid structure, were of enlightened design, with good steam and exhaust passages, while the outside Walschaerts valve gear and Trickported valves, though having short lap and travel, were good at getting steam through the cylinders. Driving on to one axle, with equal length connecting rods, the reciprocating balance was well-nigh perfect". See hammer blow below..

Two major alterations were noted in contemporary literature. The first was the application of Caprotti valve gear under the direction of Beames in 1927. The second was the design of an entirely new boiler at Derby. This boiler also formed the basis for the Patriot design. O.S. Nock presented several studies of this design; most of which favourably evaluated performance in traffic.

London and North Western Railway locomotive "Patriot". Loco. Rly Carr. Wagon Rev., 1920, 26, 189.
No. 1914 in unlined black livery, but with numerals and nameplate outlined in white.
Four-cylinder "Claughton" class locomotive, London, Midland and Scottish Railway. Loco. Rly Carr. Wagon Rev., 1923, 29, 253. illlustration
No. 5971 Croxteth showing painted in standard (former Midland Railway) livery with number on tender and "LMS" on cab side. Number on smokebox door. For inspection by Directors. Picture courtesy of F. Dingley, superintendent oof motive power LNWR or "A" Division

Hammer blow

Gribble, C. Particulars of locomotives employed in the tests and of others examined for the Committee. Appendix D. Department of Scientific & Industrial Reserch. Report of the Bridge Stress Committee. London: HMSO, 1928.

Balancing of 4-cylinder Engines
"An interesting comparison can be made between the balancing of the GWR 4-cylinder simple (Star class) and the LMS Claughton engine Series 5900. Whereas in the GWR engine the inside and outside connecting rods drive different axles, in the LMS engine they all drive the same axle. In the former engine an appreciable proportion of reciprocating parts was balanced separately for the outside and inside cylinders respectively, with the result that the hammer-blow of each driving axle was considerable, but that on the front axle was opposed to that on the second axle. In the Claughton engine the reciprocating parts of the four cylinders mutually balance each other, without the addition of any balance weights apart from those required to balance rotating masses, with the result that the engine has no hammer-blow whatever. The system of balancing which was adopted in the GWR 4-cylinder engine did not give an advantageous result from the bridge engineer's point of view, since although the total hammer-blow was small, the hammer-blow on the driving axle exceeded that of the corresponding 2-cylinder engine (Saint No. 2906). The balancing of these engines also, has now, we are informed, been adjusted, and a different system adopted.

1927 : Caprotti valve gear : Fowler/Beames.

BRITISH application of Caprotti valve gear: a general description of the principles of this gear, with the results of tests recently made on the L.M.S.R.. Rly Engr, 1927, 48, 397-8. 6 diagrs., table.
CAM-OPERATED valve gear locomotive, L.M.& S. Ry.. Loco. Rly Carr. Wagon Rev., 1927, 33, 348-51. illus., 3 diagrs., table.
CAPROTTI valve gear. Engineer, 1928, 145, 720-1. 4 diagrs., table.
The CAPROTTI valve gear applied to locomotives. Rly Mag.. 1928, 62, 290-4. illus., 7 diagrs.
4-6-0 type locomotive with Beardmore-Caprotti valve gear; L.M.S. Railway. Engineering, 1927, 124, 292-3. 3 diagrs., table.
TESTS of locomotive fitted with Caprotti valve gear. Rly Engr. 1928, 49, 95-7. illus., 6 diagrs. (incl.s.el.), table.

1928: enlarged boiler: Fowler:
This was the boiler which formed the basis for the Patriot class wherein the significance of the boiler is considered at greater length (using the Royal Scot chassis).

"CLAUGHTON" locomotive fitted with enlarged boiler, L.M.S.R.. Rly Mag., 1928, 63, 70 + plate f.p. 1. illus., diagr.
REBOILERED "Claughton" class locomotive, L.M.& S. Ry.. Loco. Rly Carr. Wagon Rev., 1928, 34, 171. illus.
RE-BOILERED "Claughton" locomotives, L.M.S.R.. Rly Engr, 1928, 49, 333. illus., 3 diagrs.

Webs in coupled wheels
This carried over to the Jubilee class where webs were fitted to some, or all, of the coupled wheels. No. 5966/5512 was so modified. see Townsin page 174

1932: Kylala blast pipes fitted experimentally to large and small boiler types.
Un-titled reference : Rly Mag., 1932, 71, 307. illus. p.418.
Nock, O.S. The Settle and Carlisle Railway: a personal story of Britain's most spectacular main line. 1992.
pp. 125-6 give infuriatinly imprecise information on tests between Kylala-fitted No. 6001 and same locomotive without it on 3 and 4 June 1930. Writer recorded same locomotive on Leeds to Carlisle express on 30 Aug 1930: with 375 tons from Hellified but with 472 as pilot. Excellent run very free running down to Carlisle (but is far from clear whether locomotive was fitted with Kylala blastpipe at time)..

Retrospective and critical

Allen, C.J. The last of the "Claughtons". Trains ill., 1949, 2, 71-7.8 illus.
Historical survey.
Atkins, P. West coast 4-6-0s at work. 1981. Chap. 6
Brings together much that is considered elsewhere, but also a great deal which may otherwise be missed by anyone attempting to produce a comprehensive history of this design which must rank as the greatest "near miss" in terms of achieving full success. In part the failure was due to the excessive caution displayed by the aptly named Chief Engineer, E.C. Trench. Highly critical of original boiler: the tube and flue density was approximately 20% greater than that of the boiler for the rebuilt Royal Scots. The class was either over cylindered or under boilered and the cylinder bore was reduced from 16in to 15½in and 15¾in experimentally and then 15¾in was standardized. Notes that sectional drawings plus a full resume of the 1913 tests apeared in Rly Gaz., 1917 6 July. In 1921 a thermic syphon (termed water circulator) was fitted to one boiler, but according to Atkins this was not mentioned until device adoipted by Gresley and Bulleid. A diagram is included on p.55. Schemes not adopted included an improved arrangement of the boiler tubes to be in line with the succesful George the Fifth, experimental blastpipes with exhaust splitters, the possible adoption of steel firebox, tenders with roller bearings and electric lighting for the locomotive. Bowen Cooke had discussed higher boiler pressures with Churchward, but was sceptical of Churchward's assertion that higher pressures were not accompanied by higher maintenance costs.An intriguing sideline is that according to Kenneth Cantlie that the number 69 had been cast for the War Memorial locomotive Patriot:.soixante neuf was a euphemism for brothels on the Western Front and 1914 was substituted. It had been intended to name another locomotive Victory.The LNWR turntables at Leeds (Farnley Junction) and Holyhead were too short to accommodate the locomotives: the NER turntable at Leeds was used..
Barradell, M. 2222 and all that. Rly Wld, 1958, 19, 127-8. 2 illus.
Brief assessment.
Barrie, D.S.M. The locomotives of the L.N.W.R., 1897-1922. Part VI—The "Claughton" 4-6-0's. Rly Mag., 1937, 81, 278-86. 11 illus., 2 tables.
The previous five parts were by O.S. Nock.
Bowen Cooke, C.J. [discussion on], Fowler, H. Min Proc Instn civ Engrs., 1914 paper 4084
This is a key reference to the initial performance of the Claughton class on test: it should be noted that Churchward and Gresley were also present..
K. Cantlie (pp. 738-42)
Discussion on Webber, A.F. The proportions of locomotive boilers. J. Instn Loco. Engrs, 1937, 27, 738 et seq
Submitted a graph (Figure 9) which compared the boiler performance at speeds from 0 to 70 mile/h of the following locomotives: "his" 4-8-4 for the Chinese National Railways; the Lord Nelson type with new boiler; and same type with original boiler; the enlarged Claughton/Patriot boiler; the original Claughton boiler; George V; Precursor, Teutonic; Greater Britain and Jumbo (comparitive performance was in descending order as listed) .
Brooks, Mike. Naming the first LMS Pacifics, Railway Wld., 1979, 40, 79.
Names which would have been given to Prince of Wales and Claughton classes if Derby dictat to remain nameless had not been enforced
Cantlie, K. ? The L.N.W.R. "Claughtons". Rly Mag., 1938, 82, 63-4.
Editor had received two letters in response to feature on Claughton Class in October 1937 Issue. One letter writer is identified, but other was from China: hence suggestin that was from K. Cantlie. Your contributor laid emphasis on the locomotive exchange between the G.W. and the L.N.W. Railways as being the cause of the 'Claughtons'. I think that this hardly does justice to my old chief, Bowen-Cooke. He was a great admirer of Germand designs and practice. When he superheated the 'Precursor' and 'Experiment' classes he put in 24-element superheaters and thus gave them high superheat, as was then more usual in Germany than in England, for it was the idea among locomotive men that superheating was a substitute for compounding, though the French were beginning to make people see that the two could be complementary. Furthermore; even though Bowen-Cooke might have wished to compound his engines, the loading-gauge was against him, and the cylinders fitted to the 'Claughtons' were the largest diameter possible, without fouling the loading gauge. Indeed, the first 'Claughton' when running trials removed quite a lot of the platform coping at Bletchley with the corner of its front bufferbeam, which projected but little beyond or below the cylinders; the corners of the buffer-beam were then chamfered off. With 16 in. as the maximum diameter of outside cylinder, and anything more than 23 in. as impossible inside, there were only two alternatives: high pressure or a simple engine. High pressure was against BowenCooke's. ideas, and I frequently heard him object to it on account of the increased boiler maintenance; 'Churchward says it doesn't make any difference, but I'm certain that it must do.' Therefore the only possibility was the use of four high pressure cylinders, with superheat in line with German practice. Bowen-Cooke was, "I think, a trifle nettled by the constant talk of his having imitated the' Stars' in his design, and claimed that they might just. as well be described as 'Alfred the Greats: simplified and with another coupled axle. This was certainly a defensible argument, for the cylinders in both designs were in line and drove the leading coupled ax)e; the valve operation was identical — Walschaert gear came comparatively late in the design — with the rocking-levers in front of the cylinders. In the 'Stars' divided drive was used, and the cylinders were in two planes. Bowen-Cooke realised that a large multi-cylindered 4-6-0 was required for the L.N.W.R. What was more natural than that he should experiment with the nearest approach to what he wanted — the G.W.R. 'Stars'? Even if the 'Star' had been a complete failure on exchange, the 'Claughtons' would still have been built, for they were inevitable."
Cox, E.S. Chronicles of steam. London: Ian Allan, 1967.
Page 40: Table 2: an attempt to show how far advanced the three-cylinder 5XP  was ahead of the modified Claughtons, but it was only very marginal in the case of the Caprotti-fitted Claughtons.alternative draughting arrangements, including double chimney
Fowler, H. Discussion at :International Railway Congress Association 11th Session, Madrid, 1930. 2nd Section. Question 6. Improvements in the steam locomotive. Bull. Int. Rly Congr. Ass., 1931, 13, 115.
Savings achieved by the application of Caprotti valve gear.
Hall, Stanley. Railway milestones and millstones: triumphs and disasters in British railway history. 2006.
Regarded as a millstone mainly because of initial inadequate boiler
Jenkinson, David. The classic 'Claughtons' in LMS days. 538-43.
illus.: No 5944; No 5971 Croxteth; No 5979 Frobisher; No 5900 using oil fuel; No 5923 Sir Guy Calthrop; No 5994; No 5905 Lord Rathmore; No 5955 paired with an ex ROD [GC type] tender; No 5986 with a new and larger LMS boiler; No 5999 Vindictive with Walschaerts valve gear; No 6023 Sir Charles Cust with Caprotti valve gear; No 5962; No 6004 with No 8532. See letter from Peter Davis in Volume 13 page 109.
The last of the L.N.W.R. passenger tender engines. Rly Mag., 1949, 95,  324-5.
Line up at Crewe of Precursor 25297 Sirocco, Claughton 6004 and Prince of Wales 25752 (very brief history and individual illus. of each locomotive.
The L.N.W.R. "Claughtons". Rly Mag., 1938, 82, 63-4. table.
Notes on the origin of the Bowen Cooke design.
The last of the L.N.W.R. passenger tender engines. Rly Mag., 1949, 95,  324-5.
Line up at Crewe of Precursor 25297 Sirocco, Claughton 6004 and Prince of Wales 25752 (very brief history and individual illus. of each locomotive.
Mills, Bob. 'Claughton' comments. BackTrack. 1999, 13, 347-9.
Very critical of design, although does establish a few good points, notably the use of Walschaerts valve gear, the undivided drive, the high degree of superheat and the use of Trick ports. Mills considers that the locomotives were badly constructed and lacked the long travel valve gear fitted to the Stars,: Makes reference to the exchange of the Star class (4005 Polar Star) with an Experiment (1471 Worcestershire), the latter having a disastrous time on the GWR, much to Churchward's relief as Board members had considered his locomotives to be excessively expensive. Refers back to a photographic feature compiled by David Jenkinson (Volume 12 page 538) which had been laudatory: Letter by Johnson on page 508 gives reason for rapid withdrawal from service. Letter by Peter Davis refutes much of Mills' contribution (page 569). Letter by Bob Mills (14 pp. 65-6) attempts to refute much of the contribution from Peter Davis. illus.: Claughton No 154 Captain Fryatt; GWR No 4005 Polar Star on LNWR; Claughton No 6021 Bevere in original condition and No 5927 Sir Francis Dent; No 6017 Breadalbane soon after returning to service in rebuilt condition.
NEW L.M. & S.R. locomotives. J. Instn Loco. Engrs, 1928, 18, 2-6 + 5 folding plates. 2 illus., 2 diagrs. (s. els.), 4 tables.
Includes results of trials performed over Shap with a "Royal Scot", a 4P compound, a standard "Claughton" and a "Claughton" with a modified (not enlarged) boiler.
Nock, O.S. The "Claughton"class, LNWR: an analysis of their design and performance. Locomotive Mag., 1944, 50, 38-41; 60-1;
Nock, O.S. The "Claughton"class, LNWR: an analysis of their design and performance. Locomotive Mag., 1944, 50, 73-6.
In this part Nock notes their use on the Settle & Carlisle line where tthey performed well on climbing and achieved speeds up to 90 mile/h downhill due to their perfect balancing
Nock, O.S. British locomotive practice and performance. Rly Mag., 1964, 110, 289-95. table.
A table, produced from R.C. Bond's records, of coal and water consumption and general performance on Shap test runs with a Castle class,a standard Claughton, a Caprotti Claughton and the Claughton with a modified (not enlarged) boiler. The tests took place in 1926/27.
Nock, O.S. The "Claughton" class, L.N.W.R. an analysis of their design and performance. Loco. Rly Carr. Wagon Rev., 1944, 50, 38-41; 60-1; 73-6. 2 illus., l0 diagrs.
Nock, O.S. LNWR locomotives of C.J. Bowen Cooke. Truro: Bradford Barton, 1977. 112pp.
Table on page 45 compares major dimensions of Claughton and Churchward Star classes: the former had larger grates and higher superheat, but lower boiler pressure and smaller tubes. Nock considered the valve gear fitted to the Claughtons to be far simpler. Chapter 6: The first 'Claughtons' at work' opens: "The early months of 1913 proved a vintage period on the L.N.W.R.". Chapter 7: The historic trials of November 1913 is based largely on based on material published in the Engineer for February 1914 and discussion of the Fowler paper on superheating in which Bowen Cooke added details of these trials.
Nock, O.S. Three generations of West Coast 4-6-0's Claughtons – Royal Scots – converted Scots. Rly pict., 1949, 2, 76-81; 98-104.8 illus., 2 tables.
Development is emphasized.
Phillips, Hugh. The Caprotti Claughtons ~ and others. Rly Wld, 1988, 49, 710-14.
Attempts to improve the performance of the Claughton class by modifying them with Caprotti poppet valve gear, modified piston valves, larger boilers and by rebuilding them as three-cylinder locomotives. Only the last was fully effective, although the other modifications produced substantial improvements. Also notes that four locomotives were fitted with Kylala blast pipes and that four Caprotti-fitted locomotives had their crankshaft settings altered to 135°. Also observes that Swindon was informed of the advantages of the Caprotti valve gear on 7 April 1927: these included the elimination of joint pins and spindle packing; no lubrication of valves and valve motion; ease of cambox removals; freedom to drift without steam; greater steam expansion; decrease in oil, water and fuel consumption.
Pollock, D.R. The rebuilt Claughtons. Rly Obsr, 1947, 17, 186-7 + plate. 3 illus.
Powell, A.J. Were the Patriots really necessary? Steam Days, 1992, 1 (34), 307-11; 330-1.
Takes a more radical tack, than in earlier contribution by suggesting that a thorough approach to removing the limitations from the potentially excellent Claughton class would have given far better value. Like the Patriot class this would have exploited the G9½S boiler on some of the locomotives.The first task should have been to overcome the chassis deficiencies and thereby keep the boiler options open. This would have involved a full refurbishment of all 'Claughtons' at General Repairs, with new rear end frames, laminated springs throughout, mechanical axlebox lubrication, long-travel valve gear, new cabs and splashers and the modifications for a raised boiler. Thus curing the steaming problem by increasing the draught through the bottom of the grate and the rear problem. The cost of this work would probably have been £1,500 per engine or a little more. Secondly, to begin replacement of life-expired original boilers with the large G9½S type.
Robinson, E.E. L.M.S. locomotives the Claughton and Baby Scot classes (complete to February 24th 1934). Egham (Surrey), Author, [1934]. [ii], l0p. + 4 plates. 4 illus. (Class book No. 1) A stock list.
Robinson, E.E. L.M.S. locomotives alterations to "Claughton" and standard 5x class passenger engines. Egham (Surrey), Author, [1934] . 4 p. (Supplement to class book No.1: to 6th October, 1934).
Rutherford, Michael. More about measurements. (Provocations [Railway Reflections No. 16]). Backtrack, 10, 209-16.
Locomotive testing and the use of the measurements obtained, sometimes for less than straight forward reasons. Rutherford strongly asserts from the data stored at the NRM that the Claughtons were far more powerful than is frequently considered, and certainly comparable with the Castle class, the tests of which on the LMS were used to reinforce exisiting predilictions.
Sandford, D.W. discussion on Cox, E.S. A modern locomotive history: ten years' development on the L.M.S. — 1923-1932. J. Instn Loco. Engrs., 1946, 36, 163-4. (Paper No. 457)
Replacemrent of Schmidt piston rings by narrow rings on the unmodified Claughtons reduced coal consumption to that attained by Caprotti-modified locomotives.
Stanier, W.A. The position of the locomotive in mechanical engineering. Proc. Instn mech. Engrs, 1941, 146, 50-61 + 4 plates. 13 illus., diagr., 3 tables. (Presidential Address).
No. 5917 Claughton class Euston to Carlisle and return
Stuart, D.H.  The L.N.W.R. "Claughtons". Rly Mag., 1938, 82, 63-4.
There was an alteration to one of the "Claughtons" which was not recorded in the article on Claughton Class in October 1937 Issue, and this was a modification that took place at Derby about' the same time that the first engine was altered to Caprotti motion at Crewe. Derby was for achieving results similar to. those obtained with Caprotti motion by certain modifications of a standard engine. The engine concerned was No. 1093 and Derby altered the fire bars to give very greatly increased air space and also modified the valves. The results of trials of this engine, together with those of standard 4-4-0 compound No. 1073, which was tried at the same time, and the G.W.R. "Castle" class 4~6-0 No. 5000, Launceston Castle, in in the previous year, all made between Crewe and Carlisle, were as follow :-

Engine No. 5908
No. 5917
No. 1093
No. 1073
Average Tare of Train (tons) 316 316 327 298 430
Coal per D.B-H.P.-Hr. (lb.) 3.82 4.86 4.04 3.78 3.78
Lb. water evaporated per lb. coal 8.08 6.96 7.06 8.62 6.80

"From these figures," Mr. Stuart adds, "readers can draw their own conclusions. One reminiscence of the 5908/5917 trials might be mentioned. On the Crewe-Euston section, for the purpose of taking certain indicator diagrams the regulator of the ordinary 'Claughtoil ' had to be open full. The results to the eyes and ears may be imagined when it is mentioned that the smoke box vacuum rose to some 13 inches! " Mr. C. Williams, of Crewe, points out that No. 5908 was the only engine of this class to be fitted with the Caprotti valve gear prior to being converted from Class 5 to Class 5X.
Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Wealth of information: Plate 499: No 2222 Sir Gilbert Claughton. at Manchester (London Road) in January 1913 while on trial in shop grey before returning to the works for final painting and fitting of nameplates. Plate 500: Official view of No. 2222 Sir Gilbert Claughton on 27th February 1913. The lamp sockets have already been replaced by lamp irons front and rear (the one at the top of the smokebox has been accidentally painted out in the darkroom along with the background!) and there is now a cover over the reversing gear. Sandpipe windshields were fitted soon afterwards and appeared on the rest of the first batch and on later engines too. page 296: Figure 133: weight diagram of No. 2222; Plate 501 No. 2222 at Preston in 1913; page 297: Figure 134: general arrangement drawing of No. 2222 (elevation and plan); page 298: Plate 502: No. 1191 Sir Frank Ree at Crewe North shed on 7 August 1913; page 299: Plate 503: No. 163 Holland Hibbert at Crewe c1914; Plate 504: No. 1161 Sir Robert Turnbull at Crewe in 1914; Plate 505: No. 2420 Ingestre at Crewe on 17 March 1923: page 300: Plate 506: No. 1216 at Camden on 25 June 1922; Plate 507: No. 1407 L/Corpl J.A. Christie V.C. at Edge Hill on 3 April 1922 (had been named in February 1922); Plate 508 previous locomotive at Camden on 3 May 1924; page 301: Plate 509: No. 1914 Patriot at Crewe Works on 22 May 1920 showing vermillion background to nameplates, Ross pop safety valves (also had a continuously sloping grate): Figure 135: 4mm scale drawing of 2499 Patience: page 302: Plate 510: No. 2222 Sir Gilbert Claughton fitted for oil-firing at Crewe North shed in 1921; page 511: Plate 208 No. 211 fitted for oil-firing on 21 August 1926 (38 converted in 1926).
van Riemsdijk, J.T. Claughton comments. Backtrack, 1999, 13, 680.
Kenneth Cantlie, a Bowen Cooke pupil, told writer about an incident observed on the footplate when boiler pressure was falling, but superheater pyrometer reading was high, and the driver closed the damper to the superheater and boiler pressure rose. Agrees with Johnson (original feature pages 347 and letters page 508 ) that the left hand inside cylinder was probably unable to produce full power.
Webber, A.F. The proportions of locomotive boilers. J. Instn Loco. Engrs, 1937, 27, 688-725. Disc.: 726-63. (Paper No.378).
Both the standard and large diameter boilers are considered on a comparative basis.
Yeadon, W.B. A compendium of L&NWR locomotives, 1912-1949. Part 1. Passenger tender engines. Challenger, 1995.
Young, Albert in P. Ransome-Wallis. Men of the footplate (1954).
Very loyal to his original company, but accepted later locomotives including Crabs and Compounds


George the Fifth:1910-: Bowen-Cooke: LMS 3P
Two identical inside-cylinder 4-4-0s emerged from Crewe in July 1910: 2663 George the Fifth (with superheater) and 2664 Queen Mary (without): this followed the successful running of a superheated Marsh I3 4-4-2T between Brighton and Rugby in 1909. Talbot (Illustrated history) calls the George the Fifth the "most successful engines of their day" in terms of power output in relation to size and cost but this statement is not collaborated in the Webber paper (below) which suggests that the Precedent class had a more efficient boiler. This test of superheating had been suggested to Bowen Cooke by the successful running of an LBSC superheated I3 class 4-4-2 tank between Brighton and Rugby, and by his wide study of current developments in the locomotive world both in Great Britain and abroad. Otherwise, in its details it was a typically North Western engine in the Webb-Whale tradition. There were four sandboxes: the ones for the leading driving wheels are behind the footsteps, shaped like those on the 'Jumbos', and filled from above the running plate; and the rear ones are between the driving wheels inside the splashers and filled from above them. The tender is the first Bowen Cooke type, which outwardly seems little different from the Whale tender but has, in fact, been completely redesigned. It has the same 'curved top' springs and only one coal rail but the separate toolboxes have gone, being replaced by cupboards on the front of the tender. Nock makes it abundantly clear that the class represented the zenith of the inside-cylinder 4-4-0 type and that performance, especially over Shap, could be of the very highest order: indeed Nock claimed that on a weight per power output basis the performance was never exceeded.

Nine more 'Queen Marys' appeared in October 1910, as George the Fifth was the experimental engine. However, in November, more 'Georges' began to appear and in 1913-14 the 'Queen Marys' were converted to superheating, presumably at their first major overhaul, and so became identical to the 'Georges' in every respect. Careful tests over several months showed that the superheated engine used some 26 per cent less coal. Even so, the 'Queen Marys' were fine engines, being superior to the 'Precursors' . Burning of the firebox ends of the superheaters sometimes occurred and No. 2495 was fitted with external rodding to an extra regulator in the dome and a Webb-type bible valve was fitted at the smokebox end with the chimney moved forward (see Plate 472). No. 1472 was fitted with a Weir feedwater heater (Plates 473/4) (the equipment had originally been applied to No. 127 Snake and then to No. 2494 Perseus);. The second of the 'Queen Marys', No. 238 F. W. Webb, The third of the class was equally well named, George Whale

London & North Western Ry. Loco. Rly Carr. Wagon Rev., 1910, 16, 157.
No. 2664 Queen Mary: non-superheated

New express locomotives, L&NWR. Loco. Rly Carr. Wagon Rev., 1910, 16, 213-14. illus.
No. .2663 George the Fifth and No. 2664 Queen Mary:

A contrast in L. & N.W.R. locomotives. Locomotive Mag., 1911, 17, 120. illustration
Crewe had built a replica of the Rocket and contrasted it with George the Fifth No. 2155 W.C. Brocklebank

Individual locomotives

No. 5000 Coronation
This was regarded as a "special locomotive" and survived the introduction of the later Stanier Coronation, but was withdrawn during WW2 (but even then was subject to special photographs. Talbot shows the locomotive in its original splendour. Plate 464 (p. 276) reproduces the official photograph with the Royal Coat of Arms on the tender; Plate 463 shows it as No. 1800 (but without name) running trials in June 1911 and Plate 465 (p. 277) shows it in service at Manchester London Road. Nock's Precursor family has a melancholy couple of pictures on pages 138 and 143 of 25438 Coronation being dismantled in June 1940. Its introduction into service as the 5000th locomotive to be built at Crewe is recorded in Locomotive Mag., 1911, 17, 119.

Paintings etc
Le Fleming, Hugh M. International locomotives. Plate 39
Painting of No. 1360 Fire Queen

3P "George the Fifth":1925-: Hughes:
Under Hughes' direction the Whale (1904) "Precursor" and Bowen Cooke (1910) George the Fifth classes were rebuilt with Belpaire boilers.

REBOILERING L.M.S.R.locomotives. Rly Mag., 1925, 56, 220. illus.

Retrospective and critical
Allen, Cecil J.
Half-a-century of train travel—No. 14. The "Georges" and "Princes" of the London & North Western. Rly Wld, 1963, 24, 136
Records of both fast running including famous run behind No. 1595 Wild Duck which had stalled on Camden Incline recovered all time lost by Rugeley, and heavy haulage during WW1.
Atkins, P. West coast 4-6-0s at work. 1981.
Although this is mainly about LNWR 4-6-0s, the author in his chapters on the Prince of Wales and Claughton classes makes it very clear that the George the Fifth class achieved an everyday performance comparable with both
Bulleid, H.A.V. Master builders of steam. 1963. p. 141.
On the Works side there was much to impress: Crewe boiler-making came as a surprise to Stanier, who had not fully appreciated the stalwart performances put up by the standard range of L.N.W.R. boilers, on such classes as the "Precursors" and "Experiments." These boilers were decidedly on the small side for the engine duties, and the L.N.W.R. footplate teams were expected to, and did, thrash them to keep time with heavy trains. They made a rather invigorating noise, and at night looked highly pyrotechnic. But these boilers did not require excessive maintenance and had long lives; and when it came to steam-raising capacity a " Precursor" class boiler topped 20,000 lb./hr. on test. Stanier obtained some of his boiler data at first hand, and was seen followmg foreman Tizard through the fire hole of a "Prince of Wales" class boiler in Crewe boiler shop: as a result he confirmed his opinion that the better circulation in a Great Western type of boiler was a great advantage. But incidentally he found Crewe boiler-making of Swindon quality but cheaper,
K. Cantlie (pp. 738-42) Discussion on Webber, A.F. The proportions of locomotive boilers. J. Instn Loco. Engrs, 1937, 27, 738 et seq
Submitted a graph (Figure 9) which compared the boiler performance at speeds from 0 to 70 mile/h of the following locomotives: "his" 4-8-4 for the Chinese National Railways; the Lord Nelson type with new boiler; and same type with original boiler; the enlarged Claughton/Patriot boiler; the original Claughton boiler; George V; Precursor, Teutonic; Greater Britain and Jumbo (comparitive performance was in descending order as listed) .
John F. Clay and J.N.C. Law. The supremacy of the Premier Line. Studies in Locomotive Performance—No. 6. Rly Wld, 1975, 36, 182-7.
Primary aim was to summarise the excellent performance which the superheated George the Fifth class achieved on the LNWR: "The question remains how it was possible for so relatively small an engine to produce such horsepowers with such regularity. This may partly be answered by a consideration of the boiler of the George. This was bigger in some essential dimensions than it at first appears. Although the grate area was a modest 22.4sq ft, the firebox heating surface was 161.3sq ft which compares with 155sq ft of the Swindon No 1 standard boiler used on the Saints and Stars. The deep fire in the large firebox had considerable properties of heat transfer and steam raising but it needed a high grade of coal to give of its best. This is another reason for the decline in LMS days when coal was of a more variable quality. The valve and front end design allowed the steam to be used with reasonable efficiency while the relative lightness of the engine and tender and their lower tractive resistance allowed a high proportion of the total ihp to reach the drawbar. All these factors combined to make the Georges very effective motive power units in the favourable operating conditions which were normal before World War 1.
We may reject the legends-the Georges did not run 400-ton trains on 60mph start-to-stop runs on 33lb of coal per mile nor was it true, as some Midland supporters would have us believe, that it was possible to run a Midland compound on the coal a George threw up its chimney. In actual fact, taking a George and a Compound both in its optimum test condition, there was very little difference in the basic fuel consumption related to the work performed but a new Compound in the mid-I920s was certainly more economical than a run-down George. The maximum ihp outputs of the two types when new were similar except in that each gave of its best in a different speed range, with the George being predominant in the 60-70mph band and the Compound some 20mph lower. Both classes deteriorated considerably as they declined in condition. For the economic circumstances in which they were designed the Georges were admirably suitable, their ease of construction and relatively low capital cost enabling the LNWR to keep an increasing traffic moving while high dividends were earned. It was however a policy only justified within narrowly restricted limits of time and circumstance, while Churchward's policy of building engines with an ample reserve of power proved ultimately to be wiser. It would appear that Bowen Cooke himself realised this, for, when the traffic was moving satisfactorily behind the Georges, he developed the Claughton class 4-6-0 which was comparable with a GWR Star in size and potential power. Unfortunately, although these engines matched or even exceeded the GWR 4-6-0s in maximum power output under test conditions, they lacked the consistency and economy of the Swindon machines. The LNWR may not always have deserved its title of "The Premier Line" but thanks to the hard work performed on footplate and in running sheds and to some good design features, especially in the Georges, it did so during the years 1910-1916 just as it deserves it today under electrification."

Ellis, C.H. Some classic locomotives. 1949.
Perhaps the painting of Beagle (opposite page 103) on a Manchester to North Wales express with mountainous exhaust dominating the moutainous scenery says it all as the text (other than a brief technical description and some discussion on the absurd or rich names bestowed upon the locomotives) is limited mainly to an enrichment of a run recorded by C.J. Allen: "The Wild Duck, with Driver Greenhalgh, was responsible for a spectacular run timed by Cecil J. Allen in 1912. The train was the 10.50 a.m. out of Euston, weighing 410 tons full, in bad weather. The engine stuck on the greasy rail going up Camden bank and had to be assisted in rear (not a regular practice from Euston in those days). Willesden was passed barely eleven minutes late, at 55½ m.p.h., and the average speed thence to milepost 156, after which the train was checked by Crewe signals, was 62.1 m.p.h. The Rugby and Stafford slacks were strictly observed. Minimum speeds of 56¼, 58¼ and 56¼ m.p.h. were recorded at the top of the climbs to Tring, Roade and Kilsby Tunnel respectively, and pass-to-pass average, Tring to Castlethorpe (25.1 miles) was 72.5 m.p.h. Wild Duck's coal consumption on this run worked out at only 33 lb. per mile, and about 40 lb. was a fair average for these engines in their youth. It was customary to drive them with a fairly early cut off and the regulator wide open—hence the shouting exhaust which many immediately associate with the London and North Western of those days. Wild Duck's run was more fully described by its original recorder, with a log," in Railway Magazine,1935,  77, (457),  1-10. (originally published in 1912)(bold type added KPJ).
Harvey, Norman . Locomotive causerie: encounters with pre-grouping locomotives. –  3. Railway Wld., 1960, 21, 99-102
George V No. 5575 Partridge recorded on 17.50 Euston to Coventry and compared with a slower journey behind a compound No. 1053 in the late 1920s
Nock, O.S. The L.N.W.R. Precursor family : the Precursors, Experiments, Georges, Princes of the London & North Western Railway. 1966.
This is certainly one of the author's best books. It vividly captures Nock's excitement at the performance of the George the Fifth 4-4-0s: on page 91 he wrote that the class has a "special place in the eventful history of express train operation over Shap. In relation to the total engine weight, there have never been more competent locomotives on the line... There were times, indeed, when the virtually impossible was attempted.".
Riemsdijk, J.T. van. Compound locomotives: an International survey. 1994. 140pp.
Written mainly about Midland compounds, but added that the maximum sustained power outputs achieved by the Midland Compounds, Southern Schools, LNWR George V and LNER Shire (D49) 4-4-0s were all much the same, but the Midland and LNER types were rarely required to produce such outputs.
Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Plate 457; This is the official view of George the Fifth on 26th August 1910, after the engine had returned to Crewe Works for final painting, following the completion of trials. The only outward signs of the superheater are the damper lever on the side of the smoke box, which was operated by the boiler handrail, and the absence of a displacement lubricator on the side of the smokebox, instead of which there are three oil pipes to the left-hand cylinder froin the mechanical lubricator on the other side; >Plate 458 (right): Smoke box view of George the Fifth, showing the superheater header; Plate 459: OfIicial view of No. 2664 Queen Mary, taken on 26 August 1910. The only outward signs that it is an unsuperheated engine are the traditional displacement lubricator on the side of the smokebox and the absence of a superheater damper lever and mechanical lubricator. Plate 460: The second of the 'Queen Marys', No. 238 F. W. Webb, at Stockport in original unsuperheated condition but with sand pipe windshields and a Whale tender fitted with straighttopped springs, introduced by Bowen Cooke in 1911. Plate 461 (below): 'Queen Mary' No. 2507 Miles Macinnes in Crewe Works about 1912, with the original type of Bowen Cooke tender introduced on this class. It has short-tapered Whale buffers mounted on wooden pads, presumably to give the same length as the long-tapered type so as to match the length of the couplmg; page 276: Plate 462 No. 82 Charles Dickens at Manchester London Road; page 277: Plate 466 No. 2664 Queen Mary at Manchester London Road c1913; p. 278: Plate 467 superheated No. 1550 Westminster c1916; Plate 468 No. 1680 Loyalty at Manchester London Road c1915; Plate 469: No. 1481 Typhon at Rhyl c1917; page 279 No. 2106 Holyhead in plain black outside Crewe Works in July 1915; Figure 126 drawing of No. 1371 Quail: p. 280: Plate 471 No. 2663 George the Fifth at Willesden on 23 June 1922; Plate No. 472 No. 2495 Bassethound at Manchester London Road c1915; Plates 473/4: No. 1472 Moor Hen was fitted with a Weir feedwater heater (both sides of locomotive shown)
Webber, A.F. The proportions of locomotive boilers. J. Instn Loco. Engrs, 1937, 27, 688-726. (Paper 378).
An analysis of George V boiler on a comparative basis.
Williams, T. Lovatt. Some reminiscences of the footplate — II. Rly Mag., 1943, 89, 269-73.
Old Fred T—, of Crewe, was one of the surly brand of drivers and always rather aggressive in his outlook. One morning his engine, a "George the Fifth," was standing at the head of a train on No. 15 platform, Euston, and Fred was leaning over the side of the cab, surveying the scene on the platform with a cynical eye. A few minutes before starting time a rather pompous Midland official walked up Platform 15 and counted the coaches as he went along. Thirteen coaches, two "diners," —loading equal to "23½." A bit staggered by this he walked up to Fred and said, "Good morning, driver, where's your pilot engine?" Fred looked him up and' down, removed the dreadful old pipe from under his straggling moustache and remarked, "In the shed, where she ought to be. Who do you think we are, the b— Midland?" "No, but I am," replied the official, and walked off huffilv.
There was no doubt, however, that the constant heavy loading and overloading of trains was a source of grievance to the train crews. We young men were fed with ideas that the locomotive rnnning department was under the tyranny of the traffic department, who just did exactly as they pleased. Drivers frequently cited with envy the position of the G.W.R., whose locomotive department was in such a strong strategical position that the drivers just would not accept anything above the agreed loading. Whether this was actually the case I do not know, but the general opinion was that our department lacked "guts." Under the heavy duties imposed on them (as has been previously mentioned) locomotives became extremely rough, and sometimes drivers made vigorous protests about their "riding" qualities. One of the most succinct. of these reports was made by a very temperamental driver belonging to Crewe North Shed who made the statement that "riding. on this engine I can only .compare to sitting on a galvanic battery with the noise of a boiler shop thrown in." It speaks well, however, for the sturdiness of the design and the steaming capacity of the boilers that tile four-coupled types, the " Precursors" and the" George the Fifths," stood up to their duties as well as they did.

0-8-2T: 1911- [1185 class]
In 1911, Bowen Cooke introduced an eight-coupled tank engine for heavy shunting and thirty were built in 1911-17. The design was basically a tank version of the G class with a rear pony truck; but to make starting easier, the boiler pressure was increased by 5psi and the valve gear was modified slightly. Some accounts state that the firebox was recessed 6ft. into the barrel, making a combustion chamber. The tube heating surface was less than the G class, possibly due to a few tubes being omitted, but the firebox heating surface was the same, and quite possibly normal G class boilers were used as replacements later. Lever reverser, was fitted, preferable in principle to the screw type for shunting engines; these were the first LNWR engines to be so fitted since Ramsbottom invented his screw reverser. As on the 0-8-0s, the third pair of driving wheels were flangeless and there were three coupling rods, the centre one fitting outside the two outer ones, as on the G and G1 classes. According to Nock most worked in South Lancashire..

The 0-8-2 tank engine. Loco. Mag., 1941, 47, 131.
Calls them the 1185 class: originally described in Loco. Mag., 1912, 18, 49.

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Page 236: Plate 390: official photograph, taken in January 1912, of No. 289, the fourth to be built. It has a three-link coupling for handling ordinary goods wagons, vacuum brake for use in emergency on passenger trains and Cooke buffers. Plate 391: No. 1663, one of the first batch completed in 1912, seen here about 1920 with a capuchon on the chimney, probably from an 0-8-0. The standard long-tapered buffers fitted originally seem to have proved unsuitable, probably because of the throw-over on sharp curves in goods yards, and to have been quickly replaced by the standard large Webb buffers with 18in. instead of 13in. heads, as used on 2-4-2 tanks for motor train working and so available from stock; page 238: Plate 392: rear view of No. 1665 at Springs Branch; Plate 393: No. 1592 at Willesden on 3 March 1922; Plate 394 No. 1514: page 239: Plate 395: No. 736 at Willesden on 10 July 1921; Plate 396: LMS 7892.
Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002. 268pp. 288 plates. 36 diagrs.
Page 202: Fig. 34: general arrangement drawing

Propsal to fit booster in 1931. Locomotive Mag., 1931, 37, 358.
"We understand it is proposed" to fit one of the 0-8-2 tank engines with a booster operating on the trailing truck.

4-6-2T Superheater Tank: 1911-
Following the George the Fifth class Bowen Cooke produced the 4-6-2 tanks. They might appear to have been a tank version of the Prince of Wales class, but they preceded them by some months. Initially, some had superheaters and others did not but eventually all were superheated and came to be known as the 5ft. 6in. Superheater Tanks. As well as being very useful suburban tank engines, they assisted over Shap, both banking and double-heading, and worked on the Central Wales line where Precursor Tanks, with their higher axle loading, were prohibited. Other details followed parctice at that time: cab-roof ventilator, 'Jumbo' style leading sandbox and no capuchon. Water pick-up apparatus was fitted. The 'Superheater Tanks' had Belpaire fireboxes, the only entire class to do so before the 'Claughtons', probably due to the flanging blocks being available, from the two Belpaire boilers made for the Jubilees Polyphemus and Ramillies in 1904, the boiler diameter being the same and the length immaterial. The grate area of 23.6 ft2 was similar to that of the 0-8-0s. The coupled wheels were 5ft 8½ in in diameter and the maximum axle load was only 16½ tons..

Atkins, Philip. Crewe at home and away. BackTrack, 1989, 3, 11-80.
Shows diagram of proposed 2-6-4T of October 1909: Bowen Cooke's justification for this proposed new 2-6-4T is noted in a report of October 1909 and recorded desirable features gained from other railways
1. Reversing gear held by a clutch to keep the shafting steady (GN)
2. Extended smokebox (GN)
3. Sloping firebox (Caledonian)
4. Well at bottom of feed tank (Caledonian)
5. Sloping tank and coal bunker (Caledonian)
6. Some improvements in the general arrangements of the footplate (Brighton)
Atkins also notes that fifty replacement boilers were constructed between 1928 and 1934, but that the wasteful "Stanier regime" oversaw the withdrawal of the class beteen 1936 and 1939.

Nock, O.S. LNWR locomotives of C.J. Bowen Cooke. Truro: Bradford Barton, 1977. 112pp.
Page 39 log of a non-stop run from Shrewsbury to Crewe with substantial load of 425 tons with good hill-climbing and an eventual maximum speed of near 70 mile/h.
Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Page 290: Plate 489: official photograph of first of the class, No. 2665 on 10 January 1911. The lubricator on the smokebox indicates that it was not superheated. The large letters (L & N W R) on the tank were not carried in service but were applied experimentally when the size of letters was being chosen: Plate 490: official photograph of one of the first batch of 4-6-2 tanks, No. 2670, taken on 11 March 1911. Except for the details associated with superheating, the engine is identical to No. 2665. Again, the large letters (no amprisand) were never displayed in traffic; Page 291: Plates 491 and 492 show front and rear views of this locomotive; Plate 493 No. 1006 at Stockport c1915: page 292: Figure 131: wight diagram; Plate 494 No. 932 at Carnforth in 1921: page 293: Figure 132: drawing; Plate 495: No. 1366 in London Area c1916: page 294: Plate 496: No. 2273 with air vents from tanks: Plate 497 No. 346 (lettered "LMS" in "LNWR" style); Plate 498 No. 6996 (rear view in late 1930s?)



In 1919 the LNWR purchased from the government thirty 2-8-0s which had been ordered for service with the Railway Operating Division on the Western Front during WWI. All except one were built by the North British Locomotive Co. and in fact came to the LNWR as new engines, since the war ended before they could be sent to France. They were classified as 'MM', the name being derived from the Ministry of Munitions which had ordered them, and lead to the adoption of the nickname 'Military Marys' by LNWR enginemen. When the engines were first obtained, they were given numbers in the LNWR ordinary stock list but the purchase was held up and in September 1919 they were numbered in the 2800 series along with 151 other engines of the class on loan. In November 1920 the purchase was completed and they received fresh numbers in the ordinary stock list. The Westinghouse pump provided brake power on the engine and so was in constant use when the engine was working; it was not something for use on the Continent only.

Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Page 243: Plate 404:  No. 2824, which eventually became LNWR No. 2407, at Carlisle Upperby about 1920, perhaps in process of delivery to Crewe; page 244: Plate 404 No. 2100 in 1920; Plate 406: No. 2400 post November 1920; Plate 407: No. 2394 at Chester c1922: page 245: LNWR weight diagram..
Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002. 268pp. 288 plates. 36 diagrs.
Chapter 10 (pp. 221-7): The complex story of loans, purchases and acquisitions from other railways (21 from LYR, 6 from SECR and two lots of two from LSWR is told, some on passage to the dump at Gretna is told. In 1927 75 were obtained by the LMS as a cheap source of tenders. The Hillhouse men (p. 227) liked the MM class. In general the right hand drive was unpopular, an experience shared by the crews on the Nord and Est Sections in France.

Beames' 0-8-4 Tank;

Beames' 0-8-4 Tank Plate 397: The final design of LNWR eight-coupled goods engine was H. P. M. Beames' 0-8-4 tank, which appeared in the first few months of the LMS period. Whereas Bowen Cooke's 0-8-2 tanks had been designed purely for shunting, the 0-8-4 tanks were intended for the haulage of heavy goods trains over short distances, which is why greater coal capacity was provided, and they spent most of their lives in the steeply graded South Wales colliery area. They were essentially a tank version of the 'G2' with higher boiler pressure, 185psi instead of 175 psi. This is the official view of No. 380, taken on completion on 27th April 1923 and showing the engine in full lined LNWR livery but with 'LMS' on the tank sides. As on the 'G2' class 0-8-0s, the centre coupling rod was pin-jointed to the outer rods, and the buffers are parallel ones of the type which had replaced the original Whale buffers on the 0-8-2 tanks. The engine also has vacuum brakes and steam-heating apparatus for working passenger trains, a chimney without a capuchon, Webb safety valves, wire mesh on the rear windows as protection when coaling, a mechanical lubricator between the sandboxes, snifting valves at the sides of the base of the smokebox, and LMS lamp irons. The bufferbeam ends are shaped for greater clearance on platform edges, and the third pair of driving wheels are ftangeless. Reversing could be effected by either screw or lever.

LOCOMOTIVE developments on the London, Midland and Scottish Railway. Rly Mag., 1923, 53, 468-9. 2 illus.
LONDON, Mtdland and Scottish Railway—new eight-coupled tank engine. Engineer, 1923, 136, 476. illus.,diagr. (s.el.)
NEW 0-8-4 tank locomotives, London, Midland & Scottish Ry. Loco. Rly Carr. Wagon Rev., 1923, 29, 287-8. illus., diagr. (s. el.)
Erratum p.323.
0-8-4 tank locomotive; London, Midland and Scottish Railway. Engineering, 1923, 116, 570. itlus.,diagr. (s.& f.els.)

Retrospective and critical

Dunn, J.M. Memories of the L.N.W.R. 0-8-4 tank engines. Rly Mag., 1953, 99, 235-8. 2 illus.
Dunn was a fitter in the South Wales District in the 1920's and recalls the difficulty in maintaining the class.
Gregory, J.H. The 0-8-4T's of the L.N.W.R.. Rly Obsr, 1947, 17, 54-5. table.
Hadley, R. The LNWR 0-8-4T Engines. LMS Journal, (1) 48-52.
Using text, table, weight diagram and photographs, the author describes this LNWR design that did not enter service until after the Grouping. Informative letter from Philip Atkins (3) 15 of cancellation of order for further 0-8-2Ts and announcement in Locomotive of 0-8-4Ts, also diagrams of Fowler version of 0-8-4T with condensing gear (presumably for Liverpool Riverside KPJ) and boosetr-fitted 0-8-2T. See letter (5-75) from Philip Griffiths on nature of track, especially colliery sidings in Sirhowy Valley off RR.
Highet, C. The L.N.W.R. 0-8-4 tank engines. Rly Mag., 1953, 99, 424-5.
Similar experiences to Dunn's, but at Edge Hill, Liverpool.
L.N.W.R . 0-8-4 tank engines. J. Stephenson Loco. Soc., 1952, 28, 58; 68. 2 illus., table.
Parker, L.T. The Lickey banker. J. Stephenson Loco. Soc., 1955, 31, 137-8; 151-60; 285. 2 illus.
Includes notes on the trial of an 0-8-4T as a banking engine.
Talbot, Edward.  An illustrated history of LNWR engines. 1985.
Plate 398: Beames 0-8-4 tank No. 468, built in July 1923, photographed at Shrewsbury, possibly while on trial before returning to Crewe Works for final painting. The ventilator in the cab roofis raised to admit more air. Compared with the view of No. 380, Ross pop safety valves have replaced the Webb type used at Crewe since 1874, and there is no mechanical lubricator on the right-hand side.Plate 403: LMS No. 7938, originally LNWR No. 1908, pictured as running about 1940 with LMS buffers and with carriage-heating hose, snifting valve and front numberplate removed, but otherwise much as originally built.
Talbot, Edward. The London & North Western Railway eight-coupled goods engines. Gnosall: Edward Talbot, 2002. 268pp. 288 plates. 36 diagrs.
Pp. 209-20: Fig. 35 (p. 210) radial truck and Fig. 36 (p. 216) combined screw and lever reverse. Noted trials with load of 209 tons locomotive started on 1 in 34 gradient in bad weather. On p. 214 Talbot noted that they were powerful engines, but that the cabs were cramped as the tanks backed into the cabs. The trailing driving axles suffered frrom poor lubrication. On page 220 it is claimmed that the exhaust note was even (unlike the 0-8-0 types)..
Whitcombe, D.H. Monmouthshire memories. Rly Wld, 1961, 22, 144-6. 3 illus.
Difficulties experienced in operating the class.

Other locomotives

Loco Mag.. Locomotive Mag., 1907, 13, 36
The LNWR locomotive Dwarf was built by George England: it never carried a number. It was used by the engineering department as an inspection engine at Manchester. It was at Crewe for a long time before being broken up. Shannon was owned by the Sandy & Potton Railway and when taken over by the LNWR was numbered 1863.

Detail: brakes/injectors, etc.

Williams, T. Lovatt. Some reminiscences of the footplate — II. Rly Mag., 1943, 89, 269-73.
LNWR locomotives were strictly individual in their designs and fitments. In some cases this was an advantage, sometimes it was most decidedly not. The phrase "strictly individual" is used to denote the fact that articles of proprietary manufacture were not encouraged. For instance, the injectors were of LNWR design and made at Crewe, and my observation of their functioning led me to think that they were reasonably good. The fact that they were situated beneath the footplate was an important point, for in this position they did not tend to get hot and so give trouble in starting. Another point in their favour was the ability to blow steam back into the tender, which is useful in helping to reduce blowing off through the safety valves when standing in a station; this had the double advantage of  also transferring what would be wasted heat into the feed water.
On the other hand, I could never raise much. enthusiasm about the vacuum brake. Situated on the top of the firebox front, with its steam-ejector operating valve on one side and the air-admission valve on the other, and connected together by a link, the brake was certainly not in the easiest position for operation. If the semi-rotary air-admission valve tended to be stiff (and this was by no means infrequent) it. was really painful to apply force with one's arm extended at the angle of a Nazi salute. Imagine the effect of a couple of hours' shunting, for example, and remember that the shorter the stature of the driver, the worse it became.
In addition, the steam valves for the injectors were located inconveniently high (though by reason of the design of steam distributor there was no alternative position) and the blower was placed so remote from the driver that in case of an adverse signal he often had to shout to the fireman to operate it, and so prevent a sudden blowback through the firehole. In our many discussions on fobtplate arrangements these points were frequently commented on and we looked with undisguised envy at the cabs of certain locomotives of other railway companies whose drivers had the regulator extension immediately on the right hand, at waist level, and the vacuum brake operating handle placed so that it could be used from a sitting position, and with scarcely any effort.
Even our whistle lever was sited so that one was forced to stretch up to reach it, but this was countered by one of our funny men, Billy T—, who invariably tied a great length of string to the lever of any engine he chanced to drive. -In his more clownish moments Billy would sit facing the tender and pull on his string at intervals until his fireman would get thoroughly annoyed, and shout: "Look where you're going, you fool. " It sometimes seemed as though the cabs were designed to have the least possible comfort, and to complete the picture one has only to recall the hard wood contraptions which went by the name of "seats." The fact that not the slightest effort was made to shape them to the contour of the body or even to provide a simple cushion of felt, induces one to compare the conditions to those which were said to exist in the chapel of a monastery—the object being to keep the members awake! It must be said in all fairness, however, that the various fittings functioned consistently well. Occasionally there was trouble with an injector failing to pick up, and on one rather thrilling occasion, nmning down Shap from Carlisle with a " Precursor," both the injectors refused to work. The driver temporarily handed over his job to me, and he and the fireman set to work frantically to save the situation. The speed increased"to 75 m.p.h., the water was out of sight in the gauge, and the temporary driver was just giving a slight brake application when the right-hand injector came on and all was well again. Such moments are indeed anxious ones in an already arduous day.
Some of the runs from Leeds to Liverpool were pretty rough in the depth of the winter and a good example of footplate work in its grimmest form; in of her words, a great contrast to a run along the North Wales coast on a sunny summer morning. There was one extremely amusing character on this road—a driver whose first name was Alfred but whose surname I did not discover. I have.travelled with this man in some of the wildest winter weather, in snowstorms when the spectacles were completely useless, and Alfred's face with its bushy beard was almost continuously over the side. Apparently this sort of thing stimulated him in some curious way, for when there came a fair stretch of line without a signal he drew in his head, stood erect, and commenced to recite impressive passages of Longfellow at the top of his voice. I found out from his mate that he was a Longfellow fan, and the fireman complained bitterly that, not being poetically inclined himself, he often wished Alfred would "put a sock in it." Even to this day I can recall the incongruous scene most clearly. The swaying locomotive hurtling along on what was never a very smooth permanent way, the snow driving past fiercely, the fireman sitting on his seat with a disgusted expression on his face and the bearded Alfred declaiming snatches of poetry to a young fellow with a shovel in one hand and a strong tendency to burst into loud laughter.
The water pick-up gear was operated by a hand wheel on the tender front. It was an easy proposition to put this down, but sometimes a question of sheer brute strength to get it up before the tank overflowed. When this happened, coal was sometimes displaced and water poured over the front coaches of the train and into the compa.rtments if the passengers had the windows open. On double-headed trains, when passing over troughs, it was usual for the leading engine to take 'water first and then pull up the scoop to allow the second to have a turn. Once on the Carlisle run the fireman of our assisting engine in front apparently could not get the scoop up while passing over Brock troughs. The first indication we on the train engine had of this was a loud yell from our driver and the three of us crouched together well under the cab. A great torrent. of water ,was flung along the boiler and came streaming over the cab roof like a young Niagara. It was the only time that I had seen water projected so far in this way, and needless to say our train engine did not get a drink from these troughs on that particular run.
I could never. understand why some simple form of steam or vacuum-operated scoop could not be fitted, but possibly the designers shirked the extra complica.tion. In later years I had much experience with the power-operated reversing gear on the LTSR. 4-4-2 tank engines, and this seemed to be quite efficient 'and satisfactory; the moral of -the tale is that if reversing gear could be so easily mechanised, then it would not be difficult to do the same for the pick-up.
The other wheel on the tender of the L.N.W.R. engines operated the tender brake, and this was always rather a joke. Sometimes it operated. with good results and on I other occasions it did not. The reasons which contributed to this state of affairs I do not intend to go into here, but there I were times when it produced results which might have been serious, and on one famous occasion it; apparently gave rise to an.incident of the nature of a pantomime.' The witness of this rich episode was a fellow pupil who (so he said) happened to be on the spot at the time. The narrative runs that a "Precursor" dropped its motion on the slow line between Whitmore and Crewe. The gradient here is a consistently slight drop towards Crewe. The breakdown gang from that place was in attendance in the charge of one Bill D—, an amusing fellow who wore a small bowler hat and a pair of black bell-bottom trousers. Moreover, he had the best flow of language in Crewe, and that is saying a lot. When the remaining motion and rods of the "Precursor" had been removed, the breakdown engine attempted to couple up to the back of her tender for the purpose of drawing her back into a siding. Unfortunately, the "Precursor" was bumped a little too hard and she ran forward very slowly. Someone jumped on the footplate and put more pressure on the tender brake. Nothing happened. The "Precursor" continued her slow and stately way, gathering speed quietly. Bill was now at his best. He danced around on the permanent way, his bell-bottom trousers flapping about his legs, exhorting his gang ta stop her. He himself showed them how by throwing ballast on the rails in front of the progressing engine. The result was disappointingly negligible. The" Precursor" still headed for her stable at Crewe.
There was one last frenzied outburst, from Bill and he made a rush for the signal box, from where an appropriate message was sent on ahead to the effect that there was an engine running away and a road was to be prepared for it in the long sidings at Basford Hall. This was done and the outcome was a very little damage and a few wagons derailed. But I am informed that the mention of tender brakes to Bill D— was for ever afterwards a mistake—to put it mildly.
With some classes of locomotives of the former LNWR, although the train braking was quite efficient, the brake power of engine and tender running light was deplorable. There came a time when this was recognised as being a highly undesirable state of affairs and a series. of most interesting comparative tests were made between the LNWR, the G.W.R., and the Midland companies.
These tests were carried out several years after the date of these reminiscences, but the writer was fortunate enough to be present during that portion of them which entailed comparisons with the Midland Railway. Briefly, the tests consisted of running :-
(a) An L.N.W.R. engine with an L.N.W.R. train;
(b) An L.N.W.R. engine with a Midland train;
(c) A Midland engine with an L.N.W.R. train;
(d) A Midland engine with a Midland train.
A dynamometer coach was interposed between the engine and the train on each occadon, and the drawbar pull was recorded under various conditions of braking. It will thus be seen that as the result of weak engine braking there would be a surge forward resulting in a high drawbar pull reading. These trials took place between Nuneaton and Leicester and those involving the Great Western were run between Bletchlev and Oxford. The results were very useful to the locomotive design section at . Crewe and several improvements and modifications in the brake layout were incorporated at a later date.

Shed numbers
Shed numbers LNWR. Rly Mag., 1900, 7, 354
Numbers affixed to back of cab: full list: 1, Camden; 2, Willesden; 3, Bletchley; 4, Nuneaton; 5, Northampton; 6, Bescot; 7, Netherfield and Colwick; 8, Rugby; 9, Walsall; 10, Ashton (Birmingham); 11, — 12, Burton; 13, Bushbury; 14, Stafford; 15, Crewe; 16, Longsight; 17, Farnley and Wortley; 18, Birkenhead; 19, Chester; 20, Huddersfield; 21, Bangor; 22, Holyhead; 23, Warrington; 24, Peasley Cross; 25, Spring's Branch (Wigan); 26, Edge Hill (Liverpool); 27, Preston; 28, Tebay; 29, Carlisle; 30, Shrewsbury; 31, Abergavenny; 32, Workington; 33, Swansea (Victoria Station); 34, Patricroft; 35, Speke Junction; and 36, Ordsall Lane.