SteamIndex

Great Western locomotive types:
late Dean/Churchward and later
Saint class
Star class
Castle class
King
County (10XX)
111
47XX
43XX
90XX (32XX)
72XX
2-6-2T
56XX
57xx
City
steam railcars (rail motors)
See also index of loco classes
Steamindex home page
See also Churchward, Collett & Hawksworth

Although it is possible to identify designs which were purely Joseph Armstrong, or William Dean, or Churchward, or even Collett, it is difficult to isolate these in the way that one can refer to a Gresley, or to a Stanier design. On the GWR design policy tended to merge and there is the protracted Dean/Churchward period prior to Churchward taking full command. Collett remained deeply wedded to the past and the improbable Earls emerged after the Gresley and Stanier streamlined Pacifics. One cannot imagine that Churchward would have approved of these anachronisms, or the 48XX 0-4-2Ts (the Churchward steam railcars were far more advanced with their outside valve gear).

Furthermore, the Great Western differed from the other post-1923 railways in that the grouping was achieved by a process of logical absorption, rather than by amalgamation. Thus, a great continuity was preserved, especially as Churchward had established a policy of standardization and this was maintained until 1947. For this reason, the locomotives will be considered as one unit, whether built under Churchward, Collett or Hawksworth. Dean will be partly considered in this section

A great wealth of published material exists about the railway and its locomotives. MacDermot's history is one of the major works of reference on railways, but it does not extend beyond 1921 nor is locomotive development considered in depth. C.R. Clinker revised MacDermot's work, which was published by Ian Allan in 1964. In the preface he advises readers to refer to the RCTS The locomotives of the Great Western Railway for further information on mechanical development. It should be noted that the revised edition does not credit A.C.W. Lowe for the section on locomotive and carriage development. O.S. Nock wrote a sequel to Clinker's revision, which surveys the post-1923 period. This third volume does not perpetuate MacDermot/Clinker's style and locomotive development receives greater emphasis, unfortunately at the expense of less detail about commercial activities.

General studies

Allen, C.J. Great Western. London, Ian Allan, [1962?] . 72 p. 41 illus.
Concise history.
Beale, Gerry, GWR locomotives between 1900 and 1910: Chapter V in John Norris et al's Edwardian enterprise
Fortunately, the author does not keep exactly to the stated period, but extends his survey to the retirement of Churchward. Thus the 47XX class is included. There some excellent photographs excellently reproduced. Some of the most interesting depict the steam railmotors and locomotives adapted for auto-train working complete with ersatz bodywork
Behrend, G. Gone with regret. Sidcup (Kent), Lambarde Press, 1964. 193 p. + 48 plates, 83 illus., 10 tables, maps on end papers. Bibliog.
A collection of personal reminiscences.
Brewer, F.W. Modern locomotive practice on the Great Western Railway. Rly Mag., 1928, 62, 104-13; 194-8; 283-9; 388-97; 450-4: 63, 27-33; 105-8; 204-12; 294-9; 351-5; 452-62: 1929, 64, 107-16; 197-202; 341-9; 439-49; 65, 33-42; 350-6; 1930, 66, 41-9; 425-31. 103 illus., 40 tables.
A very detailed survey of development and performance from the time when Churchward's influence began to permeate Dean's work.
Brewer, F.W. Modern locomotive superheating on the Great Western Railway: its application to tank engines. Locomotive Mag., 1927, 33, 320-2.
Casserley, H.C. and Asher, L.L. Locomotives of British Railways, Great Western group; a pictorial record. London, Andrew Dakers, [19 ].114 p. 164 iIIus.
Casserley, H.C. and Johnston, S.W. Locomotives at the grouping. [v. 4]. Great Western Railway. London, Ian Allan, 1966. 144 p.+ 24 plates. 95 illus., 268 tables.
Chapman, W.G. Loco's of 'The Royal Road'. London, G.W.R., 1936. 232pp
Started in 1921/2 and grew: see also Ottley 6150 and Locomotive Mag., 1922, 28,, 350 when publication was authored by "A.J.L.W."
Clements, Jeremy. The GWR exposed: Swindon in the days of Collett and Hawksworth. OPC, 2015.192pp.
This is an important book, asnd certainly justifies altering many of the entries on this webpage
Clements, Jeremy. William Dean: the greatest of them all. Southampton:. Noodle Books. 228pp.
Reviewed J. Rly Canal Hist. Soc.,2013 (217) 53
Cook, K.J. G.W.R.locomotives, 1901-1951. Rly Mag., 1952, 98, 157-61. 4 illus., 2 tables.
The policy of standardization.
Cook, K.J. The late G.J. Churchward's locomotive development on the Great Western Railway. J. Instn Loco. Engrs, 1950, 40, 131-71. Disc.: 171-210. (Paper No. 492)
This is the most complete professional assessment of the core of Great Western locomotive development
G.W.R. locomotives, 1837-1935. Rly Gaz., 1935, 63, (Aug. 30) G.W.R. Centenary Supplement, 34-9. 12 illus., table.
[G.W.R. locomotives-stock total as at 31 st December, 1922. (including absorbed stock)]. Loco. Rly Carr. Wagon Rev., 1923, 29, 188. table.
Great Western Railway: Swindon lots - standard gauge. J. Stephenson Loco. Soc., 1946, 22, 128-9. table.
Chronological table.
Haresnape, Brian. Churchward locomotives. a pictorial history. Ian Allan, 1976.
Haresnape, Brian. Collett & Hawksworth locomotives. a pictorial history. Ian Allan, 1978. 128pp.
Holcroft, H. An outline of Great Western locomotive practice, 1837-1947. London, Locomotive Publishing Co., 1957. viii, 168 p. incl. 32 plates + col. front. (133 illus. incl. 37 line drawings: s. els.), 21 diagrs., 2 tables.
The author was trained at Swindon.
Knox, C. The un-beaten track. London, Cassell, 1944. 199 p. + front. + 6 plates. 8 illus.
A history of 1939-45 War activity.
Lowe, A.C.W. The Locomotive and Carriage Department [in :] MacDermot, E.T. History of the Great Western Railway. v. 2.1863-1921. London, G.W.R., 1931. xii, 654 p. + front. + 2 folding plates. 162 illus. (inc1. 24 ports.), 13 diagrs., 20 tables, 3 plans, 7 maps. Bibliog. (footnote references).
Pp.509-93 : This well assembled official record was superseded by the RCTS history.
Mountford, E.R. and Davies, F.K. The G.W.R. and the Railways Act, 1921. Rly Obsr, 1941, 13, 104-7.
Mainly its effect on numbering.
Nock, O.S. Fifty years of Western express running. Bristol, Edward Everard, 1954. xii, 353 p. + col. front + 60 plates, (incl. 7 folding & 7 col.) 89 illus., 15 diagrs. (incl. 12 s. els.), 144 tables.
A compendium of locomotive performance: this work receives detailed scrutiny elsewhere.
Nock, O.S. Great locomotives of the GWR. Wellingborough: Patrick Stevens, 1990. 231pp.
Nock, O.S. The Great Western Railway: an appreciation. Cambridge, Heffer, 1951. xii, 185 p. + front + 32 plates. 64 illus. (incl. 7 line drawings: s. els.), diagr., 7 tables, 12 maps.
Nock, O.S. The Great Western Railway in the twentieth century. London: Ian Allan, 1964. 212 p. + col. front. + 32 plates. 73 illus. (incl. 4 ports.), 23 diagrs., 24 tables, 3 plans, map. Bibliog.
Nock, O.S. History of the Great Western Railway. v. 3. 1923-1947. London, Ian Allan, 1967. 268 p.
Railway Correspondence & Travel Socieety. The locomotives of the Great Western Railway. Part 1. Preliminary survey.
Although this was a tremendous achievement at the time it now sometimes seems sketchy in comparison with the huge survey of LNER locomotives, especially when it is remembered that this attempts to survey all GWR locomotives, including broad gauge and all absorbed locomotives. Nevertheless, it remains the definitive source. It should be available in electronic form.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives].
Smith, C. Centenary of the Great Western Railway: some notes on its locomotives. Rly Obsr, 1935, 7, 137-41; 145-8; 165-8; 204.7; 225-9: 1936, 8, 22-3; 73-7; 146-8; 228-30; 259-61; 269-71; 303-5: 1937, 8, 11-12; 37 -8; 89-91; 99; 131-2. 12 illus., 2 tables.
Part 2 and subsequent parts were entitled "G.W.R. locomotives, 1921-934" changed in 1936 by the addition of one year to the latter date. F.K. Davies added a further article: Rly Obsr, 1938, 10, 138.
Summers, L.A. Swindon steam: a new light on GWR loco development. Stroud: Amberley. 2013. 224pp.
A refreshing reaximination of locomotive development on the Great Western.
Tuplin, W.A. Great Western steam. London, Allen & Unwin, 1958. xiv, 193 p. + front. + 16 plates. 35 illus., 12 diagrs. (incl. 6 s. els.), 6 tables.
Contains a critical assessment of locomotive development. plus notes on footplate journeys. The illustrations are accompanied by detailed captions. The following brief extract gives a succinct account of Churchward's key contributions:

excellence of basic design;
excellence of detail design;
excellence of materials used;
excellence of workmanship;
intelligence and enterprise in handling on the road;
high standard of maintenance in service.

Tuplin, W.A. Great Western steam. London, Allen & Unwin, 2nd ed. 1965. xiv, 194 p. + 16 plates. 35 i lIus., 26 diagrs., (incl. 14 s. els.), 6 tables.
This edition contains an extra paragraph and more diagrams.

Names

Coltas, Gordon. Names & nameplates of British steam locomotives. Part 2. G.W.R. & absorbed. Crosby: Heyday, 1985.
Ottley 17979

Boilers

Contemporary

Decision to fit Belpaire boiler as standard
Great Western Ry. Locomotive Mag., 1903, 9, 44
Included buying-in from R. Stephenson & Co.

Compton, J.N. Discussion on Cook, K.J. The late G.J. Churchward's locomotive development on the Great Western Railway. J. Instn Loco. Engrs, 1950, 40, pp. 203.
Criticised the short stumpy boilers (a result of Churchward's seeking the maximum degree of standardization in throat plates, etc.) which must have led to difficulties on tubing because the shortness of barrels in proportion to gas area must affect the A/S ratio or hydraulic effect. One must have an enormous number of tubes, and that would lead to too much gas area. Cook avoided the specific question and implied that Churchward was seeking the free circulation of water. Compton's observations on the combining valve received a crisp response:

Rutherford, Michael. The Great Western, boilers and The Great Bear. (Provocations) [Railway Reflections No. 15]. Backtrack, 1996, 10, 146-54.
Boiler development under Churchward culminating in that for The Great Bear set against the competitive position of the railway company at that time.

2-8-0

28XX: 1903: Churchward:
This was the first British 2-8-0 design (the Whale 2-8-0s which were merely an extension of an 0-8-0 design did not emerge until 1904). The prototype, No. 97, was vastly in advance of most other freight locomotives in use on Britain's railways at that time and was comparable to the prototype 4-6-0s being developed during the same period, which represented a similar quantum leap. A slightly modernized version, the 2884 class, was introduced by Collett in 1938. Robinson in 1911 and Gresley (with external valve gear) and Fowler (for the Somerset & Dorset Joint Railway) followed with further 2-8-0 heavy freight locomotive, although in the interim Churchward had produced his 47XX freight locomotive with larger boiler (a design which his successor ignored). Like the not quite so original freight locomotives introduced on the North Eastern Railway within the same decade these designs were to remain the primary freight locomotives within their areas until the end of steam.

Number 97 (prototype)
No. 97, Great Western Railway. Rly Mag., 1903, 13, 259.
2-8-0: in original condition: Works photograph and leading dimensions.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 34-5

28XX
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 77-8

38XX
Side window cab, outside steam pioes and improved sanding gear.
Improved 2-8-0 freight engines, G.W.R. Loco. Rly Carr Wagon Rev., 1938, 44, 166. illus.
No. 2884 illustrated
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 213-16

1945: Hawksworth: modifications for oil firing. (48XX series, briefly)

G.W.R. oil burning locomotives. Loco. Rly Carr. Wagon Rev., 1945, 51, 178-9. 2 illus.
G.W.R. oil-burning locomotives. Rly Mag., 1946, 92, 91-2. 2 illus.
Mullay, A.J. and Neil Parkhouse. Oil for coal: the plan to convert British steam locomotives to oil fuel, 1945-48. Rly Arch., 2006 (12). 4-15; 62-8.
This includes a wealth of material missed from Jones from GWR Mag (1946 Sept) and from Loco. Rly Carr. Wagon Rev, 1947, 53, March. Perhaps most interesting "new material" are the diagrams and observations by Hawksworth in the later reference in 1947.
Rutherford, Michael. Crisis? What Crisis? Coal, Oil and Austerity. Part 1.. (Railway Reflections No. 71) Backtrack, 2000, 14, 665-74.
Following a very brief analysis of the development of coal burning (from coke burning) and the problems of coal supply, especially during strikes and in the immediate Post WW2 period the author introduces oil-consuming traction on the GWR (i.e. the pre-WW2 railcars and post-WW2 steam locomotives) and the influence of Sir James Milne (a thumbnail biography is given). Illus.: No 3813 renumbered 4855 when converted to oil firing, Diagram of the GWR installed equipment in the engine, Diagram of the GWR installed equipment in the tender, Diagram of the firebox showing the extra brickwork and air inlets, The cab of 3904 aka 4972 showing the fireman's padded seat ? It also had electric light!, Col.: GWR no 3711 at one time oil fired in May 1963 (W. Potter), Oil-burning Castle no 100A1 Lloyds in April 1947 on express at Reading (H.N. James)
Rutherford, Michael. Crisis? What Crisis? Coal, Oil and Austerity part 2. (Railway Reflections No. 72) Backtrack, 2000, 14, 724-31.
Further consideration of replacement of coal by oil: including the Great Western/National Programme instigated in 1947. Illus.:GWR proposal for a wide firebox 2-8-0, GWR no 2839 as oil burner no 4808 in May 1948, Fig 1 Increased use of mechanisation in the coal industry, Hall class no 3904 [previously 4972] Saint Bride's Hall, Hall class no 3952 [ previously 6957] Norcliffe Hall at Birmingham on 16 April 1948 (John Edgington), Table 1 Steam locomotive Maintenance and running costs, See letter from L.A . Summers (15, 183) on "Hawksworth Pacific".

Performance and testing

Copsey, John. 28xxs in the north during the Great Western era. Gt Western Rly J., 1994, 2 (11), 459-70.
Brief history of the class in service, including specific tests, notably of 2806 on 11 February 1906 when it hauled 955 tons from from Severn Tunnel Junction to stoke Gifford; 1095 tons to Swindon, and 1630 tons from Swindon to Paddington. On 25 February 1906 No. 2808 on a similar working hauled 2012 tons on the Swindon to London section. In 1907 tests were conducted on the 1 in 50 Gowerton incline. On 2 March 1913 No. 2834 hauled 98 loaded coal wagons (approx 1390 tons) from Banbury to Acton.

Glenfarg tests (NBR)

Grassick, J.P. The locomotive from a footplate point of view. J. Instn Loco. Engrs., 1922, 12, 51-67. Disc. 67-104; 311- (Paper 114)
Grassick worked for the NBR, but in the manner of the time described the Glenfarg tests without actually specifying the railways concerned, but it is quite obvious where and what took place! A.C. Stamer (discussion page 99 et seq) was present at the Glenfarg tests and explains why the T3 out-performed the GWR 28XX due to its more even torque and its greater adhesive weight
Thomas, John. The North British Railway. Vol. 2. 1975
Maj Stemp, traffic superintendent of the NBR, was seaking an eight-coupled engine in 1920 and was particularly attracted by the performance of the GWR 'E Group' 2-8-0s on the heavy gradients in the West of England. The NB's most powerful goods engine had to be double-headed when its load reached 28 wagons. From charts supplied by his GWR opposite number, Mr Nicholls, Stemp saw that the GWR 2-8-0 regularly took 28 loaded wagons over the 9 miles 22 chains between Lostwithiel and Doublebois, where the ruling gradient was 1 in 58, in 29 minutes. The same class hauled Welsh coal trains of 38 wagons over 1 in 90 gradients with regularity and apparent ease. The NBR opened negotiations with the GWR on 14 December 1920. At first it seemed that there would be difficulty in obtaining a route between the engine's home territory and the Border because of gauge problems, but the difficulty was quickly overcome and on 19 December Charles Aldington, general manager of the GWR, informed Calder that he had given instructions for 2-8-0 No 2846 to be prepared for the journey to Scotland. He wrote again on 1 January 1921 to say that Churchward himself would accompany the engine.

The test was fixed for 12 January, the location being the 6 miles 53 chains between Bridge of Earn and Glenfarg where a NER 0-8-0 had been tested in 1916. The GWR engine was handed over to the NB at Berwick on 10 January and at 6.15 that evening was lodged in Haymarket shed. On the morning of the test Calder received the following letter from Aldington:

My chairman and I are keenly interested in the tests and hope they will be satisfactory in all respects. As you know Mr Nicholls and one or two other Great Western representatives will be present, but I am sorry that it is not possible for Mr Churchward or his deputy to visit Scotland just now.
Conditions could not have been worse when the special train conveying NB and GW officers and observers from other companies reached Bridge of Earn. The test stretch was blanketed in snow and a near blizzard was blowing. The NBR entrant was first at the post. With 23 loaded 16-tonners and two goods brake vans (437 ton 8cwt exclusive of the brake vans) the 0-6-0 set off but stuck only a short distance up the bank. The train had to be hauled back to Bridge of Earn. The NBR engine retired from the contest; it was quite unequal to the task allotted to it.

In the second test No 2846 in the charge of the GW crew and with a load of 29 wagons and two brake vans (552 ton 4cwt exclusive of the vans) started away easily and, without faltering, reached Glenfarg in its scheduled time of 25min. For the third test five wagons were added to the load bringing it up to 643ton 9cwt. Again the GW engine got away easily but about two miles from the start it encountered hard packed snow on the rails. It began to slip badly and was forced to a stand from which it failed to recover.

Three days after the tests Calder wrote to Aldington: Mr Nicholls will no doubt have informed you of the results of the tests. I may say that these were carried out under pretty extreme weather conditions, but they proved the superiority of your company's locomotive over ours, and will of no doubt be of great assistance to us in connection with designing engines in the future for the hauling of mineral traffic over heavy gradients. I should just like again to thank you very cordially for your kindness in giving us the loan of your locomotive.

On the following day the NB general manager called for a joint report on the tests from Stemp, Fraser (the civil engineer), Chalmers and Grassick, but it was not until 14 April that the report was forthcoming.

The NB officers generally had a high opinion of the visiting engine. They blamed its failure to complete the third test on the abnormal weather conditions and expressed their confidence in its ability to keep time with 30 loaded wagons on any NB main line under any weather conditions. They thought the GW engine could take 38 wagons of coal from Dunfermline to Aberdeen without assistance and in so doing save the company £7 6s 9d per train.

Fraser had reservations. He thought that NB track would not stand up to the regular running of such a heavy engine. He proposed a top speed limit of 25 mph with severe restrictions at many selected points. It would be impossible, he pointed out, to run the engine through the platform roads at Dundee and use would have to be made of the loops.

Chalmers, too, had reservations, particularly on the question of maintenance. He wrote:
There should be no particular difficulty in designing an engine of the power required within the limits specified but so far as repairs and renewals are concerned it is difficult to estimate such as a locomotive of the Great Western type with its enormously high boiler pressure might mean very heavy boiler repairs and consequently high maintenance cost. From the experience we have had on our own heavy engines we estimate that the increase would be not less than 30 per cent. If designing an engine however, of similar tractive power it would be feasible to considerably modify the boiler pressure without involving a greater loading effect than that produced by the engine on the test.
Stemp summed up:
From an operating point of view the adoption of a type of engine similar in tractive power to that of the Great Western engine making the test would certainly be most advantageous if, later on, the use of such an engine could be allowed on other important main lines as well as on the Aberdeen section. It would eliminate altogether the use of the banking pilots which, taken all over would mean a very considerable saving in engine power.

1948 locomotive trials:
The 28XX was the oldest design tested, but it performed well in the freight trials.

Allen, C.J. The locomotive exchanges, 1870-1948.
A retrospective assessment.

Retrospective and critical
Atkins, Philip. Odious comparisons. Steam Wld., 2015, (335), 8-14.
Annual mileage and availability on a Regional basis for 1955-61; Table 6 also includes 8F, O1 and 28XX 2-8-0 types and some comment
Bodman, Martin. Coals to Newcastle? the 'Jellicoe specials'. Backtrack, 2006, 20, 498-503.
28XX used to haul coal trains from Pontypool Road to Warrington, or to Chester, for onward transmission by LNWR and either the CR or NBR to Grangemouth and thence to the Orkneys to service the British Fleet during WW1.
Bond, R.C. Organisation and control of locomotive repairs on British Railways. J. Instn. Loco. Engrs., 1953, 43, 175-265.(Paper No.520).
Includes mileage/overhaul statistics (86,981 average annual: more than 50% better than LMS 8F) for this class.
Copsey, John. 28xxs in the north during the Great Western era. Gt Western Rly J., 1994, 2 (11), 459-70.
Brief history of the class in service, Following WW1 28XX were allocated to Chester, but most were moved away and the ROD type tended to be used on the Northern Division. In the 1930s the 28XX class tended to displace the 43XX and an important source of traffic was iron ore from the Midlands to South Wales. General arrangement drawing 2884 class Lote 321, Swindon, February 1938 No. 109101.
Hall, Stanley. Railway milestones and millstones: triumphs and disasters in British railway history. 2006.
Milestone: noting that construction continued over a very long period he commented that this reflected the "excellence of the original design"
Le Fleming, H.M. G.W.R. 2-8-0's. J. Stephenson Loco. Soc., 1953, 29, 61-2; 70-2; 87. 5 illus., table.
Le Fleming, Hugh M. International locomotives. Plate 31
Painting of No. 2849 in unlined green
Railway Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes.

30XX: Robinson (G.C.R.): (1925):
In 1925 the G.W.R. purchased 80 surplus WW1 Surplus former Railway Operating Department (R.O.D.) 2-8-Os. These were subsequently slightly modified at Swindon and received the usual G.W.R. embellishments.

[80 R.O.D. type 2-8-0s purchased by the G.W.R. from George Cohen & Armstrong] . Rly Mag., 1925, 57, 95.
G.W.R. 2-8-0 chimneys. Rly Mag., 1939, 84, 226. illus.
No. 3026 received a T.V.R. "A,' class chimney and No. 3033 was fitted with a 47XX type.

Retrospective and critical
Copsey, John. '30XX' ('ROD') class 2-8-0s in traffic. Great Western Rly J., 2014, 12, 101-18.
Also further parts and correspondence generated
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 10. Absorbed engines, 1922-1947. 1966.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 30; 31
Vaughan, Adrian. The heart of the Great Western. Peterborough: Silver Link, 1994.
Relates (pp. 45-9) how Albert King (an Oxford driver) considered the 30XX to be the "worst GWR locomotive", but had experienced a failure of the steam brake when descending a long gradient.

47XX: 1919: Churchward:
The first locomotive was built with a No. 1 boiler, but this proved to be inadequate, so in 1921 a new boiler type (No. 7) was introduced for the remainder of the order. Note how the Locomotive Magazine called it a mixed traffic design and a development from the 43XX 2-6-0. Apart from the larger boilers, the class remained substantially in its original conditions throughout its life. Diagram below from Railway Gazette British locomotive types

Contemporary

2-8-0 mixed traffic locomotive, Great Western Railway. Locomotive Mag., 1919, 25, 84-5. illustration, diagram (side elevation)
Churchward 47XX class: No. 4700 illustrated. Initial announcement (Locomotive Mag., 1919, 25, 65) had noted mixed traffic nature)

Atkins, Philip. New boilers for old.. Steam Wld, 2003, (194) 8-14.
Ten replacement No. 7 boilers were built in 1951/2, but some of the class were not fitted with new boilers until the 1960s. Number 4704 retained an original boiler until April 1961: a life of over forty years and 1½ million miles.
Cook, K.J. Swindon steam. pp. 32
The last class of actual Churchward locomotives was the 4700 2-8-0 mixed traffic, 19in x 30in cylinders, 5ft 8in coupled wheels and with a new design of boiler (standard No 7) 2251b per sq in pressure and with a firebox 10ft long outside, 30,4601b tractive effort. Built shortly after the termination of World War I, the new boiler could not be completed in time, so the first engine had a standard No 1 boiler with a lengthened smokebox. This class was built to handle the heavy fast vacuum-braked freight trains which operated over the trunk routes at night.
An early well-known train of this class was the Cocoa Train from Bristol to London, so named as it conveyed a good deal of Fry's cocoa and chocolate output daily. It had been hauled fairly regularly by the Pacific The Great Bear, then Britain's only engine of that type which had four 15in cylinders and a high capacity boiler, and which for some years worked the 6.30pm, Paddington to Bristol down and the Cocoa train up. Similar "one day transits between important towns" were being developed on other routes for which the 4700 class was built and which they could handle extremely well. When the need for such a locomotive became apparent it was an extremely easy matter to develop the 4300 class by the addition of another pair of coupled wheels. The general type of extension frames remained with cylinders enlarged to 19in diameter (by 30in stroke) and the saddle of larger radius to take the smokebox of the higher capacity boiler. The leading, driving and intermediate axle- boxes, valve gear, crossheads, slide bars and piston valves remained the same and the axleboxes with inclined planes, already fitted to the 2-8-0 and 2-8-0T engines, were provided in the new trailing position. The design provided a fitting fruition to Churchward's actual reign in showing the general foresight inherent in his original plan.
[Following a description of driving one during the General Strike, which included passing through the Weston Loop which he was uncertain whether it had been cleared for use by the 47XX Cook summarised their attributes]: There has been some speculation as to whether the class was regarded as a failure, so that only nine were built. That was not so and some years later, the Running Superintendent asked for some more and they were very nearly incorporated in the new engine building programme, but then Collett decided that although they would be a bit more costly, he would rather build 'Castles' which would also be more suitable for passenger train duplications at peak traffic periods. 47s were brought in for this work when necessary but they did tend to nose about a bit above 60 miles per hour which appeared to be due to the increased sideplay provided at the trailing axleboxes on account of the length, 20ft, of the rigid wheelbase. The trailing axleboxes had a compound arrangement incorporating inclined planes. The boxes could swing over and ride up a plane of one-in-eight either way. The inclination gave resistance to the movement and acted to return the trailing end to its normal position. From a shopping point of view they were excellent and when later techniques were developed these engines were booked for 100,000 miles between repairs which they would do. We had, however, only one spare boiler to cover the nine engines so their shopping programme had to be watched very carefully to avoid being caught with two of them requiring a change of boiler. We generally obtained 300,000 miles between heavy boiler repairs.
Durrant, A.E. Swindon apprentice. Cheltenham: Runpast, 1989. 216pp.
Relates on page 128 et seq how as a draughtsman at Swindon he worked upon modifying the Class with screw reverse, but this was not implemented. He also pondered on the reason for the 60 mile/h speed limit imposed on the class
Gasson, Harold. Nostalgic days: further reminiscences of a Great Western fireman. 1980.
On page 67 it is noted that Driver Bert Edmond said that "they 'wouldn't half roll' down the other side of Savernake" on the 23.40 down express freight first stop Tiverton Junction from Paddington Goods.
Le Fleming, H.M. G.W.R. 2-8-0's. J. Stephenson Loco. Soc., 1953, 29, 61-2; 70-2; 87. 5 illus., table.
Maidment, David. At the heart of Old Oak. Part 3. Steam Wld, 2001 (174) 20-4.
Footplate experience with 4704 on Paddington to Bordesley fast freight which climbed Hatton bank well in spite of being checked at foot, but critical of spartan cab and lever reverse, however.
Railway Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes.
Riley, R.C. The Great Western 47XX 2-8-0s. Rly Wld, 1966, 27, 16-21 +. 8 illus., table.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 119-23
Tickle, C.F. (phot.) In close up.. Great Western Rly J., 5 (35) 177.
47xx cab showing reversing lever, regulator, injector, ATC bell, etc.

Names & other proposed modifications

Rutherford, Michael Great Western 4-6-0s at work. 1995. p. 177
Noted that Collett considered names which included Behemoth and Mammoth and the possible conversion of one to a four-clinder 4-8-0, but there was insufficient room for the inside cylinders.

4-6-2

No. 111 The Great Bear
Introduced by Churchward in February 1908, partly as an excuse to exploit a boiler with a wide firebox. In Railway Reflections No. 86 Michael Rutherford postulated that it may have been intended for non-stop running to Fishguard (where the Great Western investment was huge) and for non-stop running to Truro. At that time King Edward VII had given Royal approval to long non-stop runs and the wide firebox would have given considerable gains in reliability for the routine adoption of such practice.

Contemporary

"The Great Bear". Engineer, 1908. 106, (21 February), 188. diagr.
Pacific type locomotive, Gt Western Ry. Locomotive Mag., 1908, 14, 26-7. illus., diagr. (s. el.)
No. 111 The Great Bear: "logical development of the "Star" class": noted the slight increase in cylinder diameter (from 14½;in to 15in) and the 23ft long boiler barrel with a total heating surface of 3400.81ft², including 545ft² of superheat and a grate area of 41.79ft². The total weight was 96 tons and the maximum axle load 20 tons. A bogie tender was provided. The photograph and side elevation both show the short-lived footsteps adjacent to the cylinders.

Retrospective & critical

Allen, C.J. British Pacific locomotives.1962.
Chapter 1 may be devoted to The Great Bear, but Allen was far from enthusiastic: "The Great Bear was one of the very few locomotive types that Swindon has produced, and in particular among Churchward designs, to which the word 'failure' could be applied.".
Arman, Brian. The 'Bear' at Bath. Rly Arch., 2007, (17) 39-43.
No. 111 The Great Bear: six photogrphs with Star class No. 4017 Knight of the Black Eagle in two of them. Views probably taken in 1908. The feature was followed by correspondence from David Patrick and Peter Treloar which related to the front footsteps and the longer lasting front handrail formerly associated with them..
Bagwell, Philip S. The railwaymen: the history of the National Union of Railwaymen. London: George Allen & Unwin, 1963. 725pp.
Brings out difficulty of firing large locomotive with inadequate depth of cab the inside mechanism was very difficult to reach for lubrication and adjustment
Clements, Jeremy. The GWR exposed: Swindon in the days of Collett and Hawksworth. OPC, 2015.192pp. [pages 64-7]
Reproducs some of Hawksworth's detailed working drawings, but adds little in the way of comment
Coakham, D.O. George Jackson Churchward. Br, Rly. J., 1993 (47), 352
See Brian Arman's article on the Dean/Churchward partnership (BRJ 36) there seems to have been nothing but the odd anecdote to reveal the Churchward personality. G.J.C. is mentioned both as a religious man and a considerable mathematician. Does this dispose of one legend?
On page 111 (!) of his Pictorial Record of GW Architecture (OPC 1977), Adrian Vaughan illustrates The Great Bear at Paddington. The caption relates a yarn about Churchward's interest in astrology [sic] leading him to give the 'Bear' a number which 'ancient peoples believed was associated with the Evil Eye'. This was supposed to be a means of getting his own back on the GWR directorate who allegedly insisted on him building a 4-6-2.
It looks as if Churchward's mild relaxation was the study of numerology or gematria. From what one can gather, the significance of the number 111 is in being a multiple of the 'powerful' Solar number 37, along with 666 which, in spite of its unfortunate association with the beast of Revelations, had great potency among the Gnostic Christians. A jocular remark by 'the Chief' might have been seized on and misinterpreted as it passed down the ranks. See also Rivington My life with locomotives on Nos No. 6029 King Stephen, and that Great Western historians will surely have observed the numerical coincidences here, No. 4029 having been King Stephen in the Star-King series of 1910, and the number of the 'Saint' class Saint Stephen was 2929, the latter two digits being common to the three "Stephens"..
Davis, Peter. The Great Bear. Backtrack, 2008, 22, 702
See Wrottesley below: this letter makes it clear that Collett did not like the locomotive as it might out-perform his Castle class. Also argues that Webb radial truck used rather than a Cartazzi or Bissel truck as these were to hand being used on 36XX and 31XX types
Holcroft, H. Outline of Great Western locomotive practice, 1837-1947. (1957). pp. 120-1
It could be inferred from this that the large boiler with wide firebox and large grate area in conjunction with relatively small cylinders was a reversion to the broad-gauge practice of having a large reserve of boiler power. The "Star" class could handle the longest trains that the Paddington platforms would accommodate at that time, and therefore more tractive effort through larger cylinders was not really called for. The big boiler would, however, enable higher average speed to be attained through faster running on up gradients, due to the reserve of power. In accordance with the policy of braking all wheels, blocks were required for the trailing radial truck, and I was called on to produce a scheme. As the box had 4½in. side play each way and the blocks had to follow, use of the ordinary fixed brake hanger was out of the question, nor could any attachment be made to the sliding box. The problem had therefore to be solved on original lines. After an acceptable brake gear operated by a separate 18 in. vacuum cylinder bad been produced it was worked into the general arrangement, but it came as a disappointment when the plan had to be jettisoned because the back end of the engine was coming out too heavy.
The appearance of The Great Bear brought much prestige to the G.W.R. and the resulting publicity was a great asset. In service, however, the engine did nothing very remarkable. Its axle load of 20 tons 9 cwt.. per coupled axle restricted its running to the operation of passenger and brake-fitted freight trains between London and Bristol only. The radial truck proved to be the" Achilles heel," for it frequently overheated. Side thrust on this box had to be taken up by the back of the wheel boss on each side, and as this was a point difficult to lubricate efficiently and the surfaces were in a position where ashpan dust and .grit thrown up in the four-foot could reach them, the boxes were liable to overheat on the least provocation.
[He] came upon the same sort of thing many years later, on the Southern Railway when riding with "King Arthur" class engines. Some of these had Drummond bogie tenders in which the axleboxes were on the inside of the wheels while others had the Urie bogie tender with outside bearings. The boxes of the former invariably ran at "blood heat," whereas the latter ran almost cold, due to better thrust conditions and being in a position where air cooling was more effective. If, instead of tenaciously clinging to a radial axlebox common to several classes, The Great Bear had been fitted with the modern type of radial truck with outside bearings, more might have been heard of the engine. It was also handicapped by the great length of the boiler barrel, as in the absence of a combustion chamber the tubes were unduly long in relation to their diameter and so steaming was not as free as it should have been.
It is often asked: "Why was The Great Bear built?" No official reason has ever been stated, but it seems probable that the fine performance of the" Star" class prompted the directors to seek to raise the running standards even higher, and they called upon Churchward to see what could be done in the matter. With an eye to future development, the Chief Civil Engineer had for some years been replacing the older bridges by new construction capable of taking axle loadings of 22½ tons. This had been completed only on the London-Bristol section, though the renewals were being made elsewhere with stronger bridges as a matter of routine, but this was a long-term affair and could not be completed for a number of years.
Under the very restricted route availability at the time, The Great Bear became a "white elephant" to the running department, for it was difficult to fit it into more than a few set jobs. It is said of Churchward that when he was told of the appearance of H. N. Gresley's Pacific on the Great Northern Railway in 1922, he remarked: "What did that young man want to build one for: we could have sold him ours." Two years later, after Churchward's retirement, The Great Bear was converted into the 4-6-0 type, in September, 1924, as its boiler required very heavy repairs
Kirby, George. Leamington. Gt Western Rly J., 2006, 8, 238 (letter).
Claimed that his father saw The Great Bear stuck on the curves at Leamington.
Lowe, James W. British steam locomotive builders. 1975.
Lowe considered that in-service performance was disappointing and noted thet the bearings on the rear trailing truck frequently overheated.
R.H. Mann. Odd man out! Part 1: 1903-1926. Rly Wld, 1957, 18, 105-8.
Single-purpose locomotives: 0-10-0 for Lickey Incline; Holden 0-10-0T Decapod (not illustrated), Churchward Great Bear (not illustrated here), U1 2-8-8-2 Beyer Garratt for Worsborough Incline and P1 2-8-2 with Booster
RCTS. Locomotives of the Great Western Railway. Part 8. Modern passenger classes. 1960.
Includes provocative statement that [Churchard's] dislike of "The Bear" was well known. Includes three photographs: original condition, as in 1910 (photographed J.N. Maskelyne) and in final condition with top feed and cat iron chimney.
Rogers, H.C.B. G.J. Churchward. 1975.
Rogers, H.C.B. Express steam: locomotive development in Great Britain & France. 1990.
The end of this famous and fascinating engine is stated in the following extraordinary minute of a meeting of the Locomotive Committee on 1st May 1924, which is recorded in the British Transport Historical Records

(GEN 3-62-6):

"Engine 111 The Great Bear Reconstruction.

The above engine was built at Swindon in February 1908, and had a tractive effort of 27,8001b and weighed l42tons l5cwt. Owing to its extreme weight, it was necessary for the Hanwell Viaduct (a pencilled correction in the margin says: "the old iron skew bridge over the Uxbridge Road Allen, C.J. The locomotive exchanges, 1870-1948. A retrospective assessment.151; not the viaduct) to be rebuilt before the engine could be allowed to work between London and Bristol, to which route it has been limited the whole 16 years on account of the enormous expenditure that would be necessary to strengthen the bridges to carry it on other main line routes. Mr Grierson has estimated the cost of doing so at over £500,000. The 'Castle' class of four-cylinder engine has now been produced and proved successful, the engines of which have a greater tractive effort than The Great Bear by 13½% viz 31,6251b, and as they weigh only 119 tons l7cwt, or 22 tons l8cwt less than The Great Bear, no alterations are necessary to any of the bridges. There is, therefore, no longer any reason for the continued existence of The Great Bear and as it recently came into the shops for general repair and needed new cylinders and a new boiler, advantage has been taken of the occasion to reconstruct it, so that it will be similar to one of the 'Castle' type.

"In the past, reconstructions of engines have been dealt with in the same way as repairs and have not been reported to the Directors, but it is thought desirable to do so in this case owing to the notoriety of the engine and to the fact that several British railways followed our example in adopting the Pacific type and are still building them. They have not produced a more powerful engine of less weight."

Collett must bear ultimate responsibility for this misleading and inaccurate document. No competent engineer would assess the relative power of express engines on the basis of their starting tractive effort. The statement that several British railways were building Pacific locomotives was quite untrue. Only the Great Northern and the North Eastern had built them, and of these only the Great Northern type was still being built (for the LNER) at the date the minute was written. Finally, as three years later the main lines of the GWR could take the 'King" class, with a 22+ ton axle load, it is clear that the statement about the enormous expenditure necessary to strengthen bridges was false, and that Grierson, the Chief Engineer, must either have been withholding information or had been incorrectly briefed by his own department.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 86-91
Rutherford, Michael Great Western 4-6-0s at work. 1995.
Originally published as two separate books: Castles & Kings at work. Ian Allan, 1982 and Granges and Manors at work. Ian Allan, 1985. The reprint is in effect two separate books (with indexes, etc) under one cover. See page 19.
Rutherford, M.. Railway reflections No. 15. Backtrack, 10, 146.
Great Western boilers and The Great Bear set against competitive pressures at theat time envisaging non-stop running to Truro and Fishguard.
Rutherford, M.. Railway reflections No. 86. Backtrack, 16, 64.
A Century of Pacific Locomotives. Part 2
Wheeler, Geoffrey. Fired by steam. London: John Murray, 1987.
Includes coloured plate, based on water colour side elevations of The Great Bear.
Wrottesley, Michael. 'The Great Bear'. Backtrack, 2008, 22, 547-52.
Sources listed (sadly fails to record superb photographs of locomotive at Bath and notes by Brian Arman in Railway Archive, 2008 (17) 39-43). Notes that the cab was very restricted in dimensions which must have made the locomotive both difficult to drive and to fire. Steaming was believed to be difficult, and the Author attributes this to inexperience, and the same problem was encountered with the very similar Princess Royal class until firemen developed skills. The trailing axleboxes were a major source of trouble due to overheating, and the Author cannot understand why outside boxes were not fitted as this would have alleviated the problem: the Raven Pacifics suffered in the same way. Performance is restricted to a tale related by Tuplin (source not stated) of an illicit footplate journey made by E.K. Harrison when the fireman on the 18.30 did all the stoking before leaving Paddington and let this burn through on the way west. Alan Wild considers that this was an improbable story. The colour illustrations include two "F. Moore" postcards (both with the leading footsteps removed very soon after entering service, and one with the doctored livery, i.e. crimson lake rather than chocolate and cream as in the photograph on which the painting was based; the other of the Bear leaving Parsons Tunnel was presumably pure fiction). The black & white illus. include one of the boiler with original superheater; The Great Bear (without leading footsteps) about to leave Paddington in 1909; with four cone ejectors in Old Oak Common shed in early 1914; leaving Paddington with 18.30 for Bristol in 1920; rear view in 1923; on 10.45 Paddington to Cheltenham passing Twyford c1922; smokebox being cleaned at Old Oak Common in 1923; passing Kensal Green gasworks in 1921 on 10.45 to Cheltenham (which it worked as far as Swindon) and official photograph of as rebuilt as No. 111 Viscount Churchill...

4-6-0

Dean
No. 36: 1896

The bogie goods engine for the G.W.R. Locomotive Mag., 1897, 2, 93

Ahrons, E.L. The British steam railway locomotive from 1825 to 1925. page 307
Atkins, Philip. Locos from scratch. Rly Mag., 1989, 135, 516-17.
Swindon Works built locomotive within fiour mmonths of it being ordered.

Dean/Churchward designs

No. 2601

Contemporary

New G.W.R. locomotives. Locomotive Mag., 1900, 5, 26-7.
Termed a "still more remarkable engine" [than the Camel class] was 4-6-0 No. 2601, similar to No. 36 (except that the bogie had inside bearings). Volute springs used throughout except for leading coupled wheels which were spiral. The firebox was projected into the barrel. Pull-out regulator fitted. Piston valves. Sandbox on top of boiler barrel.
7ft single No. 1000 and 4-4-0 No. 3212 fitted with Belpaire type boilers.

Rutherford (Railway Reflections: Backtrack, 1998, 12, 387) writes: "Churchward's purchasing of the French compound 'Atlantics' and the almost complete copying of Brooks Locomotive Works (USA) in his 4-6-0 No. 98 that makes him so unique."

Armin, Brian. The Dean—Churchward transition. Br Rly J., 1994 (36), 267.
Two types are considered: No. 2601 an experimental heavy freight locomotive fitted with piston valves and a combustion chamber and a boiler-mounted sandbox, and the more successful No. 100 fitted with a prototype No. 1 boiler and a steel firebox.

29XX "Saint": 1902: Churchward:
This was the first, modern 4-6-0 design to be introduced in Britain, and this and the 28XX 2-8-0 introduced at about the same time, displayed considerable American influence. In 1924, No. 2925 was rebuilt as the prototype for the 49XX class and in 1931, one locomotive was fitted with Lentz valve gear. Interestingly, Gibson (Great Western locomotive design) considers that the Hall class was less successful than the better-proportioned original design. Perhaps the most amazing aspect of the 29XX series was that so many vastly inferior 4-6-0 designs were produced on other railways (LSWR, GCR, CR, NER, and Wilson Worsdell was a great friend of Churchward, and LNWR) after this epoch-making advance. The four-cylinder Saints were a logical development for higher speeds and greater power. Tuplin gave a long analysis of the 120 mile/h escapade..

No. 100
Introduced in February 1902 and as built had a parallel boiler with raised Belpaire firebox, but had 30 inch stroke cylinders but only 6½ inch piston valves.

New six-coupled express locomotive, G.W.R. Locomotive Mag., 1902, 7, 89 + plate fp. 90
Credited to Mr Dean

Atkins, Philip. Cast in a unique mould. Backtrack, 2011, 25, 442-5.
Casting cylinders for unique (and unusual) locomotives. George Burrows produced the cylinder drawings for Churchward for the initial 30 inch stroke cylinders fitted to 4-6-0 No. 100, subsequently Dean and William Dean. Unlike the later 4-6-0s the piston valves on this locomotive were only 6½in diameter..
Railway Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 8. Modem passenger classes. 1960..

No. 98

No. 98, Great Western Railway. Rly Mag., 1903, 13, 366.
Official photograph and leading dimensions

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.

Data presented for the whole engine at 5 revolutions per second were 12.40 and 4.50, 4.82 and 4.50 for each of the coupled axles.

Working into Cornwall

Farr, Keith. The Great Western and Westinghouse brakes.. Steam Wld, 2002 (176), 51.
For the arrival of Saints in Cornwall writer cites Nock's Great Western 'Saint' class 4-6-0 (1983) which Farr calls "definitive" (Ossie's knowledge was baed on family holidays in Penzance).
Treloar, Peter. The Great Western and Westinghouse brakes. Peter Treloar. Steam Wld, 2002 (176), 51.
Guesses that Saints arrived as far west as Truro when Saltash to St Germans deviation was opened in 1908, but they may not have reached Penzance until 1921 when the timber viaduct was replaced. Cites Nock painting and includes photographic illus. (which may have been taken by Nock) of 2937 Clevedon Court leaving Penzance on up postal.

1931: Collett: No. 2935 was rebuilt with Lentz R.C. valve gear.

EXPRESS locomotive with poppet valve gear, G.W.R. Rly Engr, 1931, 52, 258, 267. illus.
4-6-0 passenger engine, Great Western Ry., with rotary cam poppet valve gear. Loco. Rly Carr. Wagon Rev., 1931, 37, 183. illus.

Cliffe, Joseph. It began with 'Turbomotive'. Backtrack, 2012, 26, 637.
Extract from longer letter:: this also applies to the Lentz gear and is a reason for the restricted passages on GWR No.2935 mentioned by Summers in his 'Rotary cam gears at Swindon' article in the same BT issue. Poppet gears can run at lower cut-offs due to high clearance volume but whether this results in better efficiency is debatable.
Cook, K.J. The late G.J. Churchward's locomotive development on the Great Western Railway. J. Instn Loco. Engrs, 1950, 40, 131-71. Disc.: 171-210. (Paper No. 492)
Hyperlink directs to point in discussion where Cook emphatically states that poppet valves were inferior to standard Stephenson motion.
Le Fleming, Hugh M. International locomotives. Plate 32
Painting of No.2935
Otway, Francis O.J. The Great Western and Westinghouse brakes. Steam Wld, 2002 (176), 51.
Notes a journey made behind 2935 Caynham Court (with rotary cam valve gear) between Swindon and Shrivenham on 4 September 1943 when the performance was considered to be "inferior"
Summers, L.A.. Swindon's acquaintance with rotary cam valve gears. Part One. Backtrack, 2012, 26, 347-51 : concluded 437 et seq
Saint class No. 2935 Caynham Court was fitted with Lentz (Lenz was correct German spelling, but was not used in connection with valve gear in English literature) rotary cam valve gear in 1931 and remained thus fitted until withdrawn in 1947. Also considers the development and use of Lentz valve gears (both oscillating and rotary forms) on other railways including the LNER, and further eccentricities in valve gear introduced at Swindon. Concluded p. 437.

1932: No. 2914 was coupled to the bogie tender from The Great Bear Pacific

[8-wheel tender from The Great Bear 4-6-2 coupled to a Saint class locomotive] . Rly Mag., 1932, 71, 390.

1937: Collett: Speedometers
Programme extended to Saint and Star classes.
More speedometers for G.W.R. locomotives. Rly Gaz., 1937. 67, 996.
Speedometers for Great Western Railway locomotives. Engineering, 1937, 144, 661.
Speedometers on the G.W.R. Loco. Rly Carr. Wagon Rev., 1937, 43, 399.

Retrospective and critical

Allcock, J.N. The Great Western Railway "Saints". Railways, 1952, 13, 180-1. illus.
Davies, F.K. The 2900 class 4-6-0's, G.W.R. Rly Obsr, 1937, 9, 163-5; 199. 4 illus.
Farr, Keith. A Centenary of Saints. Backtrack, 2002, 16,. 186-95.
Written to celebrate the centenary of the inception of what the author regards as the first "true twentieth century express passenger locomotives" with their taper boilers, high running plates, long-lap/long-travel valve gear and generous steam ports". They also formed the basis for the "Hall" class. Includes the various series, such as the "Courts", the evaluation of the Atlantic form, one experiment with rotary cam valve gear on 2935. There is also description of performance, mainly in general terms, including the alleged 120 mile/h with Collett on the footplate. Their final demise is also covered.. See page 354 for possible plans to preserve two of class withdrawn in 1951 and "late" high speed exploit by another member of class. See letter by Barker (page 415) stating that 115 mile/h was achieved by Star in bridge testing [KPJ: pity Horne did not sound] Illus.:The second Saint and true prototype no 98 at Paddington, No 100 at Old Oak Common, A postcard of No 2949 Stanford Court near Ruislip, A post card of No 2977 passing Twyford, No 2934 at Swindon, No 178 Kirkland, No 186 as an Atlantic at Bristol, No 186 in its later form as a Pacific, No 100 William Dean leaving Paddington, No 2902 Lady of the Lake at Paddington, No 2907 Lady Distain in original condition at Paddington, No 2930 St Vincent with a very mixed train, No 2949 Stanford Court passing West Drayton, No 2902 Lady of the Lake in its later form in 1930 at Paddington, No 2917 St Bernard passing Langley, No 2935 Caynham Court with a mixed train leaving Bristol, Table 1 Saints nos 100, 98, 171 principle dimensions as built, Table 2 Saints nos 172/79-90 principle dimensions as built as Atlantics and as Pacifics, Table 3 Saints/Ladies/Courts principle dimensions as built, No 2977 Robertson passing Dawlish, Table 4 The naming of the 'Saints',
GREAT Western Railway: the 4-6-0 2-cylinder passenger engines. J. Stephenson Loco. Soc., 1944, 20, 138-41. 6 illus.
Hall, Stanley. Railway milestones and millstones: triumphs and disasters in British railway history. 2006.
Milestone: "tecnically in advance of any such [express passenger] locomitive in the country"
Leech, K.H. The passing of the "Saints". J. Stephenson Loco. Soc., 1954, 30, 30; 55-6. illus.
Nock, O.S. Fifty years of Western express running. 1954.
Performance.
Nock, O.S. 4-6-0 locomotives of the G.W.R.: their inception and development. Rly pict., 1947, (2), 70-83. 14 illus., 5 tables.
Development and performance.
Nock, O.S. Great locomotives of the GWR. 1990.
Author quotes results from Bridge Stress Committee for Saint class: max. axle load 18.4 tons; hammer blow at 6 rps (86 mile/h) for whole engine 17.9 tons; on axle 6.9 tons and maximum combined 25.3 tons.
Railway Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 8. Modem passenger classes. 1960..
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 33; 34; 41; 42; 44; 48; 49-50.
Rutherford, Michael Great Western 4-6-0s at work. London: Promotional Reprint Co., 1995.
Originally published as two separate books: Castles & Kings at work. Ian Allan, 1982 and Granges and Manors at work. Ian Allan, 1985. The reprint is in effect two separate books (with indexes, etc) under one cover.
Tuplin, W.A. Great Western Saints and Sinners. 1971.
Tuplin included details of the "120 mile/h episode" on which Collett was on the footplate: The following is a precis: it was regular practice to give engines newly-built at Swindon a running-in trip from there to Stoke Gifford and back, a total distance of about 70 miles. It includes a straight descent of 9 miles at 1 in 300 to Little Somerford, followed by a rise of 6 miles at I in 300 leaving plenty of room for stopping. It was usual for every engine that behaved normally on the outward journey to be run pretty fast on the way back. The 1906 incident started a rumour of 'two miles a minute' and this became so persistent a legend that eventually someone persuaded Collett to admit that a group of 'high-ups' at Swindon had indulged in a bit of horseplay. The Railway Magazine for April 1932 states on p. 305 that, on the evidence of Mr Collett, 'The purpose of the run was to demonstrate that an engine taken straight from the shops could be run at over 100 miles per hour. Those on the footplate included Mr Collett, who was then Assistant Manager of the Locomotive Works, Mr G. H. Flewellen, who was Locomotive Inspector, and the Foreman of the Erecting Shop, Mr Evans. (The driver was Mr H. J. Robinson.) The timing for some distance by the mileposts with a stop-watch was given as 120 miles per hour, and the clocking between the signal-boxes of Lime Somerford and Hullavington was booked as two minutes for the 4½ miles. 'Mr Collett points out that; while the object of running a new engine on its first trip at over 100 miles per hour was achieved, the timing could not be regarded as accurate and that the 102.3 m.p.h. record of the City of Truro in 1904, made under the personal observation of one of the most careful recorders of his time the late Charles Rous-Marten — with the aid of a chronograph reading to one-fifth parts of a second, must remain the best duly authenticated railway speed record that this country has. yet witnessed.' As G W engines at that time were never required to exceed 90 m.p.h. in ordinary service and were put on fast jobs only after running for a week or two on slow ones, there was no technical need to know whether a newly built engine could reach 100 m.p.h. So why was it desired to kriow it? Had anyone worked out at what speed the balance weights of a Lady would lift her driving wheels off the rails in every revolution?
They wanted to know whether the Lady could run at over 100 m.p.h. on her maiden excursion. But if that was all, why go up to 120 m.p.h. where out-of-balance forces were nearly half as big again? Had they developed a kind of mob-hysteria that led them to urge the driver to go 'all out' regardless of everything? By the time they had run more sedately on to Swindon and the disciplined serenity of the Works they probably agreed that not a word of this sporting venture should be whispered to anyone. No intention of this kind is ever more than a pious hope and every student of the steam locomotive must be glad that an official statement was eventually made.
It was most imprudent to allow so many officials to participate. in what was undoubtedly a risky exploit, and these sporty souls would naturally pick a day when Churchward was away from Swindon. Had he gained any hint of such intention he would have forbidden it for one obvious reason in some such terms, as 'If you go and get yourselves all killed, where the bloody hell am I?' What really did happen on this extraordinary occasion? No one concerned would say a word while the incident was fresh in anyone's mind although it was admitted that a speed of about two miles a minute was reached. One is justified in resorting to conjecture and the fact that the engine had pole reversing gear, which is not safely adjustable at speed, makes one wonder whether this was a factor.
It is possible that the driver had found that his original pole-setting was not producing 100 m.p.h., and was thereupon persuaded by one of the officials to try a little later cut-off. If he were rash enough to attempt this (or if someone else said 'Here, let me do it') the engine might well drop into full gear and accelerate like mad. Everyone might then be so appalled by the exhaust noise of such running as to do nothing at first to neutralize it, until by the time the driver had recovered from the shock of having the pole pulled out of his hand and had got round to turning the blower on and shutting the regulator, some unabashed spirit said, 'No! leave her at that. Let's see what she'll do.'
Or did the driver try to close the regulator and find that he couldn't? A big flow of steam from the boiler will sometimes cause such a pressure drop from one side of the regulator-valve to the other that the valve becomes very hard to move and if an engine without a train is running very fast downhill with steam urging it on, its brakes won't stop it quickly, if at all.
If anything of this sort happened on Lady of Lyons, already doing nearly 100 m.p.h. down 1 in 300, there could have been plenty of lively apprehension on the footplate. With driving wheels leaving the rails eight times per second, with the engine using water so fast that the fusible plugs were at risk and with the impossibility of stopping in any distance less than about six miles, someone' had to do something. There were plenty of people there, and so two could pull at the reversing pole and two at the regulator handle while someone else made sure that both injectors were working and then, still remembering what they had come for, took some mile-post-passing times.
Evidently they did get things under control at last, with 100 m.p.h. well and truly exceeded and everybody on board still shaking; The sensible ones would realize that they'd only had what they'd asked for and had been lucky to get away with it. After the white faces there would be metaphorical red ones, and no desire to admit to anyone what danger they had produced for themselves and the engine.! This is only a guess. What a pity that they didn't have anything like the black box that aircraft now carry to take a record of what went on during alarming last moments! The nearest equivalent to it was provided by the signalmen at Hullavmgton and Little Somerford. Their evidence of roughly 135 m.p.h. tended to show that Collett's report of about 120 m.p.h. was not an exaggerated one but, he himself admitted it and disclaimed it in the same sentence. An interpretation of this is that the speed, was attained but in circumstances that did not do much credit to those,responsible for the exploit.
Tuplin, W.A. Swindon's "Saints" and "Stars". Trains ill., 1953, 6, 10-15. 5 illus.,9 tables.
Two miles a minute. Rly Mag., 1932, 70, 305-6.
A statement was made by H.J. Robinson, which was confirmed by Collett, that he had driven a Saint (Jones Steam locomotive development stated a "Star class" ) at 120 mile/h.
Waters, Laurence. Great Western 'Saint' class locomotives. Barnsley: Pen & Sword, 2017, 140pp.
Reviewed by DMA in Backtrack, 2018, 32, 254: seen very briefly by KPJ

40XX "Star": 1906: Churchward:
This 4-cylinder design was the foundation for the Castle and King classes. Several were rebuilt as Castles and others were modernized to some extent. It is amazing that the design precepts encapsulated in the Saint, Star and 28XX classes were ignored elsewhere, sometimes until the late 1920s.

No. 40 North Star: 1910 conversion from 4-4-2 to 4-6-0 and fitted with Swindon superheater.

Great Western Railway. Loco. Rly Carr. Wagon Rev., 1910, 16, 22.

Original descriptions

Engineer, 1907, 22 February.
Cited by Poultney British express locomotive development. Chapter 9: The 'North Star' — Swindon, 1906.
4-6-0 express locomotive No. 4021 "King Edward", Great Western Railway. Loco. Rly Carr. Wagon Rev., 1910, 16, 121.
As used to haul Royal Funeral train from Paddington to Windsor on 20 May 1910.
Great Western Ry. Locomotive Mag., 1910, 16, 260.
The latest engines ot the new Queen class (4-6-0) were Nos. 4033 Queen Victoria, 4034 Queen Adelaide and 4035 Queen Charlotte.
Great Western Ry. Locomotive Mag., 1911, 17, 3.
Latest 4-6-0 engines of the Queen class were Nos. 4036 Queen Elizabeth and 4037 Queen Philippa.
Great Western Ry. express engines, "Princess" class. .Locomotive Mag., 1914, 20, 260. diagram (side elevation)
Churchward Star class: Nos. 4046 Princess Mary, 4047 Princess Louise, 4048 Princess Victoria, 4049 Princess Maud, 4050 Princess Alice, 4051 Princess Helena, 4052 Princess Beatrice, 4053 Princess Alexandra, 4054 Princess Charlotte, 4055 Princess Sophia, 4056 Princess Margaret, 4057 Princess Elizabeth, 4058 Princess Augusta, 4059 Princess Patricia, 4060 Princess Eugenie.
Great Western Ry. Locomotive Mag., 1923, 29, 17.
Abbey class: No. 4063 Bath Abbey... with hollow crank axles

Hammer blow

Data presented for the whole engine at 5 revolutions per second were 2.57 and 5.13, 2.64 and 0 for each of the coupled axles (the high hammer blow on the leading coupled axle should be noted).

Speedometers
Programme extended to Saint and Star classes.
MORE speedometers for G.W.R. locomotives. Rly Gaz., 1937, 67, 996.
SPEEDOMETERS for Great Western Railway Locomotives. Engineering, 1937,144, 661.
SPEEDOMETERS on the G.W.R. Loco. Rly Carr. Wagon Rev., 1937, 43, 399.

Performance
27 July 1946: Snow Hill to Cardiff service hauled by 4058 Princess Augusta with a train load of 15 coaches or about 450 tons. Departed Birmingham 14 min late and arrived 62 min late.An overloaded 'Star', John Copsey. Great Western Railway J.,2, 519-20.

Retrospective and critical

Bagwell, Philip S. The railwaymen: the history of the National Union of Railwaymen. London: George Allen & Unwin, 1963. 725pp.
Brings out difficulty of firing powerful locomotive with inadequate depth of cab: the inside mechanism was very difficult to reach for lubrication and adjustment
Copsey, John. Lode Star. Br Rly J. 1991 (37), 323-8.
The central focus of this feature is the official engine history card (1928 onwards) for No. 4003 Lode Star (reproduced). The text also notes some of the workings (from its construction in 1907) on which the locomotive participated. Illus.: (all 4003): approaching Pilning on up South Wales express c1935; on shed at Shrewsbury? c1932; at Gloucester station? c1947; nameplate; passing Kennington Junction, Oxford with train of LNER stock c1950; at Swindon with 2301 class No. 2516 which was also preserved; in Swindon stock shed with name and number plates removed.
Copsey, John 'Stars' in Traffic Part 2. Gt Western Rly J., 2000 (34) 89-97.
Frame plan (and side elevation dated April 1919, also cross sections dated September 1906. Tables show workings.
Digby, Nigel J.L. The liveries of the Pre-grouping railways. Volume one. Wales and the West of England. Lydney: Lightmoor Pres, 2017. 96pp.
No. 4007 Rising Star on pages 44 and 45.
G.W.R. Star class: outside steam pipes. Rly Obsr, 1950, 20, 74.
Holcroft, H. (Paper No. 65) Three-cylinder locomotives. J. Instn Loco. Engrs., 1918, 8. 355-68. Disc.: 368-95; 476-91.
"They [the Stars] exerted considerable influence, and their lead is due to the fact that they appeared at the proper moment in the development of the locomotive, and also because definite mechanical advantages were cheaply bought with the minimum of added parts."
Le Fleming, Hugh M. International locomotives. Plate 43
Painting of No. 4037 Queen Berengaria alongside her consort King
Maskelyne, J.N. Churchward's 4-cylinder locomotives. J. Stephenson Loco. Soc., 1947, 23, 30-5; 206-1 2. 9 illus. (line drawings: s. els.)
Nock, O.S.. The G.W.R. Stars, Castles & Kings. Part 1:1906-1930. 1967. .
Nock. O.S. Fifty years of Western express running. 1954.
Nock. O.S. 4-6-0 locomotives of the G.W.R.: their inception and development. Rly pict., 1947, (2), 70-83. 14 illus, 5 tables.
Nock, O.S. Great locomotives of the GWR. 1990.
Author quotes results from Bridge Stress Committee for Star class: max. axle load 18.6 tons; hammer blow at 6 rps (86 mile/h) for whole engine 3.7 tons; on axle 3.7 tons and maximum combined 21.3 tons.
Nock, O.S. Historical steam locomotives. 1959. Chapter 14. Churchward's masterpiece.
Book based on preserved locomotives, but Nock fails to make any reference to the one actual preserved Star.
Patrick, David. The Great Bear [letter]. Rly Arch., 2008 (19), 52.
Note on the front footsteps and handrail fitted to Nos. 4002/8/9/4011-16 and 4018 and subsequently removed.
Railway Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 8. Modem passenger classes. 1960.
Reed, Brian. Great Western 4-cylinder 4-6-0s. Locomotive Profile No. 3.
pp.49-72: centre coloured artwork drawn by David Warner (restricted to King & Castle types: Star not covered).
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 81-5; 149.
Rutherford, Michael Great Western 4-6-0s at work. London: Promotional Reprint Co., 1995.
Originally published as two separate books: Castles & Kings at work. Ian Allan, 1982 and Granges and Manors at work. Ian Allan, 1985. The reprint is in effect two separate books (with indexes, etc) under one cover.
Tuplin, W.A. Hot work on a "Star". Rly Mag., 1956, 102, 86-90. 2 illus.
Written in narrative form: it describes a 1924 "Cornish Riviera" footplate journey.
Tuplin, W.A. No more "Stars". Rly Wld, 1957, 18, 323-4. illus.
An appreciation.
Tuplin, W.A. Swindon's "Saints" and "Stars". Trains ill., 1953, 6, 10-15. 5 illus. 9 tables.
A history of these classes.
Wheeler, Geoffrey. Fired by steam. London: John Murray, 1987.
Includes coloured plate, based on water colour side elevation of Star class locomotive: 4041 Prince of Wales as built in 1913.

Names

Early G.W.R. Chairman commemorated. Rly Gaz., 1937, 66, 1097.
No.4007 Rising Star re-named as Swallowfield Park.

Specific locomotives

No. 4021 King Edward on Royal Funeral Train at Southall on 20 May 1910. Rly Archive, 2014 (45) 75 lower

4073 "Castle": 1923: Collett:
Churchward had planned to incorporate the No. 7 boiler, which had been introduced for the 47XX class, into a new 4-6-0. This would have been based on the Star chassis. The project had to be abandoned as it would have been too heavy. Therefore, Collett had to break with some of the principles of standardization achieve greater power. The four-cylinder layout of the "Stars" was retained in e new design, but the cylinders were enlarged to 16in x 26in. The grate was also larger (29.36 ft²) and a modern, side-window cab was fitted. The new design achieved a 14% increase in tractive effort over the Stars, at the expense of a 5% weight increase. The class was highly successful in service and locomotives continued to be built until the 1950's. Some Stars and The Great Bear (the Churchward 4-6-2) were reconstructed as Castles. Return to beginning.

Short cuts: exchange trials with LNER A1 pacific

EMPIRE Exhibition — Great Western locomotive "Caerphilly Castle". Engineer, 1924, 137, plate f.p. 540. diagr. (s. el.), plan
Sectionalized diagram.
FOUR-CYLINDER express locomotive, "Caerphilly Castle", Great Western Railway. Loco. Rly Carr. Wagon Rev., 1923, 29, 254-6. 2 illus., diagr. (s. el.)
FOUR-CYLINDER 4-6-0 type, Great Western Railway. Engineering, 1923, 116, 236; 742-3 + folding plate. 2 illus., 6 diagrs. (mcI. s. el.), plan.
Includes sectionalized diagrams.
G.W.R.four-cylinder locomotive: Caerphilly Castle. Engineer, 1923, 136, 197; 202.3 illus.,diagr. (s. el.)
NEW four-cylinder 4-6-0 locomotives, Great Western Railway. Rly Mag, 1923, 53, 311 + plate f.p. 253. illus.
NEW four-cylinder 4-6-0 type express locomotives, Great Western Railway. Rly Engr, 1923, 44, 394-6. 2 illus., 3 diagrs. (s. els.)
The Castle design is compared with the L.N.E.R. Al and A2 Pacifics.

1924: No. 111
The Churchward Pacific, was "rebuilt" to the Castle specification.

Great Western Railway—;reconstruction of "The Great Bear" No. 111. Loco. Rly Carr. Wagon Rev., 1924, 30, 329. illus.

1925: No. 4009 Shooting Star
Rebuilt with Castle class boiler and cab. Loco. Rly Carr. Wagon Rev., 1925, 31, 173

1926: 4,000 gallon standard tender.

GREAT Western Ry.: "Castle" class locomotive with new pattern tender. Loco. Rly Carr. Wagon Rev., 1926, 32, 341-2. 3 illus., diagr. (s. & els.)
NEW type of locomotive tender, Great Western Railway: a self-trimming tender of new design, built at the Swindon works for use with the "Castle" class engines. Rly Engr. 1927, 48, 24-6. 6 illus., 4 diagrs., plan.
Includes sectionalized diagrams.

1927: No. 5001 Llandovery Castle fitted with 6ft 6in coupled wheels.
Great Western Ry. Loco. Rly Carr. Wagon Rev, 1927, 33, 28.

1930: No. 4000:
This was a rebuilt locomotive: the original being the prototype for the Star class.

An INTERESTING G.W.R. rebuilt locomotive. Loco. Rly Carr. Wagon Rev., 1930, 36, 274-6. 2 diagrs., plan.

1931: Rigid eight-wheel tender (experimental).

G.W.R.8-wheel tender. Rly Mag., 1939, 85,154.
Note on transfer from No. 5001 to No. 5032.

1935: Speedometers
King No. 6001 fitted experimentally on 11 March 1932, but later removed, but refitted on 19 April 1932. By 1935 whole of King class fitted mostly with British Thomas Houston (BTH) and Jaeger units, but with a couple from another source. 31 of the Castles had also been equipped mainly with BTH speedometers by 1935, Gt Western Rly J., 2011, 10, 417. See also extension of programme.

1935: Streamlining
No.5005 was fitted with a domed smokebox and other aids to streamlining. Part was removed in the same year.

STREAMLINED casing removed from around the cylinders and motion of the streamlined "King" and "Castle" locomotives due to overheating, Rly Mag., 1935, 77, 385.

1937: No. 5005:
Most of the remainder of the streamlining was removed due to increased oil consumption and bearing over-heating.
G.W.R. 4-6-0 locomotive "Manorbier Castle". Loco. Rly Carr. Wagon Rev., 1937, 43, 376. illus.
Illustration shows streamlined locomotive: paragraph noted why it was removed.

1946: 5098 series;
Hawksworth: this series incorporated three-row superheaters and mechanical lubricators.

The SWINDON "Castle" class locomotives. Rly Gaz., 1949, 90, 12-13. illus., diagr. (s. & f. els.)

Oil fuel conversion: 1946

5083 Bath Abbey and four other Castle class were converted to oil-burning

Mullay, A.J. and Neil Parkhouse. Oil for coal: the plan to convert British steam locomotives to oil fuel, 1945-48. Rly Arch., 2006 (12). 4-15; 62-8.
This includes a wealth of material missed from Jones from GWR Mag (1946 Sept) and from Loco. Rly Carr. Wagon Rev, 1947, 53, March.
Nock. O.S. Fifty years of Western express running. 1954.
pp 286-295 cover running, mainly in Cornwall, with both Castle and Hall classes fitted for oil burning: most of the running was of a high standard.
Rutherford, Michael. Crisis? What Crisis? Coal, Oil and Austerity. Part 1.. (Railway Reflections No. 71) Backtrack, 2000, 14, 665-74.
Following a very brief analysis of the development of coal burning (from coke burning) and the problems of coal supply, especially during strikes and in the immediate Post WW2 period the author introduces oil-consuming traction on the GWR (i.e. the pre-WW2 railcars and post-WW2 steam locomotives) and the influence of Sir James Milne (a thumbnail biography is given). Illus.: No 3813 renumbered 4855 when converted to oil firing, Diagram of the GWR installed equipment in the engine, Diagram of the GWR installed equipment in the tender, Diagram of the firebox showing the extra brickwork and air inlets, The cab of 3904 aka 4972 showing the fireman's padded seat ? It also had electric light!, Col.: GWR no 3711 at one time oil fired in May 1963 (W. Potter), Oil-burning Castle no 100A1 Lloyds in April 1947 on express at Reading (H.N. James)
Rutherford, Michael. Crisis? What Crisis? Coal, Oil and Austerity part 2. (Railway Reflections No. 72) Backtrack, 2000, 14, 724-31.
Further consideration of replacement of coal by oil: including the Great Western/National Programme instigated in 1947. Illus.:GWR proposal for a wide firebox 2-8-0, GWR no 2839 as oil burner no 4808 in May 1948, Fig 1 Increased use of mechanisation in the coal industry, Hall class no 3904 [previously 4972] Saint Bride's Hall, Hall class no 3952 [ previously 6957] Norcliffe Hall at Birmingham on 16 April 1948 (John Edgington), Table 1 Steam locomotive Maintenance and running costs, See letter from L.A . Summers (15, 183) on "Hawksworth Pacific".

Oil fuel conversion: 1960
According to Atkins' Dropping the fire detailed planes were drawn up in 1960 for converting some of the class to oil-burning due to the increasing cost of coal.

Performance and testing;
On the basis of fuel consumption for power output the Castle class was in advance of other contemporary British locomotives. This was due to the use of long-travel valves and to the type of boiler. This superiority was clearly shown when the Gresley Al and Castle classes were evaluated in a series of comparative road trials. Good performance was repeated when one locomotive was lent to the L.M.S. The modified Al (later A3) and Royal Scot classes were both produced as a result of these locomotive exchanges.
Later, the class was involved in high-speed running when hauling the Cheltenham Flyer express. Reference should also be made to the retrospective material (especially to Nock's and Collett's works).

1925: Al /"Castle" exchange trials.

Allen, C.J. British locomotive practice and performance. Rly Mag., 1925, 57, 47-57; 151-63.
Performance of both types of locomotive is described.
Allen, C.J. The locomotive exchanges, 1870-1948.
A retrospective assessment.
The EXCHANGE trials of locomotives. Engineer, 1925, 139,627-8.
Editorial comment on the G.W.R. report on the results of the tests.
Harrison. in Peter Towned. LNER Pacifics remembered. Chapter 9.
Observes that Castle class was near zenith of Churchward's design strategy whereas Gresley's A1 was at early stage of its development during the 1925 interchange trials. Furthermore, Driver Young (GWR) deliberately ignored speed restrictions (an interesting jibe made many years later!)..
INTERCHANGE trials of passenger locomotives on the Great Western and London and North Eastern Railways. Loco. Rly Carr. Wagon Rev., 1925, 31, 142. illus.
General details of the test runs. Results are not included.
The LOCOMOTIVE exchange. Rly Engr, 1925, 46, 199-200.
The LOCOMOTIVE trials. Engineer, 1925, 139, 492. Erratum p.519.
Editorial comment.
OFFICIAL statements as to results [by the Companies concerned]. Rly Mag., 1925, 57, 57-9.
RESULTS of interchange locomotive trials, London and North Eastern and Great Western Rlys. Loco. Rly Carr. Wagon Rev., 1925, 31, 196.
The journalistic conclusions ars of interest in that one reason given for the efficiency of the Castle locomotive is that it was fitted with a vacuum pump in place of an ejector!
Voyageur, pseud. The locomotive exchange: London & North Eastern "Pacific" v Great Western "Castle". Rly Mag., 1925, 56, 478-82. 4 illus.

1926: high speed Plymouth — Paddington runs.

G.W.R. record journeys from Plymouth to London. Rly Mag., 1926, 58, 408- 9. table.

1926/27: Euston-Carlisle test runs.

Allen, C.J. British locomotive practice and performance. Rly Mag., 1927, 60,185-96.
Allen, C.J. The locomotive exchanges, 1870-1948.
Nock, O.S. British locomotive practice and performance. Rly Mag., 1964, 110, 289-95.
Comparative tables ("Castle"/"Claughton) are quoted for fuel and water consumption during the Crewe-Carlisle tests. The material originated from R.C. Bond's records.

16 September 1931: Cheltenham Flyer: No.5000 Launceston Castle: Swindon Paddington average speed 79.6 mile/h.

Mercury, pseud. A new "record of records". Rly Mag., 1931, 69, 313-17.2 illus.,diagr., table.

5 June 1932: Cheltenham Flyer: No. 5006 Tregenna Castle: Swindon Paddington average speed 81.68 mile/h.

New record run by the fastest train. Locomotive Mag., 1932, 38, 203
A new record was set up by the Great Western Ry.'s Cheltenham Flyer between Swindon and Paddington on Monday, June 6, when the 77¼ miles were covered in 56 minutes 47 seconds, at an average speed of 81.6 miles per hour. Many consecutive miles were run at 90.3 miles per hour. Leaving Swindon at 3-48 p.m., No. 5006 Tregenna Castle (Driver Ruddock), with six coaches, weighing about 195 tons, soon reached a speed of 60 miles per hour, and at 5 miles out was running at 80 miles per hour. Near Wantage Road 92 miles per hour was reached. The train reached Paddington 11 minutes adhead of schedule. On the same day the 5 p.m. ex Paddington, with engine No. 5005 Manorbier Castle (Driver Burgess) and six coaches, one being a restaurant car, 210 tons, made a special stop at Swindon 15 minutes before time. On this run on a slightly rising gradient the highest speed attained was 86½ miles an hour.

Allen, C.J. The Great Western world-speed record. Rly Mag., 1932, 71, 99-103. diagr., table.
Allen, C.J. A super-speed record. Rly Mag., 1932, 71, 10-12.3 tables.
Baker, H. The Great Western record. Engineer, 1932, 153, 638. tables.

Descents towards Honeybourne

Allen, Cecil J. British locomotive practice and performance. Rly Mag., 1940, 86, 82-92.
This article included details of 100 mile/h achieved when descending from the Cotswolds on 1 in 100 bank between Camden and Honeybourne with Castle class on light trains, notably 5063 Earl Baldwin, 5049 Earl of Plymouth and 4086 Builth Castle.

1956 Controlled road tests with high-degree-superheat Castle

Nock, O.S.British locomotive practice and performance. Rly Mag., 1959, 105, 260-
Comparison with "King" and No. 71000.
Nock, O.S. Further W.R. locomotive developments. Loco. Rly Carr. Wagon Rev., 1956, 62, 170-1. illus., table.

Retrospective and critical

Atkins, Philip. New boilers for old... Steam Wld, 2003, (194) 8-14.
7019 Fowey Castle received new boilers in January 1953, October 1954 and July 1956; yet 7020 Gloucester Castle never received a new boiler.
Atkins, Philip. Odious comparisons. Steam Wld, 2015, (335), 8-14
Annual mileage: 60,000 in 1937 and 47,000 in 1950
Bond, R.C. Organisation and control of locomotive repairs on British Railways. J. Instn. Loco. Engrs., 1953, 43, 175-265.(Paper No.520).
Includes mileage/overhaul statistics (87,424 average annual) for this class.
Bullock, William. Swindon apprentice. Rly Wld, 1992 (62). 30-3.
Application of "streamlined" casing to 5005 Manorbier Castle and its test when 100 mile/h was achieved when approaching Swindon which was overshot. Train was met by Collett and Pole whilst Inspector 'Daddy' Dew was on footplate at controls (amazing how LNER always managed to stop unlike GWR & LMS).
Carpenter, G.W. Discussion on Cook, K.J. The late G.J. Churchward's locomotive development on the Great Western Railway. J. Instn Loco. Engrs, 1950, 40, pp. 202-3
Asked whether it was originally intended to fit the 47xx class No.7 boiler to the "Castle" class engines, as this had the same length between tubeplates and a larger diameter? It had occurred to him that the increased weight of the No.7 boiler as compared with that actually fitted to the " Castle" class engines, and consequently increased axle loading, might have been the principal objection to doing this. Cook agreed that this was so.
CASTLES and Kings : a pictorial tribute. Hatch End (Middlesex), Roundhouse Books, 1964. [96 p.] 122 illus.
Entirely pictorial.
Chapman, W.G. "Caerphilly Castle" : a book of railway locomotives for boys of all ages. 1924.
Good publicity material. Shows a "Castle" being built.
Collett, C.B. Testing locomotives on the Great Western Railway. Trans. 1st.Wld Pwr Conf., London, 1924, 4, 882-94. 8 diagrs.
Includes the results of dynamometer car tests with special reference to fuel consumption.
Cook, A.F. Raising steam on the LMS: the evolution of LMS locomotive boilers. Huntingdon: RCTS, 1999. 233pp.
Table 50 (page 217) quotes the cost of classified boiler repairs on a comparitive basis in pence per mile in 1954: 2.7 pence/mile for a Duchess as against 0.8 for an A4 and 0.6 for a Merchant Navy: the Castles did not perform well on this basis: 1.8 pence per mile.
Copsey, John. 'Castles' in traffic. Great Western Railway J., 13, 62-77.
Very well chosen photographs printed in chronological order. No. 5005 is shown as streamlined and in greatly reduced streamlined form (latter in December 1935). Allocations and typical workings.
Cox, E.S. Locomotive panorama. 1965. p. 56
No. 5000 Launceston Castle was brought over to the L.M.S. First of all used between Euston and Crewe, it subsequently ran on the Carlisle road and on this latter route was the subject of Dynamometer car trials at which I was fortunate enough to be present. Throughout the whole exercise No. 5000 performed with quiet mastery all the work on which the Claughtons, often piloted, lost time, dropped their steam pressure, and made the welkin ring with their reverberating exhaust. It displayed the full measure of the extent to which Bowen Cooke had failed to absorb the lesson of the earlier G.W.-L.N.W. locomotive exchange in 1910, which was the forerunner of the Claughton design. Train loads of 400 and 430 tons were taken on successive round trips, the engine usually gaining from 7 to 11 mins. on a schedule of 164 mins. for the 141 miles. On one occasion however 6 mins. were lost between Oxenholme and Tebay due to persistent slipping in bad weather, a rather unusual failing for this class. Pressure was held almost faultlessly throughout the series, a cutoff of 22% being maintained on the more level sections, with part open regulator. Cutoff was generally opened out to 27% at Grayrigg and 35% at Shap Summit going north and to 30 % at the latter point coming south. Coal consumption averaged 3.8 lbs. per D.B.H.P. hour, which was better than anything yet seen on the L.M.S., even on the Midland Compounds, and notwithstanding the low degree superheat then common in all G.W.R. practice. The Dynamometer Car report concluded with the words" As regards efficiency it will be noted that whereas the rates of combustion and evaporation were in no way exceptional, the coal and steam consumptions were very good and this would indicate that the efficiency of the Great Western engine results from the engine portion and economical accessory fittings, and as a result of this the boiler is well able to produce the steam required at low fuel consumption"
Fermor, Drew. GWR/BR (WR) Castle class Nos. 4073-7030 (including 3, 3 & 4 row superheater versions): a guide to the history and operation of Britain's most successful express passenger steam locomotive type. Sparkford: Haynes Publishing, 2014. 156pp.
An excellent series which this volume fails to reach the overall standard
Gasson, Harold. Nostalgic days: further reminiscences of a Great Western fireman. 1980.
Chapter 2 includes a run with Driver Burt Edmonds with No. 5027 Farleigh Castle which was far from being in prime condition on an up express fromm Swansea. When the train reached Newport two coaches were added to accommodate a Royal party and Inspector George Price joined the footplate. The train was now slightly over-weight, but Edmonds declined assistance and even with careful management of the fire three minutes were lost on the the climb to Badminton, but by very fast running subsequently all of this was regained and a surplus was gained to accommodate a permanent way slowing at Hanwell.
Gibson, John C. Great Western locomotive design. 1984.
Highly critical of the design, especially its inaccessible valve gear,its boiler proprtions and its lack of adequate superheating (see especially 108 et seq)
Grime, T. Steam locomotive performance (theoretical and actual). J. Instn. Loco. Engrs, 1926, 16, 588-652. (Paper No. 200).
Analysis of "Castle" tests.
Harding, B.J. The G.W.R."Castle" class—additional details. J. Stephenson Loco. Soc., 1965, 41, 72-3; 222.
See Leech "The Castles..." (below)
Harding, B.J. 'Castle' class engines with longer smoke boxes. Backtrack, 8, 166.
See letter from David Maidment on page 50: there was a second engine No.4093 Dunster Castle with a smokebox four inches longer than standard and this was the third 'Castle' to be equipped with a double chimney. Like No.4090 Dorchester Castle, it was fitted with a four-row superheater boiler (HC6688) and when they were rebuilt Nos.4090 (April 1957) and 4093 (December 1957) were both given new front ends incorporating new front half frames, cylinders, &c, and mechanical lubricators, making them virtually new engines. All Castles fitted with double chimneys after No.4093 (with the exception of No.5068 mentioned below) retained their standard length smokeboxes. In March 1960 No.4093 reverted to a standard length smokebox whilst retaining a four-row superheater boiler and double chimney and it was condemned in that condition in September 1964. In November 1960 No.4090 was fitted with the boiler from No.4093, which had the longer smokebox and double chimney, and ran in that condition until withdrawal in June 1963. The boiler HC7671 new to No.4090 in April 1957 was fitted to No.5068 Beverston Castle in March 1961, retaining the larger smokebox and double chimney, and this engine ran in that condition until condemned in September 1962. Boiler HC7671 was built new in 1957, whilst boiler HC6688 was one of the two-row superheater type HA built new in 1940 and converted to four-row superheater type HC in 1957.
Holcroft, H. "Castles", "Lord Nelsons" and "Royal Scots". Rly Mag., 1947, 93, 13-15; 27.3 illus.
The relationship of the three designs, with emphasis on the last named.
Johnson, Martin. An introduction to steam locomotive testing [letter]. Backtrack, 2009, 23, 702.
Response to series of articles on locomotive testing Notes that indicator diagrams taken at high speed suffer from overrun.. Considers that multiple sets of valve gear as fitted to A1 Tornado produce far more even exhaust beat than any form of derived gear whether as adopted at Swindon for Castles and Kings, by Stanier on the Duchess class, or by Gresley. Noted extremely irregular behaviour on some Castle class which was highly indicative of overrun in the outside cylinders.
Kelway-Bamber, H. Modern British railway express passenger engines. J. Instn. Loco. Engrs., 1926, 16, 1004-17.
A detailed analysis of Castle locomotive work (including on the LNER) plus a comparative study of this design, the Al and "Lord Nelson" classes. One Castle run between Paddington and Plymouth was used to estimate the resistance of engine, tender and train
Langston, Keith. British steam: GWR Collett Castle class. Barnsley: Pen & Sword Books, 2015, 240pp..
Each locomotive is illustrated: for each locomotive amjor changes are noted; plus the total mileage (which in the case of those "derived" from the Star class also includes that of the Star locomotive). Date and place of scrapping. The late changes to superheating, draughting (double chimneys, etc) are also recorded. The "streamlined" No. 5005 Manorbier Castle is examined in slightly greater detail. There are photographs of the military aircraft associated with Nos. 5071 et seq, but the castles, abbeys and people associated with class names are ignored! The illustrations, especially the colour ones, are well reproduced,
Leech, K.H. The "Castles" of the Great Western Railway. J. Stephenson Loco. Soc., 1963, 39, Supplement (September). 28 p.28illus., table.
A history.
Leech, K.H. "Kings" and "Castles" today. J. Stephenson Loco. Soc., 1957, 33, 364-8. illus., 2 tables.
British Railways modifications to draughting and superheating.
Matthewson-Dick, T. Address by the President. How they run. J. Instn Loco. Engrs, 1967, 57, 155-96.
Comparison of King, Castle, D1000 and D833 running times on Paddington to Plymouth toute.
Nock, O.S.. The G.W.R. Stars, Castles & Kings. Part 1:1906-1930. 1967.
Nock. O.S. Fifty years of Western express running. 1954.
Nock, O.S. Great locomotives of the GWR. 1990.
Author quotes results from Bridge Stress Committee for Castle class: max. axle load 19.7 tons; hammer blow at 6 rps (86 mile/h) for whole engine 3.5 tons; on axle3.5 tons and maximum combined 23.1 tons.
Nock. O.S. The "Kings" and "Castles" of the Great Western Railway. [1949].
A concise account with accent on performance.
Nock. O.S. Post-war development at Swindon. Trains A., 1958, 5-12. 4 illus., 4 tables. Includes modifications made to Castle boilers and draughting.
Poultney, E.C. Locomotive performance and its influence upon modern practice. J. Instn. Loco. Engrs, 1927, 17, 172-261. Disc.: 261-72. (Paper No. 213)
General assessment of class alongside other British and other contemporary locomotives
Railway Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 8. Modem passenger classes. 1960.
Reed, Brian. 150 years of British steam locomotives. p. 99.
Near completion of a lengthy bridge strengthening programme enabled Collett at Swindon to introduce a larger four-cylinder 4-6-0 in 1923. These Castle-class engines had an adhesion weight scarcely two tons more than that of the four-cylinder engines of 1914, but that was enough to permit a 10in lengthening at the back end and the installation of a firebox with 12 per cent more grate area and 6 per cent more heating surface. The back end of the taper barrel was only 3in bigger than that of the standard No 1 boiler and the pressure of 225psi was retained; but the cylinder volume was increased by 14 per cent.
Standard of performance was higher than Churchward had been able to attain by 1914. That engineer had set himself to get a drawbar pull of 2 tons at 70mph from his express engines; the French Atlantics were the first to give it, but by 1908 he was getting it and more from the Stars in normal service. In 1924 the Castles were measured on test to sustain 2.35 tons at 71mph, but in service thereafter they frequently exceeded that measure, and they were probably the greatest single influence in forwarding British locomotive design through the last 40 years of new construction.
Reed, Brian. Great Western 4-cylinder 4-6-0s. Locomotive Profile No. 3.
pp.49-72: centre coloured artwork drawn by David Warner (restricted to King & Castle types).
Riley, R.C. Portrait of a locomotive: No. 100 A1 Lloyd's. Rly Wld, 1981, 42, 19-22
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 149-54; 187
Tuplin, W.A. Draught in the locomotive boiler. Engineer, 1958, 205, 1223. diagr.
Includes draughting modifications to this class.
Webber, A.F. The proportions of locomotive boilers. J. Instn Loco. Engrs, 1937, 27, 688-725. (Paper No. 378).
Includes Castle on a comparative basis.
Wheeler, Geoffrey. Fired by steam. London: John Murray, 1987.
Includes two coloured plates (based on water colour side elevations) of Castle class locomotives: 4079 Pendennis Castle and 5069 Isambard Kingdom Brunel

Names

Castle class. Rly Wld, 1954, 15, 140.
No. 5017 named The Gloucestershire Regiment by the Colonel of the Regiment, Major General C.E.A. Frith at Gloucester Central station
[Castle class locomotives re-named with names of War-time aircraft]. Loco. Rly Carr. Wagon Rev., 1940, 46, 273. illus.
Nos.5071-5082 : new names Spitfire, etc: full list: No. 5071 Spitfire illustrated.
[Castle class locomotives re-named with names of War-time aircraft]. Rly Gaz., 1940, 73, 336. illus.
[Castle class locomotives re-named with names of War-time aircraft]. Rly Mag., 1940, 86, 608. illus.
"A1 at Lloyd's": naming a new G.W.R. engine. The Times, 1936, 18 February. p. 11
Naming took place at Paddington Station on 17 February 1936 and was performed by Sir Robert Horne, Chairman of the GWR who was accompanied by Nevelle Dixey, Chairman of Lloyds. The engine had been numbered "A1" and Lloyd's Coat of Arms surmounted nameplate. Driver W.H. Sparrow and Fireman A.H. Miles were on the footplate.
G.W.R. "Al at Lloyd's". Rly Gaz., 1936, 64, 356. illus.
Naming ceremony. Pike's Locomotive names (2000) makes it clear that this locomotive was always cited as No. 100 and not as 100 A1 which creates a difficulty was the locomotive Lloyds as listed by Pike or A1 Lloyds. A retospective piece (Rly Wld, 1978, 39, 110-111) written to mark the formation of the Lloyd's Railway Society shows two photographs of the locomotive in service (which clearly show the separate 100 and A1 numberplates on the cabside and the name Lloyd's: members of Lloyd's are also shown with a nameplate from, the locomotive.
Great Western Railway. Locomotive Mag., 1936, 42, 93.
As a compliment to LIoyd's Underwriters, the G.W.R. have re-named one of their rebuilt "Castle" class engines Lloyds and numbered it "100 A.1" The naming ceremony took place at Paddington on Monday, 17 Feb. and was performed by Sir Robert Horne, chairman of the G.W.R., in the presence of Mr. Neville Dixey, chairman of LIoyd's, the Committee of LIoyd's, Sir James Milne, general manager of the G.W.R., and chief officers. The nameplate of the engine is surmounted by LIoyd's Coat of Arms, and flew the official House Flag. An unusual feature of the ceremony was the display of a string of flags of the International Code between the chimney and the cab of the engine, reading "A.1 at Lloyd's." The engine was formerly No. 4009 Shooting Star (exact wording).
Great Western Railway. Locomotive Mag., 1938, 44, 67
No.4016 The Somerset Light Infantry (Prince Albert's) by General Sir Walter Braithwaite at Paddington on 18 February.
G.W.R.engine "Al at Lloyds". Rly Mag., 1936,78, 209.
G.W.R.engines named after Brunel and Gooch. Engineer, 1938, 165, 706.
Nos. 5069 and 5070.
G.W.R. locomotive named "Viscount Portal". Rly Gaz., 1946, 85, 55. illus.
No.7000.
LOCOMOTIVE named after G.J. Churchward. Railways, 1948, 9, 187. illus.
No. 7017.
LOCOMOTIVE names "The Gloucestershire Regiment". Rly Gaz., 1954, 100, 502; 534. 2 illus.
No.5017 (formerly St. Donats Castle).
LOCOMOTIVE naming ceremony at Paddington. Rly Gaz., 1938, 68, 392.
No.4016 The Somerset Light Infantry (Prince Albert's).
LOCOMOTIVE naming ceremony at Paddington. Rly Mag., 1938, 82, 304.
NAMING formality at Paddington. Rly Gaz., 1937, 66, 769; 820. illus.
No. 4037 The South Wales Borderer.
NAMING formality at Paddington. Rly Mag., 1937, 80, 462.
NAMING of locomotive "G.J.Churchward". Rly Gaz., 1948, 89, 530. 2 illus.
No.7017.
Western Region locomotive "G.J. Churchward". Loco. Rly Carr. Wagon Rev,, 1948, 54, 184; 185.
Western Region locomotive renamed. "Castle" class locomotive. Loco. Rly Carr. Wagon Rev,, 1954, 60, 69
No. 5017 has been renamed The Gloucestershire Regiment.

Individual locomotives

No. 5069 Isambard Kingdom Brunel
Ballantyne, Hugh. IKB: engineer extraordinaire. Steam Wld, 2006, (232) 24-5.
Black & white photograph of Castle No. 5069 Isambard Kingdom Brunel passing through Sydney Gardens Bath at head of 13.15 Paddington to Weston-super-Mare with two of GWR special saloons at front of train and headboard on locomotive on 15 September 1959 conveying members of Institution of Civil Engineers to Bristol for unveiling ceremony near Clifton Suspension Bridge to mark centenary of engineer's death.

Restored locomotives

No. 7029 Clun Castle
A tribute to Clun Castle. Railway Wld., 1983, 44, 305-7.

49XX Hall class: 1924: Collett:
The prototype was rebuilt from the Saint class. The main changes were a reduction in the driving wheel diameter (to 6ft 0 in) and the addition of a side-window cab. Eventually 330 engines were built from new. See also 6959 Modified Hall series.

1924: No.2925 Saint Martin:
Rebuilt 29XX and prototype for the class.

Great Western Rly. rebuilt 4-6-0 passenger engine No.2925 "Saint Martin". Loco. Rly Carr. Wagon Rev., 1925, 31, 101-2. illus., diagr. (s. & f.els.)
Rebuilt 4-6-0 locomotive, G.W.R. Rly Mag., 1925, 56, 316 + plate f.p. 253. illus.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 159-60

1928 : production series

4-6-0 type locomotives for the Great Western Railway. Engineering, 1929, 127, 23. illus.
NEW 4-6-0 "Hall" class locomotives, Great Western Ry.. Loco. Rly Carr. Wagon Rev., 1929, 35, 1-2. illus., diagr. (s. & f. els.)
Text begins "To take the place of the 4-4-0 engines which are not able to deal with the long and heavy trains of today". Also somewhat muddled about 6ft driving wheels implying that all Saint class had been so fitted rather than the prototype Saint Martin which is not mentioned. Illustration of No. 4901 Adderley Hall.
NEW 4-6-0 mixed traffic locomotive, Great Western Railway. Rly Mag., 1929, 64, plate f.p. 85. illus.

6959:1944: Hawksworth:
Changes in construction methods required modifications to the frame and cylinder layouts. At the same time slightly larger superheaters were incorporated

GREAT Western Railway: further noteson the "6959" class. J.Stephenson Loco.Soc., 1945, 21, 103. illus. (line drawing: sel.)
MODIFICATIONS to G.W.R. "Hall" class mixed traffic locomotives. Rly Gaz., 1944, 81, 456-7. 2 illus., 2 diagrs. (s. els.)
MODIFIED 4-6-0 "Hall"-class engine. Loco. Rly Carr. Wagon Rev., 1944, 50,114. illus.,diagr. (s.el.)
MODIFIED G.W.R. "Hall" class locomotives. Rly Mag., 1944, 90, 350-1. illus., diagr. (s. el.), table.

1946: Hawksworth : conversion to oil fuel.

FIRST G.W.R. oil-burning passenger locomotive. Rly Gaz., 1946, 85, 29. illus.
OIL-BURNING locomotives, Great Western Railway. Engineering, 1946, 162,
416. 8 diagrs.
Mullay, A.J. and Neil Parkhouse. Oil for coal: the plan to convert British steam locomotives to oil fuel, 1945-48. Rly Arch., 2006 (12). 4-15; 62-8.
This includes a wealth of material mised from Jones from GWR Mag (1946 Sept) and from Loco. Rly Carr. Wagon Rev, 1947, 53, March. Perhaps most interesting "new material" (from GWR Mag) is a picture of Viscount Portal with Hawksworth waiting departure of 5955 Garth Hall from Paddington..
Nock. O.S. Fifty years of Western express running. 1954.
pp 286-295 cover running, mainly in Cornwall, with both Castle and Hall classes fitted for oil burning: most of the running was of a high standard.

1947 : Hawksworth : electric lighting on two locomotives.

AIR-DRIVEN alternator for G.W.R. locomotive lighting. Rly Gaz., 1947, 87, 408; 410-11. 4 illus.
Electric lighting equipment on G.W.R. Locomotive Mag., 1947, 53, 190-1. illus.
No. 3904 St. Bride's Hall (oil burning)
ELECTRIC lighting on G.W.R.engine. Railways, 1948, 9, 73-4. 4 illus.

Testing

1948 : Inter-regional exchanges:
No.6990 worked between Marylebone and Manchester, but was prohibited as out-of-gauge from the other routes.

Allen, C.J. The locomotive exchanges, 1870-1948.
A relatively close in time assessment, but the King class was limited to the London to Leeds route.

c. 1950 : One of the class was subjected to the full scale test methods developed by S.O. Ell of the Western Region.

BRITISH Railways. Western Region "Hall" class 2 cyl.4-6-0 mixed traffic locomotive. London, British Transport Commission, 1951. [1],14, [35] sheets. 2 illus., 46 diagrs (incl. s. & f. els.) (Performance and efficiency tests with exhaust steam injector. Bulletin No.1).
Metcalfe (page 127) noted that exhaust steam injector produced savings in coal expressed in lbs coal per ft² of grate area of 9-12%

Retrospective and critical
Atkins, Philip. Odious comparisons. Steam Wld, 2015, (335), 8-14
Annual mileage; availability, and average engine miles per working day in 1958: 329 locos: 38,445; 78%; 159
Bond, R.C. Organisation and control of locomotive repairs on British Railways. J. Instn. Loco. Engrs., 1953, 43, 175-265.(Paper No.520).
Includes mileage/overhaul statistics (87,942 average annual) for this class.
Cook, K.J. Machining a locomotive valve gear: operations involved in the production of Stephenson link motion for G.W.R. "Hall" class express engines. Rly Gaz., 1936, 65, 264-8; 296-300. 10 illus., 4 diagrs.
Copsey, John. Cornish 'Halls' in the 1930s. Great Western Rly J. Special Cornish Issue. 35-9.
Workings in Cornwall in 1925 by prototype 2925 included the Cornish Riviera, allocated to Penzance in 1927. Once the Hall class proper was introduced they were allocated to Penzance, Truro and Laira (for working into Cornwall) and were used for express trains. Load limits are quoted.
Copsey, John. 'Halls' on goods in the Midlands and the North. Great Western Rly J. 5, (39), 398-403.
Allocations and workings during period prior to WW2: Illus.: 4900 Saint Martin at Tyseley on 17 May 1936; Halls under construction at Swindon in 1929; 4996 Eden Hall at Tyseley in 1932; 4976 Warfield Hall at Chester in 1936?; 4955 Plaspower Hall; Hall on class H freight passing Milton on 26 June 1938.
Nock. O.S. Fifty years of Western express running. 1954.
Nock, O.S. 4-6-0 locomotives of the G.W.R.: their inception and development. Rly pict., 1947, (2), 70-83. 14 illus., 5 tables.
Development and performance.
Poultney, E.G. Locomotive valve gears. Engineer, 1953, 196, 762-3. 2 diagrs., 3 tables.
A comparison of the Stephenson gear fitted to the "Hall" class with that of the Walschaerts gear fitted to the B.R.class 4 4-6-0.
Railway Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 8. Modem passenger classes. 1960.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 161; 220; 224

Liveries

Hancock, J. Painting of Great Western locos. Br Rly J., 1989 (28) 396.
Criticism of preseerved and model tank locomotives for painting bunker side green: it was always black. Writer joined GWR in 1927.
Mullinger, C. Bertram. Liveries. [letter] Br Rly J., 1991 (36) 307.
From 1937 the Hall class locomotives had much of their brass work painted over in green.

King class (60XX): 1927: Collett:
The "King" class was perhaps Collett's greatest achievement as it represented the ultimate in British 4-6-0 development in regard to size and power. The axle load was 22½ tons and Holcroft explains how a locomotive of this size came to be acceptable to the Chief Civil Engineer. Innitially may have been considered as Cathedral class: see Locomotive Mag., 1927, 33, 178. To some extent the design seems to have been developed for prestige reasons since the Castle class was capable of operating most of the services. Further, a modest increase in power could have been achieved by using the 47XX boiler on the Castle class. However, many other famous classes were prestige designs. King George V was one of the few British locomotives ever exhibited in the United States. Return to beginning..

FOUR-CYLINDER 4-6-0 express locomotive, Great Western Ry. Loco. Rly Carr. Wagon Rev.. 1927, 33, 206-7.3 illus.
G.W.R.new 4-cylinder express locomotive. Engineer, 1927, 144, 8-9.3 illus., diagr. (s. & f.els.)
NEW 4-6-0 type express passenger locomotives — Great Western Railway. Rly Engr. 1927, 48, 251-60. 20 illus., 3 diagrs. (incl. s. & f. els.)
NEW "Super-Castle" locomotive, Great Western Railway. Rly Mag., 1927, 61, 126-8 + plate f.p.89. illus., diagr.
RECENT express locomotives. Engineer, 1927, 144, 180.
Editorial comment on the Lord Nelson, Royal Scot and King designs.

1932: speedometers
King No. 6001 fitted experimentally on 11 March 1932, but later removed, but refitted on 19 April 1932. By 1935 whole of King class fitted mostly with British Thomas Houston (BTH) and Jaeger units, but with a couple from another source. 31 of the Castles had also been equipped mainly with BTH speedometers by 1935, Gt Western Rly J., 2011, 10, 417. See also extension of programme.

1935; streamlining:
No. 60l4 was partially streamlined, but parts of the casing were removed in the same year.

EXPERIMENTAL streamlining of G.W.R. locomotives. Rly Gaz., 1935, 62, 518. 3 illus.
STREAMLINED locomotive, G.W.R.. Loco. Rly Carr. Wagon Rev., 1935, 41, 104. illus.

1935 : Partial removal of partial streamlining.

[STREAMLINED casing removed from round the cylinders and motion of the G.W.R. streamlined King and Castle locomotives due to overheating] - Rly Mag., 1935, 77, 385.

1928: United States visit: No.6000 King George V was sent to the Baltimore & Ohio Railroad's Centenary Fair of the Iron Horse

G.W.R. locomotive "King George V" in America. Rly Mag., 1928, 62,135-8. 4 illus., diagr.

Farr, Keith. The Great Western and Westinghouse brakes. Steam Wld, 2002 (176), 51..
6332 was not the only locomotive to be fitted with the Westinghouse brake: it was fitted to No. 6000 King George V for its North American tour.

Performance and testing

1927 : Paddington-Plymouth trial run.

Trial run to Plymouth of Great Western locomotive, No.6000. Loco. Rly Carr.Wagon Rev., 1927,33,251-2.

1932 : A rather vague statement referring to tests at 100 to 110 mile/h. ["KING" class tests at 100-110 mile/h]. Rly Mag., 1932, 71, 76.

1927 21 November
Seaton, Douglas . On the footplate of the Cornish Riviera Express. Locomotive Mag., 1928, 34, 15-17. illustration, table
Driver Wimhurst, Fireman W. Howkins and Locomotive Inspector H.J. Robinson. Locomotive No. 6005 King George II. Log of performance

1948 : the inter-regional exchanges

Allen, C.J. The locomotive exchanges, 1870-1948.
King class was limited to the London to Leeds route.

Ell, Samuel O. Developments in locomotive testing. J. Instn Loco. Engrs., 1953, 43, 561-633; 729-34. (Paper No. 527)
Based on tests with King class 6001.
Ell, S.O. discussion on Tuplin, W.A. Some questions about the steam locomotive. J. Instn Loco. Engrs., 1953, 43, 671-4. (Paper No. 528).
Nock. O.S. Locomotive trials on the Western. J. Stephenson Loco. Soc., 1955, 31, 307-9.
Nock. O.S. W.R. modified "King" locomotive trials. Loco. Rly Carr. Wagon Rev., 1956, 62, 152-4. illus., 2 diagrs.

Improved King class: trials to assess faster timings to Plymouth: later Duchess No. 46237 City of Bristol was tested.
Nock, O.S. The "Cornish Riviera" trials. Rly Wld, 1955, 16, 145.

Retrospective and critical

Allen, C.J. The Western "Kings" — a valediction. Trains Ann., 1964, 5-16. 12 illus.
Assessment of the class, based on performance in service.
Atkins, Philip. New boilers for old.. Steam Wld, 2003, (194) 8-14.
A 1956 plan to rebuild eight Kings with new frames and roller bearings was abandoned although the bearings were ordered. Nevertheless, new front ends, frames and cylinders were fitted to most of the class by 1958.
Atkins, Philip. Odious comparisons. Steam Wld, 2015, (335), 8-14
Annual mileage: 60,000 in 1937 and 47,000 in 1950: Annual mileage: 285 miles per day during 1950-8
Bond, R.C. Organisation and control of locomotive repairs on British Railways. J. Instn. Loco. Engrs., 1953, 43, 175-265.(Paper No.520).
Includes mileage/overhaul (78,987 average annual) statistics for this class.
Bradley, D.L. Locomotives of the Southern Railway. Part 2. RCTS, 1975.
Quotes locomotive repair costs per mile (excluding boiler) 3.43p and boiler repair costs (0.91p) and coal consumption per train mile (43.8 lb) for 1953/4. Original source not quoted.
Castles and Kings: a pictorial tribute. Hatch End (Middlesex), Roundhouse Books, 1964. [96] p. 122 illus.
No text.
Chapman, W.G. The "King" of railway locomotives : the book of Britain's mightiest passenger locomotive, for boys of all ages. London, G.W.R., 1928. [vi], 149 p. 95 illus, 8 diagrs., 3 tables.
Publicity material.
Coffin, Rex O. Kings of the Great Western 1927-1977. Hereford: 6000 Locomotive Association. 104pp.
Reviewed Railway World, 1977, 38, 474.
Cook, A.F. Raising steam on the LMS: the evolution of LMS locomotive boilers. Huntingdon: RCTS, 1999. 233pp.
Table 50 (page 217) quotes the cost of classified boiler repairs on a comparitive basis in pence per mile in 1954: 2.7 pence/mile for a Duchess as against 0.8 for an A4 and 0.6 for a Merchant Navy: the King class boilers cost 1.3 pence/mile, which was far better than the Castles (1.8)
Cook, K.J. Swindon steam. pp. 72-3
About that time [late 1920s] there were some locomotive derailments. There was one, a serious one, of a River class tank engine on the Southern at Sevenoaks, a King, King William IV No 6002 derailed her bogie when on the down 'Cornish Riviera' at Midgham, No 4508, a 2-6-2 tank engine was derailed near Kidderminster and a 4300 class jumped her leading coupled wheels, the pony wheels remaining on the track, on Menheniot curve between Menheniot and St Germans in Cornwall, but except in the case of the Sevenoaks accident there were no casualties. [KPJ: it would seem that Cook was incorrect about 4508: it was presumably 5508, and the accident was near Bridgnorth. The incident at Menheniot has not been traced and may not have involved a reportable incident].
Copsey, John. 'Kings' on the Northern Line.Great Western Rly J., 1992, 1, 109.
Pattern of workings: also includes detailed working drawings (82100, Swindon 1927).
Detail alterations —"King class" 4-6-0's. Rly Obsr., 1936, 8, 140.
Gibson, John C. Great Western locomotive design. 1984.
Critical of the design, although argues that a more logical progression from the Star class than the Castle, but greater power led to trouble with broken engine and bogie frames. Furthermore, "there is little doubt that the Kings were significantly overweight" (page 111).
Gregson, W. discussion on Gresley, H.N. High-pressure locomotives. Proc. Instn mech. Engrs., 1931, 120, 101-35. Disc.:135-206
queried "how did No. 10,000 compare with the 250 lb per sq in in four-cylinder simple engines of the G.W.R. which had always been noted for their economical running" (Gresley did not respond!).
Hills, Richard L. Power from steam. 1989.
Here it is interesting to note that the Great Western Railway, which used Welsh coal in its locomotives, retained a narrow firebox for its most powerful 4-6-0 King class engines in 1927 whereas the London & North Eastern Railway employed a wide firebox for the 4-6-2 Flying Scotsman in 1923 because the calorific value of its coal was not so high.
Holcroft, H. The Great Western Railway and its personnel. Part 3. Engineer, 1960, 209, 634-7. 6 illus. (incl. 3 ports.)
This article includes some notes made from Sir Felix Pole's "Pole's Book" (then only available as privately circulated). This explains how Collett was unaware of the Civil Engineer's policy of bridge strengthening to take 22½ ton axle-loadings. Further the idea for this large locomotive would seem to have originated from one of the directors, namely Sir Aubrey Brocklebank.
Johnson, Martin. An introduction to steam locomotive testing [letter]. Backtrack, 2009, 23, 702.
Response to series of articles on locomotive testing Notes that indicator diagrams taken at high speed suffer from overrun.. Considers that multiple sets of valve gear as fitted to A1 Tornado produce far more even exhaust beat than any form of derived gear whether as adopted at Swindon for Castles and Kings, by Stanier on the Duchess class, or by Gresley. Noted extremely irregular behaviour on some King class which was highly indicative of overrun in the outside cylinders. Wonders why Hawksworth failed to modify class with four sets of valve gear..
Justin, R. "King" boilers. J.Stephenson Loco. Soc., 1955, 31, 33; 118.
In service in June, 1954.
'Kings' in a Welsh Valley. Trains Ill. Ann., 1961, 38.
In 1938 Nos. 6004 King Geoge III and 6015 King Richard III tests were made between Newport and Ebbw Vale with loads which eventually reached 1350 tons with a King class at each end of the trains. This feature states that the GWR was contemplating a four-cylinder design: Robin Barnes Locomotives that never were states (and shows) that a 2-10-2T was contemplated using the King class boiler (but with two cylinders).
Leech, K.H. "Kings" and "Castles" today. J. Stephenson Loco. Soc., 1957, 33, 364-8. illus., 2 tables.
British Railways modifications to draughting and superheating.
Leech, K.H. The "Kings" of the Great Western Railway. J. Stephenson Loco. Soc., 1962, 38, Supplement (November). [8 p.] 7 illus., 2 tables.
A history of the class.
Le Fleming, Hugh M. International locomotives. Plate 43
Painting of No. 4037 Queen Berengaria alongside her consort King Richard I
McCormack, Kevin. Great Western 'Kings'. Haynes. 152 pp.
Review by Phil Atkins in Backtrack, 2012, 26, 443. excellently set in a contemporary context with interesting associated archival material, of the 30 elite GWR 'King' class four cylinder 4-6·0s. This covers not only from their introduction in 1927 until their remarkably abrupt withdrawal during 1962 (already 50 years ago!), but also the subsequent restoration to full working order not only of the flagship No.6000, but also much more recently, and with considerable difficulty, of Nos.6023/4 from desperate hulk condition. In many respects the equals of the larger 4-6-2s of the LMS, LNER and Southern, the GWR 'Kings' featured in front-rank service throughout their 32-35 year normal working lives. Their final years, ie 1960-1962, when diesel-hydraulic traction was rapidly, albeit briefly as it turned out, flooding the Western scene, are particularly well documented here. Although No.6006 was withdrawn in February 1962, no others followed until June, when no fewer than seven were retired. This also coincided with the final heavy general repair of No.6002, when such repairs cost several thousand pounds, the equivalent of a very large sum indeed in today's currency, incredibly only for it to be condemned merely three months later, still leaving five remaining 'Kings' active. Many of the illustrations have not previously been published and reference is made to the enigmatic testing, of which no photographic evidence nor any definitive official explanation appears to be on record, of two 'Kings' on the Ebbw Vale branch in 1938. No reference is made to the covert increase in weight of the class over the years by at least 7 tons, thereby raising the axle load from the consistent official 22 \12 to an unacknowledged 25 tons. (This would no doubt have increased still further had the eight sets of roller bearings actually supplied for fitting to members of the class cl957, been installed). The author is in error in stating that the ground- breaking Churchward two-cylinder 4-6·0 No. 100 entered service in early 1902 fitted with a taper boiler from new, although one was indeed later fitted. It was actually the subsequent 4-6-0 No.98 (March 1903) which established both this and the remarkably enduring overall Churchward format, including cylinder- wise. A regrettable omission from the necessarily select bibliography is Michael Rutherford's excellently researched Castles & Kings at Work (Ian Allan, 1982) which particularly throws considerable light on the ancestry of Swindon's finest and also provides more dimensional data for the technically minded. At the end of the book there are stunning very recent photographs of Nos 6023 and 6024, the former liveried in the short-lived BR blue, now of increasingly distant memory, and with its original elegant single chimney restored. The author's enthusiasm and intimate hands-on knowledge of the 'King' Class comes through in this book, which is highly recommended.
Matthewson-Dick, T. Address by the President. How they run. J. Instn Loco. Engrs, 1967, 57, 155-96.
Comparison of King, Castle, D1000 and D833 running times on Paddington to Plymouth toute.
Nock, O.S.. The G.W.R. Stars, Castles & Kings. Part 1:1906-1930. 1967.
Nock. O.S. Fifty years of Western express running. 1954.
Nock. O.S. The "Kings" and "Castles" of the Great Western Railway. [1949].
A concise account with accent on performance.
Nock. O.S. Post-war development at Swindon. Trains Ann., 1958, 5-12. 4 illus., 4 tables. Includes modifications made to Castle boilers and draughting.
Poultney, E.C. Some notes on locomotive power. Engineer, 1961, 211, 205-7 2 diagrs., table. (Rly Engng Abs 14221).
Evaluation of the controlled road tests in the light of subsequent diesel testing.
Railway Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 8. Modem passenger classes. 1960.
Reed, Brian. 150 years of British steam locomotives. p. 99.
When the bridge strengthening programme was complete a further extrapolation was made in 1927 with the Kings, the most powerful 4-6-0s to run anywhere. but with the high axle load of 22½tons they had restricted route availability and only 30 were built over a span of three years, whereas 171 Castles were constructed at Swindon over a period of 27 years. The Kings were soon handling 550-ton trains from London to Westbury, 95 miles, on mile-a-minute schedules, and took up to 375 tons unpiloted over the 1 in 40 South Devon grades.
Reed, Brian. Great Western 4-cylinder 4-6-0s. Locomotive Profile No. 3.
pp.49-72: centre coloured artwork drawn by David Warner (restricted to King & Castle types).
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 168-70; 188; 189
Rutherford, Michael Great Western 4-6-0s at work. 1995.
Originally published as two separate books: Castles & Kings at work. Ian Allan, 1982 and Granges and Manors at work. Ian Allan, 1985. The reprint is in effect two separate books (with indexes, etc) under one cover. See page 19.
Stanier, W.A. discussion on Vallantin, R.G.E. Compound locomotives of the P.L.M. Rly. J. Instn Loco. Engrs., 1931, 21, 283. (Paper No. 274)
At the time of the Vallantin paper Stanier was not a member of the Institution, but: "I have turned up the records of some trials of the Castle class, and I find they give a figure of 10lbs. of coal per 100 ton-miles. The coal per i.h.p. is 2.1 lbs. We must remember in considering these figures that that is with South Wales coal which has not been damaged by transport by sea and by tranference from coal tip to ship and out of the ship again, and so on".
Topham, W.L. The running man's ideal locomotive. J. Instn Loco. Engrs., 1946, 36, 3-29. Disc.: 29-91. (Paper No. 456)
Topham admired the external axleboxes fitted to the bogie.
Tritton, Julian S. Locomotive limitations. J. Instn Loco. Engrs., 1946, 36, 283-323. (First Sir Seymour Biscoe Tritton Lecture)
On p. 296 it was claimed that the King class had exceeded the maximum dimensions for a narrow grate.
Tuplin, W.A. Draught in the locomotive boiler. Engineer, 1958, 205, 122-3. diagr.
Draughting experiments.
Wheeler, Geoffrey. Fired by steam. London: John Murray, 1987.
Includes coloured plate, based on water colour side elevations of King class locomotive: 6000 King George V: associated text notes bogie problems and original intension to name class after Cathedrals.

Names
Cathedral class see introduction

[No. 6028 re-named King George VI]. Locomotive Mag., 1937, 43, 30
That it was formerly King Henry II not mentioned.

68XX "Granges": Collett :1936 :
Parts of scrapped 43XX 2-6-0s (mainly the wheels and motion) were incorporated into this new class which was intended to replace the 43XX type.

4-6-0 mixed traffic locomotives, "Grange" class, Great Western Railway. Loco. Rly Carr. Wagon Rev., 1936, 42, 304-5. illus., diagr. (s. & f. els.)
[INTRODUCTION of the "Grange" class, G.W.R.]. Rly Mag., 1936, 79, 384-5. illus., diagr. (s.el.)
NEW mixed traffic locomotives, G.W.R.. Rly Gaz., 1936, 65, 490-1. illus., diagr. (s. el.)

Retrospective & critical

Atkins, Philip. Odious comparisons. Steam Wld, 2015, (335), 8-14
Annual mileage; availability, and average engine miles per working day in 1958: 80 locos: 37,680; 78%; 156
Copsey, John 'Granges' at Work. . Great Western Rly J., 4, 147-64.
Allocations and workings. cross section drawings, Swindon November 1937; 6810 Blakemere Grange Swindon factory 15 November 1936; general arrangement drawing and frame plan Swindon August 1936, Number 106700 Lot 308.
Gibson, John C. Great Western locomotive design: a critical appreciation. 1984.
Gibson noted a letter by John A. Trounson in Steam Railway (March 1981) which noted that the Grange class incorporated improved cylinders as compared with the Hall class. The centre line of the pistons and valves were 2½in further apart and level with the axle centres rather than 2½in above. Gibson doubted the authenticity of this statement, but it was verified by A.C. Sterndale and this would give the class a greater steamchest volume than the Hall class.
Hill, John. GWR locomotive matters. Rly. Archive, 2015, (47), 80.
Asserts that original Churchward steam circuit was better than Collett version and this was carried over from 43XX.
Nock, O.S. 4-6-0 locomotives of the G.W.R.: their inception and development. Rly pict., 1947, (2), 70-83. 14 illus., 2 tables.
Railway Corrspondence and Travel Society. The locomotives of the Great Western Railway. Part 8. Modern passenger classes.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 210
Rutherford, Michael Great Western 4-6-0s at work. 1995.
Originally published as two separate books: Castles & Kings at work. Ian Allan, 1982 and Granges and Manors at work. Ian Allan, 1985. The reprint is in effect two separate books (with indexes, etc) under one cover. See page 179 where Rutherford states that Grange class incorporated a new design of cylinder..

Illustrations

The Great Western 'Granges'. Backtrack, 2016, 30, 96-9
Colour photo-feature: No. 6800 Arlington Grange (fully lined green) on 13.55 Penzance to Newton Abbot at Gwinear Road on 15 October 1960 (Roy Patterson); No. 6824 Ashley Grange (fully lined black) on Penzance shed on 20 September 1960 (R.C. Riley); No. 6828 Trellech Grange on westbound freight at Liskeard on 9 July 1961 (R.C. Riley); No. 6811 Cranbourne Grange on empty stock going east at Teignmouth on 24 July 1958 (R.C. Riley); No. 6829 Burmington Grange arriving at Hereford on a south west to north west express in August 1963 (J.L. Champion); No. 6847 Tidmarsh Grange at Swansea High Street on 11 May 1063 (E. Oakley); No. 6851 Hurst Grange on Goring troughs probably with a South Coast to Midlands express formed mainly of green liveried stock on 25 July 1964 (N. Beckett); No. 6841 Marlas Grange on semi-fast formed of non-corridor stock near Twyford in December 1959 (Trevor Owen); No. 6829 Burmington Grange on Paignton to Wolverhampton express in carmine & cream livery at Stoke Canon on 6 July 1957; No. 6856 Stowe Grange at Great Malvern with Hereford portion of 09.15 ex-Paddington on 16 November 1963 (Roy Patterson).

Preservation

Hill, R.F. 'Granges' [Letter]. Gt Western Rly J., 2001, 39, 417.
Writer notes that 6815 Frilford Grange had been selected for preservation as it was in good condition, but that 6998 Burton Agnes Hall was selected as it was cheaper (it contained less non-ferrous metal).

New build: 6880 Bretton Grange

Website: 6880.co.uk
The "edge" that the "Granges" had is undisputed but can only partly be explained in rational engineering terms. The critical difference between the "Granges" and the "Halls" lay in the dimensions of two chambers at the front end of the locomotive's steam circuit, namely the steam chest and the steam ports. Both were larger on the "Granges" than the "Halls". The greater steam port volume is said to give better cushioning of the piston at the end of each power stroke. It is believed that this accounted for the marked absence of fore-and-aft surging on the "Granges", but which is very apparent when riding behind a "Hall" to this day! Despite great efforts to make steam locomotive development a more theoretical and less empirical discipline, it always relied heavily on experimental observation as a means of proving a design. The truth is that while the characteristics of a locomotive could be observed and measured in great detail, it was not always possible to explain why they performed as they did. The legendary free steaming of the "Granges" was just such an enigma. They were fitted with exactly the same boilers as the "Halls" (in fact, boilers were frequently swapped between the two types). The draughting arrangements in the smokebox, critical to a locomotive's steaming characteristics, were identical. Yet the truth is that time and time again the "Granges" steamed noticeably more freely, a quality much appreciated by hard pressed footplate crews. It is hardly surprising that the "Granges" came to be regarded as "The Enginemen's Engine".

78XX Manors: Collett
This was a lighter version of the Grange class. It also incorporated the wheels and valve gear from scrapped 43XX locomotives. To reduce weight, a new boiler (Standard No. 14) was used in place of the Hall type boiler used in the Grange design.

G.W.R. "Manor" class, 4-6-0 mixed traffic engines. Loco. Rly Carr. Wagon Rev., 1938, 44, 66-7. illus., diagr. (s. & f. els.)
4-6-0 mixed-traffic locomotive; Great Western Railway. Engineering, 1938, 145, 189. illus.
"MANOR" class locomotives, G.W.R.. Rly Mag., 1938, 82, 268-9. illus., diagr. (s. el.)
MIXED traffic locomotives for the Great Western Railway. Rly Gaz., 1938, 68, 273. illus., diagr. (s. el.)
NEW G.W.R.4-6-0 locomotives. Engineer, 1938, 165, 191. illus.

Retrospective and critical

Copsey, John. '78xxs' in Traffic. Great Western Rly J., 2,(15) 637-48.
Includes general arrangement drawings and notes changes in names from as originally proposed. Illus.: 7810 Draycott Manor at Leamington Spa in 1947 H.J. Stretton-Ward); 7810 on Banbury to Paddington train in July 1946 (M.W. Earley); 7810 on Swansea service formed of LNER stock near Cheltenham on 24 July 1939; 7813 Freshford Manor assisting 6010 King Charles 1 on Rattery Bank in 1954 (MWE); 7814 Ilford Manor at Cheltenham in December 1950; 7802 Bradley Manor at Shrewsbiry c1952; 7816 Frilsham Manor on Reading shed on 26 June 1964.
Gibson, John C. Great Western locomotive design: a critical appreciation. 1984.
Gibson noted that they were very poor steamers, at least by GWR standards, only capable of producing a good power output for brief periods. It was only at the end of their lives that expert modifications to blastpipe and chimney by Sam Ell produced a remarkable improvement. A shorter, more compact chassis with lighter details would have saved some of the weight, and allowed a smaller reduction of heating surface and grate area. But the groove into which design had subsided was too rigid to allow anything so logical. 'The mixture as before' seemed to be the rule in the drawing office under Collet, so the Manors were virtually the standard 4-6-0 chassis with a consider ably smaller boiler, a case of standardisation carried too far.
A recent inspection of No 7808 Cookham Manor at Didcot, where it is preserved, showed that the Manors used the Grange cylinder castings. They also had the lowered centre line, though bored out to l8in diameter instead of the 18½in. In their case though the trouble was an inadequate boiler, so they would have been unlikely to benefit in the same way. It is a great pity that although about half a dozen of the rather feeble Manors have been preserved, not one of the excellent and well-loved Granges has survived.
Nock, O.S. 4-6-0 locomotives of the G.W.R.: their inception and development. Rly pict., 1947, (2), 70-83. 14 illus., 5 tables.
Railway Corrspondence and Travel Society. The locomotives of the Great Western Railway. Part 8. Modern passenger classes.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 217-19; 224
Rutherford, Michael Great Western 4-6-0s at work. 1995.
Originally published as two separate books: Castles & Kings at work. Ian Allan, 1982 and Granges and Manors at work. Ian Allan, 1985. The reprint is in effect two separate books (with indexes, etc) under one cover. See page 19.

10XX "County": 1945: Hawksworth:
This design introduced many breaks with Churchward's traditions. Neither the boiler nor the driving wheels came within the standard range. The former was somewhat similar to the Stanier 8F type, but was pressed to 280 lb/in². The driving wheels were 6 ft 3 in in diameter. The class was intended for mixed traffic work. The initial locomotive was fitted with a double chimney. Return to beginning.

4-6-0 tender locomotive, Great Western Railway. Engineering, 1945, 160, 147. illus., diagr. (s. el.)
GREAT Western Railway ["County" class 4-6-0] - Railways, 1945, 6, 171. illus.
G.W.R. 4-6-0 locomotive. Engineer, 1945, 180, 109. illus.
GREAT Western Railway: the "1000" class 2-cyl. 4-6-0 locomotives. J. Stephenson Loco. Soc., 1945, 21, 134-6.
The NEW G.W.R. 4-6-0 No.1,000. Rly Gaz., 1945, 83, 1S9.
Editorial comment.
NEW, G.W.R. 4-6-0 locomotive, "1000" class. Rly Gaz., 1945, 83, 168-9. 4 illus., diagr. (s. el.), table.
NEW, G.W.R. 4-6-0 locomotive, "1000" class. Rly Mag., 1945, 91. 342-4. 2 illus., diagr. (s. el.)
NEW, G.W.R. 4-6-0 "1000" class engine. Loco. Rly Carr. Wagon Rev., 1945, 51, 128-9.3 illus., diagr. (s. & f. els.)

Retrospective and critical

Atkins, Philip. Odious comparisons. Steam Wld, 2015, (335), 8-14
Annual mileage; availability, and average engine miles per working day in 1958: 30 locos: 36,485; 65%; 181
Bond, R.C. Organisation and control of locomotive repairs on British Railways. J. Instn. Loco. Engrs., 1953, 43, 175-265.(Paper No.520).
Includes mileage/overhaul (87,588 average annual) statistics for this class.
Bulleid, O.V.S. Railway rolling stock and tendencies in design. Engineering, 1949, 167, 68-71; 94-5; 60. 13 illus., 4 diagrs. (s. els.), 5 tables.
Bulleid showed how little the County design was in advance of Churchward's Star class.
Cook, A.F. Raising steam on the LMS: the evolution of LMS locomotive boilers. Huntingdon: RCTS, 1999. 233pp.
Table 50 (page 217) quotes the cost of classified boiler repairs on a comparitive basis in pence per mile in 1954: 2.7 pence/mile for a Duchess as against 0.8 for an A4 and 0.6 for a Merchant Navy: County boiler repairs cost 4.9 pence per mile and were nearly twice as expensive as the majority of the classes listed.
Copsey, John. The Hawksworth 'Counties' at work. Part 1. — The Company days. Great Western Rly J., 2004, (52), 223-35.
Brief technical dsecription and specification, including a note on the non-standard boiler and driving wheel size, and the double chimney fitted to No. 1000. Originally envisaged as being 99XX series. Initial allocations and workings, mainly into Cornwall.
Copsey, John. Hawksworth 'Counties' at work. Part 2. Great Western Rly J., 2005, (53), 267-84.
Part 1 see Number 52 page 223 et seq. General arrangement drawings of locomotive and tender (side, front & rear elevations and plans).
Copsey, John, The Hawksworth 'Counties' at work Part 3. Great Western Rly J., 2005, (54), 326-43.
Part 2 see page 267. Part 1 see. Covers post-1956 period by which most of the class were based on the periphery: Cornwall, West Wales and at Shrewsbury and Chester, although some were still at Bath Road and worked into London.
Nock. O.S. Fifty years of Western express running. 1954.
Performance
Nock, O.S. 4-6-0 locomotives of the G.W.R.: their inception and development. Rly pict., 1947, (2), 70-83. 14 illus., 5 tables.
Railway Corrspondence and Travel Society. The locomotives of the Great Western Railway. Part 8. Modern passenger classes.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 221 et seq
Rutherford, Michael Great Western 4-6-0s at work. 1995.
Originally published as two separate books: Castles & Kings at work. Ian Allan, 1982 and Granges and Manors at work. Ian Allan, 1985. The reprint is in effect two separate books (with indexes, etc) under one cover. See page 19.
van Riemsdijk, J.T. Compound locomotives: an International survey. 1994.
Any two-cylinder engine, then, but especially any engine with two outside cylinders only, has to have its balancing designed to some sort of compromise. If there is little balance of reciprocating masses, then the pull at the drawbar will fluctuate widely with each rotation of the wheels, once a reasonable speed has been attained. If, on the other hand, a high proportion of the reciprocating masses is balanced, then the result will be a lack of rotative balance and at high running speeds the wheels will pound the track and may even lift clear of it for some part of every revolution. This latter phenomenon was clearly demonstrated in trials of the LMS Class 5 which have been fully reported. It should also be pointed out that this phenomenon of pounding, known as hammer-blow, is a material factor to the bridge engineer. As for the fluctuation in draw bar pull produced by two small a degree of reciprocating balance, it is worth recording that a fluctuation as great as twelve tons per revolution was measured in high speed running by BR 'Britannia' Pacifics and Class 9F 2-10-0s, and by GWR Hawksworth 'County' 4-6-0s. The trouble was apparently cured by changing the coupling between engine and tender and allowing the tender to reduce the effect on the train, but it is none the less clear that large two-cylinder locomotives should not be allowed to run as fast as modern valve gear design permits.
Waters, Laurence. The power of the Counties. Oxford Publishing, 2006. 112pp.
via RCHS Bib. 2006: No. 721. 210 illus. from photographs
Wheeler, Geoffrey. Fired by steam. London: John Murray, 1987.
Includes coloured plate, based on water colour side elevations of No. 1000 County of Middlesex in original condition

Names
G.W.R. Locomotive Mag., 1946, 52, 58
List includes Cheshire rather than Chester

2-6-0

Krugers

Contemporary

Great Western Ry. Locomotive Mag., 1903, 8, 165
Refers to Nos. 2602-2605 as "six-coupled goods engines with leading pony trucks" and No. 33 as the first of the Mogul type and Nos 2622 et seq as being fitted with taper boilers

The Krugers were extremely ugly locomotives and pose classification problems in that the first locomotive was a 4-6-0, but also susequent members of the class were 2-6-0s.
Arman, Brian. The 'Krugers'. Br. Rly J., 1990, 4, 33-40.
These were strange looking locomotives: the first (No. 2601) was a 4-6-0; the second (No. 2602) looked as if it had been intended to be a 4-6-0 but was a 2-6-0; the remaining locomotives were all 2-6-0s. The first had Ramsbottom safety valves; the first two had sand boxes placed on top of the front barrel of the boiler. All had Belpaire fireboxes with a large combustion chamber. Initially the boiler of No. 2601 was pressed to 200 psi, but gradually boiler pressures were lowered. The initial locomotive was subjected to haulage tests on the difficult route from Gloucester to Swindon over Brimscombe Bank. The class was rebuilt or renewed as Aberdare 2-6-0s.
Atkins, Philip. The curse of '03. Backtrack, 220, 24, 508-9.
Discusses several 1903 designs which had very short lives including Krugers (not illustrated)
Lewis, Ivor. The Great Western Railway 'Krugers'. Part One: The background to their design. Backtrack, 2013, 27, 272-80.
This is an odd piece about an odd locomotive design and this part examines the literature about William Dean without noting the new book by Clements, nor Gibson's Great Western locomotive design, nor Hamilton Ellis's biographical study. Jones Highland Goods is mentioned, but Wilson Worsdell's brave ventures are ignored. The article enjoys a brief excursion into technological innovation and the phrase paradigm shift is mentioned. Without question Dean attempted to shift the form of the steam locomotive, but with conspicuous lack of success: his successor, Churchward, was far more successful. Illustrations: Kruger 4-6-0 No. 2601; Churchward with his design team (including James Armstrong Robinson. John Armstrong, H.C. King, F.G. Wright, W.H. Waister, F.W. Marillier and W.H. Williams); 4-6-0 No. 36; Swindon drawing office. p. 278 Hawksworth surrounded by Drawing Office staff allegedly including A.E. Leader, G.H. Pearson, J.W. Cross, G. Burrows, W.H. Pearce, and D.E.F. Deverell. 2-6-0 No. 2602. See also letters on p. 510 from L.F.E. Coombs on Serve tubes, and from Adrian Tester.

Photographs

No. 2610 at Southall shed c1905. Rly Archive, 2014, (45) 75 upper

Aberdare

Contemporary

"Mogul" goods locomotives, Greast Western Ry. Locomotive Mag., 1904, 8, 91.
26XX No. 2674 ilustrated fitted with taper boiler and Belpaire firebox: Nos. 2662-2680. illustration

Retrospective

Arman, Brian. The Dean—Churchward transition. Br Rly J., 1994 (36), 267.
No. 33 acted as the prototype for the Aberdare goods and was fitted with 6½ in piston valves. The last batch (2621-40) were built with piston valves, but were later fitted with slide valves. They were also equipped with steam reversers.
Copsey, John. The 'Aberdare' 2-6-0s. .Br Rly J., (21) 277-90.
Mainly allocations of, and workings by, although author notes Holcroft's assessment of the class (without citing which source).
Le Fleming, Hugh M. International locomotives. Plate 22
Painting of 36XX No. and an Aberdare 2-6-0
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 18 et seq

43XX: Churchward: 1911
Class, introduced in 1911. Holcroft's Locomotive adventure is a key reference on this type as he claimed to be responsible for its inception. It is believed that, like the 45XX 2-6-2T, the class came under a cloud following a derailment in the late 1920s (the full reference has still to be re-found KPJ, but see Cook).

Contemporary

Great Western Ry. Locomotive Mag., 1911, 17, 145.
Three engines, Nos. 4301-4303, of a new class are now at work. They are of the 2-6-0 tender type, with 5-ft. 8-in. coupled wheels; the boilers are of the standard tapered pattern with Belpaire fireboxes, built for a pressure of 200 psi and fitted with the Swindon superheaters; the clackboxes are in the safety valve casings, as is now the practice on the G.W.R., and these engines are supplied with triple sightfeed lubricators of the standard adopted by the railway. The new pattern of tender, with a capacity of 3500 gallons, is provided.

83XX: Collett: 1927/8
RCTS: :in severely curved areas, such as Cornwall, flange wear on the leading coupled wheels was liable to become serious. Weight was therefore transferred to the front end, so that the pony would be forced to impart more side thrust to the main frames and bear a greater share of the flange wear. This was done by moving the buffer beam forward by a foot and fitting a heavy casting behind it. The first engines to be modified were Nos. 4351/86/95/85, which became 8300/35/44/34 respectively in November/December 1927, but these reverted almost immediately. Between January and March 1928 sixty-five engines in the 53XX series were modified, and given the corresponding numbers in the 83XX series, No. 5391 being the first. The full weights were now 11 tons 19 tons 8 cwt. + 17 tons 2 cwt. + 16 tons 10 cwt., total 64 tons — an increase of two tons — which achieved the object, but confined the modified engines to Red routes (fig. J12). By 1944 withdrawal of 4-4-0's capable of running over " Blue " routes left a gap which was filled by restoring surviving 83XX to their original 53XX condition. Twelve of them had been replaced by "Granges" in 1937-39 and fifty-one were re-converted between June 1944 and February 1945, leaving Nos. 8359/93 which were done in February 1946 and September 1948 respectively; the original 53XX numbers were restored.

Contemprary

Great Western Ry. Locomotive Mag., 1928, 34. 24
One of the 2-6-0 engines of the "4300" class has been rebuilt with a larger boiler and modified leading truck and re-numbered 8300. It is being tried in Cornwall. [KPJ: failed to see this paragraph; therefore not in Jones and reference to "larger boiler" incorrect!]

93XX: Collett: 1932:
This was a modernized version of the Churchward 43XX class, introduced in 1911. The 93XX series incorporated heavier buffer beams were fitted which made the class unsuitable for certain routes which could be operated by the 43XX.

NEW mogul mixed traffic engines, Great Western Railway. Loco. Rly Carr. Wagon Rev., 1932, 38, 116. illus.

Retrospective

Clarke, Jeremy. In praise of the moguls. Part one. Backtrack, 2020, 34, 70
A general survey of the 2-6-0 type back to Massey Bromley: implies that 43XX was a good design, but omits Chuchward/Dean Aberdare class!
Cook, K.J. Swindon steam. pp. 28-30
In 1911 the first of the mixed traffic 4300 class, 2-6-0 type with 18½;in x 30in cylinders, 5ft 8in coupled wheels, standard 4 boiler at 200lb pressure and tractive effort of 25,6701b emerged. 322 of these were built during the next ten years. They were what was known as blue engines, carrying a blue disc on the cab sides which indicated an axle loading of between 16 and 18 tons which gave them a good route availability. They became maids of all work and at one time, comprising 7 per cent of the locomotive stock they were running '4 per cent of the locomotive mileage. They extended into 5300, 6300 and 7300 series and much later a further 20 were built. Some were sent overseas for military use in World War I and shortly after the conclusion, thirty-five of them were built by Robert Stephenson and Co.
Cook, K.J. Swindon steam. pp. 72-3
About that time [late 1920s] there were some locomotive derailments. There was one, a serious one, of a River class tank engine on the Southern at Sevenoaks, a King, King William IV No 6002 derailed her bogie when on the down 'Cornish Riviera' at Midgham, No 4508, a 2-6-2 tank engine was derailed near Kidderminster and a 4300 class jumped her leading coupled wheels, the pony wheels remaining on the track, on Menheniot curve between Menheniot and St Germans in Cornwall, but except in the case of the Sevenoaks accident there were no casualties. [KPJ: it would seem that Cook was incorrect about 4508: it was presumably 5508, and the accident was near Bridgnorth. The incident at Menheniot has not been traced and may not have involved a reportable incident].
Copsey, John. '43XXs' in the West. Gt Western Rly J.,2003, 6, (47) 395-406.
Brief notes on Holcroft's involvement in this design following his visit to the USA. Extract from report by Churchward (31 December 1910) stating reasons for design and noting parts in common with 31XX 2-6-2T; extract from report by Inspector G.W. Flewllen of 22 October 1914 of footplate observation of 4316 between Plymouth and Truro which gave a better ride on curves than a Bulldog and greatly increased power for climbing. In the late 1920s excessive wear was noted on the front coupled wheel flanges on the locomotives working in Cornwall and this led to the modification of 65 locomotives with extra weight behind the front buffer beam as the 83XX series. In 1932 side-window cabs were fitted to the 93XX series.
Copsey, John. '43XXs' in the West. Part Two. Gt Western Rly J., 2003, 6 (48). 441-8.
Duties during WW2: less passenger work; war damage to several members of the class including several at Newton Abbot; military freight traffic; allocations in January 1942 and January 1946.
Copsey, John. '43xxs' in the West. Part 3. Gt western Rly J., 2004, 7, (49) 38-49.
Hill, John. GWR locomotive matters. Rly. Archive, 2015, (47), 80.
Asserts that original Churchward steam circuit was better than Collett version and this was carried over from 43XX to Grange design.
Railway Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 105-113
Sanford, D.W. (Paper 451). The relationship between smokebox and boiler proportions. J. Instn Loco. Engrs., 1945, 35, 40-53. Disc. : 53-76. 5 diagrs., 2 tables.
W.H. Hutchinson (62-3) commented upon the very small chimneys fitted to some GWR locomotives, notably the 43XX class - Holcroft replied that these were designed using the Goss formula
Tuplin, W.A. Some questions about the steam locomotive. J. Instn Loco. Engrs., 1953, 43, p.682. (Paper No. 528).
In his response to the very considerable heat generated in the Discussion of Tuplin's paper, Tuplin elected to respond to Holcroft's observations by noting that the 43XX and Hall class shared the same cylinder dimensions, whilst the 2-6-0 had a smaller baller boiler: Tuplin stated: "Yet despite apparent over-cylindering in comparison with the 4-6-0, the 2-6-0 worked very well indeed".
The '43XX' Class —useful Great Western locomotives. Backtrack, 2019, 33, 96-9.
Coloured photo-feature which includes two or three locomotives in smart liveries, including fully lined green and several in filthy condtion but none in Great Western livery. Copsey did not appear to have tackled this class in Great Western Railway Journal

Westinghouse brakes

The Great Western and Westinghouse brakes. Peter Chesson.
See No. 173: see Rutherford Great Western 4-6-0s page 247 illus of 6332 fitted with Westinghouse pump and note that two or three GWR were so fitted at anyone time..

0-6-0

2351: Dean

Clements, Jeremy. William Dean:: the greatest of them all. Southampton: Noodle, 2012.
A "biography" of the Dean Goods 0-6-0, rather than that of the designer, plus some observations on the engineer's other designs.
Hambleton, F.C. Great Western goods engine class 2361. Loco. Rly Carr. Wagon Rev., 1944, 50, 162-3.

2251: Collett: 1930:
These light-weight, modern (if an 0-6-0 at that time could be so-called) locomotives incorporated taper-boilers and side-window cabs. They were intended for working on long, lightly-built routes, especially in Wales. The last two were actually delivered after the formation of British Railways..

New goods engine, G.W.R.. Loco. Rly Carr.Wagon Rev., 1930, 36, 119. illus., diagr. (s. & f. els.)
New 0-6-0 locomotives, Great Western Railway. Rly Mag., 1930, 66, 392-3. 2 illus.

Retrospective and critical

Copsey, John. '2251' class mixed traffic 0-6-0s. Great Western Railway J., 3 (24) . 449-71.
Extremely comprehensive: includes detailed general arrangement drawings (Swindon January 1934 Lot 283 No. 102601. Class had a more restricted route availability than 2301. Correspondence on balancing shows that class suffered same problems as other late 0-6-0s, such as LNER J39, namely to be able to pull harder and travel faster than mechanism could withstand. Led to long correspondence mainly on reversing mechanisms, tenders and workings: David Rowe (26-117) on lever reverse and ROD tenders; from D. Walker (26-118) noting that 2268 never used on Westbury to Salisbury turn; notes 56xx workings from Westbury to Salisbury and 2264 frequently worked Bristol to Reading freight via Devizes. from Philip Atkins (28-239) (comment on letter by Anthony East (26-119) on balance weights (Stroudley system); from R.S. Potts (26-118) noting that Tyseley supplied 2251 to Stratford (for banking to Wilmcote or beyond) for trip working to Long Marston and on passenger workings to Leamington; also records a working to Machynlleth; Maurice Dart notes 2251 workings in Devon (page 119 Number 26). Sightings in Cornwall in 1954 (James Graham 28-239).
Gasson, Harold. Nostalgic days: further reminiscences of a Great Western fireman. 1980.
Waxed lyrical about what he termed the 22XX class: "a baby 'Castle'" (p. 23); "loved by the drivers and fireman"; "they were excellent branch goods engines; superb branch passenger engines and very handy when it came to shunting" (p. 26).
Green, C.C. Cambrian Railways, 1859-1947. London: Ian Allan, 1997. 224pp.
Notted (page 173) initial problems with broken valves and crrank axles...
Railway Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 4. Six-coupled tender engines.
Includes some blue text: it is doubtful "if their appearance caused quite a sensation": albeit that the class was unusual in being the only British 0-6-0 to be designed with a taper boiler. It does note that some locomotives were built with lever reverse, but it did not identify which ones.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 171-2
Rutherford, Michael. Back to basics: Collett's GWR '2251' Class 0-6-0. Backtrack, 2001, 15, 574-9.
Notes on the design of the 2251 class: alternatives included an up-dated Dean goods, a small 2-6-0 using the No. 10 boiler, and an 0-6-0 version of the 56XX 0-6-2T. The 2251 combined the No. 10 boiler with the 57XX chassis to ensure that antiquity prevailed. Rutherford advocates building the 2251 en masse to populate the preserved railways (and presumably British roads).See letter by Summers on page 726.

0-6-0: modifications to pre-grouping types

M. & S.W. Jn.
Ex-Midland & South Western Junction Railway Beyer Peacock 0-6-0 rebuilt as GWR No. 1005 with taper boiler, right hand drive and tender with increased capacity by fitting side sheets.

Rebuilt six-coupled goods engine for the M. & S.W. Jn. Section of the Great Western Ry. Loco. Rly Carr.Wagon Rev., 1926, 32, 174-5. illustration, diagram (side elevation)

4-4-2

De Glehn compounds

A French locomotive for an English railway. The Times, 1903, (5 January), 13.
The newspaper item notes that Société Alsatienne of Belfort was to supply the GWR with a four-cylinder compound locomotive to the design of de Glehn and Du Bousquet for trial. The item makes much of the high speed attained by these locomotives and notes that the North Eastern Railway had sent its chief traffic manager and assistant locomotive engineer to France to investigate these compounds, but concluded that the NER was unsuited to "continuous bursts of high speed",

Rumour of testing against Ivatt 990 class Atlantic.
Great Western Ry. Locomotive Mag, 1903, 9, 285.

Withdrawal of No. 102 La France.
Great Western Ry. . Loco. Rly Carr. Wagon Rev., 1927, 33, 28.

Retrospective

Allen, Cecil J. British Atlantic locomotives, revised & enlarged by G. Freeman Allan.. 1976.
Carpenter, G.W. Discussion on Cook, K.J. The late G.J. Churchward's locomotive development on the Great Western Railway. J. Instn Loco. Engrs, 1950, 40, pp. 202-3
Refered to the De Glehn compounds, and asked whether, in view of the widely accepted idea that it was not possible to accommodate large low-pressure cylinders between the frames of British locomotives without undesirable reductions in axlebox bearing surfaces, any heating trouble had been experienced with the French engines, the last two of which had 235/8in. diameter low-pressure cylinders. He also wondered whether the valve design of the French compounds, which had slide valves, was less efficient than that of the Churchward 4-6-0s, which had long travel piston valves, as the thermal efficiency of the compounds was theoretically higher. Cook replied that he did not consider that there had been any particular heating problems with the French compounds as the pressure was limited in the low pressure cylinders. The engines ran well and were efficient, but did not give a free exhaust when worked heavily. Nevertheless, a drawbar pull of 2 tons was obtained at 70 mile/h.. Carpenter recalled that the efficiency of the similar Nord 4-4-2 compounds was greatly increased in later years when the front end was re-designed and a. multiple jet blast pipe fitted and queried whether further investigations into compounding had been made at Swindon following the Marechal trials between otherwise similar, compound and simple locomotives in 1912 on the PLM Railway. This part failed to illicit a response.
Durrant, A.E. Swindon apprentice. Cheltenham: Runpast, 1989. 216pp.
Includes Swindon drawings Numbers 56767 of Western Railway France compound 4-4-0 sent to Churchward on 17 August 1899 and Number 56770 of an older Nord compound 4-4-0
Rous-Marten, Charles. British locomotive practice and performance. Rly Mag., 1903, 13, 501-8.
Notes on the purchase of the de Glehn compound 4-4-2 La France by the GWR including a record of the performance of the du Bousquet/de Glehn type in France. Also notes that the Great Eastern Railway had locomotives built by the Schneider Co. at its Creusot Works to a Sinclair design.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 38-9; 45-7
Stanier, W.A. discussion on Vallantin, R.G.E. Compound locomotives of the P.L.M. Rly. J. Instn Loco. Engrs., 1931, 21, 283. (Paper No. 274)
In this Country, where we are confined to a width of not more than nine feet across the cylinders, we have to devote a great deal of thought to the design and construction and-what is more important -the maintenance of higher pressure boilers. It is interesting that the French engines which the Great Western purchased in, I think, 1904, had 227lbs. pressure and were not superheated, and as a consequence a great deal af trouble was experienced with condensation in the receivers. They were afterwards fitted with Great Western boilers and superheaters and to a great extent the condensation difficulty disappeared ; but they were still limited, because the nine feet width restricted the size of the cylinders which could be fitted, and as a consequence the power of the engine was much too small for the increased requirements on the Great Western Railway. The Great Western had therefore to develop four-cylinder simple engines
van Riemsdijk, J.T. Compound locomotives: an International survey. 1994.
Considered that Churchward introduced ill-judged modifications, notably the under-sized pipework between the superheater and the high pressure steam chests, possibly as a failure to note that the steam flow was far greater than in his 4-cylinder simples.
Vaughan, Adrian. The heart of the Great Western. Peterborough: Silver Link, 1994.
Relates (pp. 45-9) how Albert King (an Oxford driver) fired the componds: noting the independent controls for the high and low pressure cylinders and how the soft exhaust called for a thin fire.
What our railways are doing. Rly Mag., 1903, 13, 517.
La France (de Glehn compound Atlantic) arrived at Poplar Docks on 19 October 1903 and required thirteen trucks for its journey to Swindon to be assembled.
Wheeler, Geoffrey. Fired by steam. London: John Murray, 1987.
Includes coloured plate, based on water colour side elevations of No. 102 La France in original condition painted black

No. 40 North Star: 1906
The first Churchward four-cylinder locomotive was a 4-4-2 (for comparison with the de Glehn compound 4-4-2s) and thus has to be examined "out-of-sequence".

An "Atlantic" locomotive for the Great Western Railway. The Times, 1904, (1 November), 13.
Announcement that Company was building an Atlantic locomotive to be tried against the French de Glehn compound. The news item makes much play on the speed of Atlantic locomotives, notably the Atlantic City Flyers operated by the Philadelphia and Reading Railroad, and by the run made by the City of Bath on 14 July 1903.

Allen, Cecil J. British Atlantic locomotives, revised & enlarged by G. Freeman Allan.. 1976.
Allen had no experience of No. 40 in its brief existence as an Atlantic.
Clayton, James discussion on p. 371: Holcroft, H. Three-cylinder locomotives. J. Instn Loco. Engrs., 1918, 8, 355-68. Disc.: 368-95; 476-91. (Paper No. 65)
Clayton noted Holcroft's mention of the 4-4-2 engine, the North Star, on the Great Western Railway and that this engine was fitted with the "Deeley" valve-gear. It was a good gear of the "Walschaert" type, but, unfortunately, differing from it hy having the right-hand valve-gear different to the left-hand, and so added considerahlv to the cost without making the gear any better as such, though it does avoid the use of eccentrics. It also makes a hreakdown on one side of the engine a total disablement, as the motion on either side is dependent on the other.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 39-41

Scott class: 1904-12
Russell described the two-cylinder Atlantics as the Scott class: these were built to act as direct comparisons with the French Atlantics. The initial locomotive No. 171 Albion had been built as a 4-6-0 and was converted to an Atlantic.A batch of 12 Atlantics was built as Atlantics in 1905 and given "Scott" type names. They were converted to 4-6-0s in 1912..:

Allen, Cecil J. British Atlantic locomotives, revised & enlarged by G. Freeman Allan.. 1976.
Only one record of good performance is included.
Nock, O.S. British locomotives of the twentieth century. Volume 1. pp87-8
Noted that originally experienced steaming problems which Stanier diagnosed as being due to a failure to redesign the ash pan to meet the demands of the changed rear end. Also noted that the Atlantics did not suffer from slipping on steep gradients.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 39-44

No. 180 without name at Paddington in May 1905, Rly Archive, 2015 (48), 25
Pouteau photograph

4-4-0

Hambleton, F.C. The first GWR express engine. Loco. Rly Carr. Wagon Rev., 1942, 48, ??
1894: No. 8 Gooch

Nock, O.S. Standard gauge Great Western 4-4-0s. Newton Abbot: David & Charles, 1977/8. 2v.
Part 1. Inside-cylinder classes, 1894-1910. 96pp.
Part 2. Counties to the close, 1904-1961. 1978. 96pp.
Typical Nock: mixture of development with performance.

Reohorn, John. Flowers and the City. Backtrack, 2008, 22, 14-18.
The double-framed GWR 4-4-0s produced under Dean and Churchward: from the Armstrong class which was nominally constructed from parts of former broad gauge locomotives and which shared much in common with the contemporary singles (neither type is illustrated); the Duke class with 5ft 8in coupled wheels which Hamilton Ellis labelled "Olde English" in style (see the appropriate volume of Russell for the rich variety of Olde English) through to 6ft 8in Badmintons, 5ft 8in Camels and Bulldogs, the larger driving wheel varieties of Atbaras (Atbara was the name of a Boer War battle), Cities and Flowers. The Birds were clearly not intended to fly as they had smaller driving wheels, and then there were Collett's masterpiece: the Earls, dukedogs, or should they have been Dodos? Earl Cawdor fitted with a large diameter Wilson Worsdell type of boiler is mentioned but not illustrated. Many, but all of the classes received piston valves and were superheated. The exploit of City of Truro is mentioned: see Russell.Pictorial record...v1 Figs 500 and 501. Illus.: Duke No. 3323 Mendip with original round-top boiler; Badnminton class No. 4115 Shrewsbury at Cardiff in 1922 (Ken Nunn) Bulldog No. 3405 Empire of India; Flower class No. 4156 Gardenia at Cardiff in 1922 (Ken Nunn); Atbara No. 3373 Atbara with original boiler; Atbara No. 4148 Singapore with express at Cardiff in 1922 (Ken Nunn); Bulldog No. 3383 at Dawlish on up local train on 2 September 1936 (Ken Nunn); Atbara/City No. 3705 Mauritius in 1903 state; No. 3712 City of Bristol;
Rutherford, Michael. GWR double-framed 4-4-0s. (Railway reflections [No. 39]). Backtrack, 1998, 12, 153-61.
Problems with former Brunel baulk road as had very little resilience. Gooch-style sandwich frames helped as tended to flex. Dean had to face the problem of gauge change. Had relied upon 2-4-0 for express work, but not easy to convert to 4-4-0 or 4-2-2 as slide valves were under cylinder block. The 0-4-4Ts used for express work were criticised in the wake of the Doublebois derailment on 16 April 1895 were converted to 4-4-0s (3521 class). Boiler evolution is described, as is the contribution of F. G. Wright, Chief Draughtsman from 1892 to 1896. Illus.: GW 2-4-0 No 3237; GW No 7 Charles Saunders; No 3274 Cornishman; No 3303 Marlborough; No 3312 Bulldog; No 3310 Waterford; No 3352 Camel; No 3374 which at various times was Baden Powell, Pretoria, Britannia or; Ex badminton No 3294 rebuilt and renumbered as No 4102 Blenheim; Bulldog No 3446 Goldfinch; Locomotives; No 3208 Earl of Plymouth; No 3440 City of Truro;

Dean/Churchward designs

Armin, Brian. The Dean—Churchward transition. Br Rly J., 1994 (36), 267.
There was a gradual evolution: Duke class No. 3312 Bulldog was fitted with a No. 2 boiler with Belpaire firebox; 3310 Waterford was a further advance and 3352 Camel incorporated a drumhead smokebox (1899) and represented the first true Bulldog.

Atbara class
Reohorn explains that the River Atbara was the location of a Boer War battle. According to RCTS Locomotives of the Great Western Railway Part 7 must include Avalon class (vide Locomitive Mag., 1900, 5, 133)

Contemporary

New express locomotives, G.W.R. Locomotive Mag., 1900, 5, 109.
6ft 8in 4-4-0 with Belpaire domeless boiler: No. 3373 Atbara illustrated, but no indication that this was first of Atbara class

Return of the C.I.V.'s. Locomotive Mag., 1900, 5, 188

Great Western Ry. Locomotive Mag., 1903, 9, 270
No. 3387 Roberts, one of the Atbara class, had been overhauled at Swindon and fitted experimentally with a very thick and short copper topped chimney.

The Royal visit to Devonshire. 50-1. Locomotive Mag., 1902, 7, 50. 2 illustrations, 2 tables
Paddington to Kingswear on 7 March 1902 and Plymouth Millbay to Paddington on 10 March 1902. Locomotive No. 3374 Britannia (named for occasion) driven by Burden on both main legs. Camel class No. 3357 Royal Sovereign (specially named) took the Royal Train forward to Plymouth driven by Driver Uren.

Armin, Brian. The Dean—Churchward transition. Br Rly J., 1994 (36), 267.
The Atbara class included the drumhead smokebox, the domeless parallel boiler and straight frames.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 10; 11; 156
Copsey, John. Atbaras and Flowers in traffic. Great Western Rly J., 6, (46) 341-55.
Allocations and workings
Le Fleming, Hugh M. International locomotives. Plate 23
Painting of No. 4172 Gooch

Westinghouse brakes

Treloar, Peter. The Great Western and Westinghouse brakes. Peter Treloar. Steam Wld, 2002 (176), 51.
Westinghouse brakes were fitted for hauling "foreign stock", mainly that from the GER and LBSCR. At least one Atbara: 4138 White so fitted.

Camel class
Similar to Atbara class, but with 5ft 8in coupled wheels.

Contemporary

New G.W.R. locomotives. Locomotive Mag., 1900, 5, 26-7.
No. 3352 (with number on smokebox) and name Camel on cabside illustrated. Features noted were Belpaire domeless boiler, extended smokebox mounted American-fashion on a saddle and 5ft 8in coupled wheels. 1538 ft² heating surface in tubes plus 125ft²in firebox. No. 3332 Avalon was fitted with pop safety valves. Others of the class were Nos. 3333 Brasenose, 3334 Eclipse, 3335 Etona, 3336 Glastonbury and 3337 Kenilworth.

City class: 1901-
This class is interesting in that 3440 City of Truro may have been the first British locomotive to reach 100 mile/h and possibly on the strength of this is preserved as part of the NRM Collection. It is possible to write about it in the present tense as it is still running and is in superb condition, running like a sewing machine.

G.W. Ry. engine "City of Truro" sent to York Railway Museum. Loco. Rly Carr. Wagon Rev., 1931, 37, 112.
No. 3717 formerly No. 3440.
Harris, Michael. City of Truro: a locomotive legend. 1992.
First published 1985: pamphlet
Nock, O.S. Historical steam locomotives. 1959. Chapter 13. The Great Western Cities.
Obviously overtaken by later work.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 15-17
Rutherford, Michael. City of Truro: main line centenarian. York: Friends of the National Railway Museum, 2003. 40pp
Richly illustrated. Centre page spread of detailed general arrangement drawings (side elevation and plan), also weight diagrams and very extensive bibliography. Book also represents a memento for the NRM's team involved with the locomotive's running both in Britain and in the Netherlands: reviewed by Michael Blakemore in Backtrack, 2004, 18, 253.
Shephard, Bob. City of Truro: a livery survey. Modellers' Backtrack, 1994, 3, 326-30.
Wheeler, Geoffrey. Fired by steam. London: John Murray, 1987.
Includes coloured plate, based on water colour side elevations of 3440 City of Truro

Speed record
Andrews, David. Special experimental tests — more pieces of the City of Truro puzzle. Backtrack, 2008, 22, 118-21
Another re-examination of the contemporary publications, both those made in New Zealand (The Evening Post, Wellington, 17 June 1904) and the Bulletin of the International Railway Congress (1905, pp. 2118-21 where on page 2119 it is clearly stated that: "with the advanage of previous knowledge that the experiment was to be made"). Also includes references to George Flewellen via a letter to C.J. Allen (Rly Mag., 1934 Oct.) and correspondence in The Times from John Phillimore on 9 April 1931 and 23 May 1931 where Phillimore records that Flewellen considered Gresley's Pacifics to be "ugly" and the reason for City of Truro coming off the Ocean Mail Special at Bristol was due to bad coal blocking the tube plate. The location of the "permanent way men" or "slack" is also examined. Writer asks where Churchward's quoted instruction "Withhold any attempt at a maximum speed till I give the word — then you can go and break your b— neck" originated other than on page 28 of Nock's Fifty years of Western express running (original quote corrected: Nock did not use word "bloody". Illus. of No. 3440 (all black & white): two in original condition but without indication of date (one is at Westbourne Park); remainder are of preserved locomotive at Bath on 28 April 1957; leaving Nottingham Victoria light engine on 26 August 1959 en route to Scotland for use during Scottish Industries Exhibition and with 4575 No. 5528 crossing Pensford Viaduct on 28 April 1957..
Lee, Pat. Setting the record straight on the perils of delivering a Centenary lecture in celebration of City of Truro achieving 100 mph. J. Rly Canal Hist Soc., 2005, 35, 37-42.
In part inspired by an article by Paul Binyon in The Times on 22 May 2004 that refered to the carriage of gold bullion on the train and the record was unofficial. Lee is able to dismiss both of these statements and claim (by a re-examination of the data from various sources) that a speed in excess of 100 mile/h was achieved. The actual run took place on 9 May 1904 and was an Ocean Mail Special, which was run at high speed to demonstrate that the GWR could compete with Liverpool and with the LSWR for carriage of the trans-Atlantic mail, and to pressage its timetable improvements to the West of England. Rous-Marten sources quoted are The Engineer for 1904 (13 May, 20 May and 10 June), Rly Mag., 1904 June, 1907, December and 1908 April and comment upon these by James Inglis in GWR Mag., 1922 Novemeber. Contemporary newspaper reports for 10 May 1904 consulted included those in the Western Daily Mercury and Western Morning News. Writer claims that much of evidence was reproduced in Tuplin's Great Western saints and sinners. (1971).
Setting the record straight. Stuart Chrystall. J. Rly Canal Hist Soc., 2005, 35, 132-3.
See previous Issue J. Rly Canal Hist. Soc. pp. 37-42: .suggeests that City of Truro did just reach 100 mile/h. Cites O.S. Nock's Speed records.
Setting the record straight. Bill Crosbie-Hill. J. Rly Canal Hist Soc., 2005, 35, 133.
See previous Issue J. Rly Canal Hist. Soc. pp. 37-42. Cites Jim Russell's Pictorial record of Great Western engines (v. 2 p. 12) where he statedd that Rous-Martin [sic] had noted that the Atbara class were fast runners and had published details of one achieving 97.8 mile/h down Dauntsey bank two years before the City of Truro record. Unfortunately, the Rly Mag. citation is lacking as is the writer's own recording of 96 mile/h behind the new County 4-6-0 No. 1005 in April 1946 at the same location (which is mentioned in this letter).
Lewis, John. More light on the 'City of Truro'. Br. Rly J., 1983, 1 (1) 19-20/24.
The high speed run on 9 May 1904: this is mainly concerned with the composition of the train and an estimate of its weight. This also corrects the impression that North German Lloyd Kronprinz Wilhelm may have unloaded bullion. In an attempt ot establish the amount of weight conveyed the following contemporary sources were examined: Western Daily Mercury, 10 May 1904; Western Morning News 10 May 1904, The Engineer 20 May 1904 and 10 June 1904 (reports by Charles Rous-Marten, and Ralway Magazine June 1904 (Charles Rous-Marten). Also cites the account in the Great Western Magazine for November 1922 and an article iby H.G. Kendall in Railway Magazine September 1960 which states that William Kennedy (a letter sorter) may also have timed the train. Lewis also wonders if Rous-Marten caught the special with little time to spare.
Tuplin, W.A. Great Western Saints and sinners (pp. 125-6):
The first reasonably well-authenticated speed of 100 m.p.h. on the GW was that made by double-frame 4-4-0 No. 3440 City of Truro during a reckless record-breaking run from Plymouth to Bristol on 9 May 1904. The adjective 'reckless' is included because the train ran down the helter-skelter from Brent to Totnes in South Devon at an average of 69.6 m.p.h. with a maximum of 77 and down the corkscrew from Dainton to Aller Junction and on to Newton Abbot at 57.6 m.p.h. average. But even this driver had his limit. Down the serpentine stretch from Whiteball to Wellington he got up to about 100 m.p.h. and then his nerve failed. Platelayers working en the line ahead provided a timely excuse for braking , hard at a point where continuance of full regulator opening might have pushed speed up towards 110 m.p.h. if the train had remained on the rails. But the driver evidendy had doubts about this. No published record shows any other train ever to have reached 100 m.p.h. in this vicinity. .
Whitehouse, Patrick and Thomas, David St John. The Great Western Railway: 150 glorious years. pp. 92-3.
A reproduction of the pages from the Railway Magazine for June 1904 and for April 1908. In the earlier Issue Flewellen is recorded as "Llewellyn".

No. 3297 Earl Cawdor

Armin Pouteau at Paddington. Rly Archive, 2015 (48), 24 lower contributed an interesting caption to a Pouteau photograph of No. 3297 Earl Cawdor at Paddington on a Down express sometime in 1903 or early 1904. G.J. Churchward was unquestionably a wise and astute leader of men, so when his chief assistant, FG. Wright, suggested building a boiler which differed from Churchward's rapidly developing practice, he readily agreed. Wright, who may have been influenced by McJntosh's Caledonian Railway practice, suggested a larger boiler barrel with enhanced steam space, attached to a narrower, deeper, round topped firebox, which was cheaper to construct and maintain. In addition, it utilised a deep firebed, which gave greater potential steam raising capacity, because more heat could be produced by the deeper fire in contact with the water circulating around the firebox. In this he seems to have anticipated Whale's practice on the L&NWR and Gresley's continued use of the round top firebox on the GNR and L&NER. see Locomotive Mag., 1903, 9, 174. As rebuilt, No. 3297 should have been able to 'pull a house down' but in fact it proved to be something of a damp squib (tests mentioned in Locomotive Mag., 1903, 9, 241) and spent most of the three years that it ran in this condition on lesser duties such as milk and van trains. The boiler was affixed in July 1903 and removed in October 1906. The NER style cab lasted only sixteen months and was replaced by standard design in November 1904. Earl Cawdor was built in May 1898 and withdrawn in November 1927.

Bulldog class

Contemporary

Bird class: 1909

Great Western Ry. Locomotive Mag., 1909, 15, 105. illustration
A new series of 4-4-0 locomotives, of a slightly modified Bulldog type, had been put in order, five being already built and ten more in course of construction. Known as the Bird class, The chief differences of design from that of the Bulldog class as originally built, consist in providing the standard taper boiler, with an extended smokebox 9-in, longer than in the Bulldog, and in a considerable deepening of the frame plates around the coupled wheels. The boiler has nearly 200 sq. ft. less heating surface, and the grate area is 3-ft. less, but working pressure has been raised to 195 psi.

Robert Reeves. The Great Western Duke and Bulldog 4-4-0s with drawings by K.C. Woodhead. Modellers Backtrack, 1991, 1, 111-16.
Originally known as Pendennis Castle class, later as 3252 class, but later known as Duke class after No. 3252 Duke of Cornwall. Bogies originally fitted with Mansell wheels; boilers originally fitted with steam pumps driven off crosshead. Some were rebuilt with Belpaire fireboxes and some were fitted with larger boilers and became Bulldogs. Finally Bulldog frames were combined with Duke boilers to become Dukedogs. The first Bulldog was No. 3352 Camel (later No. 3340). Frames could be curved: Nos. 3300-40; straight Nos. 3341-3440, and deep straight Nos. 3441-55. Steam reversing gear was fitted, but was replaced by screw reverse from about 1908. Illustrations: No. 3276 Dartmoor (straight nameplate) at Bodmin Road c.1906; No. 3323 Mendip (round top boiler); No. 3330 Vulcan (Belpaire boiler); No. 3260 Mount Edgecumbe (narrow cab) at Gloucester in 1932; No. 3285 Katerfelto (wider cab) at Reading in 1927; No. 3258 The Lizard (short tender) at Reading General; No. 3267 Cornishman at Didcot in February 1936; Duke class: diagram (side & front elevations); Bulldog: diagram (side & front elevations); also 2000 and 2500 gallon tenders: diagrams (side & front elevations and plans); No. 3352 Camel in original condtion; No. 3368 Sir Stafford at Swindon in 1900; No. 3341 Blasius at Reading General; No. 3409 Queensland during WW2; No. 3405 Empire of India at Old Oak Common in 1931; No. 3361 with nameplates removed after May 1927.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 1-4; 8; 76; 157
Wheeler, Geoffrey. Fired by steam. London: John Murray, 1987.
Includes coloured plate, based on water colour side elevations of No. 3403 Trinidad in post WW1 livery of unlined plain black.

Westinghouse brakes
Treloar, Peter. The Great Western and Westinghouse brakes. Peter Treloar. Steam Wld, 2002 (176), 51.
Westinghouse brakes were fitted for hauling "foreign stock", mainly that from the GER and LBSCR. Bulldogs: 3390 Wolverhampton (illustrated with Westinghouse pump); 3394 Albany; 3407 Madras; 3429 and 3434 Joseph Shaw were so fitted.

County class (4-4-0)
An interesting design as most commentators note their bad riding qualities which according to Gibson (p. 57) was due to a combination of the very short wheelbase coupled with the long-stroke outside cylinders.

Cook, K.J. Swindon steam. p. 30
Perhaps the least successful of his [Churchward's] standard types was the County class, 4-4-0, 18in x 30in cylinders, 6ft 8½;in coupled wheels, standard boiler No 4 at 200lb pressure, which rather became known as Churchward's 'Rough Riders'. Originally the first ten were numbered 3473 to 3482, later renumbered 3830 to 3839 as in the meantime thirty more had been built and become the 3800 class numbered 3800 to 3829. They were designed for working the west to north trains via Severn Tunnel over the joint LNW-GW metals to Shrewsbury. They tended to develop considerable hammering in the left hand trailing axleboxes, caused by the amount of counter-balance to the reciprocating parts of the motion which had to be concentrated in the four wheels. The short rigid wheelbase was also a factor. When in good condition they did their work excellently and I found them very sweet in riding on the curves from Maindee Junction Newport to Pontypool Road and beyond, but after 40,000 miles or so they were liable to need works attention. But they were also, for many years, on the GWR Bristol-Birmingham trains, which ran over the Midland line between Yate and Standish Junction, the largest GW engines permitted to use this track.
Fullagar, L.A. Comparison of slide bar pressures. Trans Instn Loco. Engrs., 1915, 5, 1-10. + 4 plates (diagrs.) (Paper No. 34)
The engine types selected were the GWR County class and the Great Northern large Atlantics. The former, with other of the Great Western 30in. stroke classes, were those whose oscillation at starting was noticed. The latter, which represent the other extreme in connecting rod and stroke ratio, are amongst the steadiest running engines known to the Author. For the purpose of comparison an ipdicator diagram was taken belonging to what may be termed a neutral engine of not dissimilar dimensions, LTSR 4-4-2 tank of the No. 80 class, now MR Nos. 2176-9.
Gibson, John C. Great Western locomotive design: a critical appreciation. . Newton Abbot: David & Charles, 1984.
See also page 75 where the limitations of the boiler are noted.
Nock, O.S. Fifty years of Western express running. 1954.
Pp 150-6 Nock lists some fine runs made during 1917 between Hereford and Shrewsbury and the reverse by E.L. Diamond whilst he was at school in Colwyn Bay and travelling to and from his parents in South Wales.
Nock, O.S.Standard gauge Great Western 4-4-0s. Newton Abbot: David & Charles, 1977/8. 2v.
Part 2. Counties to the close, 1904-1961. 1978. 96pp.
Typical Nock: mixture of development with performance.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 64-70
Stanier, W.A. [discussion on] Cook, K.J. The late G.J. Churchward's locomotive development on the Great Western Railway. J. Instn Loco. Engrs, 1950, 40, 131-71. Disc.: 171-210. (Paper No.492)
Mentioning the Churchward County class, Sir William said: "Churchward had built that engine with his tongue in his cheek. He knew that the front end was too powerful for the wheel base. This engine was built for working trains on the Shrewsbury and Herford line, which was a joint line with the L. & N.W., and the L. & N.W. objected at that time to the 4-6-0 "Saint" class working over it. He was not going to be told what to do by Webb! Therefore Churchward built the "County", which had plenty of power to run the service.
van Riemsdijk, J.T. Compound locomotives: an International survey. 1994.
Perhaps the best known example of this phenomenon [nosing from side to side] was the Churchward 'County' 4-4-0, which, with a long piston stroke and a short rigid wheel base, was a particularly rough rider.... but it is none the less clear that large two-cylinder locomotives should not be allowed to run as fast as modern valve gear design permits.
Vaughan, Adrian. The heart of the Great Western. Peterborough: Silver Link, 1994.
Relates (pp. 45-9) how Albert King (an Oxford driver) found the classs to be rough riding.

3265 Tre Pol and Pen
Green Cambrian Railways (p. 165) states that idea of combination of Bulldog frames with Duke class boiler was due to K.J. Cooke: it led to the cutting edge Earls

32XX "Earls" [later 90XX ("Dukedogs")] Collett :1936:
This class was constructed from withdrawn "Bulldog" class frames (double type) and "Duke" type boilers. The "Bulldog" and "Duke" classes had been introduced, by Dean, in 1899 and 1895 respectively. The parts utilized in the 32XX locomotives came from withdrawn engines. Bonavia called them a curious throwback to a relatively antique design. The new type was intended for use on lightly-built routes in the West Country and in Wales. Return to beginning..

"EARL" class 4-4-0 locomotives, Great Western Railway. Loco. Rly Carr. Wagon Rev., 1936, 42, 232. illus.
No. 3200 Earl of Mount Edgcumbe illustrated
LIGHT 4-4-0 locomotives, G.W.R.. Rly Mag., 1936, 79, 102. illus.
LIGHT 4-4-0 passenger locomotives, G.W.R.. Rly Gaz., 1936, 65, 39; 60. illus.
Editorial comment.

Retrospective and critical

Copsey, John. The 'Earls' at work. Great Western Rly J., 2 (9), 388-92.
Class emerged in 1936, and is probably better known as the 90xx class (renumbered in 1946). Allocations, workings, early withdrawals. Illustrated: 3212 Earl of Eldon at Swindon on 8 May 1937; 3201 without name. 9021 at Machynlleth in early 1950s
Davies, F.K. The G.W.R. "3200" class 4-4-0's. Rly Obsr, 1938, 10, 399-402. 2 illus.
Ebenezer, T.J. Differing clack valve positions on 'Dukedog' 4-4-0 rebuilds. Steam Wld, 2006, (233) 20.
Illus. of 9009 with clack valve near smokebox at Swindon in August 1956 and of 9014 with clack valves on top of boiler (at Shrewsbury).
General Arrangement of 'Earl' Class 4-4-0. Great Western Rly J., 1994, 2 (10) 418-20.
Dated January 1940.
Green, C.C. Cambrian Railways, 1859-1947. London: Ian Allan, 1997. 224pp.
Pp 190-1 show the nameless Earls Nos. 3210 and 3208 performing Royal duty of hauling LNWR Royal Train on 15 July 1937 between Dovey Junction and Afon Wen during Coronation Tour, following the opening of the National Library of Wales in Aberystwyth. The leading former Earl is carrying the Royal decoration carried by Great Western locomotives.
Nock, O.S. Standard gauge Great Western 4-4-0s. Newton Abbot: David & Charles, 1977/8. 2v.Part 1. Inside-cylinder classes, 1894-1910. 96pp.
Typical Nock: mixture of development with performance
Pearse, John. Nameplates, curves and vital statistics. Steam Wld, 2005, (222) 54-9.
The essential element was the transfer of the Earl names from 90XX (32XX) Nos. 3200-3212 (and the non-use of those intended for 3213-3219) to Castle class 5043 to 5063. The writer questions why the names were not moved to the Grange class, but there seems to be a consensus that only the Castle class was prestigeous enough for the name of an earl. Pearse notes the mismatch in the radius of the plates. Cites G.W. Echo No. 158 (Peter Treloar); No. 159 (Peter Timms) and No. 161 (Michael Hale); Frank Burridge's Nameplates of the Big Four and J.H. Russell's A pictorial record of Great Western engines. Volume 2. Colour illus. of nameplates of Earl of St Germans, Earl of Dudley, Eral of Dartmouth, Earl Waldegrave and Earl of Mount Edgcumbe. No photographs of nameplates in situ on 32XX
Railway Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 7. Dean's larger tender engines. 1954.
Reohorn, John. Twilight of the Dogs: development and use of the Great Western mixed traffic 4-4-0s. Backtrack, 2007, 21, 302-10.
The title introduces yet another soubriquet for the Dukedog, Earl or more correctly 90XX or 32XX class which emerged from Swindon in 1936 and consisted of Duke boilers married to Bulldog double-frames. K.J. Cooke is honoured with this major innovation. A table lists the locomotives which were melded in this way. The article also outlines the development of the two types of GWR 5ft 8in 4-4-0s which featured double-frames: the Duke class and the Camel/Bird/Bulldog class on which Churchward developed his boiler from domeless Belpaire through to coned B4 standard. The use of both the Duke class and the 90XX on the Cambrian lines is also considered. Colour illus: 9018 leading 2251 No. 2268 leaving Barmouth with express for Pwllheli in August 1958 (P.H. Wells); 9017 at Welshpool in 1956 (P.B. Whitehouse); 9018 as pilot to 6330 leaving Barmouth in 1958 (P.H. Wells). Black & white: Duke 3272 Amyas in original condition; Duke 3286 Meteor with Belpaire boiler near Southcote Junction with train for Basingstoke on 8 July 1932; Bulldog 3340 Marazion (domeless parallel Belpaire boiler) calls at Brent with 16.10 Newton Abbot to Plymouth on 23 July 1910 (Ken Nunn); 3409 Queensland with superheated D3 boiler; 3265 Tre, Pol and Pen as rebuilt with Bulldog frames and prototype; Duke 3271 Eddystone at Welshpool on 6 April 1926 (viewed from above boiler has a remarkably French look) (Ken Nunn); 9002 passing Buttington Junction withh down express on 29 February 1952; 9003 with down express near Aberdovey on 15 August 1953 (Eric S. Russell).
Riley, R.C. The Great Western "Dukedogs". Rly Wld, 1963, 24, 150-6. 12 illus.
Opening photograph Nos. 3214 and 3213 climbing Talerdigg bank on 9 April 1939 (J.G. Dewing). Text notes which locomotives were fitted with new or reconditioned boilers when "new". Not photo-journalism, but excellent resume of class with interesting photographs
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. page 148
Tipper, D.A. The Great Western "9000" class. Rly Wld, 1953, 14,175-9.5 illus., 3 tables.
Webster, V.R. The Great Western "Dukes". Railways, 1944, 5, 171-3; 185-8. 4 illus.
Includes the prototype Duke rebuild (Tre Pol and Pen) and the 32XX class.
Wood, J.D.. The GWR Dukedogs. BackTrack, 4, 221-5.
32XX, subsequently 90XX 4-4-0s: notes on names and workings. Illus: Dukedog No 9000 at Andover Junction on 14 June 1953; 9000 at Reading West on same day (railtour); 9005 at Welshpool on 12 September 1953; 9011 at Southall shed on 10 August 1952; 9014 at Portmadoc on 10 June 1954; 9011 in Swindon shed on 16 June 1957; 9014 at Afon Wen on 10 June 1954; 9015 at Witney on freight on 8 June 1954; 9021 in Stafford Road Works Wolverhampton on 19 September 1954 (locomotive in pieces); 9025 at Swindon waiting to be cut up writer noted tapered chimney but did not mention top-feed on boiler) (16 June 1957); 9027 passing 9010 at Oswestry on 9 July 1952.

4-4-0: modifications to absorbed stock.

M & SWJR

No.1121 (G.W.R. number):
Tyrrell 4-4-0 rebuilt with Swindon taper-boiler and right-hand drive.

Rebuilt passenger engine for the M. & S.W.J. Section of the G.W.R.. Loco. Rly Carr. Wagon Rev., 1924, 30, 331. illus.,diagr. (s.el.)

Retrospective and critical

Davies, F.K. The Midland and South Western Junction 4-4-0 locomotives. Rly Obsr.,
1940, 12, 1-3.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 10. Absorbed engines, 1922-1947. 1966.

2-2-2

No. 9: Dean: 1884

According to Ahrons (p. 264) this locomotive was constructed from parts (wheels and motion) from a 4-2-4T of 1881. The locomotive had 7ft 8in driving wheels, 18in x 24in cylinders actuated by rocking shafts from outside Stephenson link motion. It was rebuilt again in 1890 with 7ft driving wheels and outside bearings.

No. 10: Dean: 1886

This was similar to No. 9 and worked the same heavy trains between Swindon and Paddington, but had direct Stephenson link motion, double frames throughout and outide bearings. It was also rebuilt with smaller driving wheels in 1890.

3021 class mentioned by Fryer. One of these No. 3021 Wigmore Castle derailed in Box Tunnel and was rebuilt as a 4-2-2 in 1894.

4-2-2

3001 Class

Created by reconstruction of 3021 and remainder of class, plus a further fifty constructed from new.

Contemporary

No. 3027 Worcester: derived from broad gauge 2-2-2, formerly Thames; name changed to Worcester in 1892; rebuilt with Camel class boiler with Belpaire firebox in 1900.
Single express locomotive, G.W.R. Locomotive Mag., 1901, 6, 28 illus

As late as 1911 some of the class were receiving new Belpaire boilers
Locomotive Mag., 1911, 17, 146

Fryer, Charles. Single wheeler locomotives. 1993. Chapter 8.
"Masterpieces by accident"
Hambleton, F.C. The first 4-2-2 express loco. Loco. Rly Carr. Wagon Rev., 1944, 50, 47.
3021 Wigmore Castle: Dean rebuild of 2-2-2.

Nock, O.S. Historical steam locomotives. 1959. Chapter 6 Four famous 4-2-2 singles.
Included herein

Tank engines

2-8-2T

72XX: Collett:1934:
Many of the Churchward 52XX 2-8-OTs were redundant in the 1930's due to the recession in the coal trade. To increase the working range of the locomotives, Collett rebuilt them with larger bunkers. This necessitated the rearward extension and conversion to the 2-8-2T type.

CURRENT locomotive conversions. Rly Engr, 1934, 55, 332..
The class is incorrectly referred to as a 4-8-4T.
GW.R.heavy tank engines. Engineer, 1934, 158, 522. illus.
REBUILT eight-coupled tank locomotives, G.W.R.. Rly Mag., 1934, 75, 453. Illus.
2-8-2 heavy goods tank locomotive, "7200" class, G.W. Rly.. Loco. Rly Carr. Wagon Rev., 1934, 40, 330-1. illus., diagr. (s. & f. els.)
2-8-2 type heavy goods tank locomotive, G.W.R.. Engineering, 1934, 138, 482. illus.

1945 : Hawksworth : conversion to oil fuel.

G.W.R.oil burning locomotives. Loco. Rly Carr. Wagon Rev., 1945, 51, 178-9. 2 illus.

Retrospective and critical

Copsey, John. '72xx' goods tanks. Great Western Rly J., 1996, 3 (18) 97-111.
In 1930 a batch of 42XX 2-8-0Ts (Nos. 5275-94) had been constructed and placed into store: in 1934 these were converted into 2-8-2Ts with large bunkers to accommodate extra coal and water. Further convrsions took place in 1936 and between 1937 and 1939. Locomotive allocations and workings are described.
The Great Western's eight-coupled tanks. Locomotive Mag., 2017, 31, 672-5.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes. 1962.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 181-2

2-8-0T

42XX: 1910-1930
Many subsequently rebuilt as 2-8-2Ts (72XX)

Great Western Ry. Loco. Rly Carr. Wagon Rev., 1911, 17, 25-6. illustration, diagram (side elevation)
No. 4201 illustrated

Copsey, John. '42XXs' in traffic. Part 1. Great Western Rly J., 2009, 9 (69) 277-97; (70) 339-52.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes. 1962.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 92-5

2-6-2T

No. 99
Introduced by Churchward in late 1904 as one of initial three standard designs (the others being the 4-6-0 and 2-8-0)
The Loco. Mag., 1903, 9, 370 provided a photograph of the new ten-wheeled "double-ender" tank locomotive No. 99 (Swindon works No, 1992), designed for heavy suburban traffic. "It is in reality an enlarged edition of the eight-wheelers of the No. 11 type (2-4-2T), with outside cylinders instead of the usual inside arrangement, and six wheels coupled instead of only four" (KPJ a wild understatement). The cylinders were 18-in. by 30-in., and were provided with piston valves; the boiler, which carried a pressure of 195 lbs. per sq. in. and had its centre 7-ft. 11¾in. above the rails, was of the taper pattern with Belpaire firebox, similar to those of the "City" class; and the six-coupled wheels were 5-ft. 8-in. in diameter; there was a pony-truck at the leading end, and the trailing wheels were provided with radial axleboxes.

Copsey, John. The '51XXs' at work. Gt Western Rly J., 2013, 11, 302-19.

Photographs

No. 99 with Kruger No. 2610 at Southall shed c1905. Rly Archive, 2014, (45) 75 upper

31XX: Collett:1938
Parts, including the frames, of withdrawn 3150 class locomotives were used in the construction of this class of five engines. 5 ft 3 in driving wheels were fitted. The engines were used for banking work.

Davies, F.K. G.W.R. 2-6-2T reconstructions. Rly Obsr, 1939, 11, 35-8.
NEW series tank locomotives, G.W.R.: reconstructions of two 2-6-2 classes. Rly Gaz., 1939, 70, 512-13. 2 illus., 2 diagrs. (s.els.)
RECONSTRUCTED "Prairie" tank locomotives, G.W.R.. Rly Mag., 1939, 84, 402. 2 illus., table.
2-6-2 tank engines, G.W.R.. Loco. Rly Carr. Wagon Rev., 1939, 45, 82. 2 diagrs. (s. & f. els.)
Future numbers given as 3100-3140

Retrospective and critical

Firth, M. The G.W.R. 2-6-2T's. Rly Obsr, 1944, 15, 2-3; 15-16.
Hodge, John. 3100 Class. Gt Western Rly J., 2013, 11, 360
Outlines relative failure of this operation, but letter fails to give date
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes. 1962.

44XX: Churchward: 1904:
This class was introduced by Churchward in 1904. In 1932, No. 4402 was fitted with a Westinghouse pump for spraying oil onto the flanges. This was intended to reduce flange wear on the sinuous Princetown branch. Un-titled reference : Rly Mag., 1932, 70, 386. The class was closely associated with a small number of very teeply graded lines: notably the Princetown branch and the line from Wellington to Craven Arms.

Copsey, John. The '44XXs' in traffic. Great Western Rly J.., 2011, 10, 449-59.
Alan Hall. The evolution of the 44XX class. Great Western Rly J.., 2011, 11, 46; 58-60.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes. 1962.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 137-40

45XX: Churchward :1906/4575 : Collett :1927 :
Collett modified the side-tanks of this design, when introducing the 4575 batch in 1927. No contemporary reference had been traced of this minor alteration, but.... One of the class was involved in a derailment at Bridgnorth and the Inspecting Officer (A.H.L. Mount) suggested that the flanges should be deepened.

Great Western Ry. Locomotive Mag., 1928, 34, 262.
Refers to Nos. 5531-5 as "latest 2-6-2 goods tanks"

Beale, Gerry. 4555 — a GWR preservation pioneer. Great Western Rly J., 1995, 2, 689-98.
General arrangement drawing (side and front elevations/cross sections: Lot 226 (4555-74, Swindon November 1923). The "career" of 4555, with its allocations, etc until it was selected for preservation. Illustrations: 4564 at Newton Abbot on 29 June 1926; 4566 in September 1932; 4560 at Worcester in 1939; 4555 near Friog avalanche shelter on 28 July 1951 with Manchester to Pwllheli train; at Machynlleth in mid-1950s; at Coryton on 23 June 1962; and at Kemble on pre-purchase (for preservation) test run with Reggie Hanks, Jack Hancock and Mr Ridgway.
Cook, K.J. Swindon steam. pp. 72-3
About that time [late 1920s] there were some locomotive derailments. There was one, a serious one, of a River class tank engine on the Southern at Sevenoaks, a King, King William IV No 6002 derailed her bogie when on the down 'Cornish Riviera' at Midgham, No 4508, a 2-6-2 tank engine was derailed near Kidderminster and a 4300 class jumped her leading coupled wheels, the pony wheels remaining on the track, on Menheniot curve between Menheniot and St Germans in Cornwall, but except in the case of the Sevenoaks accident there were no casualties. [KPJ: it would seem that Cook was incorrect about 4508: it was presumably 5508, and the accident was near Bridgnorth. The incident at Menheniot has not been traced and may not have involved a reportable incident].
Davies, F.K. G.W.R. locomotives — detail alterations. Rly Obsr, 1934, 6, 165-6.
Firth, M. The G.W.R. 2-6-2T's. Rly Obsr, 1944, 15, 2-3; 15-16.
Flanges of locomotive tyres: deepening of the commencing size on the Great Western Railway for the purpose of minimising the risk of derail ments and lessening the side-cutting of rails. Rly Engr, 1929, 50, 90-3. 7 diagrs.
See below.
Lowe, James W. British steam locomotive builders. 1975.
Lowe consideres taht both the 44XX and 45XX were highly succesful and the locomotives were capable of rapid acceleration
Mount, A.H.L. Report on the derailment of a passenger train that occurred on the 13th January, 1928, between Linley and Bridgnorth. Ministry of Transport Railway accidents.., which occurred during the three months ending 31st March, 1928. London, H.M.S.O., 1928. 10 p.
The derailment was caused by excessive speed on poor track. Mount suggested that deeper flanges might have prevented the accident. According to report (available on Railway Archive website) locomotive was No. 5508.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes. 1962.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 141-6

Operation
Empty stock at Paddington: Locomotive Mag., 1927, 33, 180

5101 : Collett :1929:
This was a modernized version of the 31XX class, which had been introduced in 1903. Similar to 31XX but outside steam pipes were provided, the position of the sandboxes at the trailing end was altered, and the footsteps at the leading end were modified. No. 5101, the first of the series, left the shops at the beginning of October. A refinement was the oil box near the small end of the connecting rod. .

GREAT Western Railway new tank engines. Loco. Rly Carr. Wagon Rev., 1930, 36, 37. illus.
No. 5105 illustrated.
GREAT Western Ry.: 2-6-2 side-tank locomotive. Loco. Rly Carr. Wagon Rev., 1931, 37, 109 + folding plate. diagr. (s. el.), plan.

Retrospective and critical

Firth, M. The G.W.R.2-6-2T's. Rly Obsr, 1944,15, 2-3,15-16.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes. 1962.

61XX: Collett: 1931:
This was a high boiler pressure (225 lb/in²) version of the 5101 class. It was built for the Paddington suburban services. The 61xx had beern developed from No. 99 of 1903 which was the prototype for the 31xx class of large 2-6-2Ts. The higher boiler pressure gave rapid acceleration to the short suburban trains. .

Brewer, F.W. Great Western locomotives in 1931. Rly Mag., 1932, 71, 115-21.illus., diagr. (s. & f. els.), 6 tables.

Retrospective and critical

Chadwick, John. One good [firing] turn. Gt Western Rly J., 2005, 7, 439-42.
Based at Oxford and used to footplate work on locomotives of all the pre-nationalization railways. Comment on difficulties of preparing 61XX 2-6-2Ts due to extreme lack of space in cab.
Copsey, John. The '61xxs' in the Great Western era . Great Western Rly J., 1993, 1 (5), 178-88.
See also letters on page 264 from Roy Williams who was a fireman at Slough, claimed that class was capable of very high speed from J.E. Norris use of 61XX type plus single autocar as replacement for diesel railcar working between Oxford and Hereford during period 1938/9; and fromGeoff Goslin who queried why all the large Prairies were fitted with lever reverse in spite of clear evidence that 61XX were used on fast workings.
Firth, M. The G.W.R.2-6-2T's. Rly Obsr, 1944,15, 2-3,15-16.
Gardner, N.E.. 61XXs. Great Western Rly J., 1993, 1 (8), 350.
No. 6116 retained its smaller (5ft 3in) driving wheels in period 1938-55 when writer was employed at Slough. It was not a popular engine, due to its increased coal and water consumption.
Harvey, D.W. Bill Harvey's 60 years of steam. 1986.
Page 163: In 1949/50 Harvey enjoyed travelling on the 61XX Nos. 6129 and 6166 allocated to Princes Risborough which worked into Marylebone and commended their low water consumption.
Pugh, B.E. Memories of '61xxs'. Great Western Rly J., 5, 292-5.
On 13 October 1952 the writer was promoted to be a fireman at Slough from Swansea East Dock where he had been a cleaner. Describes a typical day with the class. Notes the lack of steps to the bunker. The trip cocks provided to access the London Transport platforms at Paddington were liable to be actuated by heaps of rubbish in the shed yard. Illus.: Slough mpd in 1940s; 6123 in BR days; 6136 on Slough shed on 8 June 1952.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes. 1962.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 198-200.

81XX: 1938: Collett:
The frames from withdrawn Churchward 51XX class locomotives were used as the basis for this class, but new boilers and front-ends were fitted. The class was intended for surburban duties.

Davies, F.K. G.W.R. 2-6-2T reconstructions. Rly Obsr., 1939,11, 35-8.
NEW series tank locomotives, G.W.R.: reconstruction of two 2-6-2 classes. Rly Gaz., 1939, 70, 512-13.2 illus., 2 diagrs. (s. els.)
RECONSTRUCTED "Prairie" tank locomotives, G.W.R. Rly Mag., 1939, 84, 402. 2 illus., table.
2-6-2 tank engines, G.W.R. Loco. Rly Carr. Wagon Rev., 1939, 45, 82. 2 diagrs. (s. & f. el.s)
Nos. 8100-8149 allocated

Retrospective and critical

Bond, R.C. Organisation and control of locomotive repairs on British Railways. J. Instn. Loco. Engrs., 1953, 43, 175-265.(Paper No.520).
Includes mileage/overhaul statistics (71,720 average annual) for this class.
Firth, M. The G.W.R.2-6-2T's. Rly Obsr, 1944,15, 2-3,15-16.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes. 1962.

Vale of Rheidol: 1924: Collett:
The two locomotives built for this 1 ft 11½ in gauge Welsh railway were a development of a Davies & Metcalfe 1902 design.

LOCOMOTIVE for the Vale of Rheidol Railway: designed and built at the Swindon Works of the Great Western Railway. Engineer, 1924, 137, 210. illus.
NEW narrow gauge tank engines, Great Western Railway. Loco. Rly Carr. Wagon Rev., 1923, 29, 306. illus.

Retrospective and critical

Boyd, J.I.C. Narrow gauge rails in Mid-Wales: a historical survey of the narrow gauge railways in Mid-Wales. 1952.
Davies, W.J.K Vale of Rheidol Light Railway. London, Ian Allan, 1964. 56 p. 47 illus., diagr., 16 tables. 10 plans, map. Bibliog.
Includes a concise guide to the motive power.
Jones, R.B. British narrow gauge railways. 1958.
Rutherford, Michael. Some reflections on the narrow gauge. Part 1.. (Railway Reflections No.129). Backtrack, 2007, 21, 242-9.
Captions (page 244) note that Walschaerts valve gear and cylinders based on Swindon steam railcars/railmotors and that the former Cambrian Railways' 2-6-2Ts "rebuilt" at Swindon were "replaced" at Swindon

2-6-2T: modifications to absorbed stock.

Alexandra (Newport & South Wales) Docks & Railway.

No. 1204 (G.W.R. number): c 1924: Collett:
This locomotive was originally built for the Mersey Railway, but was acquired by the A.D.R. in 1904. It was fitted with a taper-boiler by the GWR.

GREAT Western Railway: an interesting locomotive rebuild. Loco. Rly Carr. Wagon Rev., 1932, 38, 88-9. illus.
RCTS. The locomotives of the Great Western Railway. Part 10. Absorbed engines, 1922-1947. 1966.

0-6-4T: modifications to absorbed stock.

Barry Railway

L: 1924: Collett:
Some locomotives were rebuilt with 42XX boilers.

REBUILT 0-6-4 tank locomotive, Great Western Railway. Loco. Rly Carr. Wagon Rev., 1923, 29, 320. illus.

Retrospective and critical

Mountford, E.R. The Barry Railway class "L" 0-6-4T's. Rly Obsr, 1943, 14, 191-2.
RCTS. The locomotives of the Great Western Railway. Part 10. Absorbed engines, 1922-1947. 1966.

0-6-2T

56XX. 1924: Collett:
Most of the absorbed lines in South Wales employed 0-6-2T engines. To comply with this tradition the S6XX class was designed. It utilized many Swindon standard parts.

NEW 0-6-2 tank engines, Great Western Railway. Rly Mag., 1925, 56. 116. illus.
NEW tank locomotives for mineral traffic, Great Western Ry. Loco. Rly Carr. Wagon Rev., 1925, 31, 2-3. illus.diagr. (s. el.)
0-6-2 tank locomotives, Great Western Ry. Loco. Rly Carr. Wagon Rev., 1928, 34, 380-1. illus., diagr. (s. el.)
6650 series supplied by Sir W.G. Armstrong Whitworth & Co. Ltd..

Spit and polish. W.J. Reynolds. Rly Wld, 1957, 18, 311
56XX No. 6664 photographed at Swindon painted in lined Brunswick green

Retrospective and critical

Cook, K.J. Swindon steam. p. 61-2
During this period a rather unusual type of locomotive for Swindon production came into the picture in the form of an 0-6-2 tank engine, 4ft 7½in coupled wheels, 18in x 26in inside cylinders incorporating standard boiler No 2 at 200lb pressure and giving a tractive effort of 25,800lb, the 5600 class.
This type of locomotive was prevalent on practically all the constituent and absorbed railways in South Wales and it was considered necessary to build a powerful standard engine of this type to meet the locomotive shortage of that time. The design and construction were pressed forward hand in hand with the first of the Castles and in all a total of 200 were built, partly at Swindon and partly by contractors, a notable achievement by the latter being on an order of fifty placed on Armstrong-Whitworth at Scotswood on the Tyne who delivered steadily at five engines per week at quite a quick delivery. They worried us persistently to give them drop forgings of articles for which we had dies, but they certainly made a very good production.
Restricted clearances and curves made it necessary to fit inside cylinders and precluded the use of large outside cylinders, and the trailing pair of carrying wheels enabled a large bunker to be provided. These engines worked up and down the Welsh valleys, chiefly between collieries and ports with shipment coal but also on passenger trains, working chimney first up the valleys and the guiding wheels in front when running down bunker first increased their stability at the higher speeds.
In the design of all inside cylinder locomotives the crank axle restricts the length of journal and therefore the bearing surface of the driving axleboxes and with a powerful locomotive this is apt to cause trouble from time to time which was so on the 5600 class and some experiments were made on the crank arrangements and balancing. In general it was a very successful engine but a peculiar phenomen occurred, only on some engines, of what was very unusual on GW locomotives, namely that in their very early days they became out in their beats and the valves had to be reset. This having been done there was no more trouble. They had 8in piston valves above the cylinders pitched somewhat high and it appeared that on these particular engines there were some residual stresses in the cylinder castings which on their early working and heating up expanded the castings upward to release the stresses. After resetting the valves they remained correct. GW locomotives had extreme regularity in their exhaust beats and we certainly could not permit of anything like two beats and a wooffle which was noticeable on some rival lines.
These engines had a new design of three-bar motion crossheads one large bar at the top and two smaller ones at the side a little lower with the gudgeon pin suspended below. Space above the leading coupled axle of such engines in which to support the slide bars is restricted and the former general arrangement of four bars was not very satisfactory but the new design was a great improvement. A smaller version of this crosshead was used a few years later on new small standard tank engines.
Davies, F.K. Great Western Railway: notes on the 56XX and 66XX classes, 0-6-2T. Rly Obsr, 1935, 7, 133-4.
Gibson, John C. Great Western locomotive design. 1984
Author made a very serious allegation concerning fundamental fault in the original design of valve gear for this class which reached the productiuon stage and was only found when first locomotive steamed: see pp. 131-3:
Nock, O.S. GWR steam. 1972.
On pp. 117-18 Nock relates how A.W.J. Dymond had told him that the design work for the 56XX had been taken verry seriously at Swindon: Hawksworth had stressed that there was to be no prototype as 50 were to form the initial order. The wheelbase of the 56XX was identical to that of the Rhymney R and M classes which could share the same No. 2 boiler. The Taff Vale 0-6-2Ts required a smaller boiler (the S3 type) to be designed using the existing flanging plates and tools.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 5. Six-coupled tank engines. 1958.
Rickard, S. The "5600" class tank locomotives. Rly Wld, 1960, 21, 231-3. 4 illus.

0-6-2T: modifications to absorbed stock.

Rhymney Railway:
Several classes were modified with Swindon taper-boilers.

Copsey, John. Rhymney 'M' and 'R' 0-6-2 tanks after the Grouping. .Great Western Rly J., 2 (10) 421-9.
123 locomotives were provided by the RR at the Grouping: 101 of these were 0-6-2Ts constructed under the superintendency of John Jenkins until 1906 and C.T. Hurry Riches thereafter. All were supplid by outside builders. The M class were constructed by Robert Stephenson in 1904 and had 4ft 6in driving wheels: numbered 33/47-51. The R class (also supplied Robert Stephenson, but with higher boiler pressure in 1909) were numbered 30-2/34/46. In 1921 Hudswell Clarke supplied six with Belpaire boilers: 35-8/39/44. The earlier locomotives were fitted with Belpaire boilers and some were "improved" with No. 2 taper boilers. Lists allocations and duties of class in December 1938. Illus.: 36 on 18 May 1957 at Llanbradach on empties; 41 ex-works at Cathays shed in early 1930s; 38 at Bargoed Pits on 27 June 1957 (coal train with two enclosed hoppers for pulverized fuel: see letter Number 13 page 572 by K.N. Prince and letter from David Tomkiss in Number 14 page 616; 38 on mixed freight passing ex-Cardiff Railway 155 at Cherry Orchard on 17 May 1952 (R.C. Riley); 40 on 5 May 1951 at Radyr shed (had been rebuilt with taper boiler in October 1949)(RCR); 39 at Cherry Orchard (two views); 38 at Walnut Tree on 26 September 1957 (R.O. Tuck); 36 at Tyndall Street on 21 March 1957 (ROT); 38 at Llanbradach on 7 September 1957 (ROT)
Jones, W. The last Welsh 0-6-2 tanks. Rly Wld, 1958, 19, 328-32. 5 illus., 2 diagrs. (s. f. & r. els.)
Class AR
Mountford, E.R. Some further thoughts on the Rhymney Railway 0-6-2 tanks. Rly Wld, 1959, 20, 182-6. 6 illus., 3 tables.
Material additional to W. Jones' contribution
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 10. Absorbed engines, 1922-1947. 1966.

Taff Vale Railway:
Several classes were modified with Swindon taper-boilers.

Beckerlegge, W. The Taff Vale Railway 0-6-2T's. Rly Obsr, 1941, 13, 53-5.2 illus. (line drawings. s. els.)
Jones, W. The Taff Vale 0-6-2 tank engines. Rly Wld, 1957, 18, 176-9. 7 illus.
Mountford E.R. The Taff Vale class "A" 0-6-2 tanks. Trains ill., 1957, 10, 375-82. 4 illus., 2 tables.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 10. Absorbed engines, 1922-1947. 1966.

0-6-0ST: Armstrong/Dean/Collett

Le Fleming, Hugh M. International locomotives. Plate 20
Great Western Railway Group 0-6-0Ts at Worcester The Great Western had a long love affair with the 0-6-0T type, and for decades they formed about a quarter of the total engines in stock. Few were side tanks; most of those running in the nineteenth century were built with saddle tanks, the twentieth-century engines having pannier tanks. Many of the saddles were later converted to panniers. Swindon and Wolverhampton went their own sweet ways in the design of their locomotives, and even when each works was building 0-6-0Ts of comparable size, power, and appearance, few of the parts would be interchangeable, a crazy situation which was perpetuated under Collett who built versions of each simultaneously at Swindon in the 1930s! Wolverhampton engines were distinguishable by having a shallower valance angle under the running plate than the Swindon engines, and this basic difference persisted no matter what boilers, cabs, or tanks were fitted, so that where, as in the 1700, 1800, and 2700 series, each contributed similar engines, the origin could be plainly perceived. Swindon, in addition to the inside-framed type, built two versions with double frames, one with small and the other with larger wheels. The group here looking from left to right, consists of:
1 A Swindon heavy inside-framed locomotive, rebuilt with pannier tanks.
2 A Swindon outside-framed large- wheeled locomotive, rebuilt with pannier tanks.
3 A small Wolverhampton engine, with domeless boiler, raised Belpaire firebox, and pannier tanks. One of this strange variety lurked in the depths of Birkenhead, complete with bell for dock work, until about 1950.
4 A Swindon 'heavy', similar to the first locomotive.
5 A Swindon outside-framed small- wheel engine, still with saddle tank. Note the variety of bunker shapes, and the open-back cabs, a bleak feature which many retained until finally scrapped under BR auspices.
Railway Correspondence & Travel Society The locomotives of the Great Western Railway. Part 5. Six-coupled tank engines. Kings Heath (Birmingham), RCTS 1958.
Rutherford, Michael. Matchless matchboxes (Provocations/Railway Reflections No. 9). . Backtrack, 1995, 9, 471-7.
Development of the pannier tank locomotive from side and saddle tank precursors under Armstrong, Churchward and Collett. Author divided GWR 0-6-0 tank locomotives into five categories:
1. Large Wolverhampton tanks with inside or sandwich frames.
2. Small Wolverhampton tanks with inside frames only
3. Large Swindon tanks with inside and double frames [57xx was main manifestation of these]
4. Miscellaneous
5. Absorbed locomotives.
Argues that Classes: 1813, 1854, 2721, 57XX and 94XX formed a single procession in design, and produced a total of 1313 related locomotives. The Dean Goods (2301 class) stemmed from the 1813 series, whilst the 94xx employed the Number 10 boiler developed for the 2251 mixed traffic 0-6-0. Argues that two divergent processes were at work: standardization and diversity wrought through improvements, or the need to meet specific conditions. Two tables illuminate both trends within the classes. Some of the diversity is demonstrated by the series numbers: 67xx were fitted only with steam brakes and were intended for shunting. The 97xx series were fitted with condensing apparatus, a special form of ATC to clear electrified tracks, and trip cocks for working over the Metropolitan line to Smithfield. The 8750 series incorporated several improvements. Suggests that the design should have been adopted as a standard by British Railways. The class combined cheapness, simplicity, reliability and versatility.
As an introduction Rutherford introduced one of his "hot under the collar" topics, namely the quest for authenticity in museum exhibits.
illus.: Diagram for an 1813 class; No 1846 of Lot 60; No 1720 of the 1854 class at Newport; No 2797 of Lot 129; Locomotives; Two versions of diagram B48; A typical GW scene at Fishguard with no 5716 on shunting duty; No 5766 near Shiplake; No 9709 at Paddington; Nos. 4616 and 4631 at Folkestone; Table 1 GWR Tanks ???; 94xx No 8401 at Cardiff; No 8783 at Kings Sutton; Table 2 0-6-0T diagrams;

Class 1661

Hambleton, F.C. GWR saddle tank engines. Loco. Rly Carr. Wagon Rev., 1944, 50, 185.
Line drawings of class both as 0-6-0ST and 0-6-0PT forms

1361 class: Churchward: 1910

Webster, V.R. Unusual lineage — the Cornwall Minerals engines and their Great Western descendants. Rly Wld, 1984, 45, 17-20.
Includes fairly brief description of the Sharp Stewart 0-6-0Ts designed by Francis Trevithick to operate back-to-back (includes illustration of one in that state): later further locomotives of this type, but not back-to-back) were produced under Churchward's direction: largely to work in Plymouth Docks.

0-6-0PT:
Pannier tanks became a standard feature on G.W.R. 0-6-0 tank engines in 1909, or thereabout. These tanks permitted easy access to the inside motion. In addition, they could be fitted to locomotives with Belpaire boilers, whereas saddle tanks (which were used until 1909) could not.

1366:1934: Collett:
These were intended for dock working and were based on the Churchward 1361 0-6-0ST class.

NEW light shunting tanks, G.W.R. Rly Mag., 1934, 75, 141-2.2 illus.

Retrospective and critical

RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 5. Six-coupled tank engines. 1958.
Tuplin, W.A. Pannier tank engines of the Great Western Railway. Rly Mag., 1954, 100, 677-81. 5 illus., 3 tables.

15XX: 1949: Hawksworth:
These shunting engines featured several departures from Great Western traditions. To improve accessibility and ease maintenance, no running plate was fitted. Outside Walschaerts gear was also incorporated for the same reasons.

"1500" class 0-6-0 tank, Western Region. Loco. Rly Carr. Wagon Rev., 1949, 55, 130. illus.
NOVEL Western Region shunter. Trains ill., 1949, 2, 113. illus.
OUTSIDE-CYLINDER shunting tank engine for the Western Region. Rly Mag., 1949, 95, 390-1. illus.,diagr. (s. & f. els.)
WESTERN Region "1500" class tank engines. Rly Gaz., 1949, 91, 481. illus., diagr. (s. el.)

Retrospective and critical

Cook, K.J. Swindon steam. p. 151
0-6-0 six coupled tank engine with outside cylinders and Walschaerts valve gear. It was designed especially as a 24-hour shunting engine which did not need to go over a pit for oiling which was a very laudable objective. Ten were built of which several were employed on the empty coaching stock between Paddington and Old Oak Common Carriage Sheds. (In those good old days all main lines arrivals went out to Old Oak Common for servicing before coming back into Paddington, so there was no delay to passengers on account of late arrival of an incoming train.) On this work these little engines were very much in the public eye and Hawksworth was very surprised at the interest created and the very favourable comments.
Nock, O.S. British locomotives of the 20th century. Volume 2. 1930-1960.. 1984.
On page 104 Nock noted that Capt. Hugh Vivian, Chairman of the Locomotive Committee is supposed to have said writtren to Nock statingt "I am very glad that at last we've got a GWR engine with outside Walschaerts valve gear"
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 5. Six-coupled tank engines. 1958.

850 class

Copsey, John '850' Class saddle tanks. Gt Western Rly J., 4, (25) 40-58.
Relies upon Holcroft's Outline of Great Western locomotive practice for technical information. First batch introduced in 1874-6. Allocations and duties: 1901, 1918 and 1930.

Collett "developments"

From 1929 onwards Collett indulged in a huge programme of scrapping and replacing the small engines which formed about half of the Great Western fleet. Churchward seems to have disapproved strongly of engines without guiding wheels, and used pony trucks on his smaller tank engines, the excellent 44xx and 45xx 2-6-2 tanks.... He allowed Wolverhampton to finish buildng a large order for small 0-6-0 saddle tanks, which was completed by 1905. Otherwise he built no 6-wheeled engines at all, except for half a dozen dock tanks, the 1361 class, and they were a re-hash of and a replacement for similar engines which had come from the Cornwall Minerals Railway. Collett at first seems to have adhered to the same principle, and built many more of the 45xx class; the last hundred from No 4575 to 5574 had taper-topped side tanks holding 1,300 gallons, in place of the flat-topped ones which only held 1,000 gallons. For his big replacement programme however Collett turned right round, and built only 6-wheeled engines, all of them up-dated versions of Dean and Armstrong designs, as rebuilt by Churchward with Belpaire fireboxes and pannier tanks in the case of the 0-6-0. Gibson, John C. Great Western locomotive design: a critical appreciation. 1984. Return to beginning..

Contemporary accounts of introduction of pannier tanks

Great Western Ry. Loco. Rly Carr. Wagon Rev., 1910, 16, 135.
Note refers to 1711 and 1850 as "square-sided saddle tanks": presumably 0-6-0PT is easier than 0-6-0SSST

57XX: 1929: Collett:
There were 863 locomotives in the 57XX class. They were employed throughout the system on shunting and freight duties. Some were also used to haul passenger trains. Some were sold for use in collieries and some served on London Transport. Return to beginning.

SIX-COUPLED pannier tank engines, Great Western Ry. Loco. Rly Carr. Wagon Rev., 1929, 35, 70. illus. diagr. (s. & f. els.)
General arrangement drawing of first series of 57XX 0-6-0PT. Great Western Rly J., 1993, 1 (8) 330-1.
Swindon April 1928: side elevation and plan.

67XX series: 1930:
57xx built solely for freight working.

GREAT Western Ry. 0-6-0 pannier tank engine. Loco. Rly Carr. Wagon Rev., 1931, 37, 110. illus.

87XX (later 97XX): 1932:
57xx modified with condensing apparatus for working between Paddington and Smithfield over London Transport lines. Goslin covers the 97XX class well, but unlike most of the other classes considered the 97XX were used solely on freight workings, mainly two and from Smithfield..

[8700: locomotive fitted with condensing gear]. Rly Mag., 1932, 71, 76.

Goslin, Geoff. Steam on the Widened Lines. Volume 2: The Great Western and Southern companies. Colchester: Connor & Butler, 1998. 56pp.

Oil-firing: 1960
Atkins in his Dropping the fire refers to the experimental fitting of a GWR tank engine to burn oil fuel in 1960 and Rutherford (below) included picture of No. 3711 so-fitted in 1958 and still extant in 1963.
Rutherford, Michael. Crisis? What Crisis? Coal, Oil and Austerity. Part 1.. (Railway Reflections No. 71) Backtrack, 2000, 14, 665-74.
Following a very brief analysis of the development of coal burning (from coke burning) and the problems of coal supply, especially during strikes and in the immediate Post WW2 period the author introduces oil-consuming traction on the GWR (i.e. the pre-WW2 railcars and post-WW2 steam locomotives) and the influence of Sir James Milne (a thumbnail biography is given). Illus.: No 3813 renumbered 4855 when converted to oil firing, Diagram of the GWR installed equipment in the engine, Diagram of the GWR installed equipment in the tender, Diagram of the firebox showing the extra brickwork and air inlets, The cab of 3904 aka 4972 showing the fireman's padded seat ? It also had electric light!, Col.: GWR no 3711 at one time oil fired in May 1963 (W. Potter), Oil-burning Castle no 100A1 Lloyds in April 1947 on express at Reading (H.N. James)

Retrospective and critical

Cook, K.J. Swindon steam. p. 151-2
The 57XX numbered 853 and if the further 200 had been built [rather than 94XX] to the same design the class would have totalled over 1,000, far and away the largest class in the country....
When in the Works with me one day he [Hawksworth] had put forward this idea largely on the grounds that the drumhead type of smokebox was a better job and made the boiler more easily interchangeable. But I pointed out that the cylinders, tubeplates, smokebox front plates and extension pieces were all jig-drilled and completely interchangeable, no less easily than with drumhead smokeboxes.
As modified to 94XX class there were four distinct disadvantages, two from the production and Works maintenance angle and two from the operating point of view. Dealing with the latter aspect first, the 57XX class with the 2301 class straight-sided firebox had a narrower cab which made it eminently suitable for shunting; all the controls of regulator and brake valve were readily at hand to a driver looking outside the cab. The taper firebox of standard No 10 boiler needed a wider cab and when leaning out of the cab the driver could not reach the brake valve. There were end-less complaints about this and ultimately another pivoted handle coupled to the brake valve was mounted nearer to the cabside to give a remote control for brake application.
The mounting of the large boiler increased the total and axle weights of the locomotive and turned it from 'blue' to 'red' route classification although as its 'vital statistics' of cylinder bore and stroke, wheel diameter and boiler pressure remained the same, the tractive effort was not increased but it prevented the Running Superintendent from allocating it for a number of turns for which he desired it.
From the construction viewpoint the incorporation of the smokebox saddle on top of the cylinders required extra moulding box sections in the foundry and turned the cylinders from the small to large group and put an extra load in the foundry.
The 2301 class boiler with its straight sides was a very simple boiler to construct but above all during repair when a new firebox was necessary this could be built up and riveted by hydraulic power as a unit and inserted into the steel casing whereas the standard 10 firebox had to be inserted in sections, seam-welded in position and a good deal of hand wedge-riveting carried out.
Copsey, John. '57XXs' AT Reading. Great Western Rly J., 1993, 1 (8) 332-8.
Reading was one of their many s/hunting grounds.
Davies, F.K. The locomotives of the Great Western Railway: the 5700 class 0-6-0T. Rly Obsr, 1949, 19, 3-6 + plate. 3 illus.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 5. Six-coupled tank engines. 1958.
It should be noted that unlike the later Locomotives of the LNER this RCTS volume was published considerably before the demise of the class, members of which outlasted British Railways, notably those used by the National Coal Board and London Transport. Umpteen have been "preserved" and bits and pieces are on most "preserved railways"
Tuplin, W.A. Pannier tank engines of the Great Western Railway. Rly Mag., 1954, 100, 677-81. 5 illus., 3 tables.
A history.
Vaughan, Adrian. The heart of the Great Western. Peterborough: Silver Link, 1994.
Relates (pp. 49-51) how Charlie Turner (an Oxford driver) found the 37XX series easier to fire than the 74XX as the firegrate was square, deep and very effective, like those on the Metro tanks.

54XX: 1931: Collett:
This was a replacement design for the 2021 class. It was used for push-and.pull working (auto-working). There were also 64XX and 74XX series..

NEW "auto" locomotives, G.W.R. Rly Mag., 1932, 70, 208. illus., diagr. (s. el.)
SIX-COUPLED tank locomotive for auto-service, Great Western Ry. Loco. Rly Carr. Wagon Rev., 1932, 38, 1. illus.

Retrospective and critical

Cook, K.J. Swindon steam. pp. 86-7
A design was prepared with 5ft 2in diameter wheels, six-coupled with 16½ in cylinders, to incorporate a slightly larger boiler, but in order to test it at its work, a locomotive undergoing overhaul in the Works of 2021 class, actually No 2080 was selected for modification. In this case, existing cylinders, type of smokebox, motion and boiler were retained but 5ft 2in wheels were provided and auto- fitted. She was tried out in January 1930 and put through very extensive running tests on various services. The design, when brought up to date with the further details envisaged for the new locomotives, was considered to meet the requirements and an order was issued for 30 locomotives, 20 to have 5ft 2in wheels (to be 54XX class) and ten 4ft 7½;in wheels (64XX class). It was also decided, in order to get the first locomotive into traffic in advance of the main order, to convert but much more extensively another 2021 class. So 2062 was fitted with 16tin x 24in inside cylinders of a new design; slide valves, but incorporating saddle to accommodate the drumhead type of smokebox, cast steel crossheads for three-bar motion (similar to but smaller than on the 0-6-2T 56XX class) in order to get away from the old four-bar arrangement and also clear the leading axle, standard No 21 boiler, pannier tanks and covered in cab. This converted locomotive was actually to be number one of the lot, and numbered 5400 it was turned out in August 1930. The first of the completely new locomotives was turned out in November 1931 and batches followed, interspersed with the usual annual batches of Castles and other types. The standard 21 boiler was actually the standard 11 but with drumhead smokebox. I was anxious that as a prototype 5400 should not for ever be a hybrid, as one or two of the previous class numbers had been, so I ordered on the shops 30 complete sets of components including the main frames. A number of the new parts had to be used during the conversion and those not needed were stored with the remainder of the locomotive assemblies. When the last of the lot, No 5419 was under erection, 5400 had accumulated a good mileage and was approaching the stage when its first repair was due. So I ordered it into the 'A' Shop, had it stripped down, and in its place built up the new frames using from the old engine everything which had had to be new during the conversion and together with the remnant of the new components, completed the engine which was then fully standard with the rest of the class. 5400 had not been back in traffic more than a few weeks before one locomotive observer wrote in to say that that was not the locomotive which had been taken in.for overhaul! Such is the spotlight on all locomotive details by railway enthusiasts.
Copsey, John. The '54XX' autos. Gt Western Rly J., 2002, 6, 88-101.
Brief details of the development of steam rail-motors (steam railcars) and the services operated by them. Than ran well but were subject to oscillation going downhill. Coaling was difficult and the paintwork became stained. Auto-train (push & pull) gradually augmented and then displaced these rail-motors using the 517 class 0-4-2Ts and 0-6-0PTs of 2021 and 1076 classes. No. 2120 was specially adapted with external cladding to look like a coach. No. 2062 was modified with 5 ft 2 in driving wheels, renumbered 5400, and evaluated on Moretonhampstead services and later again assessed with five inches smaller wheels on steeper gradients in the Cardiff and Merthyr areas. The 54xx were used initially on the West London suburban services "centred" on Greenford. Workings from Westbury are also described. General arrangement drawing (5401/6400 type, Swindon, May 1931)
Copsey, John. The '54XX' autos. Part 2. Gt Western Rly J., 2002, 6, 139-50.
Crump, Bob. Autocar Work. Gt Western Rly J., 2004 (51),177-9.
Working from Reading in the post WW2 period on the Henley branch and on the mainline to Didcot and to Slough. In the 1950s working on the Marlow branch from Maidenhead. Also memories of push & pull workings on the Eastern Valley services from Newport to Brynmawr prior and during WW2: many were worked by 64XX and some sandwich formations of three cars were operated. See also letter from Ray Caston (No. 52 p. 239) concerning auto-train workings on Eastern Valley to Abersychan and Talywain (on former LNWR lines) which were suspended on 5 May 1941. Also letter from A.E. Abear on workings with 14XX from sub-shed at Staines: the worst task was fueling the bunker from a coal wagon..
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 5. Six-coupled tank engines. 1958.
It should be noted that unlike the later Locomotives of the LNER this RCTS volume was published before the demise of the class.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 193-5
Rutherford, Michael. The Ghost in the Machine. George Armstrong and the Wolverhampton heritage. (Railway Reflections No. 65). Backtrack, 2000, 14, 294-302.
The Great Western way of using old designs for new manufacture (as in the case of the 54XX and 48XX classes for auto-train work). illus.:850 class No 987, the first 0-4-2T No 517, 2021 class No 2104 at Stourbridge, Birmingham, Diagram; Steam railcars, Rebuilt 517 class No 1425 at Wood End, Birmingham, No 1426, No 2062 was rebuilt to make a prototype No 5400 for a new class, No 1925, No 5415 at Kensall Green, London, 58xx No 5813 at Bearley, Stratford on Avon, No 6422 on an autotrain at Windmill End, Birmingham, 74xx class No 7409 at Blaenau Ffestiniog, Railcar No 21 at Monmouth (see letter by York (page 430) and illustration of Winchcombe station (page 283), 16xx class No 1646. Extra information of 4-wheel passenger coach with electric lighting and (page 298) of auto-trailer No. 1 which was never a steam railmotor. (John Lewis page 430),
Tuplin, W.A. Pannier tank engines of the Great Western Railway. Rly Mag., 1954, 100, 677-81. 5 illus., 3 tables.
A brief history for this class!.
Vaughan, Adrian. The heart of the Great Western. Peterborough: Silver Link, 1994.
Relates how Charlie Turner (pp. 49-51) found the 74XX series more difficult to fire than the old Metro tanks. The 74XX firegrate was far too shallow at the back end

94XX: 1947: Hawksworth:
Taper-boiler development of the 57XX class. Most features standard with other classes, notably 8750 class and 2251 class (the No. 10 taper boiler with drum head smokebox). Locomotives were still being delivered long after the GWR had ceased to exist and there is now an extensive literature on this wasteful manufacture of a defunct locomotive type, namely the 0-6-0 shunting engine.

Bagnall shunting engines for Western Region. Rly Gaz., 1949, 91, 243. illus., diagrs. (s. & f. els.)
Bagnall series.
Heavy shunting tank engines for G.W.R. Rly Gaz., 1947, 87, 43. illus., diagr. (s. & f. els.)
Heavy shunting tank engines for G.W.R., Rly Mag., 1947, 93, 314-15. illus., diagr. (s. & f. els.)
The new G.W.R. "9400" class engine. Railways, 1947, 8, 114. illus.
0-6-0 tank heavy shunting engine, G.W.R. Loco. Rly Carr. Wagon Rev., 1947, 53, 102. illus.
No. 9409 illustrated.

Retospective & critical

Atkins, Philip. Last of a long line [rise and fall of WR 94xx 0-6-0PT's]. BackTrack, 1991, 5, 39-40.
The GWR placed orders for 200 obsolete locomotives from a variety of private builders immediately prior to nationalization (the other railways had accepted the need for diesel shunters, or more modern designs for light work).
Atkins, Philip. Dropping the fire. 1999. pp 48-9.
Notes that in 1960 there was a proposal to equip thirty of this class with oil-firing.
Cook, K.J. Swindon steam. p. 151-2
The redesign of the 0-6-0T 57XX class into 94XX class was rather a tragedy and 200 of them were built. The 57XX numbered 853 and if the further 200 had been built to the same design the class would have totalled over 1,000, far and away the largest class in the country.
When putting up the annual locomotive building programme to Paddington it was the custom to append a diagram of each class to be built. On this occasion the recommendation was, among others, for a further batch of 57XX class but when Sir James Milne, General Manager who had in years gone by been a pupil in the Loco Works at Swindon, saw the diagram he said words to the effect that in this year of grace you cannot build a locomotive with a steam dome. It was a pity that Hawksworth did not hold his ground and stick out for 57XX class but in actual fact he must have been quite pleased inwardly to modify and produce the 94XX class using a standard No 10 boiler with a drumhead type of smokebox.
When in the Works with me one day he had put forward this idea largely on the grounds that the drumhead type of smokebox was a better job and made the boiler more easily interchangeable. But I pointed out that the cylinders, tubeplates, smokebox front plates and extension pieces were all jig-drilled and completely interchangeable, no less easily than with drumhead smokeboxes.
As modified to 94XX class there were four distinct disadvantages, two from the production and Works maintenance angle and two from the operating point of view. Dealing with the latter aspect first, the 57XX class with the 2301 class straight-sided firebox had a narrower cab which made it eminently suitable for shunting; all the controls of regulator and brake valve were readily at hand to a driver looking outside the cab. The taper firebox of standard No 10 boiler needed a wider cab and when leaning out of the cab the driver could not reach the brake valve. There were end-less complaints about this and ultimately another pivoted handle coupled to the brake valve was mounted nearer to the cabside to give a remote control for brake application.
The mounting of the large boiler increased the total and axle weights of the locomotive and turned it from 'blue' to 'red' route classification although as its 'vital statistics' of cylinder bore and stroke, wheel diameter and boiler pressure remained the same, the tractive effort was not increased but it prevented the Running Superintendent from allocating it for a number of turns for which he desired it.
From the construction viewpoint the incorporation of the smokebox saddle on top of the cylinders required extra moulding box sections in the foundry and turned the cylinders from the small to large group and put an extra load in the foundry.
The 2301 class boiler with its straight sides was a very simple boiler to construct but above all during repair when a new firebox was necessary this could be built up and riveted by hydraulic power as a unit and inserted into the steel casing whereas the standard 10 firebox had to be inserted in sections, seam-welded in position and a good deal of hand wedge-riveting carried out.
Copsey, John. The '94XX' class in traffic. Gt Western Rly J., 2007, 8, 328-53.
Includes Swindon detailed working drawings (side, front and rear elevations, sections and plan) dated November 1945, and a diagram which shows the protracted construction of contract-built locomotives from Bagnalls, Stephenson, and the Yorkshire Engine Co. over the period 1949 to 1956. Includes illus. of 8403 and 8408 when new in 1947 and lettered "GWR" and of 8428 on delivery to Swindon with small BR logo and large W.G. Bagnall board on side. Duties are listed which included station pilot duties at Paddington, shunting and the haulage of short distance freight.
Davies, Bob. Banking after 'Bertha'. Backtrack, 1, 67-8.
The 94XX found a final niche on the Lickey incline where others employed latterly included Standard 9F 2-10-0s; ex-GWR 2-8-2T and 2-8-0T (former had clearance problems); 3F 0-6-0 tender locomotives; 3F 0-6-0Ts (Jinties); 94XX 0-6-0PTs and Hymek and class 37 diesel banking engines. illus.(b&w): train of tank wagons banked by a 9F and a pair of ex-GW 0-6-0-PT's; ex-GW 2-8-0 5226 descends 'light engine'; quartet of 0-6-0 PT's assist an Esso tank wagon train at Vigo.
Joy, Stewart. The train that ran away. p. 40
Highly critical of the 200 94XX ordered after it had been agreed that diesel shunters were the economic way forward
Nock, O.S. British locomotives of the 20th century. Volume 2. 1930-1960.. 1984.
On page 103: No. 10 boiler increase weight and put it in the red route availability category and thus could not perform 57XX (blue) duties.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 5. Six-coupled tank engines. 1958.
It should be noted that unlike the later Locomotives of the LNER this RCTS volume was published considerably before the demise of the class, members of which outlasted British Railways, notably those used by the National Coal Board and London Transport. Umpteen have been "preserved" and bits and pieces are on most "preserved railways"
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 226; 227
Tuplin, W.A. Pannier tank engines of the Great Western Railway. Rly Mag., 1954, 100, 677-81. 5 illus., 3 tables.
A very brief history for this class!.

16XX: 1949: Hawksworth:
These small locomotives were intended for branch line and dock working. They arrived into British Railways and ended up in some unlikely locations, notably on the Dornoch branch, north of Inverness. Atkins letter Steam Wld, 2010 (279) page 33 notes that batch of twenty authorised in 1953 (presumably during reorganiaztion hiatus at that time): this batch not noted in Jones.

A NEW Western Region engine. Railways, 1950, 11, 43. illus., diagr. (s. & f. els.)
0-6-0 pannier tank locomotive. Engineering, 1950, 169, 76. (REA 4794).
WESTERN Region. 0-6-0T —

1600 class. Loco. Rly Carr. Wagon Rev., 1950, 56, 22. illus.
WESTERN Region "1600" class tank engine. Rly Gaz., 1950, 92, 486. illus., diagr. (s. el.)
W.R. "1600" class tank engine. Rly Mag., 1950, 96, 383. illus., diagr. (s. el.)

Retrospective and critical

Pile, Kevin. Small is beautiful: the 16XX pannier tanks. Br. Rlys Ill.., 1995, 4, 517-29.
For each locomotive the MPDs to which they were allocated and the works wherwe they were overhauled are listed.
RAILWAY Correspondence and Travel Society. The locomotives of the Great Western Railway. Part 5. Six-coupled tank engines. 1958.
It should be noted that unlike the later Locomotives of the LNER this RCTS volume was published considerably before the demise of the class, members of which outlasted Britsih Railways, notably those used by the National Coal Board and London Transport. Umteen have been "preserved" and bits and pieces are on most "preserved railways"
Tuplin, W.A. Pannier tank engines of the Great Western Railway. Rly Mag., 1954, 100, 677-81. 5 illus., 3 tables.
A very brief history for this class!.

4-4-4T

Midland & South Western Junction Railway No. 27: 1925:
Collett: rebuilt with taper-boiler.

GREAT Western Ry. (ex Midland and South Western Junction): 4-4-4 tank engine No.27, rebuilt. Loco. Rly Carr. Wagon Rev., 1925, 31, 102-3. illus., diagr. (s. & f. els.))

Retrospective and critical

RCTS. The locomotives of the Great Western Railway. Part 10. Absorbed engines, 1922-1947. 1966.

4-4-2T

2221 County Tanks: 1905

Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. pp. 58-63

4600: Churchward: 1913
Solitary locomotive which Atkins states was closely associated with Hawksworth in its design: same boiler as 45XX. The engine was intended for passenger train working in Cornwall. Not being found so suitable as other types, it was stationed at Tyseley for suburban working, but it was not liked and therefore it was not repeated. Holcroft .

4-4-2 side tank engine, No. 4600. Loco. Rly Carr. Wagon Rev., 1914, 20, 120.
Allocated to Tyseley. 5ft 8in coupled wheels; 17 x 24in cylinders; total heating area 1271.86ft²; grate area 16.6ft²; 200 psi working pressure.

Atkins, Philip. 1913 — a halcyon year. Backtrack, 2005, 19, 300-5.
RCTS. The locomotives of the Great Western Railway. Part 9. Standard two-cylinder classes. 1962.
Russell, J.H. A pictorial record of Great Western engines. Volume 2. [The Churchward, Collett and Hawksworth locomotives]. page 58; 63

4-4-0PT

Armin, Brian. The Dean—Churchward transition. Br Rly J., 1991 (36), 267.
No. 1490 of 1898 was the first pannier tank. It was fitted with a Belpaire firebox. It was unsteady at speed and not suited to fast suburban services: it was too heavy for services over the Metropolitan Railway. In 1907 it was sold to the Bute Works Supply Co., who sold it to the Ebbw Vale Steel & Iron Co, who sold it to the Brecon & Merthyr Railway in 1908, who sold it to Cramlington Colliery in 1916 where it was scrapped in 1919.

GWR 4-4-0PT No. 1490. Br Rly J., 1991 (37), 351.
4mm/1ft scale drawing (courtesy Mike Lloyd): side, front & rear elevations; also main dimensions.

2-4-2T

No. 11
5ft 2in coupled wheels; 17 x 24in cylinders, 170 psi boiler pressure: fitted with a steam reveerser and a water scoop which could operate in eaither direction

Contemporary

New G.W.R. locos. Locomotive Mag., 1901, 6, 26.
The first of some new tank engines, No. 11 class, had been turned out from the, Swindon Works and was numbered 3601 (Swindon No. 1866). It was fitted with the water scoop, and the tool boxes were placed-at the side of the smokebox instead of on the top of the side tanks.

Great Western Ry. Locomotive Mag, 1903, 9, 285.
Ten No. 11 class to be built but with taper boilers

Retrospective

Armin, Brian. The Dean—Churchward transition. Br Rly J., 1994 (36), 267.
No. 11 acted as the prototype for the not very successful 1490 class of 1902. There were problems with the failure of its tanks when the water pick-up gear was used.

36XX: 1902
Production series

Contemporary

Great Western engines. Locomotive Mag., 1902, 7, 78.

Le Fleming, Hugh M. International locomotives. Plate 22
Painting of 36XX No. and an Aberdare 2-6-0
RCTS. The locomotives of the Great Western Railway. Part 6. Four-coupled tank engines. 1959.

2-4-0T

Morris, O.J. Early cab doors, GWR 2-4-0 tank engine "Prince". Loco. Rly Carr. Wagon Rev., 1941, 47, 71.
No. 2137

0-4-2T

517 class
The first 517 Class design was for a saddle tank with a 13ft 7in wheelbase and 5ft coupled wheels with a total weight inservice of about 32 tonnes. Construction started at Wolverhampton in late 1867 and the first, No. 1060 entered traffic in April 1868. 59 further were manufactured in five lots over the next two years. The last six (Nos. 1106-1111) differed in having side tanks. From No. 1046 onwards the wheelbase was increased by either 1 or 13 inches. The Class was renumbered 517-576 in July 1870. Six further lots were manufactured between September 1873 and February 1878. These were: Nos. 826-849; 1154-1165; 202-205; 215-222; 1421-1444. The total of these lots was 72 and they differed from the earlier series in having a longer wheelbase (15 feet) and were about 3 tonnes heavier.

1922: Collett:
This class was introduced in 1868. In 1922, No. 1421 was rebuilt with a domeless Belpaire boilder.

REBUILT tank engine, Great Western Railway. Loco. Rly Carr. Wagon Rev., 1923, 29, 4 illus,diagr. (s.el.)

Retrospective

Copsey, John. The '517' Class at work. Part 1. Great Western Rly J., 2010, 10, 62-80.
Copsey, John. The '517' class at work . Part 2. Great Western Rly J., 2010, 10, 122-45 .

RCTS. The locomotives of the Great Western Railway. Part 6. Four-coupled tank engines. 1959.

3571 class: 1895

Le Fleming, Hugh M. International locomotives. Plate 19
Painting of No. 3579
RCTS. The locomotives of the Great Western Railway. Part 6. Four-coupled tank engines. 1959.

48XX (later 14XX): 1932: Collett:
This was a modernized version of the 517 class- It was intended for branch line and push-and-pull (auto-train) working.

NEW 0-4-2 tank engines, GW.R.. Rly Mag., 1935, 76, 461. illus.
0-4-2 type tank locomotive, G.W.R.. Rly Mag., 1933, 72, 122-3. illus., diagr. (s. & f. els.), table.
0-4-2 light auto tank engines, 4800 class, Great Western Ry.. Loco. Rly Carr. Wagon Rev., 1933, 39, 9-10. illus.

Retrospective and critical.

Copsey, John. '48XX' auto engines. Great Western Rly J., 1997, 3 (22) 329-40.
General arrangement drawings, Swindon August 1932 (headed 1400 class): replacement for 517 class. First locomotive entered service in September 1932. Allocations and duties. Illus.: 4806 on Exeter shed c1934; 4833 at Swindon Factory in 1934; 4843 at West Drayton on Staines autocar on 22 April 1935; 4861 at Merthyr shed on 15 August 1937; 4817 on Gloucester shed on 16 May 1937; 4863 at Gloucester on 7 April 1939; 4858 at Banbury shed c1945; 1425 at Swindon on 4 July 1947 ex-works; 1413 at Gloucester c1947; 1406 at Gloucester; 1450 at Kennington Junction, Oxford c1951.
Copsey, John. 48xx Auto Engines. Part 2. Great Western Rly J., 1997, 3 (23) 411-20.
Later workings. Illus.: unidentified 14xx at Broughton Gifford Halt c1949; Marlow auto (14xx in middle) c1954; 1447 on Reading shed in 1951; 1450 at Kennington Junction, Oxford, possibly with freight for Abingdon; 1442 c1951 with possible long freight for Abingdon; 1455 c1954; 1431 and 1452 on Fishguard branch; detail on 1442 and 1419; and 1426 outside Caerphilly Works on 10 April 1960.
RCTS. The locomotives of the Great Western Railway. Part 6. Four-coupled tank engines. 1959.
Wheeler, Geoffrey. Fired by steam. London: John Murray, 1987.
Includes coloured plate, based on water colour side elevations of No. 1450, or what artist calls a Marlow Donkey: claimed to be middle chrome green, but looks like black! lettered "GWR"

0-4-0T

No. 101

Armin, Brian. The Dean—Churchward transition. Br Rly J., 1994 (36), 267.
No. 101 was designed to burn oil fuel using the Holden system and was probably intended for the Wrington Vale Light Railway. It employed Joy valve gear and a complex firebox. Locomotive Mag., 1902, 7, 175 mentions oil fuel: Locomotive Mag., 1903, 9, 14 also mentions oil fuel, covered in cab and lever reverse; WN 1969. . In 1903 it was rebuilt with a Lentz boiler and a Vanderbilt circular corrugated firebox.

1101: 1926: Collett/Avonside
These were Avonside standard products modified for G.W.R. conditions.

TANK locomotives for Swansea Docks, G.W. Ry.. Loco. Rly Carr. Wagon Rev., 1926, 32, 276. illus.
TWO recently constructed locomotives. Engineer, 1926, 142, 399. 2 illus.

Retrospective and critical

Baker, S.W. Great Western 0-4-0 dock tanks. Railways, 1950, 11, 168-9. 5 illus.
Beckerlegge, W. Great Western 0-4-0 dock tanks. Railways, 1951, 12, 57-8. 3 illus.
Additional material.
RCTS. The locomotives of the Great Western Railway. Part 6. Four-coupled tank engines. 1959.

Steam railcars (rail motors)

Great Western Ry. Locomotive Mag., 1903, 9, 241
The second steam railway motor car was running in turn with No. 1, between Chalford and Stonehouse: both were giving great satisfaction.

GWR steam rail motor. Locomotive Mag., 1936, 42, 30
Last service performed by this type was from Neath (Canal Side) to Court Sart on 11 September 1935.

Retrospective and critical

Bulleid, H.A.V.. Bulleid era. p.193
Illus. credited to J.R. Bazin: railmotor and trailer at Plympton in 1905: caption noted "They had ample power and rode well: Stanier recalled". Bazin had been sent by Ivatt to the GWR and to the TVR in 1905 to study their steam railcars.
Gibbs, Ken. The steam rail motors of the Great Western Railway. Stroud: History Press, 2015. 191pp.
Both a history of the type with many diagrams, and the project to bring back a former auto trailer into a steam rail motor
Jenkinson, David and Barry C. Lane. British railcars, 1900 to 1950. Penryn: Pendragon, 1996
Lewis, John
Great Western Railway auto trailers: Part 1. Pre-grouping vehicles. Didcot: Wild Swan, 1984, 224 pp.
Great Western auto trailers. Part 2. Post-grouping and absorbed vehicles. Didcot: Wild Swan, 1995, 186 pp.
Great Western steam rail motors. Didcot: Wild Swan, 2004. 308pp.
The above form a magisterial study: although the cart was written before the horse. Includes a great many detailed working drawings, with Swindon numbers of both the engine (motor) and carriage units. Copious notes on liveries, workins, etc..
RCTS. The locomotives of the Great Western Railway. Part 11. The rail motor vehicles and internal combustion locomotives.. 1956.

Unfulfilled projects

Hawksworth Pacific

Allen, C.J. British Pacific locomotives.1962.
Chapter 9 on projected designs does include the "Hawksworth Pacific" where Allen stated that he had seen an outline drawing and "A certain amount of work seems to have been done on this design, but this must have been unofficial, for F.W. Hawksworth, Chief Mechanical Engineer... in a letter to me has dismissed the design as no more than 'a draughtsman's dream'" [KPJ but Hawksworth was a draughtsman!].
Nock, O.S. British locomotives of the twentieth century. Volume 2. 1930-1960.
Even by Nock's meandering standards he takes an inordinate length to state that nothing was announced about the "Hawksworth Pacific". Yet two draughtsman involved had told Nock that they had received "the sketchiest instructions" from the Chief Draughtsman, F.C. Mattingley to design a Pacific to carry a boiler pressure of 280 lb per square inch, with 6 ft 3 in coupled wheels, and as many standard features as possible. Nock noted that a tractive force of 47,000 lbf was being sought and quite a lot of work was done on the boiler which would have carried a dome. The front end was a redesign of that fitted to the King class with streamlined ports and steam passages. According to Nock "higher authority" (presumably the Ministry of Transport) would not sanction such work and Hawksworth was very upset at having his Pacific vetoes whilst Bulleid got away with it. Return to beginning..
Summers, L.A. Fact, speculation and fiction in the case of Mr. Hawksworth's Pacific. L.A. Backtrack, 14, 238-44.
Contains a fairly strong case against any real work being directed towards a Hawksworth Pacific. Very long letter from E. Davies (p. 371) which compares the speculations of Nock, Rutherford and RCTS. Refers to collection of drawings at NRM for Pacific type of boiler based on that of Merchant Navy class. illus.:Speculative drawings of GW locomotives.

Crisis, what crisis? L.A. Summers.
See feature by Rutherford (14 724), Cites a letter received from O.S. Nock on 22 February 1977 wherein he states that he was informed by Sam Ell and Geoffrey Tew that they had worked on the boiler and the Chapelon-style front end, respectively of the "Hawksworth Pacific" and the proposed Pacific was strongly backed by Captain Hugh Vivian, a GWR Board member and a Director of Beyer Peacock, as well as a copper refiner.

Updated 2019-09-11