The Midland Railway’s Locomotive Engineers
The Midland Railway was well served by its four locomotive engineers. They all made significant contributions to the art of locomotive design, this being particularly so with first three.
These three also each introduced a distinctive personal style to their engines such that their creator may be readily identified. By contrast other companies developed what in effect became a ‘house style’, which resulted in more continuity in their locomotive appearance.
Two of the Midland’s greatest rivals, the LNWR and the GWR followed this path. In the case of the LNWR the style appeared in the 1870s under Francis Webb, persisting even after much larger engines were produced in the twentieth century and three locomotive engineers later.
The Great Western adopted certain features identifying its locomotives early in the twentieth century which were retained essentially in undiluted form until the end of the company’s existence half a century later.
Only after the design baton had passed from Mountford Deeley to Henry Fowler was some continuity in locomotive appearance maintained; indeed it would go on to serve the initial decade of the LMS.
On the formation of the Midland Railway in May 1844 Matthew Kirtley was appointed Locomotive Superintendent. At that time the nascent company owned around 100 locomotives, by the time of his death in 1873 there were about 1050. Some of his 0-6-0s, albeit rebuilt, survived to run in BR days but the only Kirtley engine to be preserved 2-4-0 Nº 158A had been withdrawn and set aside by the LMS in 1947.
His locomotives were noted for their robust construction which created a solid foundation for future service, rather than for innovation. Apart from the 880 and 890 classes produced right at the end of his career, all of his main designs had outside frames with the majority double-framed.
Matthew Kirtley supported by the researches of his assistant Charles Markham developed a simple method of burning coal in a locomotive firebox without smoke, which was adopted throughout much of the world and lasted until the end of steam. Changing from coke to coal added ¼ to ½ per cent to the annual dividend.
While this is quite well known, little credit has been given to Kirtley’s works practice techniques and his administration skills. In conjunction with his Works Manager John Fernie, although the title was not used then, he introduced into Derby Works over the period 1858-63 what was probably the first attempts at a limits-and-fits system of gauging in a railway-owned works.
In 1847 he became one of the founder members of the I Mech E.
Samuel Waite Johnson went to the Midland as its second Locomotive Superintendent in 1873 following the death in office of Matthew Kirtley. He stayed until he retired on the last day of 1903. His locomotives were characterized by a beauty of line allied with a mechanical precision and supported by an ‘esprit de corps’ among his men that was second to none and probably unmatched anywhere. The beauty of his engines was enhanced after he decided to paint Midland engines a deep shade of red in place of the previous green.
During his reign the number of locomotives increased from just over 1,000 to nearly 3,000. Despite this huge expansion, and in contrast to the LNWR, which built large numbers of identical engines based on a comparatively small number of wheel arrangements, Johnson, who similarly used only a handful of wheel arrangements, arranged for his engines to be created from a large number of small batches. This approach enabled design improvements to be readily introduced between each batch. As a consequence, while his engine classes each exhibited a certain style that appeared to run continuously, in practice there were subtle differences which made one batch of locomotives slightly different from the next even if they were all considered to be nominally of the same class – modellers beware! In this respect he differed from Kirtley, who also faced with a large expansion in goods traffic essentially made far fewer changes between one batch to another. Although the earlier engineer allowed the manufacturers some leeway resulting in detail differences between one maker and another.
One reason for these detail departures was because the Midland made considerable recourse to the private locomotive builders for the supply of its motive power. Apart from their earliest days its major rivals the LNWR and the Great Western produced virtually all of their engines in house.
Johnson rebuilt many of his predecessor’s engines enabling them to give decades of future services through designing a range of boilers that suited his own and Kirtley’s engines. So carefully and sympathetically was this rebuilding carried out it appeared as if he had been responsible for the complete design producing as it were inside and double framed versions of his designs as if he was unsure which frame arrangement was better.
In conjunction with his very great friend Walter Smith – a locomotive engineer on another railway company – he introduced the most successful form of compound locomotive to run in Britain. This went far in correcting the poor reputation the compound concept was held in in contemporary Britain.
Samuel Johnson was a fine example of the very best in British nineteenth century locomotive engineering; not especially intellectual in his approach but willing to develop ideas empirically in steps, insisting on simplicity combined with a fine eye for beauty.
He was President of the I Mech E for 1898.
Richard Mountford Deeley, who succeeded Samuel Johnson was arguably the most cerebral of the Midland’s locomotive engineers. He developed the compound locomotive introduced by Johnson and turned it into a machine more suited to the general locomotive practice developing in the early twentieth century when multiple-manning of engines was becoming common place instead of one-man one-engine of Johnson’s day. He ensured the compounds could be successfully driven by any competent driver. So good was his improved design that large numbers were built by the LMS having proved their worth in haulage capacity and economy trials against rival classes the company had inherited at the Grouping.
Yet in many ways it was a frustrating time for such a clever engineer. He was prevented from building his magnus opus a ‘super-compound’ 4-6-0 version of the exiting 4-4-0 machine, also his 0-8-0 for goods traffic never proceeded.
He was a nationally recognized expert on lubricants and lubrication co-writing a book with Leonard Archbutt – the Midland’s Chief Chemist – that was a considered for 25 years to be a bible on the subject. He devised a machine for testing the effectiveness (oiliness) of lubricants; also a torsion testing machine used in Derby Works as well as holding a number of patents.
Also in connexion with Archbutt, he developed a water treatment process which softened the water used by locomotives so avoiding, or at least much reducing, the scale formation in boilers.
He had wanted to try superheating in 1904 – two years before its first application in Britain on the Lancashire and Yorkshire Railway – but prevented by a Board that objected to the modest royalty fee charged by Schmidt.
The visual impact of his locomotives was more austere than that of Johnson, which combined with reduced lining and the large numbers placed on the sides of tenders and tanks gave a more severe functional outline perhaps more appropriate for the late Edwardian era.
During the nineteenth century, Midland locomotives were among the largest in Britain. A former President of the Midland Railway Society David Tee wondered how the contemporary Great Western was able to run its passenger trains. E L Ahrons made not too dissimilar comments about the locomotive provision on the LNWR of the same period.
Over a period extending from around 1880 to 1910 the Midland’s civil engineers expended a great deal of time, money and effort into adding additional running lines to its network. This entailed quadrupling existing double track lines so adding badly needed route traffic carrying capacity. Unfortunately, it appears these lines were under-engineered in so far as insufficient allowance was incorporated to accept the long and heavier locomotive of the future. This was combined in some areas by a restriction on the maximum length of lie-by sidings that could be installed which compromised train lengths.
Furthermore, unlike most other companies, many Midland four track sections of line were arranged as goods and passenger not the fast and slow traffic. Goods lines were not operated on absolute block consequently several trains might be occupying the ‘block’ from one signal box to the next. In the immediate environs of London there were six lines – fast and slow lines plus goods lines.
Taken together the Midland’s manner of traffic segregation and operating methods, especially the effective Centralized Train Control introduced in 1907, meant it could avoid building passenger locomotives larger than 4-4-0s or goods engines bigger than 0-6-0s which also got around the weak under-bridge problem.
These factors thus significantly affected the locomotive designs produced under the company’s last locomotive engineer Henry Fowler. He was however able to introduce superheating in 1910 which led to him delivering a very informative and comprehensive paper on the subject to I C E in 1913.
Locomotive design thus consolidated under Fowler but this was due almost entirely to the board keeping a tight control on expenditure and the civil engineering department’s aversion to allowing larger locomotives. To maximize goods line utilization two fully loaded 0-6-0s were coupled together to work trains of around 1500 tons also about the maximum the primitive braking and couplings could safely cope with.
Fowler initiated an extensive programme of rebuilding locomotives, usually in conjunction with Belpaire fireboxes while many of the passenger engines and the most modern design of goods engine were superheated. Superheating increased a locomotive’s haulage capacity and/or its fuel economy. It therefore could be used to produce a more powerful engine weighing little more than the equivalent saturated steam machine.
Some of Fowler’s designs e.g. 483 class 4-4-0 and the 4F 0-6-0 were destined to be built in large numbers by the LMS. They were multiplied because they were very robust, while cheap in construction and repair due to careful design and reliable fittings and auxiliaries. For example when the LMS became part of British Railways in 1948 10 per cent of the locomotives it bequeathed were 4F class 0-6-0s a design introduced by the Midland in 1911.
Samuel Johnson, Locomotive Superintendent.
J N Maskeleyne a respected writer on locomotive matters and an adherent of the London Brighton and South Coast railway observed that the rugged appearance of the front of a Stroudley engine reminded him of its creator’s countenance. It is doubtful a similar conclusion would ever be drawn between S W Johnson and one of his creations, although a case could be made for his predecessor’s work. As C Hamilton Ellis author of The Midland Railway explained:-
“Old Matthew Kirtley had been the rough, jovial, homespun sort of engineer, the man who had once been a fireman, and whose daughter married a title. We see his likeness in a group of his staff, grinning amid all their serious faces, wearing solid-looking clothes that do not fit. His engines, as we have seen, were like him: rugged, not conspicuously elegant, tremendously substantial and lasting. Mr. Johnson was as surely Mr. Johnson as Mr. Kirtley had been Old Matt. He was a pillar of the Church, with a formal beard, neatly attired in discreet dark dittoes, a man of kindly severity and meticulous habit. His work, like himself, was meticulous. A Johnson locomotive was designed like a work of art and made like a watch.”
So where had Samuel Johnson learned and honed his engineering artistry? By the time he was appointed the Midland’s Locomotive Superintendent on 2nd July 1873 he was an experienced engineer having served in various senior capacities on the Great Northern, Manchester Sheffield & Lincolnshire, Edinburgh & Glasgow/North British and finally the Great Eastern. In the course of this varied career he gained considerable experience in producing new locomotives and rebuilding existing ones. Skills which he would develop and exploit to a remarkable degree of refinement following his move to Derby.
In common with other Victorian locomotive engineers he was able to meet fully the company’s locomotive needs from only a handful of wheel arrangements. For fast passenger service he provided 2-4-0s, 4-4-0s and 4-2-2s augmented for slower trains by 0-4-4Ts. Goods traffic was entrusted to
0-6-0s together with 0-6-0Ts and 0-4-0STs – the first two of these were also sometimes to be seen on passenger traffic.
He introduced two new wheel arrangements to the Midland 4-4-0 and 4-2-2, he also brought in a third if the American Moguls are included although strictly speaking they were not his work. The first of these new arrangements appeared very soon after his arrival having very recent experience of building a pair for the Great Eastern. The latter incidentally were the first inside cylinder 4-4-0s to run on an English railway, this arrangement which was to prove very popular had appeared a few years previously in Scotland. While there was a gap following his initial two batches of Midland 4-4-0s before more appeared he was demonstrating his ability to build engines which precisely matched traffic requirements. The bogie engines will have curved better and possessed a superior steam distribution despite having similar dimensions to the 2-4-0s – but their time had not yet come.
Prior to Mr Johnson’s arrival the Midland had made very little use of tank engines, while those few the company did possess had nearly all been introduced in Matthew Kirtley’s last years to suit the requirements demanded of engines working around London. He by contrast favoured tank engines to the extent that some of his 0-4-4Ts were regular performers on some fast trains, particularly between Bradford and Leeds. Indeed his first locomotive design for the Midland – as opposed to completing some outstanding 2-4-0 orders initiated by Matthew Kirtley was for a series of 0-6-0 side tanks for goods and passenger traffic on some recently acquired lines in south Wales. This design proved so useful that it was quickly modified before being multiplied to become the company’s standard shunting engine.
This last sentence leads to another important feature of Samuel Johnson’s design philosophy which essentially comprised three elements: –
(i) Engines were ordered in comparatively small numbers – typically ten, twenty or thirty at a time only in a very few instances did some goods locomotives appear in batches of as many as fifty. Each batch usually displayed changes in design details when compared to the preceding batches.
(ii) Engines built by contractors exhibited detail differences not only when compared to their Derby-built sisters but also with those supplied by other contractors despite being built to nominally the same drawings.
(iii) Engines were long-lived and be this cause or effect, it resulted in the survivors gaining detail changes and/or improvements throughout their existence. As a consequence many engines running in BR or late LMS days had gained a combination of features that made each one unique or very nearly so.
This was Samuel Johnson’s version of a practice which had become established, albeit to a more limited extent, under Mr Kirtley. Interestingly towards the end of Mr Johnson’s tenure, as had also been the situation under his predecessor, the number of variations which could be attributed to (i) and (ii) reduced.
The engineering input necessary in the early development stages in each new locomotive class combined with ancillary developments prompted the greatest number of changes. As the design evolved to maturity there was reduced demand for modifications to be incorporated into the later batches. This honing of each design also ensured that later locomotives, even if in themselves they introduced quite radical departures e.g. piston valves, Belpaire fireboxes, the number of changes necessary before reaching design maturity reduced. However this evolutionary process was then repeated to a greater or lesser extent by his successors.
In this respect the 865 single frame 0-6-0s which carried a B boiler present an interesting example. They were built from 1875 until 1902, but to nine different general arrangement drawings despite them appearing superficially the same and having only two different wheel diameters. The last 295 (34.1%) were to the same drawing which stands in contrast to the earlier Class H. A variant comprising 110 engines yet it necessitated three different general arrangement drawings and involved four different builders.
For those who are interested in locomotive development exploring what happened on the Midland offers such devotees the opportunity of engaging in a cerebral activity resembling something perhaps akin to three-dimensional chess! Indeed, one could go so far as to venture it is almost impossible to be absolutely certain about many aspects concerning Midland locomotive design. Almost each and every component comprising a locomotive was re-considered followed perhaps by revision – sometimes more than once. To complicate the situation further some, but not all, of these later detail changes would be applied retrospectively to earlier engines during their visits to Derby and outstations for overhaul.
It is somewhat risky to be categorical about how much a Locomotive Superintendent as opposed to his Chief Locomotive Draughtsman contributed to any particular design. While it is believed Samuel Johnson had considerably more influence than some we ought not neglect some of the draughtsmen who influenced Mr Johnson’s Midland engines.
Prior to his appointment to the Great Eastern Samuel Johnson had fulfilled the same role on the Edinburgh & Glasgow but prompted by the amalgamation of this railway with the North British he left Cowlairs in 1866 for Stratford taking with him a young Scottish engineer of considerable ability – Walter Mackersie Smith.
Mr Smith’s new post on the Great Eastern was as rolling stock design engineer, although one source does credit him with becoming Chief Draughtsman. Be that as it may, either covertly or overtly he nevertheless contributed to Mr Johnson’s Great Eastern engines. When Samuel Johnson moved to Derby Walter Smith did not follow but elected instead to spend nine years in Japan as the first Locomotive, Carriage and Wagon superintendent of the Imperial Government Railways. He returned to Britain in 1883 becoming Chief Locomotive Draughtsman of the North Eastern Railway. One of his sons, who became a fellow of the Royal Society was named Samuel Walter Johnson Smith.
Mr Smith’s version of three-cylinder compounding is probably his most famous achievement – certainly in respect of the Midland and later LMS railways but this talented engineer also introduced for example a practical design of piston valve with two associated cylinder designs, and transverse water tubes in locomotive fireboxes – an arrangement much used by Dugald Drummond. Some of these, along with other details from the Smith stable (père and fils) would see service on the Midland under Mr Johnson. Apart from use on his employer’s railway some of their ideas also found favour on the Great Central.
The second engineer who was destined to influence the first 15 years or so of Mr Johnson’s Midland engines was Robert Billington. He had previously been William Stroudley’s assistant in charge of locomotive and rolling stock design and construction before becoming in November 1874 Mr Johnson’s Chief Locomotive Draughtsman. Incidentally William Stroudley had been Works Manager of Cowlairs until he left in 1865 to become Locomotive Superintendent of the Highland Railway at Lochgorm taking Dugald Drummond his with him initially as foreman erector prior to his becoming Works Manager two years later. It has been reported these moves were prompted in part at least because of a row whilst all three men were at Cowlairs. Be that as it may, it is interesting to record Samuel Johnson’s only son James, who was for a brief period Locomotive Superintendent of the Great North of Scotland Railway wed Dugald Drummond’s daughter Christine. Their third child was christened Dugald Samuel Waite Johnson.
Following William Stroudly’s premature death, Robert Billington was invited to become his successor taking up his new post in January 1890. Samuel Johnson appointed Thomas Iveson as the replacement Chief Draughtsman. He served until January 1901 being succeeded by Walter John Smith – another of Walter Smith’s sons – who no doubt was particularly useful in the introduction of the Compounds to the Midland.
Arguably Thomas Iveson made a an important but undervalued contribution to the appearance of Midland locomotives during his eleven years. The 0-6-0s, 4-4-0s, 4-2-2s, 0-4-4Ts etc built during his tenure appear to my eye at least, to have been amongst the best lookers of all.
Finally, when considering the men who influenced the appearance of Midland engines under Mr Johnson, one should not overlook the contribution made by the private locomotive builders. Although nominally these engines were built to Derby issued drawings and specifications, many of these builders managed to incorporate their own ‘stamp’ on the final product.
Of these, Edward Snowball as Chief Locomotive Draughtsman of Neilson & C° (he had previously held the equivalent position with Robert Stephenson & C° where he had been responsible for the Midland’s 130 class 2-2-2) had perhaps the greatest outside influence on Midland locomotive appearance during this period. Certainly Neilson & C° appear to have been granted a greater degree of autonomy compared to the others – or maybe the company just assumed it. Perhaps some examples might illustrate this point. The radii of components were altered from Derby’s specification or even added where none had existed before. For example the elegant reverse curve which joined the splashers of 2-4-0s and 4-4-0s was a Neilson innovation. He changed the ‘flat oval’ shape of the cut out in the outside frame behind the front steps into a circle in some tenders while the segment shaped cut outs between the wheels differed slightly from Derby practice. The Neilson Class B 0-6-0s had 16 spokes in their driving/coupled wheels in place of the 14 of the remainder. As a final example, in the case of the 5ft – 6½ins 0-4-4Ts (Class D) Neilson ignored the Derby drawing and deepened the frame between the coupled wheels so that its upper profile was proud of the platform.
Locomotive Identification
In contrast to other companies the Midland did not adopt a systematic or consistent system for identifying its locomotives. While later, in the twentieth century engines would be classified by their power and purpose e.g. 3F, 4F, 2P etc, this system was incapable of identifying a specific batch. This lack of a definitive system meant in some cases the distinction that the company did make between classes could appear rather arbitrary in some cases while in others surprisingly crude.
Disregarding a short-lived early scheme introduced by Mr Johnson, which attempted to identify all the extant Kirtley machines together with the latest Johnson designs in a continuous run of sequential numbers, three main methods were used. Derby-built engines were frequently identified by the order number allocated to that batch i.e. O.xxx, while contractor-built engines were usually, but not exclusively given a class letter designation. Lastly E L Ahrons, an important contemporary writer on locomotive matters, adopted the practice of identifying engines by the number of the first engine (usually) of that class to be built.
Of these latter three, the letter system first being adopted (probably) in late 1873 was intended to be applied to all classes. Thus Mr Johnson’s first 0-4-4T design for the Midland introduced in 1875 was known originally as the Class C but they were quickly referred to as O.1 from the order to build them issued in December 1874.Henceforward the letter designation was limited to contractor built engines with Derby built referred to officially by their order numbers. However, it should come as no surprise to note that some contractor-built batches were not given letter designations being identified in other ways. Examples being ‘Like O.232’, ‘Neilson Goods’ and ‘Like M&GN Class C’.
The method used by Ernest Ahrons was also practised to some extent by the Locomotive Drawing Office, however it elected often to use a ‘more convenient’ number rather than strictly that of the first example built. Thus the 1738 class 4-4-0 (O.554) could also be referred to as 1740 class!
While this may appear somewhat confusing it does become easier if when considering Johnson locomotives running in Johnson condition one uses their pre-1907 numbers.
[AT 4/10/2024]
