North British Railway Study Group Journal Number 140-159
Key to all Issue Numbers

NBR No. 1, an 0-6-0 originally
built in 1870, at Scotland Street.
See Euan Cameron’s article
starting on page 4 for more information.

No. 140 (August 2020)

Obituary Alasdair Alexander (Sandy) Maclean. 3
Born in Edinburgh, Sandy lived in Morningside Road in a flat overlooking the Edinburgh Suburban line. After an education at George Heriot’s School he joined the railway as a Junior Clerk at Newington. He served his two years National Service in the RAF before returning to Morningside Road as a booking clerk. He was successful in being selected for Management Training and one notable job was head of the Coaching Plant section in the Operations department of ScotRail HQ in Glasgow where he made a particular impact on revolutionising carriage cleaning with the introduction of more scientific methods. Living in Greenock, Sandy seldom strayed from his home in his latter years. "Sandy was, in railway terms, without a doubt in my mind one of the most informative and intelligent people I have ever met. Always ready to help he was a true gentleman". Includes a portrtait.

Euan Cameron. Thomas Wheatley’s ‘untypical’ freight 0-6-0s. 4-16
First batches built in 1870-1, before a larger,more homogeneous series appeared in 1873 [already discussed and illustrated in Journal issue No. 138, pp. 9-13]. These "very plain little engines shared the characteristic robustness and solidity of all Wheatley’s locomotive chassis, and after rebuilding most ran until the First World War. A few survived beyond the end of hostilities."
Wheatley rebuilds of R. & W. Hawthorn 0-6-0s of the 64 series, J110 in the Group List, 1921 diagram book 70
When Wheatley succeeded the disgraced William Hurst as Locomotive Superintendent of the NBR in 1867 he faced an urgent need to improve or replace, that part of the surviving stock of locomotives which had been built before the middle 1850s, derived both from the N.B. itself and from absorbed constituent companies. In the case of the North British itself, 70 of the first 71 locomotives had been built by R. & W. Hawthorn of Newcastle upon Tyne. The Hawthorns had a fatal design flaw which went back to their origins in the Stephenson ‘Patentee’ design, the consequences of which became worse as the engines grew larger [see Journal No. 128, pp. 3-4 for a fuller discussion, also D.K. Clark’s Railway Machinery, vol 1 p. 233]. The key document prepared at Cowlairs from 1867 onwards, known to devotees as the ‘Cowlairs 1867 List’ was essentially a record of the necessary replacement, one for one, of the vast majority of this early stock.
Some of the very last NBR Hawthorn 0-6-0s were reconstructed in such a way that their successors counted as rebuilds rather than replacements. They are described here as fully as possible, and it is up to the reader to decide whether ‘rebuilding’ is an apt description or not. The general arrangement of the original Hawthorn 0-6-0s Nos. 64-71 of 1850-1 has not survived, but it is believed that photograph of Edinburgh Waverley East End by the early photographer Thomas Begbie shows one taking water. The engine has heavy outside sandwich frames and – judging by the presence of angled supports from the outside frame to the boiler – appears to have had the slender and ineffective inner framing which stretched from the cylinder block back to the firebox. The firebox was larger in diameter than the boiler barrel, which was domeless. According to the description in the Cowlairs list, these original engines had 17-ft by 24-in cylinders, lined up to 16-in diameter in some cases, and 4-ft 9-in diameter wheels spaced 7-ft 4-in + 6-ft 8-in apart. Some accounts claim that the engines began their existence with 18-in bore cylinders but had to be reduced almost immediately. Sandwich-framed six-wheel tenders were supplied. The ‘rebuilding’ of these locomotives occurred between 1868 and 1872. In association with their rebuilding – but slightly after – a most peculiar renumbering took place, by which (apparently) 64/5/6 were renumbered 9-11, and 70 was renumbered 14. Surely Nos. 67-8 should have become 12 and 13, but that change never happened. No. 67 was rebuilt at St Margaret’s in 1868 with similar wheels and cylinders to the original, with the coupled wheelbase adjusted to 7-ft 4-in+ 7-ft 6-in. A domeless boiler was fitted, similar to the original but with a flush rather than raised firebox. The remainder, 64/9, 65/10, 66/11, 68 and 70/14 were all rebuilt at Cowlairs. The rebuilding list gives them 4-ft 6-in wheels, except that 14 received 5-ft 0-in wheels (an anomaly confirmed by photographs). In all cases the Cowlairs rebuilds had their axles spaced 7-ft 3-in + 7-ft 9-in The outside frames were dismantled and rebuilt with the large plates carrying the axlebox horns reassembled on to a new longitudinal frame.  But with No. 14 it is likely that the hornplates were completely new, as those on the rebuilt engine were trapezoidal shaped with straight sides, rather than the concave curves seen on the others.
All these rebuilds received new inside mainframes, running the length of the engine from buffer- to drag-beam, with additional bearings for the driving axle. The massive structure created by the four full-length frames secured back and front will have done a much better job of keeping the cylinders and axleboxes in correct alignment. The cylinders of all except No. 67 were 16-in bore x 24-in stroke, a standard size for Wheatley goods locomotives and used on nearly all the engines discussed in this article.
The boilers of the rebuilds were quite different from the originals, as the latter will have had a firebox too wide to fit between the inside frames. It is likely that the Hawthorn ‘rebuilds’ received something like a Cowlairs standard boiler of the 1865-1870 period, with an approximately 4-ft 0-in diameter barrel 10-ft 2-in long and a flush round-topped firebox 5' 0" long. A large proportion of Wheatley’s engines received a version of this boiler. In the pre-1871 period such boilers were typically domeless with a safety-valve trumpet over the centre of the firebox crown, with the ‘trumpet’ part located on a square base with a flat top. As first built, they possibly had Salter safety valves, but Wheatley replaced these in the early 1870s with direct-loaded sprung valves entirely enclosed in the trumpet. Some engines (10 and 68 as far as we know) received open-topped dome covers over their safety valves in place of the Cowlairs trumpets, probably purloined from other engines; but these did not enclose a steam dome.
Weatherboards were fitted, sometimes with bent-over tops, and 68 was given a facsimile of an enclosed Wheatley upper section to its cab, probably in the 1880s. The 1870s recreations of the Hawthorns carried reconstructed outside frames to different dimensions from the originals, and all new wheels, inside frames, boilers, and platework. Does this count as a ‘rebuild’ or a new construction? One reason to describe them as rebuilds may have been financial: renewing an engine with the same number as a predecessor allowed the cost to be written down to repairs rather than new building.
Given the (theoretical) age and chequered history of these rebuilds, it might surprise that Matthew Holmes gave four of them a fresh lease of life by rebuilding them yet again: yet he did. The rebuilding in this case involved a reboilering, with the addition of locomotive steam brakes, updated boiler fittings, and a Holmes round cab. No. 11 was rebuilt in 1884, 10 in 1886 and 9 in 1896. 68 was rebuilt at some point around 1900, the precise date not certain. Nos. 67 and 14 were not rebuilt. The boilers were of what is presumed to have been the same size as the 1870s versions, but with all Holmes’s characteristic fittings. The barrels were 10' 2" long and the fireboxes 5' 0" long. There were 171 tubes x 1¾-in diameter, tube heating surface of 817 ft2., firebox heating surface 83.5 ft2., total 900.5 ft2. Boilers with these identical dimensions were also fitted to the Longbacks and the No. 1 series 0-6-0s (as below) when rebuilt. The tenders attached to the rebuilds were generally borrowed from other 0-6-0 locomotives and varied greatly. It must be presumed that as the engines left Cowlairs they were simply allocated whatever tender happened to be available. Some of the tenders from R. & W. Hawthorn locomotives long outlived their original locomotives, coupled to other engines altogether; yet curiously very few Hawthorns actually kept their own tenders.
Wheatley No. 59 series of reconstructed 0-6-0s, sometimes known as Longbacks, J124 in the Group List, 1921 diagram book 71
Between 1868 and 1869 Wheatley also constructed eight double-framed 0-6-0s at Cowlairs, in many respects very similar to the rebuilt Hawthorns. These engines were assigned random numbers of dismantled locomotives and have therefore been regarded as replacements rather than rebuildings. The most striking difference between these engines, known unofficially as Longbacks and the rebuilt Hawthorns is that most of the engines built new (with the exception of 154/5) received substantial, deep, slotted frames of continuous metal plate outside the wheels as well as inside. The resulting mainframes will have been exceptionally robust, and the locomotives worked for many years. 154 and 155 had composite outside frames with hornplates riveted to a longitudinal iron beam, more in the manner of the Hawthorns but with a different profile to the hornplates. Photographs of 135, 154 and 155 survive in their original condition.
Miscellaneous locomotives rebuilt with double frames and outside cranks in the 1860s-1870s
Besides the more or less identifiable ‘classes’ of outside-framed 0-6-0s, there was also a handful of individual locomotives, mostly reconstructed from Hawthorn material, but associated with different original company owners and often with rather unclear histories.
No. 17
Double-framed 0-6-0 No. 17 appeared from St Margaret’s works in 1869. Theoretically it was a ‘rebuild’ of one of the original Hawthorn 0-4-2s supplied to the N.B. before it opened, but in this case little or nothing of the original engine can have been incorporated in the rebuild. No unambiguously identifiable image of the locomotive in its 1869 condition survives; but one may assume that the boiler and fittings were similar to those of No. 67. The outside frames, however, were of the same deeply slotted continuous plate as on most of the Longbacks, which were of course Cowlairs engines. 17 had 4-ft 7" wheels spaced 7-ft 6-in+ 7-ft 6-in. Holmes rebuilt 17 in either 1896 or 1898 (sources differ) and it was attached to a tender purloined from a Neilson 90 Class 2-4-0 of 1861 (all but one of which had by that time been scrapped). It was long associated with Thornton shed, where it seems to have been used on service trains.
No. 50
No. 50 was an exceptional survival from an earlier series of Hawthorn 0-6-0s, Nos. 47-54. It was comprehensively rebuilt in early 1869 in the same way as 67, retaining its distinctive curved outside frames. As rebuilt it had 4' 2" wheels with conventional spokes, spaced 7-ft 6" + 7' 6". It was rebuilt in 1882 (probably by Holmes although it retained some Drummond aspects to the boiler fittings) and lasted as No. 1030 to late 1910. It was attached to one of Wheatley’s scrap 4-wheel tenders with a chassis constructed of very thick wooden baulks at the sides and ends, to which strengthening plates were riveted. It served as the Carlisle Canal trip pilot for some time.
Two others of the same series, 47 and 52, were rebuilt at Cowlairs in 1874 and St Margaret’s in 1868 respectively. Neither received a second rebuilding and they appear to have escaped the attention of photographers. Nos. 137-9
137-9 were three 0-6-0s supplied by Hawthorns to the Edinburgh, Perth and Dundee Railway in 1851. 137 and 138 were rebuilt at Cowlairs in July and May 1868 respectively. They had 5-ft 0" wheels and typical Hawthorn outside frames, probably spaced 7-ft 2-in + 6' 6-in. Neither received a second rebuilding. 137 retained a massive six-wheeled Hawthorn tender, and was based at Dundee.
Nos. 249-50
Two Neilson 0-4-2s were supplied to the Edinburgh and Glasgow Railway in 1851. According to the Cowlairs 1867 List, they were replaced by two engines of the same numbers built at Cowlairs in December and October 1867 respectively: but the list describes them as 0-4-2s with 5-ft 0-in wheels spaced 7-ft 9-in + 7-ft 1-in. 250, at least, was definitely a 0-6-0 incorporating Hawthorn material, though the reported wheelbase may well be correct. In the early 1890s it was photographed at Cowlairs with a domeless boiler with lock-up safety valves over the firebox crown, and a box lower section to the cab surmounted by a partial upper section of vaguely Drummond appearance. In 1896 it was rebuilt by Holmes with his usual fixtures and fittings.
250/873/1073 had a Stephenson 4-wheel tender before its second rebuilding and a Wheatley 6-wheel 1,800 gallon tender when photographed at Ladybank in the 1900s.
No. 280
280 was somewhat unusual although resembling those discussed above in general layout. It was reportedly built at Cowlairs in 1865 as a 0-6-0 with 4-ft 9-in wheels spaced 7-ft 7-in + 7-ft 6-in. It had the usual Cowlairs boiler of the period, but the flat weatherboard had an arched top, and bulged outwards around the spectacle plates before narrowing down to join the lower section of the cab. This style of weatherboard appears also to have been used on 2-4-0s 235/6/9 built around the same time. It received a 6-wheel tender similar to that attached to 68. 280 lasted long enough to receive its 800 series number after 1895, though it was not rebuilt by Holmes. The No. 1 and 2 series inside-framed 0-6-0s built with 4-ft 2-in wheels and 7-ft 3-in + 7-ft 9-in wheelbase, 1921 diagram book 41
The final classes to be reviewed here were two batches of goods 0-6-0s of great simplicity and solidity, built for slow mineral traffic around 1870-1871. Unfortunately some inaccurate information in print makes it at times difficult to distinguish between the two series, which differed in wheelbase from their first building onwards.
Twelve 0-6-0s were built, mostly in 1870-1, with solid slotted inside frames similar to those on 0-6-0ST No. 220 [see issue No. 138 pp. 6-7] with which they shared wheel sizes and wheelbase, 4-ft 2-in wheels spaced 7-ft 3-in + 7-ft 9-in. As one of the class was given the number 1 released by the scrapping of the original Hawthorn 0-4-2, they became known as the 1 class. The Wheatley 0-6-0 so numbered retained that distinction until it became 1150 and its capital number was taken over by a Reid 4-4-2T.
According to some records, including official ones, the first of the class, No. 251, was built as early as 1867, but its design similarity to the other engines makes this rather unlikely. The 15-ft 0-in wheelbase engines had inside frames only, no brakes on the locomotive, very simple domeless boilers with safetyvalves over the firebox crown, and simple weatherboards above the cab side-boxes. The cylinders of all these engines were 16-in x 24-in. They received Wheatley’s 1,800 gallon tenders. Matthew Holmes rebuilt all twelve locomotives between 1894 and 1900, slightly increasing the wheel diameter to 4-ft 3-in with thicker tyres and raising the running plate height accordingly. The rebuilds received the same boilers as the rebuilt Hawthorns and Longbacks, and were fitted with locomotive brakes for the first time. The tenders were not altered.
Between October and December 1871 Wheatley added 6 more locomotives but these from first construction differed from the original 12. The latter six engines (known as the ‘2’ class) had wheelbases 6-in shorter than the previous twelve, at 6-ft 9-in + 7-ft 9-in. While photographs of this class in original state are extremely rare, a picture of 223 shows it to have had a boiler with a dome on the centre of the barrel. By inference from the data from the rebuilding, it may be supposed that these boilers had a barrel 9-ft 7-in – 9-ft 10-in long and a firebox 5-ft 0-in long. The 2 class later formed the basis for the much more numerous and better recorded ‘430’ class, which they resembled in wheelbase and cylinder sizes.
Matthew Holmes rebuilt all six between 1887 and 1901. They received the shorter boilers already designed for the rebuilt Beyer, Peacock locomotives formerly of the E. & G.R., with 9-ft 7-in boiler barrels and all Holmes’s usual fittings. Most if not all of these locomotives appear to have been attached to Wheatley 1,800 gallon six-wheel tenders throughout their existence.
Most of the 1 and 2 class lasted until withdrawal between 1913 and 1915, by which point they were over 40 years old. Some lasted in traffic until 1920, and 1196 ex 252, with possibly others, survived long enough to carry its duplicate number in large control numerals on the tender. By that stage most of these early locomotives were in use as yard pilots or for very short trip workings. As an example, when No. 2 was a pilot at Portobello it was also regularly assigned to a trip goods to Dalkeith. There are oral traditions concerning the allocations of many of these locomotives, but as the traditions often contradict each other, they are omitted here as unreliable. It is remarkable that well into the 20th century North British Railway yards would have seen locomotives shunting and running short trip goods, which had their origins dating back to the 1860s or even the 1850s. The N.B. was a very cautious and parsimonious railway (by and large) and Matthew Holmes in particular seems to have been determined to make good use of any serviceable material that could be found. The only critique that one might make of keeping such venerable antiques in traffic was that when so many 1860s locomotives had to be withdrawn within a few years after 1910, the N.B. was left with a shortage of locomotives, which plagued it until after grouping even despite the proliferation of more modern engines. Meanwhile these quaint old engines kept the cadre of N.B. engine photographers well occupied

Thomas Begbie photograph of NBR Hawthorn engine of the 64-71 series, at Waverley East c.1860.

4

No. 10, previously No. 65, at work, marshalling goods train. Note dome cover probably taken from another engine, and Dübs tender. 1880s?

5

No. 1016, previously No. 66, at Ladybank. This shows the locomotive as rebuilt by Holmes.

5

Wheatley Longback 0-6-0 No. 135 before rebuilding, with Dubs tender: dark olive livery applied in Drummond period (Euan Cameron coloured drawing).

6

Wheatley Longback 0-6-0 No. 135 after rebuilding by Holmes in 1894: fully-lined out Holmes livery as depicted in multiple photographs from period. Locomotive brakes (not shown here) added some time after engine rebuilt. (Euan Cameron coloured drawing).

6

NBR No. 68 at Kilsyth Old Station. View shows wealth of other detail including brake van and wagons and semaphore signals

7

Wheatley ‘Longback’ 0-6-0 No. 135 as running before rebuilding. This is the condition of locomotive seen in photograph taken at Anstruther in 1887. Note short sloped cab roof wrapped around curve of weatherboard.

8

Longback 0-6-0 No. 135 after rebuilding in 1898. Note details of construction of mainframes, perpetuated from original condition but very different from 135, and Wheatley short-wheelbase tender.

8

On left No. 135 after rebuilding with ‘New NBR engine’ alongside is Atlantic No. 878 Hazeldean

9

No. 135 in unrebuilt condition, photographed at Perth.

9

No. 155 as rebuilt by Wheatley, at Anstruther in January 1887 10
No. 1018, previously No. 17, with breakdown train at Thornton, some time between 1901 and 1914. Leftmost figure on ground possibly Christopher Cumming 10
No. 50, a St. Margaret’s rebuild of 1869. Note the distinctive St Margaret’s works plate on the frame. 11
No. 1 of 1870 in original condition. It was later rebuilt, in 1898. Note steam pipe leading down from firebox crown towards to the injector has been removed, as has the whistle, so the engine was probably under repair. Dome cover was later addition. 11
0-6-0 No. 1 as running in the Drummond and Holmes period before rebuilding. While a photograph taken of this engine in the early 1890s shows a dome cover over the safety valves, the form of safety valve cover shown here was the original, and appears in multiple other photographs of the class. 12
No. 1196, previously No. 252, with control number on tender. Possibly at Armadale in 1916, driver Thomas Marshall. (Hennigan collection) 13
No. 2 at St. Leonards after rebuilding. Note young visitor on running plate. (Hennigan collection) 13
No. 2 as running soon after rebuilding in 1888, in the livery of the period 14
No. 17 as renumbered 818 in full Holmes livery. Note low-pitched boiler, steam brake for locomotive, and tender from Neilson 2-4-0. 14

Carriages. 17

Train of four-wheel carriages hauled by 4-4-0T No. 1465 (later LNER Class D51), at Abbeyhill, heading for Edinburgh Waverley. (NBR Photo Archive 20568
6-wheel non-vestibuled 6-compartment third class carriage at Meadows Yard, carrying LNER No. 31433. (Hennigan Collection)
NBR bogie non-vestibuled lavatory semi-open third class non-gangwayed carriage No. 31246 (former NBR No. 1246), NBR 1908 diagram 6,NBR 1920 diagram 6, LNER diagram 6B, built for West Highland Railway. Vehicles originally had large ‘picture’ windows in the centre saloon which lacked ventilation, and altered to form shown. (Real Photographs)

Donald Cattanach. General Pasley and the inspections of the NBR in 1846 and 1847. 18-27
Adds much to our understanding of General Pasley and the development of early railway inspections for the Board of Trade, especially under Dalhousie. It also summarises the effects of the British climate upon a "difficult" section of the East Coast Main Line which is prone to flooding and severe coastal erosion; both factors being exacerbated by Global Warming. Includes a reproduction of the letter sent by Pasley to Dalhousie on 18 May 1846 recording his (first) inspection of the line. This is also intersting in that it also records his inspection of the tubes being manufactured in Manchester for the Menai crossing. Pasley was accompanied on the inspection by Charles Jopp, Resident Engineer, and his assistant, and by James Bell, also then in the employment of the Engineer of the line, John Miller, but who would be appointed the NBR’s Resident Engineer on the opening of the line (and, later, its first Engineer-in-Chief).
Pasley's inspection took place within a day and he found some of the structures extremely sound, but others were very poor including two bridges which were required to be rebuilt. Some included rubble stone. The line was not sanctioned to be opened. Pasley returned on 17 June 1846 to inapect the formerly unsound structures and did permit the opening on 22 July. On 31 July subsidence at Markle caused the 04.30 southbound mail to derail. Both the drivver and locomotive superintendent Robert Thornton who was also on the footplate escaped with severe bruising. Thornton also drove Pasley around by locomotive. On 28 September the area experienced torrential rainfall and the line was  severed in several places, notably at the crossings of the Eye Water and the Tower Burn south of  Cockburnspath.
Includes the Penmanshiel Tunnel tragedy on Saturday 17 March 1979, when workmen lowering the base of the tunnel were entombed in a major rockfall wwhich led to the railway being diverted onto a new alignment.

General Pasley portrait (colour) 11
The North British Railway and other lines in 1847. 21
Part of Berwickshire, showing locations mentioned in the text 21
Lamberton Holdings: part of railway wall in foreground, indicating original track alignment, swept away by landslip above farm at Lamberton Holdings* 22
Meg’s Dub on 20. June 2020: name possibly relates to arrival in Scotland on 1 August 1503 of 13-year-old Margaret Tudor, daughter of Henry VII of England, for her marriage to James IV of Scotland.. 23
Another view of Meg’s Dub. 24
Caravans occupying original track alignment at Marshall Meadows. Satellite image (Google Earth) 25

* The present alignment is much closer to the cliff face, which is covered with netting to stabilise the bank. The A1 road runs at the top of the cliff. On Saturday 20 June 2020, the 1E11 from Edinburgh to St Neots is traversing the 90mph section, nearing the Scottish Border.

[J37 0-6-0 No. 64582 with train of 20 bulks (Leith General Warehousing grain wagons) near Morningside Road]. 48
Caption suggests train may have been en route from Thomas Bernard's maltings at South Leith to T. & J. Bernard's Brewery next to Gorgie East station in 1950s.

Alan Simpson. West Fife pits and the NBR: Part 6 – the Cowdenbeath Coal Co. era. 28-37
Area comprises parishes of Ballingry and Beath and town of Cowdenbeath and village of Lumphinnan. Kelty descibed in Journal 136 and Lochgelly in Journal 139.
Predecessors to the North British Railway: The public railway system came to the area in the late 1840s with the opening on 4 September 1848 by the Edinburgh & Northern Railway (E&NR) of the line from Thornton to Crossgates. This line, which ran to the east of the present day town of Cowdenbeath, had stations locally at Cowdenbeath and at Crossgates. The E&NR changed its name to the Edinburgh, Perth and Dundee Railway (EP&DR) from 1 August 1849. The line was later extended to Dunfermline on 13 December 1849 where it made an end-on junction at a joint station with the Stirling & Dunfermline Railway later called Dunfermline Upper station [following the closure of what was the eastern part of the Stirling and Dunfermline the local Sheriff Court and a retail park are on its site].
The next local public railway development (lying between Cowdenbeath and Lochgelly) was the opening to traffic in June 1860, of a junction, called Lumphinnans Junction, between the EP&DR and a new local railway company called the Kinross-shire Railway. This later line headed southwards from Kinross, where it had made an end-on junction (and had a joint station) with yet another local line called the Fife & Kinross Railway which The EP&DR itself in turn was vested in the NBR on 1 August 1862 and this brought the NBR to Fife. Much later, the public railway layout of the area was transformed by the building of new lines which were part of the overall Forth Bridge approach railways which opened in the early 1890s. Their construction was undertaken by the NBR as part of the creation of a continuous double-tracked trunk route from the Forth Bridge to Perth (in part by upgrading existing lines but in others by building entirely new stretches of line). In Fife and Kinross-shire these lines were:
• An entirely new line from the northern landfall of the Bridge to Inverkeithing;
• An entirely new line from Inverkeithing eastwards to Burntisland where it joined the former EP&DR;
• Upgrading to main line standard the northern section of the former Dunfermline and Queensferry Railway (from Inverkeithing to Townhill Junction on the Thornton to Dunfermline line);
• An entirely new line from Cowdenbeath Junction to Kelty;
• Upgrading to main line standard of the line from Kelty to Kinross, Milnathort and Mawcarse on the former Fife & Kinross Railway section;
• An entirely new line from Mawcarse and through Glenfarg to Bridge of Earn.
The ‘Kelty Loop’ (Cowdenbeath Junction to Kelty). From 2 June 1890, a completely new main line opened which left the existing NB Dunfermline to Thornton line at a point named Cowdenbeath Junction (later renamed Cowdenbeath South Junction) on the western margin of Cowdenbeath and headed north to join the existing Lumphinnans Junction to Kelty route at what was now called Kelty South Junction. Included with this new line was a railway station named Cowdenbeath (New), located just off the High Street and almost in the centre of Cowdenbeath itself . The NBR renamed this original station Cowdenbeath (Old) as from 1 June 1890 and it remained open for passenger traffic until 31 March 1919 and from then on until 1 January 1968 for goods traffic only. These dates are per the late D.M.Lindsay’s NBR Chronology. Cowdenbeath (Old) survived in use for miners’ work trains for a period after its closure in 1919 to ordinary passenger traffic. These miners’ trains appear in the Working Timetables and typically ran from Dunfermline Upper station to Kelty via Lumphinnans Junction in the early morning (around 6 am) with a return working in the mid afternoon after the end of the day shift at the various pits. Quick mentions that an engine based at Kelty worked miners’ trains which called here long after public closure’ and refers to Locomotives of the LNER: Part 8B and on page 66 of that work it is noted that ‘Dunfermline shed also numbered a dual-fitted J83 amongst its stock, No. 9831 [it became BR No. 68478] which in addition to the normal shunting and trip duties undertaken by the class at Kelty was also able to work short-distance passenger trains (provided for miners) calling at Cowdenbeath (Old) Station long after its closure to normal traffic...
Early 20th century railway developments in the area
Lumphinnans East and North Junctions
New loops were installed in April 1901 at both Lumphinnans East and Lumphinnans North Junctions on the line from Kelty South Junction. This line was in fact the original Kinross-shire Railway and until the recent construction of the Kelty Loop had been the only route northwards from west Fife (excluding the NB’s freight-only former West of Fife Mineral Railway which ran to the north-west of Dunfermline) to Kelty, Kinross and Milnathort.
Cowdenbeath Loop (see Map 1)
The original NB main line westwards from Lumphinnans Junction to Cowdenbeath (Old) and Cowdenbeath South Junction had, by the late 1890s, a number of separate colliery sidings, such as those serving the Donibristle colliery and the Raith pits of the Lochgelly Iron & Coal Co. (see Part 5 of this series in Journal No. 139) connected to it. In addition to these there was the Cowdenbeath gasworks siding and finally, the NB’s Kirkcaldy and District Railway opened in March 1896 and ran from Invertiel Junction near Kirkcaldy to Foulford Junction near Cowdenbeath.made a junction with the original main line at Foulford Junction. As a result of the growth of traffic to and from the various pits a diversionary route was built to allow through traffic to by-pass this congested section of the original route. The new line (named the Cowdenbeath Loop on large scale OS maps) ran eastwards from Cowdenbeath North Junction (which lay north of Cowdenbeath (New) station) to Lumphinnans Central Junction, where it joined the original main line from Thornton and Lochgelly. Today, the Cowdenbeath Loop remains open for traffic on the Fife Circle trains. The opening date for this new line was January 1900 and this is given on page 106 of Thornton railway days; edited Lillian King (Windfall Books: 2000). Several other writers state that the Loop was opened to traffic in March 1919 but:
• The Cowdenbeath Loop is first shown on Sheet 40 (Kinross) of the 1 inch to 1 mile OS map revised in 1903/04 and published in 1906;
• It is also shown on the Sheet XXXIVNE of the 6 inch to 1 mile OS map revised in 1913 and published in 1920.
I suggest that the date of March 1919 was that from which all passenger traffic in the area heading either east to Thornton Junction or west to Dunfermline was re-routed via the Cowdenbeath Loop and Cowdenbeath (New) station. The Cowdenbeath (Old) station had closed to passenger traffic as from 31 March 1919 and what had been the original main line (between Lumphinnans Central Junction and Cowdenbeath South Junction) was now used mainly for freight traffic but

Cowdenbeath Coal Co. wagon, No. 856 (4-plank: 8 ton capacity? Lettered Lumphinnan Collieries, Fife C C C Ld 28
Map: NBR (later LNER) lines and private mineral lines in different colours extracted from Ordnance Survey One-inch Popular edition, Scotland, 1921-1930, Sheet 68 - Firth of Forth. 1928 29
Map: Christie & Co’s Iron Works. Edinburgh Perth & Dundee Railway main line and private mineral lines in different colours extracted from Ordnance Survey 6-inch First Edition, Fife and Kinross Sheet 31 1856. Survey date: 1854. 31
Map: Lumphinnan Iron Works and No. 1 and No. 7 Pits. NBR main line and private mineral lines in different colours extracted from Ordnance Survey 25-inch Second Edition, Fife and Kinross Sheet XXXIV.NE Publication date: 1896. Re-surveyed: 1894. 33
Cowdenbeath Coal Co. Ltd. 8 ton wagon No. 950, built RY Pickering, with spring buffers and steel underframe 34
Map: Cowdenbeath area. NBR lines and private mineral lines in different colours extracted from Ordnance Survey 6-inch Second Edition, Fife and Kinross Sheet XXXIV.SE shows the NBR’s two main routes in the area, the old line with Cowdenbeath Old Station and the new line with Cowdenbeath New station. Extracted from Ordnance Survey 6-inch Second Edition, Fife and Kinross Sheet XXXIV.SE 1896. Re-surveyed: 1894. 35
NBR coal waybill from Cowdenbeath colliery (document from Lindsay Horne Collection, donated to NBRSG Archive) 37
NBR coal waybill from Hill of Beath Coal and Fire-Clay Works (document from Lindsay Horne Collection, donated to NBRSG Archive) 37
NBR coal waybill from Donibristle colliery (document from Lindsay Horne Collection, donated to NBRSG Archive) 37

John McGregor. The lost Esplanade. 38-43
The first serious attempt to reach Fort William, by the the Fort William Railway (1862) would have diverged from the Inverness & Perth Junction at Etteridge (Glen Truim) or Newtonmore, running by Loch Laggan and Glen Spean. Thomas Bouch made a preliminary survey; but he could not persuade the landowners who had commissioned him that they must seek outside capital to supplement their own slender resources. The Glasgow & North Western Railway, a bill for which was lodged in 1882-3 was a highly ambitious and blatantly speculative failed becuse the landowners were not prepared to allow railway construction on their land. The engineer for this line was Thomas Walrond-Smith. The West Highland Railway was more fortunate in that the Napier Commission Report in 1884 had prepared the way for a railway to Fort William and onward to an Atlantic port at Roshven with the prospect of Treasury assistance. The route from Craigendoran to Fort William was engineered by Charles Forman, was approved in 1889 and opened in 1894. It is treated here very much as a fait accompli. The promoters carefully disciplined their parliamentary presentation; before their Bill came to Parliament, they obtained the North British Company’s promise of a guarantee and working agreement; and they secured declarations of support, or at worst neutrality, from every major proprietor along the route. Moreover, both public and parliamentary opinion was broadly sympathetic. A 30-mile extension to Loch Ailort (which the North British did not include in their guarantee), together with a new harbour at Roshven, was added to the West Highland Bill — and by so doing the promoters staked a claim for subsidy. Though the West Highland’s Roshven arm was rejected in the House of Lords, the Commons Committee concluded that the proposed line to Fort William was justified, both on its own merits and as a step towards a third railhead on the west coast, supplementing Oban and Strome Ferry. The Highland Company, and the Caledonian too, miscalculated their opposition — they had expected the West Highland Bill to fail comprehensively on the vexed question of government aid, and were slow to realise how far the North British were already committed.
This article's main thrust is on the Fort William terminus and a possible link to the pier by a tramway, thus creating an esplanade.
There is some suggestion that the North British sought primarily to intersect the Callander & Oban, at the same time pre-empting any independently promoted cut-off to Crianlarich which might pass into Caledonian hands. A railway onwards from Crianlarich and across Rannoch Moor into Lochaber was in itself a dubious proposition. But running powers to Oban might not be granted, and Fort William offered a tempting bridgehead at which to ‘wait and see’, pending further advance — whether to Inverness by the Great Glen, as most commentators expected, or to the west coast, if government support were first assured. To forestall the former became the Highland Company’s priority. Judging the battle lost when the West Highland Bill, shorn of Roshven, passed the House of Lords, they negotiated the Great Glen Agreement, or ‘Ten Years Truce’ and ceased their opposition in the Commons (where the Caledonian fought on to defend Oban). By this Great Glen treaty, any extension of the West Highland towards Inverness was postponed for a decade after the commencement of traffic to Fort William. One more point must be made. The Callander & Oban Company, worked at cost by the Caledonian, enjoyed a meaningful measure of autonomy; but the West Highland became, almost from the outset, ‘the North British by another name’, and the promoters’ early pledges to their supporters, all along the route, lay at the discretion of their paymaster-patron.
As authorised in 1889, the West Highland would have entered Fort William through crofting land along the River Lochy, crossing the River Nevis on a causeway to reach the old fort, the prospective site of the passenger station. A new seawall along Loch Linnhe was to carry a connecting tramway to the town pier; and, on the understanding that this would remain a tramway, the burgh commissioners framed an enthusiastic petition-in-favour. They expected to obtain an open promenade-cum-carriage drive behind the seawall, with tramway traffic limited to an unobtrusive shuttle, linking station and steamers. Edinburgh solicitors MacRae, Flett & Rennie, the principal agents for the promoters and subsequently for the West Highland Company, made no commitment in so many words but they gave every assurance that the ‘interests of Fort William’ would be kept in view. The fort was acquired by Campbell of Monzie, whose wife, Christina Cameron, had inherited the Lochaber estate of Callart and was feudal superior of the burgh and later she made over the fort site to the West Highland Company. Forman acknowledged that, in taking the West Highland across Rannoch Moor and down Loch Treig into the Spean valley, he had copied the drove-road proposed by Thomas Telford at the beginning of the 19th century.
During the interval, when it seemed that the Callander & Oban might halt permanently at Tyndrum, narrow-gauge feeder lines had been suggested, both from Oban and from Lochaber. One such was the Fort William, Ballachulish & Tyndrum Railway (1874), running by Glen Coe. This probably speculative scheme had progressed to a notice of intent — only to fade away when standard-gauge construction onwards to Oban was resumed. It would have terminated at Fort William pierhead, half-a-mile from the fort, entailing only the demolition of decayed property in the town’s west end.
The railway was  almost complete and ready for inspection, but the terminus problem remained unresolved and the Board of Trade inpector was r equested to intervene. This was Major Marindin and his methods marked a great advance since those of Pasley mentioned elsewhere in this Issue. With nothing resolved, Forman departed to accompany Marindin on his end-to-end, week-long (and prospectively final) examination of the West Highland line. MacPhee was sent in pursuit with amended proposals but returned empty-handed. Though he had found engineer and inspector at Tyndrum, they were preoccupied by an accident in which the fireman of a ballast train had been fatally injured. On the evening of Monday, 9 July the inspection party reached Fort William – and that very afternoon the President of the Board of Trade, facing questions in the House of Commons, had promised mediation. In consequence, a telegram awaited Marindin at the Alexandra Hotel – he must use his best endeavours to bring town and railway company to a new accord. Provost Young and town clerk Fraser, with MacPhee in attendance, presented themselves that night ‘at half past ten o’clock’, to request that Forman resume negotiations there and then. Not unreasonably, the engineer declined.
During Tuesday Marindin examined the entire layout at Fort William. He also revisited the line along Glen Spean, where the stations at Inverlair, Roy Bridge and Spean Bridge had received only brief attention the previous day. On Wednesday he heard both the commissioners’ submission and Forman’s counter-argument. The inspector’s verdict, announced on his return to London, showed careful balance (or skilful fence-sitting?). Without condemning the pierhead station, he judged the old fort the better site. Acknowledging that Fort William had been deceived, he proposed to interpret the town’s protection clause generously, in respect of public passage across and alongside the seawall line. But he also pronounced that a tramway as first intended could not have been made compatible with the open promenade-cum-carriage drive which the commissioners still cherished. He would have prescribed thorough fencing, or imposed restrictions severely limiting public access.
In his own mind, Marindin had resolved not to pass the barely finished West Highland for traffic before his task was two-thirds done – his interim memorandum to this effect was written at Tyndrum. However, he hinted that he would be indulgent in everything not absolutely required for safety on his re-inspection a few weeks later. Though it proved a very near thing, on Friday, 3 August he declared the line ready. With Fort William and Lochaber determined to celebrate, the foreshore quarrel was for the moment set aside. Opening day saw a double celebration in that the West Highland Mallaig Extension had just secured Parliament’s approval – though the Treasury’s input had yet to be confirmed. Thus the foreshore dispute would be resumed in a new context, which also included the collapse and precarious reinstatement (1894-5) of the Great Glen Agreement. And the terms eventually accepted by the Fort William commissioners in 1896 would be bound up with the West Highland Ballachulish Extension, authorised that same year but never to be begun. These are matters for another article.

Map: Glasgow & North Western Railway, 1882-3. Note connecting spur across Strathfillan to the Callander & Oban at Tyndrum and the crossing of Loch Leven at the Dog Narrows, not Ballachulish Ferry. Based on J & W Emslie Official Railway Map of Scotland. 1927. 38
Fort, Fort William in late 19th century: old barracks building converted to houses: retained by NBR but demolished by LNER. 39
Lucas & Aird pug at the half-demolished fort 40
Modern view showing how the railway squeezed past the Nevis Distillery, where a gable had to be rebuilt. The site has since been redeveloped 40
Fort William area, showing authorised line and Roshven extension and deviation and Banavie branch extracted from Ordnance Survey 6-inch Second Edition, Inverness-shire - Mainland Sheet CL 1904. Date revised: 1899 41
Seawall and railway c.1900. NBR’s Tweeddale Place tenements in centre of view; passageway through wall can be seen to right (coloured image) 42

Grant Cullen.  The North British Railway and the Great War: organisation, efforts, difficulties and cchievements. 44-
When, shortly after the outbreak of the war in August 1914 and the successful despatch of the first Divisions of the British Expeditionary Force to France, it became obvious that railways were to play a part of primary importance in the development of the conflict, that the war, in fact, was to be a "Railway War". More than four years later British railways had accomplished, as a result of their activities which had taxed their energies and resources to the utmost extent, and had exercised a considerable influence on the movements and achievements of the British forces, if not on the actual course and outcome of the war itself.
The construction of railways, in time of peace, to serve the purposes of war, offensive or defensive, was first advocated in Germany in the 1830s, becoming that country’s policy, although the need for organisation, directed to the provision for the building, repair, destruction and working of railways and for the regulation of military traffic in general under war-time conditions was not fully realised in Europe until after the American Civil War of 1861-1865. In 1866 Prussia established a Field Railway Section (Feldeisenbahnabteilung) which was eventually to develop into a comprehensive scheme of preparation for war by organising every possible phase of military rail-transport and leaving nothing to chance that could be foreseen and provided for in advance.
With the eastern part of its system stretching along the shores of the North Sea from Berwick-upon-Tweed to Aberdeen; with its trunk lines radiating from Edinburgh to Carlisle, Perth, Stirling, Glasgow, Fort William and Mallaig and with its direct connection with the further north of Scotland and the railways of England, through its association with the Highland Railway at Perth, its partnership with the Great Northern and the North Eastern in the East Coast Route between London (King’s Cross) and Scotland, its cooperation with the Midland in respect of the Waverley Route, via Carlisle, Galashiels and Edinburgh, the North British Railway came into immediate prominence as one of the vital means of communications in Great Britain for the purposes of The Great War.
There were, however, various special reasons that arose, which tended still more to accentuate that fact. It was considered then that it was quite within the range of possibilities that an invasion of the country by the enemy might be attempted on the East Coast of Scotland. Hence the NBR was called upon in the earliest of days to convey to their appointed destinations the troops to be amassed for defensive purposes along the coast. Many military training centres were, also, set up within convenient distances of NBR lines, their location being, no doubt, inspired to a certain extent by the idea of having more men available in case the enemy should attempt a landing. Much of this fear of invasion, particularly amongst the general populace, had been driven in the early 1900s by a series of ‘spy’ novels, the best known of which was ‘The Riddle of the Sands’ by Erskine Childers (Childers subsequently lost his life in the Irish Civil War in November 1922). As described in its author’s own words, Riddle of the Sands was written as ‘... a story with a purpose’ written from ‘a patriot’s natural sense of duty’, which predicted war with Germany and called for British preparedness. The whole genre of ‘invasion novels’ raised the public‘s awareness of the ‘potential threat’ of Imperial Germany. Although a belief has grown that the book was responsible for the development of the naval base at Rosyth, the novel was published in May 1903, two months after the purchase of the land for the Rosyth naval base was announced in Parliament (5 March 1903) and some time after secret negotiations for the purchase. The first Railway Executive Committee was constituted as follows. Sir Frank Ree (LNWR); (later Sir) Herbert Walker, London & South Western; Sir Guy Granet (Midland); Mr. F. Potter (Great Western); (later Sir) A. Kaye Butterworth (North Eastern); (later Sir) J A F Aspinall (Lancashire and Yorkshire); Sir Sam Fay (Great Central); Oliver R. H. Bury (Great Northern) and Donald A. Matheson (Caledonian). Matheson was in fact representing the North British Railway, the Highland Railway and the Great North of Scotland Railway, in addition to the Caledonian. The North British Railway remained as the ‘Railway Secretary Company’ for Scotland. Matheson had held the rank of Major in the Engineer and Railway Volunteer Staff Corps from July 1900. That the NBR acquiesced in this appointment was testament to improved relations between the NBR and ‘The Caley’ during the early years of the 20th century. The first Secretary of the Executive Committee, under Sir Frank Ree, was L W Horne, later Superintendent of the Line of the LNWR. Sir Frank Ree died suddenly on February 13th 1914 with Herbert Walker being appointed in his place. The Committee adopted for their headquarters the Westminster offices of the LNWR on Parliament Street, London and here, prior to July 1914, they had had six meetings.
The peace-time preparations of the Executive Committee further included the provision of means by which it could rely, in time of war, upon being in direct communications with the leading centres of railway communications. At first a system of wireless telegraphy was projected, but this was abandoned in favour of telephone installation. Under the direction of the Government, the Post Office authorities supplemented their ordinary London and trunk line services by providing a system of telephone wires between the Executive Committee’s offices in Westminster and all the railway termini in London, together with the general offices of the Midland Railway in Derby, the North Eastern at York and the Lancashire and Yorkshire at Manchester. Inasmuch as each centre of railway administration itself controlled a telephonic system . This comprehensive system of telephones, which was to play an extremely important part indeed in the working of the organisation was completed only the very week before the declaration of war
Sources
Official History of the Great War (Francis Edmonds) Vol. 1, 1923, originally published by Historical Section of the Committee of Imperial Defence. Reprinted by Naval and Military Press 2013. KPJ: note not Edmunds as per article: checked with York University OPAC
British Railways and the Great War (Pratt) Vol 1, originally published by Selwyn and Blount, Ltd. Available in the Classic Reprint Series of Forgotten Books. Also available to view online at https://archive.org/ details/cu31924092566128/page/ n6/mode/2up
An unappreciated field of endeavour: logistics and the British Expeditionary Force on the Western Front 1914-1918 (Clem Maginnis). Helion: 2018. Illustrations:

Donald A. Matheson, General Manager of the Caledonian Railway

45

Sir Frank Ree of the LNWR, as portrayed in Vanity Fair

46

Part 2 will be ‘War! State Control Applied – Mobilisation’ covering the full extent of their own lines, the facilities afforded to the Executive Committee for communications with every part of the country were exceptionally great.

Allan Rodgers. The Scotsman vans of the NBR – a follow-up. 49-52
Following publication of previous article on the Scotsman newspaper vans, which appeared in Journal 139, Study Group member Jim Hay got in touch. It turns out Jim has a 7mm model of one of these vans, in NBR livery, originally built many years ago by Sir Eric Hutchison, who, it will be recalled, wrote the article on the Scotsman vans which appeared in the November 1945 issue of Model Railway News. Much of the information I had included on the 1890s six wheeled vans was based on the information contained in Sir Eric’s article; so, the existence of a model built by him was clearly a source of new information about the livery of these vans as running in NBR days. So much so, that I decided a follow-up article was required to include amended illustrations of the 1890s vans showing the NBR livery I now believe they carried, based on scrutiny of Sir Eric’s model.
Illustrations (all except on from Model Railway News in colour).

Photograph of Sir Eric’s model – side and end view. Note vermillion ends and black solebar and ironwork: ends steps are also black, with upper surface in vermillion. 49
Photograph of Sir Eric’s model – part elevation. Note missing handle on the single door to the left of the ducket – clearly this is a dummy door. 50
Photograph of Sir Eric’s model – close-up of Scotsman scroll. 50
Elevation drawing of van number 257 as used in Model Railway News article of 1945. 50
Drawing: of Scotsman van number 251 in NBR livery. Allan Rodgers 50
Revised drawing of Scotsman van number 257 in NBR livery. Allan Rodgers 51
Drawing: of Scotsman van number 251 in LNER livery, as LNER No. 3251. Allan Rodgers 52

Stewart Noble Where was Helensburgh’s first railway station? 53-8
The short answer is that Helensburgh is where it was first located in 1858, that is adjacent to the  corner of Sinclair Street and East Princes Street. References to a station in George Street probably relate to a ticket platform where returning commuters left the train and joined their horse-draewn transport. A report by Major Marindin in 1892 is used to justify the permanence of the original terminus station, rather than assertions made by the author elsewhere. 

Helensburgh, showing station and surrounding area in early 1860s map extracted from Ordnance Survey 25-inch First Edition, Dumbartonshire Sheet XVII.5. Publication date: 1862. Survey date: 1860 54
As above but further east, Ordnance Survey 25-inch First Edition, Dumbartonshire Sheet XVII.6. 54
Original station, viewed from  corner of Sinclair Street and East Princes Street, with the municipal buildings in foreground. 55
Warehouse at 19 George Street 57

Operating the Caledonian Railway Andrew Boyd reviews a recent two-volume work by Jim Summers. 59
At first sight the review of a two-volume work about the Caledonian Railway must surely have no place in a Journal devoted to the NBR, so why has the Editor been prevailed upon to allow its inclusion? The answer is that they provide an excellent insight into how a major railway of that era functioned and was managed and operated. Every railway was different and each had its own management structures and methods of conducting business. However they all faced similar challenges and were all subject to the same statutory and regulatory requirements. They had to co-operate with other, often competing, companies. They exchanged traffic, complied with Railway Clearing House accounting and other obligations, managed joint lines and stations, exercised running powers over the lines of other companies and had to accommodate the running powers exercised by other companies. Students of the North British Railway will therefore find here much to interest and inform them, not least in comparing CR practice with that of other companies. Such comparisons were not always unfavourable to the NBR. Although written in an engaging style, the author has applied a professional perspective to his subject. He is well qualified to do so as in his working life he was a career railwayman who occupied senior positions within BR management. Apart from his current role as vice-chairman of the Caledonian Railway Association and as a railway modeller of some standing, he is also a member of our Group, well known to many of us. The contents encompass most aspects of managing and operating the railway. These range from the structure of top management and the duties of various grades of staff to the classification and loading of trains, operation of marshalling yards and hazards of shunting; and from the timetabling and operation of the Royal Train to the running of suburban and workman’s trains and the handling of mail, parcels, goods, mineral and livestock traffic. Amongst the aspects covered are what the author nicely describes as getting on with the neighbours; arguments with traders about demurrage charges; operation of slip carriages; timetable preparation; and the operation of a selection of principal stations including the joint station at Perth. He also covers the investigations made into the feasibility of electrifying the Glasgow Central Low Level lines in the early years of the last century and Donald Matheson’s examination of US practice in relation to coal wagons and shipment. The latter issue was one of considerable financial relevance to railways such as the CR (and NBR) that were major coal carriers.
As with his previously published book on signalling the present volumes are copiously illustrated.
The comparisons with other companies are particularly useful. In a chapter on organisation, the author explores the role of Superintendent and his duties. In 1916 the CR was, as the author puts it, ‘tinkering’ and beginning to use the term Superintendent of the Line, while around the same time the NBR was also considering its own organisation but more fundamentally. Two posts emerged on the NBR in 1917, namely an Operating Superintendent (Major Charles H Stemp), and a Commercial Superintendent, evidencing the NBR’s decision to formalise the split between commercial matters and the running of trains.
No doubt with the eye of a professional railwayman the author discusses the compilation and layout of the General and Sectional Appendix to the Working Timetables. The Appendix had become a crucial document for those operating the railway yet the CR’s final issue in 1915 was in his words ‘a weighty, inconvenient hotchpotch‘ whereas the NBR’s approach was more progressive, not allowing the imminent grouping deflect it from issuing in 1922 a ‘modern edition’ of its Appendix. He commends the innovations introduced by the NB and speculates what the CR might have adopted had it remained independent. As it was the CR contented itself with issuing in 1921 ‘an unenterprising traditional Supplement… to its hefty Appendix of 1915’.
In his chapter on Control the author notes that despite the challenges faced by the CR in operating its congested main line over Beattock and in handling the intense mineral traffic in Lanarkshire, it was the NBR which at Portobello in 1913 pioneered in Scotland the notion of a control office. Reference to these instances is not of course to suggest that the author fails to commend the CR when appropriate nor that he fails to compare the CR favourably with the practice of other companies when justified but these do illustrate the objective approach adopted by the author. Amongst the examples of the working relationships which necessarily arose between the CR and the NBR two instances may be mentioned. An edited version of an article by David Stirling first published in the SRPS magazine ‘Blastpipe’ describes the conflict between the two companies over the rebuilding by the NBR of the swing bridge carrying its Stirlingshire Midland Junction railway over the Forth & Clyde Canal. The CR had to resort to litigation to force the NBR’s hand over this issue but eventually an agreement was reached to enable work to proceed. While the line was closed for work to be done, extensive diversion of trains operated by both companies had to take place, often entailing reversals at Greenhill, Polmont and Grangemouth Junction.
A few years later an unplanned line closure resulted from a vessel colliding with the CR’s Forth Bridge at Alloa in October 1904. The line, used by the trains of both companies (as the NBR exercised running powers over it) was closed until June 1905. Co-operation resulted in the institution of a service of three CR passenger trains each way over the NBR route between Alloa and Stirling and the conveyance of CR passengers and traffic on two evening NBR trains each way on the same route. These two volumes are thoroughly recommended to all those who wish to learn more about how railways were managed and operated in the days when companies like the NBR and the CR were in business.

Macmerry Station. 60 (rear  cover)
Macmerry had quite a simple layout, but did include a run-round loop and two sidings (one with a crane), together with a passenger platform long enough for an engine and a few coaches. The branch on the eastern side of the station, diverging to the south of the map extract and heading roughly north-east on the lower right hand part of the extract, was a mineral railway serving Merryfield, Bald, Dander and Engine Pits to the north of what is now the A1 road. Extracted from Ordnance Survey 25 inch Map of Haddingtonshire IX.11. Publication date 1894, revised 1892.