9. The road to victory: transportation in the British Expeditionary Force, 1917–18

As DGT on the western front, Sir Eric Geddes was responsible for ensuring that the necessary transport arrangements were in place to support the BEF’s future military operations. In 1917 Haig’s force undertook four offensives – at Arras, Messines, Ypres and Cambrai – each of which dwarfed the battle of the Somme in 1916 in terms of the volume of ammunition fired in support of the infantry. As Ian M. Brown observed, in the second half of the conflict BEF ammunition expenditure was constrained by the working lives of the guns rather than the absence of sufficient reserves of shells.¹ In 1918 the dislocation caused to the BEF’s transport infrastructure by the German spring offensives was substantial, but insufficient to critically impair the fighting abilities of the British troops. Over the summer months a vast allied transportation effort provided the conduit through which the counter stroke was delivered and sustained through much – but not all – of the war’s final campaign.

The results of Geddes’s reorganization of transport, maintained following his departure for the Admiralty in May 1917, were played out during the so-called hundred days offensive. In the eight-day bombardment prior to the Somme, the BEF had fired 1,732,873 rounds. Eight weeks later the transport network behind the front line came perilously close to collapse under the pressure of sustaining an ammunition expenditure of 28,000 tons per week. In contrast, when British troops assaulted the Hindenburg Line eight weeks after the battle of Amiens they were supported by an artillery that was able to fire a staggering 943,847 rounds in just twenty-four hours. In the week ending 29 September 1918 the BEF expended 3,383,700 rounds, a total of 83,170 tons. In the decisive period between 8 August and 11 November 1918 the BEF poured a colossal 621,289 tons of artillery ammunition into the retreating Germans’ defences.² The war was won on the western front, and the BEF played a key role in its delivery.

The importance of those munitions being in a position from which they could be fired into the German lines cannot be overstated. Yet the role of the transportation units responsible for the movement of voluminous quantities of material during the summer and autumn of 1918 has been largely overlooked in histories of the First World War. Numerous authors have charted the technological and tactical improvements made by the BEF between the nadir of 1 July 1916 and the end of the war, but few have documented the modernizing or learning process within the context of the allies’ ability to apply those modifications effectively. The allies possessed superior access to both human and material resources than the central powers, especially following the United States’ entry into the war in April 1917. However, ‘many of those resources were in the wrong place: far away in overseas empires or the US’.³ Colossal quantities of shells were of no use to artillery commanders at the front if the Channel ports could not process them from the ships, nor the trains, lorries, wagons or barges move them inland to the guns from which they were despatched on their final journey across no man’s land. The successes of the BEF’s supply services, in conjunction with myriad long-and short-term issues that combined to reduce the German army’s effectiveness, greatly contributed to General Ludendorff’s decision to seek an armistice in 1918.⁴

This chapter examines the BEF’s operations in the second half of the war, and tracks the evolution of the DGT’s role in France and Flanders. It underscores how important the transport factor was to the operational tempo of an army engaged in material-intensive warfare, and emphasizes the complexities experienced by those – soldier and civilian – tasked with the maintenance of an industrial army in both static and mobile environments. In 1917 the organizational structure conceived and implemented by Geddes delivered unprecedented volumes of firepower to the front, but proved incapable of responding swiftly to the consequences of its success. The devastation created in large part by the artillery supplied by Geddes’s light railways proved an insurmountable obstacle in the mud around Passchendaele. When mobility did return to the battlefield the following year, the prevailing conditions engendered passionate discussion over the subordination of the civilian-created directorate to the QMG’s department within the military hierarchy. Yet as the need for civilian input of a managerial, directorial nature declined in the latter stages of the war, the divide between recognizably civilian and military tasks became indistinguishable. The transport factor – exemplified in the universal tasks of road building, railway construction and the delivery of goods from distribution centres to consumers – exerted a powerful influence over the BEF’s war-making capacity until the very end of the war.

Making a new world: transporting the war of material in 1917

Negotiations over the BEF’s fair share of the allies’ finite transport resources, discussed in the previous chapter, were not the only claim on Geddes’s attentions during his tenure as DGT. He was also responsible for making the necessary transport arrangements to ensure that the BEF was adequately supported in its military operations. However, Geddes had no influence over the location of the BEF’s 1917 offensives or upon the nature of Britain’s contribution to those campaigns. On 15 November 1916 the allies’ military leaders met at Chantilly and agreed to continue the strategy of simultaneous offensives against the central powers on all fronts in 1917. Haig and Joffre confirmed the location for operations on the western front two weeks later. Haig, although desiring an offensive in Flanders to clear the Belgian coast, accepted Joffre’s plan for a British attack – to take place between Vimy and Bapaume – concurrent with a French assault launched between the Somme and Oise rivers.⁵ Therefore, Geddes’s concerns over the directorate-general of transportation’s first winter in existence were dominated by the need to provide a transport infrastructure capable of safeguarding Haig’s troops against the problems experienced during the Somme.

The effects of Geddes’s reorganization of British transportation on the western front became apparent almost immediately. The tonnage carried by the hitherto underutilized IWT fleet increased by a third in the month after Geddes’s appointment, as 94,073 tons were carried by canal and rivercraft in November 1916. The following month, as ‘every effort continued to be made to relieve the railways and ports by utilising as fully as possible the carrying powers’ developed by Gerald Holland over the previous years, IWT provided carriage for 108,000 tons of material that otherwise would have taken up space on the overburdened railway network.⁶ Barges were increasingly loaded direct from ships in port, a practice that allowed the BEF to avoid crippling shortages when the port of Boulogne was closed for a month following the accidental sinking of the SS Araby in late December.⁷ However, as on the railways, the winter weather of 1916–17 caused significant reductions in the circulation of waterborne traffic. ‘Towards the end of January’, Holland noted in his final memorandum as director of IWT, ‘the frost was so intense that vigorous measures had to be adopted to keep open the navigation sections of the Northern Waterways system’. On 27 January, ‘in spite of all efforts to keep the ice broken up, the canals became frozen to such an extent as to bring transport operations in many places to a standstill’. Activity continued near the coast at Calais and on the River Somme – where the strong current prevented the water from freezing entirely – but IWT was unable to continue its expansion until the thaw commenced on 15 February. By March 1917 the carriage figures achieved by IWT had not only bounced back to the levels recorded in December 1916 but had surpassed them. 150,000 tons were carried by IWT in the month before the battle of Arras, and at both Calais and Dunkirk between 37 and 39 per cent of imports were loaded direct from ship to barge – reducing the strain on the limited railway facilities at both ports.⁸

The cross-Channel barge service also contributed to the alleviation of rail and port congestion. In January 1917 cross-Channel barges carried 1,635 tons per week to France. By September that figure had risen to 14,460 tons per week, and from May 1917 the service was entrusted with ammunition deliveries. The Royal Navy’s alleged resistance to, and eventual adoption of, the convoy system has dominated historical analyses of Britain’s response to the German policy of unrestricted submarine warfare in 1917.⁹ Yet the humble barge also made a telling contribution to the maintenance of the allies’ cross-Channel communications, and helped facilitate materially intensive forms of warfare. Between January 1917 and the armistice 574,127 tons of ammunition passed between the IWT depots at Richborough and Zeneghem; not a single barge was lost to enemy action in the English Channel during the war. Alongside ammunition, the service carried the IWT directorates’ stores and supplies, railway and engineering materials and even aircraft to the western front.¹⁰

The increased deployment of IWT exemplified the advantages that accrued from the centralization of transport. Geddes’s authority to choose the most suitable mode of transport for the various classes of supply demanded at the front replaced the self-interested actions of 1916 – when individual departments and services had competed with each other for the limited rail capacity available – with an organization that allocated transport according to a broader consideration of the BEF’s overall priorities. Goods for which there was a steady, largely predictable demand from the front could be sent forward by the comparatively slow barges, which freed up significant numbers of railway wagons for the carriage of more urgently required stores.¹¹ The exploitation of canal transport permitted both a larger and more responsive war effort from 1917 onwards, with clear implications for the intensity of the operations the BEF was able to conduct.

However, the BEF’s offensive preparations in early 1917 demonstrated both the key role to be played by transportation in the conduct of modern warfare and the subordinate position still occupied by those responsible for the transport infrastructure’s maintenance within the BEF’s hierarchy. As on the Somme the previous year, the transport facilities in the area designated for what became known as the battle of Arras required significant improvement ahead of zero hour. The town of Arras was served by just two standard-gauge lines, both of which had to be doubled and provided with new railheads to satisfy the BEF’s needs. An additional sixty-five miles of track and sidings were laid in preparation for the offensive using rails lifted directly from the LNWR’s lines in England, and by mid February Haig appeared satisfied with Geddes’s progress. The former recorded in his diary that the ‘energy and knowledge’ demonstrated by ‘Geddes and his men’ deserved ‘very great credit’. The British railway outlook ahead of the 1917 campaigning season, he stated, was ‘promising!’¹²

The winter frost, the German retirement to the Hindenburg Line, the sluggish arrival of the materials ordered by Geddes and Haig’s initial decision to concentrate British construction work around the Somme in anticipation of offensive operations in that area all retarded the development of the transport system around Arras.¹³ Yet for Cyril Falls, writing in the official history of the battle, the most significant challenge for the units charged with improving the transport infrastructure behind the front lay in the BEF’s continued prioritization of the supply of ammunition over engineering materials ahead of Arras. Artillery was, as Keith Jeffery acknowledged, ‘central to victory’ in the First World War and the cause of around two-thirds of the conflict’s casualties.¹⁴ With this in mind, Ian M. Brown’s pioneering work on British logistics considered available ammunition stocks to be a ‘useful gauge of the BEF’s supply state’ during the war and a valuable tool for understanding the force’s logistical operations.¹⁵ Between 17 March and the opening of the battle on 9 April, all but one of the ammunition trains requested by BEF commanders were received at railheads. ‘When going round the batteries in action’, Geddes told Falls after the war, ‘artillery officers begged him to stop bringing up ammunition because they had all they wanted and their men were too tired to offload more’. ‘This was indeed a happy state of affairs’, Falls wrote, ‘but it is certain that no engineer officer said as much respecting engineer stores’.¹⁶ During the same period only 125 out of 206 stone trains and thirty-five of the fifty-six requested trains of engineering stores ran.¹⁷

The corollary of the BEF’s focus upon the maintenance of ammunition deliveries over other supply traffic was that Geddes’s light railway policy – which had originally been conceived to relieve road transport, assist the advance across shell-torn ground and to forward stone and equipment for road repairs – had yet to bear fruit before the battle of Arras began.¹⁸ The Third Army had requested fifty miles of light railway for new construction, and a further fifty miles for repairs and extensions to connect the British system to captured German lines in November 1916. In addition, a light railway system inherited from the French by XVII Corps, in an area north of Arras ‘where roads to the front line were almost non-existent, had been developed into a ‘complete supply system’ comprising around thirty-two miles by April 1917.¹⁹ However, the slow arrival of rails and engines rendered the light railway operating companies unable to provide significant respite to the overburdened road network during the battle. Across the Third Army’s zone of operations, teams of mules and drivers from the Royal Field Artillery provided the motive power required to haul trains of six wagons over lines caked with mud, which were highly susceptible to derailments and consequent delays. Such a poorly equipped light railway system could only contribute a very small amount to the Third Army’s supply needs at Arras, with the inevitable consequence that the inadequately supported road network suffered greatly in the inclement weather that accompanied the offensive.²⁰ Falls was left in no doubt as to the consequences of the BEF’s prioritization of firepower over transportation before Arras:

Sufficient stocks of artillery were undoubtedly a key component of the war-winning, combined-arms strategy deployed by the BEF in the latter stages of the First World War. However, the subordination of supplies for the upkeep and extension of the BEF’s transport infrastructure drastically impaired the force’s ability to maintain its operational tempo and advance in strength before the Germans could regroup at Arras.

The light railway network’s rapid expansion over the spring and summer of 1917 permitted the British guns’ unprecedented consumption of ammunition, which contributed greatly to the BEF’s initial successes in Flanders. The transport infrastructure in the area tasked with supplying the operations that became known as Messines and Third Ypres contrasted favourably with the road and rail facilities available around the Somme the previous year. A double line from Hazebrouck to Steenwerck had been supplemented by the doubling of the line between Hazebrouck and Ypres in 1915, and the ROD had been responsible for traffic on the latter from November of that year. A further line between Bergues and Proven was laid in 1916 and doubled in 1917, which gave the BEF a total capacity of 180 trains per day in the battle area. A highly developed forward delivery system, which utilized both light railway and road transport in roughly equal volumes, provided the Second Army with facilities capable of removing around 300,000 tons of material from the railheads in the month preceding Messines.²² For the first time in the war light railways were deployed to carry ammunition right up to the heavy batteries before the assault on Messines, which allowed General Herbert Plumer’s force to accumulate a stockpile of 144,000 tons of shells prior to the battle.²³ Geddes may have departed for the Admiralty by the time the attack took place, but he bequeathed to his successor as DGT, Sir Philip Nash, an organization able to supply the Second Army’s artillery with each of the 3,561,530 rounds it fired into the German defences between 26 May and 6 June 1917.²⁴ Designed from the outset as a limited operation to clear the ridge, and described by Nick Lloyd as ‘perhaps the finest example of a “bite-and-hold” operation ever conducted’,²⁵ all of Haig’s objectives for Messines were in British hands within a week and the battle was closed down as attentions shifted north towards Ypres.

As the third battle of Ypres unfolded over the summer and into autumn, the limited applicability of Geddes’s industrial supply system to the conditions of the industrial battlefield – and its considerable effects upon the landscape of the western front – were gradually and graphically exposed. Events at Messines had demonstrated what was possible with a ‘massive accumulation of guns and shells, the unrelenting preliminary bombardment of trenches and strongpoints’, and the employment of new techniques designed to protect the advancing infantry and suppress the German artillery.²⁶ The brief duration and comparatively restrained aims of the offensive concealed the outcome of such colossal expenditures of firepower. At Third Ypres the BEF assembled an even larger collection of artillery pieces to support an offensive with far more ambitious objectives, both in strategic and geographical terms. Haig had extended the distance of Plumer’s prospective advance at Messines to between three and four thousand yards, but he demanded far more from the troops employed at Ypres.²⁷ The commander-in-chief’s sights were set on the capture of Roulers, twelve miles beyond the front line on 31 July, and he emphasized on an early plan for the operation that ‘there must be no halt on reaching Roulers’.²⁸ To prepare the way for the prospective advance, the BEF’s artillery fired 4,283,550 rounds during the preliminary bombardment – an increase of 147 per cent over the number that had been hurled towards enemy lines ahead of the Somme.²⁹

However, the early gains at Third Ypres – eighteen square miles of enemy territory were captured on the first day, against 3.5 square kilometres on 1 July 1916 – were superficial. As had occurred in Picardy, the exertions in Flanders failed to definitively break the German lines and the battle descended into a series of piecemeal assaults undertaken in appalling weather.³⁰ On the lines of communications towards the rear the rain had little influence. On 7 August, Nash reported to Haig that ‘everything has gone extremely well. The number of trains to rail-head has increased to an excessive number: 30 more per diem than we had calculated for, and told the French we would want!’³¹ However, beyond the railheads the weather conditions proved a considerable handicap to the BEF’s advance. Haig complained as early as 1 August about ‘a terrible day of rain’ that had left the ground ‘like a bog in this low lying [sic] country!’ The light railways and roads necessary for the sustenance of the troops were ‘steadily … pushed forward’ despite the ‘terrible wet’,³² but within a week troops could only reach the front line ‘via a thin network of duckboard tracks that had been laid across the sodden landscape’.³³ Conditions around the 18th Division’s artillery made the movement of guns even more arduous. The ooze and slime were so thick that it took the men of one battery over six hours to move a single 18-pounder gun just 250 yards.³⁴ The German gunners, whose arsenal had been increased from 389 to 1,162 pieces before the battle, targeted the main roads and added to the devastation.³⁵ On 1 September X Corps’s area commandant complained that ‘there [was] considerable difficulty in getting the traffic through owing to the constant shelling and bombing and the Labour Corps are always repairing shell holes in the roads which block the traffic’.³⁶

With the road network increasingly congested and impassable, the time had come for light railways to fulfil the role Geddes had conceived for them during his transportation mission the previous summer. As the battle continued, an increasing quantity of the goods unloaded from the 220 trains that arrived at railheads each day were sent forward on the ever-expanding light railway network. By September 1917 the BEF’s light railway system comprised 623 miles of track, almost double the mileage available during the operations at Messines (see Table 9.1). The tonnage conveyed by light railways increased by an even larger proportion over the same period, rising from an average of 95,180 tons per week in June to 210,808 tons per week in September. Light railways carried over 227,000 tons in the final week of September, which included 47,724 tons of ammunition both in support of the successful attack at Polygon Wood and in anticipation of the assault on Broodseinde.³⁷

Table 9.1. Selected weekly averages on the light railway network for typical months, 1917.

The operations at Polygon Wood and Broodseinde represented the second and third steps in a sequence of assaults designed by Plumer to replace the more ambitious and costly approach that had been followed by Gough’s Fifth Army in August and September.³⁸ Plumer’s designs drew upon his experiences at Messines in that they called for a succession of assaults, each of roughly 1,500 yards, which aimed for strictly limited objectives until the high ground of the Gheluvelt Plateau lay in British hands. Each offensive in the sequence was accompanied by concentrations of artillery fire that far exceeded even that which had accompanied Gough’s initial attack on 31 July.³⁹ The principal result of the focused, intense bombardments that took place in late September and early October 1917 was the annihilation of the Flanders landscape. The shallow nature of Plumer’s attacks meant that British troops always advanced over shattered ground, the undulating nature of which made light railway construction both increasingly difficult and material intensive. More than one-fifth of all traffic on the light railway network comprised materials for the repair and maintenance of the railways, and between fifteen and twenty men per mile were required to ensure that the system remained operable.⁴⁰

The light railway network developed and operated by Britain’s transport experts was, through its facilitation of the limited and sequential advances of September and October 1917, the architect of its own breakdown. The shells delivered to the guns by light railway created ground that was wholly unsuitable for the forward projection of railway lines as the Ypres offensive crept forward. In his memoirs Gough recorded a vivid account of the prospect that faced the British troops in Flanders in late 1917:

At Third Ypres the transport system behind the front line collapsed. Nash reported to Haig on 13 October that ‘he [had] light engines on the 60-centimetre railways sunk halfway up the boilers in the mud! Track has disappeared!’⁴² The conservative nature of Plumer’s objectives meant that the German guns lay beyond the reach of the attacking infantry and remained free to direct their fire upon the BEF’s forward communications, which exacerbated the difficulties faced by those tasked with getting urgently needed resources to the fighting troops. Cuts to the line became increasingly frequent, and ‘in certain sections it appeared to be the aim of the German command to paralyse our advance by denying the supplies requisite to enable it to be continued rather than by overwhelming the troops themselves’.⁴³ The severance of British rail lines rendered any trains on them at the time sitting ducks – unable to move forward to deliver their goods nor backwards to safety. ‘Many a train load of ammunition or other vital supply’, recalled one commentator, ‘was … shot to pieces before the lines could be repaired’.⁴⁴ The BEF’s extensive light railway network soon became ‘utterly unable to cope with the immense volume of ammunition, engineering stores, and supplies required by the troops’ at the front.⁴⁵

Light railways were incompatible with the mud and German artillery tactics, while the lack of construction materials sent forward during the battle left the BEF reliant on a road network that was wholly inadequate for the task. At Poelcappelle on 9 October the advancing British troops were supported not by the 1,295 guns that had been assembled ahead of the Menin Road assault on 20 September but by a paltry twenty-five guns. Most of the artillery was ‘stuck uselessly on the blocked single-track roads further to the rear’.⁴⁶ Brigadier-General Ludlow of X Corps rode through Ypres the following day, where he observed ‘an enormous amount of traffic’ along the Menin Road, to Hell Fire Corner – a ‘scene of utter devastation and appalling mud’.⁴⁷ Under such conditions the achievement of an operational tempo of suitably rapidity to permanently destabilize the German army was impossible. In the words of the Royal Engineers’ history of the war:

The industrial supply system devised, installed and operated by Britain’s transport experts played a key role in creating the horrific battlefield conditions experienced by those who participated in the third battle of Ypres. The transport network’s increasing efficiency may have permitted the BEF’s senior commanders to prosecute a ‘rich man’s war’ from 1917 onwards.⁴⁹ However, both Gough and Plumer chose to repeat the mistakes made the previous year: unprecedented volumes of firepower took priority over the provision of adequate supplies of human and material resources for the maintenance and development of the forward communications required to sustain the advance.⁵⁰

The BEF’s obsession with firepower created a logistical imbalance on the western front that generated particularly appalling connotations in the last month of the battle. The haunting images of men, machines and animals submerged in the mud and slime around Passchendaele provide a far more poignant demonstration of the BEF’s failure to adequately comprehend the implications of artillery-intensive warfare than the eighteen-mile-long line of trains outside Amiens the previous summer. Yet both were symbolic of the emergence of a challenge hitherto underappreciated by those charged with directing the British war effort. The improved quantity, if not quality, of the Ministry of Munitions’ output in 1916 had exposed the inadequacy of the BEF’s transport infrastructure and engendered Geddes’s reorganizations. The quicker throughput of goods at the ports, the increased number of trains arriving at railheads and the greatly expanded capacity of the transport network between the railheads and the front had each aided the creation of a logistics system that was able to deliver unprecedented volumes of material to the terminal point of the DGT’s responsibilities. When the location of the front line shifted – an inevitability if the allies were to forcibly expel the Germans from occupied territory – the implications of that reorganization were revealed.

Britain’s transport experts were not responsible for the provision of forward communications on the battlefield. Their role was confined to the supply of materials demanded by the army to the required place at the desired time. Ahead of the so-called DGT line it fell to the chief engineers of corps and divisions to carry out the construction and maintenance work necessary to move men, guns and supplies into position quickly enough to resume the offensive before the enemy recovered. Significantly, none of the senior engineers who participated in the battle of Third Ypres identified mud as the principal factor that influenced the outcome of the campaign. Instead, all were critical of the logistic, engineering, manpower and administrative components of the BEF’s approach to warfare in late 1917. The most lucid observations came from the chief engineer of the Canadian Corps, Brigadier-General William Bethune Lindsay. The Canadians had taken over from II Anzac Corps on 17 October and Lindsay had immediately abandoned construction on light railways in order to focus on road and tramway building projects. The corps demanded that both the time and manpower to ensure sufficient communications were in place before it attacked, that adequate pauses were made between assaults to allow the road network to be advanced and repaired, and that the objectives selected for successive operations took account of the deteriorating conditions.⁵¹ Between 26 October and 10 November the Canadians struggled through the morass and took the Passchendaele ridge at a cost of almost 16,000 casualties.⁵² Their material requirements – in terms of ammunition, engineering stores, road stone, planks and pit props – employed 960 lorries and countless thousands of wagons and pack animals per day,⁵³ all of which had to travel forward over a landscape obliterated by the munitions furnished by Britain’s transport experts.

The BEF’s experience at Ypres in 1917 amplifies recent calls from historians of the First World War to ‘look beyond the British army’s struggles to master the technical battlefield’, and to acknowledge the multiple practices of adaptation that occurred on both sides of no man’s land during the conflict.⁵⁴ On the western front alone the belligerent armies applied new methods and technologies proactively, and were forced to respond to the innovations of their opponents. The BEF’s material-intensive form of warfare depended heavily upon the maintenance of a dependable supply link to the front line. Light railways, a response to the problems of forward communications encountered during the Somme in 1916, represented a civilian-led attempt to improve both the capacity and flexibility of the connection between the front line and the BEF’s ever-expanding supply of munitions. However, the Germans’ redirection of artillery fire towards the disruption of British supply operations, the British army’s ongoing sacrifice of engineering and transport materials on the altar of firepower, and the horrendous weather conditions severely eroded the BEF’s ability to advance in the autumn of 1917. As Robin Prior and Trevor Wilson noted, Third Ypres ‘did not cease because it had reached some meaningful culmination. It simply came to a halt’.⁵⁵ When British troops began to move again the following year, they did so in the opposite direction.

A diminished role? Britain’s transport experts and the return to mobility

The organization responsible for the supply of those troops underwent a number of personnel changes between the suspension of Third Ypres and the start of the German spring offensives. In London, Sir Sam Fay ‘reluctantly’ accepted the post of DGMR following Sir Guy Granet’s decision to leave for America and take up the post of food controller. Fay was replaced as DOM by the soldier who had previously acted as his deputy, Colonel Herbert Delano Osborne.⁵⁶ In France, Sir Philip Nash was also replaced by a soldier rather than a civilian when he left the post of DGT to undertake his examination of allied transport resources on the western front. Major-General Sydney D’Aguilar Crookshank, a Royal Engineers officer whose pre-war experience had predominantly been acquired in India, became the BEF’s first non-civilian DGT.

However, the promotion of these professional soldiers to roles previously held by civilians was not the result of a resurgent military voice within the British war effort. Instead, the turnover of personnel within the upper echelons of the BEF’s transport services between the campaigning seasons of 1917 and 1918 was instigated by the prime minister and Sir Eric Geddes. Crookshank was first recommended as a suitable DGT by Geddes when the latter attempted to acquire Nash’s services for the Admiralty in October 1917.⁵⁷ Haig was content to see Nash leave at the time on the proviso that the change was not made until after the planned operations at Cambrai had been concluded.⁵⁸ The German-Austrian breakthrough at Caporetto scuppered Haig and Geddes’s arrangement. Nash took on the duty of coordinating the emergency movement of troops from the western front to Italy and then, in January 1918, embarked upon his examination of communications between the English Channel and the Adriatic that led to the creation of the IATC.⁵⁹

A further casualty of the ‘increasingly irate’ Lloyd George’s rejuvenation of GHQ in the winter of 1917–18 had profound implications for the directorate-general of transportation.⁶⁰ The QMG, Sir Ronald Maxwell, left France at the end of the year, his poor health offered as an excuse to remove him from active duties. Following what Haig dubbed ‘the decision of the Army Council to replace him with a younger man’,⁶¹ the forty-six-year-old Major-General Sir Travers Clarke took up Maxwell’s post and acted as QMG throughout the final year of the war. His appointment was described later by Sir Frank Fox as ‘a daring experiment on Lord Haig’s part; for he was a comparative youngster to be put into a post which was then the most anxious and onerous in the Army’.⁶² By the summer, Clarke’s onerous responsibilities had expanded to include authority over the directorate-general of transportation.

Viewed through the surviving observations of those indirectly involved, the machinations that brought the directorate-general of transportation under Clarke’s control give the appearance of a naked power grab by the new QMG. As early as January 1918 Brigadier-General John Charteris noted the existence of a ‘permanent feud’ between Crookshank and Clarke, while in April Fay recorded the appearance of ‘a set against Crookshank’ who wanted ‘to get the transportation business under the Q.M.G. again, same as it was before Geddes’s appointment’.⁶³ Haig became aware of the ill feeling between the two men during the same week, when a delegation of senior officers – with whom Clarke was in accord – openly questioned Crookshank’s abilities and recommended his replacement. Haig thought it a ‘serious matter to change a highly placed Administrative Officer’ at such a critical period of the war, and attempted to diffuse the situation by reducing the direct personal contact between the DGT and QMG to weekly rather than daily conferences.⁶⁴

Haig’s careful man-management temporarily reduced tensions in France, but the question of Crookshank’s position remained open in London. On his first day as secretary of state for war, Lord Milner told Fay that ‘he had had conversations with General Staff G.H.Q. on [the] appointment of Crookshank, [and] thought he was not good enough as Director-General of Transportation’. Consequently, Fay was ordered not to confirm Crookshank’s permanent appointment as DGT despite the latter having performed the role ever since Nash had departed for the IATC.⁶⁵ At a tense meeting in Milner’s office on 25 May, attended by most of the Army Council, Clarke attempted to circumvent Haig’s support for Crookshank and advocated the elimination of the post of DGT altogether. He stated that he ‘wanted to split everything up into separate directorates’ and return to the extant organization of 1914–16.⁶⁶

There were sound operational reasons underpinning Clarke’s suggestion, even though it appeared to augur a return to the watertight organizational structure that had retarded the development of a coherent transport policy before Geddes’s arrival in France. Clarke argued that the return of a more mobile form of warfare ‘made it clear … that the separation of the Transportation services from the Q.M.G. was a serious defect of organization and a possible source of danger. The Q.M.G. was responsible for supplies but he was not in a position to co-ordinate and control all the means of supply’.⁶⁷ As the BEF retreated into a more circumscribed area under the pressure of successive German attacks, the uninterrupted daily provision to the fighting troops of 1,934 tons of supplies per mile of front became ‘the whole question’ of the war.⁶⁸ However, with Crookshank’s status officially on a par with Clarke’s, any divergence in policy between the two men required referral to the commander-in-chief for a decision. Any delays in the decision-making process courted disaster in the fluid conditions not experienced on the western front since 1914, and were made more likely by the changed character of battlefield supply. Based at GHQ, Clarke’s direct access to the latest intelligence and the immediacy with which he received Haig’s operational priorities contrasted favourably with the location of Crookshank’s offices at the so-called ‘Geddesburg’ a few miles away – although the distance between GHQ at Montreuil-sur-Mer and the DGT’s headquarters at Monthuis was far less pronounced than that between GHQ and the IGC’s offices had been earlier in the war.

GHQ’s comparative responsiveness had profound implications for the BEF’s ability to meet the challenging mobile conditions of 1918. For Clarke the major lesson of Third Ypres had been the light railways’ inadequacy to cope with heavy and accurate German artillery fire. Consequently, he observed, ‘upon the roads and mechanical transport fell the bulk of the maintenance work of the armies’. Between the campaigning seasons of 1917 and 1918 Clarke implemented a complete reorganization of the BEF’s mechanical transport, under the principle that lorries were for common use rather than to be dedicated to specialist units. As he explained in his post-war report:

Clarke’s new mechanical transport policy came into force on 13 March.⁷⁰ The German army’s final offensive push began just over a week later.

The progress of the German operations over the following weeks and months appeared to confirm Clarke’s beliefs both in the comparative advantage of road transport over light railways and of the desirability of centralized control over transportation from GHQ rather than the separation of duties between the QMG and a DGT. By 21 March – thanks to what David Stevenson has described as a ‘masterpiece of staff work’ – the Germans outnumbered their opponents at the junction of the British Third and Fifth armies by ratios of 2.6:1 in men and 2.5:1 in guns, and had accumulated 1,079 aircraft in the sector against the allies’ 579.⁷¹ The assembled German forces commenced a preliminary bombardment of unrivalled ferocity before dawn, which focused principally upon paralysing the BEF’s rearward communications and silencing the British artillery.⁷² Both road and rail networks suffered badly at the hands of intense German fire. ‘Practically every road’ was swept with 5.9-inch shrapnel and gas, while the area surrounding the railhead at Roisel, east of Péronne, received a combination of gas, high explosive and shrapnel shells at a rate of seven or eight rounds per minute. In the rear of the Fifth Army the bombardment destroyed the track, cut telephone lines and rendered the forward section of the light railway system unworkable.⁷³ By nightfall, the troops of the Third and Fifth armies had been driven back distances of up to eight miles.

The German advance continued at pace in the days that followed, and rapidly disorganized the network upon which Geddes’s resupply system depended.⁷⁴ As light railway lines became untenable, British attentions swiftly turned to the evacuation of rolling stock or otherwise denial of its use to the enemy. Most of the equipment in the northern part of the battle zone was evacuated through Fosseux; over 300 locomotives and tractors were disabled by the removal of essential parts, and nearly 2,000 wagons were burnt by the BEF in the days after 21 March.⁷⁵ By the end of the following month the route mileage operated by light railways had been reduced from 920 to just under 360 miles, while the demands of the fighting had led to a rapid decrease in the personnel available for the network’s operation and maintenance. From a peak figure of 262,000 tons per week just before the German spring offensives were launched, the tonnage carried by the BEF’s light railways fell by over 50 per cent in the three months that followed.⁷⁶

Further back, the standard-gauge railways also presented severe complications for the BEF’s logisticians during the spring. By 29 March the Germans held Bapaume, Albert and Péronne, and within a week they were able to shell the railways around Amiens from positions on the outskirts of Villers-Bretonneux. Almost all of the Third and Fifth Army’s railheads of a fortnight before the battle had been captured, along with many miles of valuable track.⁷⁷ As the German advance continued into April the surrender of important main lines, or their proximity to the enemy, made it increasingly difficult for Clarke and Crookshank to maintain fluidity throughout the rail network (see Figure 9.1). The loss of engine depots, the effects of German pressure on the crucial Amiens–Hazebrouck forward lateral line and the increased demands for railway traffic as troops from the Portuguese, French and American armies were rushed into the fray all increased the traffic on the remaining unmolested railways behind the British forces. During April alone the BEF ran as many ammunition trains as had been run in the entirety of the Somme campaign of July–November 1916, while the loss to the allies of the Amiens–Arras line reduced the available return routes for empty rolling stock and increased congestion across the network.⁷⁸ ‘Good circulation’, Clarke explained, ‘is the essence of economical railway working; and a block at any point has an affect similar to that of an aneurism on a human artery’.⁷⁹ By May even the coastal lateral route had become seriously clogged with human and material traffic, and the continued forward progress of the German troops forced Clarke to utilize railheads far back from the fighting to sustain the BEF. However, the British forces continued to receive food, supplies and even a regular mail service despite the disruption.⁸⁰

A coastal barge service between the northern waterways and the Seine was rapidly implemented as a temporary improvisation to aid the overburdened railways in the rear, but the road network provided the crucial, widely accessible connection between the railheads and troops engaged in fierce combat during the spring. Consequently, mechanical transport – the deployment of which lay outside the DGT’s purview – became a progressively more valuable component of the BEF’s umbilical cord as the German offensives unfolded. As Clarke emphasized in his review of May 1918, effective ‘man power depend[ed], in the final result, on road power’.⁸¹ The QMG-controlled reserve mechanical transport companies were allotted to formations according to the demands of the situation, ‘kept untasked until the last possible moment’, and withdrawn into reserve – or recalled for service elsewhere – as soon as their specific task had been completed.⁸² The latter was far more common in 1918. In some cases, Clarke noted in his review of the QMG department’s work in April 1918, ‘M[echanical] T[ransport] drivers were on duty almost continuously for five days at a stretch’ during the month. ‘Under conditions of stress and danger’, he continued, ‘they had kept their vehicles on the road and assisted greatly to relieve the situation’.⁸³ ‘Not only did the new organization prove its value’ during the spring offensives, wrote Major Wilfrid Lindsell in a post-war article for the RUSI Journal, ‘but unmistakable proof was also afforded of the necessity for centralised control of all transportation services under the Quartermaster-General’.⁸⁴

Figure 9.1. The British Expeditionary Force’s rail network, April 1918.

Source: D. T. Zabecki, The German 1918 Offensives: a Case Study in the Operational Level of War (Abingdon, 2006), p. 87. Map drawn by Cath D’Alton.

The provision of separate but interlinked administrations for the operation and maintenance of an interconnected transport network had proven incompatible with the requirements of modern warfare in 1916. The directorate-general of transportation provided the BEF with an organization better equipped to service the material requirements of an industrial army, but the experiences of 1917 and early 1918 had confirmed that they were predominantly suited to the development and sustenance of a stationary army. The German spring offensives created conditions not experienced in France since the autumn of 1914, when the then QMG, Sir William Robertson, had accepted overall responsibility for the coordination of traffic on the western front. By mid April 1918 Sir John Cowans, the QMG of the forces, was explicit in his belief that the 1914 decision to subordinate the IGC had to be replicated. He wrote to Clarke that ‘the introduction of the D.G.T. was all very fine when we were in a stationary period, but now either he or his representative ought to be in your office to take your orders’.⁸⁵ Fay, in the Army Council meeting held on 25 May, managed to convince his colleagues to retain the directorate-general of transportation as a distinct body, but was unable to protect the department’s independence. The council advised Haig of their decision to subordinate Crookshank to Clarke in early June and by the end of the month the matter had been decisively settled in Clarke’s favour.⁸⁶

The status and identity of the directorate-general of transportation elicited strong reactions from Britain’s transport experts. Fay predicted gloomily in April 1918 that the organizational changes advocated by Clarke and Cowans would ‘certainly mean the importation of a great deal of feeling, with a probable loss of efficiency or interest’ among the civilians embedded within the military machine.⁸⁷ He returned to the theme in Milner’s office, where he urged the Army Council ‘that great tact must be exercised or the railway men who were volunteers in the organization of Geddes, both officers and men, might become mulish and difficult to handle’.⁸⁸ Fay’s warning implied that the patriotic good will provided to the army by Britain’s transport experts was far from an inexhaustible commodity, and suggests that civil–military relations within the British war effort remained fragile. Two civilian responses to the events of 25 May seemed to confirm Fay’s assessment. Geddes, from his vantage point at the Admiralty, labelled the decision to subordinate Crookshank to Clarke a ‘military conspiracy’. In France, Ralph Wedgwood’s stance indicated that his patience with the army had worn thin by the summer of 1918. As confidence in the DGT’s ability remained low among the professional soldiers at GHQ and in the War Office, Fay – who later told Milner that ‘nothing would induce’ him to take the post if it was under the direct command of a QMG rather than Haig – suggested Wedgwood as a potential successor to Crookshank.⁸⁹ Wedgwood, the director of docks, was eminently suitable for the role. He had considerable pre-war experience of handling freight traffic within the North-Eastern Railway’s goods department, had acted as a railway transport officer at the beginning of the war and had overseen a colossal expansion in output at the BEF’s docks since the directorate’s establishment in early 1917. However, he had become embroiled in an increasingly acrimonious dispute with the Labour Corps over the employment of unskilled workers within the docks during 1918 and told Fay unequivocally that ‘he would have nothing to do with [the position of DGT]. “The army got into a mess before, and were going to get into another now, let them get out of it in their own way.” He was’, Fay reflected, ‘deaf to any argument’.⁹⁰

Wedgwood’s emphatic refusal to take on the role of DGT, founded upon a belief that another transportation ‘mess’ was imminent, illustrates that self-interest retained a powerful influence over men’s actions throughout the conflict. His comments to Fay demonstrated that his motivations were born of a desire to avoid association with a project he anticipated would end in failure, rather than emerging from a sense of loyalty to the incumbent. Indeed, not all of Britain’s transport experts were inclined to support Crookshank as DGT. Henry Maybury, the director of roads, complained to Fay that ‘he could not get a definite decision out of Crookshank, and that things had never worked out well’ during the latter’s tenure as DGT. He implored Fay to ‘send us a boss’ in place of the vacillating soldier. George McLaren Brown was the ‘boss’ Fay identified, a man who had risen from ticket agent for the Canadian Pacific Railway to head the company’s European operations before the outbreak of war. Control of the Canadian Pacific’s extensive fleet of ocean-going vessels lay within Brown’s remit, and he had been drawn into Fay’s directorate of movements to provide ‘general control of the movement of all war material and stores to France … [and the] working of ports in the United Kingdom’.⁹¹

The friction that accompanied Brown’s appearance in France to shadow Crookshank revealed the other strand of self-interest that attended personal relationships across the British war effort between 1914 and 1918. Clarke, echoing the attitude taken by his predecessor when faced by Geddes in 1916, claimed that ‘some of the directors [in the QMG’s department] would not work’ alongside Brown. However, on this occasion the lack of cooperative spirit was not caused by tendencies towards self-preservation within the professional army. Indeed, the most vociferous critic of Brown’s possible appointment as DGT was a civilian, Brigadier-General John Stewart. Like Brown, Stewart’s background was in the railway industry and he had close associations with the Canadian Pacific. Unlike Brown, he had been heavily involved in the British army’s global war effort for much of the conflict. He had been fundamental to the organization of the Canadian Overseas Railway Construction Corps and the rapid construction of the BEF’s light railway network, had been entrusted with the investigation of railway facilities in Egypt, Palestine and Mesopotamia in 1917 and 1918 and had taken on the role of director of construction in the summer of 1918.⁹² Stewart dismissed Brown as ‘only a ticket agent’ who ‘knew nothing about railways’. Fay put the former’s comments down to a personal grudge and an ambition to secure the post of DGT for himself.⁹³

Whether a valid assessment or not, the entire episode demonstrated the continued delicacy of interpersonal relationships within the diverse, pressurised environment of the wartime British army. Britain’s transport experts were human, and susceptible to the same inclinations towards self-preservation, ambition, jealousy, fatigue and obstinacy as their military counterparts. Ultimately, neither Brown nor Stewart succeeded the professional soldier as DGT. Circumstances on the battlefield rather than in council chambers or offices behind the lines ensured that the war was over before Crookshank could be replaced.

From Amiens to the armistice: the transport factor on paper and in practice, 1918

The town of Amiens has acquired a central position in considerations of the German and allied campaigns of 1918, one which stretches beyond the battle that bears its name. Like Ypres in 1914 and 1915, and Verdun in 1916, the protection of Amiens in 1918 took on a symbolic role for General Ferdinand Foch in the wake of his appointment as allied commander-in-chief in late March.⁹⁴ Yet the town possessed a far more significant, practical importance to the BEF that spring. Amiens and Hazebrouck comprised the two principal bottlenecks in the BEF’s railway communications. Almost all of the traffic received by rail from the BEF’s northern ports passed through Hazebrouck en route to the front, while most of the traffic despatched from the southern Channel ports went via Amiens. Around 80 per cent of the allies’ north–south traffic – which in early 1918 averaged 140 trains per day, and was liable to comprise as many as 212 trains per day if a large-scale strategic movement of troops took place – either went through or skirted the town.⁹⁵ Construction of an avoiding line to improve rail capacity around Amiens had commenced in early March, but work on the nine-and-a-half-mile-long deviation had not been completed when the German offensive began.⁹⁶

The depth and pace of the German advance forced the allied high command to acknowledge the threat to Amiens in the opening week of the campaign. The relative distances from the front line of Hazebrouck and Amiens had led Haig to concentrate his defensive efforts on the former, the latter being located ‘at a greater depth than most … commanders would have believed to be vulnerable’ following the events of 1914–17.⁹⁷ Within a few days of 21 March such beliefs had been overturned. In his diary entry for 26 March Haig recorded that the attendees of a meeting at Doullens – which included, among others, the French president and premier, Foch, and Pétain – unanimously ‘decided that Amiens must be covered at all costs’ to prevent its loss to the enemy.⁹⁸ The situation appeared precarious. Construction units hitherto engaged on the avoiding line around the town were redeployed to construct defences under Stewart’s command, while Clarke began to prepare for the removal of personnel, animals and stores from the area between Amiens, Abbeville, Blargies and Dieppe. ‘Scheme X’ was ready on 31 March, and was accompanied shortly after by ‘scheme Y’ – evacuation plans that dealt with the area surrounding Calais and Dunkirk as well as that between Abbeville, Abancourt and Dieppe.⁹⁹ Henry Rawlinson, when he replaced Hubert Gough as army commander in the area on 28 March, captured the town’s significance in a letter to Henry Wilson. He wrote that there could ‘be no question but that the Amiens area is the only one in which the enemy can hope to gain such a success as to force the Allies to discuss terms of peace’.¹⁰⁰ Consequently, he sent an urgent appeal to Foch for more troops. ‘I feel anxious for the safety of Amiens’, he wrote, confiding in his diary on the same day that ‘if the Bosche attack heavily tomorrow I fear he will break our last line of defence in front of Amiens and the place will fall’.¹⁰¹

The town did not fall on 29 March. After a week in which the front line had shifted almost forty miles, the German army’s advance slowed as the transport factor began to assert itself. By 24 March the German Eighteenth Army ‘was starting to feel the effects of fatigue and stretched supply lines’ as the gap between the assault troops and their railheads widened. By the time the Germans occupied Albert two days later they had left their railheads far behind, and the front-line units had been without fresh rations for two days.¹⁰² Between the tiring attackers and their supplies lay ground shattered by the fighting of 1916 and obliterated by the Germans in their retirement during 1917. On 29 March General Georg von der Marwitz, commander of the German Second Army, described the landscape through which his troops had progressed in a letter to his wife:

The forward movement of heavy artillery, ammunition and food became increasingly difficult as the advance progressed. Under such circumstances the Germans were forced to follow a path familiar to the BEF from previous operations; the campaign paused so that guns, fresh troops and supplies could be dragged into position. By the time fourteen divisions were ready to attack on the morning of 4 April the allies had reinforced and fortified their defences ahead of the town. The German Second Army reached the outskirts of Villers-Bretonneux, around ten miles from the allies’ vital railway centre, but the town itself remained in allied hands. Ludendorff terminated the offensive the following day and redirected his energies towards operations further north.¹⁰⁴

The German high command may have won ‘nothing of value’ from operation Michael,¹⁰⁵ but their army’s failure to take Amiens did not ease allied anxieties as to the situation astride the Somme. Foch, Haig and Rawlinson all recognized the town’s psychological significance, but it was the task of civilian railway expert Philip Nash to articulate its practical importance to the allied transportation effort. On 2 April, two days before the final German push for the town began, the British section of the SWC asked Nash to examine the implications for the allies’ traffic capabilities should Amiens fall:

Only three double-tracked routes for traffic across the Somme were available to the allies, two of which crossed the river around Amiens, while two single lines with a far lower capacity provided further options to the west.¹⁰⁷

The report Nash submitted a week later illustrated just how restricted the allies’ options were. If the crucial lines around Amiens were to be rendered unsafe for rail traffic, the remaining lateral routes between the town and the coast were ill-equipped to replace them. In March 1918 the allies had moved an average of 140 trains per day along the north–south routes over the Somme. However, Nash stated that the capacity of the only routes across the river to the west of Amiens – which ran Eu–Abbeville and Gamaches–Longpré – permitted just ninety train movements per day. Furthermore, as his projections did not include any contingency for bad weather or movements required for ‘railway exigencies’ such as the return of empty wagons, the actual capacity of the lines was likely to be at least 10 per cent lower.¹⁰⁸

Nash’s investigation emphasized how the movement of men and materials across the Somme would be constrained by the loss of Amiens. However, his report also warned of the ‘much more serious situation’ that would occur in the event of a German thrust towards Abancourt. The main line that headed west from Abancourt was the crucial link in Nash’s higher estimate of allied rail capacity (see Figure 9.2). If use of the line were denied by enemy action, ‘the possibility of through movement between North and South would be limited by the capacity of the Dieppe–Eu section’ along the Channel coast. ‘This is a single line section with heavy gradients and poor facilities’, Nash explained, ‘so that only eight daily train movements in each direction can be counted upon’. To make matters worse the country surrounding Serqueux, which fed Abancourt along a line that accommodated 100 movements per day, was ‘so hilly and broken that the construction of any new connection to the main line north of the latter was ‘impracticable’.¹⁰⁹ In short, Nash’s examination highlighted that if the allies were unable to use Abancourt then they would lose access to almost their entire capacity for lateral rail movements across the River Somme – the coalition troops and French civilians located north of the river could not be supplied from the ports of Le Havre and Rouen, while the French munitions factories around Paris could not be fuelled by coal delivered direct from the Bruay-Béthune mines.

The precarious situation outlined in Nash’s report was further complicated by events that began on the day it was submitted. The next phase of the German attack forced the allies to consider the parlous state of transportation across the western front, not just in the area immediately jeopardized by the fighting. Following the launch of operation Georgette on 9 April the Germans achieved more spectacular tactical successes, and by nightfall on 11 April the leading German units occupied territory around six miles from the outskirts of Hazebrouck. By the following day the Germans were within artillery range both of the key junction on the BEF’s northern line of communications – which connected the ports of Calais, Boulogne and Dunkirk to the front line – and the Bruay-Béthune coalfield from which 70 per cent of the French munitions industry’s coal was mined.¹¹⁰ As the advance developed Nash informed the British section of the SWC that

Figure 9.2. The Somme crossings west of Amiens, March 1918.

Source: A. M. Henniker, History of the Great War: Transportation on the Western Front, 1914– 1918 (London, 1937), p. 399. Map drawn by Cath D’Alton.

This traffic represented an additional pressure on the communications across the Somme at a time when the presence of active German artillery had severely curtailed movement in the vicinity of Amiens.

The depth of the German penetration in Flanders compelled the allies into action. Within the BEF, Clarke’s hitherto localized plans for the evacuation of personnel and stores evolved into a sophisticated and detailed scheme, designated ‘Z’, for the complete abandonment of the area north of the Somme. As Ian M. Brown has noted, the BEF ‘had spent the better part of three years building this line of communication, along with its attendant infrastructure – rail improvements, depots, bases, port improvements, road improvements, light railways, ammunition depots’.¹¹² All were scheduled to be removed, destroyed or otherwise rendered inoperable to the enemy. Scheme ‘Z’ ultimately provided the QMG with a timetable for the evacuation of 250,000 men, their attendant equipment and 600,000 tons of stores from northern France in just twenty-eight days.¹¹³ While many of the preparations remained hypothetical, on 11 April the units of the locomotive repair shop at Borre near Hazebrouck – established in April 1917 with the capacity to undertake 200 heavy and 120 light repairs per year – were ordered to dismantle its machinery and remove the contents from the enemy’s grasp. ‘The lads worked so well’, recalled Colonel L. S. Simpson, ‘that in three days and nights they had got out, loaded up, and dispatched to Audruicq practically everything except the big wheel lathes and an engine or two that were not down on their wheels – this under shell fire most of the time – and I am glad to say three hundred of them turned up at Audruicq safe and sound on the 13th, followed the next day by the remainder’. The evacuation did not halt work for long. Within a week the ROD had opened up a repair shop at Rang-du-Fliers, to the south of Étaples, which was ‘turning out three to five engines a week, besides repairing metre gauge stock and making ballast ploughs, telegraph post slotting machines, and a hundred and one other things’ for the remainder of the war.¹¹⁴

The German advance also created immediate concerns for the BEF’s allies. On 12 April Haig received warning from the French president, Georges Clemenceau, that only five days’ reserves of coal existed in France.¹¹⁵ Without access to continuous supplies of coal the French munitions industry could not function. However, Nash was emphatic in his belief that it was ‘absolutely necessary, from the purely military point of view, to immediately free the single North and South lateral in Allied hands from all but military traffic. This means that the whole of the Coal traffic from the Pas de Calais to the area South of the Somme must immediately be shut down’.¹¹⁶

The manner in which the allies solved this dilemma illustrates the extent to which the French and British war efforts had become intertwined by April 1918. On 19 April, when Nash was emphasizing to the British section of the SWC that traffic from the Bruay-Béthune coalfield had to be suspended, the French minister of armaments informed him that ‘in order to make good the loss of the Pas de Calais mines’ the French would require 600,000 tons of coal per month from England.¹¹⁷ Consequently, an inter-allied meeting took place at the Ministry of Armaments in Paris on 23 April, during which experts from both nations – taking into account the higher calorific content of British coal in comparison to French coal – thrashed out an agreement for Britain to import 450,000 tons of coal per month into ports south of the Somme in the event that the Germans rendered the northern coalfields inoperable.¹¹⁸

Nash’s experiences of coalition warfare during the German spring offensives provide a startling contrast both to the terse and suspicious atmosphere that surrounded Sir Eric Geddes’s negotiations with the French in 1917 and Nash’s own attempts to coordinate transportation matters on the Italian front in 1918. The speed and depth with which the German army advanced after 21 March compelled the French and British to subordinate their insular domestic objectives for the benefit of the alliance’s survival as an effective fighting force. By late April, thanks to the energies of transport experts from both nations and a commitment to the maintenance of communications between them, the allies had managed to reduce north– south coal traffic by one-third and had taken steps to ensure the continued supply of fuel for the French economy should events at the front compel the abandonment of the Bruay-Béthune mines.¹¹⁹

However, the reduction of traffic flows across the Somme in spring 1918 was ‘merely a palliative’ and comprised only half of the parameters for Nash’s examinations.¹²⁰ Alongside his instructions to identify the potential economies available on the north–south routes behind the allied front, the SWC requested that he assess the ‘desirability and practicability’ of railway improvements that ‘would materially improve the general transportation situation’ were Amiens to be captured.¹²¹ By 9 April he was able to report that ‘certain works’ that could ‘immediately give some relief’ to the allies’ transport constraints were already ‘in hand’, while further construction efforts to increase the routes available for northsouth traffic were being contemplated.¹²²

A fortnight later that contemplation had resulted in an action plan agreed by both the British and French railway authorities.¹²³ As coalition troops continued to frustrate successive German assaults over the following months, allied engineers undertook an extensive scheme of railway construction that greatly increased the allies’ capacity to move men and materials within the truncated space between the front line and the Channel coast. British engineers improved stations on the Gamaches–Longpré line to provide the Fourth Army with more railheads, doubled the thirteen-mile-long line between Longpré and Martainville and the twenty-seven miles of line between Abbeville and Frévent, contributed to the duplication of the line south from Étaples to Port-le-Grand, and engaged in a range of improvement works behind the Flanders front.¹²⁴ In May 1918 alone the British built or reconstructed 148.74 miles of broad-gauge track.¹²⁵ Yet their efforts were dwarfed by those of the French, whose principal achievement was to survey, prepare and construct an entirely new double-line some fifty-five miles long from a position south-east of Abancourt to a new connection across the Somme.¹²⁶ The line, which had only been set out on the ground on 30 April, was linked through from end to end by 15 July and ready for traffic a month later (see Figure 9.3). As Henniker noted admiringly in the British official history, ‘a trunk line complete with engine sheds, water supplies, signalling, telephones, station buildings, etc., had been constructed in … 106 days from the date on which work started’.¹²⁷ It was, concurred the Royal Engineers’ history of the war, a ‘very remarkable feat’.¹²⁸ Upon completion the new line, combined with the other construction undertaken that summer, reduced the proportion of north–south traffic that had to pass through Amiens: three double-lines, with a capacity of 144 trains per day in each direction, provided the allies with more lateral rail options than they had possessed before the Germans had attacked. In short, by the late summer of 1918 the BEF’s railway communications were in better shape than they had been before 21 March.¹²⁹

Yet despite the comparative reduction in the town’s importance to the allies’ lines of communications, Amiens still represented the perfect location from which they could exploit their freshly installed transport options. Haig had already raised the idea of pushing the Germans beyond artillery range of the town before 18 July, when a counterattack on the Marne by troops of the French Tenth Army indicated that Ludendorff had lost the initiative on the western front.¹³⁰ The German troops who occupied the sector ‘were not the formidable fighters of the March offensive’; their defences were ‘inadequate’, their morale was poor, and the difficulties of supply across the devastated landscape were such that their rations were deemed ‘very bad and scarce’.¹³¹ Between March and the end of July the German army had suffered 977,555 casualties, which it had found increasingly difficult to replace.¹³²

Figure 9.3. The Somme crossings west of Amiens, August 1918.

Source: A. M. Henniker, History of the Great War: Transportation on the Western Front, 1914– 1918 (London, 1937), p. 400. Map drawn by Cath D’Alton.

By contrast, the transport infrastructure behind the allies facilitated the accumulation of significant human and material resources in the sector ahead of the battle of Amiens. Between 27 July and 10 August Rawlinson’s Fourth Army expanded greatly; the number of men attached to the army rose from 257,567 to 441,538, and the number of horses from 54,323 to 98,716.¹³³ The BEF drew together three cavalry and seventeen infantry divisions for the offensive, with each battalion of the latter equipped with thirty Lewis guns, eight light trench mortars and at least sixteen rifle-grenadiers. They were supported by 534 tanks, 800 aeroplanes and over 2,000 lavishly stocked guns – each 60-pounder possessed sufficient shells to fire four rounds per minute for four hours on zero day. Alongside them the French First Army comprised fifteen infantry divisions, over 1,000 aeroplanes and 1,624 light and heavy guns. Across no man’s land sat eleven depleted German divisions, supported by fewer than 400 aircraft and inadequate numbers of artillery.¹³⁴ ‘Given the relative strengths of the forces involved’, argued Rob Thompson, ‘it was clear that this assault would be victorious’ for the allies.¹³⁵

The scale and manner of the allied victory on 8 August has given the battle of Amiens a prominent position in the (especially Anglophile) revisionist history of the conflict.¹³⁶ The results of the first day have provided ample evidence to support ‘the perception that the British applied sophisticated, fire-power-based combined arms methods to overcome German defences and restore mobility to the battlefield’.¹³⁷ The BEF’s leading units had driven the Germans back some eight miles by nightfall on the first day, captured over 12,000 prisoners and 374 guns and cleared any lingering threat to the railway communications around the town. When operations were suspended after 11 August the British and French between them had inflicted 48,000 casualties and asserted the allies’ material, psychological and tactical superiority over their opponents – a predominance they did not relinquish for the remainder of the war.¹³⁸

Those advantages were entirely underpinned by, and their maintenance dependent upon, transportation. The allies’ material superiority over the Germans provided them with the means by which to prosecute a more intensive form of warfare in the second half of 1918. During the hundred days the BEF participated in a series of concentric operations, conceived by Foch as a ‘sequence of offensives, each one within the capabilities of a single army’s fighting power and logistics, … engaged at a pace that would exhaust Germany’s ability to sustain battle once and for all’.¹³⁹ To sustain pressure on the retiring German forces, the BEF and its allies had to ensure that the tempo of their operations over the late summer and autumn of 1918 were sufficient to prevent the enemy from preparing positions that could withstand the allied onslaught. Once the enemy had been compelled to fall back, a successful pursuit depended upon the timely movement of ammunition, supplies and materials across the battlefield in sufficient volume to render the Germans unable to prepare new defensive lines. Faced by a tenacious, determined opponent, the BEF could not abandon the industrial machinery upon which successful combined-arms operations relied. Therefore, as much as could be physically moved had to be shifted forward over the available roads, railway lines and waterways to maintain pressure on the enemy. Ultimately, the transport infrastructure in France and Flanders represented a fundamental component of the allies’ weapons system during the advance to victory – one that had to be exploited with greater success than the Germans had achieved in the spring.

The altered situation on the western front engendered a re-evaluation of the BEF’s transport organization. The conditions within which the BEF operated as a mobile fighting force differed significantly from those that had been in place when Sir Eric Geddes had created the directorate-general of transportation. Clarke was quick to perceive how the return of movement to the battlefield had changed his department’s task, and maintained his insistence that a military-led organization was better suited to provide transport for a mobile army than a civilian one:

The QMG’s transport policy, issued as guidelines to the BEF’s individual armies on 27 August, reflected his attitude. He recommended that work on light railways, the medium most closely associated with Geddes’s reforms of the BEF’s transport system, was to cease unless conditions for their repair were ‘specially favourable’.¹⁴¹ A combination of mechanical and horse-drawn transport would ‘take up most of the load from railheads’ to the front line as the army pressed forward. In essence, Clarke recommended a return to the transport policy envisaged within the army’s pre-war – pre-civilianized – organization.¹⁴²

There was more to Clarke’s new transport policy than his slight against the perceived blind spot of civilian specialists suggests. The QMG’s decision to recommend that the main line railways and roads received priority attention underlined both the BEF’s reliance on the former for the bulk movement of goods inland from the ports and the latter’s advantages in the fluid conditions of a mobile war. Put simply, light railways were poorly equipped to deal with the rigours of a general advance over a substantial distance. As Henniker explained in the official history:

Furthermore, the light railway organization that had been steadily developed over the course of 1917 had been severely dislocated by the German spring offensives. Some 560 miles of light railway had been destroyed or captured by the Germans, while the redeployment of light railway troops into broad gauge construction units meant that by August 1918 the BEF’s light railway directorate lacked the human and material resources that a comprehensive use of the medium required.¹⁴⁴

However, despite the medium’s limitations and the contents of Clarke’s guidance, light railways were exploited whenever circumstances permitted during the hundred days. As has been previously demonstrated for the BEF’s devolved approach to battle planning, GHQ provided army and corps commands with considerable latitude to exercise initiative in the realm of transport improvements.¹⁴⁵ In practice, Clarke requested that the BEF’s individual armies obtain permission for light railway construction during the hundred days only when the proposed projects required substantial resource commitments, affected the principal west–east trunk routes or demanded inter-army coordination.¹⁴⁶ Early in the advance, when German systems were captured in good order or existing lines could be connected to the British network with relative ease, light railways provided valuable bulk distribution lines between the standard-gauge railheads and forward dumps. During August the BEF’s advance was supported by a light railway network that conveyed a weekly average of 157,651 tons – significantly lower than the figures recorded during the previous year’s operations at Third Ypres, but a substantial increase on the 109,172 tons carried each week during May 1918. Over the course of the hundred days the mileage of light railways operated by the BEF increased by 35 per cent,¹⁴⁷ shaped into a series of ‘long antennae’ that ran west–east in pursuit of the advancing troops. The Third Army built up a particularly successful line, which crossed the St Quentin Canal and extended over thirty-five miles between Fosseux and Crèvecœur-sur-l’Escaut.¹⁴⁸

Clarke’s observations on the difference between civilian and military leadership in the field of army transportation fixated on a distinction that had become increasingly blurred by August 1918. The QMG’s subordination of the directorate-general of transportation did not mean that Britain’s transport experts lacked influence over the BEF’s operations in the war’s final campaign. Rather, their principal contribution was felt on a practical rather than managerial level. Between late 1916 and the summer of 1918 the directorate-general of transportation had been conceived, supplied with skilled personnel and industrial equipment and embedded within a military supply machine that stretched from the Channel coast to the front line; the organizational changes demanded by industrial warfare had been devised and implemented long before the battle of Amiens. After the battle, the directorate-general of transportation experienced its ‘severest test’ of the war:

That ceaseless work – rather than entailing the creation and establishment of new organizational structures and using pioneering managerial techniques to monitor the performance of a dispersed workforce – primarily comprised unloading, loading, driving, navigating, building, repairing, and operating the BEF’s panoply of transport options in support of the fighting troops.

The predominantly civilian workforce within the directorate-general of transportation responded to the changed circumstances of the war with vigour. Their endeavours were prodigious across all areas of the BEF’s transportation service in the final months of the war. At the six principal docks under the direction of Brigadier-General Ralph Wedgwood, the North-Eastern Railway’s chief goods manager, average weekly imports rose from 150,300 to 173,270 tons between August and October 1918.¹⁵⁰ At an individual level, the tonnage handled per man per hour increased by 15 per cent over the same period. Improvements also took place within the IWT service commanded by Brigadier-General Cyril Luck – the ex-Royal Indian Marine successor to the LNWR’s Gerald Holland. In October the BEF’s fleet of tugs and barges conveyed an average of 66,368 tons per week across the navigable waterways of France and Flanders, an increase of 20 per cent over the figures recorded the previous June.

Substantial as these improvements were, the BEF’s operational tempo during the hundred days primarily depended upon the speed with which the railways and roads behind the front line could be extended. The railway construction troops’ initial progress was good. By 8 September the most easterly divisional railheads available to the Fourth Army were located around Bapaume, some thirty-two miles away from their position before the battle of Amiens, and the entire length of the Amiens–Albert–Arras line was handed over to the ROD for operation the following day.¹⁵¹ From the nadir of May, when the German spring offensives had left the BEF in possession of just 220.7 miles of operable railways, by the end of September the British portion of the rail network behind the western front comprised 485.3 miles of track. In the first two months of the hundred days an average of 153 trains per day arrived into the BEF’s railheads, loaded with ammunition and supplies for onward distribution to the advancing troops.¹⁵²

The ROD, commanded by the Midland Railway’s superintendent of the line, Colonel Cecil Paget, played a critical role in facilitating the British contribution to the battles of the Hindenburg Line in late September. The BEF had fired a daily average of 4,748 tons the previous June, which had required the provision of sixteen daily ammunition trains by Paget’s division. Across September as a whole the ROD ran an average of twenty-four, and as many as thirty-three, ammunition trains per day as the force accumulated the firepower considered necessary to destroy the German army’s last major defensive system. Across August and September as a whole the BEF hurled a daily average of more than 8,000 tons of munitions across no man’s land, a period of artillery expenditure that peaked with the colossal bombardment detailed at the beginning of this chapter.¹⁵³ In the opening phase of the hundred days campaign, which climaxed when the 46th Division crossed the St Quentin Canal and pierced the Hindenburg Line, the requirements of Foch’s strategic plan rather than logistical constraints influenced the scale and timing of British military operations.¹⁵⁴

However, the efforts and endeavours of Britain’s transport experts could only ameliorate rather than eliminate the transport factor in the weeks that followed. The prodigious weight of fire employed by the BEF’s artillery during August and September was not sustained in October, as the difficulties inherent in the supply of a large, moving army over the available railway and road networks steadily accumulated. First, the pace of railway construction behind the advancing force slowed considerably.¹⁵⁵ During August and September 86.04 miles of new track were laid and 518.19 miles reconstructed. In early September Sir Sam Fay had secured agreement with the Ministry of Munitions to raise the reserve levels of track in France to 500 miles by the end of October (an increase of 20 per cent). However, Clarke recorded on 5 October that shipments of track had not kept up with the programme and that consequently ‘the stock of rails [was] very much reduced’.¹⁵⁶ Delay-action mines, lodged in railway embankments and bridges by retiring German engineers, added further complications for the railway construction troops – in the words of one commentator they ‘caused more transportation difficulties than the whole of the previous four years had produced’.¹⁵⁷ On 19 October the QMG’s diary recorded that a supply train had run into a crater, while progress in the Cambrai–St Quentin area was particularly affected by ‘a large number of accidents and the explosion of delay-action mines’ over the course of the following week.¹⁵⁸ When the Third Army successfully dislodged the Germans from defensive positions around the River Selle in a night attack on 20 October, it was ‘unable to exploit its success’ for three days due to the slow arrival of the force’s ammunition trains.¹⁵⁹ Increased congestion on the rail network meant that a daily average of 133 trains ran into railheads during October, a decrease of twenty per day that contributed markedly to a decline of one-third in the BEF’s artillery use in October 1918.¹⁶⁰

The reduced capacity and reliability of the BEF’s rail lines had profound implications for the rest of the force’s transport infrastructure. A reduction in the deliveries of stone slowed the construction of new railheads and roads closer to the front, which increased the distances to be covered by mechanical transport at the same time as it decreased the amount of material available for the upkeep of the road surface. Furthermore, when an undetected device exploded on 3 November and destroyed a bridge on the Cambrai–Busigny line that had only been repaired a fortnight earlier, it forced the divisions of IV and V Corps to draw their supplies from railheads ten miles further to the rear.¹⁶¹ At the same time the bulk of the Fourth Army’s railheads were located between twenty and twenty-five miles behind the front. Over the next week Rawlinson’s troops advanced a further twenty-five miles, but a combination of mines, destroyed bridges, and accidents ensured that few new railheads could be opened to traffic. ‘At the date of the Armistice’, Henniker recorded, ‘the only reliable railheads for the Fourth Army were 50 miles behind the Armistice line; in the north even the most advanced railheads of the Fifth Army were 30 miles behind it’.¹⁶²

The task of bridging the ever-widening gap fell chiefly upon the BEF’s fleet of lorry drivers, whose bodies and vehicles were subjected to punishing workloads as the hundred days progressed. The lorries of the 14th GHQ Reserve Mechanical Transport Company were on the road for around one hundred hours over the course of five days in late September, while Henniker recorded cases of ‘columns taking seventy-two hours to complete what should have been a daily round’. The combination of tired drivers, poor roads and long journeys – which reduced the time available for vehicles to be serviced – increased the number of accidents and breakdowns among the BEF’s pool of lorries.¹⁶³ At the end of October Clarke reported that ‘some 3,600 lorries were out of action (out of a total of about 20,000 working in Army areas) chiefly due to broken springs’,¹⁶⁴ which further degraded the supply situation at the front. Eighty lorries were required to carry ammunition for VI Corps’ heavy artillery. By 2 November only fifty-six were available to accumulate firepower ahead of the battle of the Sambre.¹⁶⁵

The BEF’s victory on the Sambre sent the German army into a general retreat. For the remainder of the war the ‘British trudged on through cold and wet across the sabotaged and booby-trapped landscape that the Germans had left behind them’. The ‘weather and logistical difficulties rather than the Germans were the main obstacle to the advance’ in the final phase of the hundred days.¹⁶⁶ Since 8 August the BEF’s spearhead formations, the Third and Fourth armies, had advanced approximately sixty and 100 miles respectively.¹⁶⁷ The repairs of mine craters in the roads could not keep pace with the advance, as congestion, delays, accidents, the demands of civilian populations in liberated territory and breakdowns across the transport network severely decreased the volume of material to reach the front. By the week ending 9 November the BEF’s transportation network echoed the paralysis experienced on the Somme two years earlier. Ten accidents and sixteen mine explosions – mostly on the critical Cambrai-St Quentin line – reduced the speed with which trains arrived at railheads, were unloaded and returned to the bases for fresh loads. Consequently, a ‘serious shortage of trucks’ existed at almost all of Wedgwood’s ports, which ‘resulted in congestion of quays and very considerable delay in discharging ships’.¹⁶⁸ Beyond the railheads, the belt of country only passable by low-capacity animal transport rapidly widened. The roads directorate, under the road board’s Brigadier-General Henry Maybury, was responsible for the construction, repair and maintenance of a network comprising 4,412 miles of road – he had, Fay reflected after the armistice, ‘left his mark all over the roads in Northern France’. However, like the majority of the men employed on supply and transportation duties in November 1918, Maybury was worn out.¹⁶⁹ Men, mostly of advanced years and lower standards of physical fitness, had been tested to their physical and psychological limits; their lorries had been ‘knocked about’ and overworked.¹⁷⁰

The demands of mobile warfare, waged by a mass army backed by the entire impedimenta that successful combined-arms operations required, could no longer be effectively met by the manpower and resources available for its supply. On the afternoon of 9 November the Third Army instructed IV, V and XVII Corps to consolidate their positions and echelon back in depth to reduce the army’s transport burden. VI Corps took over responsibility for the whole of the Third Army’s front, and was ordered to act as an advanced guard and keep in touch with the German retirement.¹⁷¹ To the Third Army’s right, responsibility for the Fourth Army’s pursuit was devolved upon Major-General Hugh Keppel Bethell’s 66th Division on the same morning. The Fourth Army had deployed over half-a-million men and animals and 2,000 guns at the battle of Amiens. Bethell’s force comprised only one infantry brigade, one cavalry brigade, seven armoured cars, a field gun battery, two sections of 4.5-inch howitzers, an anti-aircraft section, three field companies of engineers, a pioneer battalion and the division’s machine-gun battalion. It represented all that the available transport infrastructure could sustain just three months later.¹⁷² The heavy artillery, tanks, light railways, repair workshops and sundry units and services that had contributed to the successful operations conducted by Rawlinson’s forces throughout the course of the hundred days were abandoned. The transport factor made it impossible to feed a larger, hungrier force.¹⁷³ Consequently, Bethell’s force was ordered to maintain contact with the Germans the following day, but to do no more.¹⁷⁴ At 11 a.m. the next morning the war on the western front came to an end.

Conclusion

The BEF’s transport problems did not end when the guns fell silent. The reduced demand from the front for materials directly related to the conduct of military operations, such as ammunition, was counteracted by requests for the construction materials necessary to reconnect the pre-war rail links between France, Belgium and Germany. ‘There was’, recorded Clarke in his review of November 1918:

The limited train capacity of the ‘hastily reconstructed lines in the forward areas, devoid of all ordinary facilities for working, remained; reconstruction across the [devastated area] only increased the length of line over which traffic was precarious and intensified the shortage of rolling stock’.¹⁷⁶ The BEF could not advance to the German border in strength even when the military force opposing it had ceased to offer resistance.

However, the transport infrastructure that by November 1918 was inadequately equipped to move and supply an industrial army had proven itself good enough to facilitate the allied success on the western front over the previous two years. Britain’s transport experts played a key role in that success. The integrated transport directorate established by Sir Eric Geddes, populated with civilians, and embedded within the military hierarchy over the winter of 1916–17 facilitated the prosecution of a material-intensive war on a scale beyond the BEF’s capability during the battle of the Somme. The directorate-general of transportation created the circumstances in which British gunners could add significantly to the destruction wrought upon the French and Belgian landscape by the artillery of the First World War, not least at Third Ypres. What David Lloyd George referred to as ‘the campaign of the mud’ in his indictment of the BEF’s senior commanders was – at least in part – the responsibility of the prime minister’s desire to introduce civilian specialists to the administration of the British army.

The events of 1917 and 1918 demonstrate the inaccuracy of Lloyd George’s statements regarding the imposition of supposedly superior civilian methods upon a backward-looking, reactionary army. Geddes’s system was well suited to the task of supplying a stationary force, but it sank into the quagmire it had created at Passchendaele and proved ill-equipped to service the requirements of a mobile force. The BEF’s ultimately successful transportation effort during the final year of the war came about through the amalgamation of civilian and military expertise; recognisably non-military technologies, methods, and personnel were applied to the identifiably military problem of sustaining a relentless offensive pressure against a retreating opponent. By the summer of 1918 the line between civilian and military existed more in the minds of the personalities at work in the BEF’s administrative hierarchy than it did in the directorate-general of transportation’s practical accomplishments. Sir Douglas Haig ‘could launch simultaneous offensives or sequential ones on widely separated fronts – something that had been unthinkable before 1918’.¹⁷⁷ Those offensives were a crucial factor in the German decision to seek an armistice.

Yet neither Haig, nor Geddes, Travers Clarke, Sydney Crookshank, or any of the other individuals responsible for the supply of the BEF during the hundred days could entirely eliminate the realities of mobile, material-intensive, modern warfare. The German spring offensives provided a graphic demonstration that the armies of the First World War could not tear loose from their railheads, penetrate deep into enemy territory, and maintain the intensity and tempo of their operations.¹⁷⁸ The BEF’s advance could not have taken place without adequate, sustained access to supplies of food, ammunition, and myriad goods of both direct and indirect relationship to the conduct of military operations. Consequently, in the words of a staff officer who was responsible for XIII Corps’ supply arrangements throughout the hundred days, the final campaign of the war was ‘the most cumbrous steam-roller affair it [was] possible to conceive’.¹⁷⁹ Britain’s transport experts had helped to keep it in motion for long enough.

______________

¹ I. M. Brown, British Logistics on the Western Front, 1914–1919 (London, 1998), p. 174.

² The statistics in this passage are derived from J. Terraine, The Smoke and the Fire: Myths and Anti-Myths of War, 1861–1945 (London, 1980), pp. 118–19; D. Stevenson, With Our Backs to the Wall: Victory and Defeat in 1918 (London, 2011), p. 379; B. Bond, British Military Policy Between the Two World Wars (Oxford, 1980), p. 5.

³ Stevenson, With Our Backs to the Wall, p. 223.

⁴ D. Stevenson, ‘1918 revisited’, Journal of Strategic Studies, xxviii (2005), 107–39, at pp. 113–19.

⁵ W. Philpott, Anglo-French Relations and Strategy on the Western Front, 1914–18 (Basingstoke, 1996), p. 129.

⁶ TNA, WO 95/56, director of IWT, war diary, memorandum 4, p. 1.

⁷ NLS, Haig papers, Acc. 3155/109, diary entries, 22 and 26 Dec. 1916; Brown, British Logistics, pp. 156–8.

⁸ TNA, WO 95/56, director of IWT war diary, memorandum 4, pp. 2–4; A. M. Henniker, History of the Great War: Transportation on the Western Front, 1914–1918 (London, 1937), p. 218.

⁹ D. Stevenson, 1917: War, Peace, and Revolution (Oxford, 2017), pp. 67–87.

¹⁰ TNA, WO 158/851, history of IWT, p. 58; E. A. Pratt, British Railways and the Great War: Organisation, Efforts, Difficulties and Achievements (2 vols., London, 1921), ii. 1107–9.

¹¹ M. G. Taylor, ‘Land transportation in the late war’, Royal United Services Institution. Journal, lxvi (1921), 699–722, at p. 710. Adrian Hodgkin recorded that three fully loaded barges carried the equivalent of 100 wagons of 10-tons’ capacity. See IWM, Hodgkin papers, diary entry, 28 June 1916.

¹² History of the Corps of Royal Engineers, ed. H. L. Pritchard (11 vols., Chatham, 1952), v. 630; NLS, Haig papers, Acc. 3155/110, diary entry, 17 Feb. 1917.

¹³ TNA, WO 158/852, history of light railways, pp. 5–6.

¹⁴ K. Jeffery, 1916: a Global History (London, 2015), pp. 257–8.

¹⁵ I. M. Brown, ‘The evolution of the British army’s logistical and administrative infrastructure and its influence on GHQ’s operational and strategic decision-making on the western front, 1914–1918’ (unpublished University of London PhD thesis, 1996), p. 218 n. 13.

¹⁶ C. Falls, History of the Great War. Military Operations, France and Belgium, 1917 (3 vols., London, 1940), i. 546.

¹⁷ Falls, Military Operations, France and Belgium, 1917, i. 191 n. 2.

¹⁸ TNA, WO 107/69, work of the QMG’s branch, pp. 20–1.

¹⁹ Pritchard, History of the Royal Engineers, v. 296.

²⁰ W. J. K. Davies, Light Railways of the First World War: a History of Tactical Rail Communications on the British Fronts, 1914–18 (Newton Abbot, 1967), pp. 61–2; Pritchard, History of the Royal Engineers, v, pp. 293–4.

²¹ Falls, Military Operations, France and Belgium, 1917, i. 547.

²² J. E. Edmonds, History of the Great War. Military Operations, France and Belgium, 1917 (3 vols., London, 1948), ii. 39–40; Davies, Light Railways of the First World War, pp. 67–8.

²³ R. Prior and T. Wilson, Passchendaele: the Untold Story (New Haven, Conn., 1996), p. 59.

²⁴ Edmonds, Military Operations, France and Belgium, 1917, ii. 49.

²⁵ N. Lloyd, Passchendaele: a New History (London, 2017), pp. 56–7.

²⁶ Prior and Wilson, Passchendaele, p. 65.

²⁷ Prior and Wilson, Passchendaele, p. 58.

²⁸ Edmonds, Military Operations, France and Belgium, 1917, ii. 18.

²⁹ Edmonds, Military Operations, France and Belgium, 1917, ii. 138.

³⁰ J. Hussey, ‘The Flanders battleground and the weather in 1917’, in Passchendaele in Perspective: the Third Battle of Ypres, ed. P. H. Liddle (London, 1997), pp. 140–58.

³¹ NLS, Haig papers, Acc. 3155/116, diary entry, 7 Aug. 1917.

³² NLS, Haig papers, Acc. 3155/116, diary entry, 1 Aug. 1917.

³³ Lloyd, Passchendaele, p. 123.

³⁴ Prior and Wilson, Passchendaele, p. 98.

³⁵ Stevenson, 1917, p. 198.

³⁶ BLSC, Liddle collection, papers of Brigadier-General W. R. Ludlow, LIDDLE/WW1/GS0984, diary entry, 1 Sept. 1918.

³⁷ TNA, WO 158/852, history of light railways, p. 7; Davies, Light Railways of the First World War, pp. 74–5.

³⁸ For a discussion of Gough’s attacks in the opening phase of the battle, see Prior and Wilson, Passchendaele, pp. 98–110.

³⁹ Prior and Wilson, Passchendaele, pp. 115, 128.

⁴⁰ R. Thompson, ‘Mud, blood and wood: BEF operational combat and logistico-engineering during the battle of Third Ypres, 1917’, in Fields of Battle: Terrain in Military History, ed. P. Doyle and M. R. Bennett (Dordrecht, 2002), pp. 237–55, at pp. 141–2; Davies, Light Railways of the First World War, pp. 72–3; TNA, WO 107/69, work of the QMG’s branch, p. 21.

⁴¹ H. Gough, The Fifth Army (London, 1931), pp. 214–15.

⁴² NLS, Haig papers, Acc. 3155/118, diary entry, 13 Oct. 1917.

⁴³ Taylor, ‘Land transportation’, p. 707.

⁴⁴ Taylor, ‘Land transportation’, p. 707.

⁴⁵ TNA, WO 107/296, report of British armies, p. 24.

⁴⁶ Edmonds, Military Operations, France and Belgium, 1917, ii. 238; Thompson, ‘Mud, blood and wood’, p. 246.

⁴⁷ BLSC, Ludlow papers, LIDDLE/WW1/GS0984, diary entry, 10 Oct. 1917.

⁴⁸ Pritchard, History of the Royal Engineers, v. 312.

⁴⁹ G. Sheffield, The Chief: Douglas Haig and the British Army (London, 2011), pp. 101–2; Brown, British Logistics, pp. 173–4.

⁵⁰ Thompson, ‘Mud, blood and wood’, p. 250.

⁵¹ Thompson, ‘Mud, blood and wood’, p. 248. Thorough records of the work undertaken by the Canadian engineers in support of the Passchendaele offensives can be found in TNA, WO 95/1063, Canadian Corps. Chief engineer war diary, Aug. to Dec. 1917.

⁵² Lloyd, Passchendaele, pp. 266–86.

⁵³ TNA, WO 95/1063, Canadian Corps. Headquarters branches and services: chief engineer, Methods of distribution employed for ammunition, 10 Nov. 1917, pp. 1–2.

⁵⁴ W. Philpott, ‘Beyond the “learning curve”: the British army’s military transformation in the First World War’, RUSI, 2009 <https://rusi.org/commentary/beyond-learning-curve-british-armys-military-transformation-first-world-war> [accessed 16 Dec. 2017].

⁵⁵ Prior and Wilson, Passchendaele, p. 194.

⁵⁶ S. Fay, The War Office at War (London, 1937), pp. 141–4.

⁵⁷ NLS, Haig papers, Acc. 3155/118, diary entry, 5 Oct. 1917. Unfortunately, Haig’s diary does not elaborate upon Geddes’s justification for the recommendation.

⁵⁸ UWMRC, Granet papers, MSS. 191/3/4/160, Haig to Geddes, 21 Oct. 1917.

⁵⁹ PA, Lloyd George papers, LG/F/17/6/14, Geddes to Derby, 17 Nov. 1917; Fay, The War Office at War, pp. 101, 107.

⁶⁰ For a brief overview of the prominent changes to Haig’s senior staff in 1917–1918, see I. Beckett, T. Bowman and M. Connelly, The British Army and the First World War (Cambridge, 2016), pp. 346–8.

⁶¹ NLS Haig papers, Acc. 3155/120, diary entry, 16 Dec. 1917. Brown, British Logistics, pp. 180–1 suggests that Sir John Cowans, the QMG at the War Office, orchestrated Maxwell’s removal from GHQ.

⁶² ‘G.S.O.’, G.H.Q. (Montreuil-Sur-Mer) (London, 1920), p. 226.

⁶³ J. Charteris, At G.H.Q. (London, 1931), pp. 282–3; Fay, The War Office at War, p. 160.

⁶⁴ NLS, Haig papers, Acc. 3155/126, diary entries, 21–23 Apr. 1918. Crookshank, Haig discovered, resented what he perceived as Clarke’s use of the daily interdepartmental meetings as an opportunity to criticize the work of the DGT’s office.

⁶⁵ Fay, The War Office at War, pp. 189–90.

⁶⁶ Fay, The War Office at War, p. 193.

⁶⁷ TNA, WO 95/38, branches and services: quarter-master general, an explanatory review of the work of Apr. 1918, p. 2.

⁶⁸ Nash, quoted in Fay, The War Office at War, p. 190.

⁶⁹ TNA, WO 107/69, work of the QMG’s branch, pp. 2–3.

⁷⁰ TNA, WO 95/37, branches and services: Quarter-master general, Circular to all armies, 28 Feb. 1918. The necessary work of reconditioning lorries for service in the GHQ reserve companies ‘was perhaps sixty per cent’ complete by 21 March. See Taylor, ‘Land transportation’, p. 708.

⁷¹ Stevenson, With Our Backs to the Wall, p. 42.

⁷² Stevenson, With Our Backs to the Wall, pp. 53–4.

⁷³ R. H. Beadon, The Royal Army Service Corps: a History of Transport and Supply in the British Army (2 vols., Cambridge, 1931), ii. 132; Henniker, History of the Great War, p. 371.

⁷⁴ J. Boff, Haig’s Enemy: Crown Prince Rupprecht and Germany’s War on the Western Front (Oxford, 2018), pp. 211–14.

⁷⁵ Henniker, History of the Great War, p. 372; Davies, Light Railways of the First World War, pp. 90–1.

⁷⁶ TNA, WO 158/852, history of light railways, p. 22; Davies, Light Railways of the First World War, p. 92.

⁷⁷ Brown, British Logistics, p. 189.

⁷⁸ TNA, WO 95/38, QMG war diary, explanatory review of the work of April 1918, p. 1; WO 107/296, report of British armies, pp. 8–9; Brown, British Logistics, p. 189; W. G. Lindsell, ‘Administrative lessons of the Great War’, Royal United Services Institution. Journal, lxxi (1926), 712–20, at pp. 715–16.

⁷⁹ TNA, WO 95/38, QMG war diary, explanatory review – May 1918, pp. 3–4.

⁸⁰ Stevenson, With Our Backs to the Wall, pp. 66–7.

⁸¹ TNA, WO 95/38, QMG war diary, explanatory review – May 1918, p. 1.

⁸² M. Young, Army Service Corps, 1902–1918 (London, 2000), p. 121.

⁸³ TNA, WO 95/38, QMG war diary, explanatory review of the work of April 1918, p. 6.

⁸⁴ Lindsell, ‘Administrative lessons’, p. 716. Lindsell later reinforced the message in an instruction manual on the subject of military administration published in 1933. See W. G. Lindsell, A. and Q.: or Military Administration in War (Aldershot, 1933), pp. 127–38.

⁸⁵ TNA, WO 107/16, IGC, general correspondence, Cowans to Clarke, 16 Apr. 1918.

⁸⁶ Brown, British Logistics, pp. 194–5.

⁸⁷ Fay, The War Office at War, p. 191.

⁸⁸ Fay, The War Office at War, p. 193.

⁸⁹ Fay, The War Office at War, p. 205.

⁹⁰ Fay, The War Office at War, pp. 193–4.

⁹¹ W. Stewart Wallace, The Macmillan Dictionary of Canadian Biography (3rd edn., Toronto, ON, 1963), p. 85; Fay, The War Office at War, p. 47.

⁹² ‘General’s death closes colourful saga of west’, Vancouver Sun, 24 Sept. 1938, p. 3. A partial account of Stewart’s wartime contributions is given in G. W. Taylor, The Railway Contractors: the Story of John W. Stewart, His Enterprises and Associates (Victoria, BC, 1988), pp. 106–20.

⁹³ Fay, The War Office at War, pp. 202–3.

⁹⁴ Stevenson, With Our Backs to the Wall, p. 67; E. Greenhalgh, Victory through Coalition: Britain and France during the First World War (Cambridge, 2005), p. 198.

⁹⁵ D. T. Zabecki, The German 1918 Offensives: a Case Study in the Operational Level of War (Abingdon, 2006), pp. 85–6; Henniker, History of the Great War, pp. 398–9.

⁹⁶ Pritchard, History of the Royal Engineers, v. 636–37.

⁹⁷ Zabecki, The German 1918 Offensives, pp. 111–12.

⁹⁸ NLS, Haig papers, Acc. 3155/124, diary entry, 26 March 1918. Emphasis in original.

⁹⁹ TNA, WO 107/35, Amiens–Abancourt–Dieppe–Abbeville area: Measures to be taken in the event of the abandonment of, March to June 1918; Henniker, History of the Great War, p. 402; Pritchard, History of the Royal Engineers, v. 646. By 4 April Stewart was responsible for the work of 67 different units – a workforce of 22,400 men.

¹⁰⁰ Quoted in Zabecki, The German 1918 Offensives, p. 86.

¹⁰¹ Quoted in R. Prior and T. Wilson, Command on the Western Front: the Military Career of Sir Henry Rawlinson, 1914–18 (Oxford, 1991), at pp. 280, 282.

¹⁰² Stevenson, With Our Backs to the Wall, pp. 224–5.

¹⁰³ Quoted in M. Pöhlmann, ‘Return to the Somme 1918’, in Scorched Earth: the Germans on the Somme 1914–1918, ed. I. Renz, G. Krumeich and G. Hirschfeld (Barnsley, 2009), pp. 179–201, at p. 192. Emphasis in original.

¹⁰⁴ Boff, Haig’s Enemy, p. 218.

¹⁰⁵ A. Watson, Ring of Steel: Germany and Austria-Hungary at War, 1914–1918 (London, 2015), p. 521.

¹⁰⁶ TNA, CAB 25/111, German possession of Amiens: effect on transportation, Storr to IATC secretary, 2 Apr. 1918.

¹⁰⁷ I. M. Brown, ‘Feeding victory: the logistic imperative behind the hundred days’, in 1918: Defining Victory, ed. P. Dennis and J. Grey (Canberra, 1999), pp. 130–47, at pp. 134–5.

¹⁰⁸ TNA, CAB 25/111, German possession of Amiens, Nash to Storr, 9 Apr. 1918, pp. 1–2.

¹⁰⁹ TNA, CAB 25/111, German possession of Amiens, Nash to Storr, 9 Apr. 1918, p. 6.

¹¹⁰ Stevenson, With Our Backs to the Wall, pp. 71–2; Boff, Haig’s Enemy, pp. 220–2; Watson, Ring of Steel, p. 521. Dunkirk itself was also under shell-fire, ‘and in consequence could be little used’. See ‘G.S.O.’, G.H.Q., p. 255.

¹¹¹ TNA, CAB 25/111, German possession of Amiens, Nash to Storr, 10 Apr. 1918, p. 1.

¹¹² Brown, ‘Feeding victory’, p. 137.

¹¹³ TNA, WO 107/34, programme of railways: memorandum forecast of trains and railheads, 1918. For a detailed account of the wider administrative responsibilities demarcated in scheme Z, see Henniker, History of the Great War, pp. 402–11.

¹¹⁴ L. S. Simpson, ‘Railway operating in France’, Journal of the Institution of Locomotive Engineers, xii (1922), 697–728, at pp. 717–18.

¹¹⁵ Stevenson, With Our Backs to the Wall, p. 72.

¹¹⁶ TNA, CAB 25/111, German possession of Amiens, transport position in France. Memorandum by Major-General Sir P. A. M. Nash, 19 Apr. 1918, p. 1.

¹¹⁷ TNA, CAB 25/111, German possession of Amiens, transport position in France, 19 Apr. 1918.

¹¹⁸ TNA, CAB 25/111, German possession of Amiens, summary of a meeting held at the Ministry of Armaments, Paris, on 23rd Apr. 1918, to consider the requirements for the export of coal from the United Kingdom to France in certain eventualities. Sir Richard Redmayne, the chief inspector of mines and former chair of mining engineering at the University of Birmingham, led the British delegation.

¹¹⁹ TNA, CAB 25/111, German possession of Amiens, Nash to Storr, 13 Apr. 1918; telegram: G.H.Q. to Britcil, 13 Apr. 1918; note on the organisation of supply of the Allied forces operating in the area north of the Somme, 15 Apr. 1918, p. 2; extract from War Cabinet, 395, dated 19th Apr., 1918; Belin to Sackville-West, 24 Apr. 1918.

¹²⁰ Henniker, History of the Great War, p. 400.

¹²¹ TNA, CAB 25/111, German possession of Amiens, Storr to IATC secretary, 2 Apr. 1918.

¹²² TNA, CAB 25/111, German possession of Amiens, Nash to Storr, 9 Apr. 1918, p. 2.

¹²³ TNA, CAB 25/111, German possession of Amiens, Nash to Storr, 27 Apr. 1918, pp. 2–3; Pritchard, History of the Royal Engineers, v. 645.

¹²⁴ Henniker, History of the Great War, pp. 400–2; J. H. F. Le Hénaff and H. Bornecque, Les chemins de fer français et la guerre (Paris, 1922), pp. 237–8; D. Lyell, ‘The work done by railway troops in France during 1914–19’, Minutes of the Proceedings of the Institution of Civil Engineers, ccx (1920), 94–147, at pp. 109–12.

¹²⁵ TNA, WO 95/40, QMG war diary, statistical summary, Oct. 1918, p. 5.

¹²⁶ Further details of the railway construction work completed in the Somme region during this period are given in TNA, CAB 25/111, German possession of Amiens, Le Hénaff to Nash, 18 July 1918; Les armées françaises dans la grande guerre: la direction de l’arrière (Paris, 1937), pp. 669–71.

¹²⁷ Henniker, History of the Great War, pp. 400–1.

¹²⁸ Pritchard, History of the Royal Engineers, v. 645.

¹²⁹ Brown, British Logistics, p. 194; Brown, ‘Feeding victory’, p. 139.

¹³⁰ W. Philpott, Bloody Victory: the Sacrifice on the Somme (London, 2009), pp. 517–18, 520.

¹³¹ Prior and Wilson, Command on the Western Front, pp. 289–90.

¹³² Watson, Ring of Steel, p. 524.

¹³³ J. P. Harris, Amiens to the Armistice: the BEF in the Hundred Days’ Campaign, 8 August–11 November 1918 (London, 1998), p. 73.

¹³⁴ Philpott, Bloody Victory, pp. 520–1; Prior and Wilson, Command on the Western Front, pp. 309–15.

¹³⁵ R. Thompson, ‘“Delivering the goods”. Operation Landovery Castle: a logistical and administrative analysis of Canadian Corps preparations for the battle of Amiens, 8–11 August, 1918’, in Changing War: the British Army, the Hundred Days Campaign and the Birth of the Royal Air Force, 1918, ed. G. Sheffield and P. Gray (London, 2013), pp. 37–54, at p. 40.

¹³⁶ For a selection of accounts, see Harris, Amiens to the Armistice, pp. 116–17; G. Sheffield, Forgotten Victory. The First World War: Myths and Realities (London, 2001), pp. 237–41; C. Messenger, The Day We Won the War: Turning Point at Amiens, 8 August 1918 (London, 2008); N. Lloyd, Hundred Days: the End of the Great War (London, 2013), pp. 54–5.

¹³⁷ J. Boff, ‘Combined arms during the hundred days campaign, August–November 1918’, War in History, xvii (2010), 459–78, at p. 461.

¹³⁸ Boff, Haig’s Enemy, pp. 234–5; Harris, Amiens to the Armistice, pp. 103–4; Prior and Wilson, Command on the Western Front, p. 320.

¹³⁹ W. Philpott, Attrition: Fighting the First World War (London, 2014), pp. 327–8. Emphasis added.

¹⁴⁰ TNA, WO 95/40, QMG war diary, explanatory review, Aug. 1918, p. 1.

¹⁴¹ TNA, WO 95/40, QMG war diary, diary entry, 27 Aug. 1918.

¹⁴² TNA, WO 95/40, QMG war diary, explanatory review, Aug. 1918, p. 1; G. R. Winton, ‘The British army, mechanization and a new transport system, 1900–14’, Journal of the Society for Army Historical Research, lxxviii (2000), 197–212; Henniker, History of the Great War, p. 451.

¹⁴³ Henniker, History of the Great War, pp. 450–1.

¹⁴⁴ Taylor, Railway Contractors, p. 118.

¹⁴⁵ The extent to which Haig and GHQ successfully delegated authority during the hundred days has been subject to sustained examination. For a range of views, see T. Travers, How the War Was Won: Command and Technology in the British Army on the Western Front, 1917–1918 (London, 1992), pp. 175–82; Sheffield, The Chief, pp. 293–339; W. Reid, Douglas Haig: Architect of Victory (Edinburgh, 2006), pp. 449–86; Prior and Wilson, Command on the Western Front, pp. 397–8; J. Boff, Winning and Losing on the Western Front: the British Third Army and the Defeat of Germany in 1918 (Cambridge, 2012), pp. 214–15; A. Simpson, Directing Operations: British Corps Command on the Western Front, 1914–18 (Stroud, 2006), pp. 156–76.

¹⁴⁶ TNA, WO 95/39, QMG war diary, 12 Aug. 1918.

¹⁴⁷ TNA, WO 95/40, QMG war diary, statistical summary, Oct. 1918, p. 4.

¹⁴⁸ Davies, Light Railways of the First World War, pp. 98–102, 104; Boff, Winning and Losing, p. 88.

¹⁴⁹ S. D’A. Crookshank, ‘Transportation with the B.E.F.’, Royal Engineers Journal, xxxii (1920), 193–208, at p. 194.

¹⁵⁰ Unless otherwise stated, all figures in this passage are taken from TNA, WO 95/40, QMG war diary, Statistical summary, Oct. 1918.

¹⁵¹ TNA, WO 95/40, QMG war diary, sketch map illustrating how the railways followed up our advancing troops, Sept. 1918; Henniker, History of the Great War, p. 435.

¹⁵² Henniker, History of the Great War, p. 456.

¹⁵³ Henniker, History of the Great War, pp. 456–7. As Henniker noted, the traffic figures given above only included ammunition trains from the main ammunition depots. Additional traffic from the advanced depots in army areas often demanded the running of as many as 12 additional trains.

¹⁵⁴ Boff, Winning and Losing, pp. 31–2, 89–90.

¹⁵⁵ See TNA, WO 95/40, QMG war diary, sketch map, Sept. 1918. The distances between the most easterly railheads available to divisions on 8 Oct.lay on aggregate far closer to those available on 8 Sept. than those in use on 8 Aug.

¹⁵⁶ Fay, The War Office at War, p. 213; TNA, WO 95/40, QMG war diary, diary entry, 5 Oct. 1918. Fay’s concern was wholly justified. Between Aug. and Oct. 1918 the BEF’s railway construction troops laid 485 miles of new track. See Pritchard, History of the Royal Engineers, v. 659.

¹⁵⁷ J. C. Harding-Newman, Modern Military Administration, Organization and Trans-portation (Aldershot, 1933), p. 23.

¹⁵⁸ TNA, WO 95/40, QMG war diary entries, 19 and 26 Oct. 1918.

¹⁵⁹ J. Boff, ‘Logistics during the hundred days campaign, 1918: British Third Army’, Journal of the Society for Army Historical Research, lxxxxiv (2011), 306–21, at p. 320. All classes of supply were affected by congestion on the railways in late Oct. The regular daily supply trains for 61st Division arrived a day late on four occasions, while the deliveries for 16 and 19 Oct. did not get through at all.

¹⁶⁰ TNA, WO 95/40, QMG war diary, statistical summary, Oct. 1918, pp. 4, 7.

¹⁶¹ Boff, ‘Logistics during the hundred days campaign’, p. 314.

¹⁶² Henniker, History of the Great War, pp. 460–61.

¹⁶³ TNA, WO 95/454, headquarters branches and services. Deputy Director Supplies and Transport, the Fourth Army, notes on conference, 4 Oct. 1918; WO 95/40, QMG war diary, Explanatory review, Oct. 1918, p. 3; Henniker, History of the Great War, p. 461.

¹⁶⁴ TNA, WO 95/40, QMG war diary, explanatory review, Oct. 1918, p. 3.

¹⁶⁵ Boff, Winning and Losing, pp. 86–7.

¹⁶⁶ Stevenson, With Our Backs to the Wall, pp. 168–9.

¹⁶⁷ Boff, Winning and Losing, p. 36; Harris, Amiens to the Armistice, p. 295.

¹⁶⁸ TNA, WO 95/40, QMG war diary, diary entries, 2 and 9 Nov. 1918.

¹⁶⁹ Fay called upon Maybury on 27 Dec., and recorded in his diary that the latter was ‘not at all well—has nerves and cannot sleep’. See Fay, The War Office at War, p. 208.

¹⁷⁰ TNA, WO 95/454 Deputy Director Supplies and Transport, the Fourth Army, diary entries, 31 Oct. and 10 Nov. 1918; Crookshank, ‘Transportation with the B.E.F.’, p. 206.

¹⁷¹ Boff, Winning and Losing, p. 35.

¹⁷² A. Montgomery, The Story of the Fourth Army in the Battle of the Hundred Days, August 8th to November 11th, 1918 (London, 1920), pp. 260–1; J. E. Edmonds and R. Maxwell-Hyslop, History of the Great War. Military Operations, France and Belgium, 1918 (5 vols., London, 1947), v. 528.

¹⁷³ J. C. Darling, 20th Hussars in the Great War (Lyndhurst, 1923), p. 127.

¹⁷⁴ Edmonds and Maxwell-Hyslop, History of the Great War, v. 533.

¹⁷⁵ TNA, WO 95/40 QMG war diary, explanatory review – Nov. 1918, p. 2.

¹⁷⁶ Henniker, History of the Great War, p. 467.

¹⁷⁷ Brown, British Logistics, p. 179.

¹⁷⁸ M. Van Creveld, ‘World War I and the revolution in logistics’, in Great War, Total War: Combat and Mobilization on the Western Front, 1914–18, ed. R. Chickering and S. Förster (Cambridge, 2005), pp. 57–72, at p. 67.

¹⁷⁹ W. N. Nicholson, Behind the Lines: an Account of Administrative Staffwork in the British Army, 1914–18 (London, 1939), p. 215.