The majority of British underground accommodation up to and during 1916 was of shallow cut and cover rather than deep tunnelled construction.¹⁶ Where the British took over trenches from the French on the Somme in the autumn of 1915 they found very deep dugouts cut into the chalk. However, anticipating the German change in policy a year later, the British regarded them as highly vulnerable where they were close to the Germans. Those seen by the commander of the Tyneside Scottish Brigade at La Boisselle were up to 30ft deep and held as many as fifty men:

There were numerous deep dugouts in our front trench, but it would have been impossible for the men to get out of them in time to repel a surprise raid; in fact, these dugouts would have proved mere traps, so their use was prohibited, and the entrances were closed up by a network of barbed wire. The deep dugouts in the supports and reserve trenches were, however, found most useful not only for the shelter of the garrisons of the trenches, but for use as advanced dressing stations, company and Battalion Headquarters, and signallers’ dugouts, and for R.E. and ammunition dumps. (Brigadier General Trevor Ternan, commanding 102nd Infantry Brigade).¹⁷

The Somme saw a major expansion of the use of such underground accommodation behind the front line to support attacks. For the assault on La Boisselle a significant portion of the Tyneside Irish Brigade was housed in a series of around forty-five dugouts in the reverse slopes of the Tara and Usna hills 1000yds (900m) behind the British front line either side of the Albert–Bapaume road. These were at about 20ft depth and linked by underground passages, although the labour available from miners in the Tyneside units made them the exception rather than the rule.¹⁸

Following the Somme, the British used extensive dugouts for both headquarters and troop accommodation for the 1917 offensives at Vimy, Arras, Messines and Ypres.

As the work of offensive mining declined, many tunnelling companies became wholly employed on this work in the run up to attacks. For the Messines operation they built twenty-four shell-proof battle headquarters for brigade staffs and twenty-eight battalion headquarters and dugout shelters for 200 officers and 10,000 men. For the Third Battle of Ypres the tunnellers constructed twenty-three brigade headquarters, sixty-two battalion headquarters, seven artillery, seven dressing stations, twenty-three observation posts and fourteen tunnelled dugout systems incorporating troop accommodation.¹⁹ Such a system beneath the ruins of the village of Wieltje, which housed the headquarters of the 107th Brigade in August 1917, was recalled by a staff officer, Captain Cyril Falls:

Wieltje dug-outs! Who that saw it will forget that abominable mine, with its ‘town major,’ its thirteen entrances, the water that flowed down its main passages and poured down its walls, its electric light gleaming dully through steam-coated lamps, its sickly atmosphere, its smells, its huge population of men – and of rats? From behind sack-curtained doorways the coughing and groaning of men in uneasy slumber mingled with the click of typewriters. In the corridor one would fall over a runner, slimy from head to foot with mud, resting while he waited for a return message to the front line. One advantage only it had: it was safe within.²⁰

Communication tunnels were used immediately behind the front line to provide safe passage and many were dug in the Ypres Salient by both sides, although the wet sand made the construction of shallow inclined entrances difficult: in most places they were almost vertical. For the Messines operation these were constructed in four places to a total of 15,000ft length.²¹ Where the geology was more favourable in the chalk areas they were used even more extensively and the attack on Vimy Ridge on 9 April 1917 benefited more from such tunnels than from mines. The letter sent on 5 December by GHQ to Army commanders concerning the use of communication tunnels for the final attacks on the Somme was received as 1st and 3rd Armies were preparing for the Arras and Vimy Ridge attacks. Such tunnels were to be a major feature of the battles, but in many ways were quite different from the Russian saps used on the Somme. They were termed ‘subways’ and were not intended to extend far across no man’s land. Rather they were used to enable the infantry to get to their assembly trenches safely and secretly from the support trenches or from farther back. In particular the exposed Zouave Valley west of Vimy Ridge was identified as a vulnerable area as the Germans laid a barrage into the valley at the first suggestion of an attack. Six of the subways were tunnelled into the eastern slope of the valley, from 300 to 1,537yds. Mulqueen’s Company was responsible for four which, he recalled, were commenced before the arrival of the Canadians at Vimy Ridge. Blue Bull, Vincent, Cavalier and Tottenham Road were:

...named after the nearby communication trenches and a great deal of planning with the Division and Brigades preceded their construction. The 9th (Scottish) Division was in the line when Tottenham Road was started and it is probable that its location was fixed with an eye to its use if and when an attack were launched against the Ridge. Certainly before this and the other three tunnels were completed, their importance in the proposed attack became paramount.

Two of the longest were dug by Mulqueen’s former Company, 172, on what was to become the 3rd Canadian Division front: the Goodman was 1,883yds and the Grange 1,343yds (of which part is open to the public in the Vimy Memorial Park). Twelve subways in all were used for the Canadian Corps attack, measuring in total six miles. They were constructed 6ft 6in in height and were wide enough for two laden men to pass. Most had electric lighting and some were equipped with tramways used especially for bringing forward ammunition for connected heavy mortar positions. The subways were also used to provide accommodation for battalion and brigade headquarters, cookhouses and latrines. They had their own water supply, piped from two large underground reservoirs each holding 50,000 gallons, and electrical power from generators. They also provided a secure conduit for signals cables and housed signals stations. The dressing stations in the subways were 34ft by 9ft and were dry and well-lit.

Very careful attention was paid to traffic control in the light of the experiences on the Somme when the tunnels were so congested by wounded as to be unusable. The 4th Division scheme allocated military police to each subway with an officer at the eastern end and a sergeant at the western end and a telephone to keep them in communication. A plan of the subways was placed at each entrance and mid-way along, with direction boards at each junction. At the western entrance a large signboard guided troops coming across the Zouave Valley to the subway so that they could go directly to it. The officer had complete control over the tunnel traffic from nine hours before the attack. A system of traffic lights was used: when the assaulting troops were moving forward to the assembly positions only up traffic was permitted, the only exceptions being individual officers, medical personnel and runners. A red light at the tunnel entrance indicated that it was to be used for up traffic only and military police at each entrance enforced this. Thus the tunnel would be kept clear for the successive waves of the attack troops to get forward rapidly. Through one subway alone it was estimated that the traffic amounted to 9,700 men between 5 and 11 April.²²

The ends of some of the subways were used to create bored trenches using the Wombat borer. In five of the subways blown trenches were prepared using the borer, 6in diameter bores of about 200ft charged with ammonal cartridges. Three of these were actually blown at thirty seconds after zero, the longest being 195ft long and 14ft deep at the Chassery craters, where the 2nd Canadian Mounted Rifles assaulted, although it is unlikely that these blown trenches were used owing to the rapid success of the attack. The troops did not depart directly to attack from the subways, but the first waves used them to reach the assembly trenches about two hours before zero. The succeeding waves then used them where necessary. The capture of the ridge was a dramatic success for the Canadian Corps and the British Tunnelling Companies. An officer of 185 Company, whose unit had dug the Douai, Bentata and Zivy subways, wrote home enthusiastically to his father four days after the attack:

Our work had been intimately connected with the early stages of the advance, and it is satisfactory to know from G.H.Q. that its purpose (the subways) was successful. We were attached to the Canadians and all the officers with whom we came in contact were enthusiastic about the results, and undoubtedly we were responsible for saving innumerable casualties in the initial stages and I dare say more will be heard of it later on.