The attack carried out by the 109th Brigade has been officially described as a "bombing action," and such, doubtless, to a great extent it was. But there was no fear more constantly present with General Nugent and General Ricardo than that of its developing into the conventional bombing action, which progressed at snail's pace at best, at worst reached an early deadlock, and always required bombs by the dozen for every yard of ground gained. In this case four thousand yards, covered by two or three parallel lines of trenches, with numerous communication trenches at right angles to the attack, had to be taken. And speed was essential; for, much as the attack depended upon the advance east of the Canal, that in its turn could have been taken in flank by machine-gunners on the west had not the 109th Brigade kept pace with it. And so the order was--no bombing till other methods failed. At the head of the platoon columns to move up the trenches was to be not a bomber, but a Lewis gunner. A Lewis gun could be used by a man of large physique--and of these the Ulster Division still possessed plenty, if not such numbers as a year back--with a sling over the left shoulder, the gun resting above the right hip. It was heavy and clumsy, but its tremendous moral effect in such broad trenches as those of the Hindenburg System can readily be imagined. It was to be the spear-head of the attack along each line of trenches. General Ricardo had learnt the idea from the Canadians, who had employed it at Vimy. Moreover, when it was found possible, riflemen were to move on top, outside the trench. The question of artillery support was not easy, since no preliminary registration, here, or on any other part of the front of attack, was possible. Colonel Simpson, who took up his position with General Ricardo at a command post in the sunken Demicourt-Havrincourt Road, just short of the British front line, planned and controlled the artillery support in most brilliant fashion, well worthy the fame he was winning as one of the most scientific and least conventional junior artillery commanders in the British Army. It must not be supposed that the part played here, or on the whole front, by artillery was negligible, nor that the barrage was inconsiderable. On the contrary, the artillery support for the main attack was of vast weight. In Havrincourt Wood, along the rides, guns stood almost wheel to wheel. But these guns had not registered. Consequently the barrage might be expected to be somewhat ragged, and was to keep considerably further in front of the tanks and infantry than would otherwise have been the case. Still, a barrage does not cut wire. It was the tanks which were to accomplish this. Had they failed, the whole scheme would have collapsed. During the night of November the 19th, the whole line of tanks on the front of the III. and IV. Corps moved to a general distance of a thousand yards from the German outpost line.