I
We have said that in the original battle scheme, certain points of vantage, Quennemont, the Knoll, and Bellicourt, were assumed to be in our hands a day or so before the main attack on the 4th Army front was launched.
These fortified heights were of importance owing to the singular geography of this sector of the line.
All along this piece of the front, more or less parallel to the lines of the armies, runs—deep and broad—the St. Quentin Canal.
For three and a half miles, however, between Bellicourt and Vendhuille it runs underground through a tunnel.
We have seen how, in the northern part of the line, the enemy had relied upon the Canal du Nord to form the principal obstacle to an attack.
In August we had captured a document which proved that he realised that if we attacked at all in the south, and whether we attacked with Tanks or not, it would be in that three-and-a-half-mile gap that our heaviest blow would fall.
The photograph gives an excellent notion why we had to avoid certain sectors of the Canal at all costs, and Sir Douglas Haig, in his Despatch, gives an admirable idea of some of the complex features which the topography here possessed.
“The general configuration of the ground through which this sector of the Canal runs, produces deep cuttings of a depth in places of some sixty feet; while between Bellicourt and the neighbourhood of Vendhuille the Canal passes through a tunnel for a distance of 6000 yards. In the sides of the cuttings the enemy had constructed numerous tunnelled dug-outs and concrete shelters. Along the top edge of them he had concealed well-sited concrete or armoured machine-gun emplacements. The tunnel itself was used to provide living accommodation for troops, and was connected by shafts with the trenches above. South of Bellicourt the Canal cutting gradually becomes shallow, till at Bellenglise the Canal lies almost at ground level. South of Bellenglise the Canal is dry.
“On the western side of the Canal, south of Bellicourt, two thoroughly organised and extremely heavily wired lines of continuous trench run roughly parallel to the Canal, at average distances from it of 2000 and 1000 yards respectively. The whole series of defences, with the numerous defended villages contained in it, formed a belt of country varying from 7000 to 10,000 yards in depth, organised by the employment of every available means into a most powerful system, well meriting the great reputation attached to it.”
On the three and a half miles of front, where alone Tanks and artillery could cross the line of the Canal, the outpost system which everywhere protected the Hindenburg Line, was doubly reinforced, and gained a natural strength from its position on the heights, beneath which the Canal had burrowed.
Only a very “full dress” attack on so highly organised a system as the Hindenburg Line was likely to be successful, and in order to launch such an attack it was essential that we should already hold the Knoll and Guillemont and Quennemont Farms.
We have seen how in the last day or two of the battle of Epehy we assaulted the line again and again, duly captured the sector opposite Bellicourt, but how, two days before the main attack was to be launched, the Knoll and Quennemont were still in the hands of the enemy.
This state of affairs caused grave anxiety, as the whole set-piece attack was based on the idea of using this line as a “jumping-off” position.
It had been intended that the two American Divisions, which were to fight on this sector, should only be put in when this line had been secured.
It was now decided that they must themselves make a final effort to capture the outpost line before the main assault, which was due for dawn on September 29.
Therefore, at dawn on the 27th, the 27th American Division, assisted by twelve Tanks of the 4th Battalion, again attacked under cover of a creeping barrage.
[93]“The attack met with strong opposition, and the final position reached was the subject of conflicting reports from the troops engaged and from the air observers. Subsequent events showed that small parties of Americans and Tanks had reached the vicinity of their objective, and had very gallantly maintained themselves there; but the line as a whole was not materially advanced by the day’s operations.... The barrage could not now be brought back on this flank owing to the knowledge that parties of American troops, as well as a number of American wounded, would be exposed to our own fire. Also any alteration in the barrage plans, which had already been issued, would inevitably lead to confusion.”
Either, therefore, the whole main attack must be delayed, or the American divisions and some of the British troops north of them must start some 1000 yards behind their barrage, and from a very indefinite jumping-off line.
The latter course was decided upon.
[94]“The artillery start line, as originally planned, was to hold good, and the troops of the 27th American Division would form up for the attack on a line as far forward as possible, and would be assisted by an additional number of Tanks. The strength in Tanks was augmented to such an extent as should easily overwhelm the enemy resistance west of the start line. It was thought that this, with the slow rate of barrage, would enable the Americans to carry out their task.”
But there was yet one more difficulty, a serious obstacle of which we were serenely unaware. A British anti-Tank minefield, consisting of rows of buried heavy trench-mortar bombs, each holding 50 lb. of ammonal, had been put down just prior to our loss of the area in March 1918, and of this minefield no information had reached the Tanks.
It will thus be seen that the dice were very heavily loaded against success on this part of the front before day dawned on the eventful 29th of September, 1918.
The whole attack was to be on a twelve-mile front. The infantry were to take advantage of a number of foot-bridges, which our bombardment had prevented the enemy from getting out to destroy, and in some places our men were prepared to wade or swim through the water.
It was expected, however, that the chief resistance would be offered on the famous three and a half miles.
Altogether about 175 Tanks, including the new American Battalion, were to be launched, and four Corps were to be involved.
To the 9th Corps on the right, the 5th, 6th and 7th Tank Battalions of the 3rd Brigade were allotted.
In the centre, with the Australian and American Corps, the 1st, 4th and 301st American Battalions of the 4th Brigade were to fight.[95]
The 8th, 13th and 16th Battalions of the 5th Tank Brigade were to be held in 4th Army Reserve.
Almost up to zero hour on the 29th we still hoped to get news that we held the Knoll and Quennemont. But no reassuring message came through.
It was thus in a very singular world that the 301th American Tank Battalion was destined to make its debut.
[96]“The 301st’s reconnaissance before the battle was very efficiently carried out in spite of many disadvantages. The taping especially was a classic example of pluck and efficiency. It must be borne in mind that this was no quiet front, and that the attempts to take his outpost line had made the Boche exceedingly nervous and alert. In consequence, the nights preceding the battle were some of the dirtiest I’ve experienced. The Battalion R.O. (I’ve forgotten his name), one Company R.O. (Lieutenant T. C. Naedale) and a sergeant were knocked out whilst supervising the taping. Lieutenant Naedale got his wounds dressed and continued his work up till zero hour. It is worthy of mention, in connection with this incident, that each American Tank had its own tape laid out over our front line towards the Boche by the Company R.O.’s. Tank Commanders told me afterwards that they had to start fighting before the end of their tape was reached.”