Chapter XVII

HAMEL, MOREUIL, AND THE THIRD BATTLE OF THE SOMME.

I.

On the night of the 22nd-23rd June, a night raid of peculiar interest was carried out against the village of Bucquoy, north of the Ancre. Five platoons of infantry and five female tanks of the 10th Battalion were employed. The notable features were the use of tanks at night, and the remarkable skill and success with which they were hidden for a fortnight within a few hundred yards of the German outposts. One section of four tanks was camouflaged right in the open, the machines side by side and covered by their nets, in a slight depression not half a mile from the village. Another four tanks were driven into old gun-pits by the side of the Bucquoy-Hennescamp road, even nearer to the enemy. Here the two sections remained for nearly fourteen days, waiting for a favourable opportunity with the full moon. Various causes postponed the enterprise, but at length the conditions were considered propitious, and the five tanks required moved yet farther forward after dark into a sunken road skirting the eastern gardens of Bucquoy, of whose main street one side was held by our outposts and the other by those of the enemy. But now the moon failed, and again the raid was cancelled. Those responsible for it do not seem even to have thought of the danger of leaving tanks indefinitely on the edge of No Man’s Land. For two more days and nights the five machines actually remained in this preposterous position without being discovered. Movement by day was almost impossible, and extreme care had to be exercised at night. Finally, on the night of the 22nd, the raid was carried out. The infantry, after all this hesitation and consulting of omens, were held up almost at once by trench-mortar and machine-gun fire, and although reinforced, got no farther; but the five tanks pushed on right through the village, passed their objectives, killed a number of the enemy and started a serious panic, and returned undamaged. The darkness gave them little trouble, and, on the other hand, afforded them valuable protection from the trench mortars. Their experience, and the careful preliminary work carried out by the Reconnaissance officers, were of great service to the 7th Battalion when we attacked over the same ground two months later.

A more important action was fought on 4th July on the old Villers-Bretonneux battle-ground. It was the first definite offensive on our part since the German blow in March, and it also introduced the Mark V. tank. This machine, identical in shape and armament with the Mark IV. (except that the Hotchkiss light gun replaced the Lewis), possessed great mechanical improvements, primarily in the adoption of epicyclic gears, which enabled one man to drive and steer the tank, and abolished the dilatory business of gearsmen and secondary levers. Three men, in fact, were released for other work, while the tank’s power of manœuvre was greatly improved. Fitted with a 180 h.-p. Ricardo engine, in place of the Daimler, the Mark V. had increased speed, five to six miles an hour, and could be turned as easily as the Whippet. Other modifications included a second cab for observation on the roof, where the manhole was situated in the earlier types. The only serious fault was bad ventilation, which, with the more powerful engine, raised the heat inside to a pitch that at times became unbearable.

The object of the attack on 4th July was the recapture “as a necessary preliminary to any operation to disengage Amiens,”[32] of the spur which runs northward from the Villers-Bretonneux plateau toward the junction of the Somme and the Ancre at Corbie. The objectives included Hamel village and wood and Vaire Wood, and were strictly limited. The infantry to be employed was the 4th Australian Division, and a subsidiary object of the operation was to restore the confidence of the Colonials in tanks--a confidence which had been quite unnecessarily shaken by the failure at Bullecourt in the early spring of 1917, when, as a matter of fact, it was the tanks that had a grievance. That there should be no failure through lack of means on this occasion, the whole of the 8th Tank Battalion and two companies of the 13th (sixty machines in all) were allotted to the Australians. As soon as the attack was decided upon, training with the infantry was put in hand at Vaux-en-Amienois, the 5th Tank Brigade Headquarters, and a good mutual understanding was cultivated. The Australians, clinging still to old prejudices, insisted on the tanks following the first wave of attack instead of preceding it; but this adherence to a bad system did not in effect matter at all, as the Mark V.’s took the lead almost at once on the day of action, and remained ahead of the infantry throughout.

Zero was at 3.10, just after dawn. The final rather intricate approach-marches of the five tank companies, from Hamelet and Fouilloy to the starting-points behind the infantry, were covered by the noise of low-flying aeroplanes. At eight minutes before zero the tanks moved forward again, crossed the remaining 1000 yards in twelve minutes, and at zero, plus four minutes, the barrage lifted and they advanced with the infantry. The attack was a complete success. In an hour and a half the whole affair was over: Hamel village and the woods were carried, an advance of 2500 yards on a front of 7000 had been made, and 1500 prisoners, with 2 field-guns and 171 machine-guns, were in our hands. Our casualties were under 700, most of them walking wounded, and 5 tanks hit and disabled; 57 tanks reached their final objectives, and 55 had rallied by 11 A.M. All the disabled machines were salved during the two following nights, some of them being towed 9000 yards, so that the enemy was without any knowledge of the new type beyond what he had discovered in action; and this must have been the reverse of encouraging. The increased manœuvring powers of the Mark V. astonished the Germans as much as they delighted the Australians. A feature of the action was the number of machine-gun emplacements overrun and flattened out by tanks at the request of the infantry--an entirely satisfactory process of elimination. The German machine-gunners fought in many cases with extreme bravery, continuing to fire until the very moment before their guns--and perhaps themselves--were crushed into the earth by the tanks; but the unexpected rapidity of movement shown by the latter, as compared with the deliberate actions of the Mark IV., continually took them by surprise. The results of the attack, in short, were that the Australians were satisfied with the tanks, that the tank crews were satisfied with the Australians, which was quite as important, and that a sure foundation had been laid for the greater partnership of the 8th of August.

A fortnight later the 9th Tank Battalion, temporarily attached to the 5th Brigade for the purpose, took part at very short notice in an operation south of Amiens with the French 9th Corps. The third great German blow, between Soissons and Rheims, had in its turn exhausted its momentum; but to all the world it seemed still in the full tide of victory when, on 18th July, with a rapidity founded on exact calculation, Foch’s counter-attack, led by hundreds of Rénault tanks, fell on the flank of the new Salient from the Aisne to Château-Thierry, and the last phase of the war had begun. The second battle of the Marne, or the battle of Soissons, whichever it is called, was the first of that series of Allied offensives which for the next four months were to succeed one another without a pause up and down the whole immense front from the Woevre to the sea. Five days after it began, Debeney’s 1st French Army carried out a small attack north of Montdidier, with the double object of occupying the German reserves in that region and of capturing the heights which overlook the river Avre at Aubvillers and Mailly-Raineval. It was for this operation that the 9th Tank Battalion was lent. It proved a complete success, although the casualties in tanks and personnel were rather heavy, and it was of interest chiefly on two accounts: firstly, the excellent way in which the French infantry combined with the tanks, although only two or three days of sketchy training, handicapped by difficulties of language, were possible beforehand; and secondly, for the speed with which the tanks’ share in the battle was arranged and carried out. The project was discussed for the first time on 16th July at the 4th Army Headquarters. An hour and a half later, General Elles and General Courage (commanding the 5th Brigade) interviewed the French Army commander, General Debeney, and arranged with his Staff the general outlines of the move and the action. The next day the 9th Tank Battalion, far away near Bus-les-Artois, received orders to trek on the 18th to Rosel, a distance of nearly ten miles across country. Entraining at Rosel, it detrained at Contay on the 19th, and that evening trekked another two and a half miles. During the three following nights a farther distance of twelve miles was covered in stages, the final lap, on the night of the 22nd-23rd, bringing the battalion into action with thirty-five machines out of forty-two, most of the absentees being old Mark IV. supply-tanks. During this period troops of the 3rd French Division, which was to supply the infantry for the attack, underwent a brief course of training with tanks at Vaux-en-Amienois, while the Reconnaissance Officers of the 9th Battalion hurriedly surveyed the battlefield.

After the action, in which 15 tanks were disabled by artillery fire and 54 officers and men were killed or wounded, General Debeney visited the battalion and expressed in very warm terms his appreciation of its services. The 3rd French Division conferred upon it a rare distinction in the form of its own divisional badge, which every officer and man of the 9th Battalion now wears on the left sleeve.[33]

II.

Even before the French counterstroke of 18th July, the first plans for an attack by the restored 4th British Army south of the Somme were under discussion. Limited originally to the recapture of the old Amiens defence line, the scheme expanded until it assumed the dimensions of a full-dress battle--involving at least a dozen infantry divisions, a cavalry corps, and four brigades of tanks on our part, and, on our right, the 1st French Army and part of the 3rd. Our objective, roughly, was the important railway which ran southward from Peronne through Chaulnes and Roye, which meant an advance in places of from twelve to fifteen miles.

Notwithstanding the reinforcements which had been poured into France in the past two months, so ambitious a programme, to be completed within three or four days, could mean only one thing--another tank operation on a large scale, or, in other words, a second Cambrai. Tanks were the only arm which could hope to produce such a result in so short a time. And yet, if we hunt through the official account of the battle and its antecedents, we find the determining factor hardly mentioned. We read of “the brilliant and predominating part taken by the Canadian and Australian Corps in this battle....” And again, of “the fine performance of the cavalry, ... the dash and vigour of their action....” Far away, at the very end of a despatch which takes us to the Armistice, we find this significant sentence, which seems to have slipped in by mistake: “_The whole scheme of the attack of the 8th August was dependent upon tanks...._”[34]

The italics are my own, and I will not labour again, in face of this belated confession, over the old arguments and manifest facts which I have reiterated so often already. I will add only one remark, that the very fact of the omission by name of the Tank Corps in accounts of all the attacks which it led and determined from 8th August to the end of the war, is a tacit (and, I am sure, quite unintentional) admission of its influence. It had become, although no one would confess it openly, so essential a factor that its recognition was reserved for those concluding paragraphs in a despatch which praise conventionally every branch of the service, from the Artillery and Engineers to R.T.O.’s and Sanitary Squads.

Zero day, fixed in the first place for 10th August, was put forward two days at the end of July. The Canadian Corps, borrowed from the 1st Army, and in the best possible condition after the first genuine rest it had known since its original contingents landed in France, was on the right of our line, from a point opposite Moreuil to above Hangard. With it was the 4th Tank Brigade, consisting of the 1st, 4th, 5th, and 14th Battalions, the latter newly arrived from England. On the left of the Canadians, extending past Villers-Bretonneux to the Somme, was the Australian Corps, with its old friend the 5th Tank Brigade, now made up of the 2nd, 8th, 13th, and 15th Battalions, the last being also a new arrival. On the left of the Australians, again, was the 3rd Corps, to which was attached the 10th Tank Battalion by itself. The 9th Battalion was still refitting after its action with the French, and would not be available for several days. This made a total of nine tank battalions, with a tenth in reserve, all except two being equipped with Mark V. machines. The two exceptions, the 1st and 15th, had for their sins a still later novelty known as the Mark V. One Star. This was a thoroughly bad invention, being merely a Mark V. lengthened by about six feet, and designed to carry infantry or machine-gunners as well as its crew, these passengers being deposited at suitable points in the enemy’s country. It offered an enormous target, and if its supernumeraries survived the journey, and were not shot in getting out, they were usually incapable, through sickness or coma brought on by their prolonged confinement inside, of doing any useful work.

In addition to this total, there remained the 3rd and 6th Battalions of the 3rd Tank Brigade. Both these units were now equipped with Whippets, and were attached to the Cavalry Corps, which was charged with the usual magnificent schemes of exploitation, as though machine-guns and magazine rifles had never been heard of. There was also the 17th Battalion of Armoured Cars, attached to the 5th Tank Brigade, but intended to work with the Whippets and cavalry.

It will be seen that for the purposes of this operation the 2nd Brigade had ceased temporarily to exist, its units being distributed among the others. The 1st Brigade was still absent from the main theatre of war. Its three battalions were at Merlimont, practising gunnery, hoping to draw Mark V.’s, and preparing for the next offensive between Arras and the Ancre, in which they were to have their fill of fighting.

The number of tanks to be employed on the 8th August was 435, including 96 Whippets, but excluding the armoured cars. Again, as at Cambrai, there was virtually no reserve of tanks. On the second and subsequent days composite detachments had to be organised in the field. The country immediately in front was good, and still covered in part with growing corn, but east of the line from Roye to Frise, on the Somme, if they ever got so far, was the old shelled area. This, however, was easily negotiable by heavy tanks in dry weather. The preliminary arrangements were modelled on those of Cambrai--profound secrecy, reconnaissance carried out with every possible precaution, and no artillery preparation or activity beyond the normal until the barrage fell at zero. The Australians had been training with tanks at Vaux-en-Amienois, and the Canadians, after their work with the 1st Brigade at Bouvigny and Enguinegatte, at Cavillon with the 9th Battalion.

III.

Considerable anxiety was caused by the capture, on the night of 4th August, of an Australian outpost near the Roye road, as not only would this reveal to the enemy the extension of the Australian front over ground hitherto occupied by the French, but there also was a possibility of damaging information being extracted from the prisoners. The latter, however, refused to disclose anything beyond their names and units, their demeanour being held up as an example of soldierly behaviour in the German reports on their examinations. On the afternoon of the 7th, Y-day, another alarm occurred. Eighteen supply-tanks, allotted to the 4th and 5th Australian Divisions, were hidden in a small plantation north of Villers-Bretonneux. A chance shell burst in the middle of these machines, which were fully loaded with food, petrol, and ammunition; a fire was started, which drew upon the plantation a heavy bombardment; and 15 of the tanks, with all their stores, were destroyed. This calamity led naturally to a suspicion that the tanks had been discovered, but happily it proved unfounded. The Germans believed that only an ordinary dump had been exploded. The coming attack remained entirely unsuspected: the enemy, indeed was so misled by various artifices behind the 1st and 2nd Army fronts, including an early and conspicuous concentration of tanks near St Pol, that he made every preparation against a counter-offensive in Flanders and on the Lys.

During the night of the 7-8th, a heavy ground mist rose over the Somme battlefield, and became so dense by daybreak that tanks and infantry completed their assemblage and moved into action by compass. It aided the surprise, however, and was attributed in part by the discomfited Germans to a non-existent device which they described as an “artificial fog.” The only artificial element was the usual smoke barrage included in the artillery programme, which no doubt helped to increase the early obscurity and confusion. When, at 4.20 A.M. on the 8th, there appeared suddenly out of this veil more than 330 heavy tanks, followed closely by lines of skirmishers and then by the main waves of assault, the German front for twelve miles, from the Somme to Moreuil, was rent like a cobweb. A four-minute deluge of shells was the only warning, for, as the barrage lifted, the tanks and infantry were close at its heels. The advance went with a greater swing even than that of Cambrai, for the two Colonial Corps were neither tired nor weak in numbers, and the Mark V.’s were swifter and more handy than the leaders in the earlier battle. Von Hutier’s 18th Army, composed in part of rather indifferent troops, made no fight at all during the morning. So rapid was the advance that within two or three hours the majority of the German field batteries were overrun and captured; the Albert-Montdidier road, beyond Marcelcave and Demuin, was crossed before noon; officers and men were made prisoners in their messes and billets; and one entire divisional staff was taken. While the heavy tanks were crushing such sporadic resistance as was offered during this first phase by isolated posts and villages (Marcelcave, it may be noted, was captured by a tank single-handed, whose commander obtained a receipt for it from the Australian infantry), the armoured cars had rushed ahead down the two straight Roman roads which diverge from Amiens upon Noyon and St Quentin. Mark V.’s hauled the cars over No Man’s Land and cleared away some trees and other obstacles on the roads immediately beyond, after which the machines had a clear run before them. As they were a unit of the Tank Corps and manned by tank personnel, their exceptional opportunities and exploits deserve notice. General Sir John Monash, commanding the Australian Corps, gives the following account of the cars’ experiences on the St Quentin road:--

“It has already been foreshadowed that the experiences on that
day of the contingent of fifteen armoured motor-cars under
Lieut.-Colonel Carter would form sensational reading, and the
story of 8th August would not be complete without at least a
brief reference to their exploits.

“It was nearly midnight when Carter, with a staff officer, got
back to Corps headquarters to render their report. They were
scarcely recognisable, covered as they were from head to foot
with grime and grease. They had had a busy time. The substance
of what they had to tell was taken down at the time almost
verbatim, and reads as follows:--

“‘Got armoured cars through to Warfusée-Abancourt. When we
reached the other side of No Man’s Land we found that the road
was good, but a number of trees (large and small) had been shot
down, and lay right across it in places. Obstacles removed by
chopping up the smaller trees and hauling off the big ones by
means of a tank. Pioneers helped us to clear the road all the
way down. We did not come up to our advancing troops until they
were almost near the Red Line.

“‘When we got to Blue Line we detached three sections to run
down to Framerville. When they got there they found all the
Boche horse-transport and many lorries drawn up in the main
road ready to move off. Head of column tried to bolt in one
direction and other vehicles in another. Complete confusion.
Our men killed the lot (using 3000 rounds), and left them
there; four staff officers on horseback shot also. The cars
then ran down to the east side of Harbonnières, on the S.E.
road to Vauvillers, and met there a number of steam waggons;
fired into their boilers, causing an impassable block....

“‘Two other cars went to Proyart and found a lot of troops
billeted there, having lunch in the houses. Our cars shot
through the windows into the houses, killing quite a lot of
the enemy. Another section went towards Chuignolles and found
it full of German soldiers. Our cars shot them. Found rest
billets and old trenches, also with troops in them. Engaged
them. Had quite a battle there. Extent of damage not known, but
considerable. Cars then came back to main road. We were then
well in advance of Blue Line.’

“The consternation and disorganisation caused by the sudden
onslaught of these cars, at places fully ten miles behind
the enemy’s front line of that morning, may be left to the
imagination. It was a feat of daring and resolute performance
which deserves to be remembered.”[35]

The Whippets were greatly handicapped by being attached to the Cavalry Corps. The latter, moving rapidly forward in the early morning toward the line, complained that the tanks could not keep up. As soon as it came to fighting, however, the situation was reversed: it was the cavalry that could not keep up with the Whippets, or indeed do anything at all. “By noon on the 8th August, great confusion was developing behind the enemy’s lines, and by this time the Whippets should have been operating five to ten miles in advance of the infantry, accentuating this demoralisation. As it was, being tied down to support the cavalry, they were a long way behind the infantry advance, the reason being that, as cavalry cannot make themselves invisible on the battlefield by throwing themselves flat on the ground as infantry can, they had to retire either to a flank or to the rear to avoid being exterminated by machine-gun fire.”[36] One sentence in this quotation sums up the whole case against cavalry. If the Whippets were idle because “being tied down to support the cavalry they were a long way behind the infantry advance,” what, in Heaven’s name, was the use of the mounted arm? The latter in theory, and if it was to justify its existence at all, should have been a long way in front, instead of a long way behind. Cavalry behind an infantry attack are merely an annoying encumbrance, blocking up roads, losing their way, and behaving, at the same time, as if they alone are winning the battle. Colonel Buchan, tactful at the expense of accuracy, permits himself to say that “the whole British cavalry performed miracles, advancing twenty-three miles from their points of concentration.” This is merely silly, for the points of concentration were far away to the west of Amiens. Sir John Monash, on the spot, and speaking from bitter experience of the uselessness and presumption of these archaic cohorts, tries to find excuses for them and writes with great moderation, but his conclusions tally with those of every other unprejudiced observer. He says:--

“The Cavalry Corps would appear in the battle area also,
with all preparations made for a rapid exploitation of any
success achieved. The utility of the cavalry in modern war,
at any rate in a European theatre, has been the subject of
endless controversy. It is one into which I do not propose to
enter. There is no doubt that, given suitable ground and an
absence of wire entanglements, cavalry can move rapidly, and
undertake important turning or enveloping movements. Yet it
has been argued that the rarity of such suitable conditions
negatives any justification for superimposing so unwieldy a
burden as a large body of cavalry, on the bare chance that it
_might_ be useful, upon already over-populated areas, billets,
watering-places, and roads.

“I may, however, anticipate the event by saying that the First
Cavalry Brigade was duly allotted to me, and did its best to
prove its utility; but I am bound to say that the results
achieved, in what proved to be very unsuitable country beyond
the range of the infantry advance, did not justify the effort
expended, either by this gallant brigade or by the other arms
and services upon whom the very presence of the cavalry proved
an added burden.”[37]

The Whippets were peculiarly unfortunate in being so shackled on this particular day. Such opportunities lay before these machines as never fell to them again. Within two or three hours of zero the enemy, on a wide front, was thoroughly demoralised and in flight, with most of his field batteries captured; and there lay ahead, occupied by his headquarters, transport and stores, a great stretch of good undulating country served by numerous roads in fair condition. It was the day of days for Whippets. The armoured cars, few in number and confined to roads, did great things; but for a light fast tank the opportunities were endless. Colonel Buchan is in error when he says that the tanks employed that day were “mostly of the Whippet type,” for, as we have seen, there were only ninety-six of the latter. Yet if these had been let loose on roving commissions, the haul of prisoners and guns, and the general demoralisation of the enemy, would have been far greater even than was the case. With extraordinary ineptitude, the two battalions were dragged about at the heels of Cavalry Brigade headquarters, a position which alone shows an abysmal misunderstanding of their functions. Elaborate tactics for their co-operation with horsemen had been drawn up, but as the horsemen, when they attempted to advance in any numbers, were checked almost at once by a few machine-guns, the bulk of the Whippets merely cruised about them in exposed situations without being able to strike an effective blow. A few, more fortunate, were set free for their proper tasks, and showed what might have been accomplished had all been handled in so bold a manner. Some came into hot fighting about Cayeux Wood, Le Quesnel, Guilleaucourt, and Harbonnières, along the Amiens-Chaulnes railway, and, by cutting telegraph and telephone wires, increased the enemy’s confusion. One, “Musical Box,” of the 6th Battalion, had an extraordinary career. Passing through the advanced infantry line and a few cavalry patrols two or three hours after zero, it engaged a field battery which had already knocked out two Mark V. tanks. Sweeping round behind the guns, it drove off the crews and pursued them, killing or wounding nearly all. Turning north toward the railway near Guilleaucourt, the tank then busied itself with various parties of fugitives, moved on again after a short halt, entered a German hutted camp where troops were hurriedly packing up their kits, and accounted for about sixty of these; continued in chase of fresh bodies of infantry retiring eastward, shot the drivers of several motor and horse vehicles, and finally got among some horse and transport lines near Rosières, where an immense amount of damage was done. The little tank, being entirely alone, far ahead of its fellows and of the infantry, was subjected to continual machine-gun fire, and petrol was pouring down the sides and into the interior from riddled tins carried on the roof. The petrol fumes, together with the intense heat, obliged the crew to breathe through the mouthpieces of their box-respirators. Eventually, about three o’clock in the afternoon--for the “Musical Box” was in action continuously for nine or ten hours--the tank was hit by a field-gun and set on fire on the eastern side of the Harbonnières-Rosières road, beyond our farthest objective for that day. The officer, Lieutenant Arnold, managed to drag out his two men, who were overcome with the smoke and fumes. The clothing of all three was on fire. Endeavouring to escape, after rolling in the grass to extinguish the flames, the driver was killed, and Arnold and the second man were captured and brutally ill-used. The full story of this remarkable exploit was only known after Arnold’s release at the end of the war.

The heavy tanks had met with most trouble on the two wings. On the left the enemy’s stubborn defence of the Chipilly Spur, overlooking a loop of the Somme, held up the 3rd Corps until the next day, and the 5th Tank Brigade had forty machines knocked out, mostly by enfilade artillery fire from this flank. Those which passed this danger point, however, did remarkable work, which included the capture of the 51st German Corps headquarters, with most of the staff, near Framerville, where the armoured cars helped. The Mark V. Stars of the 15th Battalion in many cases delivered their unhappy loads of machine-gunners on the final objectives, but the great size of these machines made them such excellent targets that a disproportionate number were knocked out, their crews and passengers being too often burnt to death inside them. The survivors of the infantry who were carried in them were emphatic in their determination to have nothing further to do with this method of transport in action.

The 4th Tank Brigade, working with the Canadians on the right, also suffered considerable losses in the early part of the day, owing to Debeney’s Frenchmen being held up for a time along the Avre valley. The little stream of the Luce, close behind our front line, had to be crossed without the enemy’s knowledge, and had caused much anxiety, as on account of its position and the necessity for secrecy, adequate reconnaissance of it was impossible. All the tanks, except two, crossed in a dense mist without a hitch and undiscovered by the Germans. Everything went with a rush at first, and there were few casualties; but in the second phase of the attack, until the French broke through the resistance about Moreuil and were able to bring their left level with the Canadians, tanks and infantry had a hard time at Beaucourt and Le Quesnel, where the enemy brought field-guns into close action with great boldness and success. One company of the 1st Tank Battalion, caught by a battery in enfilade near Le Quesnel, lost in a few minutes nine out of eleven Mark V. Stars, all of which were burnt out with heavy casualties to their crews and to the hapless infantry volunteers on board. One tank commander swung his machine round and steered straight for the battery, but this desperate and gallant act was useless. He was killed, with most of his crew and passengers.

Such incidents, however, were only local set-backs in what, as a whole, was a victory without parallel. On a twelve-mile front we had advanced at one point to a depth of seven miles. Twenty-four hours after zero 16,000 prisoners and over 200 captured guns had been counted. Amiens and its railways were freed, and the enemy’s chief means of lateral communication in this sector, the Chaulnes-Roye line, already was within range of our field-guns. Only at Chipilly, on the left, had there been any permanent check; and that flank was in its proper alignment the next day. The greatest result of all, however, coming as it did upon the battle of Soissons and Von Boehn’s retreat, was the profound moral effect wrought upon the German soldier. In this respect the 8th August stands pre-eminent. Ludendorff himself testifies that “August 8th was the black day of the German Army in the history of the war. This was the worst experience I had to go through....” And immediately, in public and secret documents, he proceeded to point the moral. The tanks alone had done this thing; and to combat tanks every nerve must be strained and every effort directed. On 11th August, while the battle was still in progress, he issued an urgent and secret order:--

“Troops allowed themselves to be surprised by a mass attack of
tanks, and lost all cohesion when the tanks suddenly appeared
behind them, having broken through under cover of a mist,
natural and artificial. The defensive organisation, both of the
first line and in the rear, was insufficient to permit of a
systematic defence.... As a weapon against tanks the prepared
defence of the ground must play a larger part than ever,
and the aversion of the men to the pick and shovel must be
overcome at all hazards.... Especially, there must be defences
against tanks. It was absolutely inadmissible that the tanks,
having penetrated into our advance line without meeting with
obstacles or anything, should be able to push on along the
roads or beside them for miles.... The principle that a body of
troops, even when surrounded, must defend their ground, unless
otherwise ordered, to the last man and the last cartridge,
seems to have fallen into oblivion ... a large proportion of
our ranks fight unskilfully against tanks. A tank is an easy
prey for artillery of all calibres....”

Then followed a summary of the various anti-tank measures which I have already described. All this was too late, however; it was shutting the door after the steed--in this case the German soldier’s nerve and heart--was irrevocably gone. And Ludendorff must have known, ever since Cambrai, that against a massed tank attack, supported by good infantry, no army in the world could stand.

Five days later, on 16th August, Sir Henry Rawlinson added his testimony in a Special Order to the Tank Corps.

“The success of the operations of 8th August and succeeding
days was largely due to the conspicuous part played by the 3rd,
4th, and 5th Brigades of the Tank Corps, and I desire to place
on record my sincere appreciation of the invaluable services
rendered both by the Mark V. and the Mark V. Star and the
Whippets.

“The task of secretly assembling so large a number of tanks
entailed very hard and continuous work by all concerned for
four or five nights previous to the battle.

“The tactical handling of the tanks in action made calls on the
skill and physical endurance of the detachments which were met
with by a gallantry and devotion beyond all praise.

“I desire to place on record my appreciation of the splendid
success that they achieved, and heartily to congratulate
the Tank Corps as a whole on the completeness of their
arrangements, and the admirable prowess exhibited by all ranks
actually engaged on this occasion.

“There are many vitally important lessons to be learned from
their experiences. These will, I trust, be taken to heart by
all concerned, and made full use of when next the Tank Corps is
called upon to go into battle.

“The part played by the tanks and Whippets in the battle of 8th
August was in all respects a very fine performance.”

About 100 tanks had been knocked out or otherwise disabled in the first day’s fighting. In the next day’s advance 145 took part, mostly formed as composite companies, the evil effects of the absence of any reserves being felt once more. The 10th Battalion helped the 3rd Corps to clear the Chipilly Spur and the difficult gullies which abounded in that region, while the 4th and 5th Brigades fought in Framerville and Rosières and Bouchoir. A few Whippets cut loose from their leading-strings and penetrated as far as the devastated area of the old Somme battlefield; but most of these machines remained tied to the ineffective wanderings of the cavalry, who displayed their usual _penchant_ for collecting in valleys where they were certain to be shelled. Another 39 tanks were hit on this day. In the meantime Humbert’s 3rd French Army had attacked south of Montdidier. On the 10th Montdidier fell, and we were on the edge of Lihons and within 3000 yards of the railway station of Chaulnes. Sixty-seven tanks were engaged in this fighting, of which 30 were disabled by shell-fire. Another 40 machines were in again on the 11th, when Lihons was captured in part and Humbert was on the edge of the much-talked-of Lassigny _massif_, familiar to Mr Belloc’s numerous readers; and with this the share of the Tank Corps in the battle of Amiens came to an end. The battle itself was virtually over. The old tactical ideas of hammering for weeks at one point had vanished with the acceptance of the new doctrines of tank warfare. The devastated area had now been reached: the objectives aimed at had been won; and with the enemy thoroughly awakened and obsessed with the danger south of the Somme, the time had come for another blow in the new style in a different quarter.

Some idea of the wastage in tanks involved in a battle of this description, and of the infinite advantages we might have gained if we had possessed an adequate supply of these machines, may be gathered from the following figures and dates. In the four days’ fighting, a total of 688 tanks had been engaged. No fewer than 480, including supply-tanks, had been handed over to salvage. The few remaining fighting machines needed a thorough overhaul before they would be fit for another battle. There were left the 9th Battalion, with 27 tanks, which had rejoined the 4th Brigade; the 11th Battalion, with 42, which had been rushed down from the 1st Brigade to the Somme area, but had not been used; the 7th and 12th Battalions, still with Mark IV.’s (42 apiece), and the 16th Battalion, with another 42 Mark V.’s, just out from England. With the 9th Battalion brought up to strength, this would give 210 tanks as an untired reserve on 12th August. The next offensive, for which three whole brigades were needed at the very start, was due for the 21st of the month, or in little more than a week!