Because of the fog and the din of the bombardment, the leading German infantry and tanks crossed no-man’s-land and reached the British line virtually unseen. Edmonds states that wherever tanks appeared the British line was broken.⁶⁷ The young soldiers had no effective anti-tank weapons, and a number surrendered to the German infantry. When the Germans attacked without direct tank support, however, the British generally succeeded in beating them off. This was certainly the case in the sector of the 173rd Brigade of the 58th Division. On the right, the 3rd Londons repulsed all attacks and held their position all day, until the French lost Hangard yet again in the evening, forcing the battalion to make a limited withdrawal. The left battalion, the 2/4th Londons, resisted the German infantry for a time, but the enemy tanks eventually drove them back to the Cachy Switch, where they reorganised about 10 a.m. The centre battalion, the 2/2nd Londons, were compelled to conform to this movement and to fall back to a position between Hangard Wood and Cachy. The 2/10th Londons, the counterattack battalion, tried, at the battalion commander’s own initiative, to restore the situation north of Hangard Wood at 10 a.m. They were checked, short of their objective, by machine-gun fire from Hangard Wood and the commanding officer, Lieutenant-Colonel W.F.J. Symonds, became a casualty, but they were able to fill a gap in the line between the centre and left battalions of 173rd Brigade and stopped the backward movement. The Fourth Army War Diary partly attributed the lack of deep German penetration in this sector to the inexperience and generally poor quality of the 77th Reserve Division, which had come from the Eastern Front. Their progress slowed by British machine-guns, and disheartened by heavy casualties, the troops of the 77th Reserve Division came to a halt 2,000 yards short of their objective and failed to take Cachy as planned.⁶⁸

The story was different on the 8th Division’s front. Grogan’s 23rd Brigade was attacked by two groups of tanks, with infantry and flame projectors, and the front line was rapidly overrun. The three forward companies of the 2nd West Yorkshires, raked in enfilade by the machine guns of the tanks as they crossed the British trenches, were completely overwhelmed. The battalion, down to 140 men and with only its support company left in action, was driven back, first to the railway west of Villers-Bretonneux and ultimately to the reserve line in the Bois l’Abbé. The two southernmost companies of the 2nd Middlesex were also swiftly overcome, whereupon the Germans worked round the right and rear of the third, and most northerly, front-line company and destroyed it in turn. The survivors of the 2nd Middlesex similarly withdrew along the railway in stages until they reached the reserve line. Grogan’s counter-attack battalion, the 2nd Devons, waiting south of Villers-Bretonneux, had its two left companies driven in by tanks and powerful German infantry support, which opened the way for the 4th Guard Division to enter the Bois d’Aquenne. The two right companies, with two more from the 1st Worcesters, did succeed, on the other hand, in clinging to their positions in the Cachy Switch, even though the Germans in the Bois d’Aquenne were behind their left flank.⁶⁹

In Coffin’s 25th Brigade, the 2nd Rifle Brigade, who were not attacked in front by tanks, stood their ground for a time, but the tanks which had overrun the 2nd Middlesex turned north and struck the Rifle Brigade’s two right-hand companies from the rear. The left company maintained its position with accurate fire for much of the day and helped to form a defensive flank with the 2nd Royal Berkshires – the counter-attack battalion – who swung back to face south-east in order to oppose any German attempt to advance northwards from the town. The 2nd East Lancashires, defending the town itself, had been prevented by gas from keeping men actually in Villers-Bretonneux and occupied only its eastern outskirts. They too were outflanked from the south and retired to the north-west of the town where, with the assistance of some Australian troops, they were able to stop the enemy from debouching from Villers-Bretonneux in that direction. In this they received the invaluable support of artillery of the 20th Division, firing at the Germans over open sights. To the south, as the 8th Division’s infantry retired, similar deeds were performed by ‘B’ Battery of the 83rd Brigade RFA, part of the 18th Division. One 18-pounder, under Second-Lieutenant A.I. Butler, sited first behind the railway embankment near the Bois l’Abbé and later moved up onto the embankment, engaged and hit a German tank. Butler’s gun fired continuously for 7 hours and expended 1,100 rounds before it was pulled back to safety.⁷⁰

Heneker knew by 7.20 a.m. that the Germans were using tanks and Lieutenant-General Butler, at III Corps headquarters, had been informed by 8 a.m. Acordingly, Butler placed the available seven Whippet tanks at the disposal of Major-General A.B.E. Cator’s 58th Division to operate in front of the Cachy Switch. Cator was also told to employ his reserve brigade, the 174th, to restore the situation, their place in the reserve line being taken by the 54th Brigade of the 18th Division. The 8th Division had ordered forward three Mark IV tanks of the 1st Battalion, Tank Corps, from the Bois l’Abbé–Bois d’Aquenne area and assigned them to Grogan of 23rd Brigade. He, in turn, ordered them to help in the defence of the Cachy Switch, where they arrived about 9.30 a.m.⁷¹

As a result of these moves by Butler, Heneker and Grogan, the first two British counter-attacks of any real significance were, in fact, delivered by tanks. Shortly after 9.30 a.m., in front of the Cachy Switch, the three Mark IVs of Captain J.C. Brown’s No.1 Section of ‘A’ Company, 1st Battalion, Tank Corps, engaged three German tanks in the first ever tank-versus-tank action. Two of the British tanks were ‘Females’, armed only with machine-guns, and were soon disabled, but the third – a ‘Male’ tank armed with 6-pounder guns and commanded by Lieutenant F. Mitchell, hit the leading German tank three times in succession, causing its crew to abandon it and driving the accompanying infantry back. Mitchell’s own tank was hit by artillery fire but not before he had also forced the other two German tanks to withdraw. Next, about noon, the Whippets of the 3rd Battalion, Tank Corps, under Captain T.R. Price – who, by another curious coincidence, had once served with the 6th Northamptonshires in the 18th Division – swept into action past the north of Cachy and on across the plateau to the spur leading southwards in the direction of Hangard Wood. Driving through a a line of Germans in shell holes, they surprised two or three German battalions on the reverse slope, scattering them in all directions and running over and crushing many who were trying to escape. One of Price’s Whippets received a direct hit, killing its crew, and 3 other Whippets were disabled on the return journey but the detachment lost only 5 officers and men and had inflicted an estimated 400 casualties on the luckless 77th Reserve Division, whose second-phase attack towards Cachy was thereby halted.⁷²

Two more tanks – a ‘Male’ and a ‘Female’ – had been handed over to Heneker at his request at 8.40 a.m. and, during the latter part of the morning, these were used to support the 1st Sherwood Foresters, the reserve battalion of the 24th Brigade, in an attempt to clear the Bois d’Aquenne of Germans and, if possible, capture the line of road running north-south through the town. The battalion, advancing in artillery formation, reached the southern edge of the wood at about 12.30 p.m. and turned northwards, driving the Germans towards its eastern edge. The battalion commander, Lieutenant-Colonel R.F. Moore, was, however, badly wounded at this critical juncture and, deprived of his leadership, the troops were unable to press home their attack, digging in on the road running north-east between the Bois l’Abbé and the Bois d’Aquenne. Here they were shelled and suffered heavy losses, but they blocked any further German progress – this road, some 8 miles from Amiens, being the closest point to the city reached by the Germans in 1918. Although the ‘Male’ tank was hit by a German field gun, the ‘Female’ pressed on to the north-western outskirts of the town where, at about 4 p.m., it helped to clear a number of enemy machine guns and encouraged remaining elements of the 2nd East Lancashires, the 2nd Royal Berkshires and patrols of the Australian 59th Battalion to join in a spontaneous attack which advanced approximately 150 yards to a small outlying copse. Bean observes that this minor action was the only counter-attack carried out with success by the 8th Division during the daylight hours of 24 April.⁷³ The truth is that the 8th Division, by late morning, could no longer seriously be classed as a cohesive fighting force and, as a consequence of the break in the centre of the line, the Germans had occupied a pocket 4 miles wide and 1 mile deep and had taken the town.

It must also be stressed that, even when all their difficulties are taken into account, Butler and Heneker failed, during the morning, to exercise any immediate grip on the situation, and the few minor counter-attacks that had taken place were piecemeal and largely unco-ordinated, thereby merely adding to the frustration of Australian commanders such as Elliott and Hobbs. Butler’s subsequent claims that the situation was not sufficiently clear, until well on in the day, to warrant the launching of a counter-stroke, and that Heneker was ‘very rightly of [the] opinion that partial attacks would not meet with success’ are undermined by the fact that both he and Heneker permitted just such partial attacks during the morning. Indeed, in this respect, Butler’s report, submitted in August 1918, contains more than a hint of retrospective self-justification.⁷⁴

The story of how the plan for a major counter-attack did evolve as the day progressed is both illuminating and complex, throwing some interesting light upon the command and control problems faced by the Fourth Army and its subordinate formations during the crisis of 24 April. It is instructive to see how, and by whom, these various command problems were tackled and, in most cases, solved. Alerted by the German bombardment and some time before the actual German tank-infantry assault, ‘Pompey’ Elliott, as early as 4.10 a.m., had ordered the battalions of the 15th Australian Brigade to stand to arms. Within the hour, at 4.50 a.m., Elliott was issuing provisional orders for a counter-attack by battalions of his brigade. The 59th Battalion was to assault up the high ground on which Villers-Bretonneux was located and the 60th Battalion was to advance along the western spur of Hill 104 adjacent to it. On reaching the Hamelet road, these units were to turn south-eastwards. The 60th Battalion would then push on to the Hamel road while the 59th Battalion, once at the Roman road, would swing south-west to face the town. The 57th Battalion was to be prepared to move along the railway on the south side of Villers-Bretonneux in an attempt to cut off the Germans. Like Elliott, Brigadier-General T.W. Glasgow, of the 13th Australian Brigade, then in Fourth Army reserve, had also anticipated the turn of events and had similarly warned his own battalions to be ready to move.⁷⁵

Shortly after 8.30 a.m., on receiving reports that Villers-Bretonneux had fallen and that the Germans were attacking north from it, Elliott asked Hobbs, his divisional commander, whether he could now launch a counter-attack at his own discretion. He was told by Hobbs that he was only to act within his allotted area unless he received an ‘urgent request’ to assist the British on his right. If such a request was received, he could then use his discretion, keeping Hobbs informed of his actions. A little before 9 o’clock, Hobbs, who was also increasingly concerned about the 8th Division’s plight, contacted Heneker to enquire if he was going to counter-attack and to offer the services of Elliott’s 15th Brigade. Heneker’s reply, according to some sources, was off-hand and non-commital.

As the battalions of the 15th Brigade moved off to their start line, their headstrong commander, disgusted and alarmed by reports of unnecessary withdrawals by British soldiers, ordered that ‘All British troops [were] to be rallied and reformed, as our troops march through them, by selected officers and on any hesitation to be shot’ (Hobbs later curtly instructed Elliott to delete the offending sentence). By this time Hobbs was as annoyed as Elliott by the lack of communication from Heneker. He complained to Lieutenant-General Birdwood, the Australian Corps commander, that he could not obtain any information from 8th Division and did not know if that formation intended to counter-attack. Heneker had, in fact, referred the question, and the offer of the 15th Australian Brigade, to Butler and, by 10 a.m., Hobbs and, in turn, Elliott were duly told that the British were ‘endeavouring to restore the line themselves’. The battalions of the 15th Australian Brigade were to stand fast ‘pending further orders’ but to be ‘ready for instant action’.⁷⁶

Knowing the ground and convinced that their troops would ultimately be required to participate in a counter-attack, Hobbs and Elliott, who were now straining at the leash, fumed at the delays and apparent inactivity of III Corps and the 8th Division. However, on learning, at 9.30 a.m., of the fall of Villers-Bretonneux, Rawlinson, from the first, insisted that the town should be retaken as quickly as possible, before the Germans were given time to consolidate, regarding this as ‘imperative for the security of Amiens’. One of his earliest steps to this end was to order Glasgow’s 13th Australian Brigade to march south to the battle area to assist in the recapture of Villers-Bretonneux. Glasgow’s brigade, part of the 4th Australian Division, was to be placed at the disposal of III Corps for the purpose. Heneker was told by Butler at 10.50 a.m. that, on arrival, the 13th Australian Brigade would come under the orders of the 8th Division. The brigade began its 8-mile approach march at 11.15 a.m. It was intended that three of Glasgow’s battalions would take part in a projected attack to retake the town while the fourth (the 49th Battalion) was to be used to strengthen the reserve line.⁷⁷

At 10.50 a.m., now under pressure from Rawlinson, Butler ordered Heneker to involve the 15th Australian Brigade in the 8th Division’s counter-attacks and, soon afterwards, directed Hobbs to keep in touch with Heneker and ‘to act as far as possible in close co-operation’ with him, though Butler then confused matters by reporting that Heneker ‘considered he would be able to clear up the situation with his own troops’. While these exchanges were taking place, Heneker, at 11 a.m., had ordered his own 25th Brigade to prepare for an attack from the north in which the 2nd Royal Berkshires and two companies of the 2nd Rifle Brigade were to recapture the northern part of the town. Having consulted his battalion commanders, Brigadier-General Coffin responded that the 25th Brigade currently had neither the strength nor the organisation for such an operation and that, with the Germans occupying the northern edge of the town and enjoying a perfect field of fire from the plateau, any advance in daylight without proper artillery preparation would be suicidal and likely to result in heavy losses for little or no gain. The proposed attack was therefore cancelled but Coffin’s advice appears to have had an influence, either directly or indirectly, on subsequent plans. At 11.30 a.m. – some time before the attack by the 1st Sherwood Foresters was delivered – Rawlinson proposed that Villers-Bretonneux should be retaken by means of a double counter-stroke, north and south of the town. Through Butler, Heneker, who would command the operation, was instructed to contact Hobbs and to arrange for the 5th Australian Division to join in a combined attack, starting at 2 p.m. Hobbs once again offered the 15th Australian Brigade, then still in reserve at Aubigny, or two reserve battalions of the 14th Brigade, which were even nearer.⁷⁸

The evidence presented by Bean indicates that it may have been Heneker’s GSO1, Lieutenant-Colonel C.C. Armitage, who proposed that the counter-attack should be made by two brigades, ignoring the town itself but passing to the north and south and meeting to the east of it. However, as Bean points out, probably the same idea – ‘a common and fairly obvious one for attacks on villages and small towns’ – had suggested itself to most of the commanders and staffs concerned, and ‘Pompey’ Elliott, for some time, had not only been advocating a not dissimilar plan but also aching to carry it out. A further refinement was added by Heneker, who, following the advice he had received earlier from Coffin, considered that a daylight attack across the open plateau would be doomed to failure. As the moon would be full, Heneker felt that the attempt might be made more successfully that night and urged Butler to postpone the operation for a few hours. The overall idea for a two-brigade advance on either side of the town at night – having been submitted to III Corps – was approved by both Butler and Rawlinson. The latter, indeed, spent some time on the telephone to Butler, explaining how he wished the plan to be executed and sent Lieutenant-Colonel Beddington, now a member of his staff, to Heneker’s headquarters to ensure that his instructions were not misinterpreted.⁷⁹

The French did not offer a great deal of help at this point. General Debeney was informed of the plan and promised to co-operate if the attack were postponed until 25 April. This, of course, did not suit Rawlinson, who continued to demand that the operation should take place before the Germans had settled down. Debeney’s only immediate measure, however, was to order the Moroccan Division forward to a position behind the British right, where it occupied part of their reserve line near Gentelles and so freed some of the 58th Division’s own reserves. Even so, Debeney told Butler during the morning that the Moroccan Division must not, under any circumstances, be used piecemeal to counter-attack or to reinforce the front, and would only be employed, in a counter-attack, as a complete formation. When Haig visited Rawlinson’s headquarters at 12.30 p.m., on his way to lunch with Butler at Dury, he was briefed on the situation with regard to the French and directed that a telephone message be sent to Lieutenant-General Sir John Du Cane, his senior representative on Foch’s staff, asking for the Moroccan Division to join in the counter-attack rather than remaining in reserve. It was thus with some irritation that, in mid-afternoon, Haig received a written message from Foch – delivered by Du Cane – which stated patronisingly that the importance of recapturing Villers-Bretonneux would not have escaped Rawlinson and pressed the British to launch a powerful counter-attack like the one Debeney had mounted that morning. Haig lost no time in making it clear to Foch that Debeney had not attacked at all and, at 6.30 in the evening, he asked Foch to direct the Moroccan Division, or another French division, ‘to co-operate energetically and without delay’ with III Corps to retake Villers-Bretonneux. But, by this time, most of the key decisions concerning the counter-attack had been made and it was too late for the French to play a meaningful role in it, though Foch did order Debeney to co-operate more actively on 25 April and to see what could subsequently be done to relieve the units of III Corps. Rawlinson was highly critical of Debeney’s conduct during the crisis. ‘I have found Debeney v[ery] difficult’, Rawlinson complained in his diary on 24 April, accusing the French First Army commander of having ‘misrepresented his ability to use the Moroccan Div. to Foch saying he was ready to attack with it’.⁸⁰

While all this was going on, Hobbs and Elliott were becoming more and more puzzled and disturbed by the ongoing delays in launching the counter-attack. However, at around 3 o’clock on the afternoon of 24 April, Hobbs was at last informed that the 5th Australian Division had been placed under III Corps for the forthcoming operation, meaning that Butler could now call upon both the 13th and 15th Australian Brigades for the counter-attack. In a telephone call 10 minutes later, Hobbs suggested to Butler that Elliott’s brigade should strike with three battalions past the northern side of Villers-Bretonneux and in a south-easterly direction. He then rang Heneker at 3.30 p.m., when he was told that the 8th Division’s part in the attack would, in fact, be carried out by the 13th Australian Brigade, which was to make its thrust to the south of the town so that the two main attacking formations would meet on the farther side. On learning of these developments from Hobbs, Elliott expressed his pleasure that the 13th Brigade would be on his right and stated that he was willing to attack by moonlight, preferably without previous artillery preparation.⁸¹

Glasgow had initially reported in person to Heneker’s headquarters at Glisy at 1 p.m. The commander of the 13th Australian Brigade has been portrayed as tough and tactiturn. Bean, who obviously admired him, describes Glasgow as an ‘Australian counterpart of the best type of English country gentleman … but rugged as the Queensland hills’ and writes glowingly of his ‘good sense, force of will, and honesty of purpose’, Glasgow was evidently as forceful and resolute as Elliott, if less volatile. When he met Heneker early in the afternoon, the latter was unable to give him all the information he needed, simply explaining that the picture was ‘changing from moment to moment’. As he had not seen the ground over which his troops were to counter-attack, Glasgow therefore wisely insisted that he should go forward to assess the situation for himself before the plans were finalised, a proposal which Heneker accepted. There can be little doubt that the delays which so vexed Elliott and Hobbs were partly caused in the afternoon by Glasgow’s request to conduct a personal reconnaissance and, in any case, once the decision had been made to attack in the evening, Heneker was more justified than he had been that morning in deferring further action until satisfactory arrangements had been made. Though under pressure from Rawlinson and Butler and all too aware of the tricky task ahead, Heneker must have been reassured by the availability of the Australian brigades. In this respect, Ross McMullin’s depiction of an increasingly jittery Heneker during the afternoon of 24 April does not quite ring true and ‘rattled’ and ‘quailing’ are not terms which are usually associated with the 8th Division’s commander.⁸²

Upon his return to Heneker’s headquarters at 2.30 p.m., Glasgow made two or three important contributions to the counter-attack plan. One of these was to persuade Heneker to drop Butler’s scheme to launch the southern arm of the attack from Cachy and to accept instead that it should start from a north–south line between the Bois d’Aquenne and Cachy. By doing this the 13th Brigade would avoid attacking across the enemy’s front and would thereby have more protection on its right. ‘Tell us what you want us to do, Sir’, said Glasgow to Heneker, ‘but you must let us do it our own way’. Glasgow similarly dissuaded Heneker from beginning the counter-attack at the time Butler wished, namely 8 p.m., which was only a few minutes past sunset. ‘If it was God Almighty who gave us the order, we couldn’t do it in daylight’, Glasgow protested. In the end, after discussion, 10 p.m. was chosen as zero hour, only 30 minutes earlier than the time Glasgow himself had advocated. He also let it be known that, partly because the front was still semi-fluid – with the location of the German guns unknown and some British guns still settling into new positions after the withdrawal – and partly to achieve surprise, he wished to dispense with a preliminary bombardment and creeping barrage.⁸³

All of these key proposals were incorporated in the operation orders issued by III Corps at 5.30 p.m. and confirmed later by the 8th Division. In the absence of a creeping barrage, all the available artillery was to fire standing barrages, with well-defined targets, including the town, the railway south of it, and the eastern part of Hangard Wood, from 10 p.m. to 11 p.m. Then the heavy batteries were to switch to a protective barrage 500 yards beyond the objective, while the field guns fired a similar barrage 300 yards beyond it. The artillery’s contribution would be co-ordinated by III Corps. Two battalions of the 8th Division – the 2nd Northamptonshires and the Pioneer battalion, the 22nd DLI – were to be attached to the 13th and 15th Australian Brigades respectively to attack and mop up Villers-Bretonneux itself from the south and the north. On Glasgow’s right, the 18th Division’s 54th Brigade, lent to the 58th Division for the purpose, was also to advance eastwards. On this occasion the 54th Brigade would consist of its own 7th Bedfordshires, the 7th Royal West Kents from the 53rd Brigade and the 9th Londons from the 175th Brigade of the 58th Division. It could therefore hardly be described as a cohesive force with battalions who were accustomed to fighting regularly alongside each other. Furthermore, as Edmonds comments, the whole operation was to be carried out by three brigades from three different divisions rather than by a single, well-integrated formation.⁸⁴

In the late afternoon and early evening, first Glasgow and then Elliott conferred with their respective battalion commanders. At 8 p.m. Glasgow reached Blangy-Tronville, where he would share Elliott’s headquarters. This was the first real opportunity the two brigade commanders had to co-ordinate their inter-brigade arrangements. Perceiving that the line of advance of the 15th Brigade’s 57th Battalion, to the south of the town, might cross that of the two British ‘mopping-up’ battalions and cause confusion, Glasgow persuaded Elliott to make a last-minute alteration to the plan and shift the 57th Battalion to the northern side of Villers-Bretonneux.⁸⁵

The night counter-attack at Villers-Bretonneux – made in the dark, across partly unfamiliar ground and without previous artillery preparation – was fraught with difficulties but was nevertheless one of the truly outstanding Australian operations of the First World War. The 13th Australian Brigade, which started more or less on time at 10.10 p.m., was held up for a while by the wire in front of the Cachy Switch, running diagonally across the front of the leading battalions. The 52nd Battalion on the right eventually surmounted this obstacle and then overcame the German troops beyond it after a charge and what Edmonds calls ‘some sharp fighting with the bayonet’, so reaching the objective. It was, however, subsequently brought back a little, with the right extending across the Hangard Wood road towards the Domart road, in order to maintain contact with the 7th Bedfords. The 51st Battalion, on the left, was enfiladed by machine guns on the southern face of the Bois d’Aquenne and had to deal with these by using bomb and bayonet. Lieutenant C.W.K. Sadlier won a Victoria Cross during this fighting. The battalion continued to take heavy losses but the advance went on and, by the early hours of 25 April, the 13th Australian Brigade, if not on its objective, was in a position to help pinch out the town should Elliott’s brigade also achieve success.⁸⁶

The battalion commanders of the 15th Australian Brigade had not returned from their crucial final conferences until 8 p.m. and, as a result of the darkness and detours to avoid lingering gas, the brigade was approximately 2 hours behind schedule in beginning the advance. The battalions reached the first objective, over a mile away, with relative ease but opposition then stiffened. At this point, Captain E.M. Young, of the 59th Battalion, gave the order to charge and the whole line went forward with a ferocious yell, carrying all before it. The German line was broken and the Hamel road reached with the loss of no more than 150 men. While the old front line had not been retaken, the Australian counter-attack had almost encircled Villers-Bretonneux and the Bois d’Aquenne and had regained most of the ground on the northern part of the battlefield which had been lost earlier.⁸⁷

The British units participating in the counter-attack had mixed fortunes. On the extreme right, elements of the 9th Londons (Queen Victoria’s Rifles) got as far as Hangard Wood, but were unable to hold on and were forced to fall back to their start line. The 7th Royal West Kents, meeting heavy machine-gun fire from German posts in shell holes which had escaped the barrage, lost 230 men in an hour and dug in about half-way to their objective. The 7th Bedfords met little opposition over the first thousand yards of their advance and had crossed the Hangard Wood–Villers-Bretonneux road when they came under artillery fire from the right. The survivors withdrew to a position 500 yards behind the road where they found themselves virtually surrounded by Germans in nearby shell holes. Their only remaining officer, Second-Lieutenant W. Tysoe, resisted German calls to surrender. The Bedfords too held on and maintained touch with the 13th Australian Brigade until relieved in the early hours of 26 April. Indeed, as darkness was falling on 25 April, Tysoe and his force – who averaged 19 years of age – even mounted a counter-attack with the bayonet to drive off a German attempt to overwhelm their position.⁸⁸

The two battalions from the 8th Division which had been detailed to mop up the town encountered problems, and the continued firing from the town during the night gave the Australian commanders the impression that the attack of the 2nd Northants and 22nd DLI had not been conducted with the required vigour. Elliott even informed Heneker at 4.15 a.m. that he suspected that the Durhams had not moved at all, while Glasgow similarly reported that nothing had been seen of the Northamptonshires. The truth, as Bean concedes, was that the task assigned to these two battalions was far harder than Elliott or Glasgow assumed, for while ‘the thrust of the two Australian brigades was directed where the Germans did not expect it, the 22nd DLI and 2nd Northampton had to strike where the blow was expected, and the Germans were in strength with great numbers of machine-guns and fully prepared’. The 2nd Northants, on the south of the town, quickly came under artillery fire – their commanding officer, Lieutenant-Colonel S.G. Latham, being killed and the adjutant wounded – and then, some 500 yards from the railway embankment, were subjected to fierce machine-gun fire and were unable to progress further. Their casualties, by the end of the fight, totalled 295 officers and men. The Durhams likewise found the town to be strongly defended and, after efforts by two companies to approach it had been driven off by intense fire, the battalion did not try again that night. Elliott’s message to Heneker concerning the lack of progress of the Durhams had, however, prompted Heneker to send forward the 2nd Royal Berkshires of the 25th Brigade, to attack and enter the town from the north. This they did in some style at 6.30 a.m. and, with the assistance of the Australian 57th Battalion, cleared a considerable area of the town. As the light improved, the 22nd DLI managed to enter Villers-Bretonneux from the north-west. Gradually, during the day on 25 April – the third anniversary of the Anzac landings on Gallipoli – the pockets of German resistance in the town were eliminated, while parties of the 2nd Middlesex, 2nd West Yorkshires and 1st Sherwood Foresters cleared the Bois d’Aquenne.⁸⁹As Edmonds writes, because of ‘General Rawlinson’s grasp of the situation and resolution to act without French assistance, the enemy had been deprived of a considerable proportion of his gains of the morning of the 24th. He had been cleared out of Villers-Bretonneux, although the original front had not quite been recovered, and he had been foiled in his object of diverting attention before the attack on Kemmel took place and of getting nearer to Amiens’.⁹⁰

Apart from some local counter-attacks to adjust the line on 25 April – one of which involved the use of three Whippets west of Hangard Wood – the last major scene in the drama was enacted in the early morning of 26 April when the French, in the shape of the splendid Moroccan Division, belatedly made their advance. The Moroccan Division, in the words of its own war diary, was ‘in superb condition and the troops impatient to attack’.⁹¹ To restore and straighten the line in the Cachy area and to the south, the division was ordered to recapture the ground between Monument Wood and Hangard Wood. At the same time, two tired battalions of the 18th Division, the 7th Queen’s and the 10th Essex, were placed temporarily under French command and were to attack Hangard Wood. The French 131st Division would endeavour to retake Hangard village. The French unwisely rejected Butler’s advice that they should not attempt to launch the whole Moroccan Division in daylight across ground registered by the German gunners. It has to be admitted, on the other hand, that the vagueness of the British line in this sector, and the imperfect knowledge of the British troops themselves as to their exact positions, contributed to the French decision – since they, unlike the Australians and the British, were not prepared to attack from an ill-established line across open ground at night and wished to employ a barrage. When the Moroccan Division advanced at 5.15 a.m. on 26 April, under a creeping barrage but without a preliminary bombardment, resistance was met several hundred yards further west than anticipated, The barrage was lost and, having taken heavy casualties, the Moroccans were driven back by a counter-attack. The 131st Division failed to recapture Hangard village and the only success was achieved by the 10th Essex and the 7th Queen’s who, despite incurring heavy losses from machine-gun and rifle fire, retook the western part of Hangard Wood and held on there until relieved next day.⁹²

The British divisions, by then, had little more left to give. Since 5 April, the 8th Division had lost 3,553 officers and men; the 18th Division some 2,446; and the 58th Division 3,530. The casualties of the 2 Australian brigades in the second battle were 455 for the 15th Brigade and 1,009 for the 13th Brigade. However, Edmonds estimates that, with gas casualties before 24 April included, the 15th Brigade lost as many as 50 officers and 1,284 other ranks. Bean estimates the overall German casualties as being around 10,400.⁹³

The performance of the British divisions at Second Villers-Bretonneux – as on 4 April – had been variable. The battalions from the 18th and 58th Divisions, on the right, had certainly prevented any fatal German penetration in this sector and had attacked and counter-attacked, with some success, when required. The fact that this right flank, if not totally secure, did not significantly collapse in either battle undoubtedly gave the Australian counter-attacks a better chance of succeeding. The 18th Division, in particular, could be reasonably proud of its record at Villers-Bretonneux. One should take into account that the division had already been in action several days at the start of the German March offensive before fighting in both battles at Villers-Bretonneux. Besides their contribution to the defensive operations on 4 and 24 April, the infantry battalions of the division had, between them, also taken part in at least eight attacks or local counter-attacks in the period from 2 to 26 April, a higher total, I believe, than any other Allied division at Villers-Bretonneux could claim.

The 8th Division did not have its finest hour at Second Villers-Bretonneux and, as in the case of the 14th (Light) Division on 4 April, its early reverses in the face of the German attack had created a critical situation. Like the 14th Division, the 8th Division had been seriously weakened by the March fighting, though the 8th had unquestionably performed with greater distinction than the hard-hit 14th in the first week or so of the German offensive. Moreover, unlike the 14th, it had faced tanks when attacked by the Germans at Villers-Bretonneux – not an everyday experience for British soldiers in the Great War and an extremely alarming one for a good, but weakened, division which had not yet fully recovered from its previous trials nor properly absorbed its large proportion of raw young recruits. One might therefore contend that there were some excuses for its partial collapse on 24 April, while it could also be pointed out that a number of its battalions had regained sufficient cohesion, spirit and strength to participate in the mopping-up operations on 25 April. The divisional history asserts that, while it was the splendid fighting qualities of the Australians which had turned the enemy out of the positions he had gained, ‘it was the 8th Division, on which the full weight of the blow fell, who stopped the advance before its purpose had been accomplished’.⁹⁴ The 8th Division, however, was unfortunate enough to be caught up in the German offensive on the Aisne in May and it was consequently not surprising that its performance in the Hundred Days, between August and November 1918, was moderate at best. The 14th Division was another formation which had a relatively unimpressive performance level during the final British offensive. The 58th and 18th Divisions, in contrast, had success rates of 63 per cent and 78 per cent respectively in opposed attacks between 8 August and 11 November.

Australian criticisms of the British divisions at Villers-Bretonneux were thus, I would suggest, only partly justified and were sometimes based on over-hasty judgements without full knowledge of the circumstances. It is also suggested by Bean that the ‘forcefulness’ of Elliott and Glasgow was ‘not wholly relished’ by Heneker, under whose command they had fought at Second Villers-Bretonneux. From what others – such as Hubert Essame, who also served under him as adjutant of the 2nd Northamptonshires – have written about Heneker, this assumption itself appears to be at least questionable, for the commander of the 8th Division was not the type of man who could be easily bullied. Bean is possibly on safer ground when he remarks that ‘amid all the telegrams of congratulation that afterwards reached the two [Australian] brigades, there came no message from the divisonal commander to whom they had been lent’. Though Bean notes that ‘two senior members’ of Heneker’s staff rode over and thanked Glasgow for the 13th Australian Brigade’s efforts, this failure by Heneker himself to acknowledge the part played by Elliott and Glasgow and their troops was, if deliberate, a petty and mean-spirited act which does him no credit.⁹⁵ Happily, others were more than ready to give the Australians the praise that they had so richly earned. Rawlinson noted in his diary that the 13th and 15th Australian Brigades ‘did brilliantly’ while the Fourth Army War Diary similarly describes the performance of the 13th Brigade – given the lack of time for preparation and reconnaissance – as ‘exceptionally brilliant’.⁹⁶ Grogan, of the 23rd Brigade, later called the night operation of 24–25 April ‘perhaps the greatest individual feat of the war – the successful counter-attack by night across unknown and difficult ground, at a few hours’ notice, by the Australian soldier’.⁹⁷ Butler too wrote in his official report on the operations: ‘My thanks are especially due to the Australian brigades placed under my orders for the counter-attack on Villers-Bretonneux, who carried out the night attack with magnificent dash …’.⁹⁸ It is worth noting that the Australians were also willing to reverse earlier judgements and give credit where it was due. The 2nd Royal Berkshires were reported by Australian officers and NCOs who visited them on 24 April to be ‘badly shaken and probably liable to panic’, yet, after the battalion’s work with the Australian 57th Battalion in helping to mop up Villers-Bretonneux on the morning of 25 April, a message was sent by the 5th Australian Division to say ‘Well done Royal Berks’.⁹⁹

Most of the commanders involved in the operations of 24–25 April had made an important contribution either to the handling of the desperate defence of the morning of 24 April or to the planning of the successful counter-attack – or, indeed, to both. It should be recognised that Haig himself was personally in touch with the situation, having spent time at the headquarters of Fourth Army and III Corps on the afternoon of 24 April. He does not appear to have interfered in the detailed conduct of the battle, but he did what was expected of a Commander-in-Chief and put pressure on the French to co-operate, not only by attacking with the British as soon as possible but also by relieving the divisions of III Corps. Rawlinson, for his part, had quickly appreciated the need for a swift counterattack and had tried to ensure that it was carried out before the Germans had been allowed time to consolidate and move all their guns into new positions. Butler, the commander of III Corps, had to be persuaded to drop some of his own ideas concerning the direction and timing of the counter-attack, but at least he remained open to reasonable argument, while Haig – who lunched at III Corps headquarters – was impressed by the ‘quiet methodical way’ in which he issued orders and made his arrangements with Hobbs and Heneker over the telephone.¹⁰⁰ Heneker and his staff, as well as Elliott, can claim to have helped shape the plan to pinch out Villers-Bretonneux in the counter-attack, but, while Elliott and Hobbs imparted a considerable sense of urgency to the command deliberations and discussions on 24 April, Heneker’s refusal to panic, plus his readiness to listen to good ideas and act accordingly were vital ingredients in the formula for success. As Bean admits, it was Heneker, above all, who, ‘following the advice of Brigadier-General Coffin and the battalion commanders of the 25th Brigade, prevented the wasting of the 15th Brigade at midday, and induced General Butler, despite Rawlinson’s pressure, to acquiesce in postponement until evening’.¹⁰¹ Finally, Glasgow’s vital input, particularly with regard to the timing of the counter-attack and the start line for his brigade, should be re-emphasised.

In reporting on the performance of his divisions between 21 March and First Villers-Bretonneux, Butler had remarked on the lack of training or initiative of junior officers in ‘open warfare operations on a large scale’.¹⁰² Edmonds was also critical of the standards of British junior leadership in both the March–April fighting and, later, the Hundred Days. Yet, as I have argued elsewhere, I believe that, if standards of junior leadership in the British divisions was really so low, the achievements of the young conscripts in stopping the German thrusts of March and April, and subsequently in advancing to victory between August and November, would simply not have been possible. In this respect, the experience of March and April helped young officers and NCOs to shake off the habits of trench warfare and almost certainly gave those who survived a greater degree of confidence and self-reliance that stood them in good stead later in the year. The determination shown by Second-Lieutenant Tysoe, who, as the only remaining officer of the 7th Bedfords, won a DSO for maintaining his position to the right of the 13th Australian Brigade throughout 25 April, illustrates what could be achieved by relatively inexperienced officers when the challenge came. According to the historian of the 18th Division, the ‘fine example of gallantry and leadership by this young officer was entirely instrumental in holding the ground gained, with many young soldiers who were in action for the first time’.¹⁰³

The often complex chain of command at First and Second Villers-Bretonneux could have been a recipe for disaster. Australians were placed temporarily under the command of XIX Corps and III Corps and under the 8th and 18th Divisions in one or other of these actions, while elements of the 58th Division were attached to the 18th Division and vice versa between 2 and 26 April. But, despite differences of command style and personality clashes, the command system at Villers-Bretonneux could be judged to have worked remarkably well in difficult and dangerous circumstances. This was partly due to the willingness of Butler and Heneker to accept sensible advice from below, while the fact that Haig and Rawlinson were prepared to leave the tactical conduct of a critical battle largely to their subordinates made it easier for those at corps, divisional and brigade level to respond effectively and flexibly to the changing demands of the situation. Villers-Bretonneux can therefore be seen as indicative of the extent of the devolution of battlefield command and decision-making in the BEF by 1918. A number of historians, including myself, have recently argued that, so far as the day to day conduct of operations was concerned, Haig, GHQ and the army commanders were far less relevant by mid-1918 than they had been two years, or even a year, before. One could also contend that the experience of the March Retreat and of Villers-Bretonneux may have accelerated this process, for the nature of tactical command in the BEF during the Hundred Days was, in many ways, quite different to that which had existed at Arras or Third Ypres in 1917. Mistakes and failures would continue to occur in 1918, but there is now wider agreement among historians and students of the First World War that those in the BEF who were ready to learn from difficult experiences, to embrace new tactics and techniques, and to encourage greater individual self-reliance and initiative, were exerting a more marked influence on the course of operations by the time the Hindenburg Line was breached. It is surely more than a coincidence that that the British infantry divisions which performed best at Villers-Bretonneux – namely the 18th and 58th – also had a good or above average success rate in attacks during the final and victorious Allied offensive.