Other formats

    TEI XML file   ePub eBook file  


    mail icontwitter iconBlogspot iconrss icon

Battle for Egypt

CHAPTER 21 — Planning for Ruweisat

page 220

Planning for Ruweisat

THE fighting at Tell el Eisa and during the following week gave glimpses of Rommel at his tactical best. To seal the penetrations made by the Australians and South Africans, he had only elements of 382 Infantry Regiment then arriving on the battle-field, his own headquarters' Kiehl battle group with three or four tanks, Recce Detachment 3, part of Trieste Division, and 15 Panzer Division's Kirsten battle group which was sent to the coast with twelve German and two Italian tanks.

These small forces could not recapture the important tactical features that had been lost, but boldness in counter-attacking prevented the Australians and South Africans from exploiting their success. Once again Auchinleck reported: ‘I had hoped to be able to exploit this success to the west and south, but the enemy offered strong resistance and I had no reserves available with which to reinforce the attack.’1

On the 11th Rommel decided to make a counter-stroke. He ordered 21 Panzer Division to disengage unobtrusively from the Kaponga area, and on the afternoon of the 13th he launched it in successive attacks against Alamein Box with the object of cutting off the British forces stretched out to Tell el Eisa. All the attacks were preceded by heavy air and artillery bombardment but the defences held firmly. Rommel thought the failure was due to the late arrival of the infantry and he made yet another attack late the next day, the 14th, on the tip of the salient. This also failed.

Rommel's apparent concentration in the northern sector was noted by Auchinleck. He decided in the evening of the 13th that ‘on 14 July the situation may become favourable for a counter-stroke.’2 He ordered his two corps commanders to complete arrangements forthwith for an extended ‘Operation bacon’ to be carried out at short notice. Instead of bacon being contingent on 30 Corps' exploitation from the new Alamein-Tell el Eisa line, it was now made Eighth Army's main operation.

In the intention paragraph of the Army order, Auchinleck declared the purpose of the operation to be: ‘To break through the enemy's

1 Despatch, p. 365.

2 Eighth Army Operation Order, No. 96, 13 July.

page 221 centre and destroy his forces east of the track El Alamein-Abu Dweiss and north of Ruweisat Ridge.’ He gave the leading role to 13 Corps, which was ordered to secure the high ground about Point 63 by a night attack and then to exploit to the north-west. At the same time, 30 Corps was to seize the eastern end of Ruweisat Ridge to secure 13 Corps' flank, and also Sanyet el Miteiriya as well as any ground elsewhere on the corps' front as opportunity offered. Gott was ordered to fix the time the infantry of both corps were to arrive on their objectives on the ridge.

Many of the reasons for the failure of the operation, especially the disaster that befell the infantry of New Zealand Division after they had taken their objectives, fall under the heading of wisdom after the event. On the other hand, it seems impossible to avoid the conclusion that the project was hastily conceived, loosely coordinated, and abounded in examples of poor staff work on matters which might be supposed to be within the knowledge and experience of those responsible. An appreciation of this background is needed to follow the course of the battle.

Auchinleck's despatch1 deals with the fighting on Ruweisat under the prosaic heading: ‘Consolidation of the position.’ The scale of the operations is further minimised in the despatch by the opening paragraph: ‘In order to improve our position against a possible enemy offensive and to set the stage for a further attack on our own part, the Eighth Army maintained its pressure along the whole front.’

Auchinleck's order of 13 July, however, leaves no doubt that a decisive success was sought with bacon. The attack might be deemed maintenance of pressure on the enemy. But it was neither subsidiary to other operations nor preparatory to further battle. Had bacon fulfilled the Army Commander's intention as set out in his orders, there would have been nothing left of Panzerarmee to fight on the Alamein Line.

Although important, this difference between despatch and order concerning the intention is a minor matter when compared with 13 Corps' interpretation of the object. The Army and also 30 Corps' orders emphasized the destruction of the enemy. The written orders of 13 Corps contained no hint of the larger enterprise. The 13 Corps' orders said merely: ‘13 Corps will secure Point 64.’2 The objective was not mentioned as a bridgehead from which the armour would issue to smite the enemy. First Armoured Division was not encouraged in the order to play a decisive part in winning the success sought by Auchinleck. Its role, according to the order, was

1 Despatch, pp. 365–6.

2 13 Corps Operation Order, No. 140, 14 July. Point 64 is also Point 63. See page 224.

page 222 to protect the New Zealanders' southern and western flank from first light on 15 July, with exploitation to the north-west ‘if a favourable opportunity occurs after first light.’

The order was signed by a subordinate staff officer on behalf of the Brigadier General Staff. Doubt has been raised as to whether Gott saw it.1 Be this as it may, the facts are that when Gott called on Inglis at 7.30 a.m. on 14 July he confirmed the general idea and plans made at the conference two days before. The only variation was an intimation that 5 Indian Division of 30 Corps would co-operate with an attack on the New Zealanders' right flank. This confirmation led Inglis to note in his diary: ‘The whole object of the operation is that we should clear the way up to Ruweisat Ridge in order to provide an opportunity for the armour either to cross the ridge or move round the west end of the ridge to exploit.’

Concerning the protection to be given the Division by the armour, the diary note continues: ‘If the armour does exploit our infantry will be protected against counter-attack by the exploitation. In case exploitation is not possible, an armoured brigade is to be in close support2 south-west of 4 Brigade and another in 5 Brigade area to protect our infantry against armoured attack from first light.

On this base, that of the decisions at the conference on the 12th and Gott's confirmation on the 14th, New Zealand Division planned the detail of its attack.

If 13 Corps' written order watered down the Army plan, that issued by 1 Armoured Division3 emasculated the intention. The information paragraph of this order said: ‘Eighth Army is going to carry out an attack during the night 14/15 July with the object of securing Point 64 feature’, while the intention paragraph ordered: ‘1 Armoured Division will co-operate in the attack of the two Corps.’ Instructions to the division's two brigades also fell short of the plan envisaged by Army and arranged at the corps conference. Those to 2 Armoured Brigade said: ‘2 Armd Bde will be prepared to move on centre line of the inter-corps boundary with the tasks of (a) exploiting success of the N.Z. Div. to the NW; (b) countering any counter-attack by the enemy armour against NZ Div which may develop from the NE, North or NW.’ The 22nd Armoured Brigade was also told that it ‘will be prepared to move’ on a centre line from Deir el Hima to an area between Alam el Dihmaniya and Point 634

1 ‘I find it almost incredible that Gott lost sight of the main objective of his own operation and doubt whether he ever saw the “watering down” order issued by his Corps.’—Letter to War Histories from Major-General Inglis, October 1952.

It was also Gott's practice to settle important operational matters with his divisional commanders personally and without any of his staff present. This could easily lead to divergences between his verbal orders and the confirming orders issued later by his staff.

2 Author's italics in all cases.

3 1 Armoured Division Operation Instruction No. 15, 14 July.

4 The area was precisely defined by map references.

page 223 with the task of protecting ‘the southern and western flank of the NZ Div attack from first light to 15 Jul, particularly against attack by enemy armd forces.’

In these orders there was no suggestion of urgency or of rallying the division to the task of breaking through Rommel's centre and destroying his forces. In this they reflected 13 Corps' written order rather than Gott's verbal orders given at the conference on the 12th and confirmed to Inglis on the 14th. Moreover, the instruction to both brigades that they ‘will be prepared to move’ implied a waiting role and allowed the brigade commanders a discretion which, with fateful consequences, they exercised in the operation.

No one at Eighth Army or at 13 Corps appears to have noticed the marked differences between the Army and Corps' orders and, at 13 Corps, the divergences between Gott's verbal orders and those committed to writing by his staff and at 1 Armoured Division. The New Zealand Division paid a heavy price for this lack of co-ordination.1

Information held by Eighth Army and its two corps concerning the enemy varied in quantity and quality. The start line prescribed by the Army order reflected ignorance of the enemy's positions on Ruweisat. The line laid down for 13 Corps ran hard by one enemy defended locality and through another to join the line for 30 Corps which was set in rear of an enemy position. In necessary adjustments, Gott left New Zealand Division on the line it had occupied for two to three days, and General Ramsden set the start line for 5 Indian Division back about a mile on the left flank.

In the result, both infantry divisions started from much further back and consequently had much further to go in the dark than Auchinleck contemplated. Representations by the New Zealand brigade commanders that the start line should be closer to the enemy were not adopted. The effect on the artillery plan for covering the infantry on their objectives has already been noted. Moreover, the adjusted lines for the two corps did not prolong each other as the Army order intended. Fifth Indian Division's left flank was placed about a mile and a half in advance of the New Zealanders' right. It was perhaps only incidental that 5 Division's operation order gave a start line which projected 500 yards over the inter-corps boundary.

New Zealand Division had pressed for air photographs of the enemy area but they were not available. Some photographs, however, reached the Division at 7 p.m. on the 14th, but due to

1 The failure in this respect may be attributed to a practice of regarding written orders merely as confirmation of verbal instructions and therefore not calling for detailed study. More importantly, because an operation often started before the written orders were circulated, the written orders were often filed without being examined in detail.

page 224 over-exposure there was no detail in the area of the Division's operations. Good photographs might have revealed that enemy posts, located by reconnaissance and fighting patrols, which were thought to be his forward defensive localities constituted in fact his main line of resistance, and that there were no other strong-points between them and the crest of the ridge. The New Zealand Division believed that the whole of Ruweisat and the ground down to Kaponga was covered only by 15 Panzer Division and a few Italian posts. Fifth Indian Division, on the other hand, had been told that Pavia and Brescia Divisions held the ridge and its approaches, with 15 Panzer in support. The attacking infantry were expected to make contact with the enemy on a line which proved to be well behind the defended localities.

Assessments of the enemy tank strength and the dispositions of the armour were also faulty. First Armoured Division's operation order reported 21 Panzer Division's attacks on Alamein Box and then said: ‘15 Armoured Division is believed to be in the general area north of Point 63;1 strength in tanks not likely to be more than 35 German tanks. The bulk of the Italian armour is probably in the south with elements included in the German armour as makeweight.’ As a fact, less than half of 15 Division—the Kirsten group withdrawn by Rommel on the 10th to counter the Australian attack—was north of Ruweisat. The remainder of the division with thirteen tanks2 was on the southern slopes of Ruweisat, the tanks being mainly in the area to be traversed by the New Zealand 5th Brigade. The thirty-five tanks credited to the division were nearly the total runners in the whole of Afrika Korps. Fifth Indian Division had wrong information concerning the location of 15 Panzer Division, which its operation order placed vaguely north-west of Brescia Division, roughly somewhere beyond New Zealand Division's objectives.

Intelligence was further at fault in not warning the formations of the possibility of error in identifying features in the battle area. The Army had no fewer than five series of maps covering Ruweisat and at least two, possibly more, were used in the operation. Army and 13 Corps' orders named the infantry objective on Ruweisat as Point 64. The New Zealand Division order gave it as Point 63. This was not a clerical error. There were three trigonometrical points on the ridge and their designation varied according to the editions of the maps used, Point 63 in one edition being Point 64 in another. Map reading in the desert was difficult in any circum-

1 That is, north of Ruweisat Ridge and the infantry objectives.

2 Two Mark II, 10 Mark III, 1 Mark IV, 1 command vehicle—15 Division tank return, 10 p.m. 14 July.

page 225 stances. The task was not made easier when the same features had different numbers or names.1

It should be repeated that these notes and observations represent, in the main, wisdom after the event. They are the product of detailed search for the reasons why a promising project went wrong. In their sum they appear to stress the importance of checking plans and orders at all stages to ensure that there are no loose ends. Later Eighth Army, and in it the New Zealand Division, went from success to success as a result of careful planning and preparation of attacks on such positions. Ruweisat did not teach new lessons. It emphasized old ones, at a bitter price.

New Zealand Division made its final preparations for the battle at a divisional conference held by General Inglis at 10 a.m. on the 14th.

The Division was ordered to attack that night on a two-brigade front on a bearing of 320 degrees from the start line. This axis of advance passed close to and east of Point 63. Fourth Brigade's objective was defined on the sketch attached to the order as an arc stretching from the south-west corner of Deir el Shein to the eastern tip of El Mreir and covering about 4000 yards. A similar arc slightly to the east of Point 63 and about 3000 yards across its chord was defined for 5 Brigade.

The brigades were ordered to advance on two-battalion fronts with their third battalions in reserve. In compliance with 13 Corps' orders, they were to make contact with the enemy forward defences at 1 a.m. and to reach their objectives at 4.30 a.m. Each brigade was given a quota of field, anti-tank and anti-aircraft guns, machine-gunners, engineers, and signallers under command so that, in effect, they attacked as brigade groups. It was not intended, however, that they should operate independently in the accepted brigade-group fashion. Direct command was to be exercised by the Divisional Commander. Ruweisat, incidentally, was the first occasion on which the Division attacked as a division, with two brigades shoulder to shoulder and with all the resources of a division at their disposal except a reserve brigade.

General Inglis confirmed earlier proposals that the infantry assault should be made silently. This gave rise to some discussion on the artillery programme, particularly on the question of support for the infantry when they reached their objectives. The Army and Corps' plans for the artillery were designed to deceive the enemy. Up to 9 p.m. on the 14th, the guns of 13 and 30 Corps were to engage in counter-bombardment, or counter-battery, fire,

1 To avoid confusion the point will be referred to in this narrative as Point 63.

page 226 a not unusual activity. From 9 p.m. until first light they were to fire normal harassing tasks as if nothing additional were impending. After first light, by which time it was expected the objectives would have been taken, all artillery was to support New Zealand Division under the direction of 13 Corps. The programme was accepted with two amendments—maximum harassing fire up to 9 p.m. and then normal fire until 12.30, when all artillery action would cease.

The conference discussed the risk that, because of the great depth of the advance, the infantry might be supported on their objectives for a time by only part of the artillery firing at nearly maximum rang with super-charges. It was agreed that the risk would exist only for a brief space and was not likely to be dangerous because of the advance of the British armour. Nevertheless, as a precaution, specific orders were given that at least two-thirds of the guns should be ready to support the infantry on the objectives at first light. No one, of course, envisaged the possibility that the supporting arms might be prevented from moving forward. The decision reflected the confidence of the infantry in the field regiments. This confidence was not shaken when, as happened, the forward movement of the guns was held up and the infantry were deprived of their support.

As no difficulties were expected with minefields, the engineers were given only a precautionary role in their attachment to brigades. General Inglis, however, warned 5 Brigade that a suspected minefield about Point 63 would have to be investigated and cleared to permit the armour to go through. This warning may be cited as additional evidence of the understanding that the armour would be on hand to break out as soon as Point 63 was captured.

The conference was told that 5 Indian Division would attack the ridge on the New Zealanders' right but details of its plans were not available. Fifth Indian Division had been advised only during the previous night that it was to take part in the operation. All of its preparations, including reconnaissance of the battlefield and an inter-brigade relief in the rear area, had to be started from behind scratch. The division's operation order, confirming verbal instructions given to lower formations during the day, was signed at four o'clock, and though a copy was sent to New Zealand Division, it apparently arrived after the advance had started so that the Division entered the fight without knowing the plans and detailed objectives of its neighbour. This lack of knowledge is reflected in General Inglis' diary, which notes that an Indian brigade was to attack along the crest of the ridge from east to west to link with 5 New Zealand Brigade. As a fact, the Indian advance was to be parallel with that of New Zealand Division. page 227 The principal elements in the New Zealand order of battle as finally approved were:

Fourth Brigade (Brigadier J. T. Burrows): 18, 19 and 20 Battalions,

4 Field Regiment, 31 Anti-Tank Battery, 41 Light Anti-Aircraft Battery,

6 Field Company, 2 Machine Gun Company, J Section Signals, B Company

4 Field Ambulance.

Fifth Brigade (Brigadier H. K. Kippenberger): 21, 22 and 23 Battalions,

6 Field Regiment, 33 Anti-Tank Battery, 43 Light Anti-Aircraft Battery,

7 Field Company, 4 Machine Gun Company, K Section Signals, B Company

5 Field Ambulance.

Divisional Reserve Group (Lieutenant-Colonel F. J. Gwilliam): 26 Battalion, Divisional Cavalry Regiment, 5 Field Regiment, 27 Machine Gun Battalion less 2 and 4 Companies, 32 and 34 Anti-Tank Batteries, 42 Light Anti-Aircraft Battery.

The 28th (Maori) Battalion was relieved in 4 Brigade in the afternoon of the 14th by 18 Battalion, whose place in the Divisional Reserve Group was filled by 26 Battalion from 6 Brigade. The Maoris were then sent back to prepare defences on Alam Halfa, some 13 miles to the rear. Even at this advanced and supposedly decisive stage when Army Headquarters had ordered the complete and final destruction of Rommel's forces, it was still thinking of a reverse and tacitly encouraging the looking-over-the-shoulder policy. Sixth Infantry Brigade, less 26 Battalion, was still at Amiriya.

Orders in the brigades were given verbally before noon and were later confirmed in writing. On the left, Burrows decided to attack on a front of 400 yards with 18 Battalion in the lead, its right flank resting on the inter-brigade boundary. This battalion was also charged with checking the direction of the advance. The 19th Battalion was ordered to move in echelon on 18 Battalion's left, and 20 Battalion was held in reserve to move with brigade headquarters on the brigade axis, roughly in rear of 18 Battalion. The rate of advance was fixed at not more than two miles in the hour. Tools and one day's water and rations were to be carried by each man in the assaulting infantry.

Burrows ordered 4 Field Regiment to be in position on the objective at first light, the move from the start line to be made on a signal from the regiment's liaison officer with brigade headquarters. He retained three troops of 31 Anti-Tank Battery under his own command and placed the remainder, 41 Light Anti-Aircraft Battery, and 6 Field Company less two sections under the command of the field regiment. The two-pounder anti-tank guns, two sections of the engineers, and two platoons of 2 Machine Gun Company were put with the reserve battalion.

Transport was divided into two groups. The fighting vehicles, including carriers and mortar carriers, were to rendezvous at brigade page 228 headquarters at 9 p.m. under the staff captain. The B echelon was ordered to follow 4 Field Regiment to rejoin the brigade at first light. An ambulance car was attached to each battalion, reserve ambulance cars to brigade headquarters, and the remainder of the ambulance company to the B echelon.

On the Division's right, Kippenberger ordered 23 Battalion to deploy on the right, 21 on the left, and 22 in reserve. The forward battalions were instructed to cover a total front of 1000 yards, an order misunderstood by 21 Battalion which attempted to cover a front of 1000 yards by itself. The reserve battalion was ordered to follow the forward units at a distance of 1500 yards, with its leading companies widely deployed to mop up centres of resistance which might be left by the advanced troops. The rate of advance was not to exceed two miles in the hour.

Sixth Field Regiment, with 43 Light Anti-Aircraft Battery under command, was ordered to be ready to step up its batteries and to give supporting fire on the objective with two-thirds of the guns at first light. One troop of 33 Anti-Tank Battery was held with brigade headquarters, the remainder of the battery and the two-pounder anti-tank guns being ordered to move in rear of the reserve battalion. The machine-gun company was placed with the A echelon transport which, under brigade command, was ordered to follow brigade headquarters along the axis of advance. One officer and five other ranks of the engineers were put with each of the battalions.

The concentration of 4 Brigade on a front of 400 yards and 5 Brigade's comparative dispersion was the most marked difference in the respective plans. Neither brigade expected serious opposition in the advance, and both commanders wished to arrive on the objective with their units compact and well in hand. Fourth Brigade, however, was not so compressed as the order might suggest. Literal interpretation of the instructions to 18 and 21 Battalions to rest their flanks on the inter-brigade boundary would have denied the units room for manoeuvre. It was sufficient that they were in contact with each other on the boundary. In the event, 18 Battalion took ample room on its right flank and it had all it was likely to require on its left, where 19 Battalion was in echelon. Similarly, 19 Battalion had plenty of space in which to move, especially to the left.

Lieutenant-Colonel Allen's misunderstanding concerning 5 Brigade's frontage had serious consequences. Instead of occupying roughly 500 yards of the intended brigade front of 1000 yards, he deployed his A and C Companies forward with their sections at intervals of 60 yards, the eighteen sections thus attempting to cover a front of 1000 yards. Even with good light and no opposition, page 229 control by the battalion commander in an advance of over six miles would be difficult with such dispersion. It was likely to be more difficult, and proved to be so, for the platoon and company commanders. The misunderstanding was not revealed until after the battle.

Signals communications, especially in 5 Brigade, broke down in the battle, and their almost complete collapse at vital stages was later severely commented on. The plans may have been over-elaborate and, in the final haste of preparations for the battle, their technical aspects may have been beyond ready understanding by unit and formation commanders. However, they were in keeping with the optimistic spirit which pervaded the Division. Had the battle developed as expected, the signals communications probably would have functioned with perhaps only brief interruptions. The men of the Divisional Signals became involved in the reverses suffered by units and brigades. They were not the cause of the setbacks.

Three types of wireless sets were available at Ruweisat. The battalions had the No. 18 of limited range and doubtful efficiency for inter-unit communications, and the more powerful and reliable No. 11 set for communications with brigade headquarters. Brigade headquarters maintained contact with Division with No. 9 sets. It was also standard practice to link all headquarters by field telephone cable at every opportunity, including short halts.

The No. 11 set was heavy and cumbersome. Normally it was carried on a truck. For portage at least five men were required, and if the carry was a long one they needed reliefs to permit them to keep pace with the unit tactical headquarters. In 5 Brigade the battalions could not spare riflemen as carriers, and as the necessity of perserving silence in the advance to the assault forbade the use of transport, it was decided to leave the No. 11 sets with the vehicle column and rely on the No. 18 sets for unit-brigade signals. Events proved the decision to be wrong.

The cable-laying plans were complicated. Signal section cable-layers, carrying three miles of single earth-working line on a barrow drum, were to march behind the headquarters of the leading battalions paying out a line to brigade headquarters into which unit signals were to plug their lines at halts. This was simple provided the cable-layers kept pace, marched on the correct compass bearing, and the unit signallers knew where to find the line. The line was also to serve as a guide for the brigade signals office. As the office moved forward, the unwanted line in rear was to be reeled in and eventually carried forward to the advanced laying party for their further march to the objective. Concurrently, another cable party was to pay out a line connecting with Divisional Headquarters. A weakness in the plan was the effort to be economical in the use of page 230 telephone cable by incorporating some lines used in the assembly positions and in salvaging cable during the advance.

Ruweisat was to emphasize the need of extreme simplicity in signals communications, and also how easy it is during a long night advance for headquarters and signallers to deviate from plotted courses and become widely separated.

The failure of 21 Panzer Division's attacks on the Alamein Box appears to have convinced Rommel he could do no more. An appreciation of the situation attached to his battle report of 14 July says: ‘The continuous actions of 13 and 14 July proved that the two months of continuous hard campaigning had reduced the strength of the German formations to such a degree that they could not resume the offensive immediately. This fact, and the failure of the Italians, forced Panzerarmee to the decision to go over to the defensive in its present positions until the formations were completely refreshed.’

Rommel, however, did not intend to be wholly passive. He urged 21 Division to continue its attack in the north and, at midnight on the 14th, he ordered Afrika Korps, X and XX Italian Corps, and 90 Light Division to carry out a limited offensive on the southern front on 15 July. His orders noted that in the extreme south the British forces had been pushed back to Qaret el Himeimat, but that ‘there was still an enemy group of about 40 tanks with artillery and infantry at 917 right two.’ These were some of 1 Armoured Division's tanks and New Zealand Division on the start line running north-east from Alam Nayil.

The order then continued: ‘Early on the 15th Afrika Korps, co-operating closely with 20th Corps, 10 Corps and 90 Lt. Div., is to drive this enemy force back, mainly by concentrated fire of all weapons. The line 921–927 [roughly the area Deep Well to Munassib] must be won. Carry out strong recces eastward. Do not advance your main bodies past 900.1 On the southern front the passes to the Qattara Depression are to be mined.’

Afrika Korps describes these projected operations as a feint.2 However, they were not carried out. By ‘early on the 15th’ New

1 900 was one of the German thrustlines from which map references were given at Alamein. The 900 thrustline ran almost due south from the Alamein Box through Deir el Munassib

2 Afrika Korps' diary, 14 July: ‘The C-in-C arrives at battle headquarters of D.A.K. and issues the following orders: “As a feint the whole southern front is to attack at 0830 hours.” ’ At this stage the German southern flank was held by 90 Light Division and 33 Recce Unit, which had been carrying out thrust and parry operations against the small columns of 7 Armoured Division. By the night of 14–15 July, most of 90 Division's battle groups and the Recce Unit were closing on the 900 thrustline and were ready to swing north towards Munassib and Deep Well. This manoeuvre, if successful, would have brought them against the left rear of New Zealand Division.

page 231 Zealand Division's infantry had overrun Brescia's and 15 Panzer Division's forward defences and were on the top of Ruweisat Ridge. Far from resting his formations and strengthening his positions, Rommel had again to gather fragments of his German forces and throw them into the fight in a desperate effort to prevent the collapse of his centre and the destruction of his army.