i

THE closing of the gaps in the perimeter minefield in the early hours of 5 September marked the ending of Operation beresford. In the reconnaissance patrols immediately preceding the action, in the diversionary raids, and in the operation itself, some 80 New Zealanders were killed or died of wounds, 300 were wounded and 60 taken prisoner. Though 28 Battalion among the infantry suffered most, with nearly 100 casualties, mostly wounded, 5 Brigade as a whole came off more lightly than 6 Brigade. The losses in 132 Brigade were more serious. The two Royal West Kent battalions lost about 250 men each, the Buffs approximately 100, and the attached troops, including the Valentine crews, lost close on another 100, to bring the total of English casualties to nearly 700, of whom over 200 were listed as missing or prisoner.

Freyberg is on record as having apologised to his corps commander for the losses in 132 Brigade. He plainly felt his responsibility, partly for not having been firmer with Brigadier Robertson over the planning, partly for having minimised the likelihood of strong opposition in the first phase of the advance, but more particularly for having encouraged the use of the brigade in place of one of his more experienced New Zealand formations. For the last point, he knew, none better, of the parlous state of the reinforcement pool, that his men were due for, and expecting, relief and that too hard a knock might eliminate the Division from its anticipated role in Montgomery's corps d'élite. Over the planning, both he and Kippenberger had offered their wisdom and advice, but many of the New Zealanders detailed to help ‘indoctrinate’ the Kentish brigade found that, though personal relations became extremely cordial, there was a resistance to suggestions that the brigade was inexperienced in anything except page 166 the terrain, a natural resistance perhaps as the two Kentish regiments from which the brigade was drawn had distinguished records and many of the senior officers wore decorations gained in battle.

The brigade had been shown the position, estimated from reconnaissance and observation, of the enemy's line, and in fact it met this line fairly close to the point where it should have been expected. The defences were not continuous but consisted of groups of weapon pits and sangars, well spaced out both frontally and in depth according to the evidence of 26 Battalion's C Company. There is nothing to suggest that the enemy was specially prepared for an attack on this narrow sector any more than on any other sector this night, or that the fire met by 132 Brigade was anything more than the enemy's standard defensive fire. General Freyberg's responsibility in the brigade's debacle rests therefore on his failure to appreciate the degree of its inexperience. Earlier, in its occupation of the southern front of the box and in its move to the north-western sector, the brigade had shown an extreme rigidity in control and a consequent lack of initiative among its junior commanders and men, faults which could not be removed by a few days of holding unthreatened defences or by well intentioned attempts at indoctrination. These faults showed up when the brigade was passing through 25 Battalion's sector and the minefield gaps; individuals and groups, isolated in the darkness and confusion from their normal chain of command, ‘flapped’, to use the army vernacular. On the start line the rigid control was re-imposed but brought with it the necessity for an advance in close order instead of the wide dispersion which is only possible with troops trained to take the initiative and act independently. The effect of the enemy's defensive fire on the close-packed brigade was to break up the chain of command again.

The defences met by 132 Brigade were manned by Folgore Division parachutists, who proved later to be some of the best of the Italian troops in the desert. They were stiffened by a sprinkling of German parachutists of Ramcke Brigade, and these two specialist groups, undisturbed in their defences since the first day of the battle, may have offered stronger resistance than that met by 28 Battalion, whose opponents were Italian infantry of Trieste Division, supported by German anti-aircraft gunners and elements of the Kost Group (mainly 155 Infantry Regiment) of 90 Light Division, including a detachment of 900 Engineer Battalion. The Trieste troops had moreover only just taken over in Munassib from 90 Light Division, so that they were probably unfamiliar with the defences there. Word of the fighting was received about midnight at the headquarters of 21 Panzer Division, then some three miles page 167 south of Munassib, and at the headquarters of Africa Corps some three to four miles further south. The Corps immediately ordered 21 Panzer Division to ensure that it made contact with the Kost Group in Munassib, on which the division was swinging back. The divisional reconnaissance company managed to find some isolated anti-tank gunners in the depression but the attacks, first by the Maoris and then by the Valentine tanks, seem to have disorganised this stretch of the defence until after dawn. No reports appear to have been sent by Trieste Division, possibly because the headquarters of the two units concerned, 65 and 66 Infantry Regiments, were in the area overrun by the Maoris. Even the next day the Italians had only a vague idea of what had happened as their casualty report of 103 killed, wounded and missing makes evident.

Information sent direct to Panzer Army Headquarters from the Folgore-Ramcke front on the engagement with 132 Brigade brought urgent orders for the detachment of 15 Panzer Division's tanks, then in bivouac west of Alinda, to stand by ready to support Ramcke Brigade. In recording this part of the action the following day, the Panzer Army had acquired the information that the attack was made by about thirty tanks and a battalion of infantry, of whom 200 were taken prisoner, including the commander of 6 New Zealand Brigade.¹ The only really detailed account of the enemy side of the fighting lies in the diary of 135 Anti-Aircraft Regiment whose story has already been mentioned.

Apart from a certain amount of flurry in Panzer Army Headquarters when piecemeal reports of the night's fighting were arriving, Operation beresford hardly affected Rommel's plans of withdrawal. On visiting 21 Panzer Division on the morning of 4 September, he stated his feeling that a threat of attack from the north still existed, directing the division to remain where it was, that is immediately to the south of Muhafid, in a counter-attack role. When he learnt a short time later that Italian armour covering Himeimat failed to receive his instructions to remain in position until 21 Division was free to relieve it, and was already on the next leg, he let the whole withdrawal continue as originally planned. By the evening of the 4th, any slight dislocation caused by beresford had already been ironed out.

The rather meagre German reports made later on this action contain several points of interest. Rommel could not understand Eighth Army's inertia and took the lack of aggressive action to mean that the British were conforming to his criticism of slow and cumbrous reaction. By the 3rd, he felt that the British command had had time to show some reaction, and concluded from reports by air and ground observation that slightly increased activity in the New Zealand Box area might presage an attack from the north. Kesselring, anxious to regain some of the lost popularity of his air force, offered to lay on a heavy blitz on the box. This air activity, admitted by German ground troops to have been only a weak copy page 169 of what the Royal Air Force had done to them, was thought to have smashed a large concentration of troops of 10 Indian Division intended for an attack on Brescia Division, that is, on the Deir el Angar and Qattara Box sector.¹ The idea that 10 Indian Division was somewhere in the southern and eastern parts of the New Zealand Box persisted in German appreciations well after Operation beresford was over.

From prisoners the Germans acquired the notion that two brigades of 44 Division and 6 New Zealand Brigade had attacked the Folgore – Ramcke front, and on 90 Light Division's front they had captured ‘some members of the 28 Maori Battalion. They were reinforcements just arrived up with the battalion from the Sixth Reinforcements’²—rather late arrivals as the Sixths reached Egypt more than a year previously. From captured documents, the units of 132 Brigade were identified in some detail and placed ‘probably in the New Zealand Division's area’, but Italian interrogation managed to get an erroneous picture of the rest of 44 Division's dispositions by placing 131 Brigade ‘in the front line beside 132 Infantry Brigade’ and 133 Brigade still in the Delta area.³

Of the changes in the British command, interrogation of New Zealand prisoners confirmed the news that General Gott had been killed in an aircraft shot down by a German fighter, and also elicited the strange information that Auchinleck had been appointed Commander-in-Chief, India, and that Wavell had been moved to a higher command: a remarkable example of prescience as these two appointments were not made until June 1943.⁴

ii

For the New Zealand troops, the few days following beresford brought a test of morale. Rumours of relief had been rife towards the end of August, for much of the story of the Division's future role had filtered down to the man in the line. Internal rearrangements, especially the exchange of sectors between 5 and 132 Brigades, seemed to show that relief was not far off when Rommel struck. When beresford was over and 151 Brigade was seen to settle down in 5 Brigade's old defences, hopes revived.

Three of 5 Brigade's battalions remained in reserve in temporary bivouac in the open desert with nothing much to do except tidy up after the battle and watch the dogfights in the air above. The fourth unit, 22 Battalion, settled back into its old area facing east, with little expectation of being called on to man the defences. While here the battalion was unfortunate in losing its popular commander, Lieutenant-Colonel Russell, who was killed in a mine explosion as he went to help the casualties among the crew of a carrier which had been blown up in a minefield.

The heat remained excessive in the early days of September, while the fighting had left many corpses, dismembered and unburied, to swell the plague of flies. In spite of all efforts towards hygiene within the lines, of fly traps and veils of mosquito netting, no man in daytime could avoid his attendant swarm of flies, insistent for moisture from mouth, eyes, or sweat-damped clothes.

The enemy gunners, now relieved of the need to conserve ammunition as strictly as before the battle, followed Rommel's orders to keep the Eighth Army under constant harassment to deter any interference with the Panzer Army as it settled back into defence. They laid regular and methodical strafes on key areas, from mortar and anti-tank gun positions, platoon and company headquarters, in the forward defences back to battalion and brigade headquarters and the gun lines. The cookhouses in the company areas of both 18 and 25 Battalions were subjected with Teutonic punctuality to salvoes of shells at normal meal hours. The ration parties, coming in to collect their dixies of tea and stew, soon learnt to disperse and go to ground until the customary number of shells had fallen, so that few casualties occurred. The enemy shelling reached its peak on 6 and 7 September, on which two days it was almost continuous. From then on it gradually declined. The accuracy with which this shellfire fell on many previously undisturbed targets was attributed by many of the men to marked maps carried by prisoners taken by the enemy, but it may of course be possible that more targets were engaged merely because more shells were fired.

The Axis air forces, as if to make up for the lack of ardour complained of by the ground troops, also put out a burst of activity at this time. Several Stuka raids were directed at the New Zealand Box, one on the 5th bringing some fifteen casualties and considerable vehicle damage among undispersed troops and trucks in the headquarters area of 151 Brigade, and another two days later causing some damage but only minor casualties among the headquarters of 6 Brigade and 25 Battalion. On the 9th, D Company of 25 Battalion lost two men killed and five wounded from bombing. Several times in between these raids, flights of dive-bombers came over but were forced to jettison their bombs and run for home as British fighters appeared, the Kittyhawks usually engaging the covering Messerschmitts while the slower Hurricanes tried to pick off the Stukas. These German dive-bombers had by now lost most of the terror they held for British troops earlier in page 172 the war; in fact, the men were inclined to admire the Stuka pilots' bravery in running the gauntlet of the Desert Air Force's fighter patrols. Of one flight that tried to raid the area east of the box, almost all were shot down either by the fighters or, as they tried to seek safety by flying close to the ground, by the Bofors gunners.

Patrolling by 6 Brigade's battalions followed the pattern learnt by experience before beresford. No major raids were made to get prisoners, but each night seven to eight patrols of varying sizes were out reconnoitring new Axis defence diggings and harassing working parties. On the evening of the 5th, men listening to radios in the Division heard Berlin announce the capture of Clifton. With this definite news, General Freyberg decided to place his GSO I, Colonel W. G. Gentry, in command of 6 Brigade and called Lieutenant-Colonel Queree¹ to take Gentry's place.

In spite of the trials of these few September days, morale in the New Zealand Division was extremely good. The lack of worth-while results from Operation beresford and the loss of comrades in the fighting were offset by the gradually emerging details of the complete failure of Rommel's offensive. Confidence in the higher command of the desert war reached a point unknown for many months. The trying conditions combined with the Axis shellfire to keep the medical staff busy. The occasional battle casualties were well outnumbered by the steady flow of men falling sick with malaria—a hangover from the stay in Syria—infective hepatitis or jaundice, dysentery and desert sores.

Finally 5 Royal West Kents, completely lost for a time in the same dust-storm, appeared in the lines of 18 Battalion just on midnight. At 1 a.m. on 11 September, when this battalion reported that its relief was completed, 6 Brigade handed over command of its sector to 132 Brigade and the official relief of 2 New Zealand Division by 44 Division, arranged for the evening of the 10th, could be ratified. For the New Zealanders this ended what has been well described as the hard summer of 1942. The last casualty for the Division occurred when the General's staff car, travelling back along the coast road, collided with a heavy truck. The General, who was only just recovering completely from his Minqar Qaim wound and its complications, sustained a cracked rib.

iii

From the time the New Zealand Division vacated the ground occupied in Operation beresford on the evening of 4 September, the battle of Alam el Halfa can be said to have faded slowly to a finish. Hardly affected by any of Eighth Army's actions, Rommel's planned withdrawal and the deployment of the formations of Panzer Army continued until, by the morning of the 8th, all but a few patrols were behind the line along which Rommel had decided to set his new southern front. Italians of 10 Corps with a detachment of Ramcke Brigade extended the old front from the Qattara Box to Munassib, Trieste Division remained on the bend in the line on the east of Munassib, and from this point southwards to beyond Himeimat Littorio and Ariete divisions, with detachments of Folgore parachutists, were straightening out the old ‘second’ minefield and digging in behind it. The three battle groups of 90 Light Division stayed in reserve in the area of the bend, while the three German reconnaissance units and a tank detachment from 21 Panzer Division watched the southern leg as the Italians dug in. The main bodies of the two panzer divisions pulled back well to the west and sent battle groups to support the northern and central sectors of the main front. Thus the territorial gains of the Axis amounted to the occupation of the area from the Qattara Box, through Munassib and on to Himeimat, an area previously the playground of 7 Armoured Division's light columns.

As the Panzer Army fell back to these positions, the British heavy armour was kept by Montgomery's instructions close to its chosen positions west and south of Alam el Halfa. All three armoured brigades, however, made up light columns of Crusaders, Valentines or Stuarts with scout cars and field guns, which kept in contact with the enemy's rearguards by day and pulled back to laager and replenish at dusk. Several of these columns had minor engagements, particularly at dawn as they felt forward to test how far the enemy had retreated during the night. Further to the south the mobile columns of 7 Armoured Division found themselves too weak and the going too rough to ‘hustle’ the enemy as Montgomery had directed. As the Panzer Army fell back round the south of Muhafid, the contraction of the front brought much overlapping of the activities of these columns, of whom there were at least a dozen, many split into smaller patrols and all operating practically independently on a front little more than ten miles wide. The lack of co-operation, limiting their total effective striking power, permitted these columns to advance only at the pace of the German retreat.

By the evening of 4 September both air and ground observation had noted a decrease in the Panzer Army's movements. This was taken, by the more sanguine of Eighth Army's commanders such as Freyberg, to mean that Rommel considered he was out of danger and could occupy his defences at leisure. To the more cautious Montgomery it carried a threat that Rommel might have finished regrouping ready to turn at bay and renew the offensive. He caused orders to be sent to 4 Light Armoured Brigade to ‘harass the enemy's southern flank with all available troops and, if forced to withdraw owing to enemy infiltration round our flanks or by an enemy attack in force, to impose the maximum delay….’ He also persuaded the Desert Air Force, which was beginning to feel the strain of the heavy bombing programme, to lay on a two-hour raid on the concentration of vehicles observed at dusk in Munassib. This raid, on the evening of the 4th, was the last major air effort of the battle. Next morning the Panzer Army had begun to thin out and disperse its vehicles as the troops took up their final defensive positions, so that good bombing targets became progressively more difficult to find. And, as if ashamed of the Luftwaffe's poor showing in the battle, German fighters began to appear in increasing numbers, both to cover the Stuka raids intended to keep the Eighth Army from thoughts of counter-attacking while the Panzer Army settled down, and to drive off the British bombers and reconnaissance aircraft.

On 5 September the Valentines of 23 Armoured Brigade were allowed to make a cautious advance round the south-eastern corner of the New Zealand Box. By the late afternoon one regiment had joined the New Zealand Divisional Cavalry patrols who were observing the enemy from Point 100. The sight of enemy tanks in the distance to the south-west and fire from anti-tank guns kept the Valentines, acting on orders not to become engaged, hull down behind the point. Encouraged by the presence of the heavier tanks, some of the Divisional Cavalry's Stuarts worked their way forward into the eastern end of Munassib. Horrocks, on learning of this, asked Freyberg if he could move troops out again during the night to occupy the ground as far as Munassib, but Freyberg replied that such an operation was out of the question. His stand was that any attack on the Munassib area, save as a fully prepared operation with strong mixed forces, would have little value. The area on which, in his opinion, minor threats of attack might cause the enemy uneasiness and bring him to hasten his retirement was the Qattara Box – Deir el Angar front and he offered to stage raids here. Also, feeling sure that Rommel's withdrawal was permanent, he suggested that, with two extra brigades now in the New Zealand Box, the relief of the Division might begin. Horrocks, presumably after discussions with the Army Commander, refused the proposal for a raid and also stated that the time was not yet opportune for moving the New Zealanders, but that the congestion in the box should be relieved by the despatch of 132 Brigade to its parent formation, 44 Division, on Alam el Halfa, for reorganisation.

That evening the following message from Montgomery was received by the formations of Eighth Army:

The battle … has now lasted for six days, and the enemy has slowly but surely been driven from 8 Army area. Tonight, 5th September, his rearguards are being driven west, through the minefield area north of Himeimat. All formations and units, both armoured and unarmoured, have contributed towards this striking victory, and have been magnificently supported by the R.A.F. I congratulate all ranks of 8 Army on the devotion to duty and good fighting qualities which have resulted in such a heavy defeat of the enemy and which will have far-reaching results.

From this time on the Army Commander's orders were based on his assumption that Rommel would return to the defensive in the positions held before the battle. Horrocks was urged to regain the ground up to the original first minefield, using 7 Armoured Division and the light columns from 10 Armoured Division, with the New Zealand Division pressing from the north to assist where possible. Air reconnaissance had reported that detachments of page 176 tanks had been seen moving across the rear of the enemy's front from south to north and, as 30 Corps lacked armoured support, Montgomery ordered 23 Armoured Brigade to move into the northern sector.

The departure of the Valentines left the New Zealand Divisional Cavalry in sole control of the Point 100 ridge, on which any movement was still heavily shelled by the enemy. It was at this time that Lieutenant Ormond,¹ working his Stuart tank forward of the ridge and into the Muhafid Depression, came under fire from two light anti-tank guns set behind a minefield. Noticing wheel tracks through the field, Ormond reconnoitred on foot, to discover a cleared gap. He then returned to his tank and directed it through the mines. In a spirited charge against determined opposition, he disposed of both guns and their crews.

On 7 September Montgomery officially called the Alam el Halfa battle off. He decided that, in order to commence at once the major reorganisation of Eighth Army needed for his own offensive, he would be unable to spare enough troops to reinforce the columns of 7 Armoured Division so as to push Rommel back to his original line in the south. He accordingly ordered that the Panzer Army should be left in possession of the minefields from Munassib to Himeimat and that the ‘third’ and ‘fourth’ minefields in this area should be developed as 7 Armoured Division's new front. He also planned for the New Zealand Box to be extended to include some of the unoccupied ground to the south.²

In order to retain sufficient armour for support and counter-attack on both corps' fronts, Montgomery rostered his armoured formations so that each brigade in its turn could be relieved for training and re-equipping. Not many miles to the west, Rommel, only slowly losing his suspicions that the Eighth Army might eventually counter-attack, was planning a similar reorganisation but a step behind Montgomery. While the Eighth Army was preparing for offence, Rommel knew that, unless he could win the battle with the strategists in Germany, his next role would be defensive.

During August the Royal Air Force in the Middle East had been reinforced with some 181 bombers and 254 fighter aircraft, mainly Bostons, Wellingtons and Hurricanes Mark II. It had also received a token force of Mitchells (B25s) and some Kittyhawks (P40s) manned by the United States Army Air Force. There was as well in the theatre a force of Liberators (B24s), originally destined for China but held in the Middle East for the bombing of the Ploesti oil wells in Rumania. From June on, these Liberators had attacked several targets in Europe as well as the Italian fleet, and had assisted Royal Air Force Wellingtons against Tobruk and Benghazi. The medium and light bombers had been mainly directed at Matruh and the rear of the enemy's Alamein positions, especially the airfields and landing grounds around Daba and Fuka.

On the evening of 21 August the air effort was stepped up with the intention of disrupting Rommel's expected offensive. The positions on the Alamein line itself, forward installations and landing grounds, and the coastal road and railway were all among the targets. On the night of 23–24 August, 10 flare-carrying Albacores led 41 Wellingtons over the El Mreir sector, and on the following night but one, 52 aircraft, including 9 USAAF Mitchells, made the main Stuka base at Daba so unusable that it was taken some 60 miles back. Lesser raids on the intervening nights led up to the opening of the Axis offensive on the evening of 30 August, which coincided with another major effort, an attack by 44 aircraft, including USAAF Liberators, over Tobruk. The next night, and again on the night of 2–3 September, the heavies gave their attention to the Axis landing grounds, trying out two 4000 lb bombs. Throughout the Axis advance, the flare-carrying Albacores and the Wellingtons concentrated on the line of the depressions south of the New Zealand Box, also experimenting with 4000 lb bombs. During daylight the fighters, fighter-bombers, and light bombers maintained the offensive, with peak activity on 3 September when a shuttle service of Bostons, Baltimores and Mitchells (‘The Boston Tea Party’) flew from dawn till dusk over the retreating page 178 Panzer Army. The fighters on this day carried out some 200 patrols. From then on, targets were harder to find as the enemy dispersed his transport and commenced digging in, while heavy dust-storms became more frequent. On 5 September Montgomery and Coningham, the AOC Western Desert, agreed that the air offensive could now be called off.

The fighter cover provided during the battle was so constant that, while the Boston Tea Party flew almost unmolested over the Axis columns, few Axis bombers ventured over the Eighth Army's lines during daylight, except for occasional hit-and-run raids by heavily escorted Stuka formations. Against these dive-bombers the tactics evolved were for the USAAF Kittyhawks (the RAF P40s were fitted with bomb-racks and used as light or fighter-bombers) to engage the fast Me109s while the Spitfires and Hurricanes dealt with the Italian MC202s and the Stukas. In this way the slow and vulnerable Stukas, once so dreaded, were faced with such odds that their pilots would jettison their bombs, more than once on their own troops, and turn for home as soon as the Allied fighters were sighted. In the light of the number of aircraft employed and sorties flown, the total loss of Allied air forces was not unduly high.

The strategy of attrition employed over the previous weeks by air action against enemy ports in Europe and North Africa and, in conjunction with the Royal Navy's submarines, against shipping at sea paid greater dividends than were anticipated once the land battle was joined. With petrol down to a mere trickle, the Panzer Army found that not only was its cutting edge blunted but that all its movement had to be constricted. Wide dispersion of trucks and tanks cost petrol that could not be spared. Supply vehicles following the advance and the whole army in retreat were forced to use the short, direct routes through the depressions rather than risk greater consumption of petrol on more widely dispersed routes further to the south. So, as the Eighth Army sat stolidly on the defensive, the Allied bombers took the offensive against targets almost made to order. With the strong fighter cover that allowed the Boston Tea Parties to fly virtually unmolested by Axis fighters, and the flares from the Albacores that turned night into day, the Desert Air Force took full advantage of the Panzer Army's predicament. ‘The continuous air attacks resulted in a noticeable decrease in the efficiency of commanders and troops (no sleep, continuous waiting for the next bombs to fall, scattering of units, etc.,)’ is a typical example of the comments in the German reports, page 179 in which the Africa Corps' losses by air attack alone were estimated as over 100 men killed, 300 wounded, 170 vehicles (including one tank) totally destroyed, and 270 (including two tanks) damaged.¹

iv

The battle of Alam el Halfa lasted in effect less than a week. The Eighth Army's commander, his own referee, claimed a victory, but in losses inflicted and sustained the British forces could justly claim only a bare win on points. The army started the battle on 31 August with a grand total of 935 tanks, of which 693 were in the forward areas ready for action. By 8 September at least 100 tanks had become unusable, principally from enemy action, but many of these were repairable, and the final total of loss seems to have been between 40 and 50.² Of the Panzer Army's initial muster of 233 German and 281 Italian tanks, the two panzer divisions had 113 seriously damaged, of which 38 were abandoned in the withdrawal, while the Italians reported 11 of their tanks lost but left no record of those damaged.

In men, the British suffered at least 1800 casualties. This figure is made up of 1600 in actions of which records are available, with an estimated 200 to cover the casualties in the mobile columns and patrols and under bombing and shellfire.³ Against this the Panzer Army left a record of 1859 Germans and 1051 Italians killed, wounded, or missing, a total of 2910. The Axis lost also about fifty guns of 47-millimetre calibre and upwards and some 400 transport vehicles, while the number of vehicles damaged must have run over the thousand mark.

The diary of 135 Anti-Aircraft Regiment, a Luftwaffe formation attached to the Panzer Army, carried an entry on 2 September, after noting receipt of the orders for the first stage of the withdrawal, that

…The probable reasons for this decision were:—

On 7 September, when the battle was over, the diarist saw fit to alter his opinion:

In connexion with the diary entries for 2 September it now appeared that the main reason for the withdrawal of our forces was not the enemy's air superiority or the petrol shortage. The main object of the operation had been to force the enemy to come out and come to grips in an open tank battle. The enemy had not accepted the challenge but fought a delaying action and gained enough time to regroup. If our assault divisions had pushed on farther east, their right flank would have advanced past the cover afforded by the Qattara depression and the plateaus north of it, and would have become vulnerable. Enemy forces had concentrated south-east of Deir el Risw, and an enemy recce force had made a raid on the supply track. This gave rise to the conclusion that the enemy would have let the assault divisions push on east almost unopposed and would then have attacked the line of communication from north and south.

This supposition is supported by a remark of Field Marshal Rommel's to Field Marshal Kesselring on 1 September, ‘The swine isn't attacking’.

The operation was now broken off. It had cost us heavily in men and equipment with very little to show for it, and the formations had given evidence of some exhaustion and lack of drive. The morale and material effect of the enemy air attacks were mainly responsible for this.³

This opinion represented in essence the story that both sides were content to have commonly accepted. On the Axis side, embellished by the addition that Rommel had originally intended the operation as merely a ‘reconnaissance in force’, it saved Rommel's face and gave his troops an excuse for their ‘Six-day Race’.¹ For the British, it justified Montgomery's policy of caution.

Rommel had not previously been on the receiving end of massive and continuous air attack, but he was quick to realise what the British had learnt by bitter experience, that battle tactics are limited against an opponent who has command of the air.² This lesson was driven home more firmly by the shortage of petrol for, had both the fighting and supply troops had greater freedom of movement, the effect of the air attacks would have been consequently less. An operation planned for a mobile force has little chance of success if that force no longer possesses mobility. Once the two panzer divisions showed their inability, through petrol shortages, to co-operate in their attacks on the Alam el Halfa box and 22 Armoured Brigade, Rommel knew he was no longer in a position to exploit success, even if success came his way. The British perimeter had not been breached, no supply dumps of any consequence overrun; nothing of value had been, or was likely to be, ‘got from Tommy’. So long as the Eighth Army remained intact and showed no sign of aggressive action against the spearhead, there was the fear that it would turn its attention to the static front; then, not only would the necessary detachment of armour in support use up more petrol than could be spared on the long journey back from Alam el Halfa, but the defences might not be able to hold should the British unleash their unexpected air power to pave the way for a ground assault.

It was in fact lack of petrol and lack of fighter cover that blunted Rommel's last offensive in Egypt. It is to his credit that he withdrew his forces in such good order, preserving an apparently united and threatening front that deterred interference. Like so many of the Axis operations, Alam el Halfa was a gamble, but unlike earlier successes it was a gamble that did not come off.