As the story reached the bemused but doubting Germans, British bombers and British tanks added to the woes of the men in ‘The Hotbox’, and Pienaar was reported to have called Auchinleck and said that if he was to be treated as an enemy he could take Alexandria within 48 hours.

A heavy bombardment hit the South Africans in ‘The Hotbox’ at 4.00pm, and at 4.20pm 30 tanks and infantry of 21st Panzer advanced from Deir el Shein under cover of smoke. Artillery and machine-gun fire from ‘The Hotbox’ and from Robcol on Ruweisat Ridge brought them to a halt, and 1st Armoured Division’s artillery gave support. The Germans tried again at 6.00pm and again at 7.20pm, with 90th Light joining in from the north west. Just before last light both the South African brigadier and brigade major were wounded, and a new crisis developed.

We come here to a parting of the ways between the contemporary written record in the form of the divisional war diary, and events as they were remembered by Dorman Smith

1st Brigade now came under command of what the official South African history calls ‘a makeshift staff’ of a battalion commander and an intelligence officer, who ‘clamoured’ for permission to withdraw, though the divisional war diary says only that at 10.00pm they asked for tanks to help ward off a threatened enemy attack. Pienaar, equally concerned, called up the 30th Corps commander, Norrie, and told him the brigade’s flank was wide open and its position untenable unless flank protection from armour could be provided or some armour placed under command. He said bluntly that if help was not forthcoming he would pull the brigade back. He would not, he said, allow it to be overrun.

An unsympathetic Norrie said tanks couldn’t be provided, and he saw no need to withdraw the brigade. Pienaar thereupon moved up the ladder to Dorman Smith, and from him to Auchinleck. Diplomatically, Auchinleck said he would talk it over with Norrie, and for the moment that was that. Not that it was of any comfort to the temporary 1st Brigade commander, anxiously waiting in the desert, who kept asking Pienaar for a decision, pointing out that if a withdrawal was to be carried out it should be done at night, rather than in daylight.

Around midnight Norrie came back on the line with what to him might have seemed a good compromise; 1st Brigade could be pulled back and placed in reserve, but another unit would be put in its place. At this implied slur Pienaar bridled and said that if this was to be done he would regard it as a sign of a lack of confidence and he would be forced to ask to be relieved of his command. All he wanted, he said, were some tanks or, failing this, permission to move the brigade to a position further back.

Midnight is the time when decisions are made by attrition, and a weary Norrie said Pienaar could use his discretion as to where he moved 1st Brigade, but that in any case a column of the 3rd Royal Horse Artillery would be put in its place, or just to the rear of where it had been. Pienaar pulled back 1st Brigade to a less hazardous position. At dawn South African armoured cars took a look at the old 1st Brigade positions, found some 90th Light men there and took them prisoner. At 9.30am the 3rd RHA column moved into The Hotbox, but not for long. After receiving 30 minutes’ heavy shelling, it withdrew and settled down in a position to the south and a little to the rear of 2nd Brigade, where it continued to suffer from enemy shelling. South African honour was satisfied.

Dorman Smith remembered the incident differently. As he later told the story, Pienaar rang him late at night and said he intended to withdraw his division to avoid encirclement. Dorman Smith told him in no uncertain terms that he would have to stay where he was, and Auchinleck added the weight of his authority to this. Later, according to Dorman Smith, he spoke with Norrie, who said he had had a similar conversation with Pienaar.

It would have to be said that the South African version of events is more likely to be accurate, and accords with what actually happened, and the Dorman Smith version is mentioned only because it has gained some currency. Either Dorman Smith’s memory is at fault – and an instance of this occurs later – or at this late hour of the night the tired Dorman Smith simply misunderstood what Pienaar was saying. And for that matter, Pienaar may not have been entirely clear in his request. If it was a misunderstanding it was another of July’s small tragedies, because the incident further undermined confidence in the South Africans, who were left to fight in their static position all month, and when plans were made for the hoped for pursuit of the Axis forces they were assigned the sedentary task of staying behind to man the Alamein positions.

All in all it turned out to be much ado about nothing, but it was a pointer to the cross-currents and personality conflicts that bedevilled the Eighth Army right up to the closing days of the ‘old’ army.

So, looking at all that, how can it be said that one army had attacked another without the defenders really noticing, particularly when it is remembered that 1st Armoured Division and Royal Artillery had been going hammer and tongs with the panzers on Ruweisat Ridge from late afternoon? The short answer may be that despite some fairly violent exchanges, the German thrusts may have had the appearance of probing rather than attacking.

Consider the conditions, with heat haze and dust obscuring observation and objects swimming into and out of view.

In the far north the Italians were supposed to attack the west face of the Alamein Box, but the South Africans saw no great concentration of troops, merely groups who disappeared when fired on.

In the area of 90th Light’s dawn attack, troops of two South African brigades would be standing to as the sky became light behind them and in a few minutes had flooded the desert with the full glare of day. Not much heat haze yet but probably a fair amount of dust, and out of the dust emerges some enemy transport. Nothing massive. Not the menacing force covering half the desert that usually precedes a German attack. Just a largish gaggle of trucks. As the shells fall among them, the enemy lorries turn about and disappear again. Then figures of men are seen. Shell bursts balloon around them and the Vickers guns chatter, and the figures disappear, too. After a while, the British artillery, denied visible targets, gives up.

Though the air force is fairly busy, shuttling back and forth across the sky and creating havoc at unseen targets somewhere west, the rest of the day is fairly quiet until around 3.30pm, when incoming shells suddenly begin to blossom. The South Africans become watchful and wait. Outgoing shells moan overhead with a hollow roar, the sound diminishing in comforting reassurance that each projectile is aimed the other way. Dust begins to rise out there and moving figures are seen. The eager Vickers gunners let go a few belts. The Hotbox receives a pasting from shell fire, but no tanks or infantry advance towards them.

Further south, British guns and tanks sight the familiar outline of panzer turrets – ten, twenty, maybe forty of them, not the usual hordes that intimidate the infantry and terrorise rear echelons. Turret covers clang shut and the single band radio communication network buzzes with orders, requests and information. Everyone can hear so everyone knows what’s happening. And then it’s all on. Solid shot flies, bouncing off hulls or plunging a white hot lance through toughened steel and into explosively receptive shell racks and petrol tanks. A tank glows and smokes as the men inside cook, and then a blast tosses off a turret, which falls to the ground beside the blackened hull like some grotesque, huge egg cup. Other crews are luckier. Their tank slews and stops, and they know they’ve lost a track. The hatch is thrown open and they clamber out with frantic haste, and run for dear life.

There’s something odd here, though. The Jerries don’t seem as aggressive as usual, not quite as pushy, not really wanting to come through. They usually fight as though they own the place. They don’t seem to be trying too hard today.

And then at last the sun goes, a red ball that slips with visible haste behind the western horizon obscured by the dust of battle. There’s little twilight and no cool of the evening, just a rush from day to night as though darkness, the only decency left, should not be delayed.

Tanks still burn, and occasionally there’s an eruption from within. The survivors move away and form into laager and signal their supply columns to bring up more fuel and ammunition.

And what was it all about? Where were the Jerries going, if anywhere? Were they just testing, looking for a weak spot?

At his spartan headquarters, Auchinleck writes a review of the day’s events, and tells London that ‘the expected attack had not developed by last light though some enemy tanks were seen’. And the truth is that the German thrust was not what the Eighth Army was used to.

Timid infantry and tentative armour were not the stamp of the German army, not the bold assault expected of a general reaching for Cairo.

Rommel’s perception was rather different and his optimism less assured.

He wrote of the British falling back in the south and then launching a heavy counter-attack on his open flank, and of ‘violent defensive fighting’. He still hadn’t given up, but his orders for the next day, 3 July, were more restrained. He instructed his army to attack from daybreak to 10.00am to seek out weak points – an order to probe and search. Not his style at all, though he did not abandon his southern encirclement plan and he ordered Ariete and Trieste to carry out the drive to attack 13th Corps in the rear. But if there had been little of 13th Corps in the first place, there was even less now as the Eighth Army, preparing for the next day, carried out some minor reorganisation. The Indians in their remote outpost at Naqb Abu Dweis in the far south were doing no good there and were totally exposed. Auchinleck sent them back to Qarat el Himeimat to reorganise as a battle group under 5th Indian Division. He felt, also, that Kaponga Box might now become a liability, and the New Zealanders were instructed to withdraw, leaving a battle group to hold out for another day to destroy the defences and stores. Some of the brigade pulled back under cover of darkness, leaving two battalions in charge. Auchinleck wanted 6th Brigade infantry sent back to the New Zealand base camp at Maadi because under his battle group policy he considered them ‘surplus’. But a cautious Inglis decided to keep them with the division until the situation became clearer.

One day remained in this first attempt by Rommel to break through the Alamein defences – a day in which the pattern of battle changed ominously for Rommel, hopefully for Auchinleck.

It was on this day, 2 July, that it had been planned to despatch the first troop convoy from Taranto to Alexandria to make secure the Axis occupation of Egypt.