'Blau' (iii) was a German unrealised plan for a January 1944 retreat by Generalfeldmarschall George von Küchler’s Heeresgruppe 'Nord' from its investment line outside Leningrad westward more than 150 miles (240 km) to the natural defensive barrier formed by the Narva and Velikaya rivers and Lakes Peipus and Pskov in the 'Panther-Stellung' (late 1943).

Adolf Hitler expressly forbade this pre-emptive undertaking, and the Germans were then compelled to make this retreat under altogether more adverse circumstances in the the face of the Soviet 'Leningrad-Novgorod Strategic Offensive Operation'.

During the two years following the Soviet 'Moscow Strategic Offensive Operation' (5 December 1941/7 January 1942), Generalfeldmarschall Wilhelm Ritter von Leeb’s (from 16 January 1942 Generalfeldmarschall Georg von Küchler’s) Heeresgruppe 'Nord' had by comparison with the other German army groups occupied an almost stationary front. On its right Generalfeldmarschall Ernst Busch’s (from 30 October 1943 General Christian Hansen’s) 16^th Army had yielded some ground but had kept the right flank of its line anchored firmly on Lake Ilmen in the north. To the south of this lake two old Russian towns, Staraya Russa in the north and Kholm in the centre, had remained directly in front of the army since the summer of 1941. Even the Soviet breakthrough at Nevel, still farther to the south, during October 1943 was of greater significance as a portent of a possible drive to outflank the army group in the south than for the loss of ground it involved. On the left von Küchler’s (from 16 January 1942 Generaloberst Georg Lindemann’s) 18^th Army had fought three battles in the area to the south of Lake Ladoga to maintain the siege of Leningrad, and had held the Soviets to a token gain of just a few miles along the southern shore of Lake Ladoga. From the Volkhov river to the Gulf of Finland the front was reminiscent of that which had characterised that of World War I inasmuch as it was a patchwork of trenches and shell holes, the result of combat over a period of 30 months in which each side’s the gains and losses were measurable in yards rather than miles. Sited along the south coast of the Gulf of Finland to the west of Leningrad, the German heavy siege artillery transferred north from Crimea in the summer of 1942 after the capture of Sevastopol could still bring under fire all of Leningrad except the north-eastern suburbs.

For more than a year, however, the relative geographical stability of the front had not reflected the real state of Heeresgruppe 'Nord'. In September and October 1943 von Küchler had been compelled to give up three infantry divisions and Generalleutnant Emilio Esteban Infantes’s 250^th Division (spanische) (otherwise the Division 'Azul'), at the same time taking over on its right flank some 60 miles (100 km) of inadequately manned front from Generalfeldmarschall Günther von Kluge’s (from 12 October 1943 Busch’s Heeresgruppe 'Mitte'. As replacements Heeresgruppe 'Nord' had received three recently formed SS divisions with mostly non-German personnel and the Legión Azul, the latter comprising the 1,500 men General Francisco Franco y Bahamonde, the Spanish dictator, had substituted for the Division 'Azul', many of them loyalists who were looking for a chance to desert to the Soviets.

By December 1943 the line around Oranienbaum, the Soviet pocket on the south coast of the Gulf of Finland to the west of Leningrad, was being held by two air force field divisions and two of the SS divisions, and, except for the critical sectors close to Leningrad and to the north of Nevel, the rest of the front was leavened with air force field divisions and SS units newly recruited in the Baltic states. After the Soviet breakthrough at Nevel, Heeresgruppe 'Nord' had deliberately weakened its left flank and centre to strengthen the right flank. In operational terms, the position of Heeresgruppe 'Nord' had become very similar to that of Generalfeldmarschall Erich von Manstein’s Heeresgruppe 'Süd': it had been compelled to divide the bulk of its strength between the two flanks, holding the one (Oranienbaum and Leningrad) mostly for political and prestige reasons and the other (Nevel) to stave off what could become a military disaster.

During the second week of September 1943 Heeresgruppe 'Nord' had embarked on the construction of the 'Panther-Stellung' (Panther position) as its share of the so-called 'Ostwall'. The northern half of the 'Panther-Stellung' was based on natural obstacles such as the Narva river, Lake Peipus and Lake Pskov. The southern half lacked the benefit of such natural obstacles, and indeed had to be stretched to the east somewhat to cover the major road and railway centres of Pskov and Ostrov, and the link with Heeresgruppe 'Mitte' had to be shifted farther to the west after the Soviet breakthrough at Nevel. Nevertheless, when occupied the 'Panther-Stellung' would reduce the army group frontage by 25% and, unlike most of the 'Ostwall', it had by late 1943 actually begun to seem like a real fortified line. A 50,000-man construction force had improved the lines of communication to Riga and Dvinsk, and had built 6,000 bunkers (800 of them of concrete), laid 125 miles (200 km) of barbed wire entanglements, and dug 25 miles (40 km) each of trenches and tank traps. During November and December building materials were arriving at the rate of of more than 100 railway wagons per day.

In September the staff of Heeresgruppe 'Nord' had started to undertake the detailed planning of 'Blau' (iii) as the army group’s withdrawal to the 'Panther-Stellung'. The staff estimated that the million tons of grain and potatoes, half a million cattle and sheep, military supplies and other material, including telephone wire and railway track, to be moved behind the 'Panther-Stellung' as 4,000 train loads. The withdrawal itself would be facilitated by the network of alternate positions which had been built in the preceding two years as far to the rear as the Luga river. The 900,000 civilians living in the evacuation zone, particularly the men who could, if they were left behind, be drafted into the Soviet forces, raised problems. The first attempts, early in October, to march the civilians out in long columns resulted in so much confusion and hostility that von Küchler ordered the rear-area commands to introduce less arduous methods. After this, the Germans singled out the adults who would be useful to the USSR as workers or soldiers for evacuation largely by rail. During the last three months of the year the shipments of goods and people went ahead while the armies worked at getting their artillery and heavy equipment, much of which was sited in permanent emplacements, ready for movement. At the end of the year, having transported 250,000 civilians into Latvia and Lithuania, the army group could not find quarters for any more persons and ended that part of the evacuation.

The army group staff believed that 'Blau' (iii) should begin in the middle of January 1944 and be completed shortly before the spring thaw, much as Heeresgruppe 'Mitte' had completed 'Büffel' in the previous year, but on 22 December Generalleutnant Eberhard Kinzel, the army group’s chief-off-staff, informed the 16^th Army and 18^th Army that Adolf Hitler would probably not order the implementation of 'Blau' (iii) unless another Soviet offensive made it imperative. At this time, Hitler’s feeling was that in the fighting in Ukraine the Soviets had suffered casualties so heavy that they might not essay another major offensive anywhere before the spring of 1944.

Toward the end of the month it appeared that Hitler might in fact be correct. The bulge on the right flank of Heeresgruppe 'Nord' to the west and south-west of Nevel was indeed a cause for concern, but the Stavka had shifted the weight of the offensive farther to the south into the area of Vitebsk, for the time being at least. In the Oranienbaum pocket and around Leningrad General Leonid A. Govorov’s Leningrad Front and General Kirill A. Meretskov’s Volkhov Front had been ready to attack since November, but with the Soviet drive at Nevel now out of the way, Heeresgruppe 'Nord' was less worried than it had been. Intelligence reports from the 18^th Army indicated that the formations and units in the Oranienbaum pocket, in particular, had been strengthened, and also that coastal traffic between Leningrad and Oranienbaum had been usually heavy during the autumn and indeed had continued until some of the vessels had become trapped in the ice. On the other hand, almost no new units had appeared, and the Leningrad Front seemed to be depending for its reinforcements on the population of Leningrad. While an offensive at some time in January appeared probable, the longer the 18^th Army's intelligence staff considered the situation, the more it managed to convince itself that any offensive would be similar to the modest pattern of the three earlier Soviet offensives around Leningrad. On 29 December the Oberkommando des Heeres ordered von Küchler to transfer to Heeresgruppe 'Süd' one of his best divisions, Generalleutnant Ernst-Anton von Krosigk’s 1^st Division, on which the 18^th Army was depending to stabilise some of its less reliable formations in the sector of Oranienbaum and Leningrad. When von Küchler objected, General Kurt Zeitzler, the chief-of-staff of the Oberkommando des Heeres, told him that Heeresgruppe 'Nord' would not need the division, for Hitler intended now planned to execute 'Blau' (iii) and would tell him so personally on the following day.

During the mid-day conference in the Führer headquarters on 30 December, at which he expected to receive his orders, von Küchler reported on the state of the 'Panther-Stellung' and the time he would need to complete 'Blau' (iii), and also remarked that he had spoken with Lindemann, who had 'naturally' asked for his army to stay where it was even though he had been stripped of the 1^st Division. Questioned by Hitler, von Küchler replied that the 18^th Army's front was well fortified, almost too well, in fact, as the army lacked enough troops to man it completely. Hitler then ended the conference without even mentioning 'Blau' (iii). von Küchler did not fully appreciate what had happened until the following day, after he had received an order to transfer another good division to Heeresgruppe 'Süd'. Zeitzler told the army group’s chief-of-staff that Hitler had begun to falter in his decision as soon as von Küchler made the remark about Lindemann wanting to keep his army where it was, and that he thought it would take at least a week to talk Hitler around again. By the end of the day, the chief-of-staff had prepared for von Küchler’s signature a memorandum rehearsing the arguments for 'Blau' (iii)'s advisability, but that was not enough. Lindemann would have to be persuaded to reverse his opinion, as in such instances Hitler tended to take the word of the man on the spot.

On 4 January, by which date a third division was on its way to Heeresgruppe 'Süd', von Küchler went to the headquarters of the 18^th Army and, citing the need to husband the army group’s forces, almost pleaded with Lindemann to reconsider. Lindemann’s response was that his corps, divisional, and major unit commanders in the most threatened sectors were confident they could check any Soviet attack. After that, none of the army group’s arguments carried much weight. Hitler told Zeitzler he was only doing what von Küchler wanted. von Küchler and his staff could derive no comfort from the knowledge that Lindemann was probably motivated by a desire to draw attention to himself, for he had never been offered an opportunity as good as this to show, right under Hitler’s eyes, what he could do. No less disquieting for the army group was the knowledge that it was committed to repeating an error which had already been made too often in the Ukraine: the army group’s chief-of-staff told the operations chief at the Oberkommando des Heeres that the army group was marching toward disaster with its eyes open, putting formations into positions which, ultimately, could not be held.

On 14 January 1944 the Soviets launched their 'Leningrad-Novgorod Strategic Offensive Operation'. Under Govorov’s command, the Leningrad Front made the main effort. General Leytenant Ivan I. Fedyuninsky’s 2^nd Shock Army drove to the east out of the Oranienbaum pocket while General Leytenant Ivan I. Maslennikov’s 42^nd Army attempted to push to the west on the front just to the south of Leningrad.10 Against the 42^nd Army, which was the stronger of the two Soviet armies, the corps artillery of General Wilhelm Wegener’s L Corps reacted swiftly, laying down a well-placed barrage that stopped the attack before it got started. The 2^nd Shock Army fared better, and Generalleutnant Hermann von Wedel’s 10^th Felddivision (L) began to break as soon as it was hit. This was little of a surprise, but the same could not be said of the strong thrusts which Meretskov’s Volkhov Front launched the same day to the north and south of Novgorod on the 18^th Army's right flank. Novgorod had been considered a danger point, but the 18^th Army had not been convinced that the Leningrad and Volkhov Fronts would have the strength to attempt simultaneous offensives on so large a scale. As recently as 10 January, Lindemann had decided that the Soviet build-ups in the Oranienbaum pocket, to the south-west of Leningrad and to the east of Novgorod as relatively modest, particularly in terms of reserves. He had predicted that a lack of reserves would prevent any Soviet thrusts from penetrating deeply, and that the attacks in the sector between Oranienbaum and Leningrad and at Novgorod would probably be staggered in time. In fact, the Leningrad and Volkhov Fronts outnumbered the 18^th Army by at least 3/1 in divisions (55 infantry divisions, nine infantry brigades, and eight tank brigades to a German total of 20 divisions), 3/1 in artillery, and 6/1 in tanks, self-propelled artillery, and aircraft.

The Soviets had selected exactly the two locations in which the 18^th Army had the least room to manoeuvre. In the north the loop of the front separating the Oranienbaum pocket from Leningrad was only 20 miles (32 km) wide at its base, and in the south, in the area of the 18^th Army's right flank, an envelopment as little as 10 miles (16 km) deep was enough to take Novgorod and severe the German link with Lake Ilmen. As Zeitzler warned at the end of the day, the danger lay in the fact that minor mistakes could entail consequences similar to those of the Nevel debacle.

During the second and third days of the Soviet offensive, the battle seemed to be proceeding as the Germans had hoped it would. Neither Govorov nor Meretskov put in any new units, which seemed to indicate that they were operating without much in the way of reserves, and it appeared that Leningrad Front did not intend to do more than lift the Germans siege of the Oranienbaum pocket. On 16 January von Küchler told his army commanders that the Soviets had committed all their forces, and that Heeresgruppe 'Nord' could win the battle by taking limited risks in the quiet sectors. On the following day von Küchler’s optimism started to fade. Lindemann had committed his whole reserve, Generalmajor Joachim Albrecht von Blücher’s 61^st Division, to bolster the 10^th Luftwaffe Felddivision, but it was barely managing to stave off a complete Soviet breakthrough. Before 12.00 the army group informed the Oberkommando des Heeres that the fighting around Leningrad was taking a decided turn for the worse: the 18^th Army would have to begin dismantling the siege artillery during the night, and if the army group wanted to see the battle through it would have to withdraw, in the area to the south of Lake Ladoga to the 'Rollbahn' (main road) position along the route linking Leningrad and Chudovo in order to shorten the front and thereby gain two divisions. The army group had built the 'Rollbahn' position to provide against just the eventuality which was now threatening it. In the afternoon Hitler answered that he neither approved nor disapproved, but thought it would be better to give up the hold on the Gulf of Finland and withdraw the front between Leningrad and Oranienbaum. von Küchler protested that this would give the Soviets the victory and also the opportunity to advance to the south with their strength intact.

During the morning of 18 January, Lindemann reported that the fronts to the east of Oranienbaum and top the west of Leningrad were collapsing. The same was happening at Novgorod, where the encirclement of General Kurt Herzog’s XXXVIII Corps, one of the 18^th Army's four corps, was nearly complete, and the few additional battalions which the army had been able to commit would not be sufficient even to hold open an escape route for much longer. After seeing for himself how near complete exhaustion the troops at the front were, von Küchler asked and was denied permission to withdraw to the 'Rollbahn' position. In the afternoon the 42^nd Army’s spearhead, advancing to the south from Leningrad, drove into Krasnoye Selo, the former summer residence of the tsars, and severed the two main roads to the north. After that, von Küchler decided he had no option but to withdraw the two divisions on the coast of the Gulf of Finland before they were completely cut off. He informed the Oberkommando des Heeres that he intended to give the order at the end of the day regardless of whether or not he had received permission by then. At the midnight situation conference Hitler approved after Zeitzler told him that the order had already been given.

On 19 January the first stage of the Soviet attack, which was obviously only the prelude to the main Soviet offensive, ended. The task now facing the German commanders was the difficult problem of getting Hitler to accept the consequences. von Küchler’s order to withdraw had come too late to save the divisions on the coast: though some elements escaped, others were trapped and destroyed as the Soviets swept forward from the east and west. The 2^nd Shock Army and 42^nd Army then linked forces, and the appearance of several fresh divisions demonstrated that the Soviets did in fact possess more than adequate reserves. At Novgorod eight Soviet divisions encircled five German battalions, whose only hope for escape was to elude the Soviets in the swamps lying to the west of the city.

Soon after the fall of night, once Zeitzler had argued unsuccessfully for half an hour, von Küchler called Hitler and begged him to give the troops at Novgorod what would certainly be their last chance. Suddenly dropping the argument to which he had clung so stubbornly throughout the day, that Novgorod could not be given up because of its huge symbolic importance to the Soviets, Hitler finally gave his assent. On the subject of a withdrawal to the 'Rollbahn' position, however, Hitler lectured von Küchler about the demoralising effects of voluntary withdrawals, but a mere 15 minutes later Hitler called von von Küchler to authorise the movement to the 'Rollbahn' position. At 24.00 Hitler changed his mind about the latter withdrawal, but Zeitzler told him the orders had gone out and the divisions could not be recalled. Hitler had also tried to extract from Zeitzler and von Küchler guarantees that the 'Rollbahn' position would be held. Assessing the position on 20 January, von üchler declared that the tactical setbacks at Novgorod and to the south-west of Leningrad, were the result of a lack of reserves and an over-extended front. The same conditions still existed, however, and the withdrawal to the 'Rollbahn' position would free three divisions, of which two would be committed to the front to the south of Leningrad and the third to the front to the west of Novgorod. With that, von Küchler knew, his army group would have exhausted its capacity for the creation of reserves. The three divisions would be exhausted in a short time, and a Soviet operational-level breakthrough had then to be expected. von Küchler recommended that the retirement to the 'Rollbahn' position be just the first step in a steady withdrawal to the 'Panther-Stellung', pointing out that the army group was already so weakened that it would have only just enough troops to man the front when it reached the 'Panther-Stellung'.

Less than a day passed before the accuracy of von Küchler’s forecast began to become clear. On 21 January the 42^nd Army attacked toward Krasnogvardeysk, the junction of the main railway and road lines coming from the south and west. The L Corps had not had time to regroup its battered formations and units and start setting up a front. That night von Küchler flew to the Führer headquarters, and it was here on the following morning, shortly before his meeting with Hitler, that word reached him that the 18^th Army could hold Krasnogvardeysk only if it gave up Pushkin and Slutsk, also important junctions but located farther to the north. Hitler refused to consider all of von Küchler’s proposals, and brushed away everything said about Pushkin and Slutsk, the 'Panther-Stellung', and the possibility of new threats on the army group’s right flank with the statement that Heeresgruppe 'Nord' was spoiled: not having faced a crisis for more than a year, it therefore had no conception of was a real crisis was, and then added that he was against all withdrawals as the Germans would face crises wherever they were. Hitler also told von üchler that there was no guarantee that the Germans would not be defeated on the 'Panther-Stellung', and if they withdrew voluntarily to this the Soviets would not close up to the line without losing half their strength, and that it was therefore essential that the Soviets forces be bled white on their way to the 'Panther-Stellung'. Hitler’s final argument was that the battle had be fought as far as possible from the German border. von Küchler objected that the 'Panther-Stellung' could not be held if the army group was too weak to fight when it got there, but Hitler blamed all the gaps in the front on the egotism of the army groups and insisted that every square yard of ground be sold at the highest possible price in Soviet blood. Finally, demanding that the 'Rollbahn' position be held, he dismissed von Küchler Zeitzler later said the time had been bad, and that von Küchler should try again in a few days as Hitler was pre-occupied about the 'Shingle' landing that day by Allied troops at Anzio to the south of Rome, and had therefore not listened to what was said.

Meanwhile, the 18^th Army was starting to disintegrate as its troops, fighting in mud and water, approached exhaustion. Govorov and Meretskov, on the other hand, had managed since the arrival of warmer weather in the middle of the month to give their divisions one out of every three or four days to recuperate and dry themselves.

On the morning of 23 January, Lindemann gave the order to evacuate Pushkin and Slutsk and reported to the Oberkommando des Heeres could either accept his decision or replace him. During the day the 18^th Army completed its withdrawal to the 'Rollbahn' position, which the Soviets had already penetrated in several places.

On 24 January, at the headquarters of the 18^th Army, von Küchler accused Lindemann of having submitted false estimates of Soviet reserves at the end of December, to which Lindemann conceded that mistakes had been made. The belated revision of the army’s past intelligence errors was rapidly overtaken, however, by the arrival of more adverse news from the front. In the morning the Soviets entered the outskirts of Krasnogvardeysk and driven through to the bend of the Luga river to the south-east of Luga. The divisions in the 'Rollbahn' position tried to plug the gaps in the front by the commitment of rear-echelon troops. At the end of the day Lindemann reported that his right flank had lost contact with the 16^th Army and that Krasnogvardeysk would fall within 24 hours. As the loss of Krasnogvardeysk would impinge severely on the supply lines of the corps farther to the east, the army group requested permission to fall back at least to the Luga river, about mid-way between the 'Rollbahn' position and the 'Panther-Stellung' along the western sides of Lake Pskov and Lake Peipus. In the evening Zeitzler replied that Hitler’s orders were to hold firm and make the troops fight to the last. Since there was nothing else which could be achieved for the time being, Zeitzler advised von Küchler to be ruthless for a while. On 27 January, von Küchler and the other army group and army commanders on the Eastern Front attended a Nazi leadership conference at Königsberg, and here Hitler addressed his generals on the subject of faith as a guarantee of victory, demanding a strengthening of faith in himself, the Nazi philosophy and ultimate victory, and suggested that the faith of the generals was as much in need of strengthening as much as that of anyone else. During an interval in the conference, in a private talk with Hitler, von Küchler repeated a situation estimate he had submitted the day before: the Leningrad Front and Volkhov Front had committed four major attack forces to cut the 18^th Army into pieces; the Soviets were advancing on Narva from the east and Luga from the north and east; and that of the onslaught from the east took and passed Luga the lines of communication to all five of Lindemann’s corps would be severed.

Almost inevitably, Hitler’s response was to prohibit all voluntary withdrawals and to reserve to himself all decisions about withdrawal to himself. When von Küchler remarked that the 18^th Army had suffered 40,000 casualties and that the troops had fought as hard as could be expected, Hitler replied that the latter statement was not wholly true as he had heard that the army group was not fighting everywhere with as much determination as it might.

The meeting destroyed von Küchler as an effective army group commander. When he returned to his headquarters he still seemed to realise that retreat was necessary, but all he could talk about was the need to show more determination and to attack, though no one including von Küchler himself knew where the required troops were to be found. On 28 January Kinzel took matters into his own hands and told Oberst Wilhelm Hetzel, the new acting chief-of-staff of the 18^th Army, that the time had come for an order to retreat to be issued, but that the army group was forbidden to do that. The army would, therefore, have to act as if it had been given the relevant order, issuing its own orders orally rather than in writing. On the next day Kinzel managed to persuade von Küchler at least to submit a report pointing out to Hitler that the 18^th Army had been divided into three parts and could not hold any kind of a front forward of the Luga river. On 30 January von Küchler again travelled to the Führer headquarters, where Hitler finally approved a retreat to the Luga river but directed that the front then be held, contact with the 16^th Army re-established, and all gaps in the front closed. When von Küchler passed this instruction to his operations officer, the latter protested to the operations branch of the Oberkommando des Heeres that it was an order impossible to execute: one of the gaps was 30 Miles (48 km) wide, and at Staritza to the north-west of Luga the Soviets were already across the Luga river. Zeitzler later agreed to tell Hitler that the Luga line could not be held.

von Küchler had meanwhile been instructed to report back to the Führer headquarters on 31 January. At the 12.00 conference on the next day, Hitler informed von Küchler that he was relieved of his command. Generalfeldmarschall Walter Model, who had been waiting to replace von Manstein, was now given temporary command of the army group. Reacting with his usual speed, Model telegraphed to the army group headquarters that not a single step backward was be taken without his express permission, that he was flying to the 18^th Army during the afternoon of the same day, and that Lindemann should be requested to give Model once more the trust that had characterised their previous time together.

In the final days of January, the rate of the 18^th Army's losses had increased steeply. On 27 January, the northern part of the army’s front had been about 10 miles (16 km) to the north of the line connecting Narva and Chudovo along most of its length, and 40 miles (65 km) to the north-east of Narva in its western quarter. By 31 January the army’s line had been driven back almost to the Narva river in the west and slightly below the line linking Narva and Chudovo in the east. In itself this was not a surprising loss of ground, but in the interval the front had virtually dissolved. On the situation maps of 27 January the front line had still appeared as a distinguishable and almost continuous line, albeit with several large gaps; by 31 January all that was left was a random scattering of dots where battalions and companies still held short stretches of the front. The only two divisions still worthy of the name were Generalleutnant Erpo Freiherr von Bodenhausen’s 12^th Panzerdivision, which had arrived during the last week of the month, and Generalleutnant Curt Siewert’s 58^th Division arriving from the south by train. On 29 January the army group reported that as of 10 January the 18^th Army had possessed an infantry combat strength of 57,936 men: since then it had lost 14,000 men killed and 35,000 wounded, and now had an infantry strength of 17,000 men including new arrivals.

Never before had Model been offered a greater opportunity to display his skills as an improvisor, and he seized the opportunity with an enthusiasm which could not change the tactical situation but swiftly dispelled the sense of hopelessness and frustration that had been hanging over the army group. Model also possessed the advantage of Hitler’s tendency to give newly appointed commanders, especially when they were also his favourites, at least a temporarily greater latitude than had been afforded their their predecessors. Model’s first endeavours were as much psychological as military. To dissipate what he called the 'Panther psychosis', he forbade all references to the 'Panther-Stellung' and indeed abolished the designation. Past experience had shown Model that in difficult times named lines, particularly when the names suggested strength, exercised a powerful attraction on commanders and troops. The state of the 18^th Army being what it was, however, Model could not attempt to enforce his original 'no step back' order. Instead, he introduced something new in the form of the Schild und Schwert (shield and sword) concept, in which the core notion was that withdrawals were tolerable if one intended later to strike back in the same or a different direction in a kind of parry and thrust sequence. The concept seems to have been the brainchild of Hitler as a remedy for the 'disease' of falling back to gain troops to build a new defence line which in a short time would itself prove too weak to be held. It is likely that Model did not placed great faith in the concept, for he was realistic to know that while the withdrawal was usually possible the counter-offensive was not. On the other hand, he was also well enough acquainted with Hitler to know that it was always advantageous to make a retreat look like the first stage of an advance.

Model applied the Schild und Schwert concept in his first directive to the 18^th Army, which was issued on 1 February. He ordered Lindemann to pull his main strength back to a short line to the north and east of Luga. Once this had been accomplished and the 12^th Panzerdivision had finished closing the gap between the 18^th Army and 16^th Army, as had already been ordered before Model’s arrival, the 12^th Panzerdivision and 58^th Division, supplemented by as many other divisions as could be spared from the short line, would be shifted to the west of Luga for a thrust along the Luga river to establish contact with the two corps on the Narva. The first part of the directive gave the army a chance to reduce its frontage by almost two-thirds, which was necessary, and the second envisaged a gain of enough strength, which was highly doubtful, for the launch of a counteroffensive and an extension of the front 50 miles (80 km) to the west. To apply the Schild und Schwert concept on the 18^th Army's left flank was impossible. General Otto Sponheimer’s LIV Corps and SS-Obergruppenführer und Generaloberst der Waffen-SS Matthias Kleinheisterkamp’s (from 11 February SS-Gruppenführer und General der Waffen-SS Martin Unrein’s) III SS Panzerkorps, both under Sponheimer’s command, had fallen back along the Baltic coast from the Oranienbaum pocket. After 28 January they had been driven back first to the Luga river and then to the Narva river, the northern end of the 'Panther-Stellung'. They could fall back no farther without endangering the entire 'Panther-Stellung' and the important shale oil refineries near the coast about 20 miles (32 km) to the west of the river.

On 2 February, when Model inspected Sponheimer’s front, his divisions were crossing to the western bank of the river and pulling back into a small bridgehead around the city of Narva. To the south of Narva the Soviets were probing across the river and before the end of the day had established a small bridgehead of their own. Elements of Generalleutnant Otto Kohlermann’s Panzergrenadierdivision 'Feldherrnhalle', which had been diverted from Heeresgruppe 'Mitte', and one regiment of the 58^th Division, were now arriving to strengthen the front below Narva.

At every point Model was subjected to the same complaint that the men were exhausted, and at every point he reconfirmed that they would nonetheless have to see the battle through. The help the army group could give was small enough: an infantry adviser for the III SS Panzerkorps, an artillery expert to match the skilled gunners the Soviets were using, and a series of requests to Reichsführer Heinrich Himmler for experienced SS replacements, to Grossadmiral Karl Dönitz for reinforcements for the coastal batteries, and to Reichsmarschall Hermann Goring for air force personnel to be used against the partisans becoming increasingly troublesome in the German rear areas and on the lines of communication. Nevertheless, the near collapse of the 18^th Army at the end of January had exercised the effect of a temporary disengagement, in some places at least, as on the Narva river. Model’s decision to tighten the front around Luga gave the army a chance to manoeuvre and also to recover its balance.

The next move was still that of the Soviets, but it would be met on a coherent front. For a few days at the beginning of February the points of greatest pressure were on the area held by the 16^th Army, where General Markian M. Popov’s 2^nd Baltic Front pushed into the front in an area to the south of Staraya Russa and to the west of Novosokol’niki, pinning German formations which might otherwise have been available for movement to the north and, as a bonus, creating small salients as possible launch points for deep thrusts at a later time.

By 4 February the Leningrad Front and Volkhov Front had regrouped and were starting to close on the 18^th Army once more. Heeresgruppe 'Nord' informed the Oberkommando des Heeres that the Volkhov Front had massed a strong force with 200 tanks in the area to the south-west of Novgorod, and the Leningrad Front was assembling another strong force in the area to the east of Lake Samro, some 30 miles (48 km) off the 18^th Army's left flank. The conclusion which had inevitably to be drawn from this fact was that the Soviets were about top attempt the encirclement of Luga.

It was still Model’s intention to attack to the north-west, and for this he proposed alternative 'large' and 'small' solutions: the former would carry the front out to the length of the Luga river, while the latter would extend it diagonally to the northern tip of Lake Peipus. Kinzel, the chief of staff of Heeresgruppe 'Nord', later remarked to Hetzel, his counterpart in the 18^th Army, that it was gratifying just to be able to think about moves of such boldness. Whether or not either of these alternatives would actually be implemented would depend on the manner in which the battle developed over the next few days. In any event, nothing would be lost because the preliminary movements would be useful whatever the army did next.

Normally delighted by talk of an offensive, Hitler displayed no enthusiasm. In a rare personal directive to Model he cited the Narva area as most vulnerable and ordered its immediate reinforcement. In the sector between Lake Peipus and Lake Ilmen Hitler perceived the danger of the 18^th Army being pushed eastward away from Lake Peipus, and thus placed in danger of encirclement, and ordered Model to submit a request for a withdrawal to the 'Panther-Stellung' as soon as either of these eventualities became imminent. Oddly enough, therefore, after appointing the kind of daring commander he wanted, Hitler urged caution. The change probably also resulted in part from Hitler’s tendency to associate men events. It seems likely, therefore, that even before the dismissal of von Küchler, Hitler had decided that a retreat to the 'Panther-Stellung' was necessary, but had not acted then because he could not bring himself to accede to what he considered von Küchler’s responsibility for the defeat.

On 6 February the 12^th Panzerdivision completed the closure of the gap between the 18^th Army and 16^th Army, and now prepared to assemble in Pskov for an attack to the east of Lake Pskov and Lake Peipus. The 58^th Division was ready farther to the east, and the 18^th Army had called for a withdrawal on the Luga sector of the front to free three divisions in two days. In this pause, short as it was, the 18^th Army's strength had begun to increase as stragglers, men recalled from leave and soldiers released from hospital returned to their divisions. In addition, Model had ordered one in 20 of rear-echelon troops to be transferred to line duty. At the headquarters of the 18^th Army, on 7 February Model issued instructions for the first stage of the planned counter-offensive: by the movement of divisions from the north and east the army was to create a solid front between the southern tip of Lake Peipus and Luga and, with that accomplished, the army would apply the Schild und Schwert concept by employing two corps defensively in the east to check the Soviet advance from Lake Samro and one corps offensively in the west for a thrust to the north along the shore of Lake Peipus. Over the next two days the 18^th Army sought to move its divisions into position. Partisan roadblocks delayed the 12^th Panzerdivision's advance toward Pskov. The 58^th Division established a short front on the Plyussa river along the centre of the proposed new line, but the Soviets filtered past on each flank, and the other divisions would have to attack merely to re-establish the cohesion of the front. This would be no easy task as each of the divisions had only four understrength battalions and the Soviet strength was growing hourly as fresh formations and units arrived from the north-west. The swampy nature of the local terrain also caused difficulties but, on the other hand, this was probably the main reason why Leningrad Front could not bring its full force to bear more quickly.

By 10 February the 58^th Division had been split into two, and one of its regiments had been encircled. Oberst Kurt Versock’s 24^th Division, which was attempting to close the gap on the right of the 58^th Division, could make no progress and for most of the day found it very difficult to hold open the railway line linking Luga and Pskov. Although the 18^th Army tried on the following day once again to re-establish contact with the 58^th Division and to close the gap, the prospects were worsening rapidly and aerial reconnaissance had spotted convoys of 800 to 900 trucks moving to the south-east from Lake Samro. During the afternoon of the next day, the 18^th Army reported that the battle had taken a turn for the worse: the 24^th Division had been halted, Soviet armour had appeared, and both regiments of the 58^th Division had been surrounded and would have to fight their way back, probably with the loss of their heavy weapons. After the fall of night, Lindemann informed Model that the only way he could gather enough troops to close the gaps on the left flank was to withdraw the whole front to the shortest line, that between the southern tip of Lake Peipus and Lake Ilmen. Govorov had extended his right-hand pincer out to the shore of Lake Peipus and was pushing south toward Pskov, and had already pushed some formations and units far enough to the south 'to pinch the 12^th Panzerdivision in the backside'. Model reluctantly authorised the 18^th Armys retirement.

The following day brought more bad news, for at Narva the Soviets enlarged their bridgehead and established another to the north of the city. Between Lake Peipus and Lake Pskov, Govorov poured in enough troops to threaten a crossing into the 'Panther-Stellung'. So if Model was to create a front between Lake Peipus and Lake Ilmen, the 18^th Army would have to fight for it.

During the evening of 12 February, Model reported to the Oberkommando des Heeres that he still intended to take and to hold that line, and wished to know whether or not he had Hitler’s approval. The Oberkommando des Heeres’s response indicated that nobody, including Hitler, liked the idea. The high command opinion was, for once, unanimous in feeling that it was too late to establish a front between the lakes and that, in any event, it was more important to free one division for the defence of Narva and another for the defence of the narrows separating Lake Peipus and Lake Pskov. The operations chief of the Oberkommando des Heeres added that Hitler remarked on a daily basis that he did not want to risk any encirclements forward of the 'Panther-Stellung'.

At 23.00 Sponheimer reported that the Soviets had broken through to the north and south of Narva. To the north of Narva the III SS Panzerkorps had managed to close the gap in the front and even drive forward a short distance, but to the south of Narva the Panzergrenadierdivision 'Feldherrnhalle' lacked the strength to offer any form of effective resistance.

In the morning of 13 February, Model sent a new situation report to Hitler, and indicated that Heeresgruppe 'Nord' would fight the battle around Narva to its end. Model added that, should the worst happen, he would shorten the front by conceding the bend of the Narva river, but that he still believed it would be best to hold the Soviet forces between Lake Peipus and Lake Ilmen until more work had been completed on the 'Panther-Stellung'. Model received Hitler’s answer four hours later: the Narva front was to be strengthened with the greatest possible speed, and the army group was to submit a plan and timetable for a prompt withdrawal to the 'Panther-Stellung'.

For the moment it seemed that the decision to fall back to this position might have come too late to save the Narva front, to which Heeresgruppe 'Nord' released on that day, as a last resort, an Estonian brigade of SS-Obersturmbannführer Franz Augsberger’s 20^th Estnische SS-Freiwilligen Division. This brigade was the product of a draft the SS, which was responsible for foreign recruitment, had been undertaking in Estonia, Latvia and Lithuania since a time early in January. As Hitler had refused to offer the Baltic states even the promise of eventual autonomy, the draftees were of low morale and their only motivation was fear, which was of the Germans as much as the Soviets. On the night of 13 February, Sponheimer reported that the Estonians had arrived in a state of disorder verging on panic, and that some of the men had tried to desert on their way to the front. Model was therefore left with no alternative but to take troops from the 18^th Army. He ordered the 58^th Division to be transferred north after a three-day rest. This formation had lost one-third of its men and all of its heavy equipment in the encirclement battles.

On the morning of 14 February, after Sponheimer had reported that he had no room to manoeuvre and no troops with which to close the gaps, and was therefore helpless, Model asked for authority to evacuate the small bridgehead still being held to the east of Narva, which would provide him with three battalions. Zeitzler signalled his approval and also offered an infantry division from Norway. Then, shortly after daylight, came the news that the Soviets had made an amphibious landing to the north-west of Narva: later reports revealed that only some 500 naval troops, supported only by several gun boats, had been landed from Lavansaari island in the Gulf of Finland. In the report he despatched to the Oberkommando des Heeres, Model stated that the situation in the Narva area was poor, and that he had ordered the immediate abandonment of the current bridgehead. During the day the landing force was destroyed in an effort which resulted in some unfortunate German losses after German dive-bombers had mistakenly attacked a German division headquarters and destroyed several Tiger heavy tanks.

A greater threat was posed by the appearance of Soviet ski troops on the western shore of Lake Peipus in the area to the north of the narrows. The security division responsible for the area reported that its Estonian troops were melting away, and after this Model informed the Oberkommando des Heeres that he would begin the withdrawal to the 'Panther-Stellung' on 17 February and complete it early on 1 March. He would mop up the western shore of Lake Peipus in the next few days and use the first two divisions freed to cover the lake shore. He expected that as soon as the 18^th Army started to fall back, Govorov and Meretskov would attempt an envelopment round the army’s flanks as they had strong forces in position to the north of Pskov and on the western shore of Lake Ilmen.

In the two days before the withdrawal began, the Soviets made no further attempt to cross the lakes, and on 17 February Model gave a corps headquarters command of the lake sector and began to move the 12^th Panzerdivision into the area. On the Narva, the battle began almost immediately to degenerate into a vicious stalemate in which the Germans and Soviets fought and very close quarters, neither giving nor gaining. Sponheimer could not close the gaps in his front, but that Govorov was less than satisfied with his own progress was confirmed in repeated radio messages offering the highest Soviet valour decoration to the commander of the first troops to reach the road running to the west out of Narva. As the 16^th Army and 18^th Army began to fall back to the west, the Soviet armies pressed forward close on their heels. Through their networks of agents and partisans the Soviets were fully apprised of what the Germans were doing.

On 19 February Heeresgruppe 'Nord' became aware, suddenly and acutely, of an old danger which had been waiting silently throughout the last month of crises. On that day, for the first time in two months, the attacks on the perimeter of Generaloberst Georg-Hans Reinhardt’s 3^rd Panzerarmee of Heeresgruppe 'Mitte', around Vitebsk farther to the south suddenly ended, and aerial reconnaissance detected the movement of truck convoys of 2,000 or 3,000 vehicles, most of them to the north and north-west. The intelligence branch of Heeresgruppe 'Nord' calculated that two armies could be shifted to the flank of the 16^th Army within a few days. Model believed that there were two possible explanations for these Soviet moves: the first, and more probable, was that after adding to its already strong concentration in the area of Nevel and Pustoshka, the 2^nd Baltic Front would attempt to break into the 'Panther-Stellung' below Pustoshka and roll it up to the north before the 16^th Army and 18^th Army could establish themselves firmly in the position, while the second was a Soviet thrust straight through to Dvinsk and on to Riga on the east coast of the Baltic Sea to cut off Heeresgruppe 'Nord' in the Baltic states.

Model also had to take into consideration that the Soviet activity on the right flank of the 16^th Army might indicate that the Stavka was becoming discouraged in its attempts to encircle the 18^th Army but, even if this was true, it led to no decrease in the Soviet pressure on the 18^th Army for, as had been foreseen, the Volkhov and Leningrad Fronts continued to press on the flanks of the army. Meretskov attempted to break through at Shimsk, to the west of Lake Ilmen, on 17 February. For three days, while the 16^th Army's flank pulled back from Staraya Russa, the battle to keep contact between the two armies swayed in the balance. On 20 February, as both began to pull away from Lake Ilmen, that crisis passed. Govorov reacted more slowly but also more dangerously. Pskov, which throughout the war was the primary communications centre of Heeresgruppe 'Nord', was also the hinge on which the whole withdrawal to the 'Panther-Stellung' turned. The army group could not afford to lose Pskov, but had very little area around the city in which to manoeuvre. In the swamps and forests to the east of Lake Pskov, the Leningrad Front met considerable problems in bringing its forces to bear, but on 24 February it began to exert great pressure in the area to the north of the city, and also launched probing attacks across the lake. According to intelligence reports, Iosif Stalin had summoned Govorov and ordered him to take Pskov.

By 26 February the threats at Pskov and on the 16^th Army's right flank had made Hitler so worried that he asked Model to accelerate the speed of the withdrawal to the 'Panther-Stellung'.

Toward the end of the month, on the Narva sector of the front in the north, the Germans had managed to concentrate only enough strength to effect a slight tip the scales in their favour. On 24 February, General Johannes Friessner, who had proved himself as commander of the XXIII Corps in the fighting on the junction of the 16^th Army and 18^th Army, replaced Sponheimer in command of what had been, since 27 January, the Armeegruppe 'Sponheimer' and was now redesignated as the Armeeabteilung 'Narwa'. By then elements of Generalleutnant Harry von Kirchbach' 214^th Division were beginning to reach the front. Though fresh, these troops still required seasoning before commitment to the hardest-fought sectors of the front, but could be used to relieve experienced troops in the quieter parts of the line. Adopting the practice of what he called 'mosaic work', Friessner cut into the extreme tip of the Soviet bridgehead to the south of Narva and drove the Soviets there into two small pockets. Although the Soviets ignored the crushing weight of the artillery and small arms fire and kept pouring in troops through the open ends of the pockets, the danger of their reaching the coast was averted.

On 1 March Heeresgruppe 'Nord' took the last step back into the 'Panther-Stellung', and the Soviets immediately made it clear that they were not considering any pause in their offensive efforts. To the north of Pustoshka two armies fell on General Gustav Höhne’s VIII Corps, and to the town’s south of the town two armies fell on General Thomas-Emil von Wickede’s X Corps. The Leningrad Front massed two armies to the south of Pskov and committed more formations across the Narva river, attacking out of the Soviet bridgehead to the north, north-west and west. For a week the battle surged backward and forward along the whole front of Heeresgruppe 'Nord' and, with the exception of a number of small local losses, the Germans managed to hold their line. On 9 March the 2^nd Baltic Front increased its pressure on the 16^th Army's right flank and made a determined effort to secure a breakthrough.

On 10 March Heeresgruppe 'Nord' had to cope with a politically unpleasant but militarily insignificant result of the winter’s disastrous fighting: the commander of the Legión Azul, which was fighting as an element of Generalleutnant Hellmuth Priess’s 121^st Division in the XXVIII Corps, and the Spanish military attaché visited Model to tell him the legion was being called home: Franco was not turning away from Germany, they told Model, but wanted to gather all his best troops to resist any Anglo-US invasion of Spain. As the Spanish legion had been as troublesome in the rear areas as it had been ineffectual at the front, the loss to the army group was not significant, but nonetheless indicated that Germany’s supposed friends had started to appreciate the way the wind was now blowing.

In the middle of March, the 2^nd Baltic Front was still battering the 16^th Army's flank as the Leningrad Front probed for openings around Pskov and Narva. But the weather had now turned against the Soviets. After a winter which had bee, by Russian standards, very mild, the spring thaw and attendant rasputitsa mud had arrived earlier than usual, and almost immediately 12 inches (0.305 m) of water covered the ice on the lakes. The 16^th Army reported that Soviet tanks were sometimes sinking to the depth of their hulls into the mud. Against a weaker front the Soviets might just have been able to continue their advance, as indeed they were doing farther to the south in Ukraine, but the 'Panther-Stellung', which was all that remained of the 'Ostwall', was living up to German expectations.

On 28 March Hitler’s chief adjutant, Generalleutnant Rudolf Schmundt, called on Model to tell him that in a few days Hitler would name him to replace von Manstein as commander of Heeresgruppe 'Süd'. This information came at an inconvenient moment for Model, who had just completed a situation estimate in which he said that it was possible that Heeresgruppe 'Nord' would be able to transfer two divisions to Heeresgruppe 'Süd' after the front had stabilised. Model quickly revised his estimate to the effect that Heeresgruppe 'Nord' could give up five divisions and a corps staff immediately, and the 12^th Panzerdivision as soon as two self-propelled assault gun brigades and one battalion of tanks had arrived to replace it.

On the next day Model travelled to the Führer headquarters and, though still technically the commander of Heeresgruppe 'Nord', tried to use his authority to raid the army group for the benefit of his new command: Model initially informed Hitler that the army group could give up five divisions, and then raised the number to six division. In a telegram to the headquarters of Heeresgruppe 'Nord', Model stated that Hitler had ordered the transfer of six divisions, and then by telephone gave the chief-of-staff 30 minutes in which to report that the order was being obeyed. Zeitzler was finally compelled to intervene and instruct the army group not to fulfil any of Model’s orders. On 31 March, with Model now commanding Heeresgruppe 'Süd', Zeitzler persuaded Hitler to reduce the proposed transfers to one division, and that only in the near future rather than immediately. On 1 April Hitler agreed, and one day earlier Lindemann, the senior commander in the army group, had been appointed acting commander of Heeresgruppe 'Nord'.