The Tank 1917-45

First World War Battlefield Debut

Tanks began to appear on the world scene with the delivery at the beginning of June 1916 of the first British tanks, produced as the result of the order placed for them only four months earlier by the Ministry of Munitions. The tanks were simply designated Mark I and were virtually identical with ‘Mother’, except that their hulls were made of armour instead of mild steel plates and that half of them were armed only with machine guns. Like their progenitor they relied for sharp turns on a crude system adopted not on its merit but for the sake of using existing transmission components and thereby saving time and development effort. It involved putting the secondary gearbox on one side of the differential into neutral and applying a brake to the undriven track while the other track continued to be driven. This meant that four men were required to maneuver a tank: commander and driver at the front, who controlled the engine and applied the brakes, and two gearsmen, one on each side of the rear of the tank. There were also two gunners at each of the two sponsons, which brought the crew to a total of eight men.

The problem of steering the Mark I tanks and hence of maneuvering them was aggravated by the difficulty the commander had of communicating with the gearsmen, as well as the gunners, because of the noise of the engine, which was located in the middle of the hull. The engine also generated heat and emitted noxious fumes, which could make the interior of tanks extremely uncomfortable for their crews. The crews also had to endure severe jolts caused by the absence of a sprung suspension when tanks operated over broken ground. Under some conditions the tanks were slower than the infantry with which they were co-operating, although on flat, hard ground they were capable of a maximum speed of 3.7 miles per hour and had an operating range of 24 miles.

These and their other shortcomings were bound to have an adverse effect on the performance of the first tanks. However, this did not hinder their adoption and was not allowed to delay the sending of the first tanks into battle only seven months after they were ordered.

The remarkably rapid production of the first tanks was accompanied by a decision taken by the War Office on 16 February 1916 to form the first tank unit and by an increase in April of the production order from 100 to 150 tanks. In consequence, by the end of June two of the six companies into which the tank units were to be organized, and each of which was to have 25 tanks, began to train. Moreover, having endorsed the production of tanks, the General Headquarters of the British Forces in France was keen to employ them as soon as possible.

The speed with which the British Army accepted tanks and set about employing them must be credited to some extent to Swinton, who returned to England at the end of July 1916 to become assistant secretary of what was called the Dardanelles Committee of the Cabinet. On his return he discovered the existence of the Landships Committee and its work on tanks. He then took advantage of his influential position to instigate in August an interdepartmental conference to co-ordinate the activities of the Landships Committee, the War Office and the Ministry of Munitions, and he went on to promote tanks wherever an opportunity arose. For his efforts he was rewarded in February 1916 by being made commander of the tank units that were being raised in England. As such he was responsible, among others, for the somewhat peculiar decision to arm one half of the 150 tanks that were being produced only with machine guns, on the grounds that such ‘female’ tanks would be needed to protect the ‘male’ tanks, although the latter were already armed with four machine guns as well as 57mm guns, against an onrush of enemy infantrymen!

Swinton was also the first in the British Army to write about how tanks might be employed. He did so originally in a memorandum entitled ‘The need for machine gun destroyers’ that he submitted on 1 June 1915 to the General Headquarters in France. In this he suggested that ‘armoured machine gun destroyers’ should be used in a surprise assault on enemy positions with the object of destroying hostile machine guns and thus paving the way for the attacking infantry. He subsequently elaborated his ideas in a paper entitled ‘Notes on the employment of tanks’ written in February 1916. In this he again defined the principal role of tanks as clearing the way for infantry assaults by destroying hostile machine guns. He envisaged therefore a somewhat specialized, limited role for tanks and did not contemplate their employment beyond the confines of trench warfare.

In both cases Swinton warned against the premature employment of a few tanks and advocated 100 being used in a surprise assault.8 But even before the first tanks were built the commander-in-chief, Sir Douglas Haig, became keen to use some in the forthcoming offensive on the Somme. As it was, the earliest that any tanks could be made available was August 1916, when two companies were sent to France. Once they were there the General Headquarters decided to use them to bolster an attempt to revive the offensive on the Somme, which had by then stalled. In consequence the two companies were moved to the front and on 15 September 1916 took part in a large-scale attack on German positions in what became known as the Battle of Flers-Courcelette.

The tanks were distributed over the front of ten infantry divisions and were used in twos or threes to attack enemy strong points in support of the assaulting infantry. Forty-nine tanks were available but only 32 reached the starting line; nine then led the attacking infantry, engaging the enemy with their guns and machine guns, while nine others assisted in a similar way in clearing up pockets of resistance. Of the remaining 14, nine broke down and five became ditched.

Overall, the performance of tanks in their first battle was not a great success and the contribution they made to the progress of the Somme offensive, which was limited to an advance of about one mile, was very modest. However, bearing in mind the primitive nature of the first tanks and their attendant shortcomings, the fact that they were, at best, only three months old, and the inadequate training of their crews, the participation of tanks in the Somme offensive was a very considerable achievement.

Nevertheless, the use of the tanks on the Somme has been widely criticized as premature, mainly on the grounds that more might have been achieved had their debut been delayed until more were available. On the other hand it has been argued in its defence that tanks had to be put to test in battle at an early stage to gain experience with them. However, it is not evident that some of the lessons that were brought out by the early employment of tanks, such as the need for adequate training of the crews, could not have been foreseen in advance of Flers-Courcelette.

Although tanks had not achieved what had been hoped of them, their first action was considered to have justified their existence in the eyes of Haig. In consequence, a meeting was held at the War Office only four days after the first tank action at which it was agreed that an order be placed for 1,000 more tanks. However, because of some confusion, the order was not confirmed until 14 October and it did not begin to bear fruit until March 1917, when the first of the new Mark IV tanks was completed. Eventually 1,015 tanks of this type were produced. In the meantime, to keep the factories going, an order was placed for 100 Mark II and Mark III tanks, which were very similar to the original Mark I type.

Following their debut on the Somme, the use of tanks was confined to a few small-scale actions until the Battle of Arras in April 1917, for which 60 tanks became available. These were dispersed among the attacking infantry formations, and although they were successful in a few local actions many became bogged down in a terrain made impassable by heavy rains. Even worse conditions were encountered in the next major engagement of tanks, in the Third Battle of Ypres, or Passchendaele, in July to October 1917. This was fought in an area of reclaimed swampland turned into a sea of mud by a combination of artillery bombardment and heavy rains. The number of tanks available had risen to 216 and included some new Mark IVs, which incorporated a number of improvements on the Mark Is, including better armour. However, they were again dispersed among a number of infantry divisions and the terrain severely restricted their movement, which helped enemy artillery to knock many of them out while others became bogged in the swampy ground.

By the third day of the Ypres offensive the commander of the Tank Corps, Brigadier General H. J. Elles, recognized the futility of the further use of tanks in it and suggested that the remaining tanks be withdrawn for use en masse in suitable terrain. At the same time his chief of staff, Lieutenant Colonel J. F. C. Fuller, came up with the idea of a ‘one-day tank raid’ – a surprise spoiling attack carried out over suitable ground without the customary preliminary artillery bombardment. Fuller’s memoirs imply that this led to the Battle of Cambrai, which became the first successful large-scale tank attack. In fact, the Battle of Cambrai came to be a much bigger operation than Fuller had originally envisaged and others were involved in its conception, including the commander of the Third Army, General Byng, in whose sector of the front the battle took place.

All available tanks were assembled for the battle, the total being made up of 378 fighting tanks, 54 older tanks carrying supplies, 10 radio and cable communication tanks and 34 other tanks for clearing the ground of barbed wire for the planned follow-up by the cavalry and carrying bridging for it. The assembly of the 476 tanks and of the supplies of fuel and ammunition required by them was carried out in great secrecy, and for the attack on 20 November 1917 the tanks were drawn up in a single line in front of seven miles of British trenches. As they moved forward, tanks crushed the barbed wire covering enemy positions and subdued enemy machine guns by fire, clearing the way for the infantry that followed them. The supporting artillery, which totalled 1,000 guns, did not open fire until the tanks began to move so as not to alert the enemy and the attack thereby achieved complete surprise.

Led by tanks, the attacking forces broke through the defences of what was known as the Hindenburg Line and advanced up to 7 miles, which was more than the total advance made in the three months of the Ypres offensive and which was achieved at the cost of far fewer casualties. The attack demonstrated clearly how effective tanks could be as assault vehicles when used in numbers over suitable ground, even though 112 of them were destroyed by the end of the battle by enemy artillery fire.

However, the success of the initial attack was not exploited, the tanks being too slow to do so and the cavalry too vulnerable to machine guns. Moreover, when the German forces counter-attacked ten days later they recovered most of the ground. The Battle of Cambrai ended therefore in deadlock and its outcome made the German General Staff consider tanks less of a threat than it might have done. But it did not prevent a further expansion of the Tank Corps from three to five brigades.

In the winter of 1917–18 the five brigades were spread out to form a defensive cordon some 60 miles behind the British lines in anticipation of a major German offensive. When that came in March 1918, the tanks were used piecemeal and in the ensuing retreat many were lost, having to be abandoned when they broke down or ran out of fuel. In consequence they proved relatively ineffective.

Most of the tanks used were Mark IVs, which, like the Mark I, were designed for assaults on enemy trenches and were too slow as well as having too short an operating range for effective employment under the fluid conditions created by the German offensive. However, there was a new British tank more suited to them, the Medium A or Whippet, which made its battle debut on 26 March 1918. It was lighter than the earlier tanks, weighing 14 tonnes instead of the 27 or 28 tonnes of the Mark IVs, and it had a maximum speed of 8.3 miles per hour compared with 3.7 miles per hour, as well as an operating range of 80 miles compared with 35 miles of the Mark IV. It was also more maneuverable as it was driven by one man who could steer it by changing the speed of its two 45hp engines that separately drove its two tracks – ostensibly a simple method of steering a tracked vehicle, but one which required very considerable dexterity on the part of the driver as the engines were easy to stall.

The Medium A also departed from the rhomboidal configuration of the earlier tanks, its tracks being surmounted by a fixed turret that contained a crew of four and mounted four Hotchkiss machine guns. Because of its greater mobility Medium A scored some local successes, but shared with the others the consequences of the ineffective, dispersed employment.

The fortunes of the Tank Corps did not revive until the Battle of Amiens, fought on 8 August 1918. This was the second major British tank battle and it was larger and more decisive than Cambrai. The whole of the Tank Corps was assembled for it, except for one brigade that was still equipped with Mark IV tanks. The other brigades had by then been re-equipped with the new Mark V tanks. These were basically similar to the Mark IVs but they had somewhat more powerful engines and could be driven by one man, as they had an epicyclic gear steering system instead of the crude method involving four men by which the earlier tanks were steered. In consequence they were much more maneuverable. In addition to the 324 heavy tanks there were 96 Medium A tanks, and with supply tanks and spare vehicles the total assembled for the battle amounted to 580 tanks.

As at Cambrai, the tanks were assembled secretly and attacked en masse along a 13 mile front without a preliminary artillery bombardment. The surprise tank assault overwhelmed the German defences and led to a major breakthrough, during the course of which the German Army suffered heavy losses. General von Ludendorff, who was effectively the German commander-in-chief, described 8 August 1918 as ‘the black day of the German Army’.

However, the breakthrough at Amiens was achieved at the cost of many tanks destroyed by the German artillery. As a result of the losses, by the second day of the battle the number of tanks available for further action was down to 145. Moreover, the success of the initial attack was again not exploited because the Mark V tanks were still too slow, being only marginally faster than the Mark IVs. The Medium A were faster but they were tied to the cavalry, which was expected to exploit the breakthrough but proved incapable of operating in the face of machine guns, as it did at Cambrai.

Nevertheless, the Battle of Amiens led to the beginning of a slow retreat of the German Army that went on until the end of the war three months later. During this period tanks attacked successfully on a number of occasions but their attacks were generally on a relatively small scale, involving at most 40 to 50 tanks, because of their shortage after Amiens. This was aggravated by further losses inflicted by German artillery and by the increasingly mobile nature of the operations, to which the available tanks were not suited. About 175 were assembled at the end of September for the assault on the Hindenburg Line, but only 37 could be scraped together for the final tank attack on 4 November 1918.

Three weeks before the Battle of Amiens, the French Army also carried out a large-scale tank attack at Soissons, which became another major success for tanks. This was preceded by a series of smaller scale actions by French tanks, the first of which took place on 16 April 1917 as part of an offensive on the Aisne. By then 208 Schneider and 48 Saint Chamond tanks had been produced of the total of 800 ordered a year earlier, and 160 of the Schneiders were considered ready for action, although not all of them had been fitted with the additional armour found to be necessary because of the introduction by the German Army of armour-piercing machine gun ammunition in response to the use of tanks by the British Army.

When French tanks began to be developed, it was thought that they would be used to break through enemy fronts by surprise assaults carried out without preliminary artillery bombardments. But by the time they were built the British Army had began to use tanks and the German Army responded to them by digging wider trenches that the Schneiders, let alone the Saint Chamonds, could not negotiate. In consequence it was decided that they should not be used to lead infantry assaults but to support the infantry beyond the effective range of the supporting artillery. In other words, they came to be regarded as mobile close support guns, which was in keeping with the designation given to the French tank units of ‘artillerie d’assaut’.

A total of 132 tanks was assembled for the attack on the Aisne, which failed, with tanks contributing little to the limited penetration of enemy positions. The tanks, almost all of which were Schneiders, had difficulty negotiating trenches and the shell-cratered ground, and 57 were destroyed by enemy artillery.

The inauspicious debut of the French tanks was not followed by another tank action until October 1917, when 64 tanks took part in the Battle of Malmaison. This time they successfully supported the infantry, although they still operated in small groups, and only eight were knocked out by enemy artillery. No further action took place until after the German offensive in March 1918, in anticipation of which French tanks were held behind the front line for counter-attacks, the total of operable tanks amounting by then to 245 Schneiders and 222 Saint Chamonds. They were at first employed piecemeal in a number of local counter-attacks under conditions of mobile warfare for which they were no better suited than the contemporary British tanks, although they were more maneuverable than the Mark IVs, being driven by one man and having sprung suspensions. But the most significant of the counter-attacks, which was carried out on 11 June 1918 by a force of 144 Schneiders and Saint Chamonds and infantry in the Matz valley, was successful in halting an enemy advance, albeit at the cost of 69 tanks.

In the meantime, in June 1916, the French High Command was informed of the British development of tanks and Colonel Estienne, who was about to be given the task of organizing the first French tank units, was sent to England to investigate. After being shown British Mark I tanks Estienne came back with the idea that there should also be a much lighter tank, which he saw as an armoured, machine gun armed infantryman who could operate over all types of terrain. During the following month Estienne put his idea to Louis Renault, who took it up with enthusiasm and proceeded to design a two-man light tank to meet it. By November 1916 Estienne was sufficiently confident of Renault’s design to write to the Commander-in-Chief, General Joffre, as he did at the inception of the development of the Schneider tank, recommending that as soon as a prototype of the light tank was approved 1,000 be ordered. Joffre was in favour of it, but the technical branches of the Army and the Ministry of Armament again raised various technical and bureaucratic objections and although in February 1917 Renault received an initial order for 150 tanks the follow-on order for 1,000 tanks was cancelled in April, albeit temporarily. To make matters worse, Joffre was then replaced by General Nivelle, who was less favourably disposed towards tanks, and it was only in October 1917 when he, in turn, was replaced by General Pétain that 2,380 tanks were added to the earlier orders for 1,150.

Fortunately, Renault and his company continued to work on the tank, in spite of problems including that of the supply of armour plates, which had to be imported from England for lack of sufficient industrial capacity in France. As a result, a prototype was built in April 1917 but the first production tank was not completed until September.

The light tank became known as the Renault FT, or simply as the Renault. It was very different from the French and British tanks that preceded it and in several respects represented a major advance on them. In particular, it was the first tank to have its armament mounted in a rotating turret. Moreover, its general configuration became and remains to this day the norm for most tanks. Its features included the location of the driver in the front of the hull, a weapon compartment surmounted by the turret in the centre and the engine compartment in the rear of the hull, separated from the crew by a bulkhead. As in most modern tanks, the track driving sprockets were also at the rear.

In battle order, the Renault weighed 6.5 tonnes but, in spite of its light weight, it had hull armour 16mm thick and even 18 or 22mm thick turret armour, which was thicker than that of the much heavier British tanks and sufficient to defeat the armour-piercing bullets of the German machine guns. It was also marginally faster than the British Mark V, having a maximum speed of 4.8 miles per hour, but it was not as fast as the Medium A.

Although the Renault was conceived as a machine gun tank, in April 1917 Estienne decided that some should be armed with a short-barrelled 37mm cannon instead of the Hotchkiss machine gun. Such a cannon was being used by the French infantry as a close support weapon and after modifications it was successfully mounted in the Renault, a proportion of which were subsequently armed with it. The cannon fired a full range of ammunition, including armour piercing and canister as well as high explosive rounds, and the tank could carry up to 240 of them.

The rather odd calibre of the cannon mounted in the Renault and later adopted for tank and anti-tank guns produced in several countries originated with the 1868 St Petersburg Convention that, on humanitarian grounds, defined the minimum permissible weight of high explosive shells. This led Benjamin Hotchkiss in France to design a cannon that fired shells of the prescribed weight whose calibre came to be 37mm. The cannon was adopted by the French and several other navies for the defence of large ships against the contemporary threat of high speed torpedo boats, and although after a time the naval use of 37mm Hotchkiss cannon declined, other cannon of this calibre came to be used on land.

Originally the intention was not to use the Renaults until they could be employed in some number. However, a German offensive against the French front in May 1918 called for the deployment of all available resources. In consequence two battalions of Renaults were rushed to the front at the end of May, although the formation of the first was only completed earlier in the month. As soon as they reached the front in the region of the Retz forest, 21 tanks charged the advancing enemy to gain time locally for the defence. Following this hasty debut Renaults were confined to assisting the defence of the Retz forest by a series of small scale counter-attacks, which were carried out at the cost of 70 tanks destroyed or severely damaged out of the total of 210 initially held by the three Renault battalions involved in these counter-attacks.

The Renaults did not come into their own until the counter-offensive launched by the French Army in the region of Soissons on 18 July 1918. All the available French tank units were assembled for it, their strength amounting to about 225 Schneiders and Saint Chamonds and six battalions of Renaults with a nominal strength of 432 tanks, or a total of more than 600 tanks. This was even more than the total number of tanks assembled four weeks later by the British Army for the attack at Amiens, but the French tanks were generally lighter.

As at Cambrai, the attack was carried out by surprise, without a preliminary artillery bombardment, and succeeded in disrupting the enemy front. It took place over the fronts of two French armies, over one of which it was led by almost all the available Schneiders and Saint Chamonds, while the three Renault battalions allotted to it were held in reserve for the exploitation of a breakthrough. On the front of the other army, the attack was led almost entirely by the three other battalions of Renaults with about 200 tanks.

From then on Renaults were used increasingly to lead or to support infantry attacks by a series of small scale actions rather than in massed assaults. In spite of tanks lost in battle, the number of Renaults in use grew rapidly as a result of the large scale orders placed for them, which had risen to 4,000 tanks and resulted in the actual delivery of 3,177 tanks by the Armistice of 11 November 1918.36 The numbers of tanks that were being produced made possible the formation of an increasing number of tank units, which were being created in the last four months of the war at the astonishing rate of almost one new Renault tank battalion per week. As a result, by the end of the war the French Army had as many as 24 battalions of Renaults in addition to equipping two US tank battalions with them.

The large number of tanks that the French Army came to use during the war contrasted sharply with the few the German Army deployed. The difference resulted in part from the late start of the development of tanks in Germany, which was only taken up after the appearance of the first British tanks in 1916.

Yet a Holt tractor similar to those that later became the basis of the development of tanks in Britain and in France was demonstrated to Austro-Hungarian and German military authorities in 1912 and 1913 respectively. The demonstrations were arranged by L. Steiner, a Hungarian engineer and land owner, who in 1910 ordered a Holt tractor for use in farming but then demonstrated its ability to haul heavy guns as well as becoming a Holt dealer. The gun-hauling demonstrations were successful and the Austro-Hungarian authorities acquired some Holt tractors before the outbreak of the war in 1914, but the German authorities dismissed the tractor Steiner demonstrated as of ‘no importance for military purposes’.

It was only in November 1916, two months after the debut of the British tanks on the Somme, that the German War Ministry purchased a Holt tractor from the Austro-Hungarian War Ministry and invited Steiner to Berlin for discussions with J. Vollmer, who was to become the designer of the first German tank. By then, in October 1916, the German War Ministry had set up a committee to draw up the specification of a tank, which was then designed with remarkable speed by the end of December. Earlier that month an order was also placed for the production of 100 tanks, the first of which was ready for use by October 1917.

The tank was called A7V after the designation of the committee that initiated its development. It consisted in essence of a large box hull mounted on a tracked chassis based on that of the Holt tractor acquired from Austria. The hull was riveted from plates 30mm thick at the front and 15mm at the sides, which was considerably more than the thickness of the plates of British tanks, but also made it heavier, its weight in battle order being 33 tonnes. In spite of this, A7V had a relatively high maximum road speed of 8 miles per hour, but, like that of the French Saint Chamond, its obstacle crossing ability was very limited. It was armed with a captured Russian 57mm gun mounted in the front of the hull and two machine guns in each side and the rear of it. Otherwise the most noteworthy feature of A7V was its large crew of 18 men, which set up a record not surpassed since by any other tank.

As tanks were produced, three detachments of five A7Vs each were formed, and they took part in the German offensive that broke through the British lines in March 1918. One of the detachments went into action for the first time at St Quentin on 21 March and all three were engaged three days later at Villers-Bretonneux, where they spearheaded infantry assaults with considerable success.

At Villers-Bretonneux, the A7Vs were also involved on 24 April in the first ever tank versus tank battle when they ran into some British tanks. The latter were initially two Mark IV ‘female’ tanks armed only with machine guns that were forced to withdraw, damaged and unable to fight back, when an A7V fired its 57mm gun at them. Then a Mark IV ‘male’ arrived on the scene firing its 57mm guns, causing the A7V to run onto a side slope and overturn. This historic incident provided an early illustration of the need to arm tanks so that they could fight other tanks, as well as an indication of the indifferent performance of the A7V over uneven ground.

The three A7V detachments continued to be employed right up to the end of the war, but their impact was very limited because of the small number of tanks they could deploy. Although they absorbed the whole of the production of A7Vs, this only amounted to 20 tanks of the 100 originally ordered. The shortage of indigenous tanks made the German Army use captured British Mark IV tanks, with which it formed six detachments of five tanks each by the end of the war and was planning to form six more. However, even if these plans had been implemented the number of tanks the German Army had would only have been increased to about 75.

Post-War Anticlimax

The initial use of tanks by the British and French armies was followed during the latter part of the First World War by considerable further development of them and ambitious plans for their large-scale production and employment. But all these activities and plans were drastically scaled down or abandoned when the hostilities on the Western Front ended in November 1918.

In Britain the downturn was made very evident by what happened to the medium tanks that followed the original Medium A. The first was Medium B, which had already been designed in 1917 and 550 of which were ordered by November 1918. However, the orders were then cut down and only 80 were built. Even more drastic reductions were made in the case of Medium C, orders for which had risen by September 1918 to as many as 3,230 tanks only to be cancelled when the war ended. Not more than 36 or 48 were eventually completed.

Both tanks combined in a lighter form the rhomboidal track layout of the British heavy tanks with a fixed superstructure, or turret similar to that of the Medium A, in which were mounted four machine guns. They were heavier than the Medium A, weighing 18 and 19.5 tonnes respectively, and neither was faster, but they were easier to drive by one man, having a single engine.

Medium C was considered to be the best of the British tanks produced during the war and the only one to remain in service for some time after the war – until 1925 in fact. However, like all earlier British tanks it still had unsprung track rollers that resulted in a very harsh ride over broken, hard ground and limited its maximum speed.

This shortcoming was only rectified in the last British tank to be conceived during the war, the Medium D.

Medium D stemmed from experiments carried out by an engineering unit of the Tank Corps commanded by Major P. Johnson, which was charged with making improvements to tanks. Its principal objective was a major increase in their speed, and by installing a much more powerful Rolls-Royce aero engine in an existing heavy tank the unit demonstrated that on suitable ground it could attain 15mph, or almost four times its normal speed, in spite of its unsprung tracks. This as well as other experiments led to a conference on 28 April 1918 at the headquarters of the Tank Corps attended by Colonel F. Searle and Colonel J. F. C. Fuller, who were its chief engineer and chief of staff respectively, at which plans were made for the design of a new medium tank capable of a speed of 20mph. The task of developing such a tank was entrusted to Johnson and it was designated Medium D.

A month later Fuller produced a paper entitled ‘The Tactics of the Attack as Affected by the Speed and Circuit of the Medium D Tank’. In it Fuller proposed that the speed and range that the Medium D tanks were expected to have be exploited by making them burst through enemy lines and advance well beyond to attack enemy headquarters to bring about a collapse of the enemy’s command system. This would create confusion and the disorganized enemy forces would then be crushed by an assault by heavy tanks and infantry.

Fuller’s proposal was accepted in a modified form by the War Office and in July 1918 led to a ‘Memorandum on the Requirements for an Armoured Striking Force for an Offensive in 1919’, which was endorsed by the chief of the Imperial General Staff and which was approved by the Allied Commander-in-Chief, General Foch. The memorandum called for no fewer than 10,500 British, French and American tanks, or almost four times the total wartime British production, which amounted to 2,636 tanks.

However, the grandiose plans for the large-scale production and use of tanks were abandoned when the war ended. In consequence, Plan 1919 was never put to test. In fact, it could not have been put into effect in 1919 as its key element, the Medium D tank, was nowhere near being ready for use, and it could not have been even if its design had been entirely satisfactory, given the time required to develop and to produce it. The much-lauded Plan 1919 was not therefore entirely realistic.

In fact, by the time the war ended development of the Medium D had only reached the stage of a wooden mock-up. Subsequently the construction of ten was ordered, although only seven appear to have been completed, the first in mid-1919 and the last in 1920. On trials they exceeded the required speed of 20mph and they could be made to float, but they also incorporated some questionable and troublesome design features. These included an unusual suspension with a steel cable connecting all the track rollers on one side to a single spring and a very odd type of track with wood-faced track plates that could swivel to conform to the ground. Originally the designers also expected the functions of the tank commander and of the driver to be performed by one man, which was hardly practicable, and they were only armed with machine guns, although the installation of a 57mm gun was considered.

Nevertheless, in January 1920, Fuller, who was by then dealing with tank affairs at the War Office, recommended the adoption of the Medium D and also of a Light Infantry Tank, which still had to be built. The latter was a lighter version of the Medium D that weighed 7.5 instead of 13.5 to 14.5 tonnes and had laterally flexible ‘snake tracks’ with lubricated ball joints between the track plates. When tested in 1922, the Light Infantry Tank attained what was then a record speed for tracked armoured vehicles of 30mph. Whether an infantry tank needed to be so fast may be doubted, but Fuller produced a curious argument that it had to be fast to be able to protect the infantry as a destroyer protects a convoy at sea.

In addition to the Light Infantry Tank Johnson also designed an even lighter tank, the 5.5-tonne Tropical Tank, which was intended for use in India and had a peculiar configuration with two slightly staggered side-by-side machine gun turrets, like the Austin armoured cars produced earlier to Russian requirements.

However, development of the Medium D and its offshoots came to an end in March 1923 when the government-run Department of Tank Design and Experiment that Johnson headed was closed and work on tanks was passed on to industry. Whatever the reasons for this change of policy, the decision to abandon Medium D was not very surprising as after five years of development it had still not reached the stage of being ready for use, as even Fuller who was its chief protagonist admitted.

As a result of it all, the number of tanks built in Britain during the five years that followed the end of the First World War was reduced to a mere handful once the leftovers of the wartime Medium B and C programmes were completed. During the same period the number of tank units was also reduced drastically, from 26 tank battalions in November 1918 to five battalions in March 1920.

In contrast, the French Army retained a relatively large tank force. In fact, for several years it was the world’s largest, although equipped almost entirely with the Renault FT light tanks. Having already ordered 4,000 of them, in 1918 the French Army was planning to acquire more, which would have brought the total number of Renault FTs ordered since production began to 7,800. The number actually ordered by the Armistice of 1918 reached 4,635. As a result of this and production not ceasing immediately, in 1921 the French Army had no fewer than 3,737 Renault FTs, in spite of losing a number in battle and supplying or selling some to other armies.

But as soon as the hostilities ceased, the French Army relinquished the claim to its share of the Anglo-American-French heavy tank programme that it had originally hoped would provide it with 1,285 tanks. It also stopped the production of 2C heavy tanks, 300 of which were on order. Only ten of these 68-tonne tanks were eventually completed in 1921, to become the heaviest tanks in use for almost two decades. During this period there were claims, repeated in more recent years, that the French Army developed even heavier 74-tonne 3C tanks armed with a 155mm gun. But all that actually happened was an experimental installation in 1923 of a 155mm howitzer in place of the 75mm gun in one of the ten 2C tanks.

The German Army’s plans for a large-scale production of tanks were brought to an abrupt end in 1918 by its defeat. One of the tanks affected by this was the A7V-U, which was based on the components of the original German tank but had a rhomboidal track layout similar to that of the British heavy tanks to remedy the poor cross-country performance of the A7V. A prototype of it was built by June 1918 and 240 were subsequently ordered for delivery by June 1919, but none of them was built.

Similarly, none of the light tanks that the German Army was planning to use advanced beyond the construction of prototypes. The first was LK-I, at 7-tonne vehicle based on a large passenger car chassis that was fitted with tracks, an armoured body and a small machine gun turret. Its development started in September 1917 and although it did not proceed beyond the construction of a prototype, the latter served as the basis of the next light tank, the LK-II, which was similar but had a fixed turret with a 57mm gun. An order for 580 LK-IIs was issued in June 1918 but only two prototypes were built before the war ended. LK-II was to be followed in 1919 by LK-III, which was a further development of the previous model, but none was ever built.

As early as June 1917 the German High Command also issued an order for 10 super-heavy breakthrough tanks, which were designated K-Wagen. They came to weigh no less than 148 tonnes and were armed with four sponson-mounted 77mm guns. Two were nearing completion in a Berlin factory when the war ended and were destroyed by the Allied Control Commission.

While the war was still going on the German High Command expected, somewhat optimistically, that production of tanks would rise in 1919 to 4,000 light and 400 heavy tanks. However, the end of the war not only brought all the plans and expectations to an end, but Article 171 of the Versailles Treaty imposed in 1919 on Germany by the victorious Allies prohibited it having any tanks. As a result, all that came out of the German wartime tank programmes were the components of the LK-II, which were incorporated in the ten Strv/21 tanks built in Sweden in 1921. Several writers have claimed that some of the surviving A7Vs were given after the war to the Polish Army but there has been no evidence that would support these claims.

The United States started to build tanks later than Germany but had the advantage of being allied to Britain and France, which were already producing them when it entered the war in the spring of 1917. It could therefore draw on their experience and even obtain some tanks from them. Thus, when the US Tank Corps was formed in 1917 its first two battalions were equipped with Renault FTs supplied by the French Army and the third battalion was equipped with British Mark V tanks.

Ambitious plans were drawn for the expansion of the US Tank Corps to 45 battalions, which were to be equipped with Renault FT light and Mark VIII heavy tanks. The latter were the last and at 37 tonnes the heaviest of the British rhomboidal tanks and they were to be produced in France at a factory specially built for the purpose following an agreement between the British and US governments. A total of 1,500 Mark VIII was to be produced there during 1918. Of these the first 600 were to go to the US Army while the French government, which had joined the agreement, claimed the remaining 900.18 In addition 1,500 Mark VIII tanks were to be produced in the United States. But the Armistice caused all these production plans to be abandoned and eventually only 100 Mark VIII tanks were assembled in 1920 in the United States. Similarly, plans to produce 1,375 Mark VIII tanks in Britain were abandoned and only 11 were completed there.

To meet the requirements of its light tank battalions, the US Army awarded contracts to three American companies for the production of 4,440 of a US version of the Renault FT. Some were completed by the Armistice but none arrived in France before it occurred, and the contracts were subsequently cut down with the result that only 952 tanks were completed as the Six Ton M1917 light tanks. Still more light tanks were to have been produced as the result of an order for 1,000 Ford Mark I tanks, which were similar to the Renault FT, but only one of these was completed.

The only other country to embark on the production of tanks during the war was Italy, which was no exception to the drastic cuts in the orders for tanks that were made when the war ended. The tank Italy produced was another version of the Renault FT, which was designated Fiat 3000 and 1,400 of which were ordered in 1918 from Fiat and Ansaldo. But after the Armistice the order was reduced to 100 tanks, which were built between 1919 and 1921.

The large numbers of tanks that the major belligerents were planning to produce in the latter part of the war attested to the importance attached to them at the time. By the same token, the drastic reductions in the number of tanks after the war reflected a contemporary decline in their standing as well as the state of the post-war economies and the changes in the political situation. Once this happened it would be several years before tanks recovered their importance.

In the meantime tanks aroused considerable interest around the world, although without gaining commensurate military recognition. During the war their use was confined to Western Europe, except for a few British Mark I and Mark II tanks used in the Second and Third Battles of Gaza in 1917. However, after the war their use spread worldwide as they were acquired by different countries, but invariably in small numbers.

In almost all cases the tanks were the French Renault FTs, which were produced during the war in the largest number and which were the only tanks available in quantity after the war. They were generally well regarded and were well suited to the close support of the infantry, which dominated the contemporary tactical thinking. Moreover, they were relatively simple and economical to operate.

The largest number of the Renaults went to Poland. It amounted to about 120 tanks, which formed the equipment of a regiment organized in 1919 by the French Army as part of a Polish corps created in France to aid the newly independent Polish Republic. The tanks took part in the Polish-Russian War of 1920–21 but were not suited to its fluid character and made little impact on it by their small-scale actions.

The 120 tanks given by France to Poland constituted, for a time, the fourth largest tank force in the world, which highlights the small size of the tank forces after the First World War. Further illustration of this is provided by Italy, which had the next largest tank force of 100 Fiat 3000s. Belgium, which followed in terms of size in Europe, had a force of 49 tanks, while Finland bought 32 Renault FTs in 1919. Still smaller numbers were acquired by half a dozen other European countries, including Switzerland, which bought two, and Sweden, which bought one for evaluation.

A few more were delivered farther afield, with Brazil ordering 12 in 1919 when the Imperial Japanese Army procured a similar number of Renaults from France as well as some Medium A tanks from Britain. Soviet Russia also acquired some Renaults by capturing tanks sent by France to support the anti-Bolshevik forces during the Russian Civil War, to which were subsequently added 15 copies built at the Sormovo plant by a remarkable piece of ‘reverse engineering’. The Red Army also captured 25 Mark V heavy tanks sent by Britain to support the White armies. But all this only brought the total number of tanks possessed by the Red Army in 1923 to 77.

Only France had a large number of tanks for some time after the First World War, having been left with more than 3,000 Renault FTs. This was more than the total number of tanks that the rest of the world had at the time, and together with the prestige that the French Army enjoyed after the war made its ideas prevail for several years. But its stock of wartime tanks was a wasting asset, and after a time other ideas emerged elsewhere and gradually gained strength, even though the number of tanks outside France remained small.

Britain’s Lead and Failings

The aftermath of the First World War produced a variety of ideas and opinions about the future of tanks. At one extreme were views that tanks would be of no further use. At the other extreme there were claims that in the future existing armies would be replaced by fleets of tanks.

An example of the former attitude is the often quoted remark made by Major General L. Jackson, Master General of the Ordnance, in December 1919 at the Royal United Service Institution that ‘The tank proper was a freak. The circumstances which called it into existence were exceptional and are not likely to recur’. To some, evidently, the usefulness of tanks was confined to trench warfare and the latter was not expected to occur again.

The other extreme is exemplified by a paper entitled ‘A tank army’ written during the war by Captain G. le Q. Martel, who assisted Colonel Fuller at the headquarters of the Tank Corps. It described a future army composed almost entirely of different types of tanks corresponding to the principal types of contemporary warships. Fuller himself adopted similar ideas and immediately after the war began to write of ‘tank fleets’ and of battles that in future would ‘more and more approximate to naval actions’.

The appeal of the naval model is understandable in view of the fact that warships represented an earlier form of mobile weapon platforms, which is what Fuller rightly recognized tanks to be.4 However, the environments in which warships and tanks operated were obviously very different. Armies could not, therefore, be expected to operate on land as warships did at sea. Nevertheless, as late as 1931 Fuller was still forecasting that tanks ‘will operate on somewhat similar lines to a fleet at sea’.

The policy that armies generally adopted was to accept tanks but only as an auxiliary of the infantry and operating at the pace of the latter. A possible advance on the prevailing stance was indicated by a study produced in 1919 by General Estienne at the request of the French commander-in-chief. In it Estienne foresaw the existing Renault FT light tanks being replaced by more powerful chars de combat that would play a leading role in future battles. Two years later Estienne enlarged on his views in a lecture delivered in Brussels in which he spoke of the potential strategic and tactical advantages of a future mechanized army of 100,000 men that would include 4,000 tanks and armoured infantry and was capable of moving 80km in one night. However, his views were ignored. In particular, while he and a few other French Army officers advocated the creation of an independent tank arm, the headquarters of the artillerie d’assaut that provided tank units with a degree of autonomy were abolished in 1920. Instead, tank units were put under a subdivision of the Infantry Department, which stultified further tactical and technical development.

A similar situation arose in the United States, where the wartime Tank Corps was abolished under the National Defense Act of 1920 and tanks were assigned to the infantry, becoming its auxiliaries. In keeping with this the General Staff declared in 1922 that ‘the primary mission of the tank is to facilitate the uninterrupted advance of the rifleman in the attack’.

Apart from France and the United States only Britain had the means at the time to develop the use of tanks further. In consequence it was left to the British Army to take the lead in the development of a more mobile and effective use of tanks.

The lead that the British Army took was due to a large extent to the conjunction of two events. One was the establishment of the Royal Tank Corps, which in 1923 succeeded the wartime Tank Corps and became a separate arm, due to a considerable extent to the efforts of Fuller. Its strength amounted to only four tank battalions and some armoured car companies, but its status provided a degree of freedom to explore new methods of operation free of the constraints of infantry tactics.

The other event underlying the British Army’s lead was its acquisition of tanks that were in advance in several respects of other contemporary tanks and that lent themselves to the development of new tactics. One of their features was a considerably higher speed than that of the earlier tanks, which was a consequence of them being designed as an alternative to Johnson’s Light Infantry Tank ordered in 1920 at the instigation of Fuller. This prompted the War Office department responsible for the procurement of equipment to order another light tank from the Vickers company. Experimental versions of both tanks were built and tested by the end of 1921, with Johnson’s proving capable of more than 20mph while Vickers’ was slower than the wartime Medium C (which had a maximum speed of 7.9mph), as Fuller gleefully observed in his memoirs.

However, in other respects Vickers’ tank was better, in spite of the praise bestowed on Johnson’s model. In particular, its general configuration was superior, being more like that generally adopted for tanks later, and it was the first British tank to have a rotating turret, while Johnson’s still had a fixed superstructure with a fighting compartment similar to that of the Medium D, which was considered unsatisfactory by General Elles, the wartime commander of the Tank Corps. Moreover, Johnson’s tank was only armed with machine guns while Vickers’ also had a 47mm gun. Otherwise both tanks represented an advance on the wartime British tanks in having sprung suspensions instead of rigidly mounted track rollers, but that of the Vickers tank was more robust.

Low speed, the one major shortcoming of the Vickers tank, was due to its use of an unconventional hydrostatic transmission. This type of transmission was used successfully on warships, but Vickers’ designers did not appear to appreciate how inefficient it would be when used to drive vehicles, with the result that much of the engine power was dissipated as heat and far less of it was consequently available to drive a vehicle.

As a result of its poor automotive performance Vickers’ original tank was abandoned in 1922, as was Johnson’s development work. However, in the same year Vickers came up with a second tank and this was adopted by the War Office as Vickers Light Tank Mark I, although it became far better known by its later name of Vickers Medium.

The first Vickers Medium was delivered in 1923. It weighed 11.75 tonnes and looked as if it had been hastily put together by mounting the turret of a Rolls-Royce armoured car on the chassis of a high-speed artillery tractor. However, it retained the best features of Vickers’ original design, which included a 47mm gun mounted in a rotating turret that was large enough to accommodate not only a gunner but also a tank commander free to exercise tactical control and ensure a more effective use of the tank. At the same time Vickers Medium was almost as fast as Johnson’s tank, its nominal maximum speed being 18mph but in practice it was capable of more than 20mph.

Eventually 166 Vickers Medium Tanks Marks I and II were built for the British Army, which was just enough to equip the tank battalions of the Royal Tank Corps, and they were the only new tanks to appear in quantity anywhere in the world between the end of the First World War and 1929. During this period they were also by far the fastest tanks in service, their maximum speed being almost four times that of the typical contemporary tank, which was still the Renault FT. The Royal Tank Corps was therefore uniquely well equipped to develop new, more mobile methods of employing tanks and to some extent these were driven indirectly by its tanks.

At first new ideas about the employment of tanks came primarily from Fuller, who wrote extensively on the subject. The writings started in 1919 with an essay which won a Royal United Service Institution competition and in which Fuller proposed a ‘New Model Army’ built around the capabilities of tanks. The divisions of this army were to incorporate 12 infantry battalions each with an integral tank company as well as a divisional tank battalion and two regiments of horse cavalry. This amounted to a surprisingly gradualist proposal for the future use of tanks, although ultimately Fuller expected tanks to replace infantry and cavalry.

The publication of Fuller’s essay led to a meeting with Captain B. H. Liddell Hart and a long association between the two. In 1922 Liddell Hart followed Fuller by also writing about a ‘New Model Army’ but proposed a more practical organization for its divisions, which were to have separate tank and infantry battalions – the latter in armoured carriers – and no horse cavalry. But he did not differ greatly from Fuller in expecting further mechanization to lead to ground forces being ‘composed primarily of tanks’. However, he did not propose to dispense entirely with the infantry, a small contingent of which would be retained as ‘land marines’.

Like Fuller, Liddell Hart wrote extensively on the use of tanks and related matters, and both assisted the development of new methods of employing tanks through personal contacts and by publicity, particularly in the case of Liddell Hart who in 1925 became the military correspondent of The Daily Telegraph. Their writings made Fuller and Liddell Hart well known internationally, and on the strength of their writings they came to be regarded as the apostles of mechanized warfare.

However, the actual development of new and more effective methods of using tanks was carried out by others. It began with a memorandum written in 1924 by Colonel (later Brigadier) G. M. Lindsay, the inspector of the Royal Tank Corps, in which he proposed the establishment of an ‘Experimental Mechanical Force’. As nothing happened, Lindsay repeated his proposal in another memorandum, which he submitted to the chief of the Imperial General Staff, General Milne, through Fuller, who was Milne’s military assistant. Milne agreed with Lindsay’s ideas and consequently an Experimental Mechanized Force was assembled in 1927 on Salisbury Plain. Fuller was offered the command of it but rejected the offer because of dissatisfaction with some of the administrative arrangements, losing thereby the opportunity of putting some of his ideas into practice.

In spite of this, the composition of the Experimental Mechanized Force reflected the ideas of Fuller as well as those of Lindsay, who wanted it to be predominantly a force of tanks and other armoured vehicles. Its principal components were therefore a battalion of Vickers Medium tanks and a mixed battalion of armoured cars and tankettes, which were supported by four batteries of motorized and one of self-propelled artillery and a motorized engineer company. There was no infantry in it except for a motorized machine gun battalion that could only play a relatively passive role of holding ground.

To some extent the Experimental Mechanized Force was a scratch collection of the available units and its vehicles were of several different types, which made it difficult to co-ordinate the action of its components. Nevertheless, it was the first mechanized formation ever to be assembled and its organization and operational trials aroused considerable interest in Europe and the United States.

Experiments in which the Experimental Mechanized Force took part in 1927 were followed by others during the 1928 training season, by which time it was renamed the Armoured Force, but it was then dissolved. A conclusion drawn from the experiments that had been carried out was that the unarmoured components of the force were a drag on the armoured ones, which reinforced the idea that mechanized formations should consist almost entirely of tanks.

This idea was embodied in the first armoured force manual entitled Mechanised and Armoured Formations and popularly known as the ‘Purple Primer’, which was issued by the War Office in 1929. The manual was drafted by a Royal Tank Corps officer, Lieutenant Colonel C. Broad, and envisaged a future army that would include light and medium tank brigades consisting very largely of tanks as the principal mechanized formations. Their proposed composition severely limited the ability of the brigades to carry out independent operations but, nevertheless, an ‘all-tank’ brigade became the basis of further experiments.

A tank brigade was actually formed on a provisional basis when experiments were resumed during the 1931 training season. It consisted of three mixed battalions of light and medium tanks and one of Carden Loyd machine gun carriers, which were used in lieu of light tanks of which there was a shortage. The relatively homogeneous composition of the brigade made it easier to develop new methods of controlling and maneuvering tank units, which involved among other things pioneering the use of radios, which began to be available in 1929. By the end of the 1931 training season the brigade demonstrated that it could maneuver as a whole and not merely operate as so many individual tanks.

The Tank Brigade was assembled again in 1932, and after a break was reassembled and put on a permanent footing in 1934. For the following four years it constituted the only mechanized formation of the British Army and contained most of its tanks. In the course of its existence it made important advances in the technique of mobile mechanized operations, but it was clearly not a self-contained formation of several complementary arms capable of a variety of offensive and defensive operations, instead being only capable of strategic maneuvers that seem to have been expected to yield success without too much fighting.

The emphasis on operational mobility rather than tactical effectiveness based on fighting ability, which characterized the atmosphere in which the Tank Brigade was created and developed, also applied to the design of British tanks during the 1920s and 1930s.

The first tank to come after the Vickers Medium was the outcome of an apparent if only temporary revival of interest in trench warfare by the War Office, which in 1922 asked Vickers to produce the design of a heavy tank that would replace the wartime Mark V. It was to be turretless but have a hull-mounted 47mm gun and small sponsons with machine guns, which made its configuration resemble that of the wartime Mark VI designed in 1917 but never built, and showed that the War Office was still thinking in terms of the original types of tanks. In contrast Vickers offered as an alternative a very original design. This was accepted, the resulting tank being given the A.1 designation and later called Independent. Its principal feature was that it had as many as five turrets: a main turret mounting a 47mm gun and a machine gun and manned by a crew of three, and around it four small, one-man machine gun turrets. A.1 was not the first tank to have more than one turret, as the French 2C heavy tank already had a second turret at the rear of its hull and US Model 1921 and Model 1922 experimental medium tanks had a small machine gun turret on top of their main turret. However, A.1 was the first tank to have more than two turrets. As such it aroused considerable interest but only one other tank, the Soviet T-35, followed its example in having five turrets, although several other tanks built later had three turrets.

Vickers Medium Tank MK II

Auxiliary turrets apart, the general configuration of the A.1, which incorporated a driver’s station in the front of the hull and an engine compartment at the rear, represented a considerable advance on that of the Vickers Mediums. But, in spite of its weight of 32 tonnes, its main armament was not more powerful than theirs and its armour was only slightly thicker than that of the Renault FT light tank. However, it was relatively mobile, having a maximum speed of 20mph.

Only one A.1 was actually built, but it attracted attention around the world when it appeared in 1926 at a large-scale demonstration of armoured vehicles staged for the benefit of the British government and Commonwealth prime ministers. The ‘Independent’ name later given to it led to suggestions that it was intended for strategic strikes carried out independently by mechanized forces, but there is no evidence of this.

The construction of the A.1 was followed by the development of a new medium tank, which was designated A.6 but came to be known generally as the ‘Sixteen Tonner’. The A.6 was designed at Vickers to an outline specification produced by a committee of the Royal Tank Corps of which Fuller was a member, but followed the general configuration of the A.1, or Independent, and like the latter was actually designed by C. O. Woodward working under the general direction of Sir George Buckham. However, it had only two auxiliary machine gun turrets instead of four. Its main turret again mounted a 47mm gun and was large enough not only to accommodate the optimum size crew, consisting of a commander, gunner and loader, but also an observer, whose inclusion was an indulgence in view of the extra space and weight that this involved.

The first two of three prototypes of the A.6 were ready in 1928 and it was generally highly regarded. In fact, a 1930 War Office document described the A.6 as ‘probably the best medium tank in the world’. In spite of this the A.6 was not adopted by the British Army. Instead, a decision was taken in 1928 to base on it a new Mark III medium tank. This turned out to be very similar to the A.6 except for the main turret, which had a crew of three instead of four and a bustle to house the radio that had come into use.

Two Medium Mark III tanks were built by the Royal Ordnance Factory at Woolwich in 1929 and one by Vickers in 1931. Trials of them were successfully completed by 1933 but General A. Brough, who became Director of Mechanization in 1932, decided to abandon the development of the Mark III because it was considered too expensive to produce in any quantity, particularly in the prevailing economic circumstances. Instead, he decided to develop a simpler and less expensive medium tank. His decision has been severely criticized later, to the extent of being called ‘a fatal mistake’. In fact, a similar decision to build a simpler and less expensive medium tank than the Mark III had already been taken in 1928 by Brough’s predecessor. It resulted in the A.7 tank, two of which were built by the Royal Ordnance Factory by the end of 1929.

The A.7 very sensibly dispensed with the auxiliary machine gun turrets, which were replaced by a single machine gun simply mounted in the front of the hull and operated by a gunner sitting alongside the driver, while the main turret was manned by a crew of three. This meant that the configuration of the A.7 was basically the same as that adopted later during the Second World War for most tanks and in advance of that of the Mark III Medium. In other respects, such as armour and main armament, the A.7 did not differ from the Sixteen Tonner and the Mark III. It could therefore have been developed into a medium tank that was as effective as the Mark III but was simpler, lighter and should have cost less to produce. However, it was not adopted, although a number of its features were incorporated later in other tanks.

When the decision was taken to abandon the Mark III, the development of a simpler and less expensive medium tank was started afresh in 1934 at what had become Vickers Armstrongs after Carden Loyd Tractors’ takeover in 1928. It was carried out under the direction of Sir John Carden, who came to be highly regarded as the designer of the Carden Loyd machine gun carriers and light tanks. Carden decided that the new tank should still have two auxiliary machine gun turrets, like the Sixteen Tonner and the Mark III. For its main turret he adopted more wisely a three-man turret similar to that of the A.7 and designed a tank that did not differ from the latter in terms of its 47mm gun main armament, armour and maximum speed, but had the looks of the Sixteen Tonner and the Mark III.

When the prototype of the new tank appeared in 1936 under the A.9 designation nobody seemed to like it. There was a strong case therefore for the development of another tank that would replace the Vickers Mediums, which were becoming obsolete but were still virtually the only gun-armed tanks of the Royal Tank Corps. However, instead of concentrating on the development of a better medium tank, the available engineering resources were split up by a decision to divide tank units into two separate categories.

One of them was to provide close support for the infantry, and in 1934 one battalion of the Royal Tank Corps was separated from the rest and assigned to this role. At the same time Vickers Armstrongs were asked to produce a tank specifically for infantry support. The initial response to this demand was the A.10 tank, which was very similar to the A.9 but had armour up to 30mm instead of 14mm thick and was relieved of the auxiliary machine gun turrets. But although its armour was thicker than that of the medium tanks it was not considered sufficient for the infantry support role. In view of this and the contemporary financial restrictions, Carden proposed a very different type of tank that would be much more heavily armoured and at the same time cheap to produce. The idea of such a tank was accepted in 1935 and led to the A.11 infantry tank, which appeared in prototype form a year later.

The A.11 was a slow, 11-tonne vehicle with a one-man turret mounting a single machine gun. Its frontal armour was up to 65mm thick, which put it ahead of most other contemporary tanks, but otherwise it was a throwback to the First World War, conceptually little different from the Renault FT. Nevertheless, it was the type of tank favoured by General Elles, who in 1934 became the Master General of the Ordnance and as such was able to direct tank development. The A.11 was consequently adopted as Infantry Tank Mark I and Vickers Armstrongs proceeded to produce 139 of it.

However, the shortcomings of the A.11 quickly became evident and in 1936 a decision was taken to design a successor to it, which became the A.12 and later Infantry Tank Mark II and which was usually called Matilda. Design of the A.12 was carried out by the Royal Ordnance Factory in collaboration with the Vulcan Foundry and was based on the A.7 mentioned earlier, except that it had no hull machine gunner. Because no sufficiently powerful engine was available at the time in Britain, the A.12 followed the example of the A7E3 version of the A.7 in being powered by two bus-type diesels geared to a common output. It also incorporated novel features of its own, such as dispensing with the angle-iron frame to which armour plates were previously riveted in all other British tanks. Instead its castings and plates were bolted together, thereby saving weight. At 78mm its frontal armour was thicker than that of any other contemporary tank and made it immune to the existing anti-tank guns. The armour made it weigh 26.5 tonnes, which was more than the weight of any British tank since the A.1 Independent, but neither this nor its low maximum speed of 15mph detracted from its effectiveness. In fact, ignoring the limitations of its intended role, the A.12 Matilda was the most successful British tank design of the 1930s.

The one major shortcoming of the A.12 was its main armament, which consisted of a 40mm 2-pounder gun that succeeded the obsolescent 47mm 3-pounder at the time of its development. As a weapon against enemy tanks the new gun was comparable to the best of the contemporary tank and anti-tank guns, but this was achieved by firing solid, armour-piercing shot that was relatively ineffective against anti-tank guns, weapon emplacements and similar targets. What was needed, particularly for a tank that was to support the infantry, was high explosive ammunition, but this was not provided for the 40mm gun, although Renault FT already had high explosive ammunition for its 37mm gun 20 years earlier.

An even better solution would have been to arm the A.12 with a larger calibre, dual purpose gun. A small proportion of the A.12, which were designated ‘close support tanks’, was in fact armed with a 3in. (76.2mm) howitzer instead of the 40mm gun, and this was provided with high explosive rounds, but its main role was to fire smoke shells.

Another problem with the A.12 Matilda was the lack of experience in the development of tanks and the limited resources of the Vulcan Foundry, which was entrusted with its production because the only experienced producers of tanks, Vickers Armstrongs, were already fully occupied with other work. As a result only two A.12 Matildas were completed by the outbreak of the Second World War.

While the A.11 and the A.12 Matilda were being developed for infantry support, tanks were also needed for the other category of tank units that formed part of the mobile armoured forces. By 1937 these took the form of the Mobile Division, which incorporated the Tank Brigade but was not an ‘all-tank’ formation of the kind Fuller and some of the protagonists of mechanized forces advocated. However, it was not an effective all-arms combat formation either. In fact, it was still considered to be a mobile force for sweeping flanking maneuvers rather than direct confrontation with the enemy’s main forces. In this respect it could be, and was, regarded as a successor of the cavalry divisions, and its role was limited to that to which horse cavalry was reduced during the 19th century. All this influenced the characteristics of the tanks developed for the Mobile Division and its successors.

The most powerful tanks already being produced for the Mobile Division were the A.9 medium tank, which was re-branded a ‘cruiser tank’, and the A.10, which was not considered to have sufficient armour for an infantry tank and which became a ‘heavy cruiser’ although it was only 1.75 tonnes heavier than the A.9. The A.9 had a maximum speed of 25mph, which was not considered fast enough for the Mobile Division, and this applied even more to the 16mph maximum speed of the A.10. However, in the absence of other candidates, both were adopted for limited production. The first was delivered in 1939 and eventually the number built totalled 295 tanks.

In the meantime another and more mobile cruiser was developed following a visit to the Soviet Union in 1936 by Martel, who was by then Assistant Director of Mechanization at the War Office. During the visit Martel attended Red Army manoeuvres and became impressed with the Soviet BT tanks and in particular with their suspension. It was apparently new to him, although the experimental tank built in the United States by J. W. Christie on which the BTs were based had already attracted widespread interest in 1928 when it set up a speed record of 42.5mph. This prompted the US Army to order five tanks from Christie in 1931 and to take over two more ordered by the Polish government, which had defaulted, while the Soviet authorities reacted even earlier by ordering two chassis in 1930. However, it was only eight years after the appearance of Christie’s high-speed tank that Martel took note of it and proceeded to advocate the development of a cruiser tank based on it. To speed this up, Martel arranged the purchase of a vehicle that Christie still happened to have by the Morris car company, and its head, Lord Nuffield, undertook to develop the new cruiser tank. A new company called Nuffield Mechanization was set up for this purpose with the approval of General Elles, who was still Master General of the Ordnance and who wanted to create competition for Vickers Armstrongs, who until then enjoyed virtual monopoly in the production although not in the design of tanks.

Nuffield Mechanization worked with remarkable speed and, although they had not produced tanks previously, built the first prototype of the new tank within 12 months of receiving an order for it. The tank, which was designated A.13 and later Cruiser Tank Mark III, was very different from Christie’s tanks. In particular, it had a very different and much more sensible configuration similar to that adopted earlier for the A10E1 version of the A.10 and almost simultaneously for the A.12 Matilda. The only thing in common with Christie’s tanks apart from the suspension was the Liberty engine, a First World War American aircraft engine whose production was revived by the Nuffield organization. This V-12 engine developed 340 horsepower, which made it more powerful than any engine available for British tanks since the A.1 Independent of the mid-1920s and provided the A.13 with a high power-to-weight ratio of 24hp per tonne. As a result the A.13 was faster than all earlier British medium or cruiser tanks, having a nominal maximum speed of 30mph and in practice being capable of almost 40mph.

The armour of the A.13 cruiser was still no thicker than that of the Sixteen Tonner or Carden’s A.9, but its maximum thickness was doubled to 30mm on the second version. The need for heavier armour led to the idea of a ‘heavy cruiser’, which originated with the A.10, and the design of two different versions of such a tank was ordered in 1938. One of them, the A.14, was designed with the co-operation of the London, Midland and Scottish Railway (LMS) company and the other, the A.16, was designed by Nuffield Mechanization. They differed in engines, transmissions and suspensions but their general configuration was the same, and in addition to the main turret both had two auxiliary machine gun turrets, like the Sixteen Tonner, after which those who ordered them apparently still hankered.

The maximum thickness of armour of the A.14 and A.16 was 30mm, which by the time they were built was not more than that of the second version of the A.13. They did not, therefore, offer any advantage so far as armour protection was concerned, nor were they better armed as their main armament still consisted of a 40mm gun. In consequence the development of both was, very sensibly, abandoned.

However, when the A.14 was abandoned LMS were asked to design a simpler cruiser tank with the same layout and Christie suspension as the A.13 but with armour up to 40mm thick. This tank was called Covenanter and had a special, 12-cylinder horizontally opposed engine and a significantly lower silhouette but was still armed with a 40mm gun. Nuffield Mechanization designed its turret, but instead of participating in its production offered in mid-1939 to design their own version of a ‘heavy cruiser’ from the basis of the A.13 and using the Nuffield Liberty engine. The offer was accepted and in August 1939, just one month before the outbreak of the Second World War, an order was given for the production of the tank, designated A.15 and later called Cruiser Tank Mark VI Crusader. The A.15 resembled the Covenanter but was larger and somewhat heavier, weighing 19 compared with 18 tonnes. It was armed with the same 40mm gun as the Covenanter and, showing how hard old habits die, it still had one auxiliary machine gun turret, which resulted in it having a crew of 5.

The eight medium and cruiser tanks developed between 1934 and 1939 differed from each other in several respects except for their main armament, which in all cases consisted of the same 40mm 2-pounder gun. This showed that, by comparison with all the effort devoted to the development of engines, transmissions, suspensions and other components, little attention was paid to the development of more powerful armament. In particular, no attempt was made to arm any of the medium or cruiser tanks with a dual purpose 75 or 76mm gun comparable to those mounted in medium tanks that were being developed by then in at least two other countries.

To be fair, in some of the medium and cruiser tanks the 40mm gun was replaced by what was originally called a 15-pounder mortar and then a 3.7in. (94mm) howitzer and later by a 3in. (76.2mm) howitzer. But these were limited purpose weapons intended primarily to fire smoke shells, as already mentioned in connection with the Matilda infantry tank, although they were also provided with some high explosive rounds. However, they were not provided with armour-piercing ammunition. They were not therefore comparable to the dual purpose 75 or 76mm guns mounted in Soviet and German tanks with which they were often wrongly equated. The latter were, admittedly, short-barrelled low-velocity weapons, but they could still knock out contemporary tanks, if only by smashing their relatively thin armour by the sheer mass of their projectiles. At the same time they could effectively engage anti-tank guns, machine gun emplacements and similar targets with high explosive shells.

The lack of medium calibre dual purpose guns did not seem to concern those involved with the development of British medium and cruiser tanks, who thought more in terms of sweeping mobile maneuvers rather than of fighting hostile armoured forces and, even less, of engaging in all phases of offensive operations. In consequence, they only expected tanks to be armed with ‘one small gun and several machine guns’, to quote a contemporary opinion. Such views contributed, among others, to the repeated attempts to develop tanks with additional machine gun turrets. As to the calibre of tank guns, a 1937 Tank Brigade report endorsed by the General Staff stated emphatically that a gun larger than the 40mm 2-pounder was not required.

To make matters worse, the 40mm gun was not provided with high explosive ammunition with which it could engage unarmoured or ‘soft’ targets with some degree of success. In contrast to the corresponding 37mm tank guns used in other countries, the only ammunition provided for the 40mm gun was solid shot, which was good at perforating the armour of the opposing tanks but not against other targets.

The effectiveness of the 40mm guns and of the earlier 47mm guns was reduced further by the adoption by the Royal Tank Corps of the practice of firing on the move, which kept up the tactical mobility of tanks and at the same time emulated the warships that served as their model. In fact, the influence of warships on tank gunnery extended from the adoption of naval training devices to the performance in at least one of the tank training exercises of the classic naval gunnery maneuver of ‘crossing the T’ (that is sailing in line across the path of the enemy fleet to bear the maximum number of guns on it), although the relevance of the latter to tank warfare was doubtful. Some rightly questioned at the time the ability of tanks to fire accurately when moving over rough ground. However, in spite of this, firing on the move instead of at the halt continued to be favoured, but it was only after the Second World War, when stabilized gun controls were developed, that it became effective.

An entirely different aspect of the development of tanks in Britain was that of the light tanks. It originated with the ideas that emerged after the First World War about the use of very light armoured vehicles to help the infantry advance in the face of opposition. Very similar ideas had already led to the development in France of the Renault FT light tank, but what began to be considered in Britain during the early 1920s were even lighter vehicles. To further the development of this kind of vehicle, Major Martel built in 1925 in his own garage a very light one-man half-track. This was followed by an enlarged two-man version, eight of which were built by Morris Motors for use by the Experimental Mechanized Force in 1927.

Interest aroused by Martel’s vehicle encouraged another private venture, which was the construction of a one-man wheel-and-track vehicle by J. Carden and V. Loyd, who were then running a large garage in London. Their original vehicle was enlarged into a tracked two-man version, and eight were also ordered for the Experimental Mechanized Force.

After the 1927 trials it was decided that what was needed were two different types of light tracked armoured vehicle. One of them was a fast-turreted reconnaissance or scout vehicle for use by the tank battalions of the Royal Tank Corps. The other was an open-top machine gun carrier for use by the infantry. By 1928 Carden responded to these requirements by designing the Carden Loyd Mark VII light tank, a 2.5-tonne two-man vehicle with a turret mounting a machine gun, and the Carden Loyd Mark VI, a small low silhouette two-man machine gun carrier weighing about 1.7 tonnes.

The Mark VI led eventually to the development of the Bren Gun Carrier, which the British Army used on a large scale during the Second World War. During the 1930s several other armies also adopted versions of the Mark VI with head covers or a raised and enclosed superstructure as ultra-light weight low-cost light tanks. However, their capabilities were extremely limited and they could only be justified as training machines.

On the other hand, Carden Loyd Mark VII became the forerunner of a series of Vickers Carden Loyd light tanks, which came to be the most numerous British tanks after the mid-1930s and commercial versions of which were sold by Vickers Armstrongs to several countries. They were mechanically successful, being relatively reliable, and capable of speeds of up to 35mph, and together with the Mark VI they earned their designer, who became Sir John Carden, the high reputation mentioned earlier. But their fighting capabilities were restricted by their armament, which in most cases consisted only of a single rifle calibre machine gun. This might have been adequate for policing the North West frontier of India, where some of the light tanks were employed, but it was ineffective even against other light armoured vehicles.

It was also realized, in contrast to the attitude that prevailed in France, that the one-man turrets of the original Vickers Carden Loyd light tanks expected their occupants to perform too many tasks, particularly in rapidly changing mobile operations. In consequence, Light Tank Mark V, which was introduced in 1934, was provided with a two-man turret that enabled the functions of the tank commander and of the gunner to be separated so that they could operate a tank more effectively. Mark V and the very similar Mark VI were also armed with a Vickers 0.5in. (12.7mm) heavy machine gun in addition to the usual 0.303in. (7.7mm) rifle calibre machine gun. But no significant improvements were made to the chassis, which remained much the same as that of the Mark IV, with the result that the larger turret made the marks V and VI top heavy, looking as if they would topple over at the slightest provocation. Martel rightly argued at the time that the Mark VI was too short, which implied that it did not have sufficient length of track on the ground for a good ride over rough terrain, and that it was overloaded. Nevertheless, a 1936 memorandum by the secretary of state for war claimed that the Mark VI was ‘superior to any light tank produced by other nations’. Moreover, light tanks kept being produced and by the outbreak of the Second World War the number built had risen to 1,002.

In fact, the Mark VI was inferior in several respects to light tanks being produced elsewhere. One of them was the L.60, a 7.5-tonne tank developed by 1934 in Sweden by the Landsverk company with the help of German engineers, which was armed with a 20mm cannon and specimens of which were sold to Austria, Hungary and Ireland before it was developed further for the Swedish Army into the Strv m/38 armed with a 37mm Bofors gun. By 1935 the Czech company Ceskomoravska Kolben Danek also started producing 50 TNH light tanks armed with 37mm guns for Persia (now Iran), which became the forerunners of the TNHP tanks taken over by the German Army in 1939 and used by it successfully as PzKpfw 38(t) during the early stages of the Second World War.

What is more, as early as 1928 Vickers Armstrongs brought out a tank armed with a 47mm gun and a coaxial machine gun, which they designed on their own initiative prior to taking over Carden Loyd Tractors and the light armoured vehicles that the latter were developing. The tank was the 7.4-tonne Type B version of the Vickers Six Ton Tank, which had a single two-man turret in contrast to the Type A version that had two side-by-side one-man machine gun turrets, like some of the early armoured cars. The 47mm gun of the Type B was short barrelled but of the same calibre as the guns of the Vickers Medium tanks, the Sixteen Tonner and all the other British medium tanks up to the original version of the A.9. Because of it Type B was greatly superior in terms of gun power to all the Vickers Carden Loyd light tanks. At the same time its armour protection was similar to that of the contemporary medium tanks and its production cost was considerably lower. Development of this type of tank might therefore have been a better investment for the British Army than all the multi-turreted medium tanks or the light tanks armed only with machine guns, particularly at a time of financial stringency that is often blamed for the shortage of well-armed British tanks on the eve of the Second World War. In fact, the British Army did consider it only to reject the Vickers Six Ton Tank, apparently because of its slow-motion double bogie suspension.

However, rejection of the Vickers Six Ton Tank by the British Army did not discourage eight other armies from buying it. One was also borrowed by the US Army and after being tested at the Aberdeen Proving Ground was virtually copied in 1932 as the T1E4 experimental light tank, which represented a major step forward in the development of US light tanks. This eventually led to the M3 or Stuart light tanks, which the British Army was glad to receive from the United States in 1941.

Two of the armies that procured Vickers Six Ton Tanks went further and produced copies of them in quantity. One was the Polish Army, which bought 38 Six Ton Tanks in 1931 and subsequently developed an improved single turret version armed with a 37mm Bofors gun, 120 of which were produced by the outbreak of the Second World War. The other was the Red Army, which in 1930 signed a contract with Vickers Armstrongs for the delivery of 15 Type A tanks, copies of which began to be produced in the Soviet Union as T-26 tanks a year later. As many as 1,626 were built by 1934 but production was then switched to the single turret model, which was armed with a 45mm gun and was obviously much more effective. By the outbreak of the Second World War in 1939 the total T-26 models that had been built rose to about 8,500, making Vickers Armstrongs’ Six Ton Tank in Soviet guise the most numerous tank at the time in the world.

Tank Development in Europe and America

The development of more mobile and effective methods of using tanks initiated after the First World War by the British Army was not followed by other armies for a number of years. However, while other armies did not advance beyond the use of tanks in support of the infantry, they proceeded to develop them and then to produce them in increasing numbers.

French tanks

The prime example of this was the French Army. By 1926 it decided that it wanted three new types of tanks. One was a light tank of 13 tonnes for close support of the infantry, which would be in effect a successor of the Renault FT. The second was a ‘battle tank’ of about 20 tonnes armed with a 75mm gun, which would co-operate with the lighter tanks in defeating more serious opposition, including enemy tanks. The third type was to be a heavy tank weighing up to 70 tonnes.

The French Army’s requirements were anticipated by the Renault company, which developed the NC light tank that was a somewhat heavier and faster version of the Renault FT. The French Army ordered two in 1923, but it did not adopt them and only some were subsequently sold to Japan and one to Sweden. However, in 1928 one NC was modified to meet the French Army’s requirements for a light tank and a year later was transformed into the prototype of the D1 light tank, ten of which were delivered by Renault in 1931. Subsequent orders resulted in the production of more D1 tanks, which by 1935 totalled 160.

D1 was a tank of 14 tonnes with armour up to 30mm thick and with a turret mounting a short-barrelled 47mm gun as well as a coaxial machine gun. It was also provided with radio for inter-tank communication. In all these respects it represented a considerable advance on the Renault FT. But its turret was still occupied by only one man. This meant that its occupant was expected to load, aim and fire the two weapons mounted in it as well as to command the tank, which was bound to have an adverse effect on its performance on the battlefield. D1 did have one more crewman than the Renault FT, but he sat in the hull and only operated the radio.

Although the armour of D1 was thicker than that of most contemporary medium tanks, in 1930 the Directorate of Infantry demanded the development of a new tank based on it but having even more armour. A prototype of such a tank with armour up to 40mm thick was built by Renault in 1932 and two years later an order was placed for the production of 50 tanks of this type, which was designated D2. However, no further orders were issued, partly because of mechanical problems and partly because of a decision to divert production effort to the manufacture of more powerful tanks armed with 75mm guns. At about the same time D2 as well as D1 were reclassified as medium tanks and the infantry demanded that light tanks weigh less than was specified in 1926, implying a weight of 6 to 8 tonnes.

The infantry’s new requirements were issued in 1933 and led to a competition won by Renault, who built a prototype a year later. This was submitted to the usual development trials, but before they were completed a decision was taken in 1935 to adopt the tank because of the deteriorating political situation brought about by the re-armament of Germany and in particular the remilitarization of the left bank of the Rhine. In consequence an order was placed in 1935 for 300 tanks under the name Char léger modèle 1935 R, usually abbreviated to R 35. Further orders followed and eventually by May 1940 the total number of R 35 produced amounted to about 1,200.

R 35 weighed 10 tonnes but had armour up to 40mm thick, which was thicker than that of most contemporary tanks, and was one of the first tanks to have much of its hull as well as its turret made of castings instead of the less efficient method of riveting armour plates on to a frame. It was not very fast, its maximum road speed being 12.5mph, but it could be argued that this was sufficient for a tank intended for close support of the infantry. What was much more difficult to defend was its one-man turret, which was open to the same criticism as that of the NC tank, and its main armament, which consisted of the same short-barrelled 37mm gun as that mounted 17 years earlier in the Renault FT. The need for a longer barrelled 37mm gun that would be effective against enemy armoured vehicles was only recognized in 1938, when one was adopted, but this happened too late to arm the R 35, although it was mounted in its ultimate development, the R 40.

Until light tanks began to be armed with the longer barrelled 37mm guns, fighting enemy tanks was, so far as the infantry was concerned, primarily the task of the ‘battle tanks’, which came to be represented by the Char B. Work that led to this tank began as early as 1921 under the direction of General Estienne. It resulted in the first instance in the construction by industry of five different prototypes which, nevertheless, had one feature in common, namely a hull-mounted 75mm gun, like the original French tanks. Each also had a turret but armed only with a machine gun.

Experience with the five prototypes led to a new design, which followed theirs in incorporating as the main armament a hull-mounted short-barrelled 75mm gun. Three prototypes based on this design were ordered in 1926 and the first of them was completed three years later. Trials of the prototypes were successful, but in 1930 the minister for war called for an improved battle tank, which resulted in a series of changes including an increase in the maximum thickness of armour from 25 to 40mm and the replacement of the machine gun turret by one with a 47mm gun – the same in fact as that adopted for the D2 tanks. In its modified form the battle tank was finally adopted in 1934 under the designation B1, and an order was issued for its production, although initially of only seven tanks. Further small orders followed and by 1937 these brought the number of B1 tanks to 35 – just enough for one tank battalion.

In the meantime demand arose for further increases in armour protection, which led to studies of new battle tanks, but in the end it was decided to proceed with an improved version of the B1, up-armoured to 60mm and fitted with a new turret with a more powerful 47mm gun as well as a more powerful 300hp engine. The improved tank was designated B1 bis and the first 35 were ordered in 1936, but only 137 were produced by the outbreak of the Second World War in 1939. However, the number built eventually rose to about 340.

Char B1 bis was in some respects an impressive tank. In particular, it was well armoured, which was reflected in its weight of 32 tonnes, and this made it heavier than almost any tank in use in the late 1930s. It was also heavily armed. But the effectiveness of its armament was reduced by the way in which it was mounted. This applied especially to its hull-mounted 75mm gun, which could be elevated but not traversed independently of the hull, so that the whole tank had to be turned to aim it. In consequence the driver also had to act as the gunner, and he was also expected to fire a machine gun fixed in the front of the hull. The other weapons, which consisted of the 47mm gun and coaxial machine gun, were mounted in the turret that was occupied by the tank commander, who was handicapped by having to perform too many functions, as were the occupants of the other one-man turrets of French tanks. In addition to the driver/gunner and commander/gunner, the B1 and B1 bis tanks also carried a loader of the 75mm gun and a radio operator, but they were seated in the hull and only performed their respective functions.

Turning the B1 and B1 bis tanks to traverse their 75mm guns was greatly facilitated by the development of a double differential steering system with a hydrostatic steering drive, which provided very fine control over turning movements. The steering system of the B1 was in fact well in advance of others but it could not compensate for the difficulty of driving a tank and firing its 75mm gun at the same time. Such an arrangement might have worked at low speeds and in simple tactical scenarios and was in keeping with the origins of the type B tanks, which were conceived as tanks that would work closely with the infantry in breaking through enemy defences. But it was not suited to more mobile warfare in which the situation could change rapidly and when there were moving targets, such as enemy tanks.

Mobile operations were not, of course, the domain of the French infantry and of the tanks with which it was entrusted in 1920. Any development of more mobile tanks and their use had to come therefore from the cavalry, although its activities were originally restricted to armoured cars.

French cavalry became involved with armoured cars soon after the outbreak of the First World War, and when the war ended it had 205 of them, newly built on American White truck chassis as well as 67 older Renault and Peugeot armoured cars. The White cars remained the cavalry’s principal armoured cars well into the 1930s, but as early as 1921 development began of new armoured vehicles for it. One outcome of this was a four-wheeled Panhard TOE 165/175 armoured car, more than 50 of which were produced between 1929 and 1932, with one half being sent to Morocco, which was then a French protectorate.

A different response to the cavalry’s requirements was a half-track armoured car designed in 1923 by Citroën in collaboration with the Schneider company and incorporating rubber band tracks developed in Russia by A. Kegresse, a French engineer who had been in charge of the tzar’s garages. Sixteen of the Citroën-Kegresse-Schneider armoured cars were produced by 1925 and they were also sent to Morocco, but the performance of the Kegresse tracks proved disappointing. However, two years earlier a team of five Citroën cars fitted with Kegresse tracks made the first automobile crossing of the Sahara and this boosted the reputation of the Kegresse tracks, which came to be highly regarded because they were quieter than the conventional metallic link tracks and because they were claimed to have a relatively long life of about 2,000 miles on roads. There was therefore good reason for Citroën to persevere with them and to design another half-track armoured car with Kegresse tracks, for which they received an order for 100 in 1925. Schneider became the prime contractors for their production, and together with about 90 of the modernized White-Laffly version of the wartime armoured cars they became the principal armoured vehicles of the French cavalry in the early 1930s.

Twelve additional P.16 Schneider half-track armoured cars were ordered in 1930 and a year later they were followed by an order for 50 Citroën-Kegresse armoured cars of a lighter type. These were the last to have Kegresse band tracks. This was due to the appearance in Britain of the Carden Loyd short pitch metallic tracks. A Carden Loyd carrier with this type of track was tested in France in 1930 and was found to be as fast on roads, and better off the roads, than a Citroën-Kegresse vehicle with band tracks. In consequence Renault designed a small two-man supplies carrier on the lines of the Carden Loyd Mark VI armoured carrier to meet a requirement for such a vehicle by the French infantry, which adopted it in 1931. The Renault chenillette was a peculiarly French vehicle, which was produced in quantity, so that there were already 700 in 1936 and eventually the total built reached 6,000, to the detriment of the production of more effective armoured vehicles.

Having designed the chenillette for the infantry, Renault also used it as the starting point of the development of a new armoured vehicle for the cavalry. This 5.5-tonne vehicle not only had the Carden Loyd short pitch metallic track but also the general configuration of the early Vickers Carden Loyd two-man light tanks, which meant that it had a one man turret mounting a single machine gun and an engine compartment alongside the driver.

By the time this vehicle began to be developed in 1931, the cavalry decided to divide its armoured vehicles into three categories. One consisted of autos-mitrailleuses de découverte, or AMD, whose function was long-range reconnaissance and which were wheeled. The second category consisted of autos-mitrailleuses de reconnaissance, or AMR, whose function was tactical reconnaissance. The third category consisted of autos-mitrailleuses de combat, or AMC, which were expected to be able to fight enemy armoured vehicles. An outcome of this was the adoption by the cavalry of its first fully tracked armoured vehicle, which was called AMR Renault Modèle 1933 or AMR 33, and 123 of which were ordered with deliveries commencing in 1934.

In spite of being designated an auto-mitrailleuse, AMR 33 was in fact a light tank, but it was not designated as such because tanks were supposed to be the preserve of the infantry. It was followed by AMR 35, which was somewhat larger and heavier and was armed in some cases with a 13.2mm heavy machine gun and finally with a high-velocity 25mm cannon instead of a rifle calibre machine gun, which armed most of the 200 vehicles that were produced.

By the time AMR 35 began to be developed in 1933 the cavalry had progressed from simply including an armoured car regiment in its horse cavalry divisions to assembling a fully motorized formation, which a year later became its first division légère mecanique, or DLM. Its organization foreshadowed that of the armoured divisions but its intended role was limited, being similar to that to which horse cavalry had been reduced. However, even then the DLM required vehicles more powerful than AMR 33 and 35. In consequence the cavalry issued a new requirement in 1934 for what was still called an auto-mitrailleuse de combat but which was to have armour up to 40mm thick and be armed with a high-velocity 47mm gun.

This led a subsidiary of the Schneider company, the Societé d’Outillage Mécanique et d’Usinage d’Artillerie or Somua, to design a 19.5-tonne tank designated S 35, which came to be regarded as one of the best of its period. In addition to its armour and armament it incorporated a number of novel features, including a hull made of only three large castings and a double differential steering system with a mechanical steering drive that was ahead of its time. It also had an adequately high road speed of 25mph. However, S 35 suffered from the same major weakness as all the other French tanks built until then, which was its one-man turret. In consequence its occupant had to act, once again, as a gunner and loader as well as the tank commander. S 35 did have a third crewman who sat next to the driver, but he only operated the radio.

The first 50 S 35s were ordered in 1936 and further orders brought the number produced by the outbreak of the Second World War to 261 and eventually to about 416. In addition to the S 35 the cavalry also wanted a light tank for the DLM, and to meet this requirement it adopted the H 35 produced by Hotchkiss, which was originally designed for the infantry but had been rejected by it in favour of the R 35. Having a maximum speed of 23mph H 35 was considerably faster than the R 35, but it was otherwise very similar to it, to the extent of having the same turret with the same old short-barrelled 37mm gun as well as a machine gun. However, an improved H 39 version had a longer barrelled 37mm gun. Like the R 35, the H 35 was first ordered in 1935 and a total of 400 was built; it was then succeeded by the H 39, 680 of which were built by May 1940.

The cavalry acquired one other tank, which was designed by Renault to meet the same requirement as the S 35. This tank, the AMC 35, had the same armament as the S 35 and was as fast, but the maximum thickness of its armour was only 25mm. Nevertheless, it was superior to the S 35 in having a two-man turret so that its commander did not have to double as a gunner and could therefore employ the tank more effectively. Industrial problems that followed the nationalization of the Renault armoured vehicle manufacturing facilities in 1936 delayed its production so that the first was not completed until the end of 1938, and of the 100 eventually built 25 went to the Belgian Army.

The only other French tank to have a turret occupied by more than one crewman was the 68-tonne 2C conceived during the First World War. Six of the ten tanks completed in 1921 were still in service in 1940, although work on improving them ceased in 1932. They were eventually blown up by their crews to prevent them falling into German hands without ever going into action.

US tanks

Like the French Army, the US Army retained heavy tanks designed during the First World War for several years after the conflict ended. In this case the tanks were the Mark VIII, the last of the rhomboidal tanks, 100 of which were completed in 1921. They were not withdrawn from service until 1932. The US Army also resembled the French in continuing to use what were copies of the Renault FT, 952 of which were built between 1918 and 1919.

In keeping with the policy of the US War Department that the role of tanks was to assist the infantry, tanks were distributed between ten separate companies assigned one each to the infantry divisions and to three tank battalions. However, in 1928, after he observed the maneuvers of the British Experimental Mechanized Force, the US Secretary of War, D. F. Davis, directed the US Army to develop a mechanized force. Elements of such a force were duly assembled at Fort Meade, its core elements consisting of a battalion of the Renault FT-type M1917 light tanks and a battalion of Mark VIII heavy tanks. However, the available tanks were too slow to emulate the British experiments and after a few weeks the force assembled at Fort Meade was disbanded.

The only positive outcome of the Fort Meade experiment was that it led two years later to the assembling of another, although small, mechanized force at Fort Eustis. Apart from ten armoured cars, the armoured component of this force consisted of a company of light tanks, 11 of which were still of the M1917 type but four others were the new T1E1 light tanks. The new tanks were three times as fast as the others, which indicated that better tanks were being developed, although they still had to come into service.

The US Army had in fact embarked on the development of new tanks in 1919. Its original objective was to develop a medium tank similar to the British Medium D. The initial outcome of this was the 18.6-tonne Medium Tank M1921, which was completed in 1921. It was not as fast as the Medium D as its maximum speed was only 10mph, but it had a more sensible general configuration, which included a separate driver’s station at the front of the hull and a three-man rotating turret mounting a short barrelled 57mm gun and a coaxial machine gun. It also had a second machine gun in a small turret on top of the main turret, which constituted an original if questionable design feature.

Before a second medium tank was designed, information was received about the use in the British Medium D of a cable-connected single spring suspension system, and it was decided to incorporate it together with Johnson’s peculiar tracks with pivoted wooden shoes in what became the M1922 medium tank. Not unexpectedly, the suspension and tracks proved unsatisfactory on trials that began in 1923, and the M1922 medium tank was abandoned in favour of the third prototype, which was designated Medium Tank T1. This tank, which was completed in 1927, reverted to a more robust multi-bogie suspension and looked like the M1921 tank, but had a more reliable engine as well as a more advanced skeleton type track. In the course of its trials the T1 had its 57mm gun replaced in 1928 by a short-barrelled 75mm gun, and in that form it could have placed the US Army well ahead of others. Instead it was succeeded by a very different front-engined T2 medium tank, which had an ungainly appearance similar to that of the British Vickers Mediums. Like the latter it had a turret-mounted 47mm gun but this was not apparently considered sufficient as it also had a 37mm gun mounted in the front of the hull alongside the driver. T2 was designed down to a weight of 14.2 tonnes specified by the infantry and it was powered by the same 338hp engine as the Mark VIII heavy tank, which resulted in it having a high power-to-weight ratio of 24hp per tonne and a maximum speed of 25mph. Otherwise there was nothing to commend it and its development was abandoned around 1932.

In the meantime the infantry, which had originally agreed to the development of a medium tank, had become more interested in light tanks. This led to the construction in 1927 of the T1 light tank prototype, which was approved by the chief of infantry and was followed by four similar T1E1 tanks. The T1 was a two-man front-engined, 7-tonne tank, which looked like an agricultural tractor with a turret. Its main armament consisted of a short barrelled 37mm gun, which was the same as that of the M1917 light tank, but it had a maximum speed of 17.5mph and was therefore significantly faster than the latter, as already mentioned.

Little further progress took place until 1932 when one of the T1E1 tanks was rebuilt into the T1E4, which followed the lines of the Vickers Six Ton Tanks, one of which had been tested in the United States at the Aberdeen Proving Ground. This involved relocating the engine at the rear of the hull and adopting a Vickers-type bogie suspension and short pitch tracks. T1E4 became the basis of further development of US light tanks, but this was split between the infantry and the cavalry and was challenged by vehicles built by an independent developer of armoured vehicles, J. W. Christie.

The cavalry, which until then had confined its activities to armoured cars, became involved with tanks as the result of a decision taken in 1931 by the US Army Chief of Staff, General D. MacArthur, to disband the mechanized force assembled at Fort Eustis and assign the development of mechanization to the cavalry. By law tanks remained the preserve of the infantry and to enable the cavalry to use them its tanks were called ‘combat cars’ and not tanks, even though they might be the same vehicles as the tanks used by the infantry.

Christie became involved with tracked vehicles as early as 1918, when his company built a ‘motor gun carriage’ for an 8in. howitzer. This was the first of his ‘convertible’ vehicles, which could run on tracks or, if the tracks were removed, on its road wheels and thereby offered the prospect of vehicles that would be able to operate on roads on wheels and would only put on tracks when about to go into action off the roads. In principle this promised to reduce track wear, which was a major contemporary problem.

The encouraging performance of Christie’s first convertible vehicle led to an order for a convertible tank, which he designated M1919, but which was not completed until two years later than this date would suggest. It did not prove a success when tested and consequently Christie modified it at his own expense into the M1921 model. The modifications included springing the front two of its wheels and the removal of the turret, from which its 57mm gun and coaxial machine gun were transferred to the front of the hull – a weapon installation that Christie came to prefer. However, the 1921 version did not prove more successful than the original and was abandoned in 1924.

Undaunted, Christie persevered and in 1928 produced another convertible vehicle, which attracted worldwide attention because of its remarkable performance. It was a turretless vehicle weighing 7.8 tonnes and therefore was lighter than Christie’s previous models, but in spite of this it was powered by a much more powerful 338hp Liberty engine, which provided it with an exceptionally high power-to-weight ratio of 43.3hp per tonne. In consequence it was able to achieve speeds of 70mph on wheels and 42.5mph on tracks, which were much higher than those of any tank built until then. It also had a novel independent suspension with four large road wheels on each side of the vehicle individually sprung by coil springs, which allowed them to rise as much as 280mm from their normal position when going over rough ground and consequently made it possible for the vehicle to move over it at higher speeds.

After protracted negotiations, Christie’s 1928 vehicle was accepted by the US Army’s Ordnance Department and he received an order for seven similar vehicles, the last of which was delivered by his company in 1932. Of the seven, three went to the infantry as T3 medium tanks and were armed with short barrelled 37mm guns while the other four went to the cavalry as T1 combat cars and were only armed with 0.5in. machine guns as their main armament. All seven had their guns mounted in one-man turrets and weighed 9.5 tonnes, which made the infantry classify them as medium tanks.

The seven vehicles derived from Christie’s M1928 represented a major advance in the mobility of tanks but were deficient in other respects, such as having one-man turrets. To overcome some of the deficiencies, the infantry asked for an improved version of the T3 medium tank with a two-man turret and a machine gun in the front of the hull operated by a fourth crewman. Five of the resulting T3E2 tanks were ordered in 1932 but not from Christie, because of further disputes with him. Three years later further development took place with the building at the Rock Island Arsenal of a new T4 medium tank. This retained Christie’s independent suspension and the ability to operate on wheels or tracks, but like the T3E2 it had a shorter pitch track, which generated less vibration and noise than Christie’s original plate tracks. It weighed 12 tonnes and, as it was powered by a 268hp Continental engine, it had a good power-to-weight ratio of 22hp per tonne and a maximum speed of 23.9mph. Like the T3E2 it had a four-man crew and armour up to 16mm thick.

In all these respects the T4 medium tank compared well with other contemporary light/medium tanks and could have been a good basis of further development, particularly if the US Army had dispensed with its ability to run on wheels and concentrated on the best of Christie’s ideas, which was his independent suspension. This is what the British Army was about to do at the time and what the Red Army did a few years later. However, the only development of the T4 was the replacement on a version of it of the two-man turret by a crude, box-like superstructure or ‘barbette’ with machine guns mounted in its four sides, which resulted in the T4 having a total of six of them, one of which was a 0.5in. machine gun. Nothing more powerful was mounted in the basic turreted version of the T4, which was completely out of keeping with its potential. Nevertheless, 16 T4 were built, as were three T4E1s with their primitive ‘barbettes’.

T4 was the last of the medium tanks based on Christie’s ideas built for the US infantry. However, some more vehicles based on them were built for the cavalry. In addition to the T1 combat car, they included the T2 combat car, which the cavalry specified following Christie’s 1928 demonstration but which he would not agree to build. In consequence a prototype was built in 1931 at the Rock Island Arsenal, but its performance proved unsatisfactory. The T4 combat car was very similar to the infantry’s T4 medium tank and like some of the latter had its turret replaced ultimately by a ‘barbette’.

The development of tanks took a new and different turn in 1933, when the US secretary of war directed that future combat cars as well as light tanks should weigh not more than 6.8 tonnes and have a speed of at least 30mph. This led to the construction at the Rock Island Arsenal of the T2 light tank and T5 combat car, which were demonstrated in 1934 at the Aberdeen Proving Ground. The two vehicles were very similar, having the same four-man crew as well as the same armament, consisting of turret-mounted 0.5in. and rifle calibre machine guns and another machine gun in the hull by the side of the driver. They also had the same power train. The principal difference between them was that the T2 light tank had a single two-man turret and, like the T1E2, a suspension and short pitch track similar to those of the Vickers Six Ton Tank, while the T5 combat car had two side-by-side one-man turrets, like the less successful version of the Vickers Six Ton Tank. It also had a new suspension with four road wheels per side sprung in pairs by volute springs, as well as a new type of track with rubber block, double pin links and rubber bushed track pins. The suspension and tracks of the T5 combat car proved superior to those of the other vehicle and became models for most US tanks built from then on until the end of the Second World War.

Demonstration of the T5 combat car was followed by the replacement of its two turrets by another primitive ‘barbette’ superstructure, but eventually sense prevailed and a single two-man turret was used for the final version, which was adopted by the Cavalry and put into production in 1935 as the M1 combat car. By 1937 a total of 89 M1 combat cars was built and they became the basic equipment of the first permanently organized US mechanized formation, the 7th Cavalry Brigade, which was created in 1938 at Fort Knox, Kentucky.

Before this happened the cavalry made one last attempt to develop a Christie-type convertible light tank, which appealed because of its smoother operation off the roads and its high road speeds. An M1 was consequently transformed in 1938 into the T7 combat car with three large pneumatic-tyred road wheels per side. When tested the T7 proved faster than the M1, but the latter, which had a maximum road speed of about 45mph, had sufficient operational mobility to meet the requirements of the cavalry and cost less to produce and to operate. Development of the T7 was therefore abandoned after one year.

As a result of the superior performance of the suspension and tracks of the T5 combat car, the T2 light tank was retrofitted with them, which made the two vehicles even more like each other, and the modified T2E1 was adopted and put into production as the M2A1 light tank. But only nine were built in 1936, when the infantry took the surprising decision to fit its tanks with two side-by-side turrets, which the cavalry had already rejected having tried them on the T5 combat car. Moreover, twin turrets were also being abandoned elsewhere, having been recognized as inferior to single turrets: the Red Army had already stopped producing the twin turret version of the T-26 light tank and in 1937 the Polish Army stopped developing the twin turret version of the Vickers Six Ton Tank in favour of the single turret model.

Except for its two turrets, the resulting M2A2 light tank was very similar to the M2A1 and had approximately the same weight of 8.7 tonnes and the same 16mm maximum thickness of armour. A total of 237 vehicles was produced between 1935 and 1937, and they became the most numerous US tanks before the outbreak of the Second World War. The M2A2 was followed in 1938 by the M2A3, which was similar to it but had thicker 22mm armour. It was consequently somewhat heavier, weighing 9.5 tonnes, but was still capable of a maximum road speed of 36mph. However, its production was limited to 73 vehicles.

Progressive improvements of a stable design, which remained basically unchanged from the T5 combat car to the M2A3 light tank, produced vehicles that were robust and reliable as well as highly mobile. However, in one respect there had been no progress, namely in the armament of the tanks, which still consisted of nothing more powerful than 0.5in. machine guns, when elsewhere comparable tanks were already being armed with high-velocity 37mm guns. It was only at the beginning of 1939 that an M2A3 was modified into an M2A4 and armed with such a gun. But production of the M2A4 only began in May 1940, and consequently none was available when the Second World War broke out.

The inadequate armament of the light tanks was not compensated for by better-armed medium tanks, as their development was not resumed until 1937. This led to the construction a year later of the T5 medium tank prototype and its evolution into the M2 medium tank, 18 of which began to be built in the summer of 1939. Mechanically the M2 medium tank was, in essence, a scaled-up version of the successful M2 light tank, with a lengthened hull and six instead of four road wheels per side. But its superstructure was a curious combination of a ‘barbette’, which continued to have a peculiar attraction for US tank designers, and a two-man turret mounted on top of it. The turret was armed with a high-velocity 37mm gun and a coaxial machine gun, while the ‘barbette’ had a small sponson with a machine gun in each of its four corners, which together with two machine guns fixed in the front of the hull made up a record total of seven machine guns. Fully equipped and with a crew of five the M2 medium tank weighed 17 tonnes, which was on a par with other contemporary medium tanks, but its development was overtaken by the outbreak of the war and its only contribution was to provide a chassis for other tanks.

Italian tanks

As in the United States, a medium tank was also being developed on the eve of the Second World War in Italy. But its background was not as extensive as that of the US M2 medium tank and it was not followed by equally important developments. However, its antecedents were shared to the extent that the first tanks produced in quantity in Italy were copies of the Renault FT, like the M1917 light tanks produced in the United States.

The Italian tank was the Fiat 3000 and 100 of them were completed in 1921. Nothing more happened until 1929 or 1930, when 48 tanks were modified into Model 30 armed with a medium-velocity 37mm gun instead of the twin machine guns of the original Model 21 version. Little change had also taken place in the role of the Italian tanks, which remained confined to close infantry support, as did the contemporary French and US tanks.

The first tentative move towards a more effective use of tanks took place during manoeuvres in Piedmont when Fiat 3000 were tried, with the inevitable result that they were not considered agile enough for use in mountainous terrain, although they had a maximum road speed of 14mph or almost twice that of the Renault FT. Attention then switched to the British Vickers Carden Loyd Mark VI tankette, which was considered to have sufficient agility for operation in Northern Italy, a fact that was of particular concern to the Italian Army. A Mark VI was consequently demonstrated in Italy with the agreement of the British military authorities and in 1929 the Italian Army ordered four Mark VI from Vickers Armstrongs, following this with the acquisition of 21 more vehicles.

The 25 Carden Loyd Mark VI type vehicles were adopted by the Italian Army as Carro Veloce 29, or CV 29, and on their basis Ansaldo and Fiat proceeded to develop an improved model that was adopted by the Italian Army in 1933 as CV 33. This was followed by an order for about 250 and in 1935 by a second order for about 500. As a result CV 33 became in the mid-1930s the principal and in effect the only Italian tank, as the only other was the by then obsolete Fiat 3000.

CV 33 was a turretless vehicle only 1.28m high and weighing 3.4 tonnes; it was manned by a crew of two and was armed with a single or, later, twin machine guns, which had limited traverse. Mechanically it was the most successful of the various derivatives of the Carden Loyd Mark VI, but its capabilities as a fighting vehicle were still extremely limited. It was also a very questionable choice for a vehicle intended for use in mountainous terrain, which generally provides few opportunities for the maneuvers that vehicles with limited traverse weapons have to perform to use them effectively. The harmful consequences of this choice were demonstrated in 1935 during the war in Ethiopia, when a platoon of six CV 33s was ambushed in the Dembeguina Pass and, unable to maneuver, was destroyed by the Ethiopians.

Some of the CV 33 were assigned to the cavalry to support it as carro veloce or fast tanks, although they were not exceptionally fast, having a maximum speed of 26mph. Others were issued to the tank regiments for infantry support and were designated carro d’assalto or assault tanks, although they could only act as such in particularly favourable circumstances where they could exploit their low silhouette and speed because their armour was only 13.5mm thick. In neither guise were they able to fight other tanks, which, curiously, does not appear to have been considered at the time by the Italian Army.

The inability of the CV 33 to fight other tanks was not a problem during the 1935–36 war in Ethiopia, in which 498 CV 33s were ultimately deployed. However, the neglect of this capability resulted in the Italian corps sent to support the Nationalists during the Spanish Civil War of 1936–39 having no tanks to counter Soviet tanks armed with 45mm high-velocity guns that were operated by the Communist forces, although the corps had two battalions of CV 33s.

Nevertheless, the Italian Army ordered a new version of the CV 33, which went into production in 1936 as the CV 35. The two versions were ultimately designated L3/33 and L3/35 and the total number of them produced for the Italian Army reached a total of 1,395.24 About 400 more were produced for export to about 11 different countries as far afield as Brazil and China as well as Hungary.

The first step towards the development of more effective tanks was taken in 1935 when Ansaldo built an 8-tonne turretless tank armed with a hull mounted short-barrelled 37mm gun, which was to replace the obsolete Fiat 3000 still regarded with little justification as a ‘breakthrough’ tank. It constituted a singularly inept design and, deservedly, was not adopted. But its running gear was used in the prototype of the next tank, which began to be tested in 1935 and which, after further development, was adopted in 1939 as the M 11/39 medium tank.

The M 11/39 still had a hull-mounted 37mm gun, but this was longer barrelled and therefore potentially more effective against armoured vehicles. In addition it had a one-man turret with twin machine guns. The limited traverse of its main armament was in keeping with its breakthrough role, which implied leading set piece infantry attacks against enemy positions. However, this was a severe handicap in mobile manoeuvre warfare. In relation to such warfare it was conceptually similar to some extent to the French Char B, although it was far less powerful. In some other respects it was comparable to the US M2 medium tank, and like the latter was beginning to be produced when the Second World War broke out. But by the time Italy entered the war in June 1940 the number of M11/39s that had been completed still amounted to only 70.

Soviet tanks

What progress was made in Italy and the United States and indeed in any other country in the 1930s was dwarfed by the development of tanks in the Soviet Union. At first the only tanks the Red Army had were British and French tanks captured during the Civil War and the 15 copies of the Renault FT. But in 1928 production began of the first Soviet designed tank, the T-18 or MS-1. This was in effect another, somewhat lighter, 5.2-tonne version of the Renault FT with a more supple suspension. Although it was the first indigenous tank to be built it was produced in quantity, with the result that a total of 959 was built by the time its production came to an end in 1931. Moreover, while production was still under way, nine T-18s were sent into action in 1928 against Chinese forces on the Manchurian frontier as the result of a dispute about the Chinese Eastern Railway.

Soon after launching the development of the T-18 light tank in Leningrad, the Red Army recognized the need for a heavier tank and in 1927 set up a facility for the development of one at Kharkov. The first tank to come out of this was the T-12, a 20-tonne vehicle that was essentially a scaled-up T-18 but with a 45mm instead of a 37mm gun and with a small machine gun turret on top of the main turret, like the US M1921 and M1922 medium tanks. A prototype of the T-12 was completed in 1929 but after trials it was abandoned in favour of a major re-design of it designated T-24. This had the same general configuration as the T-12, including a small machine gun turret on top of the main turret, but also had an additional machine gun in the front of the hull operated by a gunner, sitting alongside the driver, who brought the crew of the tank up to five instead of four men. The first T-24 was completed in 1931 and during that year 24 more were built.

The design of the T-24 was basically sound but its performance was not apparently entirely satisfactory, and it was abandoned while the Red Army turned to foreign models. The change was driven by the ambitious plans of the Red Army for a rapid and large-scale acquisition of tanks, and it was decided in 1929 that such plans could be more successfully fulfilled by exploiting foreign expertise. In consequence, I. A. Khalepskii, the head of the Red Army Directorate of Mechanization and Motorization, went in 1930 on a ‘shopping tour’ of Europe and the United States. As a result, the Red Army purchased two tank chassis from J. W. Christie, which were exported from the United States as ‘agricultural tractors’. It also purchased, with the approval of the British government, several vehicles from Vickers Armstrongs, including 15 Vickers Mediums, 15 Vickers Six Ton Tanks, 8 Vickers Carden Loyd amphibious light tanks and 26 Carden Loyd Mark VI tankettes.

By the time they were purchased Vickers Mediums were obsolescent if not obsolete and they were quickly relegated to a training role, although surprisingly some were employed in 1941 in the Finnish-Russian War. But the others became models for armoured vehicles that were mass produced for the Red Army.

The first of these was the T-27 tankette, which was based on the Carden Loyd Mark VI. Production of it began in 1931 and by the time it was terminated in 1934 amounted to a total of 3,328 vehicles. T-27 was a 2.7-tonne turretless two-man vehicle armed with a single machine gun. Its effectiveness as a fighting vehicle was extremely limited, as was that of all the other tankettes based on the Carden Loyd Mark VI, but it was of use as a training vehicle.

The Vickers Six Ton Tank, which was copied as the T-26, was a very much more effective vehicle, although the 15 which the Red Army purchased were of the original type with two side-by-side turrets. However, after 1,627 copies of it were built between 1931 and 1934, production switched to the much more effective single turret version, which was armed with a high-velocity 45mm gun and weighed 9.6 tonnes. This became the standard infantry support tank of the Red Army, and to fulfil the requirements of this role several hundred were produced each year until 1941, by which time the total number built reached 11,218, making T-26 the most numerous tank to be produced until then.

An even more successful purchase proved to be that of the two Christie tank chassis. These became the basis of a series of bystrokhodny or ‘fast’ tanks, which provided the Red Army with tanks that were more mobile than all but a few of their contemporaries. They also became the basis of further successful tank development. The first of the Christie-based tanks, which were designated BT-2, were completed in 1931. They weighed 11.3 tonnes and like their Christie prototypes could run either on their tracks, at up to 32.5mph, or on their wheels after the removal of the tracks at up to 45mph. But they could not run on their wheels on soft ground or over obstacles because their drive was confined to the two rear wheels, which meant that on wheels they were, in automotive parlance, only 8×2 vehicles, and because the road wheels had relatively narrow, solid rubber tyres. On the other hand, when they retained their tracks, their Christie-type independent suspension allowed them to ride over rough ground at higher speeds than other tanks.

Production of BT-2 was followed in 1933 by that of BT-5, which retained the same general configuration, including a two-man turret, but was armed with a 45mm instead of a 37mm gun. The latter was produced under licence from Rheinmetall and its replacement by a high-velocity 45mm gun put BT-5 ahead of other contemporary tanks. Further developments of the BT series included the adoption of welding instead of the riveting of armour plates on the BT-7, which began to be produced in 1934. BT-7 was also fitted with a much shorter pitch track, which reduced the vibration and noise commonly associated with the Christie-type running gear, and 155 BT-7As were armed with short barrelled 76mm guns instead of the 45mm guns. The final version of the series, designated BT-7M and later BT-8, was powered by a newly developed V-2 diesel instead of the gasoline engines of the earlier tanks, which more than doubled its operating range. All the modifications led to the weight of BT-7M rising to 14.65 tonnes, but it was still capable of a road speed of 39mph on tracks. Prototypes of BT-7M were built as early as 1938 but production of it did not start until after the outbreak of the Second World War, and when it came to an end in 1940 the total number of the BTs produced came to 8,122, making them the second most numerous Soviet tanks.

Although tanks that were mass produced in the Soviet Union during the 1930s were based on foreign designs, this did not prevent further indigenous development. One outcome of it was the T-28 medium tank, which in some respects resembled the British A.6 Sixteen Tonner and in other respects was a continuation of the work on the T-24. Like the British tank, the T-28 had three turrets: a large three-man turret and a small one-man machine gun turret on either side of the driver’s station. The large turret of its prototype mounted a 45mm gun, which so far as its calibre was concerned was comparable to the 47mm gun of the Sixteen Tonner. But by the time the prototype was completed in 1932 German medium tanks, which were being tested secretly in Russia, were already armed with 75mm guns, and the Russians followed their example and armed the production version of the T-28 with a 76mm gun. The gun originally mounted in it was short barrelled but in 1938 this was succeeded by a longer barrelled model, which was more effective as an anti-tank weapon. The armour of the T-28 was also improved during the course of its production, which increased its thickness from 30 to 80mm and with it its weight from 28 to 32 tonnes. Production of the T-28 ended in 1940, by which time about 600 are estimated to have been built.

Concurrently with the development of the T-28 medium tank, the nascent Soviet tank industry also developed the T-35 heavy tank. Work on it began in 1929–30 and its design appears to have been inspired by the British A.1 Independent, as it had the same general configuration incorporating five turrets. But while the main turret of the British tank was only armed with a 47mm gun, that of the T-35 had a 76mm gun. Moreover, two of the T-35’s auxiliary turrets initially mounted 37mm and later 45mm guns and not only machine guns, as did its two other turrets and all four auxiliary turrets of the Independent. To operate all these weapons the T-35 required a crew of 11 men.

A prototype of the T-35 was completed in 1931 and production of it began in 1933. It continued to be produced in small batches until 1939, by which time about 60 had been built. During the course of its production the weight of the T-35 rose from 42 tonnes of the prototype to 50 tonnes and eventually to 55 tonnes because of increases in the thickness of its armour and other changes. This made the T-35 the next heaviest tank in service at the outbreak of the Second World War after the obsolete French 2C. But its large size made it difficult to maneuver, so much so that it could not be turned when stationary because of the long length of its tracks in contact with the ground in relation to their spacing.

In addition to those already mentioned, Soviet industry also produced a considerable number of amphibious and other light tanks. They resembled Vickers Carden Loyd two-man light tanks armed with machine guns and were in general equally ineffective.

Once production of all the Soviet tanks began in earnest in 1932 it continued at the rate of more than 3,000 vehicles per annum until 1939, except for 1937 when it fell to one half of what it was in the other years. That year saw the beginning of Stalin’s campaign of terror, which resulted in the execution of three of the Red Army’s five marshals and many of its officers as well as others. Among those who perished were I. A. Khalepskii, who brought in foreign models to accelerate Soviet tank development, Professor V. I. Zaslavsky, who directed the design of the first purely Soviet tank, and A. O. Firsov, the head of the Kharkov tank design bureau, who was blamed for the troubles with the transmission of the BT tanks.

Nevertheless, in spite of the chaos and the loss of experienced personnel caused by the campaign of terror, new tanks began to be developed in 1937 when the Red Army asked the Kharkov plant to design another ‘convertible’ tank to replace the BTs. This led to the A-20, an 18-tonne tank still armed with a 45mm gun in a two-man turret that could, like the BTs, run on wheels or on tracks. However, to make it better able to operate off the roads on its wheels than the earlier tanks with their Christie-type 8×2 wheel drive, the A-20 was provided with a novel shaft drive to six of its eight wheels, so that it could operate as an 8×6 vehicle. But this ultimate attempt to keep alive Christie’s idea of a ‘convertible’ tank was doomed to fail because the road wheels of the A-20 still had relatively narrow solid rubber tyres that severely restricted its ability to move over soft ground. Moreover, the head of the Kharkov design bureau, M. I. Koshkin, came to advocate an alternative design of a tank that operated only on its tracks. The alternative, which was designated T-32, was consequently simpler, easier to produce and could have thicker armour without being significantly heavier. In fact it weighed 19 tonnes while having armour up to 30 instead of 20mm thick, and was armed with a 76mm instead of a 45mm gun. Prototypes of both the A-20 and the T-32 were completed by mid-1939 and by the end of that year an up-armoured version of the T-32 was accepted as the T-34 medium tank. A prototype of it was built a month later and 115 vehicles were completed by the end of 1940.

The T-34 constituted a major step forward in tank development, but its features were not as original as is often portrayed. However, this did not detract from the overall success of its design. In fact, a wise selection of proven features or components was responsible for much of its success, and they were available because of the progressive, evolutionary character of Soviet tank development.

Some of the T-34’s features can be traced, through the BTs, back to the vehicles that Christie built in the United States ten years earlier. This applies in particular to its independent suspension and the tracks driven through their guide horns instead of toothed sprockets. The excellent 500hp V-2 diesel engine that powered it had been used already in the BT-7M, while the adoption of a 76mm gun as the main armament, when medium tanks in other countries were still being armed with 37, 40 or at most 47mm guns, followed the use of guns of this calibre not only in the T-28 medium and T-35 heavy tanks but also in BT-7A and even in some T-26s, all due to the early recognition by the Russians of the importance of gun power.

The one feature of the T-34 that is most often cited and praised as if it were novel and even revolutionary was its sloping armour. In fact, armour had been inclined to increase its effectiveness on a number of earlier vehicles, including experimental models of the BT series and armoured cars, such as the German SdKfz 231 six-wheeled armoured car produced between 1932 and 1937. The armour of the T-34 was undoubtedly effective, but this was due to it being thick as well as sloped. In particular, its frontal hull armour was 45mm thick as well as being inclined at 60° from the vertical, which implied a horizontal shot line thickness of 90mm. On the other hand the A-20, which had very similar sloping armour that was, however, only 20mm thick, had a horizontal shot line thickness of only 36mm, similar to the armour of other contemporary tanks.

As well as enjoying several advantages, T-34 also suffered from a number of shortcomings. The principal one was its turret, which was cramped and made more so by the adoption of a 76mm instead of 45mm gun. The turret had a two-man crew with the commander also acting as the gunner, which prevented him from effectively controlling the tank and who was further handicapped by poor vision from within the tank.

Soon after issuing the requirements that led to the T-34, the Red Army also requested a competitive development of a new heavy tank. It still envisaged a multi-turreted vehicle, and this led to the design of two large tanks of more than 50 tonnes, which had two turrets arranged in a very peculiar way. Thus, one of the turrets, which was armed with a 45mm gun, was mounted above the driver’s station, while the other, which was considerably larger and carried a 76mm gun, was mounted behind the first and towered above it, raising the overall height of each tank to more than 3m. Prototypes of the two tanks, which were designated T-100 and SMK, were built in 1939, but so, at the instigation of its designer, was a single turret version of the SMK, which was by far the most sensible of the three vehicles and which was adopted as the KV-1 heavy tank.

Production of KV-1 began in 1940, like that of the T-34, and it was armed with a similar 76mm gun as well as being powered by a higher rated version of the same V-2 diesel engine as the latter. At 43 tonnes it was significantly heavier than the 26-tonne T-34 and its frontal armour was thicker but, as it was only slightly inclined, it was not more effective. In fact, its only advantage over the T-34 was its larger and more roomy three-man turret. Otherwise T-34 made KV-1 unnecessary and it might have become a ‘universal’ tank, as its successors did several years later. However, at the time the Red Army was still wedded to the concept of the separate categories of medium and heavy tanks, as well as to that of the third category of light tanks for close infantry support or reconnaissance.

German tanks

A similar policy of fielding three different categories of tanks was pursued at the time in Britain and France, but it was not adopted by the German Army when it began to develop tanks again in the mid-1920s. In spite of being forbidden by the Versailles Treaty to have tanks, the German Army awarded contracts in 1927 to three companies – Daimler Benz, Krupp and Rheinmetall – for each to design and build in secret two tanks in the 16-tonne class armed with 75mm guns. To disguise their development they were called Grosstraktoren or ‘large tractors’. A year later a contract was also awarded to Krupp and Rheinmetall for each to develop two light tanks of about 8 to 9 tonnes armed with a 37mm gun, which were called Leichttraktoren or ‘light tractors’.

Although the tanks could be designed and built in secret in Germany, they could not be tested there without attracting the attention of the British or French authorities. However, a way round this was found as the result of a rapprochement between Germany and the Soviet Union, which led to the setting up of a joint German-Soviet test and training centre in Russia, at Kama near Kazan. All six Grosstraktoren were shipped there in 1929 and trials of them went on until 1932.

Grosstraktoren were well armed with short-barrelled 75mm guns mounted in two-man turrets, in addition to which they had a one-man machine gun turret at the rear of the hull, like the French 2C heavy tanks. They followed the example of the original British heavy tanks in locating the commander of the tank in the hull, alongside the driver, which severely restricted his field of view and which was not repeated in the design of any other tank. They were relatively fast, having a maximum road speed of 25, and as they were relatively large in relation to their weight of 16 tonnes they had sufficient buoyancy to float and could propel themselves in water at up to 2.5mph by means of two screws. But their ability to float indicated that they were not heavily armoured, the maximum thickness of their armour being only 14mm, like that of other contemporary tanks.

Leichttraktoren were also sent to Kama for their trials, which began in 1930. They looked somewhat old fashioned with their front-engined configuration resembling that of the LK-II light tank, which was about to be produced in Germany when the First World War ended. Their best feature was a two-man turret mounting a high-velocity 37mm gun, which was very similar to that made in Sweden for the L 30 wheel-and-track tank built later by the Landsverk company. They were also in the forefront of the use of radios for inter-tank communication having, like the Grosstraktoren, a radio operator sitting next to the driver.

All ten tanks sent to Kama were shipped back to Germany in 1933. In the meantime discussions with the Russians, who had already embarked on the development of the T-28 medium tank, led to a decision to follow the Grosstraktoren with a Neubaufahrzeug or NbFz, a new vehicle of 20 tonnes still armed with a low-velocity 75mm gun but mounted together with a high-velocity 37mm gun. It also had two auxiliary machine gun turrets, one in front and one behind the main three-man turret. The latter represented a major improvement on the Grosstraktoren, as it had the commander located in it and was provided with an observation cupola. Contracts for the development of the NbFz were awarded in 1933 to Krupp and Rheinmetall, and two mild steel and three armoured prototypes were completed by 1936. But after trials they were relegated to a training role. However, in 1940 the three armoured NbFz were landed in Oslo during the German invasion of Norway. Their appearance on that occasion led to rumours about a new category of German heavy tanks of 36 tonnes, armed with 75 or 105mm guns and called PzKpfw V and VI, which were perpetuated in British and US Army handbooks about enemy forces issued in 1940 and 1941, as well as other publications. In fact, development of Nb Fz never proceeded beyond the five prototypes.

Development of the Leichttraktoren was also followed by a new requirement, in this case for a light vehicle with a rear mounted engine. In 1931 Krupp were awarded a contract for such a vehicle, which was to weigh only 3.5 tonnes and which was called Kleintraktor, or ‘small tractor’. To gain time its design was based on that of the British Carden Loyd light tanks and consequently three chassis were purchased from Vickers Armstrongs in 1932.49 The Kleintraktor chassis designed by Krupp was accepted and in 1935 Krupp received a contract for 135, while each one of five other firms received a contract for three more chassis to provide them with experience of producing tanks. The chassis were given the cover name Landwirtschaftlicher Schlepper, or LaS, meaning ‘agricultural tractor’. However, they were subsequently fitted with armoured superstructures and turrets and towards the end of 1934 the Henschel company built the first three complete tanks. They became PzKpfw I after Hitler repudiated the terms of the Versailles Treaty in 1935, and the original order for 135 vehicles was quickly followed by another for 450 vehicles. By June 1937 when production ended the total number built amounted to 1,493 tanks.

Originally the Kleintraktor was to be armed with a 20mm cannon, but in 1932 General O. Lutz, the Inspector of Motorized Troops, decided that the tank which was to become PzKpfw I should be armed with machine guns only. This may have facilitated quantity production and speeded up the equipping of troops with tanks, but it did not make PzKpfw I very effective. General H. Guderian, who was Lutz’s chief of staff, defended the decision many years later, writing in his memoirs that PzKpfw I was only a ‘training tank’ adopted pending the development of more effective models. But there is no evidence that it was regarded as such at the time.

By 1934 General Lutz appears to have recognized the shortcomings of PzKpfw I and, according to Guderian, decided to order a second stopgap – a light tank armed with a 20mm automatic cannon. As a result the Ordnance Office of the German Army issued orders for the production of such a tank, which became the 7.9-tonne PzKpfw II, the first of which was completed in 1936. By then other tanks of its kind were already being armed with 37mm or even, in the case of the Soviet T-26, 45mm guns. The 20mm cannon of PzKpfw II enabled it at least to perforate the armour of enemy light tanks or even that of some medium tanks but not of others, particularly as armour began to grow on the eve of the Second World War. Nevertheless, PzKpfw IIs were produced in quantity so that by the outbreak of the war in 1939 there were 1,223 of them, or almost as many as there were PzKpfwIs.

Tanks that Lutz and Guderian really wanted were the PzKpfw III and IV. Like the NbFz that preceded it, PzKpfw IV was armed with a low-velocity 75mm gun mounted in a three-man turret. But, very sensibly, it dispensed with the auxiliary machine gun turrets as well as the 37mm gun mounted alongside the 75mm gun. Instead, it had a machine gunner-cum-radio operator alongside the driver. This meant that its general configuration was the same as that originally devised in 1929 for the British experimental A.7 medium tank and which was still the best choice, as shown by its widespread and successful adoption during the Second World War. PzKpfw IV became the most powerful tank to be issued to the Panzer Truppen before the war, although its armour was originally only 15mm thick. The contract for its development was issued to Krupp in 1935 and production of the first batch of 35 began in 1937. This was followed by an order for 176 more tanks, so that by the outbreak of the war the German Army had 211 PzKpfw IVs.

The PzKpfw III began to be produced at about the same time as the PzKpfw IV but at an even lower rate. It had the same general configuration but its main armament was a high-velocity 37mm gun instead of a low-velocity 75mm weapon. It was designed to be a vehicle in the 15-tonne class but once it began to be produced its weight approached that of PzKpfw IV, making the two tanks very similar except for their main armament. Their concurrent production was therefore very questionable, particularly from the point of view of logistics. Nevertheless, Guderian considered a tank of the PzKpfw III kind to be the right vehicle for the three light companies of a tank battalion, while tanks of the PzKpfw IV type would equip its medium tank company. The assignment of PzKpfw IV to the medium tank companies of tank battalions showed once again that they were not in the same category as the British close support tanks, as a number of writers have claimed. In fact, they were much more versatile and effective, being able to engage enemy tanks as well as other targets, which the close support tanks were never able to do.

Only ten PzKpfw III were initially produced in 1937 and by the outbreak of the war in 1939 there were still only 98 in the German Army inventory. However, by then the German Army had acquired a number of other tanks armed with 37mm high-velocity guns as a result of the annexation of Czechoslovakia in 1939. The number of Czech tanks acquired by the German Army has often been exaggerated, but the total of 469 armoured vehicles of various types that the Czech Army had included 298 LT vz 35 tanks armed with 37mm high-velocity guns. These 10.5-tonne tanks were taken over by the German Army as PzKpfw 35(t) and 202 of them were in its inventory when the war began. In addition, production of new Czech TNHPS tanks, which were also armed with 37mm guns, continued after the annexation and about 100 were completed by the outbreak of the war as PzKpfw 38(t).

Polish and Swedish tanks

Apart from Japan, the only other countries to develop tanks before the Second World War were Poland and Sweden. Poland produced 120 7TP tanks based on the Vickers Six Ton Tank, and it also produced 440 TK3 and TKS turretless tankettes developed from the basis of the Carden Loyd Mark VI. In 1936 the Polish Army also started working on a convertible tank with a Christie-type running gear, which led to the 10TP, a 12.8-tonne tank with the same two man Bofors turret and 37mm gun as the 7TP. However, its development had not advanced beyond the construction of a single prototype when an end was put to it by the German invasion of Poland.

Development of tanks in Sweden benefited from the links between Swedish and German industries, and in particular from the takeover by the German Gute Hoffnungs Hutte steel company of the Swedish Landsverk company. As a result of this, in 1928 Landsverk began working on the development of armoured vehicles, which could not be pursued at the time in Germany because of the restrictions imposed on it by the Versailles Treaty. The first outcome of this was an unusual vehicle that could run on wheels or tracks, but not by converting from one mode of operation to the other using the same running gear like Christie’s convertible tanks. Instead it had separate sets of tracks and wheels that could be used alternately. This vehicle was the idea of J. Vollmer, the designer of the German tanks of the First World War and supervisor of the construction of the first Swedish tank, the Strv 21, which was based on the German LK II light tank prototype. The wheels-and-tracks tank was designated Landsverk 5, or L 5, weighed 7 tonnes and was armed with a 37mm gun. Gute Hoffnungs Hutte sent it in 1930 to the German-Soviet test centre at Kama, while Vollmer tried to have it adopted by the Red Army. However, after three years of negotiations the Russians rejected it in 1930.

This did not put an end to the development of the wheels-and-tracks tank, as a modernized version of it was ordered by the Swedish Army in 1931. Landsverk designated it L 30 and the Army Strv fm/31. It was delivered in 1935, but although it proved capable of 47mph on wheels as well as 22mph on tracks the Swedish Army did not proceed with it beyond trials. However, in 1931 it ordered three L 10s or Strv m/31s, which was a much more sensible, tracks-only version of the L 30 and which was delivered in 1934 and 1935. They weighed 11.5 tonnes and were armed with a 37mm Bofors gun mounted in a two-man turret, which, together with their general configuration, put them on a par with the best of the contemporary tanks of their kind.

In the meantime Landsverk developed the L 60, which had a configuration similar to that of the L 10 but which was more advanced from the automotive point of view, having an independent suspension with torsion bar springs. Specimen L 60 armed with 20mm cannon were sold to Austria, Ireland and Hungary, and in the case of the latter 200 were later produced under licence. In Sweden the L 60 was developed into the Strv m/38, an 8.7-tonne tank with a three-man crew armed with a 37mm Bofors gun that constituted an effective, modern design. Sixteen were ordered by the Swedish Army in 1937, but further orders, for the very similar Strv m/39 and m/40, were not issued until after the outbreak of the war.

Creation of the Armoured Forces

When tanks were slow and few in number, as they were in general during the 1920s, their effectiveness was confined to leading small scale infantry attacks in face of enemy machine gun fire and suppressing it by their own fire. But as they grew faster and potentially more numerous, prospects emerged of creating a new type of armed forces based on tanks that was more mobile and more effective than the existing armies consisting of infantry and cavalry.

Some, and in particular Fuller and Liddell Hart in Britain and Estienne in France, foresaw tank-based armies in the early 1920s. What is more, in the following years tentative steps were taken in Britain and then in a few other countries towards the creation of mobile forces based on tanks. But in most armies the view was held that tanks were only an auxiliary to the infantry and that their role was to support it closely, and this view remained firmly entrenched in military doctrine.

The principal exponent of this doctrine was the French Army. Its tanks were a part of the infantry and were organized into independent battalions. When equipped with light tanks these were to be allocated in the event of offensive operations to infantry divisions on the scale of one battalion per division and were designated chars d’accompagnement. More powerful tanks, which were called chars de manoeuvre d’ensemble, were to lead more concentrated attacks along the main line of thrust of an infantry division or corps. However, until 1931, when the first D1 tanks were produced, the French Army had nothing other than the Renault FT light tanks, which were only suitable for close infantry support. It had no means therefore of implementing or even exploring other uses of tanks.

It was therefore only in 1932 that exercises were carried out to examine some of the other possibilities. But they did not lead to any significant progress so far as the employment of tanks by the infantry was concerned. On the other hand, they were followed by a major advance in the more mobile use of tanks by the cavalry.

Five years earlier French cavalry began to motorize some of its horse mounted units by converting them into dragons portés – riflemen carried in unarmoured Citroën-Kegresse half-tracks. This led to each horse cavalry division having a battalion of dragons portés as well as four battalion-size horse cavalry regiments. Even earlier, in 1923, it was proposed to provide the cavalry with a light tank, and in 1931 a requirement was actually issued for one, which became the AMR 33. Then in 1934 the cavalry formed a division légère mécanique, or DLM, which became one of the first formations of its kind to be placed on a permanent footing.

When fully established, the DLM consisted of a brigade of tanks, a three-battalion regiment of dragons portés and a regiment of artillery as well as other, supporting units. Eventually, by 1939, it was equipped with 220 tanks, which included 80 S 35 (Somua) medium tanks and 80 H 39 (Hotchkiss) light tanks as well as 60 AMR 33 or 35 light tanks that were part of the dragons portés battalions. However, to start with it only had the AMR 33, with which it began to be equipped in 1934.

Its organization shows that the DLM had the makings of an armoured division, but it was not regarded as such. Instead, it was considered to be a successor of the horse cavalry divisions and therefore inherited the latter’s limited roles of reconnaissance, screening operations and delaying actions carried out for the benefit of infantry formations – the roles to which horse cavalry was reduced by the beginning of the 20th century. The evolution of the DLM from the basis of horse cavalry carried with it therefore a historical impediment, which it would have to overcome if it were to become an armoured division capable of independent offensive action.

As it was, the conversion of cavalry into tank units was strongly opposed by many cavalrymen who continued to believe in the effectiveness of horse cavalry and who resented having to give up their horses. The creation of the DLM was actually opposed by the inspector of cavalry as well as the inspector of infantry and others. In the circumstances little further progress could take place.

An exception to the prevailing attitude was provided by Colonel Charles de Gaulle, who in a book entitled Vers l’Armée de Metier published in 1934 advocated the creation of a professional army of 100,000 men. The core of this army was to be six mechanized or armoured divisions, each of which was composed of a brigade of tanks, a brigade of infantry and two regiments of artillery. Such an organization was similar in principle to that already adopted for the DLM and the German panzer divisions, and de Gaulle’s book had no impact on the development of armoured forces in France nor, contrary to numerous claims, elsewhere. However, de Gaulle’s ideas prompted Paul Reynaud, who was later to become prime minister, to advocate in 1935 in the Chamber of Deputies the creation of an armoured force capable of offensive action. But the response to it by the Minister of War, General Maurin, was that the creation of such a force would be ‘useless and undesirable’.

A year later the French Commander-in-Chief, General Gamelin, was also against the creation of armoured divisions. The most he would recommend was the formation of a second DLM, which was put into effect in 1937, while a third DLM was only created in 1940, after the outbreak of the Second World War. However, at a meeting of the War Council in October 1936 Gamelin admitted that the French Army lacked the necessary offensive capability possessed by the German armoured divisions, which led him to recommend that the problem be studied. The sequel to it was that the Army Council agreed to the creation of two armoured divisions in December 1938. But their constituent elements were not assembled until the outbreak of the war and the two divisions did not come into being until January 1940.

Each of the armoured divisions or divisions cuirassées (usually abbreviated to DCRs) contained two tank battalions with a total of 68 B1 tanks, two battalions with 90 H 39 light tanks, one battalion of motorized infantry and a regiment of artillery, as well as supporting units. They had therefore some of the characteristics of combined arms armoured formations. But the French Army doctrine did not envisage their use as an independent mobile striking force. Instead they had a more limited offensive role, which was to be carried out under the control of infantry corps or other large formations within the prevailing concepts of continuous fronts and methodical operations. In fact, the DCRs were primarily concentrations of the infantry’s heavy tanks, which were not very suitable for mobile operations, and their low proportion of infantry to the 158 tanks limited their ability to operate independently.

The DCRs did not therefore provide the French Army with armoured divisions capable of offensive mobile operations but with limited purpose formations, like the DLMs, while the majority of its tanks remained dispersed in separate battalions assigned to close infantry support.

The situation arrived at by the French Army on the eve of the Second World War was duplicated in some respects by the British Army. In particular, like the French, the British Army divided its tanks between the support of the infantry and the carrying out of the role previously performed by the cavalry. But the French Army reached this situation by simply perpetuating the traditional division of armies into infantry and cavalry, whereas the British Army regressed to it after trying to develop a more effective use of tanks.

The turning point came in 1934 when the Tank Brigade, which had carried out the pioneer experiments in the more mobile use of tanks, was put on a permanent basis. Later that year a decision was also taken to divide tanks between army tank battalions, which were to be allocated to infantry divisions to support them like the French chars d’accompagnment, and a mobile division, which was considered to be a successor to the cavalry division, like the French DLM.

However, the composition of the Mobile Division was to be very different from that of the DLM. One of its major components was the Tank Brigade, whose commander Brigadier P. C. Hobart would have preferred it to be composed mainly of tanks. But others, in particular Brigadier G. M. Lindsay, who set off the sequence of events that led to the creation of the Tank Brigade, came to advocate its combination with a motorized infantry brigade. Such a combination was actually tried during the 1934 maneuvers, but the experiment proved to be a fiasco and there was to be no infantry in the Mobile Division. On the other hand it was to contain a cavalry brigade, which constituted its other major component. The cavalry brigade was to consist of a regiment of light tanks and three regiments mounted in trucks, which, like the dragons portés, would have provided the equivalent of the motorized infantry that the Mobile Division needed. But this was not a role congenial to the cavalry, which preferred to be mounted in light tanks if, very reluctantly, it had to give up its horses. A small-scale experiment at converting cavalrymen into motorized riflemen was initiated in 1935 but by 1937, when the Mobile Division was finally established, this approach was abandoned and a decision was taken that its cavalry regiments be equipped with light tanks.

As it came into being, the Mobile Division retained the Tank Brigade but with three instead of its original four tank battalions. On the other hand its cavalry component was increased to two brigades, each with three regiments, which were, in effect, light tank battalions. This meant that the Mobile Division had a total of nine tank battalions with a nominal strength of more than 500 tanks, which resulted in an unbalanced organization with too many tanks in relation to the other components of the Division and in particular its two motorized infantry battalions. Moreover, most of its tanks were light tanks armed only with machine guns.

Some of the shortcomings of the Mobile Division were rectified in 1939, when its name was changed to that of an armoured division and the number of its brigades was reduced to two. One of them was called the Light Armoured Brigade and consisted of three regiments or battalions equipped with a mixture of cruiser and light tanks, while the other was called the Heavy Armoured Brigade and had three regiments equipped with cruiser tanks. Between them the two brigades were to have 108 light and 213 cruiser tanks. This was a more practical number of tanks than that originally proposed, but the divisional troops were also reduced to one motorized infantry battalion and a small artillery regiment. In consequence, although the number of tank battalions was reduced from 9 to 6, the proportion of infantry to tanks became even smaller, almost as if the ‘all-tank’ ideas were being subscribed to again. Nevertheless, the General Staff planned in 1939 to have three such divisions as well as five brigades of army tank battalions.

Before much of the plan could be implemented, the war broke out, and the British Army entered it with its armoured forces divided between two armoured divisions, which were still being organized, and about three brigades of army tank battalions. The division of British tanks between the two categories of units persisted even after the striking success in France in 1940 of the German armoured forces, which eschewed such a division. Their success led the British Army to plan the formation of seven more armoured divisions which came into being between September 1940 and August 1941. But their basic organization still followed that of the Mobile Division and their intended role bore traces of the latter’s origin as a successor of the horse cavalry division. In fact, as late as May 1944, an Army training pamphlet still declared that armoured divisions are only ‘designed for exploitation after the enemy’s position has been broken’.

However, in practice, in Libya in 1941 and 1942 and in Normandy in 1944, the armoured divisions were not confined to exploitation but acted as versatile fighting formations. Their organization changed also in 1942 from the two armoured brigade pattern inherited from the Mobile Division to a single armoured brigade with three tank and one infantry battalions backed by a three-battalion infantry brigade. The change reduced the tanks per division from the 386 they had in 1941 to about half that number. But at the same time it increased the number and proportion of the motorized infantry in belated recognition of its importance as a complement to tanks, which even Fuller, who has been regarded as the principal exponent of the ‘all-tank’ concepts, had recognized seven years earlier.

Towards the end of the war the number of British armoured divisions declined to five, although each now had 294 medium and light tanks. The number of (army) tank brigades also declined but to a lesser extent, so that there were still eight of them compared with the maximum of 11 in 1942. Thus right up to the end of the Second World War the British Army continued to divide its armoured forces into two separate categories according to the traditional division of armies into infantry and cavalry, which other armies ceased to do.

Like the British and French armies, the Red Army also came to divide its tanks between supporting the infantry and more mobile operations. But for several years after the First World War the only tanks it had were British and French tanks captured during the Civil War and the few native copies of the Renault FT. Little attention was paid to them and in 1922 some were even sent to the Ukraine to help with agricultural activities. Nevertheless, some members of the Red Army, such as K. Kalinovskii and V. Triandafillov, began to consider the future of tanks and their deliberations led to the PU-29 Field Service Manual of 1929, which became for several years the main statement of Soviet tank doctrine. It stated that the main role of tanks was paving the way for the infantry, but added that they should be divided in action between PP or close infantry support tanks, and DD or ‘long range’ tanks that would thrust into an enemy’s positions, attacking his artillery and disorganizing his rear echelons.

The concept of DD tanks became a major feature of Soviet tank doctrine and was expanded into the idea of deep thrusts by mechanized forces, which was then elaborated into the concepts of deep battle and deep operations. The latter has been described as the ‘brainchild’ of Marshal M. Tukhachevski but, apart from the contribution to it by others, its origins can be traced back to centuries of offensive raids carried out on the plains of Eastern Europe by forces of cavalry. The most recent examples of them were the raid carried out in 1919 during the Russian Civil War by the White Cavalry led by General K. K. Mamontov, which inspired the subsequent operations of the Red Cavalry, and the break through the Polish front in 1920 during the Polish-Russian War by the Red Cavalry Army led by S. Budennyi. There was even a motorized forerunner of the deep thrusts in the raid carried out in 1920 by a Polish force consisting of armoured cars, truck-borne infantry and artillery, which penetrated more that 40 miles in a day into enemy held territory to seize the important railway junction of Kowel in the rear of Tukhachevski’s troops retreating from the gates of Warsaw.

However, for all its historical antecedents, the concept of deep thrusts by mechanized forces and the employment of tanks in the DD role did not prove easy to implement. To test some aspects of it, a decision was taken in 1929 to form a mechanized regiment, which was quickly expanded into a brigade consisting of a two-battalion tank regiment with a total of 60 tanks, an infantry battalion and an artillery battery. The brigade was put to test during the 1930 maneuvers but its performance proved disappointing, partly because the MS-1, or T-18, tanks that were used were too slow, particularly in the role of DD tanks.

More mobile tanks, such as the BTs, did not begin to be produced until a year later, but they then became available in increasing numbers as a result of the large-scale production programme launched in 1927 as part of Stalin’s first Five Year Plan. Under this plan the Red Army was urged to acquire quantities of the latest equipment and it responded by asking for about 1,500 tanks. But the Politburo tripled their number. The demand for such a large number of tanks was partly driven by grossly exaggerated estimates of the numbers of tanks possessed by other armies, in particular by the Polish Army, which was regarded as the principal enemy.

As large numbers of tanks began to come out of the factories, it became possible for the Red Army to create an increasing number of armoured units. Thus in 1932 it decided to organize two mechanized corps, each of which was composed of two mechanized and one motorized infantry brigades. The mechanized brigades consisted of three battalions of tanks, one of machine guns and an artillery battery, the tanks being T-26s in the case of one of the two brigades and BTs in the other. In total each corps had 490 tanks. In addition to the mechanized corps, each infantry division was to receive a battalion of 57 tanks for close support and each cavalry division was to have a regiment of 64 tanks. There were also independent tank brigades at the disposal of the High Command.

In 1934 two more mechanized corps were created, and the corps became the Red Army’s mobile strike force on which would rest the execution of any deep operations. One of the corps put on an impressive appearance at the 1935 Kiev maneuvers, in which more than one thousand tanks were said to have taken part. They demonstrated that the Red Army was well ahead of others so far as the size of the armoured forces was concerned – in fact, it already had more tanks than all the other armies put together. They also demonstrated that the Red Army was ahead of others in creating mechanized or armoured formations.

But how the mechanized forces were to be employed was a matter of continued debate. In their existing form the corps were considered to be unwieldy, and in response to criticism the number of tanks in each was reduced. Their T-26 tanks were also, very sensibly, replaced by BT tanks, which were far more suited to them. However, the employment of tanks in the DD role and in particular their co-operation with other arms had still not been fully developed. What is more, because of the difficulty of implementing it, there was growing dissatisfaction with the whole concept of deep battle of which the mechanized corps were an integral part.

Before any of the issues could be resolved, Tukhachevski and many others involved in the development of Soviet armoured forces were executed in 1937 in Stalin’s reign of terror. Inevitably, this had an adverse effect on the concepts of the mobile offensive employment of mechanized forces with which those executed were associated.

At about the same time Soviet tanks were used in combat, first in the Spanish Civil War and then in the clashes with Japanese forces on the Manchurian and Mongolian borders. Their deployment in Spain began with the delivery of 50 T-26s with Russian crews only three months after the outbreak of the Civil War in July 1936. Other deliveries followed and by the end of the year the Soviet General D. Pavlov, who assumed command of the armoured units of the Republican Army, was able to organize an armoured brigade with four tank battalions having a total of 230 T-26 tanks. Ten months later this was followed by the formation of a division with two brigades of T-26 tanks and a regiment of 50 newly delivered BT-5s.

The total number of Soviet tanks delivered to Spain amounted to 331. This represented a large tank force by contemporary standards, and the Soviet tanks outnumbered as well as outgunned the 106 German PzKpfw I tanks sent to aid the Nationalist forces of General Franco. However, the Republican Army failed to capitalize on its tank strength. On one occasion, in preparation for the Battle of Brunete in July 1937, it assembled 130 tanks, but they were used in a dispersed manner and suffered heavy losses, as they did on other occasions when they were used in small packets poorly co-ordinated with other arms.

A year later units of the Red Army clashed with Japanese forces at Lake Khasan, 70 miles from Vladivostock. The Japanese were eventually forced to withdraw, but because of its poorly co-ordinated frontal attacks the Red Army suffered relatively heavy losses, including that of 85 T-26 tanks out of the 257 that were employed. In August 1939, just before the Second World War broke out in Europe, another clash occurred, this time on the border of Outer Mongolia at what is described as Khalkhin-gol in Russian accounts and as the Nomonham Incident by the Japanese. This time the Red Army deployed a force commanded by General G. K. Zhukov, the future Deputy Supreme Commander of the Soviet Army, which included six mechanized and tank brigades with a total of more than 500 tanks, primarily BT-5 and BT-7. Zhukov employed them as a mobile striking force on the lines envisaged previously for the mechanized corps to carry out a double envelopment by which they defeated the Japanese forces.

But the success of this operation did not restore the faith of the Red Army leaders in the concept of deep battle and in an independent use of mechanized forces. A major factor in this was the view derived from the Spanish Civil War that tanks had become very vulnerable to anti-tank weapons – a view which was widely held at the time. It was argued therefore that tanks needed to be supported by infantry and artillery and could only achieve success if closely tied to them and not in more independent, mobile operations. Pavlov, who returned from Spain to become the head of the Armoured Force after two of his predecessors were executed, even went so far as to claim that the whole concept of DD tanks was flawed.

All this led the Red Army to begin a reorganization of its armoured forces in 1938. The name of the four mechanized corps was changed to tank corps and they were deprived of their motorized infantry brigades and supporting troops. Similarly, the tank brigades were deprived of their organic machine gun battalions. As a result the tank corps became almost ‘all-tank’ formations and were no longer capable of carrying out independent mobile operations. But these changes did not satisfy men like Pavlov who considered the existence of the tank corps to be pointless. In consequence the Main Military Council of the Red Army decided in November 1939, two months after the German invasion of Poland, to disband the tank corps. At the same time it directed that tank brigades should co-operate more closely with infantry and cavalry divisions. This meant that, so far as the concepts of the employment of tanks were concerned, the Red Army regressed to where it was ten years earlier when the use of tanks was subordinated to the infantry.

Before the reorganization took place the Red Army had four tank corps, 24 independent tank brigades, and 11 tank regiments as well as battalions of T-26 tanks attached to infantry divisions and regiments of BT tanks attached to cavalry divisions. The total number of tanks that all these units and formations had was still greater than that of all the other armies. But the Red Army was no longer in the forefront of the development of mechanized or armoured forces.

The lead in this field was seized in the mid-1930s by the German Army, in spite of the handicap that for more than a decade it was forbidden to have any tanks by the terms of the Versailles Treaty. However, the latter could not prevent studies of the future of tanks. These were carried out against the background of Germany’s strategic situation, which included the possibility of war on two fronts and the relatively small size of its army restricted by the Versailles Treaty. This led the head of the German Army, General H. von Seeckt, to adopt the policy that a small army had to be highly mobile and to favour offensive warfare. At first von Seeckt assigned an important part in mobile operations to the cavalry, but he and other German officers took an increasing interest in motorized troops and in tanks, slowly recognizing the possible use of tanks in mobile warfare. This was inspired in part by the appearance in 1924 in Britain of Vickers Medium tanks, which represented a considerable advance on earlier tanks in terms of speed, and subsequently by the maneuvers in 1927 of the Experimental Mechanized Force on Salisbury Plain.

An indication of the views that were emerging is provided by a rare but widely consulted book on tanks written in 1925 by an Austrian officer, Captain F. Heigl, who considered that the faster tanks that were appearing were eminently suited to mobile warfare and should not be restricted to operating at the pace of the infantry. In his second book published a year later Heigl went on to argue that, instead of supporting the infantry in the customary frontal attacks, tanks would be better employed in mobile operations against the flanks and the rear of an enemy. He concluded that in the future armies would consist of tanks accompanied by infantry in light armoured vehicles and by artillery equipped with tracked self-propelled guns.

Similar views were gradually adopted by others. They included Captain (later Major) H. Guderian, who became interested in tanks in the early 1920s and who came to play a leading role in the evolution of their tactics and organization. In the course of his activities Guderian became a prominent exponent of the mobile use of tanks and an ardent advocate of their concentration in large, independent armoured formations.

According to his memoirs, in 1929 Guderian reached the view that to be fully effective tanks had to be supported by other arms but brought up to their level of mobility. This view may have been influenced to some extent by the importance of inter-arm co-operation, which had been instilled into the German Army by von Seeckt and was in striking contrast to the ‘all-tank’ views favoured in Britain at the time. It led to the balanced, all-arms composition of the panzer divisions, which made them superior in this respect to other armoured formations.

The combined arms approach was adopted after the 1932 German Army maneuvers by General O. Lutz, who became Inspector of Motorized Troops and made Guderian his chief of staff. The maneuvers were still carried out with dummy tanks but Lutz proceeded in the following year to argue that tanks should be concentrated in large, independent mechanized formations that would become the core of an offensive strike force. Steps towards the achievement of this objective were taken in 1934 when, as part of the German Army’s expansion that followed Hitler’s rise to power, approval was given for the creation of three panzer divisions. This was followed by the assembly for the maneuvers in mid-1935 of an improvised panzer division and the calling into being of three panzer divisions in October of that year, with Guderian being given command of the 2nd Panzer Division.

The organization of the panzer divisions was worked out in 1934 and was based on a tank brigade with two two-battalion tank regiments backed by a motorized infantry brigade with a two-battalion truck-borne regiment and a motorcycle battalion, an armoured reconnaissance battalion, an anti-tank battalion, an artillery regiment and an engineer company, as well as supporting units.43 Each of the division’s 16 tank companies was to have 32 light tanks, which, together with command tanks, would have given it a total of 561 tanks.

Such a large number of tanks would have made the division unwieldy, but it was a common characteristic of the early ideas on the composition of armoured formations, as shown by the British Mobile Division and the early Soviet mechanized corps. As many as 500 tanks also featured in a detailed proposal for a tank division made in a book by another Austrian officer, General L. von Eimannsberger, which was published a year before the first panzer division came into existence and provided another indication of contemporary ideas. Like the panzer division, Eimannsberger’s tank division combined tanks with a motorized infantry brigade and a regiment of artillery that, with remarkable foresight, he expected to act in an anti-aircraft as well as ground roles, as the German 88mm anti-aircraft guns later did so effectively from the Spanish Civil War onwards.

When the panzer divisions were actually being formed and began to receive more effective equipment than the PzKpfw I light tanks, the number of their tanks was brought down to a more manageable level by a reduction in the number of companies per battalion to three and of tanks per company to 22 or 19. As a result by 1939 each division was to have 324 or 328 tanks. These were to include the more powerful PzKpfw III and PzKpfw IV, but although the allocation of the two tanks was relatively modest there were still not enough of the former to meet the requirements of the panzer divisions, particularly as their number grew to six, two more being created in 1938 and one in 1939.

Nevertheless, in spite of the shortage of medium tanks, the six panzer divisions constituted a highly effective mobile force. Its effectiveness was based to a large extent on the balanced composition of the panzer divisions and the close inter-arm co-operation between their units, as well as methods of command and control that were superior to those of other contemporary mechanized formations. The last involved the first large scale use of radios for inter-tank communication: even the light two man PzKpfw I tanks were fitted with radio receivers, and other tanks, starting with PzKpfw II, had transmitters as well as receivers. The extensive use of radios is sometimes ascribed to Guderian, who was a signals officer in the First World War, but provision for radios was already incorporated in the Leichte as well as Grosstraktoren, which were designed before Guderian became involved with tank development.

The panzer divisions made the most of the majority of tanks the German Army had by concentrating them so that they could be used massed, as Guderian repeatedly advocated. The only exception to this were the four light divisions created by the German cavalry, which, as in other armies, originally opposed mechanization and then proceeded to develop motorized formations of its own. Each of the resulting light divisions contained one or two tank battalions and what were in effect three to four motorized infantry battalions, as well as the usual divisional troops. Guderian opposed their creation instead of additional panzer divisions, but after the 1939 campaign in Poland, in which they were found to lack offensive power, they were converted into panzer divisions. However, as there was a shortage of German medium tanks, the 1st Light Division, which became the 6th Panzer Division, was partly equipped with ex-Czech Army PzKpfw 35(t) and the 2nd and 3rd light divisions, which became the 7th and 8th panzer divisions respectively, were partly equipped with newly built Czech PzKpfw 38(t) tanks, all of which were armed with 37mm guns similar to those of the German PzKpfw III.

A far greater shortage of medium tanks afflicted the armoured forces that were being created in Italy. In fact, none was available for them until shortly before Italy entered the war in June 1940.

The first step in the development of Italian armoured forces was taken in 1936 when a Brigata Motomeccanizzata was formed in Sienna. It consisted of a battalion of tanks, a two-battalion regiment of bersaglieri or light infantry, and an artillery battery. A year later it was renamed Brigata Corazzata, or armoured brigade, and became the first fully mechanized formation of the Italian Army. A second brigade was formed at about the same time in Milan and in 1939 the two brigades were transformed into the Centauro and Ariete armoured divisions. One more division was created later in the same year by the conversion of the Littorio infantry division.

The organization of the three armoured divisions followed that originally adopted in 1937 for the first armoured brigade. This meant that each had a tank regiment, a bersaglieri regiment with one battalion of motorcyclists and one of truck-borne riflemen, an artillery regiment and an engineer company. The tank regiment had a nominal strength of four battalions but in practice it had only two.

In theory the three divisions had a well-balanced combined arms organization. However, until 1940 the most powerful tanks available for them were the 50 obsolete Fiat 3000B tanks armed with medium-velocity 37m guns. Otherwise the only armoured vehicles with which they could be equipped were the turretless machine gun armed CV 33 or CV 35 tankettes. In consequence, the three divisions were incapable of carrying out effectively offensive mobile operations or the Guerra di rapido corso that was proclaimed by Mussolini and that became official doctrine in 1938.

The effectiveness of the Italian armoured divisions would not have been increased to any great extent if some of their tankettes had been replaced by M 11/39 medium tanks when they began to be produced in 1939. It was only in mid-1940 when a derivative of the M 11/39 with a turret-mounted 47mm gun began to be produced that the Italian armoured units were provided with a tank whose characteristics approached international standards.

In the meantime some Italian tank units had been engaged in combat, first in Ethiopia and then in Spain, where two battalions were sent to support the Nationalist forces in the Spanish Civil War. The intervention in Spain involved the sending of 149 CV 33 or CV 35 tankettes, 56 of which were eventually lost. On several occasions they led the advances of the Italian expeditionary corps, the Commando Truppe Volontarie or CTV, but they were helpless when they encountered Soviet T-26 tanks, which outgunned them completely. One particularly noteworthy action in which they took part was the failed offensive of the CTV at Guadalajara in March 1937, which was widely claimed at the time to have demonstrated the ineffectiveness of mechanized forces. In fact, the offensive was not carried out by mechanized forces but by four divisions of infantry supported by only four companies of tankettes.

Nevertheless, the erroneous perception of what happened in Spain became the basis of arguments against the development of mechanized forces. This happened, among others, in the United States, where the development of mechanized forces was largely in the hands of the cavalry, whose more conservative officers regarded it, like their counterparts in other armies, as primarily a threat to horse cavalry. The development began with the short-lived Experimental Mechanized Force, which was inspired by the experimental British force of 1927 and which was assembled at Fort Meade in the following year. This was followed by another experimental force assembled at Fort Eustis in 1930 but which was disbanded a year later, when the decision was taken to mechanize one horse cavalry regiment. But this regiment only began to train with tanks in 1933 when it was transferred from Texas to Fort Knox in Kentucky, where it was equipped with ‘combat cars’ or light tanks. In the meantime, in 1932, a mechanized cavalry brigade was called into being, but it did not become effective until 1938 after a decision was taken to mechanize a second cavalry regiment.

Much of the mechanization of the cavalry was due to the efforts of Colonel (later General) A. R. Chaffee, who became the commander of the mechanized brigade and who persuaded the chief of cavalry to recommend to the War Department in 1937 the formation of a mechanized cavalry division. However, until the outbreak of the Second World War the 7th Cavalry Brigade (Mechanized), to give it its full title, with its two cavalry regiments and a nominal total of 112 combat cars remained the only mechanized formation of the US Army.

Further progress only took place in 1940 when a motorized infantry regiment was added to the cavalry brigade, as Chaffee had long advocated, to make the brigade more operationally independent and effective. Then during Army maneuvers carried out in Louisiana in the same year, the cavalry brigade was combined with a provisional tank brigade made up of the infantry’s tanks to form an improvised armoured division. Shortly afterwards, spurred by the success just scored by the German armoured divisions in France, the US War Department decided, in July 1940, to create an armoured force with two armoured divisions based on the 7th Cavalry Brigade and the infantry’s Provisional Tank Brigade.

At that stage the US Army had a total 464 tanks, including 18 obsolescent T4 medium tanks.54 But now each of the armoured division was to have an armoured brigade with two regiments of light tanks and one regiment of medium tanks, containing a total of 273 light and 108 medium tanks. In addition, the armoured brigade included an artillery regiment and was supported by a two-battalion armoured infantry regiment, an additional artillery battalion and an engineer battalion, as well as a reconnaissance battalion. The organization of the armoured divisions was clearly based on that of the 7th Cavalry Brigade, but on maneuvers in 1941 it was found to be unwieldy. In consequence, in March 1942, shortly after the United States entered the war and before the armoured divisions went into action, their organization was changed. This involved the elimination of the brigade echelon and the introduction under the division commander of two combat command headquarters that could control any combination of the division’s units. It also involved a reduction in the number of armoured regiments to two, but each of them now had two medium and one light tank battalions, which provided the division with almost the same total number of tanks as before but reversed the ratio of light to medium tanks, there being 143 of the former and 232 of the latter. At the same time the infantry regiment acquired a third battalion, which increased the proportion of infantry to tanks, and the artillery was reorganized into three battalions.

In addition to improving their organization the armoured divisions grew in number, three more being activated before the United States entered the war. As defined in a contemporary armoured force manual, their role was ‘the conduct of highly mobile ground warfare, primarily offensive in character, by self-sustaining units of great power and mobility’. There was no question of them being confined to exploitation or other limited roles, although the manual recognized that they were ‘especially suited’ for it.

Panzers and their Second World War Opponents

The outbreak of the Second World War on September 1, 1939 brought into immediate action a large number of tanks, as all six regular and one provisional panzer divisions as well as four light divisions took part in the German invasion of Poland. Between them these 11 formations had 2,682 tanks1 out of the 2,980, excluding command tanks, which the German Army had at the time.

While almost all the available tanks were concentrated in the armoured formations, the panzer divisions were distributed among corps consisting primarily of infantry divisions. Nevertheless, they spearheaded the rapid thrusts that resulted in the envelopment and destruction of the strategically exposed and inadequately armed Polish forces in less than four weeks. The speed with which the campaign was conducted led to it being called blitzkrieg, or lightning war. This name has since come into widespread use to describe a particular kind of warfare, although it was not a German military term but merely a catchword which the Western press picked up and started using even before the fighting died down.

The cost of the campaign to the German armoured forces was the loss of 231 tanks. Most of them were PzKpfw Is and IIs, but even PzKpfw IIIs proved vulnerable to Polish 7.92mm anti-tank rifles as well as Polish-made 37mm Bofors anti-tank guns. Not surprisingly, PzKpfw IV was singled out by Guderian as a highly effective weapon that should be produced in quantity. On the other hand, commanders of PzKpfw II complained about the inadequate vision provided by the single although rotatable Zeiss periscope with which their model and many other tanks were fitted at the time. As a result PzKpfw IIs were provided with a ring of eight fixed periscopes around the commander’s hatch, which set a new standard in all-round vision from within tanks.

German tanks encountered little opposition from Polish tanks as there were few of them and the tanks that were available were not very effectively employed. The largest units were three battalions of tanks, two of which had 49 7TP light tanks each. They were used separately and fought split up into companies without adequate logistics support, as a result of which a number of tanks was destroyed by their crews when they ran out of fuel and ammunition. The third battalion was equipped with 49 R 35 tanks, which was all the Polish Army was able to procure from France before the war. This battalion was held in reserve and was eventually ordered to cross the frontier into Romania without ever firing a shot. Ironically, the final employment of Polish tanks came after the campaign, when the Germans refurbished 21 captured 7TP tanks and equipped Hitler’s escort battalion with them.

An entirely different by-product of the campaign in Poland has been the myth of Polish cavalry charges against German tanks. It arose out of a charge on the first day of the war by two squadrons of a Polish cavalry regiment, which was misrepresented in some German accounts as having been carried out against tanks. The charge was actually against infantry, but the myth of it being carried out against tanks has persisted into the 21st century.

While few tanks opposed the German tanks in Poland, even fewer opposed the Red Army when it invaded Finland in November 1939. In fact, the Finnish Army only had 26 Vickers Six Ton Tanks and not all of them had yet been armed with 37mm Bofors guns. On the other hand the invading Soviet forces had about 1,500 tanks. However, their frontal assault on the Karelian Isthmus failed, as did the offensive operations on other parts of the front, and they suffered heavy losses in tanks. But after the failure of their original offensive, Russian forces reorganized and launched another assault on the Finnish defences. This time they employed about 1,330 tanks, which attacked in close co-operation with the infantry and overwhelmed Finnish defences, leading to an armistice in March 1940.

Soviet tanks were mostly T-26s, which proved vulnerable to Finnish 37mm Bofors anti-tank guns, as they did three years earlier in Spain to German 37mm anti-tank guns, because of their relatively thin armour and inept employment. This was equally true of the BTs, almost all models of which were used against the Finns. The Soviet forces also used T-28 medium tanks, 97 of which were destroyed, and also T-35 heavy tanks with five turrets, several of which were also destroyed.

During the first offensive the Red Army tested two of its new KV-1 heavy tanks as well as its unsuccessful, multi-turreted T-100 and SMK competitors, all of which proved immune to Finnish 37mm anti-tank guns, and during the second offensive it also deployed the recently developed 52-tonne KV-2 armed with a 152mm howitzer. The new T-34 medium tank was also to be tested on the Finnish front, but did not arrive until after the armistice. For its part the Finnish Army captured a total of about 600 armoured vehicles, and the recovered T-26s became its principal tanks.

The one-sided employment of large numbers of tanks that characterized the German invasion of Poland and the Soviet assault on Finland came to an end on May 10, 1940 when the German Army launched its offensive against the Netherlands, Belgium and France. Since the campaign in Poland, German armoured forces had been reorganized, as a result of which the four light divisions were converted into panzer divisions so that there were now ten of them. Moreover, panzer divisions were concentrated into panzer corps and two of the panzer corps were combined to form a panzer group.

However, the tank inventory had only risen slightly to 3,379 tanks and the number actually deployed by the ten panzer divisions was 2,574, which was fewer than the number used against Poland. Of this total, 523 were still the light PzKpfw I armed only with machine guns, which had already proved deficient in Poland as well as Spain, and the number of PzKpfw IV had only increased by 69, in spite of Guderian’s recommendation mentioned earlier. The only significant improvement was an increase in the number of PzKpfw IIIs from 98 to 329.

The French Army, which bore the brunt of the German offensive, had approximately the same number of tanks, namely about 3,650.16 But whereas German tanks were concentrated in the panzer divisions, one third of the French tanks, which consisted primarily of R 35 light tanks, were distributed between 25 independent battalions spread out over the French front stretching from the Swiss border to the English Channel. The heavy B1 and B1 bis tanks, of which there were about as many as there were German PzKpfw IVs, were allocated to the three divisions cuirassées, or DCRs, but the first two of them only began to be organized when the war broke out eight months earlier and the third was created less than two months before the German offensive. In consequence their organization was incomplete and their units had little opportunity to train together, let alone practice mobile operations. In addition to the more modern tanks, there were also seven battalions of obsolete Renault FT tanks and one battalion of six 68-tonne 2C heavy tanks that should have been relegated by 1940 to a museum.

The only fully organized and trained mechanized formations of the French Army were the three divisions légères mécaniques, or DLMs. Two of them formed a cavalry corps commanded by General R. Prioux, which provided the traditional cavalry screen for the French forces moving into central Belgium, where the main German thrust was expected to come. In the course of performing its mission, Prioux’s corps met two advancing panzer divisions and engaged them in what was the first tank versus tank battle of the Second World War. The battle took place east of Gembloux, after which it is generally called, and involved around 400 French and 600 German tanks. The former included about 160 S 35s (Somua) medium tanks, whose frontal armour was not only thicker than that of the German tanks but virtually impervious to their guns, while the 47mm guns of the S 35 were superior in terms of armour penetration to the 37mm and 75mm of the German PzKpfw III and IV, although not to the extent that is sometimes claimed. But, like other French tanks, the S 35s were severely handicapped by having one-man turrets, which overtaxed their crews. This was aggravated by the poor vision from within the S 35 as well as other French tanks, which restricted the situation awareness of their crews and together with the lack of radios in other French tanks inhibited co-ordinated action. All this contributed to the operation of French tanks in small, isolated groups, which was noted by German tank crews and helped them to outmaneuver French tanks.

Nevertheless, the cavalry corps accomplished its mission, albeit at the cost of 105 tanks, and fell back, but its tanks were then distributed along a defence line established by infantry divisions, despite complaints by General Prioux. In the meantime the French High Command was taken completely by surprise by the advance of the panzer group of five divisions through the Ardennes Forest, which was considered to be a major obstacle to mechanized forces. The panzer group, which included a corps commanded by Guderian, crossed the River Meuse and broke through the French front at Sedan, while two other panzer divisions, one of which was commanded by General E. Rommel, crossed the Meuse north of it. After the breakthrough, the panzer divisions advanced rapidly towards the Channel and cut off French and British forces in Belgium from the rest.

Farther north the remaining panzer division invaded the Netherlands, and after four days of fighting the Dutch Army, which had no tanks, capitulated.

On the French front, the scattered battalions of R 35 tanks could offer little resistance to the onslaught of the panzer divisions. What is more, not only were the R 35 battalions used piecemeal, but their tanks, like most other French light tanks, were armed with short-barrelled low-velocity 37mm guns dating from 1918, which their opponents described as ‘worthless’. The three DCR were held in reserve behind the front line in the region of Chalons, and in response to the German offensive the 1st was sent to Charleroi in Belgium, where it became involved in some heavy fighting with Rommel’s 7th Panzer Division. During the fighting some of its B1 tanks were surprised while refuelling, while others were abandoned after they ran out of fuel, and the division was destroyed in the same piecemeal fashion as it was employed. The 2nd DCR was wasted by being spread out in small units or even single tanks to guard the crossings of the Oise River. The 3rd DCR was sent to attack the southern flank of Guderian’s corps, but instead was dispersed into defensive positions and was then committed piecemeal to the defence of Stonne.

One more DCR, the 4th, was assembled hastily during the course of the campaign, and under the command of Colonel (later General) de Gaulle attacked the advancing panzer divisions from the south, at Montcornet and Laon, and then attacked the German bridgeheads over the Somme near Abbeville, but the attacks only achieved local tactical successes. The bridgeheads had also been attacked, two days earlier, by two brigades of the British 1st Armoured Division, which had just landed in France to protect the right flank of the British Expeditionary Force. The two brigades were sent into action without infantry or artillery support and were repulsed, losing many of their tanks. The only other tanks the BEF had, apart from the Mark VI light tanks of the seven divisional cavalry regiments, were those of the 1st Army Tank Brigade, which consisted of two battalions with a total of 58 Mark I and 16 Mark II (Matilda) infantry tanks. Backed by two battalions of infantry and accompanied by the 3rd DLM, the Tank Brigade struck Rommel’s panzer division near Arras, inflicting considerable casualties, and was only brought to a halt by the fire of the divisional artillery and 88mm anti-aircraft guns, as German 37mm anti-tank guns proved ineffective against the thick armour of the Matildas.

The attack at Arras was the largest tank action carried out by the BEF before most of it was evacuated from Dunkirk, together with the troops of the French First and Seventh Armies, after abandoning its remaining tanks. While the evacuation was taking place, the panzer divisions were regrouped, and after the fall of Dunkirk they attacked again, breaking through the defence line set up along the Somme and Aisne rivers by General M. Weygand, who replaced General G. Gamelin as commander-in-chief of the French Army. In an attempt to restore some of its armoured forces, the French Army re-created the three DLM with personnel evacuated via England from Dunkirk and even created two new DLM, the 4th and 7th. After the German breakthrough, tanks of the 7th DLM, as well as remnants of the 3rd DCR, put up a stiff fight in the region of Juniville, but like the other DLM, it had by then only about 20 tanks and could do little to prevent the defeat of the French Army.

After the Armistice was signed on 22 June 1940, some of the cavalry regiments were re-formed in the unoccupied part of France. However, by agreement between the German authorities and the Vichy government, their equipment was confined to a total of 64 Panhard 178 armoured cars, with a reserve of 28, and they had their 25mm guns removed, leaving them armed only with machine guns.

The German armoured forces suffered a total loss of 770 tanks during the campaign in France, excluding command vehicles, most of which – 611 – were destroyed during the first month of the campaign. The armour of PzKpfw III and IV was found to be inadequate and inferior to that of the French tanks, but German tanks increased their survivability by mobile tactics, which reduced their chances of being hit by French guns. The guns of the German tanks, and in particular the 37mm L/45 of PzKpfw III, which was regarded as the principal anti-tank weapon, proved to be inadequate and ineffective against the frontal armour of S 35 and, even more against the B1bis tanks. In the circumstances, the most effective weapon against French tanks was the 75mm L/24 gun of PzKpfw IV, in spite of being short barrelled and having a low muzzle velocity.

The campaign in France led inevitably to demands for a larger calibre gun for PzKpfw III, which were in fact anticipated by the development of a 50mm L/42 gun. According to his memoirs, Guderian wanted such a gun as early as 1932, but at the time the chief of ordnance and the inspector of artillery considered that a 37mm gun would be adequate and would ensure commonality with the infantry’s contemporary 37mm anti-tank guns.

In consequence the first PzKpfw III with a 50mm L/42 gun was not produced until July 1940. By then another and considerably more effective 50mm gun, the L/60 with a longer barrel and a higher muzzle velocity, had been developed to replace the 37mm anti-tank gun. Hitler saw it and ordered that PzKpfw III be armed with it. But in April 1941 he found that his order had not been implemented and insisted that it be put into effect immediately, which according to Guderian it could have been and would have placed PzKpfw III ahead of most contemporary tanks. As it was, the first tank with the 50mm L/60 was not produced until December 1941, and tanks with the shorter barrelled 50mm L/42 continued to be produced until 1942. In addition, no decision was taken until November 1941 to produce a more powerful 75mm gun than the L/24 mounted in PzKpfw IV, although the armour-piercing performance of the latter had fallen behind not only that of the 50mm L/60 but also of the 50mm L/42. Yet three different experimental vehicles with a more powerful, long-barrelled 75mm gun were built by Rheinmetall to an order issued by the Ordnance Office in 1934.

Soon after the campaign in France, Hitler ordered that the number of panzer divisions be doubled. As a result ten new panzer divisions were created by the beginning of 1941, but the increase in the number of the divisions was achieved at the expense of the number of tanks in each. Thus, when the reorganization was complete, no panzer division had a tank brigade with two regiments any longer, but had a single regiment with two or three battalions. Each battalion had a medium tank company, generally with 20 but in some cases with 30 or even 36 PzKpfw IV, and two or three light tank companies equipped mainly with PzKpfw IIIs or PzKpfw 38(t)s. In consequence, the strength of the divisions varied from 145 to 265 tanks.

The reorganization was barely completed when six of the panzer divisions spearheaded the invasion of Yugoslavia and Greece in April 1941. Once again they were largely responsible for a rapid victory, which resulted in the surrender of the Yugoslav Army after only 11 days of fighting and of the Greek Army six days later. The cost reported by five of the six divisions amounted to a loss of 56 tanks.

In the meantime the Italian army in Libya threatened to invade Egypt, and the British forces stationed there launched an offensive against it. This involved a battalion of 45 Matilda infantry tanks, which led the assault by an infantry division on a series of fortified camps established by the Italian army after its initial advance. At about the same time the British 7th Armoured Division, which was equipped with a mixture of A.9, A.10 and A.13 cruiser tanks as well as Mark VI light tanks, attacked other objectives. In total the British forces had 275 tanks. The Matildas proved impervious to Italian anti-tank weapons and completely outclassed Italian M 11/39 tanks, 23 of which were knocked out in one of the camps. In the final stages of the offensive, in February 1941, cruiser tanks of the 7th Armoured Division attacked retreating Italian units, which included new M 13/40 tanks. Unlike the M 11/39, these had turret-mounted 47mm guns that were about as good as the 40mm guns of the British cruisers, and their armour was thicker. But they went into action in small packets and by the end of the day 112 had been knocked out or had been abandoned by their crews.

This completed the destruction of the Italian army in Cyrenaica, which prompted Hitler to dispatch to Libya the 5th Light and 15th panzer divisions under the command of General Rommel to bolster the Italian army in Tripolitania. The latter had already been reinforced by the arrival of the Ariete armoured division, which was followed by the German 5th Light Division. The disembarkation of this division in Tripoli was completed a month later, when its tank strength reached 151 tanks, including 61 PzKpfw III and 17 PzKpfw IVs. Then, without waiting for the 15th Panzer Division, Rommel decided to take the offensive and in two weeks drove British forces back to the Egyptian border, destroying their 2nd Armoured Division.

A counter-offensive code-named Battleaxe was mounted by the British forces in June 1941 after a convoy brought reinforcements from Britain, which included 135 Matildas and 82 cruiser tanks. Among them for the first time were Crusaders, which were more heavily armoured than the earlier cruisers. The German forces had also been reinforced by the arrival of the 15th Panzer Division, and the counter-offensive was repulsed with the loss of 92 British tanks compared with 12 German tanks. The principal cause of the failure of the British counter-offensive was the institutionalized division of tanks between infantry support and more mobile roles, which led to a dispersal of tank units and their tendency to act by themselves that contrasted with the very effective co-operation of German tanks with anti-tank guns, which included 88mm anti-aircraft guns used in a ground role.

In November 1941 what had become the British Eighth Army launched another offensive under the code name Crusader, for which 756 gun-armed tanks had been assembled while 259 more were held in reserve and 231 were in two armoured divisions undergoing training. The tanks that were assembled included 336 cruisers, which by then were mostly Crusaders but still included a number of A.13s and even 26 A.10s, and they also included 225 infantry tanks consisting not only of Matildas but also of Valentines.

The Valentine was the last of the British tanks designed before the war, and it differed from the others in not being designed to a War Office specification but in having been originated by Vickers Armstrongs. It was based on the proven chassis of what was originally the A.10 infantry tank but had frontal armour 60 to 65mm thick, which made it second in this respect only to the Matilda and as good as the French B1 tank. Not to overload the chassis, its weight was kept down to 16 tonnes and as a result of this it was provided with a turret for only two men, instead of a three-man turret like other British and German tanks. The War Office objected to the turret but, nevertheless, production of the Valentine was ordered and the first was completed in May 1940, when Britain badly needed tanks. Eventually, 8,275 Valentines were built, including 1,420 built in Canada, and their production exceeded that of any other British tank of the Second World War. Apart from being numerous, Valentine was also more reliable than other contemporary British tanks, which was attributable to it being developed by the only British company with several years’ experience in the design and production of tanks.

Like other contemporary British tanks, Valentine was armed with the 40mm 2-pounder, and some time after it was deployed in Libya comments appeared that this gun was inferior to the guns of the German tanks, and, by implication, that it was responsible for the reverses suffered by British tank units. In fact, its armour penetration was slightly greater than that of the 50mm L/42 gun of PzKpfw III as well as being greater than that of the 75mm L/24 of PzKpfw IV.34 However, the armour-piercing projectiles of German tank guns from the 37mm gun onwards contained an explosive charge with a delay fuse, which made them more deadly when they penetrated armour than the solid shot fired by the 2-pounder and which has been ignored in almost all the accounts of the fighting in North Africa.

Tanks assembled for Operation Crusader also included 195 US M3 light tanks. They were the first of the large number of American tanks supplied during the Second World War to the British Army, in which they were called Stuarts after the Confederate cavalry leader of the American Civil War. The design of the M3 or Stuart was somewhat dated and it had a cramped two-man turret, but the armour penetration of its 37mm gun was slightly greater than that of the German 50mm L/42. It was also fast and proved very reliable, thanks to the extensive development work that preceded its production. At the same time its armour protection as well as armament were comparable to those of the British cruisers. It was therefore regarded as a ‘light cruiser’ and the whole of one British armoured brigade was equipped with it.

When Operation Crusader began, the two panzer divisions of what became the Afrika Korps had a total of only 145 PzKpfw III and 38 PzKpfw IVs, while the Italian armoured units had 146 M 13/40s. Nevertheless, they managed to repulse the initial attacks of the British tank units, which were committed in a dispersed and disjointed way that nullified their overall numerical superiority. By comparison, German formations acted in a more coherent fashion and, as before, very effectively combined the action of their tanks with that of the anti-tank guns. However, in the end Rommel’s much depleted forces were forced to retreat to the border of Tripolitania. But only two weeks later and having received reinforcements, which brought up its strength to 77 PzKpfw III and ten PzKpfw IVs, Rommel’s Afrika Korps went on the offensive and drove British forces back to the Gazala line in Cyrenaica. During the four months’ lull that followed, both sides built up their tank strength. On the German side the number of tanks rose to 242 PzKpfw IIIs, including for the first time in Libya 19 Model Js armed with the long-barrelled 50mm L/60 gun, and 38 PzKpfw IVs, while the number of Italian tanks amounted to 230 M 13/40s. On the British side the number of tanks rose to a total of 850, backed by a reserve of about 120 and 300 more held in Egypt. Tanks available to the British forces included for the first time 167 American-built Grants armed with a medium-velocity 75mm gun, which was superior in terms of armour penetration to German tank guns except for the 50mm L/60 to which it was equal, and it fired high explosive as well as armour-piercing projectiles, which enabled British tank units to counter for the first time the threat of anti-tank guns.

The tactical effectiveness of the Grant’s 75mm gun was somewhat reduced by it not being mounted in a turret but in the hull and therefore having limited traverse. Grants did have a turret, but this only mounted its secondary armament of a 37mm gun. Because its 75mm gun was mounted in the hull, some Francophile historians have suggested that its design was inspired by the French B1. In fact, it had nothing to do with the latter, having originated in 1939 with an experimental T5E2 version of the contemporary US medium tank, which had a 75mm howitzer mounted in its hull instead of a turret with a 37mm gun. Subsequently T5E2 became the only available basis on which medium tanks with a 75mm gun could be quickly produced when the use of PzKpfw IV armed with a 75mm gun in the 1940 German campaign in France showed that the US Army urgently needed a tank with a similar armament. In consequence, a new M3 medium tank was developed from the T5E2 and was ordered in 1940 not only for the US Army but also, in a modified form, for the British Army, which called its version General Grant while the US version was called General Lee, after the Confederate commander. Prototypes of the M3 medium tank were completed in May 1941 and deliveries from production commenced only two months later. Eventually the total number of the US M3 medium tanks and of the British version that were produced amounted to 6,352.

Whatever their shortcomings, Grants provided the Eighth Army with tanks that were better armed than any it previously had. It also enjoyed, once again, numerical superiority. In spite of all this, when the Afrika Korps attacked the Gazala line the Eighth Army was defeated piecemeal, and having lost most of its tanks had to retreat into Egypt. It was pursued to within 60 miles of Alexandria when it halted the enemy advance by a series of counter-attacks at Alamein, where another battle took place three months later that changed radically the course of the war in North Africa.

In the meantime, on 22 June 1941 the German Army invaded the Soviet Union. The invasion was spearheaded by four panzer groups, each of which consisted of three to five panzer divisions and which between them contained 17 out of the existing 20 divisions. The panzer groups drove deep into Soviet territory and in a series of envelopments inflicted enormous losses on Soviet forces. They were only halted in the winter of 1941 at the gates of Moscow and Leningrad and deep in the Ukraine by a combination of their own exhaustion, Soviet counter-attacks and the weather.

When the invasion began, Soviet armoured forces were in a state of flux. The successes of the German armoured forces in Poland and in France led in July 1940 to a reversal of the earlier decision to disband large mechanized formations. There were now to be eight mechanized corps and in February 1941 Soviet High Command called for 21 more to be formed. Each of the mechanized corps was to consist of two tank and one motorized infantry divisions and to have 1,031 tanks. Each of the tank divisions was to consist of two tank regiments with a total of 375 tanks, one motorized infantry regiment and a battalion each of reconnaissance, anti-tank, anti-aircraft, engineer and signals troops.

The organization of the new formations had hardly settled when the German forces struck. Moreover, the leadership of the Soviet armoured forces had not recovered from the murderous purges of the previous four years. Many of the Soviet tanks were also claimed to be in need of overhauls or at least of spare parts that would enable them to operate for any length of time. But, for all the shortcomings of its armoured forces, the Red Army had a total of 24,000 tanks, according to what Stalin himself told Harry Hopkins, the US president’s personal representative. Post-war Soviet accounts put the total at a slightly lower figure of 22,600, but in any event by the end of 1941 the Soviet Army had lost 20,500 of them, which means that during the first phase of the German-Soviet war virtually the whole of the pre-war Soviet tank strength was wiped out.

The 17 panzer divisions that were largely responsible for this remarkable achievement started the campaign with a total of only 3,266 tanks, including command tanks. The most numerous of them were PzKpfw IIIs, 707 of which were now armed with the 50mm L/42 gun, but 259 were still armed with the 37mm tank gun that had already proved inadequate during the campaign in France. There were also 625 PzKpfw 38(t)s and 155 PzKpfw 35(t)s armed with similar Czech-made 37mm guns. The most powerful tank continued to be the PzKpfw IV, which was still armed with the short-barrelled 75mm L/24 gun, but there were only 439 of them.

Most of the Soviet tanks were T-26s and BTs which were armed with 45mm guns that were as good as the German 50mm L/42, but their armour was relatively thin and the vision from their turrets was confined to a single rotatable periscope that limited their commanders’ situation awareness, so much so that Finns fighting them a year earlier observed that they appeared to be ‘blind’. The situation was aggravated by the design of their two-man turrets in which the commander acted as the gunner, in contrast to other tanks with two-man turrets, like the British Valentine, in which the commander acted as the loader and had therefore a better chance of observing what was going on around him.

German PzKpfw III and IV tanks, with their three-man turrets and a commander free to observe the tactical scene, could therefore outmaneuver Soviet tanks, and they proved more than a match for them.

However, on the first day of the invasion some panzer divisions also ran into the Soviet KV and T-34 tanks, which were a complete surprise to them and caused considerable consternation because they proved almost immune to the panzers’ guns. Yet the two new Soviet tanks had been in production for more than a year and by the time the Soviet Union was invaded as many as 636 KVs and 1,215 T-34s had been produced. Moreover, Soviet authorities made no particular secret of the existence of the T-34, as a month before the invasion they allowed a well-known American photographer, Margaret Burke-White, to visit a tank school outside Moscow and take pictures of the T-34 that were then published in the United States in the widely read Life magazine.

However, in spite of being a nasty surprise to the panzer divisions and superior to their tanks in terms of armour protection as well as gun power, the deployment by the Red Army of the KV and T-34 had no effect on the overall course of the campaign. This fact was obscured for a number of years by contemporary Soviet propaganda, which falsely claimed that the T-34 was only deployed when the German forces were approaching Moscow and that it was responsible for them being driven back.

The appearance of the T-34 inevitably led to demands by German tankmen for new and more powerful tanks, and consequently a special commission consisting of the leading German tank designers visited Guderian’s panzer group in November 1941 to assess the situation at first hand. Soon afterwards contracts were awarded to the Daimler Benz and MAN companies, which had done some studies since 1938 of a 20-tonne tank, for the development of a new 30-tonne tank armed with a very long barrelled 75mm L/70 gun. In May 1942 Hitler opted for the MAN design, and after trials of prototypes the first two production vehicles were completed in January 1943.

The new tank, which was called Panther, outgunned the Soviet T-34 and had thicker armour. It was also larger, having a five-man crew, and as a result of this and its thicker armour it was also heavier, weighing 43 tonnes. In spite of this, it performed well over soft ground due to its wide tracks and a suspension with interleaved road wheels that spread the load over the earth, and its design scored well from the point of view of what was later called ‘fightability’. In fact, it came to be widely regarded as the best medium tank of the Second World War, although it suffered at first from mechanical problems due to its hurried development.

Production of the Panther was preceded by that of another powerful German tank, the 570-tonne Tiger armed with an 88mm L/56 gun. The development of this heavy tank was not begun, as is sometimes claimed, in response to the appearance of the T-34, but can be traced as far back as 1935 when the Ordnance Department first considered a 30-tonne tank armed with a 75mm gun that would be effective against French 2C, 3C and D heavy tanks. This was not a well-informed objective as 2C was already obsolete and 3C and D heavy tanks never existed, but in 1937 the Henschel company was asked to design a 30-tonne DW or breakthrough tank. By 1940 a 30-tonne tank was also designed by Ferdinand Porsche and in 1941 Krupp was awarded a contract for the development of a turret mounting a tank version of the 88mm L/56 anti-aircraft gun that had proved so effective in a ground role in Spain and in France. This was followed a month before the invasion of the Soviet Union by an order issued to Porsche and to Henschel to develop a tank in the 45 tonne class, which they did on the basis of their earlier 30-tonne tank designs. Porsche, who was apt to adopt novel but not always very practical ideas, produced a tank that had problems with its electric transmission and novel suspension and this led to Henschel’s tank being selected and produced as the Tiger.

As soon as they were ready, Hitler foolishly ordered four Tigers to be used on the Leningrad front, where they first saw action in October 1942. They were employed over unsuitable swampy terrain and one had to be abandoned in a peat bog from which it was recovered intact in January 1943 by the Russians, who were consequently not only forewarned of the existence of the new tank but were able to assess its characteristics in detail. In spite of this inauspicious debut, Tiger I or Model E became for a time the most powerfully armed tank in the world as well as having thicker armour than the British Matilda and the Soviet KV, and the 1,354 that were produced took a heavy toll of enemy tanks.

While the production of Tiger I and of the Panther was getting under way, a more immediate answer to the new Soviet tanks was found in PzKpfw IV re-armed with a long-barrelled 75mm L/43 gun instead of its original 75mm L/24. As a result PzKpfw IV not only caught up with Soviet tanks, which were re-armed two years earlier with a 76mm gun, 41.5 instead of 30.5 calibres long, but outperformed them. The first of the re-armed PzKpfw IV was produced in March 1942 and it remained effective until the end of the war, by which time 7,419 had been produced.

When the decision was taken in November 1941 to arm PzKpfw IV with the 75mm L/43 Hitler decided that the Sturmgeschutz or assault guns should also be armed with it. Sturmgeschutz, or StuG for short, were originally developed as a result of the acceptance by the German High Command of the policy advocated before the war by General Lutz and Guderian of concentrating all the available tanks in mobile formations and not allocating any to infantry support. This led the infantry to demand an armoured vehicle that could provide it with close support assault and anti-tank artillery. An order was consequently issued in 1936 for the development of such a vehicle and the first was produced in 1940. It was based on the chassis of the PzKpfw III and was armed with the same 75mm L/24 gun as the PzKpfw IV but mounted in the hull.

StuG was in effect a ‘turretless tank’. Because it had no turret, it had a lower silhouette and thicker armour in relation to its weight, as well as being cheaper to produce than a tank. It was less suitable for mobile warfare because of the limited traverse of its armament, but when armed with the 75mm L/43 it proved to be a highly effective anti-tank vehicle, so much so that it was credited with the destruction of 20,000 enemy tanks by 1944. On the eve of the invasion of the Soviet Union the German Army had 391 StuGs, and subsequently their number rose steadily. By the end of the war, a total of 9,409 had been produced and in spite of losses there were still 3,831 in use, making StuG the most numerous German armoured fighting vehicle at the time. Except when there was a shortage of tanks in the latter part of the war, StuG were not issued to the panzer regiments but were organized into separate battalions, which were used primarily to support infantry divisions.

When StuG and PzKpfw IV, armed with the 75mm L/43 and later L/48, began to be introduced in the spring of 1942 and were followed by the first Tigers and then by Panthers, the German Army reversed the situation in which it found itself when it invaded the Soviet Union and ran into the new Soviet tanks. It now possessed qualitative superiority that was to last until the end of the war.

In contrast, the Red Army did not for a time make any major changes to the tanks it had already developed, but concentrated on producing the maximum number of them to make good the losses suffered in 1941 and to regain numerical superiority. The continued production of a virtually unchanged T-34 is particularly noteworthy in view of the recognition of its shortcomings, which were brought home by the evaluation of two PzKpfw III purchased in the summer of 1940 when the relations between the Soviet Union and Germany were still amicable. Compared with the PzKpfw III, the T-34 had superior armour and armament, but its cramped two-man turret was obviously inferior to the three-man turret of the German tank and it lacked the latter’s commander’s cupola, which provided good all-round vision. The torsion bar suspension of the PzKpfw III was also found to be superior to the Christie-type coil spring suspension of the T-34. As a result, a new T-34M tank was hastily designed incorporating a three-man turret and torsion bar suspension. Two prototypes began to be assembled in March 1941, but three months later the Soviet Union was invaded and further development of the T-34M, which is seldom mentioned in all the writing about the T-34, was abandoned.

Large scale production of the T-34 continued, although it suffered a temporary setback when the Kharkov plant where it originated was threatened by the German advance and a decision was taken in September 1941 to evacuate it as well as other plants, including the Leningrad plant producing KV heavy tanks, to the Urals. For a time this left the Stalingrad plant as the only major producer of T-34s, but a most remarkable industrial effort resulted in the first T-34 being produced in the Urals as early as December 1941.

In spite of the temporary interruption of tank production and the staggering losses suffered during the first six months of the war, the Red Army had 7,700 tanks at the end of 1941. This compared well with the total of 5,004 tanks that the German Army had at the time. Some of the Soviet tanks were in the Far East facing a possible Japanese threat while some of the German tanks were being sent to North Africa, but nevertheless the Red Army continued to enjoy considerable numerical superiority over the German Army. This became much more marked during 1942 when Soviet industry produced a total of 24,668 tanks, including 12,527 T-34s. As a result of this and in spite of further heavy losses, by the end of the year the Red Army had 20,600 tanks, whereas the German had only increased the number of its tanks to 5,931, although it had also increased the number of StuGs to 1,039.51 During 1943 the Red Army lost almost the equal of that year’s output of 24,000 tanks and assault guns, which included 15,833 T-34s. But in the following year production exceeded losses, and by the end of it the number of tanks and assault guns the Red Army had rose to 35,400. The number the German Army had also increased, but only to 12,451, and by then its panzers were facing not only Soviet tanks but also thousands of American and British tanks in Western Europe.

Their numerical inferiority did not prevent German armoured forces destroying more Soviet formations when these counter-attacked around Kharkov in May 1942 and later at Rzhev. But when they took part in the German offensive in June, Hitler split them between an assault on the industrial city of Stalingrad and an equally misguided drive aimed at the Caucasus oilfields, which overstretched their resources. This helped the Red Army to break through the German front in November 1942 and led to the encirclement of Stalingrad, where the remnants of the Sixth Army, including three panzer divisions, surrendered in January 1943. However, a month later panzer formations under the command of Field Marshal E. von Manstein smashed another Soviet offensive in the Donets basin and at Kharkov in what became a classic example of maneuver warfare.

The German High Command then conceived the idea of an offensive code-named Zitadelle against a Russian salient around Kursk, which was to use the revitalized panzer formations to destroy a large number of Soviet divisions and thereby weaken the offensive capabilities of the Red Army. Guderian and other generals objected to it and even Hilter had qualms about it, but the offensive went ahead in July 1943. Seventeen panzer divisions were assembled for it with a total of about 2,450 tanks and assault guns. They included 133 Tigers and 184 brand new Panthers. But the offensive took little advantage of the mobile warfare skills of the panzer forces. Instead, they were made to assault where the Red Army expected them and where it had prepared extensive minefields and other defences backed by about 2,950 tanks. In consequence they became involved in a battle of attrition and failed to achieve the planned encirclement of the Soviet forces, although they inflicted heavy losses on them.

Particularly intensive fighting took place near the railway junction of Prokhorovka, which has been described since as the greatest tank battle. In fact, it was a meeting engagement between the 2nd SS Panzer Corps, which had 294 tanks and assault guns including 14 Tigers, and the reinforced 5th Guards Tank Army, which had about 850 tanks. The latter were mainly T-34s but included 260 T-60 light tanks, which were easy targets for German guns, while the T-34s were completely outranged by the Tigers and consequently charged to close with them. In spite of this, by the end of the day the 5th Guards Tank Army had lost as many as 600 of its tanks, 334 of which were completely destroyed, while the SS Corps suffered a total loss of only 36 tanks and assault guns. These figures refute the description in some books of the Battle of Prokhorovka as a ‘death ride’ of the panzer divisions. In fact, over the whole of the Kursk salient the German Army lost 278 tanks and assault guns, including 13 Tigers and 44 Panthers, compared with a total loss of 1,254 tanks suffered by the Red Army.

However, German offensive operations were stopped after the Battle of Prokhorovka by Hitler, who became concerned about the Anglo-American landings in Sicily that had just taken place, and decided to withdraw the SS Panzer Corps so that it could be transferred to the West. The remaining panzer formations retained their qualitative superiority and the ability to score tactical successes and to inflict heavy losses on their enemies. But Zitadelle was their last major offensive on the Eastern Front. In its aftermath strategic initiative passed into the hands of the Red Army, which became increasingly proficient at the offensive operations that came to dominate the latter part of the war in Eastern and Central Europe.

At the beginning of the war, the Red Army had 30 of the mechanized corps it started creating in 1940, but most of them were quickly destroyed and in July 1941 they were officially abolished. Instead the remaining tank units were reorganized into independent brigades that were confined to close support of the infantry. Each brigade had 46 to 93 tanks made up of a mixture of KVs, T-34s and whatever light tanks were available. But as the Red Army began to regain its strength, it re-created four tank corps in March 1942. Initially each had two tank and one motorized infantry brigades, but a third tank brigade was added later, which brought their strength up to 98 T-34s and 70 light tanks. At the same time they dispensed with the KVs, which were not mobile enough for them and which were organized into independent tank regiments that would be used for infantry support.

By the end of 1942 the Red Army already had 28 tank corps. It had also created eight mechanized corps, each of which had one tank brigade and three mechanized brigades consisting of three motorized infantry battalions and a tank regiment, and each had a total of 100 T-34s and 104 other tanks. The tank and the mechanized corps were well designed for mobile operations of limited scope, but for larger scale penetrations of the enemy front and encirclement several would have to be combined, which led to the creation in May 1942 of the first two tank armies that corresponded to the German panzer corps, just as the Soviet tank corps corresponded to the panzer divisions.

The reorganization of the Soviet armoured forces did not prevent their defeat in the Battle of Kharkov in May 1942, but they played a major role in the encirclement of Stalingrad and after the Battle of Kursk they led the offensives that restored Soviet control over Ukraine, Belarus and the Baltic states. During this period new types of armoured vehicles came into use, starting in 1943 with the SU-122, a ‘turretless tank’ on the lines of the German assault guns, which consisted of a 122mm howitzer mounted in the hull of the T-34. It was relatively ineffective and was quickly superseded by the SU-85, which was very similar except for being armed with a long-barrelled 85mm gun. Adoption of the 85mm gun was prompted by firing tests carried out with the Tiger captured on the Leningrad front, which brought out the need for a more powerful weapon than the 76mm guns of the contemporary Soviet tanks to defeat its 100mm thick frontal armour. About 2,050 SU-85s were produced by the autumn of 1944 when the SU-85 was succeeded by the SU-100, which was similar except for being armed with a long-barrelled 100mm gun. The gun of the SU-100 was an adaptation of a naval gun, just as the 85mm gun of SU-85 was an adaptation of an anti-aircraft gun, which speeded up its development and facilitated the production of about 1,200 by the end of the war. The guns of SU-85 and SU-100 made them effective as tank destroyers, and the T-34 chassis on which they were based provided them with the mobility required for operation with armoured forces.

Before SU-85 and SU-100 were developed, a small number of another turretless assault gun was used at the Battle of Kursk. This was the 45.5-tonne SU-152, which represented the second and much more sensible attempt to mount a 152mm howitzer on the KV tank chassis than the first, which involved mounting it in a huge turret. The resulting KV II was used in 1940 in the assault on Finnish defences, but proved unsuitable for more mobile warfare and disappeared shortly after the German invasion in 1941.

New types of tanks were also developed during 1943. One of them was the new version of the T-34 armed with an 85mm gun mounted in a three-man turret, the first of which was issued to Soviet tank units in March 1944. It was still inferior to the German Panther so far as its gun performance and frontal armour were concerned, but it outnumbered the latter, 18,000 having been produced by the end of the war compared with 5,966 Panthers, not all of which were of course available for the Eastern Front. New heavy tanks were also developed, first by re-arming the KV with the same 85mm gun as that originally mounted in the SU-85 to produce the KV-85, of which only 130 were built in 1943. It was followed by Iosef Stalin, or IS, which had a more heavily armoured KV chassis and dispensed with the fifth crew member, who was the hull machine gunner. IS-1 was armed with an 85mm gun but IS-2 was armed with a 122mm gun. The gun was once again an adaptation, in this case of an artillery gun, and with it the 46-tonne IS-2 matched the Panther and the Tiger in gun power, but its rate of fire was slow and it only carried 28 rounds of ammunition. IS-2 began to be produced by the end of 1943 and to be issued in the spring of the following year to independent heavy tank regiments, which were used to support medium tanks by their fire.

To counter IS-2, the German Army had the 68-tonne Tiger II, as well as the turretless Jagdpanther armed with a 88mm L/71 gun that was longer barrelled and more powerful than the 88mm L/56 gun of Tiger I. Eventually 489 Tiger IIs were built, but they were completely outnumbered by IS-2, the production of which amounted to 3,207 tanks by the end of the war. The German Army also had the Jagdtiger armed with a long barrelled 128mm gun, the first of which was built in October 1943 but whose production was disrupted by air raids, so that only five were in the hands of the troops in June 1944. The 70 tonne Jagdtiger was the most powerfully armed and the most heavily armoured vehicle of the Second World War, having 250mm thick frontal armour, but only 77 were built.

The Red Army made full use of its numerical superiority by launching simultaneous offensives along different parts of the Eastern Front and destroying separately parts of the German Army. In this it was helped by Hitler’s disastrous strategy, which required German forces to hold on to their positions instead of being allowed to operate more freely. In particular they were expected to hold on to cities and towns designated Feste Platze, or fortresses, which were to break the momentum of the Soviet onslaught. What this did instead was to split German forces between isolated strong points in which they could be more easily encircled and destroyed piecemeal. These methods contributed, among others, to the destruction of the Army Group Centre in June 1944 in Belarus, which came to be regarded as an even greater catastrophe for the German Army than Stalingrad.

The offensives of the Red Army brought it in April 1945 to the gates of Berlin, which it stormed, delivering a coup de grâce to Hitler’s Reich. The forces that assaulted Berlin included four tank armies and a total of 6,250 tanks and assault guns. Determined resistance in urban terrain far less suitable for the operation of armoured forces than the plains of Eastern Europe exacted a heavy toll on the Soviet armoured units, which lost 1,997 tanks and assault guns, more than the 1,519 the German defenders had when the assault on Berlin began.

As they were being overwhelmed on the Eastern Front, German forces were also gradually overcome in the west. The process began at the Second Battle of Alamein in October 1942, when the British Eighth Army commanded by General B. Montgomery attacked the German and Italian forces that had advanced into Egypt. The latter included two panzer divisions with a total by then of 211 PzKpfw IIIs and IVs and two Italian armoured divisions with 280 M 13/40s. The Eighth Army facing them had three armoured divisions, two of which were reinforced by the attachment of a second armoured brigade, and two independent armoured brigades. In total, therefore, it had seven armoured brigades and 1,441 tanks backed by a reserve of 1,230 tanks held in Egypt in depots, workshops and training units. The disparity in the resources was therefore considerable and even greater than these figures would indicate because, of the German tanks, only 30 were PzKpfw IV armed with the long barrelled 75mm L/43 gun, whereas tanks deployed by the Eighth Army included not only 170 Grants but also 252 newly arrived US-built M4 medium tanks, which the British Army called Shermans.

Shermans were armed with 75mm guns that were somewhat better at penetrating armour than the 75mm guns of the Grants, although not as good in this respect as the 75mm L/43 guns of PzKpfw IV. However, unlike the hull-mounted guns of the Grants, those of the Shermans were mounted in turrets, which made them tactically more effective, and they also fired high explosive as well as armour-piercing ammunition unlike British-built tanks, most of which were still armed with 40mm 2-pounders that only fired solid shot.

Taking advantage of its newly received tanks and its numerical superiority, the Eighth Army wore down the tank strength of the German and Italian forces in a series of attacks that on the 13th day of the battle forced them to retreat, at which stage they were left with ten German and no Italian tanks.

The M4 medium tank, or Sherman, which the British Army first used at Alamein was developed as a result of decisions taken by the US Army as early as August 1940, even before the M3 medium tank and its British Grant version were designed, to follow them as soon as possible with a tank also armed with the 75mm gun but mounted in a turret. Not to delay production, the M4 used basically the same chassis as the M3 medium tank while its general configuration followed that of PzKpfw IV. A pilot model was completed in September 1941 with series production beginning in February 1942. Apart from light tanks, the M4 became almost the only tank used by the US Army up to the end of the Second World War, by which time a total of 49,234 were produced. It also became the principal tank of the British Army.

The Sherman was used to an increasing extent by the British Army after the Battle of Alamein not because there was a shortage of British tanks but because of their shortcomings. In fact, the number of tanks produced in Britain in 1940 was the same as that produced in Germany and in 1941 it overtook the latter, producing 4,811 compared with 3,114. Even more were produced in Britain in 1942 when the annual output rose to 8,611 tanks, which was more than twice the number produced in that year in Germany.

Unfortunately, much of the considerable British production effort was misdirected or even wasted. The extreme example of this is the Covenanter cruiser tanks, 1,365 of which were produced but none of which was considered fit for battle.

The failings of the Covenanter were due to a considerable extent to the company responsible for it having little experience of tank design. A similar situation existed in the case of other tanks, such as the A.13 and Crusader cruiser tanks, which acquired a reputation for unreliability when used in North Africa. Some of the problems were aggravated by the way in which tanks like the Covenanter and Crusader were rushed into production, and these persisted because rectifying them would have interfered with the production of the maximum number of tanks that was demanded after the defeat of France in 1940, when the British Expeditionary Force lost about 700 tanks. The perceived shortage of tanks that drove the demand for the production of the largest possible number of tanks was exaggerated by Churchill, who stated two years later in the House of Commons that ‘we had … in the United Kingdom less than 100 tanks’. In fact, production records that have come to light since then indicate that in spite of the tanks lost in France and some 300 sent to the British forces in Egypt, there must have been still at least 700 tanks in Britain.

There was also a perception that British tanks were outgunned by German tanks. Actually, the 40mm 2-pounder gun of the British tanks could perforate thicker armour than the 37mm and 50mm L/42 guns of most German tanks, and it was only the long-barrelled 50mm L/60 introduced in 1942 that was superior to it. Where the British tanks were consistently deficient was in not having a gun capable of firing effective high explosive as well as armour-piercing ammunition, like the 75mm gun of the PzKpfw IV even in its original short-barrelled L/24 form. When the British tanks finally advanced beyond the 40mm gun, they went no further than the 57mm 6-pounder with which the Crusader was re-armed in 1942. The 6-pounder was a very effective anti-tank gun, as good in fact in this respect as the long-barrelled 75mm L/48 gun of the PzKpfw IV, but as a high explosive firing weapon it has been described as ‘useless’. Thus it was only with the arrival in 1942 of the American-built Grants and then of Shermans armed with 75mm guns that British armoured units were equipped with tanks capable of firing not only armour piercing but also effective high explosive ammunition.

Yet in 1943 and even in 1944 the General Staff and the War Office were unable or unwilling to accept that both types of fire should and could be delivered by every tank. They accepted, rather reluctantly, that some British tanks might be armed with ‘dual purpose’ guns, but expected that others would specialize in one or the other of the two functions. This would have perpetuated the specialization that bedevilled British tank development, manifesting itself in the division into infantry and cruiser tanks and the arming of tanks with the 40mm 2-pounder whose ammunition was only effective against other tanks. It was only during the last two years of the war that this tendency to over-specialize began to fade away.

In the meantime, the Eighth Army chased what was left of the Afrika Korps out of Egypt and pursued it across Cyrenaica into Tripolitania, where it was reinforced by the remaining Italian armoured division, the Centauro. However, the latter was still only equipped with M 13/40s or the very similar M 14/41 tanks, which were by then completely outgunned by tanks like the Sherman. After some delaying actions, the German and Italian forces retreated into Tunisia until they reached the Mareth Line of fortifications built before the Second World War by the French against an Italian invasion from Libya.

Two months earlier, in November 1942, Anglo-American forces landed on the Moroccan and Algerian coasts of what was then French North Africa and after overcoming some French resistance advanced on Tunisia. The German High Command reacted by landing a panzer division as well as other troops, including some Tiger tanks, in Tunisia. Having built up their strength, the German forces in Tunisia severely mauled the US 1st Armoured Division that had advanced from the west to the Kasserine Pass, destroying more than one hundred of its tanks, which included Lees, the US equivalent of the British Grants, as well as Shermans. They then turned against the British Eighth Army but were repulsed at Medenine. The Eighth Army subsequently stormed the Mareth Line, in which Valentine tanks played a prominent part, but after one more major action in Tunisia units equipped with them were provided instead with Shermans. The Crusaders, with which the British 6th Armoured Division was partly equipped when it landed in Algeria, were also replaced at the time by Shermans.

Towards the end of the campaign in Tunisia, which ended with the surrender of the German and Italian forces in May 1943, British forces were reinforced by two brigades, or about 300 Churchill infantry tanks. These 39-tonne tanks were designed during the ‘Phoney War’ that preceded the fall of France in 1940, when tanks were expected to have to operate over shell-torn ground similar to that encountered during the First World War. They were relatively slow but better armoured than the Matilda infantry tanks. However, in spite of their weight, they were originally only armed with 40mm 2-pounders, although the Mark I version also had a 3in (76.2mm) howitzer mounted in place of the hull machine gun. But before they were deployed in Tunisia they were re-armed with the 57mm 6-pounders, which, together with their ability to operate over difficult ground, made them effective in the close mountainous terrain in which the Tunisian campaign was fought. On the other hand, the 52 Tigers that were sent by the German High Command to Tunisia were misused there when they could have been better employed on the Russian Front, where the open country made their 88mm guns much more effective.

After their victory in Tunisia, Anglo-American forces invaded Sicily and then advanced slowly up the Italian peninsula, where the terrain generally restricted movement off the roads. The scale of tank operations was therefore limited, and they commonly consisted of small scale actions in close support of the infantry. By the same token, the incidence of tank versus tank fighting was low, although Allied forces included a considerable number of armoured units. These included an armoured division and eight separate tank battalions in the case of the US Fifth Army and, eventually, three armoured divisions and two independent tank brigades in the case of the British Eighth Army. The opposing German forces included, at different times, one or two panzer divisions and an independent heavy tank battalion with up to 45 Tigers, as well as another battalion equipped with 76 Panthers that first went into action against the Allied landing at Anzio in February 1944.

All the British as well as US tank units in Italy were equipped with Shermans, except for a number of US-built M5 light tanks, which were a direct development of the earlier M3 light tanks, and the Churchill infantry tanks of the two British independent tank brigades. The same types of tanks were used by the Allied forces that landed in Normandy in June 1944, except for some British armoured units that were equipped with a new type of cruiser tank, the 27.5-tonne Cromwell, instead of Shermans.

Development that led to the Cromwell started in 1941 with the design of the very similar Cavalier and then Centaur cruiser tanks. Both were intended to be more heavily armoured successors of the Crusader, but were powered by the same Nuffield Liberty engine as the latter and were armed with the same 57mm 6-pounder as Crusader III. However, Centaur was subsequently fitted with a more powerful 600hp Meteor engine, which was a de-rated unsupercharged version of the Rolls-Royce Merlin aircraft engine that powered the Hurricane and Spitfire fighter aircraft of the Royal Air Force, and this, together with the Merritt-Brown transmission proven in Churchill tanks, converted it into the Cromwell and enabled the latter to overcome the reputation for unreliability acquired by British tanks. Except for its early versions, which were still armed with the 6-pounder, Cromwell was armed with a 75mm gun that fired the same ammunition as the 75mm gun of the Shermans.

In several respects Cromwell represented a considerable advance on earlier British tanks. However, so far as its principal characteristics – its gun and its armour – were concerned, it was no better than the Soviet T-34 that had been introduced three years earlier. When the writer brought this out some years later, the originator of the British cruiser tank development, General Martel, took exception to it and ignoring the facts claimed that the T-34 was ‘far inferior to the Cromwell’. The Russians did not think so, for when they were offered Cromwells in 1943 under the military assistance programme they turned them down. Instead they asked for more Valentines, which the Red Army was using as light tanks. A total of 2,394 of them was sent from Britain to the Soviet Union in addition to all but 30 of the 1,420 Valentines produced in Canada, although about 300 were sunk en route in Arctic convoys.

Like the Shermans, Cromwells were outgunned by the German tanks, but a more immediate problem facing British and US armies was that of landing on heavily defended beaches and then breaking through the coastal defences. This called for tanks that could be launched from ships and swim ashore, a capability foreshadowed in 1924 during a US Marine Corps exercise when an armoured amphibious vehicle built by J. W. Christie swam from a battleship to a Puerto Rican beach. In 1931 Vickers Armstrongs built two prototypes of the first successful amphibious light tanks, the A4E11 and A4E12, which were copied in the Soviet Union as the T-37 and T-38 and about 4,000 of which were produced for the Red Army between 1933 and 1939. But they were small 3-tonne two-man tanks armed with a single machine gun that could only swim in calm inland waters. Heavier tanks could not be made to float except by attaching large pontoons to them, which was not very practicable, until an ingenious system was devised in Britain by N. Straussler, a Hungarian engineer who had previously designed armoured cars for the Royal Air Force and for the Netherlands East Indies Army.

Straussler’s system involved the use of a canvas floatation screen that, when erected, provided the necessary buoyancy and when collapsed enabled a tank to operate in the usual way. In water a tank could propel itself at up to 6mph by means of two propellers driven by its tracks, which came to be called ‘Duplex Drive’ or DD, by which tanks fitted with the floatation system are generally known. The first to be modified into a DD tank was a 7.5-tonne Tetrarch light tank, which was tested in 1941. It was followed by the conversion into DD tanks of some 600 Valentines, which were only used by the British Army for trials and training, and then by the heavier 30-tonne Shermans, which in their DD form equipped three US as well as three British and two Canadian battalions or regiments that were earmarked for the assault landings in Normandy. In the event, four out of the eight units did not swim ashore but, because of rough seas, were taken directly on to the beaches by the landing craft. The fortunes of the other four units varied considerably: one of the US battalions, on Utah beach, landed all but one of the 30 tanks launched into the sea from landing craft, but of the 29 tanks launched by another US battalion 27 sank well short of the Omaha beach they were to assault.

The British, but not the US, Army also made considerable use of tanks modified to perform special tasks, and together with the DD tanks they formed the 79th Armoured Division. Its units included three regiments of Shermans fitted with mine-detonating flails, which were called Crabs, and three regiments of Assault Vehicles Royal Engineers, or AVREs, which were Churchill tanks re-armed with spigot mortars that fired large demolition charges. AVREs were also made to carry fascines, or large bundles of brushwood, which were used as in the First World War to fill trenches for crossing them, and they also carried assault bridges and rolls of hessian carpet that were unrolled over patches of soft ground that was difficult for vehicles to cross.

Tanks of the 79th Armoured Division, which led the assault in the British sector, won the fire fight on the beaches and enabled the infantry to follow on to their objectives at relatively low cost. An exception to this were three battalions of searchlight tanks, code-named Canal Defence Lights or CDL, which played no effective part in the Normandy campaign. Development of searchlight tanks began well before the Second World War and they were originally expected to dazzle the enemy, or to ‘attack by illumination’ as General Fuller described it, arguing, somewhat naively, that they were a means of winning wars. In fact, they were only used once or twice in the closing stages of the war for night illumination. Much the same applied to the CDL tanks of the US Army, which followed the British example and raised six battalions of them. But development of the CDL tanks proved to be a fiasco and the resources devoted to them would have been better spent elsewhere.

Once they created a bridgehead in Normandy the Anglo-American forces faced the reaction of the German forces and in particular of the panzer formations stationed in France. The latter had a total of 1,673 tanks and assault guns made up of 758 PzKpfw IVs, 655 Panthers, 102 Tigers and 158 StuGs. All outgunned British and US tanks, except for some Shermans armed with the British 76mm 17-pounder that outranged the 75mm L/48 gun of the PzKpfw IV and was comparable to the 75mm L/70 gun of the Panther. However, German tanks were deprived of some of their advantage by the hedgerows of the Normandy bocage, which restricted the range at which targets could be engaged. The overall effectiveness of the panzer formations was also reduced by their piecemeal deployment and by Hitler’s irrational operational orders.

Nevertheless, panzer formations inflicted severe losses on the Allied forces and checked a thrust out of the bridgehead, called Operation Goodwood, by three British armoured divisions with a total of about 700 tanks. But in the end they succumbed to the superior numbers of Allied tanks, backed by massive aerial bombardment. In the US sector, five armoured divisions with a total of about 1,500 tanks broke through at St Lô, while on their left the British Second Army attacked with three armoured divisions and two armoured brigades, or more than one thousand tanks, and a week later the First Canadian Army attacked with two more armoured divisions and two armoured brigades. In the meantime Hitler issued an order for a counter-offensive against the flank of the American advance, which proved disastrous as it exposed the attacking German forces to envelopment and led to them being trapped in the Falaise Pocket. Many of the German troops managed to escape, but most of their equipment was lost. More of what was left was subsequently lost when the remnants of seven panzer divisions retreated across the Seine under aerial attack, so that they were only able to bring out about 100 or 120 tanks.

The 11 or 12 Allied armoured divisions that broke out of the Normandy bridgehead and then advanced rapidly across France to the Belgian and German borders were all equipped with Sherman and M5 light tanks, except for the British 7th Armoured Division, which was equipped almost entirely with Cromwells, and two other British and one Polish armoured divisions, which had one regiment of Cromwells in addition to three regiments of Shermans. Both Shermans and Cromwells were armed with 75mm guns that could not perforate the frontal armour of the German Panthers and Tigers even at point blank range, while the latter could perforate theirs at 2km. To some extent Allied tanks were able to redress the balance by exploiting their numerical superiority and mobility to attack the more vulnerable sides of the German tanks. But qualitatively German tanks were superior.

The need for a more powerfully armed British tank had been recognized two years earlier and led to the development of the Challenger armed with the 76mm 17-pounder anti-tank gun. The new tank was, in effect, a lengthened Cromwell with a large, clumsy turret and did not prove entirely satisfactory. Nevertheless, 200 were ordered in 1943 and some were later used by the Cromwell equipped regiments as ‘tank killers’. In the meantime it was found that the 17-pounder could be squeezed into the turret of the Shermans, and this proved to be a better way of using it. In consequence, employment of the 17-pounder was concentrated on the Shermans and tanks re-armed with it, called Fireflies, were issued to British tank units on the scale of one 17-pounder tank to three 75mm gun tanks. Initially only 84 were actually deployed and by the end of the second month there were still only 235 of them in the field. However, by the end of the war the British 21st Army Group had 1,235 Shermans with 17-pounders compared with 1,915 others still armed with 75mm guns, and they provided it with tanks which were at last as well armed as the Panther.

The 17-pounder was made even more effective by the introduction towards the end of the campaign of Armour Piercing Discarding Sabot (APDS) ammunition with projectiles consisting of a hard-high-density tungsten carbide sub-calibre shot within a pot-like aluminium carrier or sabot that separated from the shot at the muzzle. In spite of the loss of some of the kinetic energy imparted to the projectile by the gun to the sabot, most of it was still in the shot, which because of its smaller cross-sectional area penetrated more of the target than a conventional full calibre projectile.

Tungsten-cored ammunition was actually provided for the 37mm guns of German tanks as early as the 1940 campaign in France, but in their case the carrier did not separate from the shot and the velocity and penetration of the latter fell off rapidly with range. This type of ammunition was called Armour Piercing Composite Rigid (APCR), and after 1940 was used by German tank guns on the Eastern Front and in North Africa, but on a limited scale because of shortages of tungsten.

APDS ammunition was superior to APCR because its performance did not fall off as rapidly with range. It was first provided in Normandy for the 57mm 6-pounders that were still mounted in some of the Churchill tanks. However, most Churchills were by then armed with 75mm guns so this had little impact on the situation. It was only when APDS began to be provided for the 17-pounders of the Fireflies that its effectiveness began to tell. In fact, its armour penetration was 40 per cent greater at 1,000m than that of the conventional APCBC (Armour Piercing, Capped, Ballistically Capped) ammunition, although its dispersion and therefore loss of accuracy limited the range at which it could be effectively used.

For all this, the re-arming of the Shermans with the 17-pounder was only a makeshift solution that was adopted pending the development of a new and equally well-armed tank. Challenger failed to become one and before another attempt was made to develop it the General Staff opted for another stopgap, which was a derivative of the Cromwell armed with a new lower powered version of the 17-pounder called the Comet. Four regiments were equipped with it and it saw some action in the closing stages of the war. In the meantime, in May 1944, a decision was finally taken to develop another cruiser armed with the 17-pounder. Six prototypes of this 42-tonne tank called the Centurion were rushed to Germany in May 1945, which was too late for them to see any action, but the Centurion became one of the most successful British tanks ever built.

At about the time the Challenger began to be developed in Britain in 1942, US Ordnance also saw the need for tanks to be armed with a gun more powerful than the 75mm gun of the Shermans and started to develop such a gun. This led to a 76mm gun with a higher muzzle velocity and therefore greater armour penetration, but its adoption was not pursued with any urgency largely because the Army Ground Forces commanded by General L. J. McNair, which controlled the acquisition of equipment, regarded armoured forces as no more than a reincarnation of the 19th century cavalry that should be used for exploiting the successes won by other arms and not to fight enemy armoured forces. Tanks were not to be armed therefore to fight other tanks, which were to be fought instead by units of tank destroyers, such as the M10, which had a more powerful 3in. gun mounted in open-top turrets on less heavily armoured M4 medium tank chassis. The tank destroyers were much favoured by General McNair, and his views on the limited, exploitation role of tank units were shared on the eve of the landings in Normandy by some of the senior US Army commanders, including General G. Patton, who considered the 75mm gun-armed M4 tanks entirely adequate for the exploitation role.

It was eventually agreed that one third of the M4s should be armed with 76mm instead of 75mm guns, but the first of the 76mm gun tanks were only produced five months before the landings in Normandy and none took part in them. However, once US armoured units came up against German tanks, it became obvious that tanks had to be able to fight other tanks and there was an urgent demand for tanks better armed than the M4s with their 75mm guns. In consequence, M4s with 76mm guns were rushed to Europe and the commander of the US 12th Army Group even asked for tanks with the British 17-pounder. As it happens, none was available, and even when the 12th Army Group reached the Belgian border only 212 of its 1,579 M4s or Shermans were armed with 76mm guns. But by the end of the war the number of Shermans with 76mm gun deployed by the US forces in Germany rose to 2,151, out of 4,123, or to just over one half of the total.

The armour penetration of the 76mm gun was still considerably less than that of the Panther’s 75mm L/70 and of the 17-pounder, but it was at least slightly better than that of the 75mm L/48 gun of the PzKpfw IV. However, during the final months of the war its performance was improved by the introduction of APCR or High-velocity Armour Piercing (HVAP) ammunition, which increased armour penetration at 1,000m by 46 to 53 per cent compared with its standard armour-piercing ammunition.

Views responsible for the late deployment of Shermans with 76mm guns also contributed to the delays in the development of a new and more powerful US tank armed with a 90mm gun. The Ordnance Department began to consider the installation of such a gun on the Sherman in 1942 and a year later the Armored Force requested 1,000 Shermans armed with it. But Ordnance rejected this request in favour of a new tank that was still armed with a 75 or 76mm gun, while Army Ground Forces objected to it on the grounds that a powerful gun would encourage tanks to fight other tanks and thus divert them from the exploitation role!

In consequence, a series of experimental tanks was built with 75 or 76mm guns while the Army Ground Forces continued to favour Shermans armed with 75mm guns. By May 1943 Ordnance recommended that some of the experimental tanks be armed with 90mm guns, and in spite of opposition from the Army Ground Forces 50 tanks armed with them, designated T25E1 and T26E1, were built a year later. Shortly afterwards US armoured units landed in Normandy, and as the shortcomings of their tanks’ 75mm guns became painfully obvious the Armored Force requested that high priority be given to the production of the T26E1, recommending that 500 be built. Army Ground Forces refused to approve this but eventually 250 were ordered. Twenty of the first 40 to be produced were shipped to Europe in January 1945 as M26 Pershing heavy tanks, and they saw some action in the last two months of the war, by the end of which there were a further 270 in Europe. Production of the 41-tonne Pershing continued until the end of 1945, when it had reached a total of 2,428.

The 90mm gun of the Pershing represented a significant advance on the 76mm and even more on the 75mm guns of the Shermans, but in terms of its armour penetration it was still not quite as good as the British 17-pounder or the German 75mm L/70, and it was completely outclassed by the 88mm L/71 of the Tiger II heavy tank that first saw action a year earlier. But only about 100 Tigers could be mustered in December 1944 for the abortive Ardennes offensive, which was the last major effort of the panzer forces in the West. The German High Command could still assemble ten panzer divisions in March 1945 for a counter-offensive against the Soviet forces in Hungary, but the number of tanks at its disposal was generally considerably smaller than those of the opposing armies as a result of the lower scale of their production in Germany.

The differences are clearly shown by the totals of tanks produced in the different countries during the Second World War, which also illustrate the scale on which tanks were used during that conflict. Thus the total number of tanks produced in Germany from 1939 to 1945 amounted to 24,242. During the same period the number of tanks produced in Britain was 30,396. The corresponding figure for the Soviet Union was 76,186. The number of tanks produced in the United States was even higher, being 80,140. The total for the three countries fighting Germany was consequently 186,722 tanks, or almost eight times the number of German tanks.