The coming into being of tanks, the first and the most important of the tracked armoured vehicles, is commonly described as a unique event inspired by the ideas of one man, who is usually identified as Lieutenant Colonel E. D. Swinton, an engineer officer of the British Army. Such accounts bear little relation to what actually happened, although Swinton himself professed that it was true. He made this very clear in his memoirs, written several years after the first British tank was built, in which he described himself as its ‘originator’ and included a chart showing how the ideas leading to it, or ‘seeds’ as he called them, sprang from him.1
In fact, tanks were the outcome of the ideas and activities of several men, and any contribution that Swinton might have made to their evolution was small and indirect. Moreover, the construction of the first tanks was preceded and based on developments that took place during the previous half century.
The most important of these developments was that of the tracked running gear. Its use in steam traction engines was proposed as early as 1858 or 1859 in patents taken out in the United States. By 1867 at least one steam-powered tractor was actually built there with its two rear wheels as well as the front steering wheel replaced by short tracks.2 However, the development of tracked tractors did not begin in earnest until 1904, when B. Holt replaced the rear wheels of one of the steam traction engines produced in California by his company by tracks. Because it retained its front steering wheels, Holt’s original tracked tractor was what would later be called a ‘half-track’, as were other tractors produced by him until 1912.
In the meantime, in 1905, a further advance took place when Richard Hornsby built the first fully tracked tractor in Britain. It was shown in 1905 and 1906 to the Mechanical Transport Committee of the War Office, which was sufficiently impressed to conduct official trials in 1907. A year later a Hornsby tractor even took part in a revue at Aldershot before King Edward VII, and in 1909 the War Office ordered a somewhat smaller tractor designed by Hornsby to military requirements, which has been preserved to this day at the Tank Museum at Bovington in Dorset.3 This tractor was tested at Aldershot, but after about 1911 the British Army lost further interest in tracked tractors, although the tractor built to its requirements was exhibited in London at the beginning of 1914.
Seeing no opportunities for further business in Britain, Hornsby sold his tractor patents to the Holt Manufacturing Company in California, which started to produce tracked tractors on a regular basis in 1908 and subsequently exported a few of them to Europe in addition to those sold to American farmers.
While it lasted, the interest of the War Office in the Hornsby tractors was prompted by their possible use for hauling guns. But in 1908 a member of the Mechanical Transport Committee, Major W. E. Donohue, suggested that instead of being towed a gun be mounted on a tractor and provided with some form of protection. This amounted to a proposal for a tracked self-propelled gun, but Donohue’s suggestion was not pursued.4
Remarkable as it was, Donohue’s proposal was not the first of its kind. Five years earlier a French artillery officer, Captain Levavasseur, put forward a scheme for a self-propelled 75mm gun on an armoured tracked chassis. The scheme was considered by the French Artillery Technical Committee, but the latter came to the conclusion that animal traction was preferable for guns and finally rejected the scheme in 1908.5
Other proposals made before the outbreak of the First World War in 1914 did not fare any better. The most interesting of them was made by Captain G. Burstyn, an Austrian officer who apparently saw a Holt tractor in 1911 and this inspired him to design a ‘motorgeschutz’, a tracked armoured vehicle with a turret mounting a cannon and with rollers on extended arms to assist the crossing of trenches. Burstyn offered his design to the Austro-Hungarian War Ministry but the latter rejected it.6
In Russia, V. Mendeleev, the son of the famous scientist, is reputed to have started working in 1911 on the design of a tracked armoured vehicle armed with a 120mm naval gun, but his work did not advance beyond drawings.7 Another design originated in Australia, where L. E. de Mole, a civil engineer, produced drawings of a tracked armoured vehicle. He submitted his design to the British War Office in 1912 but the latter showed no interest in it.8
The idea of a tracked armoured vehicle had clearly emerged in more than one country before the outbreak of the First World War in 1914, but failed to arouse the interest of the military authorities in any of them. To be fair, few tracked tractors from which tracked armoured vehicles were to be derived had been built and their characteristics or even existence were not widely known. What is more, armies did not fully appreciate the growing importance of heavy weapons and the extent to which their effectiveness was constrained by their limited mobility, based as it generally was on horse traction. In consequence they were less receptive than they might have been to the idea of tracked armoured vehicles, which offered to make heavy weapons more mobile and to a far greater extent than armoured cars had done already.
Once the war began, heavy weapons and in particular machine guns and field guns, until then regarded as only an adjunct to the rifle armed infantry that constituted the bulk of the armies, proved dominant. At the same time horse traction failed to provide them with the mobility necessary for offensive action. All this favoured static defence in which heavy weapons needed little, if any, mobility and in which they could be used to full effect. The effectiveness of static defence was further increased by the use of entrenchments and barbed wire.
The outcome was deadlock, particularly on the Western Front in France, with neither side being able to break through the other’s defences by the traditional massed infantry attacks. The immediate problem became that of finding a way that would enable the infantry to continue to attack in the face of machine guns and barbed wire. In response to this came proposals for armoured assault vehicles that would pave the way for the infantry by attacking enemy machine guns and by crushing the barbed wire.
The first in Britain to think of using tracked vehicles to solve the problems created by the onset of trench warfare appears to have been Swinton, and to this extent he deserves credit for originality. But what his thoughts were is not very clear and, whatever they were, any influence they might have had on the development of the first British tanks was indirect and limited.
According to his own account written several years after the event, the idea of a tracked armoured vehicle came to Swinton suddenly in October 1914 when he recalled the description of a Holt tractor sent to him four months earlier. This occurred while he was travelling to England from France where he was acting as the sole official correspondent with the British Expeditionary Force (BEF). On arrival in London he discussed his views with Lieutenant Colonel M. Hankey, the secretary of the influential Committee of Imperial Defence.9 There is no record of what was actually discussed but it seems to have included the possibility of converting tracked tractors into some kind of assault vehicle, and such a vehicle is alluded to in a letter written a month later by Swinton to Hankey.10
A clearer outcome of the contacts is a device described in a memorandum written by Hankey in December 1914 and this, according to Swinton, embodied what he had put to Hankey.11 But what Hankey proposed was only a large heavy roller pushed by a single, engine-driven track and provided with an armoured cab for a driver and a machine gun. As Hankey explained, the object of this device would be ‘to roll down the barbed wire by sheer weight’.12 What he was proposing was therefore a specialized wire crusher and not an armoured fighting vehicle.
A copy of Hankey’s memorandum was sent, among others, to the First Lord of the Admiralty, Winston Churchill, who after receiving it wrote to the Prime Minister, H. H. Asquith, saying that he agreed with Hankey about the use of ‘special mechanical devices for taking trenches’. However, he saw them in a different form from that envisaged by Hankey, namely that of steam tractors that were armoured and fitted with machine guns as well as tracks. The prime minister then took the matter up with the Secretary of State for War, Lord Kitchener, and this led to the formation of a War Office committee that was to consider the possible adaptation of tracked tractors to the role of assault vehicles. As part of its activities the Committee arranged tests of a Holt tractor over an obstacle course made up of trenches and barbed wire entanglements in February 1915. The tractor, handicapped by being made to pull a heavy trailer, failed to cross some of the trenches, which led the Committee to conclude that what was proposed was impracticable and to abandon further consideration of it.
Prior to the test the Committee received a memorandum written by Captain T. G. Tulloch, an artillery officer recommended by Swinton, which more clearly than hitherto envisaged armoured vehicles armed with machine guns that could move across country and over barbed wire to attack enemy trenches. The vehicles were to be based on Holt tractors but coupled in pairs to form articulated vehicles – which Tulloch appears to have been the first to propose in 1911 in sketches of a large articulated armoured vehicle.13 However, the ideas contained in Tulloch’s memorandum were not taken up. But its title, which was ‘Land Ship’, became the designation of the first British tracked armoured vehicles and also reflected the influence of naval ideas on contemporary thinking.
In the meantime Churchill had Hankey’s idea of using large rollers to crush barbed wire put to test by the Royal Naval Air Service, which he controlled. The test was carried out using steam rollers, which proved incapable of climbing the slightest slope, and the whole idea was abandoned. As a result Churchill turned his attention to other ideas. These were put to him by officers of the RNAS who were seeking vehicles more capable than their armoured cars, which were becoming ineffective with the onset of trench warfare.
The earliest of these ideas came from Flight Lieutenant T. G. Hetherington, who in November 1914 proposed to Commodore M. Sueter, the director of the Admiralty Air Department who was in charge of the RNAS, the construction of a giant three-wheel armoured vehicle armed with 12in. (309.8mm) naval guns. Sueter realized that this was not a practical proposition but was prepared to pursue the idea of a scaled down version, which he submitted to Churchill in January 1915.
The vehicle that was now being proposed was still very large, the front two of its three wheels having a diameter of 40 feet (12m). Moreover, it was to have three turrets, each with two 4in. guns, while its weight was estimated to be 300 tonnes. In its proposal it was described as ‘a cross-country armoured car of high offensive power’, which showed that the aspirations of the RNAS officers were sound. But their judgement of what was practicable was not. It is not surprising therefore that when Churchill referred the proposal to the Admiralty it was turned down by one of its experts, Admiral Sir Percy Scott.14
However, Hetherington’s idea of a large three-wheeled vehicle was not entirely irrational. Such a vehicle with two driving wheels with a diameter of 10m, or almost as large as those of Hetherington’s proposed vehicle, was actually being built in Russia in 1915. Its construction was promoted by M. Lebdenko, the head of the experimental laboratory of the Russian War Ministry, and was allegedly supported by the tsar. But it was abandoned as impracticable without being completed.15
Although it was not practicable and in spite of its rejection by Admiral Scott, Hetherington did not abandon his idea of a large-wheeled vehicle and got a chance to put it directly to Churchill at a dinner on 14 February 1915. This time Churchill showed greater interest in it and referred it to the Director of Naval Construction, E. H. T. d’Eyncourt. The latter concluded that the proposed vehicle would weigh 1,000 tonnes and was not therefore practicable. But he did not reject the concept of a large-wheeled armoured vehicle, and suggested that Hetherington’s vehicle be replaced by a smaller one.16 Churchill agreed and ordered the formation of a committee chaired by d’Eyncourt to pursue the ideas put to him. This came into being on 20 February 1915 under the title of ‘Landships Committee’.17
While these events were taking place, Sueter was pursuing his interest in the use of tracks, originally for mobile armoured shields to be pushed by infantrymen in front of them like wheelbarrows. Then, having apparently turned against big wheel vehicles, he began in February 1915 to consider the design of a tracked 25-tonne armoured vehicle armed with a turret-mounted 12-pounder gun. The design was produced in collaboration with B. J. Diplock and incorporated the latter’s Pedrails, which were the only tracks made at the time in Britain.18
By the time the Sueter-Diplock design was drawn, the Landships Committee had come into existence and shortly afterwards was informed about it. In consequence, the committee had two different designs to consider, one being that of a wheeled vehicle derived from Hetherington’s proposals but with wheels having a diameter of 16 feet (4.9m) instead of 40 feet, and the other being a tracked vehicle with Pedrails. The committee reported on them to Churchill, who, relying on the advice of d’Eyncourt that the two proposed designs were viable, took it upon himself to authorize on 26 March 1915 the construction of six-wheeled and 12-tracked vehicles.19
The order for the wheeled vehicles went to the William Foster Company in Lincoln. Their work suffered a series of false starts and the order issued to them was cancelled in June 1915 without a vehicle being built.20 Work on the tracked vehicles produced more tangible results. It was directed initially by Colonel R. E. Crompton, an experienced engineer who pioneered the use of steam traction engines in India in the 1870s and was then involved with their use in the South African War, who was appointed a consultant to the Landships Committee.
Working with another engineer, L. A. Legros, Crompton produced a design similar to that conceived by Sueter and Diplock. But, instead of having a turret with a 12-pounder gun, Crompton’s vehicle was to be an armoured carrier capable of carrying as many as 50 or even 70 men, which came to be considered at that stage the tactical purpose of what were beginning to be called ‘landships’. Otherwise it was to have the same peculiar configuration as that adopted by Sueter and Diplock, which consisted of a long rigid chassis resting on two wide Pedrail tracks arranged in tandem, each driven by a separate engine. The vehicle was to be steered by turning the track assemblies in relation to the chassis, which meant that its turning circle was impractically large. Crompton realized this even before the vehicle was built, which it eventually was in a reduced form by the Stothert and Pitt company in Bath, and proved unsatisfactory.21
The failings of the original design became clear to Crompton after a visit to France on 21 April 1915 when he concluded that the vehicle he was designing would not be able to negotiate the bends in roads and village streets. In consequence, he decided to abandon the original design in favour of an articulated vehicle that would have a smaller turning circle and therefore be more manoeuvrable. The idea of an articulated tracked vehicle was not entirely new, as one had already been proposed by Tulloch, and it was actually implemented by Diplock who exhibited an articulated tracked truck in London in 1913.22 But no other articulated tracked vehicle had been built and it was going to take more than 40 years before one was successfully developed.23
Each section of Crompton’s second articulated vehicle was still to have only one Pedrail track. But by May 1915 he recognized the shortcomings of the complicated and heavy Pedrail tracks, which had not been used successfully in any vehicle, and recommended the purchase of American tractors with lighter and proven tracks. As all the available Holt tractors had been earmarked already for the British Army for gun towing, the Landships Committee ordered two similar ‘Creeping Grip’ tractors from the Bullock Tractor Company of Chicago.24 Following this Crompton started working on his third design, which was similar to the second design but with each section of the articulated carrier having two Bullock tracks.
In addition to recommending the purchase of the two Bullock tractors, Crompton also arranged the purchase of a lighter American tractor produced by the Killen Strait company. This tractor was almost unique in having a short track instead of the steering wheel that almost all contemporary tractors still had. This was not of any mechanical merit, but the two main tracks of the Killen Strait tractor were very effective for their day. As a result the Killen Strait tractor was used very successfully in June 1915 to demonstrate for the first time the ability of tracked vehicles to negotiate barbed wire entanglements and other obstacles in front of the Minister of Munitions, Lloyd George, Churchill and others whose support was essential for the development of landships.25 The Killen Strait tractor was subsequently fitted with the hull of an RNAS Delaunay-Belleville armoured car, and this combination became in July 1915 the first, albeit only experimental, tracked armoured vehicle.
When the two Bullock tractors arrived in England they were coupled to test Crompton’s concept of an articulated vehicle. Tests of the coupled tractors were carried out in July 1915 and showed that an articulated vehicle would be difficult to develop.26 In consequence, the Landships Committee decided to abandon further work on it and at the same time terminated the appointment of Crompton as its consulting engineer. Instead of the large articulated troop carrier that was aimed at until then, the Landships Committee decided to develop a smaller vehicle with a rigid, one-piece hull and a turret-mounted cannon. An order for such a vehicle was issued on 29 July to the William Foster Company, which had worked previously on the wheeled landships. It proceeded with remarkable speed and had the vehicle built and running by 6 September 1915.
The vehicle built by Fosters, which came to be known after its managing director W. A. Tritton, corresponded to one half of Crompton’s third design, that is the articulated carrier with lengthened Bullock tracks. In the interest of speed Tritton used the engine, gearbox and differential of a heavy wheeled artillery tractor that his company was producing. He also copied the method of steering used by the Bullock as well as other contemporary tractors, which involved a pair of steering wheels but mounted behind instead of in front of the vehicle. However, the vehicle could also be steered by braking one of the output shafts of the differential – a method first used by Hornsby 10 years earlier.
Tritton’s vehicle was only an experimental machine with a box hull of boiler plate and a fixed dummy turret. But it provided, at last, a sound basis for the mechanical development of landships. When it began to be tested its Bullock tracks proved unsatisfactory and it was rebuilt with a longer and stronger type of track. The new type of track was designed by Tritton, whose prior experience of track development was confined to a single half-track tractor called ‘Centipede’ built by his company in 1913. Nevertheless, the track proved successful and its performance was crucial to the further development of landships.
When it was modified and fitted with the new type of track, Tritton’s vehicle became known as ‘Little Willie’ and in that form has been preserved at the Tank Museum at Bovington. It was successfully demonstrated in December 1915, and because of its longer tracks could cross wider trenches than in its original version, whose 4ft trench crossing capabilities proved inadequate when it was first tested in September 1915.27
In the meantime the Landships Committee had established contact with the War Office and on 26 August 1915 received from it a set of requirements to be met by landships. These requirements were derived from three memoranda submitted by Swinton to the General Headquarters of the British Forces in France between 1 and 15 June and sent on to the War Office on 22 June.28 In the memoranda Swinton set out his ideas concerning ‘machine gun destroyers built on the caterpillar principle that would lead infantry assaults on enemy trenches’ and suggested, among others, that they should be able to cross trenches 5ft wide. This was subsequently incorporated in the War Office requirements which the Landships Committee received on 26 August.29
However, on 29 June Swinton wrote again to the General Headquarters tentatively suggesting that the trench crossing requirement be increased from 5 to 8ft.30 This suggestion obviously arrived too late to be included in what was sent to the War Office seven days earlier and was not part of the requirements received by the Landships Committee.
Whether Swinton’s 8ft trench crossing requirement was ever made known to the designers of the landships is an open question. If it was it could only have been after they received the War Office requirement for a 5ft trench crossing capability on 26 August, by which time they were bound to have started thinking of improving on the design of the first landship since a full size wooden mock-up of the second landship was already built by 19 September, when Swinton first saw it.31
Nevertheless, after seeing the mock-up of the second landship Swinton declared that it was ‘the actual embodiment of my ideas and the fulfilment of my specification’. In fact, the only feature of the second landship that might have provided some justification for such a claim was its ability to cross wide trenches. But, although this was similar to what Swinton had belatedly suggested to the General Headquarters, the second landship was designed without reference to him and the general ideas it represented were already embodied in the first landship, which, as Swinton himself acknowledged, was not built to his specification.32
As to Swinton’s more general claim, quoted earlier, to have been the ‘originator’ of the tank, Churchill, who was closely connected with its development, rightly observed ‘there was never a person about whom it could be said “this man invented the tank”’ and added, with some justification, that the tank was a child of the RNAS.33 But when Sueter quoted this in a book in which he tried to put on record the achievements of the RNAS, Swinton wrote on the margin of his copy of the book now in the possession of the present writer ‘It was not. EDS’, refusing to let go of the claim that he assiduously fostered for many years.34, 35
The second landship was actually designed by Tritton and Lieutenant W. G. Wilson, an engineer seconded from RNAS to Fosters who was to become well known for his designs of epicyclic transmissions. Its salient feature was the novel layout of the tracks, devised by Wilson.36 This provided the tracks with a high, upturned nose and a high return run so that they went around the body of the vehicle instead of being below it. The upturned nose was inspired by the high parapets of enemy trenches, and together with the long length of the tracks provided the second landship an exceptional trench crossing capability that greatly exceeded the War Office requirements. It also gave the second landship its characteristic rhomboidal silhouette.
The configuration of the vehicle did not lend itself to the installation of a turret and the original idea of arming it with a turret-mounted gun was therefore discarded. Instead the vehicle was armed with two guns mounted in sponsons projecting out of its sides – a contemporary mode of mounting the secondary armament of the larger warships. As the Army was short of suitable guns, the vehicle was armed with 57mm 6-pounder naval guns that the Director of Naval Ordnance promised to supply in sufficient quantity. In addition to the two 57mm guns the vehicle was also armed with three machine guns.
In other respects the second landship followed Tritton’s machine. In particular, it was powered by the same 105 horsepower Daimler engine and had the same Foster wheeled tractor transmission as well as a pair of steering wheels at the rear. It also ran on the same type of unsprung plate track as its forerunner. The thickness of its armour, which was simulated by soft steel plates, varied from 6 to 12mm and fully laden it weighed 28 tonnes.
Development of the vehicle proceeded with remarkable speed: a mock-up of it was built in September 1915, its design was completed in October and the vehicle itself was completed by 26 January 1916. It was at first referred to as the ‘Wilson machine’ but later as ‘Big Willie’ and ‘Her Majesty’s Land Ship Centipede’, and eventually as ‘Mother’ as it became the progenitor of the British heavy tanks of the First World War.
By February 1916 the Army finally decided that it wanted what by then had began to be called tanks rather than landships for reasons of secrecy. However, its decision was only taken after trials of Mother at Hatfield in January and February 1916 during which it successfully negotiated all obstacles. This included trenches 9 feet wide, while the official requirement was still that it should cross trenches 5ft wide and not what Swinton ultimately suggested.37 Its performance convinced most of the military and civilian officials who attended the trials of its potential value, although Lord Kitchener dubbed the tank ‘a pretty mechanical toy’.
However, the decision to go ahead with the production of tanks similar to Mother was, curiously, left to the General Headquarters of the British Forces in France. Its representatives attended the trials at Hatfield and recommended the acquisition of tanks, although only 40 were subsequently asked for. On hearing of this ridiculously small number Swinton, who had returned to an influential government post in England in August, persuaded the War Office to raise the number of tanks to be produced to 100 and on 12 February 1916 the Ministry of Munitions authorized their production.
Thus the evolution of tanks in Britain reached the end of its experimental phase and entered that of production and use in the field.
Concurrent with the development of the tank in Britain and with no knowledge of it, the tank was also developed in France. That this should have happened is not entirely surprising since the two countries faced the same military problems and possessed or had access to similar technological resources. Nevertheless, it is remarkable that some of the steps in the development were taken in the two countries at almost the same time. This applied, among others, to the decisions to produce tanks, which were taken in Britain and in France within days of each other.
The French decision was bolder as it called for the production of 400 tanks. However, no prototype of the French tank had yet been built and the French production order took several more months to implement than the British.
As in Britain, the development of the tank in France was preceded by proposals for the use of special devices to overcome the problems created in 1914 by the onset of trench warfare and in particular that of attacking trenches protected by barbed wire. The first of them appears to have been a proposal made in November 1914 to convert a road roller into an armoured vehicle that would crush barbed wire. The proposal was actually implemented and the resulting Flot-Laffly roller was tested, but only to be rejected as being impractical, like the rollers considered in Britain by Hankey and by Churchill. Another proposed barbed wire crusher consisted of the Boirault apparatus – a bizarre device consisting of six large, 4 metre by 3 metre linked skeleton panels engine-driven by chains to move forward as if it were a hexagonal wheel or six-link track. On the advice of the French commander-in-chief a ministerial commission rejected this device outright, but it took a demonstration carried out 5 months later to make another commission, representing the technical services of the Army, reach the inevitable negative conclusion.
A different method of breaching barbed wire entanglements, which amounted to the use of a large wire cutter mounted on a wheeled agricultural tractor, was proposed in November 1914 by J. L. Breton, an influential deputy to the National Assembly. This method was tested in July 1915 and, in spite of the inadequate performance of the tractor over broken ground, the Ministry of War ordered the construction of ten Breton-Prétot wire cutters.38
Much more practical results followed the development of an armoured car on which the Schneider armament company of Le Creusot embarked towards the end of 1914. By mid-January 1915 its design had been drawn, and later that month a Schneider engineer involved with the project, E. Brillié, travelled to Aldershot in England to see the 75hp Holt tractors that had just been acquired by the British artillery for hauling heavy guns. During his visit Brillié also learnt of the existence of a new, smaller 45hp ‘Baby Holt’ which, unlike the heavier 75hp model and other contemporary tractors, dispensed with the steering wheels. As a result of the visit Schneider ordered one tractor of each type, receiving both from the United States by the beginning of May 1915.
Trials of the tractors at Le Creusot showed that the Baby Holt was the more manoeuvrable of the two and that it could form the basis of a tracked armoured car superior to the wheeled cars. Its ability to move over rough ground and to negotiate obstacles was demonstrated in the presence of the President of the Republic, and in July Schneider started to design an ‘armed and armoured tractor’ based on a slightly lengthened Baby Holt chassis. However, work on it was stopped in September because of the intervention of Breton and the technical branch of the Ministry of War, which made Schneider redirect its work to the design of a vehicle that would carry modified Breton-Prétot barbed wire cutters more effectively than they were originally. Trials of a cutter mounted on a Baby Holt tractor carried out in December 1915 raised doubts about its effectiveness but, even before they took place, an order was given to Schneider by the minister of war for ten tracked armoured vehicles to carry the cutters.
Further tests were carried out in January 1916 and showed that the Breton-Prétot cutters were not needed for opening passages through barbed wire because a tracked vehicle could crush it down by its weight.39 Nevertheless, more time and effort might have been wasted on the wire cutters because of the political influence of Breton, who continued to promote them. However, this was prevented by the arrival on the scene of Colonel J. E. Estienne, who managed to bring the work at Schneider back on to a more fruitful path and provided a new drive for the development of tracked armoured vehicles.
Estienne was a technologically minded artillery officer who, among other things, was a pioneer in the use of aircraft for directing artillery fire. When the war broke out, he was given the command of an artillery regiment and was allowed by the minister of war to take with him a section of two aircraft that he had organized and that made his regiment the only one during the first two months of the war to have its own spotter aircraft.40
Once the war began, Estienne turned his attention to the mobility of the artillery and after only a few days of operations is reputed to have told the officers of his regiment that victory would belong to whoever of the belligerents was the first to mount a 75mm gun on a vehicle capable of moving over all types of terrain – the 75mm field gun being the basis of the French artillery and probably the most successful gun of its time.41 This was a remarkable recognition of the need to make heavy weapons more mobile, but Estienne did not see how this could be done until October 1915 when on the Somme front he visited a neighbouring British division and saw Holt tractors being used for towing guns.42 This led him to the idea of an armed and armoured assault vehicle and to write, on 1 December 1915, to the French Commander-in-Chief, General Joffre, asking for an audience to present his ideas.
On the strength of his reputation for novel ideas, Estienne was summoned to the General Headquarters on 12 December 1915, when he explained in some detail the vehicle that he had in mind to General Janin, who was in charge of army materiel. The vehicle that he envisaged was to be tracked and weigh 14 tonnes, with armour 15 to 20mm thick and an armament of a 37mm cannon and two machine guns, and it was to be manned by a crew of four.43 It was called a ‘cuirassé terrestre’ or ‘land ironclad’, which was virtually the same as the ‘landships’ designation applied to the first British tanks and reflected, once again, the contemporary influence of naval ideas.
Eight days later, with the approval of General Joffre, Estienne began to investigate who could produce the vehicle he proposed. He first called on Louis Renault, the head of the car company that still bears his name, but the latter already had as much war work as he could deal with. Estienne then met Brillié, who was working on the vehicles ordered to carry the Breton-Prétot wire cutters instead of the tracked armoured car he had originally conceived, and found much similarity between their ideas. In consequence Brillié agreed to study Estienne’s proposal. Having secured industrial collaboration, Estienne wrote again, on 28 December 1915, to General Joffre suggesting that Schneider be given an order for 300 to 400 vehicles. Joffre responded favourably, recommending further study and, when Estienne reported to him on the situation, decided on 31 January 1916 to order 400 cuirassés terrestres, which by then were to be armed with a short-barrelled 75mm gun instead of the 37mm cannon originally proposed.44
The actual issue of the order to Schneider was the responsibility of the technical branch of the army, which resented Estienne by-passing its bureaucratic procedures and demanded further trials. These were successfully carried out, but had it been left to the technical branch there would have been further delays. However, Estienne appealed once again to Joffre and as a result on 25 February 1916 the minister of war approved the issue of an order to Schneider for 400 vehicles that, for the sake of secrecy, were called ‘tractors’.45
The first Schneider tank to be built as a result of the order was completed by the beginning of September 1916. Like that of the British ‘Little Willie’, its simple box hull was still of mild steel. The supply of armour plate caused delays in the production of more tanks, but even before the first Schneider tank was built the technical branch of the army, smarting from being left out of its development, decided to promote the construction of another tank. In pursuing this objective it found a willing partner in the company of Forges et Aciéries de la Marine et d’Homecourt, commonly known as Saint Chamond, which was a bitter commercial rival of the Schneider company, and it won the approval of Breton, who had become the head of an inventions committee in the Ministry of Armament. In consequence Saint Chamond proposed a tank designed by Colonel Rimailho that was quickly accepted by the minister of war, who issued an order to Saint Chamond on 8 April 1916 for 400 vehicles. All this was done without reference to Estienne or even Joffre, who was not informed by the minister of the order until 27 April 1916.46
Like Schneider’s, the Saint Chamond tank was based on the Baby Holt tractor but had a lengthened track. It had a more powerful armament in the form of a full-size 75mm gun and thicker armour, which contributed to it weighing 23 tonnes instead of the 13 tonnes of the Schneider. It also incorporated a novelty in the form of an electric transmission, but its design was seriously flawed because the front of its hull overhung the tracks to such an extent that it dug in whenever it tried to cross a trench of any width.
The prototype of the Saint Chamond tank was completed at the beginning of September 1916 at approximately the same time as the Schneider tank. In consequence, the development of tanks in France started not with a single model, as in Britain, but with two different vehicles built almost simultaneously.