tration of artillery and the expenditure of an immense amount
of ammunition in a prolonged bombardment from guns of all
calibres was not necessarily effective against such defences as
the Germans had created, and could not insure that the assault-
ing infantry would not be held up by uncut wire and suffer ter-
rible loss from concealed and protected machine-guns. Some
other means of dealing with the numbers of these weapons
which would survive a bombardment even of the heaviest
nature was necessary. As a result of the performance of the
experimental tank in Feb. 1915, G.H.Q. France came to the




FIG. I. Mark I. Tank.

conclusion that such machines might be of some assistance, and
made a demand for 4O. 1 This number was increased by the
British War Office to 100, and supply was undertaken by the
Tank Supply Committee of the Ministry of Munitions, com-
posed of members representing interests directly concerned,
under the chairmanship of Maj. (later Sir) Albert Stern. Orders
were placed on Feb. 12, and production was commenced at once
with the utmost secrecy, under great difficulties of labour and
material owing to the great national effort then in full swing to
produce munitions of all sorts.

During this initial period of manufacture various minor im-
provements of the original design were incorporated. The

1 The first order placed by the French army headquarters, without
awaiting the construction of an experimental machine was for 400
machines, and this was increased to 800.



sample machine, in accordance with the original specification
for a machine-gun destroyer, was armed with two 6-pdr. guns
and three Hotchkiss machine-guns. It was soon realized that,
though effective for its purpose, this armament was deficient in
man-killing fire-power for self-defence in the event of attack by
large numbers of men, and it was decided to arm a certain pro-
portion of tanks with four Vickers machine-guns instead of two
6-pounders. They were known as " females," the gunned
machines being known as " males." The total number under
construction was 150, half males and half females.

The Mark I. Tank (see figs. land 2). Details of this machine are
given in the table on p. 697, but since all the succeeding machines
were a development of it, some further description is given. Its
main characteristic was the all-round track, which gave it its climb-
ing power. This was perpetuated in all the British heavy tanks,
though in the medium, or " Whippet," tanks evolved later, the same
result was obtained by the projection of the tracks which were not
under the machine. The essential value of both forms of construc-
tion was that whatever the fore and aft angle of the machine with
the horizontal a gripping surface was presented. This was achieved
by the high, curved bows, rendered possible by combining the body
and the chassis of the machine and using the sides of the body to
support the tracks. The length was 32 ft. 6 in., over all, and 21 ft.
5 in. without the tail. The width was 13 ft. 9 in.; the height 8 ft.
\ inch. Power was given by a six-cylinder sleeve valve Daimler
engine of 105 H.P. ; transmission consisted of a cone clutch, primary
(two speed) gear box (controlled by the driver) differential secondary
(two speed) gears, actuated by hand levers, one to each track, and
chain drives to the driving sprockets at the rear of each track. There
Were thus four speeds ahead. This system necessitated a " gearsman "
at each side, and was inconvenient and clumsy. The petrol feed was
by gravity. There was no silencer. The armour plate varied from
12 mm. thickness in front to 6 mm. at the top and on the belly, the
only part proof against the German " K " (armour-piercing) bullet
being that 12 mm. thick. The armament of the male tank comprised
two 6-pdr. guns, one on each side and four Hotchkiss machine-guns,
and that of the female tank two Vickers machine-guns on each side
and three Hotchkiss machine-guns. Owing to the pressure of time
certain details in the design and equipment of the machine were
adopted in order to employ material which was in production and
quickly obtainable without waiting for special manufacture. The
power unit, including the engine, with gear-box and differential
was adopted because it was the standard power unit of an existing
tractor and already in production. The 6-pdr. gun was adopted not
because it was considered the best for dealing with machine-guns
(the 2-pdr. automatic gun was originally specified), but because the
Admiralty had a large number of these guns in hand and promised to
supply the requisite number to the army. The 6-pdr. proved, how-
ever, to be a very good weapon for its purpose. The male tank car-
ried 324 rounds of 6-pdr. common shell for the 6-pdr. gun and 6,272
rounds of S. A. A.; the female carried 31,232 rounds of S.A.A. 8
The crew consisted of one officer and seven other ranks. The driver
sat with the officer in the conning tower in front. The total weight
of the male tank when loaded was 28 tons; of the female 27 tons.
The average speed across country was 2 m. per hour, and the radius
of action, nominally 23 m., averaged 12 m. over rough ground. The

* It was originally proposed to carry a certain proportion of case
shot. This was not done, but it was found necessary later to do this.




26'5-Appnx

FIG. 2. Mark I. Tank.



682



TANKS



tank could cross trenches up to II ft. 6 in. in width and could climb
a vertical height up to 5 feet. Six of the first establishments of tanks
were equipped with a wireless set capable of sending and also to
some degree of receiving. 1

Two features of the Mark I. tank were not perpetuated in later
patterns, except in the first gun-carrier machines. One was the tail.
This consisted of a pair of wheels carried by a frame pivoted at
the stern of the machine which for ordinary steering could be ac-
tuated by the driver by means of wire cables. For sharp turns, which
were effected by driving on one track alone, they could be raised off
the surface of the ground by a hydraulic ram at the back of the
tank. The weight of this tail attachment also served to ease the rate
of descent of the tank after crossing a summit, and the extra length
it gave to the whole machine increased the width of the gap which
could be crossed. It was found in actual practice that the complica-
tion and liability to damage the tail was not compensated for by its
advantages, and its use was abandoned after the first actions. In
both the male and female Mark I. machines the 6-pdr. and Vickers
machine-guns were mounted in sponsons to give as far as possible
arcs of fire up to direct ahead and astern. In order to reduce the
width so that the tanks could be carried by rail these sponsons were
removable and could be unshipped for travelling, when they were
carried on small wheeled trollies. The inconvenience of this system
caused it to be abandoned, and in later patterns of tank the sponsons
were so designed that when travelling they could be swung inwards
and housed in the width of the tank, or could be unbolted and slid in.

In March 1916, measures were taken to provide the personnel
to handle the new weapon, and an establishment was framed for
a unit. For secrecy this unit, under the command of Col. Swin-
ton, was raised and formed, as a portion of an existing service,
under the name of the " Heavy Section," subsequently changed
to " Heavy Branch," of the Machine-Gun Corps. This was to
provide the personnel for the 150 tanks then under construction,
without any reserve of machines or man-power. At first the
organization was for three battalions of 50 tanks each, but this
was altered to six companies of 25 tanks each, each company
consisting of four sections of six tanks and one spare tank.
Each section was formed of three male and three female
machines, and was subdivided into three sub-sections of one
male and one female tank. A specially constructed and equipped
mobile field workshop was allotted to each two companies. To
assist in the formation of this unit a nucleus of officers and men
were transferred from the existing Motor Machine-Gun Corps;
officers also being obtained from the cadet battalions, and from
France, and other ranks being enlisted from the motor trade.
Technical personnel of all ranks was supplied by the Mechanical
Transport Branch of the Royal Army Service Corps.

The first headquarters of the Heavy Branch were at Bisley,
where, since there were no machines, the training was of a pre-
liminary nature confined to discipline and gunnery and the use
of the Vickers and Hotchkiss machine-guns. Training in gunnery
was carried out by means of borrowed guns, and entailed the
sending of the men to Salisbury Plain and to the Naval School
of Gunnery at Whale Island. So soon as the tanks began to be
delivered from the contractors, the training in driving, tactics
and shooting from tanks etc., was carried on in a secret area at
Elveden in Suffolk, where a facsimile battlefield had been pre-
pared. The whole of this work was carried out under immense
difficulties as regards time and the need for secrecy, the main
underlying idea of all the preparation being that the role of the
unit was to assist and help the infantry. By the beginning of
Aug. several machines had been delivered, and a certain amount
of training in their use had been carried out.

Meanwhile, the Somme offensive having come to a standstill
in spite of the power of the British artillery then available,
it was decided to use the tanks, or whatever of them were ready,
in the renewal of the attack. Two companies of the heavy
branch, 50 tanks with 10 spare machines, were accordingly con-
centrated in France for this purpose by the end of Aug., and
training was continued preparatory to taking part in operations.
Friday, Sept. 1 5, was to mark the appearance of the tank in war-
fare, when the secret of the new weapon which had been so
carefully kept would be revealed and the weapon itself put to
the test. The whole production of the unit up to this time was
a remarkable feat. Not only had a number of entirely new

1 This scheme was also abandoned and later found necessary.



machines been manufactured sufficient for 60 to take the field
within six months of the order for them having been placed at a
time of great industrial stress, but the secret of their creation,
which was known to thousands, had been so well kept that they
did actually come as a surprise to the enemy.

It was to assist in the further advance of the British right flank,
which had begun so successfully at the opening of the battle 10
weeks before, between the Somme and the Ancre that the tanks were
to be thrown into the fight. The IV. Army was to break through the
enemy's front between the Combles ravine and Martinpuich and
seize Morval, Les Boeufs, Gueudecourt and Flers. On its left the
Reserve V. Army was to attack and gain Martinpuich and Cource-
lette while the French were to press on its right. The cavalry were to '
follow up through the gap which it was hoped would be created and
seize the high ground about Rocquigny-Villers au Flos-Riencourt-
lez-Bapaume. Two companies of the tanks were engaged, the bulk i
with the IV. Army, the rest with the Reserve Army. The general ;
idea of their tactics was that they should start so as to reach their !
objectives five minutes before the infantry. They were to act in
small detachments of two or three machines against the strong i
points in the enemy's defensive system, lanes being left for their
advance in the artillery barrage commencing at zero hour.

The tanks advanced at dawn in a slight mist and came as a com-
plete surprise to the enemy. The operations, of those with the XV. j
Corps of the IV. Army were the most successful; but for various '
reasons the results of the employment of tanks was somewhat dis- I
appointing. Of the 49 machines taking part 32 alone reached their i
starting points, 9 pushed ahead with the infantry and caused con-
siderable loss to the enemy and 9 others, which did not catch up the
infantry, did good work in dispersing of the enemy still holding out ,
at isolated^ spots ; of the balance of 14, 9 broke down and 5 became
" ditched." (Ditching was usually caused either by a tank getting !
into such a position in a deep and wide crater or trench that its
engine power was not sufficient to pull it out, though the tracks ;
gripped, or by weight of the machine being taken by its belly on hard
ground, in which case the tracks revolved without biting.) One i
tank gave remarkable help to the infantry held up in front of Flcrs
by wire and machine-gun fire, when by its action it caused the sur- i
render of 300 Germans and enabled the infantry to move on. Another
destroyed a field gun. On Sept. 25 and 26, 13 machines acted with the
IV. and Reserve Armies. Of these nine were ditched in shell craters,
two reached the village of Thiepval and stuck there. But again, as a
set-off to mishaps, one single tank on the 26th performed a remark- j
able feat which demonstrated the potentialities of the machine.
Within one hour, and at the expense of five British casualties, it
made possible the capture of a strongly held, well wired, trench
(the Gird trench) some 1 ,500 yd. long and strengthened by numerous
strong points, which had held up a whole brigade of infantry since
the previous evening. The Germans suffered heavy loss, and 8
officers and 362, other ranks, surrendered. On Nov. 15, at the battle
of the Ancre after heavy rain, of five tanks that went into action, all
became ditched, two machines doing very valuable work before this
happened. Next day, in an attack on a field work south of Beau-
mont Hamel, one machine out of three employed was put out of ac-
tion by shell fire, and two became ditched. The latter, however,
were able to bring so effective a fire on the strong point that it sur-
rendered and 400 prisoners were collected by the tank crews. But,
whatever their defects, the tanks had passed with ease through
all entanglements and had destroyed many machine-guns, which
weapons, indeed, were practically powerless against it.

The employment of the tanks in Sept. 1916 was contrary to
the views of those who had originated the Arm, who were
responsible for its production and had most studied its action.
They held that the utmost value should be obtained from the
new weapon and that the secret of its existence should not be
given away until a surprise attack could be carried out on a
sufficiently extensive scale to give a chance of achieving a de-
cisive success. In this sense the launcl|jng of the tanks was a
repetition of the error made by the Germans when they released
gas on a small section alone on April 22 1915. Whatever may
have been the urgency at that time of reviving the momentum
of the Somme offensive, which had died away after weeks of
great endeavour and immense sacrifice, and of raising the moral
of the tired troops, and whatever might have been the success
of the new weapon, it is doubtful if the small number actually
employed could have given a result to compensate for the pre-
mature disclosure of the secret, which in potential value was
equal to that of the 42-cm. howitzers and the poison-gas of the
enemy. Again, not only was a small number of tanks used, but
they were employed in driblets in different directions, instead of
together in as great a mass as their available number would
allow. As an experiment this trial of the tanks was, no doubt,



TANKS



683



productive of valuable lessons, but they were obtained at the
cost of discounting the future. It was in the face of a consider-
able amount of scepticism that the machines took their place in
the battle-line, and those who did not believe in them, without
looking beyond the lack of any startling success achieved on a
great scale, were not slow in their condemnation. But one
thing the tank had definitely proved: The machine-gun and the
barbed wire entanglement no longer ruled the infantry battle. In
the tank, still experimental and imperfect as it was, they had
found their master. At last attacking infantry could meet the
defence on more than equal terms.

On the part of the tanks themselves there was mechanical failure
of machines, which were the first of their kind, tried under more
difficult conditions than those which they had been designed to meet.
The specification in which they had been constructed had been based
on the requirements of the spring of 1915 at a time when the pro-
longed shelling, such as in 1916 and later rendered whole tracts al-
most impassable even by infantry, was undreamt of. At the Somme
the tanks had to cross a dry " crumped " area, and at the Ancre a
combination of " crumped " area and morass. There were, in addi-
tion, breakdowns due to faulty design and to the unexpected wear
and tear of certain parts of the machinery, which only experience
could have shown should have been of special material and which
were, moreover, practically worn out by practice and demonstrations
before the action. Failures were also due to the lack of time for the
complete training and practice of the crews ; to insufficient prepara-
tion in the way of reconnaissance and supply services, which for
such complicated machines necessitated very complete organization.
On the other hand, many of the commanders at whose disposal
tanks were placed had no idea of their capabilities, weaknesses or
limitations. Nevertheless, the outstanding brilliant exploits of single
machines which did not break down, the great saving of life for which
the tanks as a whole were responsible, and the demoralizing effect
of their appearance on the enemy and the great encouragement
afforded to the British infantry, clearly showed that the machines
were sound in principle, only needing improvement in detail, and
that those handling them required more experience in this new
form of warfare.

As a result of its trial it was decided that the new Ann should
not only continue, but should be expanded to a force of 1,000
tanks. After the fighting on the Ancre the unit did not again
go into action till April of the following year, which allowed of a
period for expansion, reorganization and training, and the incor-
poration of improvements to machines. 1 The expansion of the
Heavy Branch of the Machine-Gun Corps was to be on the follow-
ing lines: there was to be a fighting headquarters in France and an
administrative headquarters in England. 2 The six original com-
panies were to be expanded to four battalions in France and the
two companies at home to five battalions, or nine battalions in
all. The unit therefore reverted to its original battalion organiza-
tion. It was eventually to comprise three brigades, each of three
battalions, each battalion of three companies and a workshop,
each company consisting of four fighting sections of five tanks
each and a headquarters section of four tanks, or 72 tanks to a
battalion. The number of tanks per section was shortly after-
wards reduced to four, making 20 per company or 60 per battal-
ion. These brigades were formed at the beginning of 1917, and
the organization continued unchanged until June, but though the
personnel was being raised and trained as far as possible on this
extended scale, the supply of tanks did not keep pace, so much
so that on April i, when it was decided that the tanks should
take part in the battle of Arras, only 60 Mark I. and Mark II.
machines could be counted on for action.

In Feb. 1916, when the original Mark I. tank was still in its
experimental stage, designs had been got out for a Mark II.
machine and a Mark III. machine, both of which were slightly
improved Mark I. tanks. After the Somme battle certain mod-
ifications which were found necessary were incorporated, and in

1 The most important improvement was the improvised " un-
ditching beam " which was introduced to assist a tank to extricate
itself when stuck. In the later types a specially designed contrivance
was part of the equipment of the machine.

2 Lt.-Col. (afterwards Maj.-Gen.) H. J. Elles, R.E., was appointed
colonel commanding the unit in France, Bng.-Gen. F. Gore Anley
being appointed administrative commander of the tank training
centre in England on Nov. 9. In May 1917 Brig.-Gen. Anley was
succeeded by Maj.-Gen. Sir J. E. Capper. The training centre was
moved to Bovington in Dorset.



Oct., in order to keep up continuity of supply, orders were given
for 50 machines of each type (making 250 tanks ordered in all)
pending the decision on the design for the 1,000 asked for by the
army, in Sept. The chief improvements consisted of the aban-
donment of the tail, as already mentioned; the addition of
" spuds," or fins, to the track plates, to give a better grip; and
the provision of improved rollers. The Mark III. machines
were also intended to have thicker armour to resist the " K "
bullet. In regard to the 1,000 new tanks preliminary orders
to collect material were given to the manufacturers in Sept. At
the end of the month when it was learnt that these machines
could not be ready by March i (partly owing to the manufac-
ture of spare parts necessary for the existing tanks), the demand
was cancelled, but was renewed two weeks later. This is men-




FIG. 3. Mark IV. Tank.

tioned to show that military opinion as to the advisability of pro-
ceeding with the new weapon was not unanimously favourable.
The 1,000 machines were to be of the type known as the Mark
IV., which was a much improved Mark I. tank. It was then
anticipated that this number could be delivered by the end of
June. Owing to various difficulties this estimate was not ful-
filled, but sufficient machines reached France in time to equip
two battalions for the attack on Messines in May; and it was
the standard British tank during 1917 and 1918.




FIG. 4. Mark IV. Tank.

The Mark IV. tank (see figs. 3 and 4). The first advance from the
original weapon was the same in size and general design, with certain
improvements. There was no tail. The track rollers were better.
The sponsons could be housed within the tank for rail transport.
The Lewis machine-gun was installed instead of the Hotchkiss
machine-gun (this was not an improvement and was reversed
later). A better radiator was fitted and also a silencer. The width of
the driver's cab was increased to allow of wider track shoes. The
petrol tanks were placed outside the tank at the stern. A short 6-
pdr. gun was adopted. Detachable spuds were fitted to the tracks,
and unditching gear provided. The entrances and exits were more
accessible.

During the Verdun offensive early in 1916 it had occurred to
those responsible for the production of the fighting tank that
one reason why the Germans had been unable to keep up the
initial pressure of their attack was their inability to bring up
their artillery and ammunition over the shelled and entrenched
area so as to keep pace with their advance; and that if they had
been in possession of guns mounted on self-propelled carriages,
or carriers pn caterpillar tracks, it would have assisted them
greatly. A design was therefore prepared in July of a " gun-
carrying " tank (see fig. 5), to carry a 6o-pdr. gun or 6-in. how-
itzer which could be fired from the tank if necessary or con-
veyed by the tank and fired from the ground. Of these machines,



TANKS




FIG. 5. Gun-carrying Tank.



48 were made, delivery in France commencing in July 1917.
They appear to have been employed as much' for the conveyance
of ammunition and stores as for the purpose for which they were
designed. In Dec. 1916, also, the design of the " Whippet,"
the Medium Mark A. tank, (see figs. 6 and 7) was commenced.
Of this type 200 machines were produced, delivery in France
starting at the end of 1917. It was evolved to be complementary
to the heavy tanks and to meet the demand for a speedier,
handier machine which could be produced in large numbers.
Its main points were its increased speed, nominally 8 m. per



hour on the flat, and an average across-country of 5 m. per
hour, and its lightness, one-half that of the Mark I. machine.
It was also easier to transport by rail. Each track was driven
independently by a four-cylinder 45 H.P. Tylor engine. Trans-
mission was by cone clutches to gear boxes of the constant mesh
type, giving four speeds forward and one reverse. The design
was largely due to Sir W. Tritton. Thus, at the end of 1916, in
addition to the first type of heavy tank in the field, measures
were in hand to supply a much improved pattern of that ma-
chine (Mark IV.) and also a lighter type the " Whippet."