Produced by Brian Coe, David E. Brown, and the Online
Distributed Proofreading Team at http://www.pgdp.net (This
file was produced from images generously made available
by The Internet Archive)









The Daily Telegraph

WAR BOOKS




MOTOR TRANSPORTS IN WAR




  The Daily Telegraph

  WAR BOOKS

  Cloth
  1/- net
  each

  Post
  free
  1/3
  each

  HOW THE WAR BEGAN
  By W. L. COURTNEY, LL.D., and J. M. KENNEDY

  THE FLEETS AT WAR
  By ARCHIBALD HURD

  THE CAMPAIGN OF SEDAN
  By GEORGE HOOPER

  THE CAMPAIGN ROUND LIEGE
  By J. M. KENNEDY

  IN THE FIRING LINE
  Battle Stories told by British Soldiers at the Front.
  By A. ST. JOHN ADCOCK

  GREAT BATTLES OF THE WORLD
  By STEPHEN CRANE
  Author of “The Red Badge of Courage.”

  BRITISH REGIMENTS AT THE FRONT
  The glorious story of their Battle Honours.

  THE RED CROSS IN WAR
  By M. F. BILLINGTON

  FORTY YEARS AFTER
  The Story of the Franco-German War. By H. C. BAILEY.
  With an Introduction by W. L. COURTNEY, LL.D.

  A SCRAP OF PAPER
  The Inner History of German Diplomacy.
  By E. J. DILLON

  HOW THE NATIONS WAGED WAR
  A companion volume to “How the War Began,” telling how the world faced
  Armageddon and how the British Army answered the call to arms.
  By J. M. KENNEDY

  AIR-CRAFT IN WAR
  By ERIC STUART BRUCE

  HACKING THROUGH BELGIUM
  By EDMUND DANE

  FAMOUS FIGHTS OF INDIAN NATIVE REGIMENTS
  By REGINALD HODDER

  THE RETREAT TO PARIS
  By ROGER INGPEN

  THE RUSSIAN ADVANCE
  By MARR MURRAY

  THE SUBMARINE IN WAR
  By C. W. DOMVILLE-FIFE

  MOTOR TRANSPORTS IN WAR
  By HORACE WYATT

  THE SLAV NATIONS




  MOTOR TRANSPORTS
  IN WAR

  BY

  HORACE WYATT

  ADVISOR ON HEAVY MOTORS TO “THE AUTOCAR”;
  CONSULTING EDITOR OF “MOTOR TRACTION”;
  HON. SEC. OF THE IMPERIAL MOTOR TRANSPORT COUNCIL, ETC.

  _ILLUSTRATED_

  HODDER AND STOUGHTON

  LONDON NEW YORK TORONTO

  MCMXIV




CONTENTS


                                                      PAGE

        INTRODUCTION                                     7

CHAPTER

     I. THE SCOPE OF THE MOTOR VEHICLE                  11

    II. THE IMPORTANCE OF THE MILITARY MOTOR            26

   III. TRIALS AND MANOEUVRES                           41

    IV. EXPERIENCES OF MOTORS IN ACTIVE SERVICE         59

     V. MOTOR AMBULANCE WORK                            77

    VI. THE TRANSPORT OF AMMUNITION AND ARTILLERY       94

   VII. ARMOURED CARS AND OTHER MILITARY MOTORS        102

  VIII. THE PROVISION OF MILITARY MOTOR TRANSPORT      117

    IX. A COMPARISON OF NATIONAL CONDITIONS            132

     X. BRITISH SUBSIDY TYPE MOTORS                    148

    XI. TRANSPORT MOTORS OF CONTINENTAL ARMIES         166

   XII. EMERGENCY MEASURES ON THE OUTBREAK OF WAR      182




INTRODUCTION

BY THE EDITOR OF “THE AUTOCAR”


We have been told, and rightly, many times within the last few weeks
that the present war is unique, not merely on account of the vastness
of the contending armies, but also on account of the power of the
weapons employed. In fact, the war has very properly been described
as an engineer’s war, and such, indeed, it is, as the engineer is
wholly responsible for the tremendous development in every warlike
instrument which has taken place since 1870. He is responsible, too,
not merely for the development, but for the invention of wholly new
methods of offence and defence. But his influence does not end here,
and it is not merely in the firing line that one sees the influence
of the engineer: even as this war is the first occasion on which
modern weapons, explosives and projectiles have been tested on the
grand scale, so, too, is it even more emphatically the first occasion
on which motor transport has been thoroughly tested at all. While the
recent Balkan war provided a practical test of many of the weapons
used in the great war to-day, motor transport played only a very small
part in it; and it is very extraordinary that an innovation of this
kind should be truly tested for the first time upon such a stupendous
scale. It is the motor car, the motor van and the motor lorry which
have rendered the rapid movements of the present war possible; it is
not yet realised to the full how great have been the services of motor
transport in the supply of ammunition and food to the troops, and in
the rapid conveyance of the wounded to the hospitals. No one is better
qualified than Mr. Horace Wyatt to deal with this new and important
branch of modern warfare. In his capacity formerly as Editor and now as
Consulting Editor of _Motor Traction_ he has studied the question from
its inception: from its small beginnings in British army manoeuvres
many years ago right up to the present time he has followed the subject
with the closest attention. Not only so: he has personally investigated
the work performed by motor transport in the _grandes manoeuvres_ on
the Continent. I have had the good fortune to work closely with him for
many years, and it puts me in a position to say that his knowledge of
the subject is unique both in detail and in general, so that readers
of the present volume may rest assured that facts and facts alone are
dealt with in its pages.

                                                      H. W. STANER,
                                                Editor of _The Autocar_.

  Coventry,
  _October, 1914_.




MOTOR TRANSPORTS IN WAR




CHAPTER I

THE SCOPE OF THE MOTOR VEHICLE

    Early History--The Industrial Motor--The Motor ’Bus and Motor
      Cab--Steam Lorries and Tractors--Petrol-Electric Vehicles--Daily
      Mileages and Fuel Consumption.


When we remember that the motor vehicle as we know it to-day is the
result of a development not more than a quarter of a century old, its
enormous influence upon the character of modern warfare must indeed be
regarded as remarkable. Especially is this so in view of the fact that
progress has not in the main been dictated by military considerations,
but almost entirely by the requirements of private individuals and of
peaceful trading concerns. The case is very different from that of the
aeroplane and the airship, which from the very moment that they began
to appear as practical possibilities, were recognised as having far
greater potentialities in connection with warfare than in any other
sphere. The whole science of flight has been studied to a great extent
from this point of view, and the Government Departments concerned, in
all civilised countries, have recognised the necessity of keeping in
touch with and encouraging the movement, and have realised all along
the nature of the work to be done by the flying corps.

On the other hand, the use of the motor vehicle was extended in
the first instance mainly as a sport, and as a new occupation for
well-to-do individuals of a mechanical turn of mind. There is an
attraction about speed in all forms, and consequently, it was on this
point that attention was for many years concentrated. Furthermore,
developments were influenced to no slight extent by changes of fashion,
and the need of satisfying the requirements of people who were not
necessarily qualified to direct progress into the best possible
channels. The motor vehicle was used as a luxury, and exploited as
a means of bringing into being new forms of sport, for many years
before it acquired sufficient reliability or worked with sufficient
economy to justify its employment on economic grounds. The industrial
motor industry is, in fact, at the present day only about ten years
old. In the first instance, one of the principal factors in securing
the occasional use of motor vans was the advertisement value of a
rather unusual type of vehicle, which naturally attracted considerable
attention wherever it went. A little later mechanical transport was
adopted by a limited number of firms, not on account of any superiority
in economy or reliability over old systems of delivery, but rather with
a view to extending the area embraced, and so gaining an advantage
over competing concerns more than sufficient to balance the increased
cost involved by the employment of vehicles by no means cheap either
as regards first cost or operating expenses. Once the industry was
established, however, its rapid growth was inevitable, since it was
found possible to construct vehicles the employment of which was more
than justified on purely economic grounds. The line of least resistance
was found in connection with public services and hackney carriages
for the conveyance of passengers, while in the carriage of goods the
new means of transport had to compete with cheap if slow systems
of delivery by horsed vehicle, and with the railways which, if not
offering a direct method, at least offered a very cheap one when a
large volume of traffic had to be handled.

In the other sphere, competition was limited chiefly to the horsed
’bus, the horsed cab and the tram car, and the last named was under
a disadvantage in some quarters, since conditions exist in parts of
London and in various other cities extremely unfavourable to the
complete employment of railed transport on the roads. The motor cab
was assisted in driving the horse cab off the streets by the stupid
conservatism of the old-fashioned cab driver, who refused point-blank
to employ the taximeter, and so to forego the advantage which he had
obtained by keeping his fare in a certain amount of ignorance as to the
proper legal charges to which he was entitled. The promoters of the
early motor-cab companies took advantage of this state of affairs, and
introduced the motor cab and the taximeter simultaneously. The vehicle
itself had the attraction of novelty and the advantage of greater
speed, while its early popularity was still more directly due to the
taximeter giving an accurate check of the amount payable on every
journey. In this sphere, consequently, the victory of mechanical over
horse transport was rapid and inevitable. Simultaneously, the motor
omnibus made steady, if not quite such speedy, progress. Its advantage
in speed over the horse ’bus was at first the determining factor, but
after improvements in the mechanism, giving increased comfort and
reliability, it was able to get the better also of the electric tram
in spite of the advantage possessed by the railed vehicle of larger
carrying capacity, which of course tends towards reduced operating
costs per passenger carried. The inflexible nature of a tramway system
has been the principal factor in securing the popularity of a free
road vehicle, and at the present moment the motor omnibus is able to
compete directly with great success against the electric tram car. So
it came about that passenger transport was very rapidly converted to
mechanical power. If London is taken as an example, we find that at
the present moment over 95 per cent. of passenger transport is carried
on by mechanical vehicles, while certainly not more than 15 per cent.
of goods transport has yet been similarly diverted. Nevertheless,
the motor vehicle for the carriage of goods has made great progress,
particularly in this country.

[Illustration: A GROUP OF MOTORS IN THE SERVICE OF THE FLYING CORPS AT
A RECENT REVIEW IN FRANCE.]

Throughout its history, it has been greatly helped by the prior
existence of the steam traction engine. From these heavy and slow
machines, suitable only for limited use in particular spheres, have
been developed two very useful classes of lighter steam-propelled
machines coming under the provisions of the Motor Car Acts. The
first is the steam tractor, which is merely a small edition of the
traction engine, able, on account of its lighter weight, to travel
at considerably higher speeds. The other is the steam lorry, which
is an extremely valuable machine for the carriage of anything up to
about six tons of goods at speeds of about five miles per hour. From
the five-ton steam lorry there has more recently developed a lighter
type of steam vehicle in the shape of the three-ton lorry, generally
running on rubber tyres, and so entitled legally to travel at much
higher speeds. The great economy of steam motors made it absolutely
necessary for the makers of internal combustion industrial vehicles to
study every possibility of reducing operating costs. They had on their
side advantages as regards higher speed possibilities and more complete
independence of fuel supplies. The steam motor of ordinary type cannot
be conveniently designed to carry with it fuel and water supplies
adequate for very long journeys. On the other hand, the steamer has the
great advantage of being able to exert tremendous power at low road
speeds. The steam engine is more flexible and more capable of standing
a heavy overload than the internal combustion engine. Even if it is
brought almost to a standstill, it can go on applying the full steam
pressure behind its piston during every stroke. Given adequate supplies
of fuel and water, it is an admirable and very economical machine for
all sorts of rough and heavy work. Curiously enough, the steam lorry
and the steam tractor have been essentially British developments, and
as such they have done much to bring the British industrial petrol
vehicle up to its present high standard of perfection.

The essential differences between a tractor and a lorry should here be
noted. The tractor is designed merely to haul a load, while the lorry
is primarily intended to carry its load. In the first case, the engine
and the load-carrying vehicle are two separate units coupled together;
in the second, they form one unit. The latter is the more convenient
arrangement so far as manoeuvring in enclosed spaces is concerned,
since a good deal of skill is needed to back a tractor train with
accuracy. Also, the steam lorry uses its load to increase the adhesive
power of its driving wheels. On the other hand, the steam tractor can
itself be doing useful work, while some of its load-carrying vehicles
or trailers are being loaded or unloaded. By providing two sets of
trailers, it can be kept usefully employed and need not waste time at
its terminal points. Moreover, if it is required to work under very
difficult conditions, it is a great advantage to be able to unhitch the
engine from the trailer. If, for example, the bed of a river has to be
crossed and the wheels sink into loose sand, the tractor is unhitched
and run through without its load until it reaches solid ground. When it
is brought to a standstill, its engine is employed through the medium
of wire rope gear to drag the loaded trailer slowly but very surely
out of its difficulty. Thus, for cross-country work, the tractor has
much to recommend it, and it is not surprising that the success of the
five-ton steamer has led to systematic endeavours to perfect internal
combustion tractors possessing all the same advantages, and also
self-contained for long journeys as regards fuel and water supplies.

Mention has already been made of the fact that, when a tractor is used,
the load does not assist the adhesion of the wheels. This constitutes,
as it were, an artificial limit to the tractive power, and has
naturally caused some designers to consider methods by which the engine
power of a tractor could be applied not only to one pair of wheels but
to all the wheels, so that the whole weight of the engine itself can be
used to secure adhesion.

The four-wheel drive is not common in commercial service, as it has
only been found necessary under a limited number of very severe
conditions. A good deal has been done, however, in this direction,
particularly in France. The resulting vehicle need not be purely a
tractor. In fact, we often find heavy lorries employed not only to
carry a substantial load, but to haul an additional lighter load in
a trailer. As a rule, these trailers have iron-tyred wheels, but for
service in which economy of engine power is more important than economy
of money, rubber tyres are usually fitted, since they have the effect
of reducing the power absorbed in hauling the trailer by about 25 per
cent.

Another development which is due mainly to the difficulties of adopting
the internal combustion engine for the haulage of heavy loads without
shock, is the petrol-electric system. In this system the power of the
car engine is used to drive an electric dynamo. This dynamo generates
current which is either supplied direct to electric motors or else
stored in a battery of accumulators, the former method being the
better and more likely to survive. Sometimes one electric motor is
used, taking the place of an ordinary gear box, and driving the back
wheels through a universally jointed shaft and a differential gear. In
other cases, two balanced electric motors are employed in or near the
driving wheels. In others again, two motors are used, each driving
through shaft and differential gear to one axle of the vehicle, and
so providing an electric four-wheel drive. Another arrangement is the
provision of four electric motors, one for each wheel. The vehicle is
controlled through the medium of a “controller”; that is to say, an
apparatus which, by the movement of a handle, varies the electrical
connections and so makes the installation suitable for providing either
a big torque at low speeds, or a comparatively light torque at high
speeds. Electrical machinery is also in a sense self-regulating, and
consequently a well-designed petrol-electric transmission is tantamount
to the provision of an infinitely variable change speed gear. One of
the strongest arguments against the petrol-electric method is that,
when the machine is running fairly light and fast, the electrical
machinery involves certain unnecessary power losses. Consequently,
systems have been devised in which mechanical and electrical drive
are combined, the latter only operating the vehicle under conditions
equivalent to an increase of load on the engine.

Efforts have been made for many years past to evolve a satisfactory
internal combustion engine using paraffin or some other heavy and
comparatively cheap oil in place of petrol. While these attempts have
by no means failed, the practical results are up to the present more or
less limited to the use of paraffin fuel in tropical or semi-tropical
countries, where the higher temperature facilitates its employment.
Among the disadvantages of paraffin are difficulties in starting up,
a tendency to soot up the sparking plugs, the need of more frequent
cleaning of cylinders, and a certain amount of disagreeable smell,
partly due to the creeping of the liquid through every available
crevice.

So far as the ordinary petrol van or lorry is concerned, various types
have been developed to meet a variety of commercial needs. A certain
number of light vans are run on pneumatic tyres, but the solid tyre is
preferred wherever economy is more important than speed. It of course
goes without saying that, if a chassis is to run on solid tyres, it
must be of substantial construction, and so designed that its mechanism
will not be injured by the fact that the solid tyre is not so capable
as the pneumatic of absorbing small vibrations.

A very popular type of motor van is designed to carry about 25 or 30
cwt. These machines are capable of speeds up to about 25 or even 30
miles per hour in emergency, and can average comfortably 14 to 16
miles. Under reasonable conditions, they can cover daily journeys of
100 to 120 miles. Among larger types the 3-tonner predominates. This
class of machine can be generally used for daily journeys of 70 to 90
miles, averaging perhaps 11 or 12 miles per hour. It usually consumes
petrol at the rate of about 1 gallon to 8 miles, though better results
are obtainable under good conditions. There are also a large number of
5-ton petrol lorries in commercial service. These can be advantageously
used to cover 60 or 70 miles a day, consuming about 1 gallon of petrol
to every 6 miles run. The motor cab runs about 20 to 25 miles on a
gallon of petrol, and the motor omnibus about 7 to 10 miles. This
question of fuel consumption is, of course, distinctly important in
military service, when adequate supplies are only maintained at the
right points with considerable difficulty.

In later chapters some account is given of the attempts made by various
governments to influence the development of motor traction into the
directions dictated by their military needs, but this brief sketch of
the general trend of events will be sufficient to indicate the present
position, and to provide the necessary knowledge for the appreciation
of the facts and considerations to which we shall now turn.




CHAPTER II

THE IMPORTANCE OF THE MILITARY MOTOR

    The Opinions of German and British Military Experts--The Old and
      New Methods of Transport and Supply--How Troops in the Field are
      Fed.


Although we, in Great Britain, have developed the industrial
motor vehicle almost entirely with a view to the improvement of
communications in time of peace, various circumstances, which will
be referred to in more detail in a later chapter, have led other
countries to fasten their attention more firmly on to the application
of mechanical power to military needs. Very considerable sums of money
have been expended during the past five or six years with this end in
view, and such expenditure could only have been justified if a full
study of the probable course of a great war under modern conditions
had led to the conclusion that the motor is something more than an
accessory and convenience, but is rather one of the prime essentials of
success. In order to prove that this view is, in fact, held by those
who have devoted their whole time to the study of modern warfare, one
need go no further than the now famous or notorious book on _Germany
and the Next War_ by General F. von Bernhardi:

    “In a future European war ‘masses’ will be employed to an extent
    unprecedented in any previous one. Weapons will be used whose
    deadliness will exceed all previous experience. More effective and
    varied means of communication will be available than were known in
    earlier wars. These three momentous factors will mark the war of
    the future.”

From this statement it is clear that, even if only improvement in
means of communication is considered, the motor vehicle forms one of
the three greatest factors in moulding the course of modern warfare.
Railways have been available in many previous wars, and there can be
no doubt that the reference to more effective and varied means of
communication is occasioned almost entirely by the development of
motor vehicles suitable for use in the transport and supply columns.
Simultaneously, both of the other prime factors are affected by the
introduction of motor vehicles. Road motors can assist materially in
massing men rapidly at any desired point, and mechanical power is
absolutely essential for the transport of guns of enormous calibre, the
employment of which in the field is only in this way rendered possible.

Quoting again from the same authority we get an idea of the bearing of
our subject upon a military theory now universally accepted as true.

     “The commander who can carry out all operations quicker than
    the enemy, and can concentrate and employ greater masses in a
    narrow space than they can, will always be in a position to
    collect a numerically superior force in the decisive direction;
    if he controls the more effective troops he will gain decisive
    successes against one part of the hostile army, and will be able
    to exploit them against other divisions of it before the enemy
    can gain equivalent advantages in other parts of the field.... If
    the assailant can advance in the decisive direction with superior
    numbers, and can win the day, because the enemy cannot utilise
    his numerical superiority, there is a possibility of an ultimate
    victory over the arithmetically stronger army.”

Taking this statement in conjunction with the well-known German theory
that safety only lies in offensive warfare, we realise immediately the
incalculable importance of the introduction of any new system which
will give to large bodies of men the powers of more free and more rapid
movement. When armies are increased beyond certain numerical limits,
it becomes absolutely necessary for them to depend upon supplies
brought up regularly from the rear, and not upon the uncertainties of
living upon the country.

    “Improved means of communication facilitate the handling and
    feeding of large masses, but tie them down to railway systems and
    main roads, and must, if they fail or break down in the course of
    a campaign, aggravate the difficulties, because the troops were
    accustomed to their use, and the commanders counted upon them.”

We have here a complete recognition of two most important points. The
first is, that the use of motors in the transport and supply columns,
if successfully carried on, represents an enormous advantage, which may
even allow ultimate victory to come to a numerically inferior army. In
the second place, we have the acknowledgment that any breakdown in the
service for which the motor vehicles are responsible, will be fatal to
success.

A military correspondent of the _Daily Telegraph_ has recently
emphasised the same point. He has pointed out that hitherto the massing
of an army of about a quarter of a million men has represented the
probable limit of possibilities, and that even then such numbers could
only be massed for a short period. The Russo-Japanese war, in which
larger numbers were engaged, has by no means disproved this theory,
since it partook of the nature of a siege rather than that of a field
campaign. At the present moment, the enormous numbers dealt with
envolve certain limitations in movement, the scope of which is dictated
by the distribution of railways and of roads. Without motor transport,
the rate of movement of huge armies would be necessarily very slow, the
radius of action from railhead would be small, and the daily movement
of the troops would be strictly circumscribed for more reasons than
one. The effect of the introduction of motor transport is somewhat
similar to that which would be obtained if the railway could, in a few
hours, be extended in any direction along any made road for a distance
of about forty or fifty miles. The delivery of supplies, as it were in
retail, to the troops must still be carried out by horse transport,
since motor lorries are not suitable for continuous use where made
roads do not exist. The comparatively slow movement of horsed vehicles
even now affects the rate of progress of an army. When huge bodies of
men are in motion, the depth from the front to the rear of the army is
very considerable, and at the end of the day the supplies have to be
brought up from the rear to the front in time to enable the whole force
to be fed.

[Illustration: PART OF BIG FLEET OF “ALBION” LORRIES PURCHASED BY THE
BRITISH WAR OFFICE.]

[Illustration: A GROUP OF BRITISH “BERNA” LORRIES TAKEN OVER BY THE WAR
DEPARTMENT.]

[Illustration: A FLEET OF “THORNYCROFT” LORRIES REQUISITIONED FOR
SERVICE.]

[Illustration: A FLEET OF “HALLFORD” LORRIES CALLED UP IMMEDIATELY ON
THE OUTBREAK OF WAR.]

The use of transport and supply motors does not amount merely to the
employment of a large number of these machines for miscellaneous
duties, but rather corresponds to bringing into existence a new link
in the chain of the main system of supply. The existence of railways
behind the army is assumed. At some safe point along the railway is
formed the base, and from this base stores are brought up to a point
known as “railhead,” This is the point where, for the time
being, military rail traffic ceases. It is evident that railhead is
a variable quantity, liable to move forward or backward from day to
day. The main accumulation of stores is at the base, and the stock at
railhead at any moment consists only of sufficient to meet one day’s
requirements. Before the introduction of motor transport, the whole
of the supplies from railhead had to be taken by horsed vehicle, and
subsequently distributed in the same way among the troops. Under the
new method, motor lorries carry the supplies up to a place called
“re-filling point,” which is a movable point situated from day to day
in the most convenient position possible to arrange, with a view to the
distribution of supplies by horsed vehicle to the army.

In the old system, the transport vehicles worked in _echelons_. The
first of these, with the baggage and supplies for a day, followed so
closely behind the troops as to be able to join them every night.
The next, half a day’s march behind, carried supplies sufficient to
replenish the first column daily. Further back again were other
_echelons_ carrying on the same scheme. This meant that the whole of
the roads for enormous distances behind the forces were encumbered
by transport. Between railhead and the army there were many links
involving endless possibilities of confusion, and consequently shortage
in supplies. Moreover, food came up to the troops very slowly from the
base, and it was impossible to supply a regular stock of fresh meat and
bread.

The advantage of the new system is based on the speed capacity of the
motor vehicle, a supplementary point being an enormous reduction in
the length of a column carrying a given quantity of supplies. It is,
however, the higher speed of the motor which has the greatest effect,
since it enables many columns to be replaced by one. In the words of
Colonel Paul: “One _echelon_ of mechanical transport can do the work
of five _echelons_ of horse transport, and one column will suffice to
connect the horse transport immediately behind the troops with the
railway.”

The result is to facilitate operations of troops up to a distance from
railhead represented by half of a full day’s work for the motors. The
simplest way of appreciating the result obtained is to take an actual
example. Under the new system, on, let us say, Tuesday evening, the
soldier at the front is provided with a hot meal of fresh meat, cooked
by the regimental travelling kitchens on the march. This food had been
handed over to the kitchens on Monday evening by the distributing
horsed vehicles, which had received it sometime during Monday at the
re-filling point a few miles back from the motor supply column, which
had left railhead perhaps 50 miles from the front in the small hours of
Monday morning. Previously, the supplies had been brought down by rail
from the base, and in this way the food which the soldiers are eating
on Tuesday night, was probably to be found in the neighbourhood of the
base on Sunday afternoon in the shape of live animals.

Working out the scheme from a rather different point of view, the
soldier on the Tuesday night is in possession of Wednesday’s supply
of bread and cheese, and an emergency ration of preserved meat in
case of any delays or breakdown in the transport service. The horsed
vehicles are at the time empty, and are returning to meet the motors
at re-filling point. The motors by this time are back at railhead
waiting for Wednesday’s supplies to be discharged from the railway
trains. At about three o’clock on Wednesday morning the motors will be
loaded and ready to start. Their speed capacity will enable them easily
to catch up with the distributing horsed vehicles before the end of
Wednesday’s march, and to tranship their supplies at re-filling point
for distribution on Wednesday evening.

The whole system is, in reality, very simple, and it enables large
armies in the field to be supplied daily with fresh meat and bread
instead of being dependent on food brought up slowly and in many
stages, and for that reason necessarily of a character less nutritious,
and much more liable in the long run to cause illness among the men.
At the same time, the big carrying capacity of the motor has served
to clear the roads behind the army of an enormous block of vehicles
essential in the past, but now no longer necessary. In connection with
this point, Colonel R. H. Ewart, D.S.O., representing the Indian Office
at the Imperial Motor Transport Conference of 1913, gave some very
interesting figures, which may be quoted as an extreme case:

     “Up to 1910 the reports show that there were nearly
    five-and-a-half million bullock carts in British India alone, and
    in all our wars up to date, we have had to mobilise a very large
    number of these carts for our line of communication work. We find
    that when moving in large bodies, the utmost speed we can rely upon
    is about one-and-a-half miles an hour. We have worked out that it
    takes six bullock carts to move in eighty days what one 2-ton lorry
    can transport in ten. The bullock carts take up twice the room on
    the road for a given load, and in the matter of establishment--a
    question which you will all realise in the time of war is a very
    serious one--it takes thirty-five men, drivers, artificers and
    supervisors to look after what one man could do with a lorry.”

From these figures it will be seen what an enormous saving is effected
by the use of motors, even if we only take the point of view of the
feeding, maintenance and payment of the men actually employed in the
transport columns themselves.

The impossibility of imposing upon horsed vehicles the necessity for
gaining fifty or even thirty miles in the course of a day, in order to
catch up by the evening with an advancing army after leaving railhead
in the morning, is perfectly obvious. The truth of the statement
already made that the use of motors for transport and supply work is
a necessity and not merely a convenience in modern warfare, is thus
made clear, and under the circumstances, readers who are perhaps more
attracted by the more showy, but less essential, uses of motors in
war, will understand that a consideration of the subject of this book
must necessarily be devoted very largely to the organisation and
_matériel_ of the supply columns.

The class of vehicle most commonly favoured for the work of feeding
troops in the field is the 3-ton petrol lorry, capable of covering
eighty or ninety miles in a day, and if need be of travelling under
fairly favourable conditions at twenty miles an hour. Behind very
mobile troops, such as cavalry, preference is sometimes given to
lighter lorries rated to carry 30 cwt. or 2 tons, and capable of
rather higher rates of speed and rather bigger daily mileages. Some
European Powers favour for general work lorries carrying 4 or 5 tons,
and in addition capable of drawing an extra 2 tons or so upon a
trailer. In every case, the internal combustion vehicle is preferred on
account of its independence upon frequent renewals of fuel and water
supplies. However, steam tractors are often used for various classes
of specially heavy work, as, for example, for drawing the travelling
workshops which have to be established at the movable base of the
supply columns at railhead.




CHAPTER III

TRIALS AND MANOEUVRES

    Early Tests of Steam Lorries--Lord Kitchener’s Views on Motors in
      the South African War--British W. D. Tractor Trials--The Carriage
      of Troops by Car--The Army Manoeuvres of 1912--Recent Trials in
      England, France, and Germany.


Naturally, the motor vehicle could not be entrusted with work of the
first importance in time of war without previously going through a
period of encouragement and probation. Some fourteen years ago, motor
cars and cycles began to be used in small numbers during military
manoeuvres in Great Britain and elsewhere.

In the French manoeuvres of 1901, cars and motor tricycles were
employed for transporting staff officers, and for scouting work. The
motorists who lent the cars were entrusted with the duty of driving
them, and were granted certain privileges on that account. Results
were on the whole satisfactory.

In the same year, the British War Office, as a result of experience
gained in South Africa, were encouraged to conduct trials of motor
lorries. The entrants were five in number. Four of these were steam
lorries, the makes represented being the Foden, Straker, and two types
of Thornycroft. There was only one entrant of an internal combustion
engined machine. This was a Milnes-Daimler modelled on the German
Daimler cars, and having a four-cylinder engine rated at 25 h.p., with
ignition by low-tension magneto. Fuel was supplied by pressure of the
exhaust, and the car had a channel steel frame and large built-up steel
wheels. Even at that comparatively early date, the Foden lorry was, in
general appearance, very similar to the standard steam lorry of to-day.
It was, of course, fitted with a locomotive-type boiler, this being
a practice which has since been adopted by almost all manufacturers
of this class of machine. The Thornycroft lorries had vertical
boilers, and the one type was representative of standard practice,
the other being rather a peculiar machine driven from the rear. The
Foden and standard Thornycroft were most successful in carrying out
the very arduous road tests imposed, which involved a large number of
particularly steep hills. The Foden was by far the most economical in
water and fuel. The trials ended by cross-country tests in the Long
Valley at Aldershot, and during these the Foden was unfortunately
driven by accident into a deep ditch, with the result that its front
axle was broken. Consequently, the standard Thornycroft received the
first award and the benefit of subsequent small orders from the War
Department, although at the time there was a rather strong feeling that
the Foden ought also to have been recognised.

It was not until about two years later that, in the publication of
evidence given before the commission appointed to inquire into the
conduct of the South African war, the opinion of Lord Kitchener on the
utility of motor transport in its then state of development was made
public. His views were expressed as follows:

    “We had (in South Africa) about forty-five steam road transport
    trains. As a rule they did useful work, but questions of weather,
    roads, water and coal distinctly limited their employment as
    compared with animal transport, to which they can only be regarded
    as supplementary. The motor lorries sent to South Africa did
    well. Thornycrofts are the best. They will in the future be found
    superior to steam road trains as field transport.”

From this it will be seen that the main result of South African
experience was to indicate the superiority of the comparatively light
self-contained motor vehicle over the heavy traction engine.

In 1903, a considerable number of cars and cycles supplied by members
of the Motor Volunteer Corps were used in the British manoeuvres. The
cars employed numbered forty-three, and averaged about 12 h.p. They
were used mainly for staff work, and were very fairly effective.
The attempts to use them for the carriage of searchlights were not
very successful. Some thirty motor cycles were employed for carrying
despatches, and behaved on the whole splendidly. Mr. J. F. Ochs, in
describing, during a lecture at the Royal Automobile Club, the results
obtained, made a somewhat prophetic statement in his remark that, “If
Mr. Marconi could perfect his invention, how useful a car fitted with
it would be.”

While the undoubted utility of motors for staff work and for scouting
was recognised at least as a certainty of the future, progress in
comparatively heavy military transport was for some years after
this limited. The military authorities were averse to the use of
petrol-driven cars, on account of the supposed danger of employing so
inflammable a fuel. Efforts were made to use paraffin, but results
were not particularly satisfactory. The Mechanical Transport Companies
at Aldershot went on experimenting with and developing the use of
steam vehicles, and particularly of steam tractors, which came to
be regarded as, on the whole, more suitable for rough work than
self-contained lorries. By 1906, the mechanical transport sections
were in possession of adequate tractor-drawn workshops, to support
the varied fleet of mechanical vehicles available for a variety of
purposes, as well as the staff cars, a limited number of which had been
purchased by the War Department.

Arrangements had also been made for giving the drivers and mechanics
some theoretical as well as practical knowledge, and the movement had
in fact formed itself into the nucleus of what it was then supposed
would be required; namely, an organisation providing for military
service a large number of 5-ton steam tractors, and a limited number of
cars and motor cycles for staff and scouting duty.

For some time, efforts to procure for army service some really reliable
internal combustion tractors running on paraffin were continued. In
February, 1909, trials were held at Aldershot, in connection with
which a considerable premium and valuable prospective orders were
offered as an inducement to manufacturers to turn their attention to
this class of machine. The entrants, however, only numbered three.
One of these was a substantial four-cylinder Thornycroft paraffin
tractor, which performed well throughout and was ultimately successful
in obtaining an award, though it does not appear that the type has
since been adopted in any quantity. A very singular machine which did
wonderful work for its power was a Broom and Wade single-cylinder
paraffin tractor of about 20 h.p. The work of hauling a heavy
military trailer with a load of about 6 tons was, on occasions, too
much for this machine under the sometimes very arduous conditions
under which the trials were carried out. The third entry hardly came
within the scope of what the War Department wished to encourage. It
was a Stewart-Crosbie steam tractor with a two-cylinder compound
double-acting engine giving 40 b.h.p. at 600 r.p.m. It was able to meet
the stipulations as to capacity for carrying fuel and water supplies,
since the boiler was of the vertical central-fired water-tube type
working at 200 lbs. pressure, and supplying to the engine superheated
steam which had been passed through coiled tubes in the furnace.

During the greater part of the trials the roads round Aldershot were
covered with a thick coating of snow, which constituted a serious
difficulty for iron-tyred tractors when using public thoroughfares in
time of peace. In emergency, it would of course have been possible to
fit spikes, or grips of some kind, to the wheels to prevent skidding
and slipping, but this could only be done at the risk of great injury
to the roads which did not appear justifiable under the circumstances.
All three tractors were provided with means for employing their engine
power through the medium of a wire rope, and this had to be utilised
in some cases to get the loads up some of the very steep gradients
encountered. The trials terminated in an extraordinarily difficult
test across the Long Valley. Here much loose sand was negotiated
successfully, and afterwards the engines were required to get their
loads across a deep swamp. The Thornycroft was the most successful,
but even the little Broom & Wade machine managed to carry out the work
by means of a system of pulleys applied to its wire rope gear. It was
curious to watch this little engine dragging its big load in a trailer
which had sunk almost to the wheel tops in mud and water. Occasionally,
turf and mud had to be dug away from the front of the trailer, and when
it was in motion the wheels were actually rotating slowly in the wrong
direction under the influence of the pressure of a continuous supply of
weed-bound mud surging over their tops.

These trials, if not satisfactory in attaining their main object, at
least helped to demonstrate that practically nothing is impossible to
a soundly constructed motor vehicle, if properly equipped for rough
work. Intermittently, experiments have been made at Aldershot with
various machines of more or less peculiar construction. Among these may
be remembered the Pedrail tractor, the wheels of which carry a number
of articulated feet which, as the machine progresses, plant themselves
one after another squarely upon the ground. The necessary mechanism
was, however, too complex to render anything of the sort suitable for
extensive military use, and the same trouble probably applies to the
Caterpillar type of tractor, in which the wheels are surrounded by a
track in the form of a sort of endless chain, which lays itself as the
machine moves upon the ground, and distributes the weight over a large
area. Passing over rough country, a tractor of this sort rolls like
a ship at sea, but is very seldom in any real difficulty, even when
traversing ditches or fairly low hedges. Steering has to be effected by
allowing the wheels on one side to over-run those upon the other, with
the result that the engine turns with a sort of skidding motion.

An interesting test of the value of the motor car in war was carried
out by the Automobile Association on March 17th, 1909. The Association
made an offer to Lord Haldane, then Secretary of State for War, to
transport a battalion by motor vehicles to any coast town that the
War Office might consider a possible scene of invasion. The point
ultimately selected was Hastings. For the purpose of the scheme it was
assumed that a sudden concentration of troops at Hastings had become
necessary, and that a battalion of the Guards was about to entrain in
London, when information was received that a portion of the railway
line had been destroyed by spies or agents working on behalf of the
enemy. Under such circumstances, the battalion could only be sent by
road. On the date named, a battalion of infantry at full war strength,
over 1,000 officers and men, with machine guns, ammunition, medical
stores, tools, food, water, baggage, blankets, and other impedimenta
amounting to some 30 tons, was distributed among 286 touring cars and
about 50 motor lorries.

The cars were lent and driven by members of the Automobile Association,
and several manufacturers of heavy motor vehicles provided the
necessary number of lorries for carrying the guns and stores.

The battalion was a composite one, consisting of officers and men
of the Grenadier and Scots Guards from Chelsea Barracks, Wellington
Barracks, and the Tower. The programme, which entailed picking up the
men at their respective barracks, joining up the three columns at the
Crystal Palace at 10 a.m. and arriving at Hastings soon after 1 p.m.
was carried through successfully, and within half an hour of arrival
the battalion with its full equipment was marched along the sea-front.

The experiment aroused considerable interest in military circles in
this country and abroad, particularly so in Germany, where a number of
newspapers published full particulars and a plan of the route taken.

In 1908 the German Army Department adopted the scheme which it has
since enforced for securing military transport, and from that time
onwards annual trials have been held, generally in the late autumn,
and over heavy and mountainous roads. In this they have differed from
the majority of the annual trials held in France, first of all under
the auspices of the Automobile Club of France, and later directly by
the military authorities. Our neighbours have shown a tendency to
make the routes selected somewhat easy, and not to test the vehicles
over unduly severe gradients. The German scheme was re-considered at
the end of 1912 as a result of the experience obtained up to that
time. The trials of 1912 were over a distance of about 1,300 miles,
including roads through the mountains of central Germany. The distance
covered each day by the 4-ton lorries drawing additional 2-ton loads
on trailers was about 60 miles. Subsequently, the newer regulations
prescribed more strict limits of axle weight, in view of uncertainty
as to the strength of the roads and bridges which would have to be
negotiated. A minimum engine power of 35 h.p. was prescribed, and
gradients of one in seven had to be taken with full load and equipment.
An interesting point of the new German regulations is the provision
of a belt pulley somewhere on the driving shaft for the purpose of
operating machine tools. Another point is the stipulation that the
brakes of the trailing vehicle shall be capable of being operated from
the driving seat of the lorry. A certain degree of standardisation was
at the same time introduced.

In the same year, a big step towards the proper utilisation of motor
transport for military work was taken by an extensive experiment made
in this direction during the British Army manoeuvres. The use of
mechanical transport was subsequently referred to by the King as one
of the special features on that occasion, and the opinion was very
generally expressed that the rather sudden and early termination of
the manoeuvres was due to the unexpected effect of motor transport
in increasing the mobility of the troops, and bringing the opposing
forces into contact with one another with startling rapidity. Even
so late as 1912, a certain number of military authorities were still
very doubtful as to the advisability of relying on the motor vehicle
in active service, but the manoeuvres in question undoubtedly proved
the case, although the difficulties of operating mechanical transport
for the first time on an extensive scale were increased by the fact
that the machines available were of all sorts of makes and types, no
attempts at standardisation having been possible. Many of the machines
hired for the occasion were in very poor condition, and did not compare
favourably with those owned by the Government. Consequently, the
difficulties of working in convoy at short intervals were accentuated.
All the motor transport was concentrated on one side, and the armies
dependent upon it were kept well supplied daily with fresh meat, the
opposing forces being dependent on horsed transport and chilled meat.
Motor buses were on one or two occasions during the manoeuvres utilised
with great success for the rapid movement of fairly large bodies of
men. These manoeuvres probably represented the last appearance of
traction engines for any military use other than the haulage of very
heavy guns, or other kinds of quite abnormal work, not forming any part
of the regular system of supply and transport.

During the last few years trials have been held at irregular intervals
by the War Department for the purpose of testing the suitability of
various specially constructed motor lorries for recognition under the
subsidy scheme, the nature of which is explained in detail in a later
chapter. The last trials of this kind took place early in 1914. An
official report published in June stated that results had been very
successful as regards both the number of entrants, and the general
standard of excellence of the vehicles submitted. The average speeds
both on easy and on hilly routes were well above those specified.
Radiators were found to be amply large to be effective even in the
hottest weather. The Mechanical Transport Committee reiterated their
opinion that one of the two systems of brakes should act upon the
propeller shaft. The average fuel consumption of the competing cars
was exceedingly good, working out at 52 gross ton miles per gallon.
The best result was about 63 gross ton miles per gallon over a distance
of about 200 miles. On the whole, it is evident that the cars were
very satisfactory, since it was stated that there appeared to be no
necessity to make any serious alterations in specifying for future
requirements.

The most recent French trials were hardly completed when war broke out.
They were as usual well patronised, but not calculated on the whole to
try the machines to the utmost. It was intended in subsequent years
to introduce new and more stringent regulations, but the opinion was
fairly generally expressed among manufacturers that the Government in
doing so were differentiating their own needs too far from ordinary
business requirements, and that it would be impossible to find a market
for the types indicated. Early in the year, another series of trials
of considerable importance was held in France, for the purpose of
testing new types of four-wheel driven tractors. These machines are
needed particularly for the haulage of artillery, and further reference
to them will therefore be deferred until that subject comes up for
consideration.




CHAPTER IV

EXPERIENCES OF MOTORS IN ACTIVE SERVICE

    The South African War--The Italian Transport in Tripoli--The Balkan
      Campaigns.


Although mechanical transport was employed during the South African
war, the experiences then gained must not be applied with too much
rigidity to the conditions of the conflict taking place in Europe.
In South Africa, a considerable number of traction engines were put
into service, while steam motor lorries were also used. Colonel R. E.
Crompton, C.B., who was in charge of the British transport columns, has
described how “De Wet, knowing the country, destroyed bridge after
bridge until the roads and the railways were only islands, disconnected
by things called ‘deviations’--horrible places, full of dead animals,
horse transport, animal transport of all kinds, which had died there,
simply because there was practically no road.... The fact that we
were able, even though we had broken engines, to repair them from our
spares, so that the dead engines became live engines, so impressed
Lord Roberts that he felt that we were at the birth of real, practical
mechanical military transport with all the advantages it gives.”

There can be no doubt that the experience obtained during the South
African war pointed directly to the use in the first case of steam
tractors, and later--when they could be sufficiently perfected--of
internal combustion tractors with a bigger radius of action. These
conclusions resulted not only from the inherent conditions of military
service, but also from the local conditions of the country in which
this particular war took place.

Reviewing the possibilities of South Africa in times of peace, Mr. W.
W. Hoy, the General Manager of the Government Railways and Harbours,
while approving of the use of light passenger and goods vehicles
up to 2 or 3 tons capacity, lays stress on the desirability of the
light paraffin tractor for easy services on good roads, and the heavy
paraffin tractor for cross-country work with trains of trailers each
carrying from 12 to 25 tons of goods. If we admit that a country in
which these represent the main normal requirement cannot be safely
taken as indicating accurately even the war requirements of other
countries, we are reduced for practical experience to the Italian
campaign in Tripoli and the recent wars in the Balkans. Italy is
one of those countries in which commercial motor transport has not,
owing to unfavourable local conditions, made any great progress. As
a result the war was begun without any provision having been made in
this direction, and the authorities were at first very sceptical as
regards the desirability of employing motors at all in connection with
the operations of the army. After much discussion, two light lorries,
fitted with twin pneumatic tyres on the back wheels, were sent out on
trial. These served very rapidly to convince the staff officers of the
superiority of the system over horse transport. Consequently, thirty
more light Fiat lorries were sent out as promptly as possible, and
these were followed by larger consignments, bringing the whole fleet in
use up to the number of about 200. Arrived at Tripoli, the cars were
slung off the transport ships on to big pontoons, and towed to the
quay. From that point they were immediately employed for the transport
of all kinds of war material, as well as provisions and forage. They
were further utilised for the conveyance of large bodies of troops to
the front, and for carrying wounded to the hospitals and dead to the
improvised cemeteries. Most of the country over which they operated
was entirely devoid of roads, and consisted chiefly of rough loose
desert strewn with rocks and treacherous sandy hills. These peculiar
conditions account for the type of vehicle selected for employment.
Heavier lorries on solid tyres would no doubt have experienced even
greater difficulties in negotiating country of this class.

The following extract, from a full account published by the
manufacturers of the uses to which their vehicles were put, will serve
to give an idea of the varied employment of military motors:

    “At the battle of Zanzur, on June 8th, 1912, fifty-four vehicles
    took part and were divided into four columns under the personal
    command of Capt. Corazzi. Ten were under the command of an officer
    at the disposal of the Medical Corps; a second column, under
    the command of Lieut. Milani, carried a load of barbed wire and
    netting, sand bags and shovels; a third column, in Lieut. Bosio’s
    charge, carried also 800 spades, 600 shovels, sand bags, and
    barbed wire; and a fourth column of fourteen lorries, under Lieut.
    Marocco, took a large quantity of dynamite and other explosives in
    addition to pioneers’ tools.

    “The first column to move were the ambulances, which left Tripoli
    at two o’clock, and at 3.30 came out of the outer redoubt at
    Gargaresh to follow the fighting column and to work under the
    instructions of a surgeon-captain. The other columns left Tripoli
    about three o’clock, and at 4.15 at Gargaresh, about 5-1/2 miles
    from Tripoli, they formed up in a square about 350 yards in front
    of the redoubt under cover of a hill, waiting for orders. At 5.30
    they advanced, and leaving cover of the hills, moved forward
    about 2-1/2 miles beyond the batteries. The nature of the ground
    changed as the columns approached a sandbank, which had until
    then protected them from the enemy’s fire. The passage over this
    sand dune was extremely difficult, as the cars had to proceed in
    single file at walking pace, exposed to a violent rifle fire.
    Proceeding round the extreme north of the Arabo-Turkish trenches
    the columns reached the Marabotto of Abd-el-Gelil shortly after the
    arrival of the third battalion of the Fortieth Fusiliers, mountain
    artillery, and a company of pioneers, and proceeded with the work
    of fortification. When the columns returned to Gargaresh, and while
    the Rainaldi Brigade was engaged against overwhelming forces of
    the enemy, one of the motor columns, acting under Lieut. Milani,
    was ordered to load provisions, whilst the other two were told off
    to join the ambulance section. In the very line of fire the motors
    brought succour to the wounded, conveyed some seventy disabled
    soldiers to the temporary hospital at Gargaresh, and carried forty
    dead to the cemetery. At Gargaresh the order arrived to convey to
    Marabut the provisions and luggage of the 6th and 40th regiments of
    the line. The three motor columns therefore re-formed, one going to
    Tripoli to load provisions and returning to Marabut, the other two
    being loaded up with luggage. The three columns then returned to
    Tripoli.”

After two years of incessant service, and notwithstanding the “emery”
effect of the fine sand which was carried in clouds by the wind and
penetrated everywhere, it is generally understood that the Italian
military motor fleet maintained reliable services throughout the
war, and that the individual machines were in surprisingly good
condition when their service was completed. Results were, at any rate,
sufficiently satisfactory to justify the Italian Government in placing
considerable further orders, with a view to increasing their motor
columns. This war was probably the first event which enabled the motor
vehicle to prove itself in practice absolutely essential as a military
implement. A Tripoli newspaper summed up the value of the experience
obtained:

     “Many people will have asked themselves how it was possible for
    the Lequio Division to live, march, fight, and win with a base
    of operations distant from 70 to 200 miles, with rapid and long
    deviations which were almost of daily occurrence, in a country so
    barren and inhospitable that man and beast would perish if they
    were left for only two or three days without provisions.... The
    motor lorry provided the solution of the problem; by its use in a
    few hours provisions were brought from the stores and bases to the
    fighting column, having been conveyed possibly hundreds of miles,
    and, further, by its means not one day passed without the troops
    having bread, wine, and coffee. The motor lorry was ubiquitous; it
    transported ammunition or succoured the wounded, fetched fodder for
    the horses and other animals, or money for the troops and for the
    Arabs; it brought new boots for the soldiers or delivered urgent
    messages, as well as being used for the transport of troops from
    the various bases right up to the first fighting line in battle.
    Only the advent of the autocar rendered possible many of the
    daring moves of this war, as it solved the difficulties of desert
    transport.”

As regards the uses of motors during the various Balkan campaigns, the
only reliable and available information appears to be that contained in
a series of articles contributed by Capt. A. H. Trapmann of the _Daily
Telegraph_ to the columns of _Motor Traction_. At the commencement
of the war in 1912, there were less than 100 motor vehicles in
Greece, and some sixty of these--the property of Greek subjects--were
immediately commandeered. The machines formed a fleet very far from
ideal, representing cars of all makes and sizes, many of them suffering
from negligent treatment or unskilful handling, and some very near the
termination of a chequered career. The officers entrusted with the duty
of purchasing the machines were completely ignorant of their value or
qualities, and the drivers into whose hands they were subsequently
put consisted mainly of people who could, or said they could, drive a
motor car, though the great majority did not profess to possess any
knowledge over and above that required for travelling with reasonable
safety and certainty, assuming the mechanism of the cars to give no
trouble at all. The better machines were chiefly allotted to the
various generals and their staff officers, while some of the worst were
fitted up with lorry bodies for the transport of goods.

After the fall of Salonica, the Greek objective was Janina, connected
with the port of Preveza by an excellent road about sixty-three miles
long. Directly Preveza fell into Greek hands, the authorities were
faced with the problem of provisioning an army, in the first instance
consisting of 15,000 men and gradually augmented to 60,000, operating
against a fortified town in a totally barren country intersected by
huge mountain ranges. The front of the army extended for about a
hundred miles, and only one good road was available from the base to
the centre of the advanced positions. Under these conditions, the
authorities realised the possibilities of motor transport, and about
thirty motor lorries, mainly obtained from Italy, were shipped to
Preveza and put into service. It was found that each lorry could,
in three hours, carry to the front about enough food for 1,000 men.
This, however, was not the only problem. The army was absorbing on
the average one ton of ammunition per day for every thousand men. The
lorries were only capable at the best of handling 2-ton loads, and
consequently were kept more than fully occupied. Moreover, the road,
though good in certain portions, was in others particularly dangerous,
being very winding and hewn for the most part out of the side of a
precipice. Heavy traffic and heavy rains contributed to make the
conditions yet worse, and under the circumstances, it is not surprising
that very serious accidents occurred, and that by the end of the first
six weeks only nine out of the original thirty lorries were still upon
the road. It then became necessary to replenish the supply, which was
managed in one way or another, and the service was maintained with
enormous difficulty under conditions of false economy, which dictated
considerable purchases of unreliable secondhand machines. Even so,
the results served completely to convince Captain Trapmann that motor
transport was the only solution of the supply problem in warfare.

It seems that a similar opinion was forced upon the Greek military
authorities, since one of the first moves when the second campaign
became inevitable in 1913, was the purchase of one hundred motor
lorries. This step, while good in itself, was inadequate, since no real
provision was made for the supply of competent and responsible drivers,
for adequate supervision, or for completely equipped workshops. Many
of the drivers were well-to-do enthusiasts who had volunteered for
service, and who very soon came to regret that they had done so.
It is one thing to drive a good touring car and to fall back upon
professional assistance whenever trouble occurs, but quite another to
handle and maintain a heavy motor lorry without competent backing and
under thoroughly bad conditions of service. Some 50 per cent. of the
motor fleet was usually out of commission, and the staff of the repair
shops were so incompetent that it was seldom that a car once taken to
pieces was ever fit for the road again. The following extract from
Capt. Trapmann’s account gives some idea of the difficulties which had
to be overcome:

    “The strategy and tactics of the campaign against Bulgaria landed
    the Greek headquarters at Doyrani on July 8th, and there nearly
    two-thirds of the Greek motor service was concentrated on July
    10th. Greek headquarters decided to move sixty miles west to Hadji
    Beylik along the railway, and the vital question was how the cars
    for the service of the staff, and the lorries for the army service
    were to accomplish the journey. The single-line railway track was
    impossible on account of unbridged gaps, and also because the
    railway was in urgent demand for transport. The only semblance of
    a road was a mule track two feet wide, which led for the most
    part through a tangle of vegetation, and occasionally amidst a
    wilderness of rocks and stones. Eventually it was decided literally
    to force a road by sheer weight. The lorries took turns at leading,
    raced full speed for twenty yards, and then bashed their way
    through the jungle. After fifty yards or less the lorry would be
    brought to a standstill by the accumulation of rubbish piled up in
    front. This would be cleared away, the car would back, then start
    on a fresh charge. When a lorry got seriously damaged it would
    be replaced by another and taken in tow by a third. Sometimes
    explosives had to be used, and small rivers were bridged by the
    simple expedient of placing tree trunks in them until a car could
    cross. It was bumpy work.

    “In Macedonia a road of any sort was a luxury, the best roads
    could not compare as regards surface with a fourth-class English
    roadway, whilst as often as not the motors had to make their own
    road as they went along. It must be remembered, also, that driving
    in war time is very different from under peace conditions. Bridges
    and culverts have usually been destroyed, telegraph lines sag
    across the road, and at night time are apt to get entangled in the
    driver’s neck with dire results. I, myself, have seen a goodly
    number of motor smashes, one when a temporary bridge gave way
    under an overloaded lorry, another when a contact mine exploded.
    The worst accident I remember, however, took place soon after the
    fall of Janina. A very old and depreciated lorry was being used to
    convey passengers down to Preveza, a distance of sixty-three miles,
    by a mountain road which for half its length was cut on the edge of
    a precipice. At one of the awkward places on the road the steering
    gear broke, and the car with its human freight dashed over the
    cliff and fell into the river.”

The conclusions reached by Captain Trapmann as a result of very
exceptional opportunities of observing military motor transport
under active service conditions should be of considerable value. His
catalogue of desirable features is as follows:

    (1) Clearance from ground in order to enable a car to pass over
    rock-strewn stretches.

    (2) An adjustable cow-catcher in front for use at night on good
    stretches of road, on which, however, dead or wounded horses or men
    may be lying.

    (3) An inclined bullet shield of light steel to protect the front
    of the radiator from casual sniping.

    (4) A stout iron hook or ring in front and behind for towing
    purposes, especially when a river has to be negotiated, the bridge
    over which has been destroyed.

    (5) Solid tyres with a set of non-skid chains which can be fitted
    when occasion arises.

    (6) A wire grappler to preserve the driver from the danger of
    sagging telegraph wires hanging across the road.

While the experiences detailed in this chapter are, comparatively
speaking, on a very small scale, and consequently results cannot
confidently be applied in anticipation to a war of immensely greater
magnitude, they have at least served to show that even unavoidable lack
of experience, or avoidable lack of competence, cannot prevent the
motor vehicle from being a very valuable asset behind an army in the
field. The Tripoli and Balkan campaigns proved not only the necessity
of employing motors for the work of the transport and supply columns,
but also the possibility in so doing of saving the lives of very
many wounded men who, when dependence was placed on slower methods,
frequently died from exposure on their way down from the front to the
base.




CHAPTER V

MOTOR AMBULANCE WORK

    Considerations of Design of Emergency Ambulances--Points to
      be borne in Mind--Some Examples of Practical Designs now in
      Service--The Work of Motor Ambulances at the Front--Scouring the
      Battlefields--How the British Red Cross Society gets its Fleet.


Among military uses of motor vehicles, the motor ambulance probably
comes next in order of importance after the transport and supply
waggons. Evidently with the motor ambulance must be grouped cars
suitable for use in carrying wounded men who are not obliged to be
transported in a recumbent position, and even ordinary touring cars
when employed, as they are being somewhat extensively at the present
moment, for taking convalescent men on health-giving motor trips. This
last is a quite useful class of work in which even those motorists
can participate who are only able to offer their services and those of
their cars in the vicinity of their own homes and at specified hours.

In times of peace, the motor ambulance proper is, so far as its chassis
is concerned, more akin to an industrial vehicle than to a touring car.
The heavier examples, in some cases, run on solid rubber tyres, and
in others on twin pneumatics, while the lighter types are fitted with
single pneumatics of heavy section. In detail, the chassis is simple
and strong, and well adapted to be put under the charge of a driver
of only average mechanical ability. The principal points are that the
vehicles should be silent in running, not liable to derangement and
extremely well sprung. Owing to the first consideration, worm-driven
chassis are particularly suitable for this class of work, and owing to
the second a slightly modified light van chassis is generally to be
preferred to the highly-refined but more complicated touring car.

[Illustration: AMBULANCE DRILL OF THE CITY OF LONDON BRANCH OF THE
BRITISH RED CROSS SOCIETY, USING CARS FITTED WITH THE B.H.S. SPRING
STRETCHER-SUSPENSION.

LIFTING THE STRETCHER.]

[Illustration: LOADING THE STRETCHER INTO THE CAR BODY. IT IS
AFTERWARDS RAISED FORE AND AFT AND ATTACHED BY STRAPS TO THE SPRING
HOOKS PROVIDED. NOTE THE MEN RUNNING TO THE FRONT TO LIFT THE FORWARD
HANDLES.]

[Illustration: A TOURING CAR EQUIPPED WITH A FOUR-STRETCHER AMBULANCE
BODY BY MESSRS. BROWN, HUGHES AND STRACHAN, TO THE ORDER OF THE DUCHESS
OF WESTMINSTER.]

In time of war, the ambulance chassis is, roughly speaking, anything
big enough and sufficiently reliable that can be made available. For
example, motor omnibuses can be without much difficulty adapted to
this class of work, while touring cars are often quite suitable. The
qualifications in the latter instance are fairly ample engine power,
thorough reliability and strength for working over rough road surfaces,
very strong springs and ample wheelbase, so that the ambulance body
shall not overhang the rear of the chassis to too great an extent.
To form the complete vehicle, what is wanted at such times is not
necessarily a luxuriously-equipped conveyance, but is rather a quite
light and simple body sensibly constructed to bear its load, and
capable of standing any amount of jolting without either its component
parts shaking loose among themselves, or the body as a whole becoming
insecure in its connections with the chassis.

As regards the interior equipments, in most instances all that is
needed is provision for readily fixing in place two or four stretchers
as the case may be, and also for loading the stretchers on to and
unloading them off the body without difficulty, and without unnecessary
discomfort to the patient. The standard types of ambulance body
approved by the British Red Cross Society consist of simple but stout
wooden frameworks with all the joints reinforced by angle irons held
by bolts through the wooden members, and not merely by wood screws,
which are liable to work loose. Over this framework is stretched a
cover of waterproof canvas that has been treated with rubber, while the
front and back of the vehicles are covered in by waterproof curtains
of similar material, capable of being drawn aside or raised quite
easily so as to enable attendants, with the minimum of difficulty, to
lift the loaded stretchers into the vehicles. Medical experts who have
experience in the carriage of wounded men do not appear to be entirely
in agreement as to whether the stretchers in a motor ambulance should
be rigidly secured to the vehicle body, or should be carried by some
form of springing supplementary to that of the car itself. In some
examples of ambulances in regular use in this country, additional
springing is provided by suspending the body from the chassis by means
of semi-elliptical or complete elliptical springs. In many others, no
springing other than that of the vehicle itself is interposed between
the stretcher and the ground. One point at least on which there is
universal agreement is that on no account must any rolling motion of
the stretcher relative to the vehicle body be permitted, as motion of
this kind causes acute discomfort to the patient, and often leads to
physical effects similar to those occasioned by the rolling of a vessel
at sea.

Probably side loading is the ideal method of getting stretchers on to
an ambulance car, but it is difficult to realise the ideal in the case
of a simple and fairly cheaply constructed body. Consequently, the
system of end loading is far more common. In this case, the stretcher
is generally slipped in along the floor of the vehicle, and when right
inside the car, is raised to the necessary elevation to allow it to
be secured in position. The lower stretchers are afterwards slipped
in and similarly secured. A design in which this is possible is more
convenient than one in which the upper stretchers have to be raised to
their full height before the operation of sliding them into the car can
be attempted.

[Illustration: _“The Autocar” photograph._

A FRENCH MILITARY SEARCHLIGHT MOTOR. THE SEARCHLIGHT IS CARRIED ON AN
UNDERFRAME, WHICH CAN BE LET DOWN IN THE MANNER SHOWN. THE ELECTRICITY
IS SUPPLIED FROM THE MOTIVE POWER OF THE CAR.]

[Illustration: _“The Autocar” photograph._

AN OPEN TOURING CAR CONVERTED INTO A MOTOR AMBULANCE.]

A type of fitting which has been adopted for some of the two-stretcher
ambulances of the British Red Cross Society is that known as the L.X.R.
It consists of a simple steel-tubed framework, the corner members
of which are slotted to take the ball ends of cross bars from which
the stretchers are slung by very short ropes and straps. When in
position in the slots, the cross bar ends bear on the tops of strong
spiral springs which relieve the stretcher of a certain amount of
vibration but at the same time, not being free to move other than in
the vertical, do not set up any rolling motion. When these fittings,
which are manufactured by Messrs. Simonis, are used, no weight is
carried from any portion of the ambulance body except the floor.
Consequently, the remainder of the construction can be very light and
merely designed to support the waterproof covering. For four-stretcher
bodies the British Red Cross Society have at the time of writing been
employing two main types. In one of these, the stretchers merely slide
in, the upper ones on to shelves and the lower ones along the floor,
and are secured quite rigidly in position. In the other type, a system
of spring suspension has been adopted. This latter system--evolved
for the Society by Messrs. Brown, Hughes and Strachan--adapts itself
to the construction of a simple, strong, but quite inexpensive body,
to the main members of which are bolted iron arms which can be easily
arranged so that they can be swung to one side while the car is being
loaded, if it is considered that there is any risk of their interfering
with the ease of the operation. Each of these iron arms has a flattened
end, bored to take a vertical iron rod, the lower portion of which is
formed into a hook, while the upper portion carries a heavy spiral
spring concealed in a neat casing. The stretchers are carried from the
hooks by means of quite short leather straps, connecting the hooks
with the stretcher handles. It will be noted that the springs allow
of no movement other than one in a purely vertical direction, and
consequently that practically no rolling should result from the use of
this system, which has the advantage of giving an additional spring
suspension at a very small increase in the cost of the complete body. A
very considerable number of ambulances built on these lines have been
supplied by the makers to the Red Cross Society.

It is impossible to lay too much emphasis on the desirability of using
for ambulances, chassis with long wheelbase, in which the stretchers
are as far as possible carried between the wheels, and the patients
thereby protected from direct road shock. It is not to be expected
that short wheel-based chassis carrying ambulance bodies with a big
overhang at the rear, will prove durable over the broken roads of the
countries in which war is taking place. If wheelbase is not sufficient
to allow of the fitting of a four-stretcher body without these grave
disadvantages, the only thing to do is to put up with the smaller
accommodation of a two-stretcher body. The usual arrangement in this
case is to extend the body right forward towards the dash on the left
of the driver, and so to push the stretchers a couple of feet further
forward, the space inside the body behind the driver serving for the
carriage of luggage, for an attendant, or for one or two wounded men
who are not very seriously injured. One of the great dangers in this
arrangement is that of obstructing the view of the driver towards his
left. This is particularly serious when the car is for use in countries
where the rule of the road is the reverse to our own, and where traffic
in the opposite direction has to pass on the side upon which a free
view is obscured. A possibility is to take in the space below the
driver’s seat and that alongside of it, and to run in the patients
on their stretchers, feet first, alongside of one another on the
floor of the conveyance. In this way, about a foot of length can be
economised, and with a two-stretcher body of this type it is of course
not necessary for the super-structure to be either strong or high.

Another important point, if the cars are to go abroad and to be used
under bad conditions of road surface, is that any ordinary simple
method of attaching the body to the chassis must be very carefully
examined before it is approved. Something more than average security is
needed.

A fair number of touring cars are being changed into motor ambulances,
not by replacement of the body but by its adaptation. This method has
the disadvantage that it renders the old body subsequently useless for
other purposes. Further, it is likely to cause delay, since every case
has to be considered on its individual merits. Also, unless the chassis
is a long one, the adaptation will almost certainly involve a big
overhang.

[Illustration: _“The Autocar” illustration._

A TYPE OF EXTEMPORISED MOTOR AMBULANCE FAVOURED BY THE FRENCH AND
BELGIANS.]

These notes will serve to give the necessary information to those who
may wish to equip motor ambulances for any kind of use during the
war, and there does not appear to be any need to go into details of
all the various other varieties of ambulance body, many of them very
beautifully fitted and designed, but also very expensive. One other
type may, however, be mentioned, since it is being employed extensively
by the French and Belgian Governments. This consists essentially of a
stout floor carrying two iron frameworks of inverted V shape. Between
these two and stretching fore and aft is an arrangement similar in
principle to a squirrel cage, or to a water-wheel with four floats. The
place of each float is taken by the necessary apparatus for the support
of a stretcher, provision being made that all the four stretchers
retain their horizontal position whatever the position of the framework
supporting them. The stretchers can be loaded in from the side to the
bottom position, and the apparatus swung round so that this operation
is continued, the stretchers after being loaded being subsequently
raised by the rotation of the frame. It is stated that in Antwerp and
elsewhere this type of ambulance has been used extensively, and is
found to be very comfortable and very easy to construct.

Turning to the work for which the ambulances are being employed, much
of this is of an obvious character. Ambulance services are evidently
needed both at the military hospitals, and also further back at the
big base hospitals of the Red Cross Society on the Continent. They are
wanted again at all the various hospitals in this country to which
wounded men are brought. They are employed, for example, in London, to
meet the hospital trains and carry from the stations those men who are
not able to be conveyed in ordinary cars.

The requirement of motor ambulances nearer the front is almost
limitless. In the system of the R.A.M.C. in service, wounded men are
first removed by regimental stretcher-bearers to the “aid post,” where
medical attention is first given to them. Thence, they are carried by
the bearer sections of the field ambulance--and possibly, if roads
permit, by motor ambulances--to the advance dressing stations, whence
after treatment they are taken by the military ambulance waggons to
meet conveyances from the clearing hospital, which is usually situated
somewhere near the railhead. Upon this hospital falls the duty of
avoiding all overcrowding nearer the front, and this must be done by
employing all available means of transport. Evidently, motor ambulances
are the most suitable kind of conveyance for this work, since they
afford a reasonable degree of comfort to the patient, and even if their
speed capacities cannot be utilised to any great extent while they are
carrying wounded men, advantage can be taken of them while returning
empty towards the front for further load. Once the patients have been
taken as far back as the field hospital at railhead, their subsequent
conveyance to the Red Cross hospitals, or any other required points,
can be carried out by train supplemented by local motor ambulance
services from the termini to the hospitals.

Another and less obvious type of service is that which involves
thorough patrolling of all those districts in which battles have taken
place, with a view to ascertaining whether any wounded men are still
remaining in the villages and along the country-side, where they may be
given thoroughly kind, but possibly somewhat unskilled, attention by
the civilian inhabitants. Another duty of the drivers of the ambulances
carrying out this work is that of setting on foot minute inquiries with
a view to finding out whether any men killed in battle have been buried
by civilians without any record having become available which would
serve as a basis for certain information which can never be so terrible
as an almost hopeless state of suspense. This class of work, of course,
has to be carried out over roads which have in many cases been badly
broken up by heavy military traffic, and possibly even intentionally
destroyed by a retreating enemy. Consequently, it puts a very severe
strain on every portion of the chassis and body of the ambulance,
and makes the fact that the whole of the motor vehicles at present
employed by the Society have been freely given or lent by their owners
without reservation and without charge all the more noteworthy.

For some time past, the Society has been shipping ambulance cars, and
also touring cars, to the Continent as rapidly as means of transit have
permitted. The requirement seems to be enormous, but even so there does
not appear to be any likelihood of the supply falling short of it.
Many motorists have placed not only their cars, but their own services
ungrudgingly at the disposal of the Society. The usual practice is for
the Society, after accepting a car for service, to undertake to have a
suitable ambulance body put upon it in place of its own. In some cases,
however, motorists have even taken this charge upon themselves, and
whatever may be the disadvantages of dependence upon volunteer service,
it can at least be said that in this case such dependence has served in
some measure to show how many men, unable for one reason or another to
take up military duties, are only too anxious to expend their energies
and their money on any object of national value in connection with
which they are able to be of use.




CHAPTER VI

THE TRANSPORT OF AMMUNITION AND ARTILLERY

    The System of Ammunition Supply--The Traction Engine--French
      Four-Wheeled Tractors for Hauling Guns--German Gun-Carrying
      Motors.


The system of maintaining ammunition supplies for troops in the field
is very similar to that already described in connection with the supply
of food. Stocks of ammunition are kept at depôts of the Army Ordnance
Corps at various points between the base and railhead, to which they
are forwarded as required. From railhead they are brought forward daily
by the motor lorries of the divisional ammunition parks to a convenient
re-filling point, where they are transhipped on to horsed ammunition
carts for distribution in detail. With the troops entirely dependent
on the motor vehicle for the maintenance of their supplies both of food
and of ammunition, there is no need to labour the enormous importance
of mechanical transport in modern warfare, or the terrible consequences
which would follow anything approaching a general failure in the
reliability of the machines used.

For the haulage of very heavy guns some form of engine power is of
course essential, and the ordinary steam traction engine provides the
most obvious solution, since it is to be assumed that if the roads
and bridges to be traversed are sufficiently strong to bear the gun
itself, they will also bear the engines which haul it. The big traction
engine is a very British product, and it is interesting, if not quite
satisfactory, to note that the huge siege guns of the German army are
stated to be hauled by engines of British origin. For lighter guns,
the steam tractor, or small traction engine, can be employed, but very
many efforts have been made to dispense with its service in favour of
an internal combustion tractor, less dependent on constant renewal of
fuel and water supplies.

[Illustration: A BRITISH “MARSHALL” INTERNAL COMBUSTION TRACTOR
SUITABLE FOR THE HAULAGE OF ARTILLERY.]

[Illustration: _“The Autocar” photograph._

A FRENCH PETROL-ELECTRIC FOUR-WHEEL-DRIVE TRACTOR, THE WHEELS OF WHICH
ARE SPECIALLY SHOD TO FACILITATE TRAVELLING OVER ROUGH GROUND.]

In the brief sketch already given of the principal trials and
manoeuvres in which motor transport has figured, some indication has
been afforded of the attempts made by the British Government in this
direction. The most consistent, and probably the most successful,
efforts have however been made by our neighbours, the French, who
have for this special purpose given much encouragement to the
development of internal combustion tractors driving all four wheels.
The movement dates back about three years, and owes its origin to the
need for hauling 155 m/m siege guns along roads and across country.
The first type of tractor produced to meet the requirement was the
Chatillon-Panhard. The weight of the first type with full load was
in the neighbourhood of 22 tons, and more recently efforts have
been made to evolve lighter types weighing about 14 tons with their
load, the belief being that these rather smaller machines would be
of great general utility. Some very important trials in this
connection took place in France early in 1914, four types of tractor
participating, namely, the Latil, the Schneider, the Chatillon-Panhard
and the Renault. The first-named is a development of the Latil type of
lorry, in which the engines drive the front wheels, and the whole power
plant is concentrated on to the fore carriage, the back wheels and the
platform being really nothing more in principle than a two-wheeled
trailing vehicle. An extension of this system involves the use of three
differential gears, one for each pair of wheels, and a third as a
balance gear half way along the vehicle from which the drive is taken
fore and aft through longitudinal shafts and worm gearing. All four
wheels are steered as well as driven.

In the Schneider, the drive is taken from a gear box containing
two sets of sliding gears through cardan shafts to the front and
back axles, and alternatively, when required, to a capstan enabling
the engine to haul through the medium of a wire rope. In the
Chatillon-Panhard, the transmission is so arranged as to involve no
universal joints, and only one differential gear. This is mounted on a
transverse countershaft, and the power is taken to the wheels through
bevel gears at the ends of the countershaft, and four diagonal shafts
driving in their turn auxiliary shafts upon which are bevels engaging
with similar bevels on the wheels. The Renault is a very simple machine
of its type. The drive is taken fore and aft from the gear box by
cardan shafts leading to differential gears on the front and back axle.
Either one of the differentials can be locked when desired to help
the machine to find its way out of difficult positions. Yet another
machine which is available to the French Government, though it did not
take part in the trials mentioned, employs electrical machinery in
place of the usual mechanical transmission gear. The engine drives an
electric dynamo, which supplies current to four electric motors, one
for each wheel. On the whole, the French four-wheel driven tractors
have performed very well under severe tests, and it is stated that
approximately 300 tractors of one or other of the types mentioned
are available for military service, though it is possible that this
estimate is somewhat exaggerated.

[Illustration: _“The Autocar” illustration._

A KRUPP MOTOR GUN-CARRYING LORRY, SHOWING THE RAMPS UP WHICH THE GUN
HAS BEEN HAULED SERVING ALSO THE PURPOSE OF HOLDING IT SECURELY IN
POSITION.]

For the rapid transport of light artillery various special machines
have been devised, providing either for the carriage of a gun upon the
platform of a motor lorry, or for the construction of a gun-carrying
vehicle forming one complete unit. In this branch of development the
Germans have shown the most initiative, and Krupps have got out several
interesting designs. In all of these strong motor lorry chassis are
used. A usual system is to fit, by hinging to the back of the chassis,
strong ramps up which the gun may be hauled, either by the power of the
motor engine or by other means. When on the platform, the gun wheels
sink into depressions formed to take them and also bear up against
shaped vertical stops. When the gun is in place the ramps are swung
over, and are so designed that their ends can then be conveniently
attached rigidly to the vertical stops, the ramps themselves also
bearing against the gun wheels and holding them quite secure; or in an
alternative method, the ramps are arranged to grip the axle of the gun
carriage.

Special designs are for motor vehicles capable of a good turn of speed,
and arranged to carry guns especially intended for fighting against
aeroplanes or airships, and consequently so arranged as to allow of
their muzzles being swung up until the gun assumes a vertical position.

Other special arrangements of motor vehicles are those providing for
the carriage of machine guns, but this type perhaps comes more properly
under the title of an armoured motor car.




CHAPTER VII

ARMOURED CARS AND OTHER MILITARY MOTORS

    The Utility of the Armoured Car--Improvised and Other Types--The
      Uses of Touring Cars and Motor Cycles--Specially Equipped Military
      Motors.


While everyone was aware that the heavy motor would play a very
important part in the great war in connection with the transport of
supplies and artillery, and that touring cars would be largely used
for staff purposes, the enormous extent to which the armoured car
has been employed has probably come as something of a surprise. We
were, of course, aware of the existence of such machines, but up to
the present there had been no real proof of their utility. The German
military authorities had evidently, prior to the war, come to their
own conclusions on the matter, and the Belgians, whether they were
prepared in advance or not, were at least very prompt in following
suit. It is doubtful whether, when the war broke out, our own War
Department was in possession of a single armoured car, but fortunately
this is a type of machine in connection with which deficiencies can
be very rapidly made up, and at the time of writing, while no certain
quantitative information is available, we are at least aware of the
existence of British armoured cars on the Continent, and we know that,
for example, London motor omnibuses have been equipped for this class
of duty. In this, and in most cases, the improvised armoured car is
merely an ordinary vehicle with some simple form of body covered over
by armour plating, and with the more vital portions of its mechanism
to some degree similarly protected. Thus, some protection is provided
for the radiator and steering gear. The dash is covered by steel
plating extending upwards to protect the driver, while the platform
or body behind is protected by vertical or sloping steel plates.
Certain examples of German armoured cars that have been captured
answer sufficiently well to this description, but naturally enough
in a country where the supply of industrial vehicles is more or less
inadequate, the touring car has been selected for adaptation. It has,
of course, the advantage of extra speed, and the disadvantage of the
vulnerability of pneumatic tyres. Some of these improvised armoured
cars merely carry men armed with rifles, while others are equipped with
light machine guns. In many instances, searchlights, or very strong
head lamps, are fitted. The latter can, of course, be operated by the
now familiar electric system, a small dynamo and a battery of cells
being carried upon the car. The most convenient arrangement for a more
powerful searchlight is to be found in the use of a petrol-electric
vehicle in which the ordinary change-speed gears are dispensed with,
and their place taken by electric machines. The car engine drives
a dynamo which generates electric current. This is supplied to an
electric motor from which the power is transmitted to the rear
wheels. The arrangement is tantamount to an infinitely variable change
speed gear, and the important point in connection with the subject
under consideration is that the whole or part of the engine power can
be used for generating electric current, which can be easily applied to
a searchlight on the car.

[Illustration: _“The Autocar” illustration._

A “CHARRON” ARMOURED CAR WITH MACHINE GUN.]

[Illustration: _“The Autocar” illustration._

A “SCHNEIDER” ARMOURED CAR WITH QUICK-FIRING GUN.]

Armoured cars carrying powerful lights are no doubt very effective
for reconnoitring during the night. The lights can be switched on
quite suddenly and some damage inflicted upon an enemy, and the light
switched off again before there is time for fire to be returned or the
machine to be located effectively. The car can then be moved rapidly to
some other point, and the manoeuvre repeated. In general, the armoured
car has been used as a kind of advance guard in front of the screen
of cavalry which performs the double duty of concealing the movement
of its own infantry and locating the forces of the enemy. Just as the
Germans employed cavalry for this purpose to an unprecedented extent in
the Franco-Prussian war of 1870, and so gained an enormous advantage
over their opponents, so the attempt has been made on this occasion
to utilise a still more rapid and effective means towards the same
ends. As already suggested, any advantage that may have accrued from
the utilisation of a new method of this kind need only be a purely
temporary one, since any desired quantity of the enormous number of
available machines can--if it appears desirable--be converted into
armoured cars for our own use or that of our allies at very short
notice. One would imagine that events will prove that the armoured
motor is valuable as an irritant rather than as a means of locating
hostile forces or screening one’s own, since both these latter ends are
very much affected by the use of aeroplanes, which make it practically
impossible to move large bodies of men secure from observation, and
correspondingly easy to gather fairly accurate information as to the
whereabouts and strength of the enemy.

So far we have touched only upon extemporised armoured vehicles as
distinct from those actually designed in the first instance for
this specific duty. The first armoured car was produced as long ago
as 1896. The design was published in _The Autocar_ only a week after
the Act which permitted a motor car to exceed four miles an hour on
British roads, and to dispense with the man walking in front with
a red flag, came into force. The suggestion emanated from the late
Mr. E. J. Pennington, and the machine, the mechanism of which was
necessarily somewhat primitive, was arranged to carry two small machine
guns the cranks of which were to be driven by the car engine with a
view to increasing the rapidity of fire. In those days, the machine
gun was usually hand-cranked. At intervals since that time various
designs had been brought out, the general principle being to employ
a completely armoured car, the driver of which would in most cases
be in considerable danger of accident, owing to the way in which the
protection provided for him obscured the view of the road on either
side. In the majority of cases, the vehicle was designed to carry one
machine gun mounted in a turret above the roof. Thus, in a design
got out by the Charron Company some time back, the machine gun was
contained in a rotating turret a little way forward of the rear axle.
The joint between this turret and the roof of the car involved a flange
and a thick rubber ring. The turret rotated on a central vertical
shaft, and on this shaft was a screw wheel. The effect of turning this
wheel was to raise or lower the turret. When the machine gun had been
drawn into the right position with the turret raised, the wheel was
turned bringing the turret down hard against the rubber ring which held
it securely, and prevented it from shaking about while the gun was
being fired.

[Illustration: _“The Autocar” illustration._

AN ITALIAN DESIGN FOR A MOTOR BATTERY OF QUICK-FIRING GUNS. THE SIDES
SWIVEL ROUND TO FACE IN ANY DESIRED POSITION.]

In another design, coming from the Creusot Works of Messrs. Schneider,
provision was made for a larger machine gun, carried in a substantial
turret projecting from the car roof, and mounted upon rollers running
on an inwardly projecting ring on the lower fixed portion of the
turret. This ring was toothed on its inner side and engaged with a
gear wheel enabling the turret, and with it the gun, to be swung round
into the desired position. The gun itself carried a seat, and the
gear for rotating the turret was connected with pedals, so that a man
sitting on the gun could rotate it and the turret by the action of his
feet, keeping his hands free for the refinements of aiming and working
the weapon.

Reference may be made to one more design for which an Italian officer
was responsible. In this case, the vehicle formed a kind of moving
battery of machine guns, mounted so as normally to point out of the
sides of the car. Each half of the body, however, was capable of
being swung round on hinges either at the back or at the front, and
castors were provided to facilitate its motion. Thus, when the car was
stationary, it was possible to swing round the whole of its armament so
as to face the front, rear, or either side.

[Illustration: _“The Autocar” photograph._

A “MINERVA” ARMOURED CAR WITH MACHINE GUN.]

If the general conclusion as to the utility of armoured motor cars
bears out the impressions formed in the earlier portions of the war,
there can be little doubt that these more comprehensive designs will
receive in the near future consideration which has been denied to them
in the past, and that types of armoured car will be evolved as much
more effective than the extemporised patterns as our armoured cruisers
are when compared with converted merchant ships.

Without devoting too much space to the consideration of machines
which are as yet merely proposals and not actualities, brief mention
may be made of a design recently got out by a British engineer, and
representing in a sense the last word in armoured cars, since it is
in no sense a make-shift, and provides for the complete protection of
the driver and every item of the mechanism. The car is, of course,
completely enclosed, and from its roof projects an armoured turret
containing two machine guns. The driver gets his view of the road only
through louvres in front and in the side doors. The lines of the car
consist of a series of curves which are preferred to flat surfaces,
in order to increase to a maximum the possibility of deflecting any
bullets which strike the vehicle. Even the radiator and the tyres are
armoured. The former is situated against the dashboard, and has above
it a cover in the shape of a cupola through which the air is drawn down
by a fan round the vertical tubes. Each wheel is built up of two steel
discs, one inside the other, and an air tube covered by strong fabric
is placed between the two. The outer disc is allowed sufficient freedom
of movement to enable the arrangement to approximate the pneumatic tyre
in effect, while being completely protected from puncture from any
cause.

It is reasonable to suppose that the near future will see considerable
developments in the armoured motor car in two directions, namely, in
the direction of the vehicle designed and constructed throughout for
a specific purpose, and also in the direction of the lightly armoured
fast touring car available for staff and scouting purposes.

This last brings us to the subject of a very valuable sphere of
activity of motors in warfare. There is, however, but little to be
written on this point, since the general use of cars by staff officers
from the commander-in-chief downwards may be taken for granted, and the
employment of fast vehicles for scouting purposes and by officers of
the Intelligence Department is equally self-evident. For the carrying
of dispatches and other such work, the motor cycle is being found
extremely useful. This, the lightest class of motor vehicle, is also
used in conjunction with its heavier relations. Motor cyclists, who
are usually skilled mechanics, are attached to all the heavy motor
transport columns, their duties being to scout ahead, to keep the units
of the column together, and also to assist in the event of any roadside
trouble.

Motors of all kinds are extensively used in connection with the flying
corps. To each squadron of aeroplanes a number of motors are attached
for various duties. Some may act as first-aid machines, and for the
carriage of spare parts. Other larger and heavier motors are employed
for the carriage of partially dismantled aeroplanes. Others, again, are
fitted up as workshops to help in the important work of repairing and
keeping in tune the engines and mechanisms of the aeroplanes.

As regards other important uses of motors in warfare, brief mention
should at least be made of the cars fitted with wireless telegraphy
equipments, and portable searchlights, and also of motor field
kitchens.




CHAPTER VIII

THE PROVISION OF MILITARY MOTOR TRANSPORT

    Systems of Direct Purchase and Subvention Compared--The Advantages
      of the Latter--The Importance of Standardisation and Workshop
      Equipment--The Limitations of the Subsidy Scheme.


Having decided definitely that a complete system of motor transport
must be employed primarily in order to secure greater efficiency and
freedom of movement of troops in the field, the next step is to decide
upon the best means of securing the availability of the necessary
number of suitable vehicles in time of war. Evidently, the simplest
procedure would be to depend solely and entirely upon the power of the
Government to commandeer or requisition the required supplies.

At first sight it may appear that nothing more is needed, but any such
conclusion would be highly erroneous. If we were to examine the fleet
of any large motor omnibus, motor cab, or motor haulage concern, we
should almost inevitably find that the vehicles employed were almost
all of one make and commonly of one type. If an operating company has
in the first instance decided to adopt a particular make of vehicle,
and if subsequent improvement in design reduces the efficiency of the
original type as compared with others of the market, then the natural
move is not to change from one manufacturer to another, but to increase
or partially renew the fleet by the purchase of new vehicles of the
same make but of a more modern model. The change from the old to the
new type does not involve alterations in by any means every part of
the mechanism, but only in those parts which have in any way shown
themselves capable of improvement. In the event of renewals being
required, it is not then necessary to stock an entirely new set of
spare parts for all portions of the car, but only to get in spares for
those parts, the design of which has been changed and improved. In this
way, the necessary stock of spare parts is so far as possible reduced,
and the work of maintaining the cars is in a similar degree simplified.
Almost every type of motor vehicle has its own peculiarities, and it
is evidently easier for a mechanic to undertake the maintenance of a
certain number of machines all of one make than to keep in running
order a similar number of miscellaneous vehicles varying essentially
from one another.

Then again, the standardisation of one make is an advantage, because
for purposes of maintenance the number of workshop appliances required
is reduced to a minimum, and it is possible in some cases to obtain
machines specially adapted for turning out in quantity some particular
part which figures largely in the maintenance of the fleet.

Yet another advantage is that the driver of any one car can, without
danger or loss of efficiency, be put on to any other car, if his own
is undergoing repair or overhaul, while the work of those departments
concerned with the storing and issuing of parts is greatly simplified,
and the accommodation required for the efficient operation of the whole
concern is reduced.

If these arguments apply to an industrial organisation working under
normal conditions, they apply still more strongly to a hastily enlarged
temporary organisation evolved in time of war. Moreover, in the latter
case the unreliable running of a fleet of cars does not represent
merely a temporary financial loss or a diminution in prestige. Its
result must inevitably be to cause, among the troops behind which
the motor column is working, a lack of necessities either in respect
of food or of warlike materials. In either case, the result is
immeasureable and the consequences may well prove fatal.

Then again, military motor vehicles are required to work under
peculiarly arduous and trying conditions. The very nature of their
service implies frequent long runs under the worst possible conditions
of weather and road surface. They may have to employ lanes or bye-ways
or even routes which can hardly be described as roads at all, and added
to this is the almost certain fact that the tracks over which they work
will have been materially injured intentionally or otherwise.

Those who live in the vicinity of any important military centre must
be well aware of the damaging effect that heavy military traffic has
upon the roads, even if well-constructed, in view of the inevitable
nature of the traffic. When plying on country roads never intended
for such use, the transport motors themselves will soon break up the
road surface. These considerations serve to show that the liability
to breakdown is much greater in military than in civilian transport,
and coupled with this is the certainty that the facilities for
conveniently carrying out repairs and overhauls must necessarily be
extremely limited. The transport columns are supported at their base by
travelling workshops manned by skilled mechanics and containing small
selections of those tools likely to be of the most general service. The
equipment of these workshops must be reduced to a minimum in order to
secure their portability, and it is highly important, if possible, to
prevent jobs coming in which cannot be satisfactorily tackled with the
machinery at the disposal of the mechanical staff.

It will thus be seen that everything points to the extreme
inadvisability of depending upon motor transport and supply columns
formed of a miscellaneous collection of vehicles of all types, all
makes and all ages. Looking at the other side of the question, the
ideal conditions are reached when every vehicle of the column is
identical and represents the very best make and type, and when all
the drivers are thoroughly acquainted with the idiosyncrasies of this
particular type of vehicle, and the mechanics responsible for repairs
equally experienced as regards every feature of the mechanism of the
machines.

To secure something approaching true standardisation in a fleet of
transport vehicles, either one or two alternative methods may be
adopted. The first and most obvious is that of direct purchase by
Government. At the moment of writing, this system is being extensively
adopted in Great Britain, and doubtless also in all other belligerent
countries in which it has been possible to keep suitable motor
manufacturing works in operation. Such steps are, however, being
taken, in order to meet a great emergency which has arisen before
alternative schemes have had time fully to mature. The establishment
of an army in time of peace is very much less than it is in war time,
and in time of war an army must be far more self-supporting than it
is in time of peace, when considerable quantities of supplies can be
brought regularly by civilian contractors to depôts where the troops
are stationed, and the military authorities require only to secure
the distribution of such supplies in detail. In time of war the whole
of the supplies must be delivered in bulk to a very limited number of
points, and from that time onwards the military authorities must be
responsible for what may be described as their wholesale as well as
their retail distribution.

Added to this are a number of other considerations, as, for example,
the fact that when on active service the scale of rations of the men
is increased, and supplies of warlike stores are rapidly expended and
have to be perpetually renewed. It is clear, then, that if the method
of direct purchase alone is depended upon, either the supply of motor
vehicles will be immensely greater than the useful requirement in times
of peace, or else facilities must be created for increasing their
supply instantaneously when mobilisation occurs, or the organisation of
new armies becomes essential.

Now, as in the present instance, it is possible after a war has
begun to provide for a steady and considerable supply of transport
motors to be handed over to the military authorities week by week,
provided always that the process of manufacturing is not seriously
interfered with either by the propinquity of military operations, or
by the need of drafting men in excessive numbers from the works to
the active forces. In our own case, it is quite within the bounds of
possibility to produce motors for the transport columns of new armies
just as rapidly as it is possible to make the personnel of those new
armies effective. This fact, however, does not cover the difficulty
occasioned by the necessary increase in transport facilities for the
standing army directly war breaks out. It has been suggested that the
difficulty might be overcome if the War Department were to purchase
large numbers of suitable motor lorries, and to employ the greater
part of them in time of peace for the carriage of general goods. This
scheme has the advantage that it not only provides the necessary fleet,
but simultaneously trains the necessary drivers; nevertheless, it has
the grave drawback that profitable employment of the kind required
could not be found unless the Government were to enter into serious
competition with haulage and delivery companies. It has been proposed
also that numbers of suitable motors might be used normally in the
service of the Post Office, and transferred on emergency to the War
Department, but this again is open to objection. The Post Office
fleets would have to be renewed hurriedly and under difficulties, and
a certain amount of disorganisation would almost certainly result.
Furthermore, the number of vehicles which could be usefully kept in
service by the postal authorities is small compared with the increased
military requirements occasioned by the outbreak of war.

We may take it, then, that the principle of maintaining in the
possession of the War Department in time of peace sufficient motor
vehicles to fill the whole of the needs in time of war is unworkable
except at enormous cost, since the majority of the vehicles could be
put to no useful work and would merely deteriorate and become obsolete
and, therefore, comparatively speaking, valueless were they to stand
idle. The whole of such a fleet would have to be replaced every three
or four years, and if this were not done an enemy equipped with more
modern vehicles would possess a marked advantage, since--though the
motor industry has now assumed enormous proportions--it is still so
young that progress in design is by no means stationary.

We now come to the question of whether it is possible to maintain
in time of peace only the number of vehicles actually required, and
to fill up the requirement in excess of this number as promptly as
possible, but nevertheless with some delay, when war breaks out.
On this point, Captain A. E. Davidson, R.E., a former Secretary of
the Mechanical Transport Committee of the War Office, has given the
following very definite opinion in a paper read by him at the Imperial
Motor Transport Conference, in his official capacity as representative
of the War Department:

     “Emphasis must be laid on the necessity for obtaining the
    transport immediately. The army which can mobilise in the shortest
    space of time gains an immense advantage by being able to take
    the initiative before the opposing armies are prepared, and the
    army which mobilises most rapidly will be able to gain a decisive
    advantage. This question has now been so carefully worked out in
    detail that the complete mobilisation of an army can be arranged
    for within a period that is reckoned in hours.”

[Illustration: _“The Autocar” photograph._

A FRENCH MOTOR WORKSHOP, ESPECIALLY EQUIPPED FOR THE SERVICE OF THE
FLYING CORPS.]

[Illustration: A GERMAN WORKSHOP CAR, CLOSED TO TAKE THE ROAD.]

[Illustration: TWO VIEWS OF PARTS OF THE BRITISH “KARRIER” SUBVENTION
TYPE LORRY, INDICATING THE EFFORTS MADE TO FACILITATE INSPECTION AND
REPLACEMENTS.

THE ENGINE VALVES WITH INSPECTION COVER REMOVED.]

[Illustration: HOW THE AXLE SHAFTS CAN BE WITHDRAWN. BY REMOVING THE
TOP AND BOTTOM CASING, THE WHOLE OF THE FINAL GEARING, INCLUDING THE
DIFFERENTIAL, CAN BE REMOVED WITHOUT JACKING UP THE VEHICLE OR TAKING
OFF THE LOAD.]

To meet this requirement, the additional motor transport columns must
also be capable of being mobilised with similar rapidity, and we are
forced back either on to the last resort of commandeering any vehicle
that comes handy, or else on to the preparation of a scheme which will
provide a substantial reserve of vehicles of approved make and type,
able to be made available at any instant at a few hours’ notice. Such a
scheme evidently involves the payment to the owners of these vehicles
of some sum intended to make up to them such loss as may result from
their liability to have their vehicles immediately commandeered. These
payments, moreover, must provide that the War Department shall have
the right of periodical inspection of the cars, so that they may
be well informed as to their condition, and may have certain knowledge
as to whether they are being properly driven and maintained in such
a way as to make them useful units of a fleet on active service. A
scheme of this sort is called a “subvention” or “subsidy” scheme, and
it is very generally admitted that such a scheme forms an essential
part of the organisation of transport and supply in every country
in which the civilian use of heavy motor vehicles is sufficiently
extensive to make the principle of subsidy applicable on a working
scale. Clearly, the amount of the subsidy which is offered to owners of
motor vehicles of a suitable type must depend, in the first instance,
on the conditions accompanying the payments. If--as to some extent in
the case of Great Britain--the subsidy scheme applies only to vehicles
of types which would not be employed for trade purposes were definite
encouragement not offered by the Government, the payments must be more
than sufficient to balance any disadvantages resulting from the use of
the subsidy type vehicle, as well as the inconvenience of undergoing
inspections.

Again, if the War Department makes various stipulations as to features
to be embodied in the design of subsidy vehicles, it is more or less
certain that these stipulations will entail manufacturing expenditure
resulting in an increase in the sale price of the machines as compared
with the price of ordinary models of similar carrying capacity.
Thus, the subsidy must also be sufficient to cover any increase in
first cost to the user. If this increase is, let us say, £50, and
the inconveniences entailed by adopting the type result in a loss of
efficiency estimated, let us say, at £30 a year, the subsidy, if it
is to form any real inducement, must evidently amount to a payment on
purchase of about £60 at the least, followed by a payment of, let us
say, £40 a year for three or four years.

In countries where heavy motor vehicles are not--unless some abnormal
encouragement is given--sufficiently extensively used for trade
purposes, the subsidy must of course be considerably higher. If the
conditions of service are so unfavourable to the use of mechanical
transport as to convince the trader that in changing, let us say, from
twenty horses and five waggons to a couple of 3-ton motor lorries, his
expenses will be increased by £100 or £200 a year, the scheme must take
this prospective loss into account. In that case, the scheme becomes
something more than mere subsidy, and partakes more of the nature of a
scheme designed artificially to encourage the use of a particular form
of transport solely on account of its utility to the Government in case
of war.




CHAPTER IX

A COMPARISON OF NATIONAL CONDITIONS

    The Fortunate Position of Great Britain--Causes of Rapid
      Development of Motor Transport--The Big Influence of
      London--Position of the Movement in European Countries.


In considering how far any form of subsidy scheme has been, or could
be, truly successful, we have to take into account first of all the
national and local conditions governing the use of motor vehicles in
ordinary commercial service, and it is satisfactory to be able to
record that a consideration of this subject leads to the conclusion
that the position of Great Britain is peculiarly advantageous, inasmuch
as the number of industrial motor vehicles in service is vastly in
excess of the total requirement of the British Army, a state of affairs
a parallel to which does not exist in any other European country.

The economical use of motors in trade and industry depends in the
first case very largely upon the quality and quantity of the national
roads. Great Britain is fortunate in the possession of the finest road
system in the world. We are not limited as regards motor haulage by
the absence of thoroughfares between our industrial or residential
districts, and it is possible to deliver goods to practically every
house or even cottage in the kingdom, without having to traverse
anything worse than a short distance of rather rough country lane or
private track.

There is no doubt that London has been very largely responsible for
the enormous development of motor transport within the British Isles.
It is generally considered that Paris represents the nearest approach
to London for purposes of comparison. Nevertheless, the population of
Paris is only about half that of London, and the area within which that
population is included is only about one quarter. In other words, the
density of population of Paris is double that of London, which means
that the average distance to be traversed in delivering goods to a
given number of people is much smaller in Paris than it is here. Now
the motor vehicle is able to show superior economy over horse-drawn
traffic mainly where it is able to make use of its capacity for speed
and its ability to cover, without tiring, long distances in the course
of a day. For house-to-house deliveries the motor is at a disadvantage,
since while it is standing waiting before a door it represents a larger
idle capital than the horsed cart, and the investment of this larger
capital can only be justified if it results in the vehicle performing
in a given time a far larger amount of work than would be possible if
horses were used.

Taking, for example, the case of a 2-ton motor van capable of running
about 100 miles in the course of a working day, the ideal condition is
represented by a run under full load from the warehouse or store to
some point about 50 miles distant. Here the whole load is delivered,
and a complete return load is found. Such conditions are seldom
available in practice, but the nearer it is possible to approach to
them, the more likely is the motor to prove a profitable investment.
On the other hand, supposing the car to be used for house-to-house
work involving, let us say, 100 deliveries in the course of a day with
a total distance covered of only about 10 miles, the motor may cover
this distance in traffic in something like an hour, whereas a horsed
vehicle might take two hours. The saving in that case is comparatively
small, and represents, let us say, only an additional 10 deliveries or
an advantage of 10 per cent. extra in the work done in the day. On the
other hand, the cost of the motor is very much higher, and it is more
than likely that on economical grounds the operating concern would not
be justified in adopting mechanical transport.

Applying these examples of extreme cases to the general proposition,
it is quite evident that both the larger population and the more
scattered distribution of population of London make the metropolis a
far more favourable nursery for the industrial motor than, let us say,
Paris, or, for that matter, Berlin or Vienna. The great London houses
have found that their conditions of delivery into outlying residential
districts have been on the whole very favourable to motor transport,
which they have consequently adopted extensively, favouring as a rule
vans carrying loads varying between 25 cwt. and 3 or 4 tons, according
to the nature of the goods to be handled. By establishing motor
services they have been able in many instances to dispense with local
distributing depôts in the environs of London, and they have found it
possible greatly to extend their areas of direct delivery. One of the
consequences has been that people resident 20, 30 or even 40 miles out
of town are now able to place orders at big London houses, and to have
the goods delivered direct to their own doors the same day, or at the
latest on the following day. This delivery is effected without any
unnecessary handling, and without any of that delay which must result
if the railway is used as an intermediary.

By thus extending their field of operations, the big London houses
have come directly into competition with the larger trading concerns
centred in towns some distance from London. These local concerns have
found that they were losing business to the London houses, and have
been compelled in the interests of self-preservation to endeavour to
retain that business by offering equally good and prompt facilities
for delivery. Even so, some portion of their trade is necessarily lost
to them, and they are compelled to seek new fields. In order to do
this and to resist competition so far as may be, they are practically
forced to adopt motor transport, and in their turn to extend their
area until it embraces other towns and villages at a greater distance
from the metropolis. Thus, the influence of London steadily spreads
outwards encouraging the adoption of motor transport in other towns.
A similar phenomenon takes place in a smaller degree round all of our
very numerous big industrial cities, with the result that the motor van
and the motor lorry have become familiar objects in every part of the
country, and have, so to speak, acted as a moving advertisement of
their own utility.

This process, coupled with the comparative excellence of our roads,
has favoured the general adoption of motor haulage by traders of all
classes throughout the country. The railways have in consequence
felt the effects of the competition of the motor vehicle, and have
retaliated by putting themselves into possession of considerable
fleets, in order to secure the rapid distribution of the goods
entrusted to them for delivery. In some instances, railway companies
have established services in country districts to act as feeders to
their branch or main lines. Simultaneously, the general development,
initiated in the first case by private enterprise, has become so marked
and has proved so conclusively the reliability of the heavy motor, that
Government Departments--notably the General Post Office--have been
impressed with the great possibilities of the new transport, and have
adopted motor vans for long distance services as well as for local
distribution of mails in great cities, as being more direct as well
as more economical than the old arrangements with the railways. This
applies particularly to the carriage of parcels.

Side by side with this development has come the astonishing progress
of the passenger motor vehicle. Here again, London has been the big
moving influence. The greatest city in the world has grown from small
beginnings according to its own sweet will. It has not been laid out,
as have younger towns, with any clear scheme in view for meeting the
growth of traffic requirements. Here again, the nearest parallel is to
be found in Paris. Portions of that city are still similar to, or even
worse than, London in this respect, and are traversed only by narrow
and winding roads laid out on no intelligible scheme. Paris, however,
has the advantage that for the past one hundred years definite methods
of improvement have been pushed. Control has remained vested in the
same departments, and the policy has been continuous. Consequently,
the network of small streets has gradually become subordinate to an
admirable system of main thoroughfares of great width and beauty, at
the intersections of which are wide open places generally utilised for
the erection of some of those fine monuments so dear to the French
nation.

Other and newer towns, of which Berlin may be taken as a fine typical
example, have been from the first almost wholly constructed in
accordance with a definite town planning scheme, and in all their later
developments the tramcar has been regarded as a necessity of passenger
transport, and every provision has been made to ensure that a complete
system of railed traffic should be in every way facilitated, and so
far as possible prevented from injuring the natural and architectural
beauties, which must at all cost be maintained in the interests of
trade prosperity as well as from æsthetic reasons.

In such cases, the motor omnibus has from the start come into direct
competition with the electric tramcar, the latter being generally
supported by enormously influential vested interests, the strength of
which has been such as to cripple, or almost entirely prevent, the
introduction of public service motors. Both in Paris and in London,
conditions for one reason or another have opposed the universal
adoption of railed transport in the streets. So far as London is
concerned, the tramcar is useful and possible in suburban districts,
and as a means of bringing people to within a short distance of
the central areas. Beyond that point, its extension is probably
impracticable, and is certainly open to very grave opposition, which
has up to the present prevented the completion of anything approaching
a comprehensive tramway system from north to south or from east to
west. The central area a few years ago was served by the horsed
omnibus, and it was with this vehicle and not with the tramcar that the
motor omnibus in its early stages had to compete. Consequently, it was
given a good opportunity of proving its desirable qualities and was not
hopelessly handicapped by being set, when in very early and imperfect
stages of development, directly against a more or less perfected system
of passenger transport on rails. London has thus proved to be the
world’s biggest nursery of the motor omnibus. Its early imperfections
caused plenty of grumbling and a certain amount of inconvenience, but
it was realised all along that it was only a matter of time before it
would oust the horse omnibus from the streets.

In Paris also, the motor omnibus has been given fair chances, and has
proved its worth. It has been employed partly on routes involving
narrow roads and stiff gradients where trams would be dangerous, and
partly on other routes the natural beauties of which are so pronounced
that it was generally felt that the laying of tramway lines, or the
creation of any permanent blots such as are occasioned by the erection
of standards and wires, would be altogether a desecration.

The motor omnibus, after passing through its novitiate, has proved
in the most practicable possible way the advantages of road motor
transport to the general public. It has hit the short distance traffic
of railways very hard, and has compelled these latter in self-defence
to inaugurate motor services of their own, especially in country
districts not well fed by the railways themselves. It is impossible to
say to what a great extent the development of motor transport is the
result of the anomaly under which the road passenger traffic of London
is not controlled by the local governing authorities, who possess
in other cities licensing powers reserved in London to the Chief
Commissioner of Police. Other great British cities, as, for example,
Manchester and Birmingham, for many years refused to allow the motor
omnibus to prove its worth for fear of competition with tramways owned
by the municipality, which, being itself the licensing authority, could
refuse to give facilities under which any competition with its own
concern could come into existence. It is only lately that the prolonged
experience of London has proved to all these authorities the enormous
utility of the motor omnibus, and its spread to every great city has
either become an accomplished fact, or an inevitable development of
the near future. Paris, with its smaller fleet of motor omnibuses, has
not exerted a similar influence in anything approaching a similar
degree throughout the provincial towns in France; and Berlin, which has
very nearly tabooed the motor omnibus altogether, has done practically
nothing towards the encouragement of motor transport in Germany. It
is rather a curious fact that this policy should have been maintained
in Berlin for so long, seeing that for many years past the German
Government have been paying huge sums in the shape of subvention, with
the sole object of encouraging the national use of trade motors. In
all probabilities, the process of ocular demonstration on the streets
of the capital would have been more effective than the whole of the
expenditure that has been incurred.

From what has gone before, it will be seen that circumstances have
all worked together to cause the development of motor transport of
Great Britain to be far more rapid than in other countries. Added to
this is the undoubted fact that the genius of the British engineer
is best expressed in something substantial and durable. The heavy
industrial motor is more typical of British tendencies than is the
light fast car. As regards the latter we may be good imitators, and
may be well able to keep on equality with competition. As regards the
former we can do more, and we have shown ourselves able to lead the
world and to produce finer industrial motors than can be obtained
in any other country. Even the progressive engineers of the United
States acknowledge that they must draw their inspirations in this
movement from Great Britain, and are not infrequently to be found in
this country studying what has been done, and learning lessons which
they will apply at home and which may serve to bring them into strong
competition with us, but are very unlikely to make them our superiors.

The general result of all these influences has been, as already stated,
that Great Britain is the only European country in which the industrial
motor is, in times of peace, used in numbers greatly in excess of the
possible military requirements of our own forces. Consequently, the
problem before our War Office has not been to encourage the use of
heavy transport, but to direct the tendencies of design and popular
taste into the channels in which they could be made to fall into line
most completely with military requirements.

Next after ourselves, France is fortunate in the possession of the best
road system of any European country, and this has helped industrial
motor development to progress with fair rapidity, though not at such
a speed as to enable the country to be self-supported as regards its
needs in military motor transport. Germany, with a less complete road
system, involving in many parts very severe gradients, has had more
difficulty still in filling its military requirements, while Austria
is in a position somewhat akin to that of Germany from this point of
view. Consequently, in all these three great countries, the subvention
scheme has had to be rather a scheme for encouraging the use of
motor transport of any kind than an attempt to direct designers into
any particular channels. Conditions in Belgium are somewhat akin to
those obtaining in France; and, generally speaking, other European
countries are so badly served by roads and so unfavourably situated
as regards their requirements for the haulage and delivery of goods,
that the development of industrial motor transport on a large scale has
been out of the question, and consequently, the establishment of any
subvention scheme would have proved futile. We are now in a position,
with a fairly clear conception of the conditions obtaining in each
of the countries concerned, to consider in more detail the nature of
the schemes evolved in the interests of military transport, and to
ascertain how far those schemes have been brought to successful issues.




CHAPTER X

BRITISH SUBSIDY TYPE MOTORS

    Early Subsidy Schemes--Amount of Subsidy--Standardisation of
      Driving Control--Important Mechanical Features--Provision for
      Working in Convoy--The “Chain-Drive” Controversy--The Present
      Position.


In view of the peculiarly advantageous circumstances detailed in the
previous chapter, the British subsidy scheme is from the pecuniary
point of view less imposing, and from the practical point of view far
more comprehensive than any other scheme yet attempted elsewhere. The
British War Department favours in general the use of vehicles intended
to carry in active service useful loads of about three tons, but as the
machines also have to take four men and a considerable quantity of kit
and stores, they correspond to ordinary industrial vehicles of four
tons capacity. There is a parallel, but much smaller, requirement, for
30-cwt. vehicles corresponding to the ordinary commercial two tonner
and capable of higher speeds than are desirable with heavier machines.
These lighter cars are intended for use behind mobile and fast-moving
troops, while the heavier type are for the service of the infantry,
and for the carriage of ammunition supplies. In each case, the total
amount of subsidy paid is from £110 to £120. A portion of this takes
the form of a cash payment when the vehicle is accepted, while the
remainder is an annual payment spread over a period of three years,
and is conditional on periodical inspection revealing the satisfactory
condition of the machine subsidised.

The agreement also provides that in the event of the War Department
desiring to commandeer the vehicle in time of war, a very liberal price
shall be paid to the owner, this price being of course dependent to
some extent on the age of the machine. The correct amount is arrived
at by deducting from the first cost of the car a certain regular
percentage intended to represent the depreciation in value during every
half year of normal service. After this deduction has been made, the
resulting figure is increased by an agreed percentage of itself, the
final result being that the price paid for any car under about two
years old is very near, if not quite equal, to the original first cost.

The first move made in the direction of a subsidy scheme in Great
Britain dates back to 1908. At that time it was believed that light
steam tractors best filled military requirements, and a number of these
were registered, the nominal payment of £2 per annum being made to
their owners. The industrial petrol vehicle was at this time passing
out of its period of probation, and it was not long before the military
authorities came to the conclusion that it represented a more useful
type, in view of the nature of the particular emergencies against which
they were chiefly called upon to guard. The ordinary steam-propelled
vehicle was open to certain objections, the principle of which is the
fact that its carrying capacity of fuel and water is limited, and the
latter must necessarily be replenished at fairly frequent intervals.
This is no great drawback in commercial work, but might be a very
serious matter indeed in times of war, when men and horses have a prior
claim on a possibly limited water supply.

In 1911 a scheme of subsidy for transport and supply motors was
authorised, and in view of the fact that no such scheme could be
matured until it had been in force for several years, a provisional
scheme was temporarily adopted, under which total sums ranging from
£38 to £52 were paid to owners of vehicles generally of about three
tons capacity. The present subsidy scheme was finally put into force
in 1912 after the War Office experts had conferred on many occasions
with representatives of the leading manufacturing interests. The
main objects of the scheme are plainly stated in the War Office
specification as follows:

    (1) To make the manipulation and control of all vehicles the same:
    and

    (2) To minimise the number of spare parts which must be carried
    in the field, having regard to the number of different makes of
    vehicles of which the transport columns of the army would be
    composed.

As to the first of these two stipulations, one would think that there
could hardly be two opinions, though in point of fact it has been
argued in some quarters that no standardisation of driving control is
in any way essential. Such a view one imagines can only be held by
those who are personally so fortunate as to have that mechanical knack
which allows a man to take charge of any car, however different it
may be from those to which he has been previously accustomed, and to
drive it with perfect safety and efficiency even at night-time and over
unknown and bad roads. We cannot assume that every transport driver
possesses this instinct, and we must therefore agree that the object
aimed at by the War Department is one which deserves every support. It
is, in fact, worth while to go into a little detail on this point, in
order to show what care has been taken to make everything as easy as
possible for the driver.

First, as regards the hand control. The steering wheel must provide for
76° of movement of the front wheels of the car; that is to say, 38° of
lock from the normal position on either side. In reaching the maximum
lock, two complete turns of the steering wheel must be made. Thus, a
driver who is used to getting a certain effect by turning his wheel
through a certain distance, will find if he is put on to a subsidy car
of another make that the same effect is still produced by the same
amount of movement. This, of course, helps to make him immediately a
safe driver in traffic or at sharp corners. The four-speed change-speed
gear is operated by means of a lever in a gate. This gate must be
formed of two slots and two selectors, the reverse being a continuation
of the first speed slot. The hand-brake lever must be arranged to push
on, and must be well away from the change-speed lever and to the
right of it. Furthermore, the brake lever must be 6 inches longer,
and must have a plain cylindrical handle, whereas the change-speed
lever is finished off with a circular knob. Thus, any confusion
between these two levers in sudden emergency, or in the dark, is
completely prevented. The throttle and ignition levers must be placed
underneath and to the right of the steering wheel, their movement being
independent of any movement of the steering column. Increased engine
speed must in each case be produced by moving the levers forward. The
total movement must be 90°, and the handles must be at right angles
to the main axis of the vehicle when in the centre of their travel.
This, again, means that a driver who is accustomed to producing a
given effect upon his engine by a given movement of the ignition or
throttle is in no way confused by being put on to a different car. The
only point in which any variation in driving control is permitted is
in regard to the accelerator pedal, the presence of which is optional.
However, if a foot accelerator is provided, it must be so combined
with the hand throttle that on releasing the accelerator, the throttle
valve returns to the position set by the hand lever. The clutch- and
foot-brake pedals are marked C. and B. respectively, and the clutch
pedal must be on the left, the brake pedal on the right, the travel
in each case being about 3-1/2 inches. From these details it will
be seen how carefully every item in the driving control has been
considered, with a view to facilitating the work of drivers who may
have to be moved frequently from one machine to another. As to whether
the arrangement adopted is a good one, we have strong evidence of a
favourable character in the statement of Mr. J. E. Thornycroft, to the
effect that his firm has adopted the War Office system of control as a
standard for all their vehicles whether of subsidy model or not.

At a rather later date it may be well worth while to consider whether
experience has not demonstrated the possibility of combining with
a full subsidy scheme a modified scheme insisting only on vehicles
of the right load capacity, and the adoption of all details of the
standard control. In this way, if the true subsidy model continues to
be somewhat unpopular amongst commercial users, a very fairly adequate
reserve will be brought into being without any considerable trouble.

Turning now to the second of the two main points which the War
Department have attempted to cover, we find the fulfilment of the
scheme beset with much greater difficulties. Every attempt to
secure standardisation in the parts of vehicles of different makes
must necessarily entail expenditure by the manufacturers, both in
getting out new designs and also in arranging for economical workshop
processes. Such expenditure is only justified commercially if the sale
of the resulting products is sufficient to secure profitable trading.
Some allowance, of course, must be made for the added prestige accruing
to a firm licensed to build for the War Office, but this is in itself
insufficient encouragement. Consequently, every stipulation made with a
view to standardisation must be covered by subsidy grants, sufficient
to cover both manufacturer and user in respect of any additional cost
or disadvantages in operation. The War Department has recognised the
impossibility of asking for complete standardisation, and so killing
the individuality of different types of vehicle. The tendency of any
such process would be to throttle normal competition, and to prevent
progress. Consequently, standardisation can only be attempted in
certain respects where it appears for one reason or another to be
highly desirable. For example, radiators are notably liable to damage,
and consequently the connections of the radiator to the machine are
standardised to allow of ready replacement of this essential as a
whole. The radiators are mounted on trunnions, the bearings of which
are in halves, and the positions and dimensions of inlet and outlet
connections are definitely fixed. As an additional safeguard, a stout
cord in the form of a bar or tube has to be placed across the front
of the radiator. The engines are not standardised except in certain
details, as, for example, the method of fastening on and driving
the magnetos, which are arranged with a view to rapid removal and
replacement whole. No high degree of standardisation is possible
without incurring great expense and other disadvantages so far as
concerns the design of the clutch and the change-speed gear, though the
ratios of the latter are determined for reasons which will be explained
later.

As regards final drive, some degree of standardisation is possible so
far as the axle arms and bushes are concerned. The bearings of the
front wheels are standardised so as to make the wheels interchangeable,
and the diameters of both the front and rear wheels are fixed. Makers
and users are left with a fairly free hand as regards the type of
body to be fitted to a subsidy machine, but in general this must be a
lorry body with detachable sides and ends at least two feet high, and
carrying a frame to take a complete overhead cover. Use is also found
for a certain number of box vans.

Many provisions are made in the subsidy scheme with a special view
to making the vehicles suitable for working in convoy. In the first
place, it is stipulated that all engines shall be fitted with governors
which shall automatically control their speed to 1,000 revolutions
per minute. This prevents drivers from racing their engines and
over-speeding their cars when running light, and also determines a
definite top speed, which should be almost exactly equal for all
subsidy machines, as it were. Coupled with this, are the provisions as
regards change-speed gear. The engine governor having limited the speed
of the 3-ton type to 16 miles per hour, and the speed of the 30-cwt.
type to 20 miles per hour, the next step is to secure that, when road
conditions require the use of lower gears, the whole of the vehicles
in the convoy will need to change gear practically at the same point,
and when on lower gears will run practically at the same speeds. With
this object in view, it is stipulated that the ratio between top and
bottom gear shall be about five to one, giving bottom speeds of about
three and four miles per hour respectively for the two classes. This
provision, taken in conjunction with the stipulated engine dimensions,
should secure that every vehicle is capable of tackling a gradient of
one in six on normal road surface whether fully loaded or empty. It
will be seen that the point kept in view has been the desirability of
arranging everything so that drivers of a large number of machines
working in convoy--that is to say, running along the same road at short
intervals behind one another--shall not have any difficulty in keeping
their proper distance, and so in preventing any risk of collisions.
For similar reasons the fitting of a ground sprag is made compulsory.
The object of the sprag is to prevent a vehicle from running backward
if, for example, the driver misses his gear on a steep gradient. The
fitting is not necessary on ordinary commercial machines, since the
brakes will almost certainly hold the car before it has run more than
a few yards, but in the case of one vehicle in a long convoy, those
few yards may mean a collision with the following car, and a hopeless
muddle causing serious delay. In view of the possibility of some one
car failing, every chassis has to be fitted with towing hooks at
the ends of the side members of the frame fore and aft, so that each
machine can, if required, either be towed or tow another, and any “lame
duck” will not obstruct the whole convoy, but can be dragged along to
some point where it can be towed out of the way, and the rest can be
allowed to proceed.

Other points in the scheme have in view particularly the fact that
the machines must be required to operate on very bad and possibly
on very hilly roads, and may even be needed to make detours across
country when roads are destroyed or entirely blocked. The chief points
in this connection are the provision of high ground clearance, which
must in no case be less than 12 inches, the stipulation as to size
of wheels, which are rather larger than usually found on commercial
vehicles, and various stipulations ensuring adequate protection of
the mechanism from mud and dust. Generally speaking, these provisions
constitute one of the principal difficulties in making the scheme
successful, principally for the reason that large wheels imply
a rather high loading platform, and this is objected to by many
commercial concerns on the ground that it makes loading and unloading
more difficult. On the other hand, larger wheels are better for the
roads, and consequently it is within the bounds of possibility that
trading concerns will presently be compelled by law to put up with
their disadvantages. From the point of view of motor users in the
Colonies, high ground clearance and large wheels are very valuable and
even essential features. Consequently, the subsidy scheme works in well
with many Colonial requirements, and manufacturers whose machines are
accepted by the War Department probably depend for success of the type
to no little extent upon the likelihood of orders from abroad.

[Illustration: _“The Autocar” photograph._

SAURER MOTOR FIELD KITCHEN.]

In the interests of the proper guarding of the mechanism from mud and
dust, the experts of the War Department have considered it necessary to
stipulate that all cars must be driven on the live-axle system either
through bevel or worm gear. In the first instance, only the former was
accepted, but more recently the worm drive, as exemplified by the
products of its pioneers, Messrs. Dennis Bros., has met with approval.
The point to be noted is that the chain drive--which, in the opinion
of many manufacturers and also of many users both at home and abroad
is unequalled for heavy work--is definitely debarred. Considerable
antagonism to the scheme as a whole was thus created, and one is forced
to the conclusion that the reasons for refusing the chain drive were
not solely connected with the question of protecting from mud, but were
possibly more concerned with the fact that, while a chain drive is
probably quite as economical in maintenance as any live-axle drive, it
requires rather more frequent adjustments and attention, and possibly
small renewals.

In commercial practice, this consideration carries little or no weight,
but on active service a breakdown is none the less serious because it
is caused merely by a breakage to one link in a chain and not by the
entire dislocation of the whole transmission gear. One gathers that, at
the present moment, chain-driven machines are in fact being used in
the service of the British Army, and in the interests of the success
of the scheme in the future, the hope may be expressed that practical
experience will show that any fears as to unreliability of this type
of transmission from the military standpoint will prove entirely
groundless.

When the great war broke out, the position was briefly as follows.
The subsidy scheme had not been in force for a sufficient length of
time to secure through its means the complete fleet required. The War
Department itself owned upwards of one hundred lorries of subsidy
type, and a limited number were to be found in civilian service. Among
these, the Leyland, which was the first type accepted for subsidy,
was probably the most prominent. It has been necessary, therefore, to
depend not only on a supply of machines of subsidy type, but in the
first instance on the requisitioning of other cars of suitable load
capacity, and more lately on the steady purchase of new vehicles of
types which, if not according exactly to the subsidy requirement,
approximate to it sufficiently to secure reasonable facilities for the
easy repair and maintenance of the machines of the transport and supply
columns, both of our existing Expeditionary Force, and also of the new
armies now in course of formation.




CHAPTER XI

TRANSPORT MOTORS OF CONTINENTAL ARMIES.

    The French Scheme--Notes on French Vehicles--Benzol and Alcohol
      Fuels--The German Scheme, Difficulties and Results--Austria,
      Italy, and Russia.


The French subvention scheme, for reasons already explained, has to be
more comprehensive in its financial clauses than that in force in Great
Britain. Without going into details, it may be summed up in the general
statement that the subsidy paid in respect of a lorry of about 3 tons
capacity aggregates about £300, spread over a period of four years.

[Illustration: _“The Autocar” photograph._

A GERMAN LORRY FITTED FOR REPLENISHING THE SUPPLY OF GAS IN THE
ENVELOPES OF ZEPPELIN AIRSHIPS.]

The French Government have specialised for many years past in machines
of this and somewhat lighter load-carrying capacity, and more recently
they have made serious and fairly successful efforts to encourage
the employment of powerful vehicles in which the engine power is
arranged to drive all four wheels, and which can be used either as
lorries or as tractors, or as a combination of the two. The genius of
the French motor engineer, in the opinion of the writer, expresses
itself better in the high-speed touring car than in the industrial
vehicle. There are excellent examples of the latter to be found, but
the average quality of the products of well-known manufacturers is
almost certainly not equal to that of corresponding British firms. A
comparison of the relative importance of the two national industries
was possible to those who had opportunities of visiting the Industrial
Vehicle Show in London in 1913, and subsequently of inspecting the
exhibits in the annexe of the Paris Motor Salon later in the same
year. On these occasions, as well as during previous opportunities
of watching French subvention vehicles undergoing trial, the writer
formed the opinion that in many cases the various features of design
in any particular vehicle of French origin are peculiarly unequal. In
some portions of the chassis we find adequate or even unnecessarily
great strength; in others, unduly light construction and a certain
disregard of details making for safety in operation. In many instances
the steering mechanism is unnecessarily exposed, and placed very far
forward so as to be liable to injury in the case of slight collision
or passage over any considerable obstruction. The chain drive is
very popular among French manufacturers. The chains are usually not
protected by cases, and in very many instances an attempt is made to
obtain through the medium of the chain a very large gear reduction,
resulting in the use of absurdly small chain pinions, which will
certainly need frequent renewal under the conditions of rough service.
In some instances again, the chains themselves are too light for
durability. There is also a certain disregard for accessibility of
the engine and clutch, and a tendency to employ pneumatic tyres on
vehicles designed for heavy loads which would be carried with far less
risk of roadside trouble on a rather more substantially constructed
solid-tyred vehicle with a good springing system.

Admitting that the French Government could not stipulate any degree of
standardisation until they had first obtained a numerically adequate
supply of vehicles, one would have thought it possible at least to do
something towards standardising the driving control. In some French
subvention models, the hand-brake lever is nearer to the driver
than the change-speed lever; in others, the opposite arrangement
is adopted. Frequently, both levers are of equal length and almost
indistinguishable to the touch, which must make it far more dangerous
to put a new driver on to a subsidy car when required urgently for
night work.

The French subvention trials have been held annually, usually in the
months of August and September, and have not been as a rule of a
very arduous character. While accompanying the competing vehicles,
the writer has been forced to the conclusion that the object of the
authorities was rather to pass for subvention any reasonably efficient
machine, than to weed out a considerable number and depend only on the
most durable. As a rule, during these trials, the competing lorries
are parked at Versailles, from which centre they run out daily over a
limited number of routes, generally of a very easy character so far
as gradients are concerned. An interesting and potentially valuable
feature of the annual French trials has been the compulsory use on
all the cars of a variety of fuels. On some days petrol has been
used, on others benzol, and on others again a half-and-half mixture
of benzol with denatured alcohol, which latter for practical purposes
may be regarded as the same thing as methylated spirits. In this way,
the French Government have endeavoured to make themselves at least
partially independent of any temporary stoppage in the imports of
petrol, though so far as we can see at present no such stoppage is in
the least degree likely during the present war. Benzol can of course be
produced in limited quantities in this country and in France, and if
the emergency arose, the supplies of benzol could be greatly increased
at the expense of simultaneously laying up stocks of other products not
at the moment marketable.

As regards alcohol, a considerable quantity of beet is grown in
France, from which either sugar or alcohol can be produced. As a
rule, this beet is used mainly for sugar manufacture, since this is
the more profitable method of employing it, but in emergency it could
be utilised for the production of a very fair quantity of commercial
alcohol, thus, roughly speaking, doubling the available stock of
home-produced fuel.

The results of these tests have been, on the whole, very interesting.
In almost every case, benzol has given better results than petrol,
while the benzol-alcohol mixture has given results in some cases rather
better than petrol, and in other cases not quite so good. On the
average, the mixture has shown itself approximately equal to petrol,
so far as consumption is concerned. By visiting Versailles in the
early hours of September mornings when the temperature was fairly
low, the writer satisfied himself that the use either of benzol or of
the mixture did not constitute any serious difficulty in the way of
starting up the engines. Moreover, the general absence of offensive
smell or smoke seemed to indicate satisfactory combustion of the fuels.

As to the results of the French subvention scheme, the fact that
the regulations have recently been made more severe, and certain
restrictions as to horse-power, weight, etc., introduced, seems
to indicate that the number of vehicles available at the time of
the outbreak of war must have been at least approaching the number
estimated as required. The last series of trials were only just over
when war broke out. In these trials some sixty vehicles competed,
representative of a considerable variety of makes and types, including
a small number of Colonial lorries of special design, and one or two
tractors. On the whole, the vehicles went through the trials well, and
the opinions of experts who were present were all to the effect that
great improvement was noticeable in mechanical details as compared
with previous years.

Incidentally, it may be mentioned that the big fleet of the Paris
General Omnibus Co. forms a very useful and conveniently concentrated
supply of substantial cars available either for the rapid carriage
of troops, or--by the substitution or conversion of bodies--for the
transport columns. Numerically, however, the Paris omnibus fleet falls
very far short of that of London, while from the mechanical point of
view the vehicles are of heavier construction, and one would imagine
less easily handled on narrow and winding country roads.

In Germany, a motor transport subvention scheme was inaugurated in
1908. At that time a limited number of German manufacturers were
producing considerable quantities of heavy motor vehicles, more
especially for export, but it was becoming evident that some very
substantial encouragement would be needed to make the home market
sufficiently active to be of any real utility to the War Office.
Consequently, a scheme was got out which was openly stated to be
“a scheme for popularising the use of mechanical transport,” or,
in other words, a scheme for persuading business houses to adopt a
species of transport which, without Government aid, would represent
an uneconomical and consequently undesirable feature of an industrial
concern. The German Government decided in favour of heavy motor
lorries, capable of carrying 4 tons and hauling an additional 2 tons
on a trailer. These trailers, contrary to usual commercial practice,
are fitted with rubber tyres, since this addition is found to ease the
work of hauling by some 25 to 30 per cent. The total subsidy for a
subvention train consisting of a power lorry and rubber-tyred trailer
amounts to something in the neighbourhood of £450 spread over a period
of five years. The choice of a heavy type of vehicle was probably
justified by the need for limiting the length of the transport columns
destined to accompany enormous armies. At the same time consideration
has evidently shown that there are grave disadvantages to the use of
such heavy cars, and recent regulations have provided more stringent
stipulations as to maximum weight.

When the subvention scheme had been in operation for five years,
figures were got out indicative of its results up to the end of March,
1913. During this period 825 army trains were subsidised, namely,
743 in Prussia and the other states whose armies are under Prussian
control, and 82 in Bavaria. In addition, some 400 lorries of very
similar types were sold in Germany outside the scheme, making about
1,200 trains available for use at that time. Allowing for increase in
the interval which has since elapsed, we may perhaps put the total
available at the outbreak of war at about 1,600. Captain Davidson
estimates that the German Army requires for transport purposes about
2,000 of its trains, but this figure presumably does not take into
account the needs of the whole of the Landwehr and Landsturm. It is
admitted that the normal British Expeditionary Force requires about
1,000 3-ton vehicles, which would correspond in capacity to about 500
of the German trains. Consequently, 2,000 of the German trains would
apparently only be about sufficient for an army four times the size
of our Expeditionary Force. Similarly, the estimate that France needs
about 5,000 vehicles of the 3-ton type apparently does not take into
account the complete mobilisation of reserves.

The manufacturing concerns which have figured most largely in the
German scheme are the German Daimler, the Büssing, the N.A.G., and
the Gaggenau. These four have all been participating in the scheme
from the start, and about ten other manufacturers have more recently
fallen into line, while in Bavaria only three manufacturers have been
building to official requirements. The states the industries of which
have enabled the strongest support to be given to the scheme are
Brandenburg, Saxony, the Rhine Province, Würtemberg, Westphalia, Baden,
and Alsace-Lorraine. No less than 41 per cent. of the total machines
enrolled are normally used in the brewing trade. In this connection,
an official report from Bavaria is rather instructive and amusing:

    “There are so many breweries in Bavaria, and these are so densely
    distributed, that there is no need anywhere to convey beer for long
    distances. Hence there are practically no vehicles employed.”

This seriously expressed implication that beer is the only really
essential commodity seems to show that lack of humour which appears to
be a national characteristic of the German race.

Next after the brewing interests, but far behind in their practical
support of the Government scheme, come concerns engaged in the
transport of goods for export, followed by those concerned in brick
transport, flour manufacture, carriage of building materials,
agricultural work, and haulage of iron and steel goods.

In endeavouring to estimate how far the existing fleet meets the
requirements of the Germany Army, we have to remember that it consists,
at least partially, of machines that have been in service for several
years, and that consequently may not be equal to any long strain
under peculiarly difficult conditions. It must be presumed that the
German Government has made provision for the continued manufacture of
considerable numbers of heavy motor lorries throughout the war, and
has not permitted the leading motor works engaged in this class of
production to be too far denuded by the mobilisation of their men.

The Austrian subsidy scheme is along the same lines as that in force
in Germany, but favours a lorry of slightly lower carrying capacity,
probably in view of the mountainous nature of many of the frontier
roads. The total amount of subsidy payable is in the neighbourhood of
£360 spread over a period of five years. The scheme was inaugurated
some time after those of France and Germany, the first trials being
held towards the end of 1911. Certain parts of Austria are well
provided with roads, so that there is a fair field for the commercial
use of motor transport. A large number of vehicles, not of subsidy
type, but no doubt capable of being made useful for light work in time
of war, are used for the carriage of mails in Hungary. In the Austrian
Tyrol, there are numbers of motor services for the carriage of mails
and passengers, but on the whole Austria is probably not very well
provided with mechanical transport. Her manufacturing industry is
limited, and she imports in fair numbers from her neighbour, Germany.

Italy can only find very small use for heavy motor vehicles in
commercial service, and consequently it would be futile as yet for
the Government to depend upon anything in the nature of a subsidy
scheme. During the Tripoli campaign, a considerable number of rather
lightly built lorries were obtained by direct purchase and proved very
serviceable. Probably they are not of a type which would be by any
means ideal in a European war, though they were doubtless the right
thing for work over loose sandy tracks where heavier machines might
well have become inoperative.

Russia also has no subsidy scheme on account of its comparatively
poor industrial development, and also the very inadequate quantity
and quality of its roads. For such vehicles as are used, the country
is dependent upon import, while the army must depend solely on direct
purchase from foreign manufacturers. It is rather interesting to note
that out of about 2,000 industrial motor vehicles exported by Germany
during the year 1913, no less than 25 per cent. went to Russia,
practically the whole of these being known to represent Government
orders. Russia has been buying motor lorries for military use from
British firms for many years past. An engineer who accompanied one of
the first vehicles supplied from this country, describes the roads over
which the car had to work during its official trials as follows:

     “The road was covered with fine sand, banked up a few feet above
    the level of the surrounding country, in which the wheels of
    peasants’ carts had cut ruts about 12 ins. to 14 ins. in depth. The
    gauge of these ruts being narrow, it was necessary to drive with
    one pair of wheels in the ruts, the other pair meanwhile cutting
    ruts of their own. At intervals planked bridges had to be crossed.
    These were old and unsafe; therefore, it became necessary to lay
    down a temporary track of boards to distribute the weight over as
    many planks as possible.”

At first sight it would appear that under such conditions the purchase
of motor lorries by the Russian Government represents a waste of money,
but the facts are explained by a credible story circulated within a few
days of the outbreak of war to the effect that the Austrian military
attaché a day or so before leaving Petrograd expressed surprise that
so many motors were being mobilised. “Your roads are so bad,” he said.
“Yes,” was the reply, “but yours are good.”




CHAPTER XII

EMERGENCY MEASURES ON THE OUTBREAK OF WAR

    The Work of the Chief Motoring Organisations--The Requisitioning
      of Vehicles by the War Department--Arrangements for subsequent
      Supplies.


A noteworthy feature in connection with the mobilisation of the British
Army on the outbreak of war was the energy with which the great
motoring organisations took up the duties of rendering the private
motorist, so far as might be possible, available for the service of
the Government. The Royal Automobile Club sent out a circular letter
to some sixty provincial Automobile Clubs affiliated to the parent
body, asking that their members owning motor cars should be requested
immediately to register them with the R.A.C. for the use of the War
Office and Admiralty, in case they should be wanted. This scheme was
simultaneously furthered by means of advertisements in the principal
organs of the London and provincial press, and posters were placed in
all the Club’s officially appointed hotels and garages throughout the
Kingdom. Registration forms asking for particulars of cars, and an
indication of the nature of the service for which they would be made
available, were rapidly prepared and widely distributed with admirable
results. These machines were placed at the disposal of the War Office
and Admiralty both for home and for foreign service, and in many cases
their owners made their own services available for facilitating or
accelerating the urgent business of the country by providing officials
with a ready means of rapid transit.

In addition, the Club kept in constant communication with the British
Red Cross Society, and has put at the disposal of this Society
the use of the R.A.C. annexe at 83, Pall Mall, for office, store
and organisation purposes. As the need for the provision of motor
ambulances for foreign service became urgent, the Club gave invaluable
assistance to the Society by keeping up a constant flow of cars
unreservedly given or lent by their owners. Cars have been supplied
for this and other purposes through the R.A.C. organisation from
practically every county in England, Scotland, Ireland and Wales. All
members and associates of the Club have been asked to assist the work
of recruiting by carrying on their cars cards urging prompt enlistment.

The Automobile Association and Motor Union was equally energetic and
prompt. The Association immediately conveyed to the War Department an
offer to help the Government to the fullest extent of its resources,
and upon the acceptance of this proposal communicated at once with
its 92,000 members--owners of cars, light cars, cycle-cars and motor
cycles--inviting them to volunteer their services, and to that end to
forward the fullest possible particulars. The response to this appeal
resulted in the enrolment of the names of about 20,000 motor owners,
and within a few days large numbers of these were being utilised not
only by the War Office, but by municipal and other bodies all over
the country. The earliest mobilisation took place on the Doncaster
race-course, where about 150 cars assembled in a few hours in response
to telegrams. This fleet remained concentrated for some days, but
the eventuality against which it was intended to provide did not
materialise. Large numbers of the A.A. members were employed during the
first two or three weeks of the war to guard telephone and telegraph
lines and cables, until permanent arrangements could be made for this
service. Hundreds of motor cycle and car members undertook long spells
of duty by day or night under the supervision of the post office
officials.

In connection with the conveyance of wounded, the Association placed
fleets of cars at the disposal of the chief military centres throughout
the country, its members holding themselves in readiness to go out at
any hour during the day or night, to carry wounded from the railway
stations to the hospitals. In a considerable number of cases, the
motorists so employed undertook, at their own expense, to convert their
cars into ambulances, and a large number of machines so transformed
were sent across the channel to work behind the firing line. Vehicles
were also forthcoming in plentiful numbers to meet refugees, and
take them to their temporary homes. Hundreds of motor cycle members
volunteered for dispatch-carrying work, and the committees of the
National Service League and other recruiting bodies in all parts of
the country were supported by cars, light cars, and motor cycles with
side cars, ready to pick up recruits and convey them to the enlisting
depôts. At normal times, the Association employs on the main roads
of the country over 500 road patrols, whose duties involve continual
cycling over their appointed beats from daybreak until dark. These men
were evidently ideal recruits for the cyclists’ battalions required
for scouting work. Over 250 of them enlisted in various regiments, or
rejoined their old regiments, while a picked body, over 100 strong,
was formed into the first two companies of the 8th Essex (Cyclist)
Battalion, under the command of the Secretary of the Association,
Captain Stenson Cooke, who was formerly a member of the London Rifle
Brigade.

[Illustration: _“The Autocar” photograph._

ONE OF A FLEET OF “MAUDSLAY” MOTOR BUSES COMMANDEERED BY THE WAR
DEPARTMENT, AND FITTED WITH LORRY BODIES.]

The Commercial Motor Users’ Association undertook the enrolment of
men competent to serve as motor transport drivers, and also formed
on behalf of its own members a kind of transport exchange. A similar
scheme on rather broader lines was handled by the Imperial Motor
Transport Council, the idea being that while some business concerns
would experience difficulties in effecting deliveries owing to their
horses being requisitioned, others--owing to the disorganisation of
trade--would have suitable facilities standing idle. In that event
considerable trouble might be saved by bringing into existence some
machinery capable of establishing contact between the two groups.

The Council also undertook work in assistance of the Motor Ambulance
Department of the British Red Cross Society, and circularised its
oversea members with a view to assisting the maintenance of British
export trade in motor vehicles.

At the outbreak of war, steps were immediately taken by the War
Department to secure for service all the motor lorries of subvention
type working for commercial houses. These not being numerically
sufficient for the whole needs of the army, several thousand other
motor lorries of approximately the same carrying capacity, but of
varying types, were requisitioned somewhat hastily. The quality of
the fleets thus formed was variable, even though a process of weeding
out at the ports of embarkation did something towards securing
uniformity. In the same way the urgent need of employing many thousands
of transport drivers naturally led to the enlistment of men of
varying capabilities. Drivers handling lorries or ’buses are in some
instances required to be fairly capable mechanics. In others, any
interference with the mechanism of their machines is discouraged, and
they are taught to be entirely dependent on the mechanical staff at
their headquarters. Such men, while thoroughly skilled in handling a
vehicle, are not really fully qualified for the business of a motor
transport driver in active service.

Very considerable numbers of London motor omnibuses were taken off the
streets and converted into ambulances or lorries, and similar vehicles
have also been used for the transport of troops and other purposes.

As soon as matters had had a little time in which to settle down, it
became apparent that the Government did not intend to rely on the
system of requisitioning to make up the wastage of their fleet in
service, or to provide transport for Indian and Colonial troops, or for
the new armies in course of formation. For this purpose large regular
orders were placed with many of the leading manufacturers, and in some
instances these orders amounted to taking over practically the entire
output. No exact figures are available as to the rate at which, during
the early stages of the war, the Government took delivery of new motor
lorries, but there is little doubt that the weekly supply ran into
three figures, and that a continuance of very substantial orders will
be necessary right up to the conclusion of hostilities.

In European countries, the comparative shortage of industrial
motor vehicles rendered necessary a more wholesale programme of
requisitioning. Thus, for example, Paris was promptly denuded of the
whole of its fleet of motor omnibuses, about 1,100 in number. A few
years ago, the old double-deck type of motor omnibus, at one time used
in Paris, was discarded in favour of a long-bodied single-decker,
capable of carrying up to about forty passengers. These machines are
so designed to the requirement of the Government as to be capable of
being transformed rapidly into waggons for the carriage of meat. The
windows are replaced by wire-gauze screens, the seats removed and the
handrails fitted with hooks. Alternatively, the ’buses can be equally
easily adapted for the carriage of wounded, by simple fittings from
which stretchers or hammocks can be slung.

[Illustration: _“The Autocar” photograph._

A LARGE NUMBER OF DAIMLER LORRIES HAVE BEEN TAKEN OVER BY THE MILITARY
AUTHORITIES. THE EXAMPLE SHOWN IS FITTED WITH A SHIELD CONTAINING A
SHEET OF TRIPLEX GLASS WHICH WILL STOP A RIFLE BULLED AT 100 YARDS.]

During mobilisation, numbers of motor vehicles were employed in
France to transport troops, and, moreover, those of the Paris ’bus
type are of undoubted utility for this purpose whenever it may become
necessary to transfer moderately large bodies of men rapidly from one
point to another, where convenient railway communication does not exist.

All the Continental countries involved in the war made strict provision
against the export of motor vehicles of any kind, while even in Great
Britain an order was, for a period, in force, prohibiting the export
of heavy industrial vehicles. It was, in fact, realised in advance in
all quarters that a war of such magnitude and involving the employment
of such huge numbers of men, could not conceivably be fought along the
lines anticipated and subsequently realised, unless full dependence
were placed upon motor transport in the first case for the provision
of food supplies, and as a corollary for a similar service of warlike
stores, for the carriage of wounded, for scouting, and for enabling
commanders and staff officers to travel with sufficient rapidity
and freedom to make it possible for them to realise with sufficient
accuracy the essential facts with which they were called upon to deal.


_Wyman & Sons Ltd., Printers, London and Reading._




TRANSCRIBER’S NOTES:

  Text in italics is surrounded by underscores: _italics_.

  Inconsistencies in hyphenation have been standardized.

  Archaic spelling that may have been in use at the time of publication
    has been retained.