scale that administrative arrangements, lines of fire, and trench systems could be shown upon it. Such a map did not exist except in the Belgian area, and elsewhere had to be made. The earlier war surveys were made upon the plane table on a scale of 1/20,000 and were completed up to the British trench lines. The subsequent de- velopment of air-photography, and the discovery of the manuscript sheets of the cadastral communal surveys, made it possible to com- pile reliable maps, not only of territory in British occupation, but of all that portion of north-eastern France occupied by the German armies.
The trench zone was mapped on a scale of 1/10,000, and forward and back areas at 1/20,000. In all 6,000 sq. m. were surveyed.
A reliable map of the physical and artificial features of the country is not sufficient in itself. The positions of defensive works of the enemy batteries, of the points of administrative importance, and of many other objects must be shown on frequently recurring edi- tions. From 12 to 20 editions embodying such information on various scales were kept up for the actual area of operations, whilst during the progress of a battle daily editions were brought out.
To cope with this volume of printing it was found necessary to provide for the rapid reproduction of trench maps at the Ordnance Survey Office at Southampton and for small but complete litho- graphic establishments in each army and at G.H.Q. Zinc plates were prepared by the Vandyke process or by helio-zincography, and flatbed printing machines, preferably motor-driven, were used for printing in the field. The French and German armies had well- equipped printing trains which were used for increasing printing facilities in important areas. The French, for example, dispatched a printing train to Italy in 1917. Towards the end the Americans employed a well-designed printing plant in lorries.
A scale of issue of about two copies per officer engaged, of each important map, was maintained, but experience showed that this scale should be increased. During the whole period of operations about 34,000,000 maps were printed for issue to the British troops, of which about two-thirds were printed by the Ordnance Survey in England and a third in France.
For the use of large scale maps in trench warfare a well-thought- out system of coordinates, based on a suitable projection, was needed.
It must be possible to read off at sight the coordinates of any desired point from a " g r '"l " or network of lines printed on the map. All the armies had such systems. Experience proved that for ease and accuracy of reference the "grid" should be in squares, the sides of which can be divided decimally by eye. The artillery often desire coordinates of the same accuracy as the surveyor i.e. on the scale of nature. The system adopted should therefore be based upon the coordinates used by the surveyor, with an easy form of abbreviation to be used by all arms to define map positions. Accuracy of bearing from any one position must be maintained together with as near an approach to linear accuracy as possible, whilst the system must allow of extension over the whole area of operations.
The considerations which influence the choice of a reference "grid" have already pointed to the desirability of an orthomorphic projection. For the conduct of surveys constantly in progress it is equally important. Computations must be cut down to the minimum, and this is best secured by working on a projection in which the position on the ground and the position on the map can be calculated in one process without sensible error.
From the cartographic point of view the question is not so important, and there are many projections which would provide a sensibly accurate map over large areas.
It was decided late in the war to adopt an orthomorphic pro- jection with two standard parallels which had recently been adopted in the French armies, but the decision came too late, and the Bonne projection continued in use in the British army until the Armistice.
Surveys for Artillery Purposes. Other things being equal, that artillery will dominate its adversary which has the quickest and most accurate knowledge of hostile battery positions and which can open most quickly an accurate and unexpected fire upon them. The construction and calibration of guns and howitzers, the homo- geneity of ammunition, and the measurement of those atmospheric elements which affect ballistics are involved; but accurate survey of the relative positions of gun and target is essential.
The positions of British heavy batteries were, therefore, fixed with a theodolite, each battery was supplied with a chart or "artillery board " on which the map was pasted down, in sections, upon a zinc or three-ply wood surface, and special " bearing pickets " were inserted in numbers in the battery zone. The bearings from these pickets to surrounding objects, suitable as reference objects, were tabulated and distributed.
The positions of hostile batteries were also surveyed with as nuch dispatch and accuracy as possible by one or other or both of the following methods:
(a) Intersection of three or more rays observed upon the flash -f discharge, the reflection in the sky of this flash, or, upon the smoke puff from the muzzle. This operation, commonly known as flash spotting, was carried put by units, each of which manned four survey observation posts, or instrument stations, and one headquarter post. Each section had its own internal telephone system and was in direct ommunication with an artillery headquarters.
Observations were directed by an ingenious controlling exchange upon hostile guns in succession. This method, independent of the
map, gave very good results but depended for success upon good visibility. Flash-spotting units could survey the positions of the bursts of our own shell and were, therefore, also used extensively for ranging and for calibration.
(b) Sound-ranging. A sound, in still uniform air, spreads outwards from its origin with an equal velocity in all directions, and the sound waves may therefore be likened to the ripples spreading outwards on a pond from the point at which a stone has fallen into it.
If we now imagine a row of surveyed pegs, more or less tangential to a ripple, projecting from the surface of the water, if we measure the times at which the ripple strikes each peg and know the velocity of the advance of the ripple we shall have all the data necessary for fixing the position of the origin of the disturbance. All armies engaged on the western front had some system for determining the position of a gun from the sound of its discharge, embodying the above principle. It is unnecessary here to describe the British system, except in so far as the survey of the sound-receiving stations is concerned^ A sound-ranging base was generally, though not invariably, laid out on the arc of a circle, the centre of which lay in the zone of the enemy's heavy artillery. There were usually six sound-receiving stations about 1,000 metres apart. The coordinates of the selected spot for each station were computed in the office, and the surveyor had to find and mark the corresponding points on the ground. Errors of more than one metre in position resulted in sensible errors in sound-ranging.
Sound-ranging is naturally unaffected by bad visibility, but is put out of action by a moderate wind blowing from base to gun and is interfered with by any high wind. As in the case of flash spotting, a good telephone connexion to the artillery headquarters is essential.
Surveying by Air-photography
Air-photography, or, to be more precise, photography of the ground from the air, has been recognized as a possible method of survey since the middle of the igth century.
Experiments in photographing the ground from balloons had been made by Col. Laussedat, Major Elsdale and others from 1859 onwards, but air-photographs played no part in any important survey before the outbreak of war in 1914. During the course of the war the difficulty of producing maps on medium scales was enhanced by the inaccessibility of the most important areas. It was due to this fact, and to the development in 1915 of photography from aeroplanes, that large areas in many theatres of war were mapped by the aid of air-photographs. But no full examination of the possibilities was made, and for peace sur- veys the method still remains in the experimental stage.
Optical Principles. Provided that the optical axis of the camera is vertical at the moment of exposure, the resulting photograph of a flat level area will be an accurate plan at a scale determined by the equation
4-
where f is the focal length, h is the height of the camera above the ground at the moment of exposure, s is the representative fraction of the scale.
Such photographs will be called vertical photographs. As a rule, however, the photograph is not exactly vertical, but the axis is tilted at an angle to the vertical. The photograph then becomes an inclined perspective view. (See Fig. I.)
Negative
( U//WC* ItHHVI ""
FIG. i.
If the direction and magnitude of the tilt of the axis were recorded at the moment of exposure it is obvious that the photograph could be projected optically or photographically on to the plane of the map, although it would remain unfixed in position and orientation. If the direction and magnitude of tilt are unknown then the pro- jection on to the required plane, or " rectification," is secured by comparing the relative positions of four surveyed points on the
