the eye of the keenest observer. The use of the camera almost eliminates the personal factor. All that is necessary is that the observer should be able to manipulate the camera and have sufficient knowledge of the map to take the photographs at the correct moment. The image of the ground on the plate is very quickly developed and printed, and studied under favourable conditions away from danger of enemy fire. Aeroplane photo- graphs became more and more important during the World War. Cameras were improved; the men whose task it was to interpret the photographs after they were taken became more and more expert; and the possibilities of their work became more fully realized. At first several difficulties had to be faced, the chief of which were mist and vibration. The mist in the atmosphere rendered the photographs taken so indistinct as to be almost useless. But by the use of filters (that is pieces of tinted glass or celluloid placed next to the camera lens), it was found that photographs could be taken through a mist which the human eye could not pierce. Vibration caused more diffi- culty. The danger that, through the throbbing of the engine, the camera might be tilted while the photograph was being taken, was finally eliminated by the use of rubber cushions which absorbed the vibration of the aeroplane.
In addition to the important photographic reconnaissance, other types of air reconnaissance were used during the war. Each dawn and dusk, and at other times when conditions made it necessary, reconnaissances were carried out by powerful machines. Their principal duties were to keep close watch of road and rail movements of the enemy. Since these movements mostly took place during the night, dawn and dusk were the times when results were most likely to be obtained. The day- to-day results of the reconnaissance were plotted in " activity maps " which made it possible to gauge the normal movement in any railway line on the battle-front, and so, with the aid of collateral information, to establish conclusions as to significant abnormalities. Another type peculiar to position warfare was the trench reconnaissance. This was carried out by an aero- plane flying low over the trench lines. It had for its object the discovery of the state of the enemy's defences, what portions of the line he was holding, the location of machine-guns and trench mortars, and all those numerous details a knowledge of which was required in carrying out an attack or raid on strongly organized positions.
" Artillery " air reconnaissance was primarily intended to counteract the effects of the enemy's artillery fire on one's own troops. Flying backwards and forwards along the battle-front, the airman watched carefully for the fire of enemy batteries. When a battery opened fire, its position was signalled by wireless, and it was promptly engaged by the counter battery.
One of the chief difficulties with which a commander has to contend during an attack is the difficulty of knowing quickly how far the attack has succeeded and to what distance his troops have penetrated the enemy lines. The aeroplane there- fore has a useful function in maintaining contact with attacking troops. In the World War machines carrying out this duty were called " Contact Patrols." They were not intended for fighting purposes, but to determine the position of the attacking troops. In open warfare the aeroplane landed in or near the forward positions and received an account of the situation from the troops themselves. But when this was impossible pre- arranged signals of various kinds were given by the foremost troops to show their position. A good airman worked without any help from the ground troops. He would fly at a low alti- tude, sometimes as low as 50 ft., along several miles of front, and place correctly the position of all the forward troops. To deter- mine the position of the enemy and also to discover in what direction counter-attacks were maturing the airman had to rely on his powers of observation. Such work was often very dan- gerous, since it necessitated flying low over the enemy. Various signals were devised by means of which the aeroplane could inform the infantry of an expected counter-attack. The duty of watching for such threats was sometimes assigned to a separa'te counter-attack patrol; but more usually it fell to the contact
patrol itself. Having obtained his information the pilot's object was to give that information as speedily as possible to those who needed it. He did this as a rule by dropping a mes- sage or a marked map at a prearranged station; or he landed at a headquarters and explained the situation in detail to the com- mander; or he returned to his aerodrome where an officer of the intelligence staff interrogated him and telephoned or telegraphed his information to those whom it concerned.
Detecting Agencies. Differing from reconnaissance princi- pally in the absence of " contact," but otherwise analogous, are those means of obtaining information which may be called collectively detecting agencies. These are visual or instrumen- tal, and in some cases a combination of the two. V'sual observa- tion for intelligence purposes differs from the ordinary watch- ing duties of sentries in that it is an organized service partly or wholly under intelligence control for the observation and recording of all enemy activity within the range of vision of the front-line observation post, the tree or belfry behind the line, or the captive balloon in which the observer is stationed. Its records go into the common stock of tactical intelligence material, its work is facilitated by a special equipment of maps, tele- scopes, etc., and its various elements are so placed and co- ordinated that the exact location of the enemy activity recorded (e.g. digging) can be fixed by intersection. In its most precise form, observation becomes " flash-spotting," that is, the loca- tion of an enemy battery position by simultaneous observations of a gun-flash from two, three or more visual posts provided with goniometers and connected electrically with a central where the result' of the intersections is plotted. Flares, Very lights and searchlights to facilitate night observation, are aids to defensive sentries, not to " positive " intelligence.
Detection by instruments (other than the usual flash-spotting in which, after all, the quickness and accuracy of the observer is the main thing) is automatic. Instruments are disposed to receive, transform and transmit impulses from outside, and the human element (except in instruments of the geophone class used in mining) is only introduced at the " central " or exchange station i.e. at the point of synthesis, and not in collection. Such are sound-ranging installations, wireless interception and direction-finding apparatus, and electrical listening-posts. AL contribute to the common stock, and each affords collatera checks called by the French reconpements or " intersections ' on the data provided by the rest, or by reconnaissance proper, or from other sources.
Interpretation. It remains briefly to outline the way in which these means are used to answer the three questions that interest the command at the front: (i) What are the enemy's disposi- tions, (2) what and where are his defences and other installations, and (3) what are his intentions? The third of these questions is really the interpretation of the other two. It depends on military knowledge, on flair, and especially on an exact appre- ciation of what constitutes normal and what abnormal " activ- ity." The first and second questions only concern us here.
Enemy Dispositions in the World War. The Allied and Ger- man armies on the western front faced each other with only a small space of ground between them. Except when an attack was in progress contact was maintained by frequent raids into the enemy lines, by means of which prisoners and documents were captured. The units to which prisoners belonged was revealed by their pay-books and identification discs. German prisoners were, moreover, usually willing to state the units who were occupying the line, as well as the general dispositions for holding it. In trench warfare, then, provided that raids were frequently and successfully carried out, the problem of identi- fying the troops in line was not difficult.
The problem of the grouping and location of enemy reserves is far harder. It is similar in open and position warfare with the important exception that in open warfare the proportion of undeployed, undisclosed reserves is, on the whole, higher. Even if the intelligence service of an army is able to locate the re- serves of the enemy, it does not follow that it will be able infal- libly to predict the enemy's intentions or area of attack. The
