This page needs to be proofread.
AIR DEFENCE
93


at the outset, is to be " strong everywhere," and consequently " strong nowhere." Such a policy involves dispersion of available strength over unimportant localities, reduction of control, loss of cohesion in effort, extravagance, and the achievement of a minimum of efficiency. The close defence of the localities which are important to the state is the only sound policy. For the defence to be effective, the attack must be met and defeated at the right height and outside the line from which it can achieve its object. Such at least must be the aim of the defence, however difficult it may be of achievement. That is to say, the defence must be outside the objective of the attack. This necessity plunges the matter at once into difficulties with what is known in the army as the " chain of command." An army works by definite boundaries shown by real or imaginary lines on the ground. The air knows no boundaries. It follows, therefore, that those units of air defence formations which are tied to the ground must be sited and organized for purposes of command with no regard to those imaginary territorial boundaries necessary to the ordinary army of the ground, and solely with regard to the whole area in which the vulnerable point or points are situated. In one respect the sea has an important bearing on the nature of aeroplane attacks. The risk of being shot down on the return journey while still over water, with little or no hope of rescue, tends to make a circumspect pilot fly high over his objective, even if this be some distance inland, as he must evade detection till he has gained such a start over the pursuers as will enable him to pass the sea in safety even with a damaged engine. This was apparently the policy of the Germans during the raids on London in Sept. 1917. Whether, in any given case, the pilot will thus sacrifice some of the effectiveness of his attack in order to give himself better chances of a safe return, will depend on his personal character, the traditions of his corps and the free hand or limiting instructions that he receives from his superiors. From the point of view of the defence this has its drawbacks. It is difficult to decide a priori, or even during the progress of the attack itself, as to the probable height of the enemy when the basis of the decision is practically conjecture. Another effect of the seacoast on anti-aircraft defences may be to limit them in area. The defences must extend over an area outside the vulnerable point; but, in cases of ports on the open sea, that area is limited to the ranges of gun and searchlight on the edge which borders on the sea. Some typical instances of the use of the various instruments of defence may now be considered. The defence has to provide against attacks both by day and by night. By day the instru- ments of defence and their adjuncts are: the machine-gun in the air, the heavy gun on the ground, the sound locator, and the observer post. By night the machine-gun in the air must be manned by a crew specially trained in night fighting, and in addition there is the searchlight. By day and by night the object of the defences is to break up the enemy attack and destroy it in detail. By day the massed attack must be broken up by gunfire before the aeroplanes on the defensive are launched against it; this entails guns outside the defensive aeroplane patrols, which again are outside the vulnerable point. Then in support of the aeroplanes (i.e. in rear of them) more guns again are required to repel such of the attackers as succeed in penetrating the aero- plane patrol area. And lastly, throughout the area of the vul- nerable point itself, provision must be made for attacking by gunfire any hostile machine which may succeed in penetrating so far. The attack will probably be audible and visible throughout the greater part of its course. In certain conditions of thick cloud or haze it may be invisible from the ground, but this fact, though increasing the difficulties, does not alter the disposition of the defences. By night the attack is broken up in an entirely different manner. Both attacking and defending machines being in darkness, the attack is, as it were, reconnoitred by the search- light, and the targets selected by the latter are isolated for engagement by the apparently simple process of keeping them illuminated. Unless the searchlights succeed in their object, the attack is invisible. It is not possible as a rule to illuminate several targets in a searchlight beam simultaneously, although during the war as many as five have been held in the beam simultaneously for a few minutes; nor is it likely that any method of illuminating a formation of, say, 22 machines simultaneously, for any length of time, would be practicable. The outer ring of guns, therefore, would normally remain inactive by night unless the absence of a defending aeroplane gives an opportunity for a gun to engage an enemy target. By day and by night the aeroplane in defence can only move a certain maximum distance on patrol without running the risk of allowing an attack to slip past in rear of it ; the aeroplane also requires a certain minimum distance on one side or other of its patrol line in which to manoeuvre and bring its enemy to battle. Suppose for the purposes of illustration these measurements be taken at 15 and 10 m. respectively. The aeroplane patrol area, and the battle and pursuit area, must be kept as clear as possible of gunfire areas and areas containing vulnerable points of any size. The width of the gunfire area will depend on the probable height at which the attack is delivered. Assuming that the latter is 10,000 ft. and that the gun can command a horizontal range of three miles at that height, the belts of gunfire may be taken at six miles in width. Observer posts must be between 70 to 100 m. away, as has been shown, in order to gain time for the defences to get into position, If they are to meet the attack as it comes in and not bring it to account merely as it is returning home. In the case of a vulnerable area represented by a circle of a radius of 5 m., the area immediately outside that will be a belt for gunfire from 3 to 4 m. in width; the next a belt of 10 m. for the aeroplane battle and pursuit area; then one of 6 m. for the outer gunfire area; and a final belt from 45 to 75 m. wide covered with a network of observer posts, each of which can be from 10 to 15 m. from each other. This arrangement provides for the problem of defence by day. By night it is necessary to consider the disposition of the searchlights, and it will have been seen that one of their functions is to indicate the approximate position of attacking aircraft. To be of any value they must be able to do this throughout the vulnerable area, the adjacent gunfire area, the battle and patrol area, and for a sufficient distance outside the latter (say 4 to 5 m.) to enable the aeroplanes patrolling in defence to move into position to meet the attack. This gives the total area through which searchlights must be disposed, the projectors being at the angles of triangles whose sides measure approximately 2,500 to 3,000 yards. Owing to accidents on the ground, trees, houses, railway stations and the like, the actual distribution of search- lights throughout the area often appears to be indiscriminate; it is inadvisable as a rule to place a searchlight nearer than from 200 to 500 yd. from a gun. Again, by night, the difficulties of determining the height of the attack are so great, that it becomes necessary' to dispose the aeroplanes in defence at different heights. Assuming this difference to be 1,000 ft., and that there are five machines one above the other, with the lowest at about 8,000 ft., the highest will be at 13,000 feet. The degree of endurance to be expected of a pilot flying on patrol at night may not exceed a tour of two hours in the air. These data, combined with a knowledge of the average lengths of the summer and winter nights, will be sufficient to give some indication of the minimum numbers of machines and pilots required in the problem of night defence. The number by day is also affected by the probable frequency and size of the attacks. It will now be easy to realize the enormous scale of defences required if any appreciable degree of efficiency is to be attained. A simple diagram will illustrate this general disposition of defences. Few "vulnerable points" are as symmetrical as those in- dicated in these diagrams, but the principle illustrated can be applied to areas of almost any shape.