this is most apparent when it is remembered that the vulnerable parts of every force are its flanks. The mobility of cavalry is of course best brought into play in a moving defensive, that is to say, in a rear-guard or flank-guard action. The addition in the middle of the war of 12 automatic rifles to a regiment has greatly increased the fire-power of the arm, but it is to be hoped that in the future some weapon will be evolved which will be less susceptible to inaccuracies and stoppages than the Hotch- kiss. It is a waste to use cavalry in a continuous defensive line if other troops are available. The proper dispositions for a cavalry defence are: (i) a chain of localities, with gaps between them, held principally by automatic weapons; (2) reconnais- sances far out to the front and flanks, to give ample warning of hostile approach; (3) a large proportion of the force in mobile reserve; (4) careful arrangements for withdrawal to a second position if it should become necessary. But, above all, cavalry- men must not be afraid of a very extended line, relying upon greater mobility and moral superiority to compensate for numerical weakness. The soundness of this doctrine needs no more proof than that furnished by the operations of the British cavalry corps in the first battle of Ypres.
The principles governing the conduct of a reconnoitring detachment will appear simple if it is remembered that the information required is, in each locality, ultimately obtained by one pair of eyes or one pair of ears. The r61e of the rest of the detachment is that of a conveyance, to enable that pair of eyes or ears to arrive at a place whence it can see or hear, and to allow of the news being sent quickly back to the commander. A reconnoitring detachment, then, fights only in order to arrive at its destination or to keep a road open for messages going back. Its strength is regulated by the amount of opposition which is anticipated, and by the number of messengers which will be required. Sometimes it will act by stealth, when its strength will be reduced to a minimum. More often, however, even if it may be possible to get forward without opposition, the difficulty of maintaining a channel of communication will enforce a fighting role on a reconnaissance; and, as the result of recent experience, cavalry opinion has veered very decidedly towards strong detachments, not less than a troop, and often as much as a squadron or more. It should here be mentioned that since the commander of the force alone knows how much importance he attaches to obtaining information, he, and not a subordinate, should decide upon the strength of a reconnoitring detachment. In principle, the order " Send out a patrol" should always be followed by a statement of the strength, and this should be insisted upon in all tactical exercises.
The advance of a reconnoitring detachment, like that of all other cavalry bodies of every size, is conducted on the principle of successive objectives, or bounds. This principle, which seemed before the war to appertain exclusively to cavalry, has now been adopted by infantry also; but, on account of the difficulty of maintaining control with fast-moving troops, it is particularly important that it should be thoroughly well under- stood by cavalry. The idea underlying the principle is that of getting as quickly as possible through specially dangerous areas. A series of objectives is chosen, the occupation of which by the enemy might cause inconvenience; such objectives may be a position covering a defile, a village, a cross-road in enclosed country, a simple eminence giving a good field of view, or some other feature of tactical importance. The distance between them varies according to the size of the force; for instance, a squadron acting as vanguard would be given objectives intermediate to those thought necessary for the brigade following as main body. The main body does not leave one objective till the advanced guard has reached the next; the advanced guard does not move on till the main body is close up to it, but meanwhile prepares the advance to its next objective by sending forward reconnnais- sances. The movement resembles in fact that of a caterpillar. This principle has very great advantages, though it tends to some extent to retard movement. In addition to providing security for the main body, it gives to the commander of the advanced guard a definite tactical feature for which to fight
should he meet the enemy unexpectedly, and so helps 'him to make up a plan. Also, in the case of two parallel moving bodies, it ensures the heads keeping more or less level with one another.
The commander of a reconnoitring detachment is given his orders in the form of instructions, the framing of which is an important matter. Once despatched, he will have to act wholly on his own initiative; it will not be possible to overtake him to give him supplementary instructions, and he will often be faced with situations entirely unforeseen. He must therefore be given full information respecting the situation at the moment and the intentions of the commander. It is a truism, but one which is often overlooked, that, in order to get definite answers, one must ask definite questions. Instructions such as " to make good " a certain locality, " to work round," or " to clear up the situation " will result in undecided action and vague infor- mation. If negative information is required, the places from which or times at which it is to be sent in should be stated. The reconnaissance commander must know how long he is likely to have to stay out and how far he is to go, whether he is or is not to remain in observation when the enemy is met with, and what he is to do in the event of his reaching his farthest objective without meeting the enemy. These two lat- ter points particularly are often omitted.
Not only reconnaissances, but all other cavalry detachments also, go out to greater distances and are left more to their own resources than is the case in the infantry. For them also, there- fore, clear and far-seeing orders are essential if they are to per- form their task satisfactorily. And besides this there are certain other matters of staff work which require special attention in cavalry formations. For instance, arrangements for even a simple march have to be particularly carefully thought out by reason of the fact that the mounted men can move slightly faster than the light transport of the formation, and very much faster than the heavy transport; motors, too, complicate the problem. The question has always to be decided: Should the transport move with brigades or what is called divisional- ized? If the latter, how is it to be assembled before and dis- persed after the march? Should.it move off first, allowing the mounted men to overtake and pass it, or last, which usually entails a very late arrival at the destination? In infantry formations the fighting troops and the transport move at the same pace, so that comparatively little difficulty arises. Again, horses have to be fed and watered, and should be offsaddled whenever possible; consequently, unless horsemastership con- siderations are overruled and the formation is kept in a state of instant readiness, cavalry take longer to get on the move than infantry. This drawback can be minimized by enacting that a portion of the force must be prepared to move off instantly and that the rest must be at some particular length of notice. Much can also be done, on occasion, by issuing warning orders, with the object of shortening the length of the orders to move when the time comes to write them. In the future, too, the cooperation of the component parts of the mobile mixed force will require staff work of a very high order.
Lastly, the most brilliant conception of a r&le, the highest degree of tactical skill, the most unerring staff work none of these will suffice unless the leader has the requisite personality. It is rare indeed that such a personality is found, and the fail- ures of cavalry in the past can be traced more often to the short- comings of the commander than to any lack of efficiency else- where, or to conditions of ground and armament. It is essential, if the formation is to retain its dash, that the commander should be entirely confident of the r&le of cavalry. Now, during periods of stationary warfare, entailing inactivity for the mounted troops, it requires a character of exceptional firmness to retain that confidence to the fullest possible extent. But the principal stumbling-block consists in the facts that a cavalry detachment once sent out can be recalled with difficulty only, and that a mounted attack once launched cannot be recalled at all. Evi- dently mistakes must often be made, and the strength of the enemy sometimes underestimated. The cavalry leader will fail if he ponders too much on contingencies; he will not succeed
