Reconnaissance occupies a prominent position amongst these various duties. There will not be a battle every day. Not in every fight will considerations of ground and other circumstances allow of the cavalry taking part in a great decision as a mounted arm. Not always will an effective pursuit by the cavalry mass be possible. It must be remembered that in order not to render itself too weak to carry out its proper rôle cavalry must not expose itself to heavy loss in battle without sufficient reason. Only the possibility of a very great success can justify the risk of staking the whole cavalry force in a decisive charge during the battle. It is true that efforts will have to be made to drive the hostile cavalry from the field to facilitate operations against the flank and rear of the enemy's line of battle. But such operations will generally be limited to fire action, and in a decisive battle the weakness of the cavalry would probably debar it from such undertakings. Engagements, also, that are fought with a view of opening a way for reconnaissance will not be of frequent occurrence, and must only be expected during occasional crises. They will, however, be of decisive value for the whole future conduct of the arm and its operative success. The same may be said of raids against the enemy's communications. Such undertakings may exercise the greatest influence on the course of a campaign, but can only be undertaken under specially favourable conditions.

On the other hand, the everyday task of the cavalry which goes hand in hand with all these various engagements and enterprises is reconnaissance proper carried out by reconnoitring squadrons and patrols. This is the daily bread of the cavalry, a duty that throughout a war should never cease to be performed, even if the main body of the cavalry has been driven from the field by the enemy. It demands, therefore, the highest training, the wisest economy of force and systematic arrangement, if it is not to become ineffective or useless.

In close and continual relation to the above is the further task of safeguarding and screening the army. Reconnaissance itself provides a measure of security, but it is always liable to miscarry, and can never secure the army against hostile observation. Reconnaissance must therefore be supplemented by a special system of security and screening which, however, demands the most careful organisation, and greatly increases the difficulty of husbanding limited forces and of keeping the troops fit by not overworking them.

Intimately connected to the services of security and reconnaissance is that of communication, which, owing to the great distances to be traversed in modern war and the necessity of early transmission of intelligence, has become, at the same time, of the highest importance and of the greatest difficulty. Here also is a daily duty for the cavalry in which a thorough training is necessary.

All these considerations must be kept in view in arranging the training, and a clear conception should be formed as to what is to be aimed at in great as well as small matters. Only thus will results be obtained which will stand the test of war.

It is only natural that in our army the greatest stress is at present laid on the duties of the army cavalry. These are so obviously new and important for the arm that it is easy to understand how more time is devoted to them in peace training, and how the task of the divisional cavalry has come to appear of less importance. It might indeed be contended that the training which the latter has hitherto received in garrison and at manœuvres would suffice for its needs.

It is my opinion, on the other hand, that such a conception is faulty. The importance of the divisional cavalry has in no way diminished, and one is not justified in assuming that any lesser demands will be made upon its efficiency than on that of the larger independent formations. The exact contrary is the case, and I consider it opportune now, when all interest is centred in the army cavalry, to lay particular stress on the extraordinary importance of the divisional cavalry.

As in the newest phases of military development, the general value of cavalry, according to my opinion, particularly in the organism of an army, has increased, so also has that of the divisional cavalry. Higher demands are made of it, and it must therefore be capable of greater performances.

The circumstances of modern war demand that great masses of mounted men shall be used as army cavalry and concentrated in the decisive direction, thus weakening a great part of the front of the army in cavalry, in order to ensure superiority at the decisive point. It is in accordance with the universal law of military success that a concentration of force at the decisive point can, under most circumstances, only be ensured by a corresponding weakening of force in other places. The front of the army, therefore, can never be covered throughout its whole length by the army cavalry, but at the same time it will never be possible to entirely denude of cavalry that front or flank in front of which no army cavalry may lie. This would be simply impossible, for every body of troops, however disposed upon the strategic front, requires cavalry for the service of close reconnaissance, of security, and of screening against surprise or against hostile observation.

Furthermore, even those bodies of troops in immediate advance of which the great cavalry masses are on the move require their own cavalry, not only for the purpose of ensuring and maintaining communication with the army cavalry, but also that they may not be entirely denuded of cavalry when, as the hostile armies approach each other, the cavalry masses clear the front and concentrate towards a flank.