considered desirable to keep the 17th Lancers in the center of it, in order that they should not interfere with the infantry fire, and that when the enemy was repulsed, they should be launched out upon them, and this was done with perfect success. It is, however, contrary to the instincts and traditions of cavalry to be shut up in a square, and, where practicable, I should think cavalry outside a square, even at some distance out of the way of the infantry fire, acting with horse artillery, would very much disturb the attacking bodies of the enemy, and perhaps attract away a portion of them, and they could be brought up, when called upon to do so, to carry out the pursuit.

In the first campaign on the east coast of the Soudan, on the advance to El Teb and afterward to Tokar, squadrons were sent in front and on the flanks of the square with scouts thrown out to feel the way and obtain information, while the main body of the cavalry was echeloned on the rear and flank opposite an angle of the square in the most suitable manner to avoid any interference with its fire. During the action it remained in this position until after the first attack on the square, when it moved away past the square on the outward flank and acted on the enemy's rear and engaged their reserves until the action was over. During the desert march in the Nile expedition, the 19th Hussars, by its scouting, protected the square and gave it timely notice of the approach of the enemy.

In a country where a great deal of bush abounds the effective charge of cavalry on to groups of savages is very much curtailed. The Arabs throw themselves on to the ground behind the prickly bushes, the ranks are opened out as the horses avoid the thorns, and the men get no chance of using their swords; but although much execution is not achieved under these circumstances, the natives have great fear of cavalry, and they are prevented from attacking elsewhere. When their attention is thus occupied, horse artillery and machine guns might make great havoc among them. At the action of Tamai, where the ground, from the rocks and ravines of the neighboring mountains, was unsuitable for cavalry charges, when one of the infantry squares was broken, the cavalry advanced, and one squadron of the 10th Hussars, dismounting, helped to create a favorable diversion by pouring fire into the flanks of the attacking Arabs.

My remarks would, I think, be considered incomplete if I did not touch on the question of cavalry charging squares, as this point is always made very prominent in all discussions on the action of cavalry. I therefore must not pass it by. I will say at once that I think it most undesirable, even under favorable circumstances, that cavalry should charge a formed square and men armed with the breech loader. At best the gain can be but local and partial, while the loss to the cavalry—an arm so difficult to keep up in numbers—must be disastrous, and it seems to me that if cavalry by its appearance can force infantry to form square, it has done enough, and that the artillery, infantry, and machine guns should do the rest. The necessity might, however, arise, and by looking at the past we see its possibility. At Langensalza two Prussian squares were broken by the Hanoverian cavalry, and the major part taken prisoners.

We have only to turn to the recent campaigns in Egypt to see the effect of determined rushes of men, intending to succeed, charging on to squares carefully formed on ground affording shelter, with an enormous amount of fire being poured upon them. It will be said that these men were fanatics, but our cavalry, too, have been, and will be, fanatics in a charge; and I still think, if the necessity recurs, that an attack, properly conducted on favorable ground, one troop charging on the corner of a square, followed by another at double distance, others in echelon on both flanks immediately following, the whole charging with the greatest impetuosity, intended to win, will break down any square that Englishmen are likely to meet with. If we look back again on the past, we will find many instances of British cavalry not being called upon in vain to make a sacrifice. At Talavera, the 23d Light Dragoons, supported by the 1st Hussars of the German Legion, advanced against the French squares. In their impetuous rush they came upon a hollow cleft covered with long grass, eight feet deep, and eighteen feet broad. Too late to pull up, the foremost rode headlong into it, some tumbled in, others over it, some rode boldly at it and gained the other side. Still they went on, swept past the infantry columns, and fell upon a brigade of French chasseurs.

At Balaklava 670 British horsemen were launched against an entire wing of the Russian army. The brigade, at first in two lines, the 11th Hussars, 17th Lancers, and 13th Light Dragoons, followed by the 4th Light Dragoons and 8th Hussars, advanced down a gradual descent of three-quarters of a mile; the Russian guns vomiting shell and shot upon them, one battery bearing on their right, another on their left, and all the intermediate ground covered with riflemen. The guns were charged and forced through, the forces drawn up in rear were overpowered. They then had to turn, and, retiring up hill, ran through the same gauntlet. In the Sikh war, at the battle of Ferozeshah, the 3d Light Dragoons charged the enemy's entrenchments at a point defended by some of their heaviest batteries. When within 250 yards the regiment moved at speed under a destructive fire of grape and musketry, and pressing forward at the charge entered the enemy's camp and captured the whole of the batteries.

Cavalry attacks have been made with success after dark, and the advantage, of course, is gained of obviating opposing fire. Prince Kraft mentions that after the battle of Mars-la-Tour, the cavalry division, re-enforced by the divisional cavalry, rode forward to complete the advantages gained. It was almost night, and fault has been found with making the attack in the dark. If the ground is well known a night attack may be advisable. While criticising it, we have to think of the feelings of a half-defeated army about to bivouac being attacked by unknown forces in the dark. In this case, at Vionville, the enemy did not wait for a second, but withdrew, and abandoned the whole field of battle. Prince Kraft quotes the attack of Blücher at Gross-Gorchen and a cavalry attack at Loon. During the first Egyptian campaign the Life Guards made an attack by moonlight at Kassassin.

I have now, I think, touched lightly on some important cavalry duties on a campaign. In some points perhaps these remarks may appear contradictory. How to combine keeping cavalry in reserve for any great action it may be called upon to perform, while using it unsparingly to assist on the battle field, if the necessity arises. It may, however, be noticed that, much as they may be criticised, few cavalry commanders have been severely blamed when they have thought it best to take the bolder course. To insure to cavalry the power of carrying out its duties successfully in war, organization and practice in peace is most essential. Infantry may suddenly be increased without much deranging its action in the field, but cavalry cannot be hurried into an increased augmentation. In tactics simplicity in every evolution and rapidity in execution are the most important principles. This simplicity of drill, I think, might be assisted if our squadrons were divided into four divisions, zuges, or pelotons. When squadrons have 48 files in the front rank there might be four of these, while weak regiments with 36 files could drill equally as well with three divisions. This system, introduced by the late Gen. Valentine Baker into the English service for a time, and now used by all European countries, was found to work well.

I think the whistle could be carried with advantage by all cavalry officers. For advanced work attention can be drawn by it without being heard at a distance like a bugle. In movements the commanding officers would find it useful to call the attention of leaders