Equitation/Chapter 30
CHAPTER XXX
THE PASSAGE BACKWARD
THE passage backward follows from the piaffer, and therefore presupposes a horse educated to the perfect state of assemblage and equilibrium.
A horse at the slow piaffer—which is, of course, the only form of the piaffer considered by the scientific equitation— balances itself on the same spot, all four legs flexing at the knees and hocks, but without gaining ground. The center of gravity is, therefore, midway of the body, and exactly under the seat of the rider. Under these conditions, the horse is like a large ball which rests upon a smooth and level surface, with which it is in contact, only at one end of a diameter. Evidently, the slightest force applied at the other end of this diameter will send the ball rolling in the direction of the force. So, in the piaffer, a force applied alternately on the two sides of the center of gravity makes the horse receive its weight alternately on its two diagonal bipeds. As the center of gravity shifts to the right, the left diagonal biped is raised, and vice versa.
If, then, under these conditions, the rider leans forward, the horse must move forward, under the operation of the same law. But if, when the horse is lifting his legs in diagonal alternately upon the same spot, the rider's weight is inclined backward, the 
When once the piaffer is obtained, the backward trot follows without much difficulty; but the movement needs moderation, and should begin with a few steps at first, the number increasing with practice. (Figure 38.)
The speed of the backward trot is not the test of its execution. A three-inch step, taken equally by each diagonal biped, and with the same cadence, tempo, and elevation as for the piaffer, is proof of a better equilibrium and a better training, than is any precipitate rush rearward in which the horse avoids the state of equilibrium by moving as it pleases. The air should always seem to be executed without exertion and without compulsion. The horse balances itself with an easy action of the limbs in diagonal, moves backward, returns to the piaffer, changes into the passage, returns to the piaffer, takes the backward trot. The rider's hands are immobile. The position of his body, as it swings like a pendulum into the correct place, is the force which actuates the mechanism.
With this animal mechanism, the backward trot is in perfect accord. The movement is entirely natural, when it is done in equilibrium from the piaffer. But if it is obtained by severity of hand, spurs, or whip, it becomes precisely contrary to the horse's nature. It is then dangerous to the rider, because the horse, pulled backward by the bridle, may rear and fall.
However, the trot backward cannot properly be considered a gait of the horse. It is serviceable only for perfecting the equilibrium, and for suppling the entire hind hand, most especially the coupling.
THE GALLOP ON THREE LEGS
IN the gallop on three legs, the horse uses both hind limbs; but only one in front, and holds the other in the air. Before the movement is asked, the horse must already be able to maintain a complete and permanent equilibrium during the ordinary gallop, to execute the jambette at the diagonal effect with great precision and with complete extension of the front leg, and to gallop, not terre-a-terre, but very slowly. (Figures 39, 40.)
The movement is asked by decomposing the air into its elements. The horse gallops slowly in assemblage. The rider stops it, and by means of the right snaffle rein and the left spur, asks immediately the jambette. After the jambette, the horse is allowed to walk. Again the gallop, the stop, and the jambette immediately. These three are repeated for whatever time is needed to calm the horse, and to teach it to keep straight when stopped and giving the jambette.
When the horse has mastered this exercise, the gallop is asked immediately after the jambette, without the intervening walk. From the gallop, the horse is stopped as before, made to give the jambette, and then started again at the gallop. Again, stop, jambette, start. Never change the lead; always keep working on the same side.
After a certain time, it always comes about that the horse executes the jambette just before it comes to the stop, partly of its own volition, and partly at the effects of the rider's hand and legs. The great point is, then, to seize upon this first single step of the gallop combined with the jambette or, in other words, of the gallop on three legs. When you have one — one only — caress with all your heart and send to the stable.
The next day, the same procedure. The horse, as before, does one step of the gallop with the jambette held. Once more, caress, dismount, caress again, and to the stable.
After a few days, get two steps of the gallop on three legs; then the next day, four. Continue in this way, but do not ask too much. When the horse does, let us say, five steps at the lead which he has been taught, change the lead and commence from the beginning precisely as before. Do not accept the slightest degree of confusion or mistake. Lean the body forward on the side of the jambette and push the horse forward with the legs.
Fillis advocates using the left leg to secure and maintain the jambette, and also to continue the gallop. I have, at various times and with different horses, obtained the jambette by holding the right snaffle rein in the right hand, high, and the curb reins low in the left in order to maintain the horse's head near the perpendicular, while my legs confine themselves to the effects needed for the gallop.
It is evident that, to obtain the gallop on three legs, the horse must be morally and physically perfect, or else have been adequately developed by its previous training. Moreover, the rider must himself possess delicate equestrian tact, and have perfect control over his effects. Even then, he will not always be successful, unless he has already educated several horses in the scientific equitation.
The gallop on three legs is a beautiful demonstration of the power of the man's effects over the animal; but it is of use only for this purpose and in the manege. Outside the manege, the air has no value whatever. It is, then, reasonable enough to teach the air to the manege horse, but not to horses that are for other service; and in general I think that the strength of the horse and the tact of the rider are better spent on more useful movements. I even go so far as to say that the gallop on three legs is a source of danger both in the case of a beginner and of a master who is training an animal for some one else to ride. For if the rider of a horse trained to the gallop on three legs is not a thoroughly competent esquire, he will not always use exactly the correct means to obtain the change of lead at the gallop, the change of direction, or the stop. He may, in that case, start the horse to galloping on three legs — to its great confusion.
Moreover, during the gallop on three legs, the horse is completely on his haunches. The hind legs carry all the weight, advance by very short steps, and always very close to the ground. Therefore, unless the horse is sent forward by the weight of the rider and by a strong effect of legs and spurs coupled with great tact of hand, the creature is exactly in the position to rear high. The gallop on three legs, like the gallop backward, demands a combination of favorable conditions as to both horse and rider that is in practice pretty difficult to find.
Considering, then, the danger to the horse's hocks and to its temper, and the peril to the rider, I cannot feel that the usefulness of the gait at all compensates for the wear and tear on the one or the risk to the other. Fillis has, indeed, executed the air most brilliantly, on the different occasions when he has exhibited his horses. I have performed the feat with several different animals. But, on the whole, the game has not been worth the candle.
THE GALLOP A TEMPO
THIS form of the gallop is a slow canter, in which the lead changes rhythmically from one biped to the other with each completion of a fixed number of steps. For example, the horse gallops ten steps to the right, and then on the eleventh it changes and gallops ten steps to the left. On the twentyfirst step it returns to the right-hand lead; and so on.
The difficulty is for the rider to keep count of the steps, since the air demands for its correct performance that the number shall always be exactly the same. Moreover, at its best, the movement requires the change of lead at every step — one stride with lead to the right, then the change to the left, then one stride with lead to the left, and again the change back to the right, thus continuing indefinitely. Naturally, this demands thorough training for the horse and the highest equestrian tact from the rider.
Both Baucher and Fillis have performed this air with remarkable evenness of rhythm. Fillis, also, once upon a time, laid a heavy wager with certain amateur horsemen, who denied the possibility of the gallop a tempo, that he would ride from the Arc de Triomphe to the Place de la Concorde, with a change of lead after every step. The grand master won.
In training a horse for this air, the change of lead should be at first only once in every twenty steps. Afterwards, with the greatest patience and moderation, the number is reduced progressively. The exercise demands great energy from the horse, which must throughout remain perfectly calm. Whatever the number of steps between changes of lead, this must always remain unvaried. 
THE GALLOP BACKWARD
For the gallop backward, the horse must be of perfect conformation, especially in its hind quarters, and must be educated to the point where it can interpret almost imperceptible effects of the rider. Its equilibrium and assemblage must be perfect—the sine qua non of this air, since the gait is very precise and the beats equal and uniform—and its strength must be sufficient to sustain without apparent exertion the gallop terre-à-terre. (Figure 41.)
In the gallop terre-à-terre, as in the piaffer, the horse is like a ball resting on one pole and movable by the slightest force. If, then, the rider's effects, by their lack of equality, timing, fineness, or uniformity, disturb this perfect equilibrium, the gallop terre-à-terre becomes impossible. But if the rider's effects are precisely correct, the horse will continue to gallop on the same spot, like the ball resting on a pole. Under these conditions, if the rider's weight shifts on the seat to throw the center of gravity backward of the perpendicular around which the whole mechanism has centered, the horse will be forced to move backward in order to prevent falling.
Meanwhile, of course, the rider, by his effects, must continue to maintain the equilibrium and the gait of the gallop. If either is disturbed ("evaporated" is the expression I use with my pupils), the horse loses either its equilibrium and then its gallop, or else its gallop and then its equilibrium. In either case, the movement becomes dislocated and impossible.
But the swing of the rider's body should never be a stiff inclination backward of a rigid spine. The weight is, at the beginning, immobile upon the saddle. Then for the change, the rider's spine plays back and forth, flexing at the coupling between the sacrum and the last lumbar vertebra, in time with each beat of the gallop and at the precise instant when the horse's two hind feet are off the ground, and the right fore leg only is bearing the weight—assuming that the backward gallop starts from the gallop terre-à-terre on the right lateral biped. This translation of the weight by the flexion of the coupling is to be repeated at each beat of the stride. Meanwhile, the rider's legs have to sustain the equilibrium and to hold the contact of the horse's mouth with the bits.
If, now, the rider, as he swings his weight, merely closes his fingers, without moving his hand, the horse will gallop backward, one step only, but still one step. That obtained, stop everything, yield everything, and caress. When the horse has become calm, forward again at the walk and the terre-à-terre at the same hand as before. Be quiet yourself; flex your spine; finger. Another step backward. That is enough for the time being. Dismount; and to the stable. The next day, the same progression.
After a few days, you will be able to obtain three or four backward steps. When the horse executes these calmly at the hand at which it was first taught, change the lead and repeat the same work at the new hand. Always keep the horse straight and forward. Better work near the wall, as this will aid in keeping the straight position.
If the horse is to be completely educated in the scientific equitation, it is better to teach the gallop backward before the gallop on three legs. Otherwise, the horse may give the gallop on three legs when asked for the gallop terre-à-terre. You cannot punish it for a mistake like this, and the result is confusion. But if the horse has thoroughly learned the terre-à-terre and the backward gallop, it is a far easier matter to push it forward against the contact, and so change from the terre-à-terre to the gallop on three legs, than to restrain it from the gallop on three legs to the gallop terre-à-terre.
In beginning either the terre-à-terre or the gallop backward, do not accept from the horse the slightest sign of being behind the hand. If you feel this at all, use your legs vigorously and push the animal forward upon the hand. The rider can always detect this tendency to stay behind the hand; and should correct it by giving three minutes of good, energetic promenade trot. For this purpose, I prefer the trot to the gallop, since at the gallop one lateral biped tends to get more work than the other, unless the rider takes pains to change hands. In any case, the gallop does not give so complete a disposition of the animal's forces as does the trot.
The "in hand" for the gallop backward is between the "upon the hand" and "behind the hand." A horse upon the hand lifts its front legs too high and its hind legs not high enough. But if the rider livens it by the action of his own legs, the horse rears or points forward. If the horse is behind the hand, the fore legs do not lift sufficiently, and the tempo of the gallop is not exact. It is, however, not possible to describe completely the sensation which comes to the rider's hand, and only by experience can the rider determine whether he is right or wrong.
In fine, then, perfect equilibrium, terre-à-terre, perfect equilibrium, flexion of the rider's coupling, fingering, moderation, and good fortune. The backward gallop proves uncommon suppleness on the part of the horse, together with great strength in the haunches. On the part of the rider, it proves high equestrian tact. Yet the position which the horse takes and the action of its legs are far from graceful, and the utility of the air is debatable. It risks the soundness of the horse's hocks, and it is certainly not worth attempting by a beginner, who has to spoil several horses physically and morally before he attains to the tact and the accuracy of seat essential to the gallop backward without danger.
And yet, for any rider, experience with the gallop 
I give (Figures 42, 43) as illustrations of the movement, Fillis mounted upon "Germinal," and myself upon "Why-Not," in order that the reader may compare the leg action of the two horses at the same gait. "Germinal" is fifteen hands, three inches high: "Why-Not" is sixteen hands, three inches. Although the backward gallop is the last refinement of equilibrium possible to the horse, it is in itself pleasant neither for the horse nor for the spectator. "Why-Not" is the fourth animal from which I have obtained it, not for my own satisfaction, but for the sake of making a picture for this book, in which I set forth nothing that I have not myself done.
And now, finally, at the end of this last chapter on horse gymnastics, I beg the reader to review the illustrations, and to compare the several pictures of "Why-Not" before his training and at the various stages of his development during the course and at the end. These photographs prove amply the muscular improvement accomplished during the horse's education.