for, if I did so, he might think that I was encouraging him to get excited. I walk him, well in hand, and then I begin again the lesson, which I do not finish until he has done the changes steadily.

The pitfall in the changes of leg exist entirely in the preparations which the breaker generally makes to obtain them. Of course, I do not allude to the preparation of the horse, which is indispensable, but only to the movements of the breaker before the exact moment when he wants the horse to change. It is extremely important that he remains quiet while the horse is taking his strides, during which the animal should not change. Thus, during the first five strides he should not make the slightest movement. He ought not to change his " aids " until the sixth time, and should then change them together and with great precision. If at the fourth or fifth time he prepares to make the horse change, he will be taking a wrong course, because the horse, being very sensitive to the " aids " and attentive to the demands of his rider, will become disunited. If he touches the reins, the forehand will change, and if he alters the position of his legs, the hind quarters will change. But as there will be no harmony between the " aids," the horse will not change his legs.

It is wrong to blame the horse for becoming disunited, because in such cases the rider almost always provokes him to do so, by not remaining perfectly quiet until the decisive moment.

It also happens that the horse, thinking that he is going on all right, does not attend to his rider's demand, and anticipates the order, or what he thinks is the order. Therefore I cannot too strongly advise the rider to keep absolutely quiet during the first five strides of the canter.

It happens with all horses, that when they have learned to do the changes easily, they will do them of their own accord, and without an indication from the rider. In this case the