Although I have ridden horses for fifty years, I did not obtain perfect rassembler until the last ten years. It is true that I worked for a long time according to the somewhat inexact data of Baucher. But the fact is that for many years I continually felt the rassembler escape me, by the displacement of the centre of gravity, whether to the front or to the rear. I had to greatly refine my tact and consequently my " aids " to obtain the complete rassembler, and to preserve it with a maximum amount of propulsion.

But this is not all. There are not only direct movements, but also those to the side and when turning. In these movements one leg always predominates, in which case the impulsion which comes on the bit is not equally distributed between the two hands. The right leg throws more impulsion on the left hand, and vice versâ. It is therefore necessary, in order to preserve equilibrium in turning to the left, that the left hand, while remaining bound to the right hand, sends back to the centre a larger amount of force, which is all the more difficult to measure ; because this hand, while keeping up the rassembler, has to regulate the change of direction.^[1]

If we now reflect that in all work of equitation, the horse, whether going forward or keeping himself back, constantly tries to escape to the right or to the left, by his haunches or shoulders, we will see that in order to maintain perfect equilibrium we have to simultaneously perceive all the actions of the horse and all those which are being prepared, so as to combine them by opposing them, by the simultaneous action of the " aids," and to produce from them the desired ideal of equilibrium.

I said a short time ago that the rassembler in direct move-