[53] A proportion relative to the same region, and which at the outset might appear similar, is pointed out by Bourgelat (see p. 266, paragraph 4). But there exists a difference, for Bourgelat compared the length of the head, measured from the forelock to the commissure of the lips, and not that of the entire head, to the distance which separates the summit of the rump and the tip of the patella. =Two and a half times the length of the head= gives: 1. The height of the withers, H, above the ground.[54] [54] This proportion is that given by Bourgelat (see p. 265, paragraph 2). 2. The height of the summit of the crupper above the ground.[55] [55] Consequently the withers and the crupper, being the same height, are situated on the same horizontal plane. Bourgelat, on the contrary, points out a difference of level in connection with these regions. According to him the summit of the crupper is situated below the horizontal plane passing the withers, and this distance equals half of the space which separates the great angle of one eye from that of the other (see p. 269, paragraph 20). 3. Very often the length of the body, from the point of the arm to that of the buttock, although for a long time the type of Bourgelat had been set aside as a conventional model, short and massive.[56] [56] See p. 265, paragraph 2. And M. Duhousset adds to this: 'The drawing that we offer, which has two heads and a half in height and length, is that of a horse which we frequently meet with' (see Fig. 107; see also p. 279, where we again consider this question of the length of the body of the horse). 'The crupper, from the point of the haunch to that of the buttock, D, F, is always less than that of the head. This difference varies from 5 to 10 centimetres. The width of the crupper, from one haunch to the other, often very slightly exceeds its length.' MM. Goubeaux and Barrier add that frequently it equals it.[57] [57] If we refer to the proportions indicated by Bourgelat, we shall find that the proportions relative to the crupper are also indicated there (see p. 266, paragraph 4). 'The crupper, such as we have just defined it, D, H, may also be found to a fair degree of exactness, as regards length, four times on the same horse.' 1. From the point of the buttock to the inferior part of the stifle, F, P. 2. The width of the neck, a little in front of the withers to a little above the point of the arm, S, X.[58] [58] MM. Goubeaux and Barrier replace this by the following: 'The width of neck at its inferior attachment from its insertion into the chest to the origin of the withers, S, X.' Bourgelat discovered the same proportion (see p. 266, last line of paragraph 4). 3. From this latter point to below the lower jaw, X, Q, when the head is naturally placed parallel to the shoulders, E, H.[59]