POINTS FROM WHICH THE PROPORTIONS ARE STUDIED. 391 



column, but rather to the strengthening of the points of the shoulder 

 and the buttock, two points which compensate what the total length 

 AD at first sight appears to be defective in, since it measures two and 

 two-thirds heads. 



The normal length, two heads and a half, on the contrary, is not 

 always constant when the animal is short and compactly built. The 

 back and the loins of many horses meeting these general requirements 

 are weak, on account of a want of development of the croup, vertical 

 direction of the shoulder, and often also on account of the prominence 

 of the spine, poorly covered by these parts. Again, this defect is 

 judged of by the value of the scapulo-iliac distance, of which we 

 have already spoken. 



Finally, the excess of length, when it is not exaggerated, cannot 

 only be compensated, as is seen in Fig. 142, but it is to be preferred 

 to the classic length of two heads and a half, even supposing the 

 scapulo-iliac separation to be normal, if it is due to the disposition 

 of the shoulder and croup, and if it coincides with a long chest and a 

 short flank. In this case, in fact, the dorso-lumbar column is in good 

 proportion and strongly supported ; it supports a spacious thoracic 

 cavity ; the animal has brilliant and extensive action, endurance, and 

 a harmonious form. This may be seen in Fig. 143, in which subject 

 No. 2 is one-sixth of a head longer than subject No. 1, which possesses 

 normal conditions, but whose shoulder and croup are wanting in length 

 and inclination. 



From all that has been said, it is seen that we cannot investigate 

 too carefully the value of the different elements which compose the 

 length of the body. A superficial estimation of this dimension is 

 insufficient, even when it appears to be in the condition indicated 

 above. The relative condition of the back and the loins on the one 

 part, the shoulders and the croup on the other, must also be appre- 

 ciated; the apparent excess of length should not be considered, a 

 priori, as resulting from the spine, nor regarded as implying always 

 the length of the chest. We should be in danger of committing 

 errors. When assigning two heads and a half to the length, we meant 

 to imply at the same time a proper scapulo-iliac distance, a well-formed 

 shoulder and croup. The particular cases which we have analyzed 

 positively prove that it is impossible to determine absolute figures. 



Relations between the Height and the Length. What 

 we have said in regard to these proportions naturally foreshadows that, 

 with Bourgelat, we give preference to the square (cam) horse, one 

 in which the height at the withers is sensibly equal to his length, and 



