380 THE EXTERIOR OF THE HORSE. 



entirely erroneous. This is not, however, an irrefutable argument. 

 At every moment, in physiological experiments, we find ourselves face 

 to face with such difficulties, and notwithstanding the failures, notwith- 

 standing the errors every day, these experiments are renewed, and sci- 

 ence constantly goes on reaping benefits from them. Must we say, then, 

 that this idea should be renounced ? Must we- reject beforehand and 

 with determination the facts acquired for this reason alone, that they 

 may have been badly collected, wrongly studied, wrongly interpreted, 

 or that they are simply approximative ? Certainly not ; this would 

 prove a deplorable state of the scientific mind. 



Far from us be the thought of drawing this conclusion from the 

 criticisms of M. Neumann. Our distinguished colleague (and he is to 

 be praised for this) has signalized the danger ; he has appealed to cor- 

 roborating observations ; he has shown and demonstrated with proof 

 the inevitable errors into which investigators incompletely prepared 

 would fall, and we cannot too warmly approve his aim. All eyes 

 are not apt to see well, all hands are not skilled in exploring, all 

 horses are not suitable for this kind of study. Let it be well noted 

 that if the results obtained by two observers do not agree in an abso- 

 lute manner, if even their differences appear exaggerated, it is not less 

 certain that these results, for each of them, can still be compared, for 

 the chances are great that whatever error is committed is repeated 

 everywhere the same way. In M. Lemoigne's opinions and in ours 

 the differences are insignificant ; we are in accord as to his articular 

 angles. In table B (page 377) we gave only the angular relations of 

 the bones in fast horses, — that is to say, in animals with an oblique 

 shoulder, a straight arm, a horizontal croup, and a straight leg ; this 

 shows why our scapulo-humeral and humero-radial angles are a little 

 more closed and our tibio-tarsal angle a little more open, and even then 

 the opening of the last results, perhaps, from the fact that in all our 

 estimations we have considered the canon as vertical. 



Let us try now, as M. Neumann has done above in order to refute 

 the theory of General Morris, to estimate, with our angles, the vertical 

 height of the members of a horse of medium size. 



Let the anterior member be represented schematically with its axis of move- 

 ment (Fig. 139). The heights a, b, c of the segments which do not themselves 

 lie parallel with the vertical line are the sides of as many rectangular triangles, 

 whose hypothenuse and one of the adjacent angles are known. It is then easy 

 to calculate them by vulgar trigonometrical formulae, and we have : 



a = 0.41 m. sin. 60° = 0.35507 m. 

 6 =- 0.31 m. sin. 5.5° = 0.25394 m. 

 c = 0.17 m. sin. 60° = 0.14722 m. 



