Mechanical Illustrations of the Planetary Motions. 327 



which the attraction of one planet produces upon the plane 

 of another's orbit, in the eight different positions described. 



The cause, in the case of the metallic disc, is exactly the 

 same as that which, we saw, produced the precession of the 

 equinoxes. The point immediately under the magnet is at- 

 tracted by the magnet, but the effect, in consequence of the 

 rotation, takes place in advance of that point. The very 

 same cause produces the reality represented by the model, 

 in the case of the planets. The action of the magnet upon 

 a metallic plate is not really different from the action of one 

 planet upon another, as far as the plane of the orbit is con- 

 cerned ; for we may suppose, as we did before in the case of 

 the moon, every part of the plate to be a planet; and the 

 magnet influences each part, as it passes it, in the same 

 manner that the one planet influences the other. 



The circumstance of the magnet's being applied nearer, 

 and more perpendicular to the orbit, than in the case of the 

 planet, does not affect the result except in degree. It is 

 brought near in order to make the effect more apparent; and, 

 as to the perpendicular direction of its action, it may be re- 

 marked that, in the case of the planets themselves, the at- 

 traction of the disturbing planet, when not in the plane of 

 the other's orbit, may be resolved into two forces — one in the 

 direction of that plane, and the other at right angles to it : 

 the former is employed in changing the form of the orbit, 

 the latter in changing its direction : it is the latter that is 

 represented by the magnet ; and, since it is actually perpen- 

 dicular to the plane of the orbit, the position of the magnet 

 truly represents it. 



The effect of the force in the direction of the plane of the 

 orbit, in altering the form of the orbit, might also, perhaps, 

 be shown by means of a magnet acting upon a loose chain, 

 previously made to revolve as a circular ring by centrifugal 

 force ; but the result I have not found to be sufficiently de- 

 cided to be easily observable by the eye. 



I come now to my final application of the same principle 

 which has pervaded all the previously described illustrations 

 of the planetary motions. It is well known to those who have 

 studied the subject, that the theory of Saturn's ring involves a 



