438 THE INSTABILITY OF THE HOMOGENEOUS. 



all surrounding portions. And when the blow is violent 

 enough to fracture the mass, we see, in the radial dispersion 

 of its fragments, that the original momentum, in being dis- 

 tributed throughout it, has been divided into numerous 

 minor momenta, unlike in their directions. We see that 

 these directions are determined by the positions of the parts 

 with respect to each other, and with respect to the point of 

 impact. AVe see that the parts are differently affected by 

 the disruptive force, because they are differently related to 

 it in their directions and attachments — that the effects being 

 the joint products of the cause and the conditions, cannot be 

 alike in parts which are differently conditioned. A 



body on which radiant heat is falling, exemplifies this truth 

 still more clearly. Taking the simplest case (that of a 

 sphere) we see that while the part nearest to the radiating 

 centre receives the rays at right angles, the rays strike the 

 other parts of the exposed side at all angles from 90° down 

 to 0°. Again, the molecular vibrations propagated through 

 the mass from the surface which receives the heat, must pro- 

 ceed inwards at angles differing for each point. Further, 

 the interior parts of the sphere affected by the vibrations 

 proceeding from all points of the heated side, must be dis- 

 similarly affected in proportion as their positions are dis- 

 similar. So that whether they be on the recipient area, in 

 the middle, or at the remote side, the constituent atoms are 

 all thrown into states of vibration more or less unlike each 

 other. 



But now, what is the ultimate meaning of the conclusion 

 that a uniform force produces different changes throughout 

 a uniform mass, because the parts of the mass stand in differ- 

 ent relations to the force? Fully to understand this, we 

 must contemplate each part as simultaneously subject to 

 other forces — those of gravitation, of cohesion, of molecular 

 motion, &c. The effect wrought by an additional force, 

 must be a resultant of it and the forces already in action. If 

 the forces already in action on two parts of any aggre- 





