330 COSMIC PHILOSOPHY. [it. 11. 



deposited from a solution, there is a certain point op to which 

 the retention of motion keeps the crystal's molecules from 

 uniting ; but as soon as this point is passed, the motion is 

 suddenly lost, the crystal solidifies, and there is no further 

 redistribution of its particles. Conversely, when a molten 

 metal is allowed to cool until it assumes a plastic semi-fluid 

 state, its molecular motion is lost so slowly that a perceptible 

 rearrangement of parts is possible : currents may be set up 

 in it, gravity will cause it to spread out wherever it is not 

 confined at the side, and pressure here and there will variously 

 mould it. But when it becomes solid, the rearrangements 

 which occurred latest become permanent, and further re- 

 arrangements cannot be produced save by a fresh supply of 

 molecular motion. In like manner, when we come to study 

 planetary evolution, we shall find strong reasons for believing 

 that on small bodies, like the moon and the asteroids, which 

 have rapidly lost their internal heat, there has been but 

 little chance for such complex secondary rearrangements as 

 have occurred upon our relatively large and slowly cooling 

 earth. 



Even after the attainment of solidity, however, a new 

 supply of motion from without may cause some further 

 redistribution without causing the body to relapse into 

 fluidity. Thus a wrought-iron rail, which when new is 

 tough and fibrous, gradually acquires the brittle crystalline 

 texture of cast-iron, under the influence of the vibrations 

 communicated by the cars which pass over it. And the 

 magnetization of steel rods, when fastened in the meridian 

 and frequently jarred, is cited by Mr. Spencer as a fact of 

 like import. Many other excellent illustrations, gathered 

 from physics and chemistry, may be found in the thirteenth 

 chapter of the second part of " First Principles." 1 



1 Throughout this work, reference is made only to the second and re- 

 written edition of " First Principles," London, 1867. The statement of thf 

 law of evolution, as contained in the first edition, is much less complete ani 

 coherent. 



