Prof. J. Thomson on Crystallization and Liquefaction. 307 



as a general physieo-mechanical principle or axiom, and I think it 

 involves the truth of the opinion just expressed : — 



If any substance, or any sj^stem of substances, be in a coudition 

 in which it is free to change its state (whether of molecular arrange- 

 ment, or of mechanical relative position and connexion of its parts, 

 or of rest or motion), and if mechanical work be applied to it (or put 

 into it) as potential energy, in such a way as that the occurrence of 

 the change of state will make it lose (or enable it to lose) (or be ac- 

 companied by its losing) that mechanical work from the condition of 

 potential energy, without receiving other potential energy as an equi- 

 valent ; then the substance or system ivill pass into the changed 

 state. The consideration of a few cases, in some of which there is 

 not freedom for the substance or system to change its state, and in 

 others of which there is freedom, will render the meaning of this 

 more clear. 



Gunpowder may be cited as an example of a substance in a condi- 

 tion not free to change its state, although when it is made to explode 

 by a spark, it passes to an altered condition, and, in doing so, even 

 gives out a great amount of mechanical work. That is to say, that 

 on the whole it is more than free to change to the exploded state, or 

 it tends so to change, but there is some kind of obstacle at ordinary 

 temperatures, to the change, which either vanishes at a high tempera- 

 ture, or requires the application of mechanical work to begin the over- 

 coming of it. "When the change is once begun, the requisite help is 

 given to the succeeding parts by those which have gone off first. 



Again, water confined in a high reservoir is not free to go to a 

 lower one ; although a siphon, primarily filled with water, may help 

 the parts successively over the obstacle by lending to each the requi- 

 site mechanical work in advance, which it afterwards pays to the parts 

 which are to follow, besides that it gives out in its fall a great addi- 

 tional amount of power or energy applicable otherwise. Two reser- 

 voirs of water, on the same level, and having an opening between 

 them under the water surface, would represent the case of perfect 

 freedom for change of state ; and two on a level with one another, 

 but separated by a partition, would represent the case in which no 

 mechanical work would finally be either given out or absorbed by the 

 change, but in which there is not perfect freedom to change, until a 

 siphon or other means of help is applied. 



A bell hung from an axle and then turned up, and left resting 

 against a stop a little beyond its position of unstable equilibrium, is 

 not free to go down, but a slight pull will bring it over this position 

 and make it free to swing, which the work stored as potential energy 

 in the raising of it from its low or hanging position, will cause it to 

 do ; its fall till it comes to the bottom being essentially accompanied 

 by the loss of that potential energy, as such, though not as actual 

 energy, out of the system of which it and the earth are the two parts, 

 and in which change of their distance asunder constitutes change of 

 their potential energy. 



If in an atmosphere of steam resting on water at its boiling tem- 

 perature for the pressure of the steam ; as, for instance, in the inside 



