40 SODDY, Evolution of Matter by Radio-active Elements. 



direction there occurs a complementary process in the 

 opposite direction. This, if practicable, certainly offers a 

 more consistent solution of the problem. The difficulty 

 in the past to this view seems to have been the Second 

 Law of Thermodynamics, that when a change of energy 

 occurs spontaneously in the one direction, it cannot of 

 itself proceed in the opposite direction. But it would seem 

 that in this law, as stated, there is the same confusion 

 between personal inability and natural impossibility, 

 which has been pointed out elsewhere. Because the 

 chemist's atom never suffered subdivision in known pro- 

 cesses, it was in danger of being regarded as indivisible 

 with reference to all the agencies of nature, known and 

 unknown. Is not the whole question of the possibility of 

 a perpetual cycle of energy-change in the processes of 

 cosmical evolution in a somewhat similar position to that 

 occupied by the atomic theory before the discovery of 

 radio-activity ? Clerk Maxwell has shown by his ' sorting- 

 demon ' that it is very easy to imagine a process by 

 which heat of itself would pass from a lower to a higher 

 temperature, and history indicates that the resources of 

 Nature are generally much beyond the utmost capacity of 

 the human imagination. On future discoveries in this 

 connection depends whether science is or is not forced to 

 accept a limited philosophy with regard to the universe, in 

 which the idea of a creative agency is the necessary 

 starting point, and universal death the ultimate goal. 



The difficulties in avoiding this view are certainly 

 very great. We know from the energy emitted by a 

 disintegrating atom, the absorption that must take place 

 if that atom is to be reconstructed out of its constituents, 

 and this is so great that, even if heat of uniform tempera- 

 ture is considered available, it is difficult to imagine how it 

 can be furnished from its surroundings unless a consicjer^ 



