MODERN CONCEPTS IN PHYSICS — LANGMUIR 239 



He suggests that the second law of thermodynamics is the only iunda- 

 mental law of nature which provides us with any distinction between 

 future and past. One way of stating this law is that all spontaneous 

 proi esses that occur in nature involve an increase in entropy. Ed- 

 dington thus proposes that the positive direction of time can bo 

 deiined as that direction in wiiich the entropy increases. If we had 

 a system in absolute equilibrium the entropy would be constant, and 

 there would then be no arrow of time. This is in accord with the 

 fact that in such a system there are no changes with time. 



It is improbable that there are two independent fundamental fac- 

 tors which ])rovide an arrow for time, so that it would seem that 

 Eddington in having found one such factor has found the only one. 

 There are, however, grave difliculties with this view. An arrow is a 

 vector quantity which should have magnitude as well as direction. 

 Now the rate of change of entropy does not seem to give us any 

 measure of time. For this purpose we use phenomena which are as 

 nearly reversible as possible, such as the swinging of a pendulum in 

 a vacuum. 



Fundamentally entropy is a measure of randomness. A random 

 distribution of molecules in space and velocity is a system having the 

 maximuHi entropy. If we throw a pack of cards out of the window 

 and collect them from the ground they have become effectively 

 shuffled. We would not expect by this process, starting with a 

 shuffled pack of cards, to find them at the end in the order in which 

 they come from the manufacturer. The direction in which the ran- 

 domness increases thus provides an arrow for time. This arrow is, 

 however, equivalent to that involving the increase of entropy. 



It is still an open question, however, whether processes directed by 

 intelligent beings may not involve a decrease in entropy. In fact it 

 seems conceivable that the evolution of organic life on the earth is in 

 some measure fundamentally contrary to the second law of thermo- 

 dynamics. The inherent tendency of evolution seems to be to bring 

 about an ordered rather than a random arrangement of parts, and in 

 the future perhaps forms of life may evolve which cause a decrease 

 of entropy on a large scale. Are we then to have some parts of the 

 universe in which the arrow of time points in the opposite direction 

 from that in neighboring parts? 



Such speculations may seem fantastic. It is, however, I believe, of 

 the utmost importance for the chemists and the physicists to evolve 

 fundamentally sound conceptions of such things as time and entropy. 



The profound changes in physical thought, particularly those 

 represented by the quantum theory, arc rapidly bringing about a 

 revolution in physical chemistry. The third law of thermodynamics 

 involving chemical constants has changed ra(lic:dly oui- methods of 



