ON THE STATISTICAL VIEW OF NATURE. 597 



ning you choose, the greater your distance from the 

 average or most probable condition into which, in the 

 long-run, things must settle down ; the more play for 

 the equalising and levelling down of coming events. 

 The world or at least that part of the world accessible 

 to our observation, and the playground of our activity 

 shows a large amount of available energy, or, expressed 

 in a purely statistical manner, it started from a highly 

 improbable condition, and it is descending or running 

 down into a more probable or average condition. The s\. 

 doctrine of availability or of its reverse, of entropy ability" a 



r>/ theorem in 



i.e., of the loss of availability turns out to be a P robabilit y- 

 theorem of probabilities ; and the refined mathematical 

 researches of Prof. Boltzmann and others show that 

 these two conceptions can be made to cover each other. 

 Moreover, we can bring home to the popular under- 

 standing the difference between the exceptional con- 

 dition, with its large amount of available energy, and 

 the average condition, with its large amount of self- 

 destructive and wasted energy (or entropy), by the 

 simile of order and disorder. For every arrangement of 

 a crowd of things or beings which is orderly, there are 

 innumerable arrangements which are disorderly ; every 

 one knows how easily the orderly arrangement lapses 

 into disorder, and nobody expects by mere haphazard 

 or chance movements to produce order out of disorder. 

 There are thousands of ways by which a stone can fall 

 from the peak of a mountain to the lower levels, but 

 only one direction which would take it up again to the 

 top. A tree has been suggested as the picture of the 

 course that natural movements take : for the one position 



