The Limits of Physical Science. 27 



not invent a conjectural state of things, and profess, 

 by setting out the operation of known laws in such 

 supposed circumstances, to extend the bounds of real 

 knowledge. All scientific work of that kind and it 

 forms a considerable part of popular science is 

 merely imaginative ; it lies in the department of" 

 romance. 



There are bounds which ought not to be passed in 

 framing hypotheses to account for phenomena that 

 remain unexplained. It is not allowable to set up any 

 hypothesis which lies outside the possibility of being 

 established by evidence. Every legitimate conjecture 

 lies within the possibility of actual proof. It is fatal 

 to any hypothesis to show that it is incapable of being- 

 proved. Its author is bound to point out a possible 

 line of adequate evidence, or his conjecture must be 

 rejected as illegitimate. The doctrine of universal 

 evolution belongs to this class of illegitimate hypo- 

 theses ; if it were true, it could not be proved true. 

 The requirements of satisfactory proof transcend the 

 limits of human thought. 



Every attempt to formulate truths derived from 

 experiential science that shall be held to be good over 

 the whole past, present, and future of concrete being, 

 encounters insuperable difficulties. 



Man is not omniscient: he must gather his facts 

 from experience, and must interpret them by compari- 

 son and inference. At every step he comes in contact 

 with the inscrutable ; gmd his intelligence is in every 



