common vice. Natural laws are the expression of 

 universal experience, as, all men die, lead sinks in 

 water, etc., for examples, and the truth of such, 

 generalisations can only be impugned by an experience 

 as large as that which established them, 3'et, have we 

 not all seen the hesitation to believe a ghost story, 

 which set such natural laws at defiance, provoke 

 indignant remonstrance from the narrator? 



As regards the connexion which should exist 

 between evidence and probability, Huxley gives an 

 admirable illustration ; here is the gist of it. If a 

 respectable man informed you that he had seen a pie- 

 bald horse in Piccadilly, you would accept his state- 

 ment without demur. If he told you he had seen 

 someone driving a zebra in the same place, you might 

 reasonably look for some corroboration, for zebras in 

 harness are very rare, they are difficult to break — it 

 might have been a horse very like a zebra which only 

 an expert could distinguish. Was your informant an 

 expert? and so on. It might be true, but on the other 

 hand it might not, and you want more evidence before 

 finally accepting it as a fact. But if your informant 

 said he had seen someone driving a centaur you would 

 know that the statement was false, and Huxley said 

 that if Johannes Muller, whom he regarded as the 

 greatest anatomist of his time, had told him he had 

 seen a centaur, he would not have believed him with- 

 out convincing corroboration, and we must agree that 

 is a fair statement of the case. 



The origin of many superstitions might be traced 

 to the time honoured combination of the ignorance 

 of many and the impudence of one. Let us if possible 

 avoid identifying ourselves with either class as students 



