SELECTION OF A SUBJECT OF INVESTIGATION. 373 



sequence of our very imperfect power of prediction we are 

 very apt to value wrongly the importance of an unmade 

 research; some of the greatest truths have been ulti- 

 mately disclosed by investigating what previously appeared 

 to be most trivial phenomena. The discovery of static 

 electricity is said to have arisen from the circumstance 

 that a bit of amber, by being rubbed, acquired the pro- 

 perty of attracting a feather. That of magnetism was 

 probably equally simple, and is said to have been first 

 observed in a piece of loadstone. That of voltaic elec- 

 tricity also arose from an apparently trivial circumstance, 

 which has already been described in this book. There 

 are, however, cases where the investigator knows before- 

 hand, with considerable certainty, from the nature of his 

 proposed question and the conditions of his experiment, 

 that if the hoped-for positive results are obtained, they 

 must necessarily be important. The successful search by 

 Faraday for magneto-electric induction, and his unsuccessful 

 one for an experimental connection between gravity and 

 the other physical forces, were instances of this kind. And 

 there are other cases where the investigator knows before- 

 hand that he is nearly certain to produce some new re- 

 sults, but is unable to foretell what they will be. In the 

 first of these cases he wishes to know what conditions will 

 render evident a particular new and important effect ; and 

 in the second, what effects will result from a particular 

 cause or class of circumstances. 



The investigator, in selecting his subject, has also to 

 consider whether the conditions of the proposed research 

 are sufficiently ripe, and that is another very knotty point. 

 Even Newton himself could not discover the universal action 

 of gravity when he first attempted to do so, because he 

 made his first endeavour before the conditions were ready, 

 and that could not have been ascertained without a trial. 



