454 SPECIAL METHODS OF DISCOVERT. 



search requisite to be obeyed by all men whilst investi- 

 gating the various sciences ; but how far these rules 

 of the art of discovery can be ascertained, systematically 

 arranged, and made explicit, in the present extremely im- 

 perfect state of scientific knowledge, is a difficult point to 

 determine. As scientific discovery includes the finding 

 of new truths in every branch of natural knowledge, a com- 

 plete art of discovery must be applicable to and coextensive 

 with the entire domain of attainable natural truth. A 

 classification of the modes of discovery is simply a classi- 

 fication, from a new point of view, of the history of 

 scientific discoveries. 



The most systematic arrangement of the methods of 

 discovery is probably according to the various sciences and 

 their subdivisions, and not primarily according to the 

 rules of thought or modes of mental action, because those 

 rules and modes are themselves based upon and developed 

 by our experience of nature, and therefore dependent 

 upon the laws of the various sciences. We can think in 

 discordance with nature, but we cannot usually discover 

 by means of such discordant thought. In so far as the 

 sciences are themselves similar, so far must the methods 

 of investigating them be alike ; and as they are all of 

 them evidently based upon logical, geometrical, and ma- 

 thematical laws, so must the rules of discovery in them 

 conform to those laws. It is evident, then, that each 

 method of discovery must both be in general accordance 

 with logical and mathematical laws, and be specially 

 adapted to the particular science and branch of science in 

 which an investigation is being made. The general method 

 of discovery, so far as it is of a logical character, has 

 already been described in this treatise, and forms the essence 

 of the subject matter of many of the preceding chapters. 



The particular circumstances under which discoveries 



