160 NOVUM OROANUM 



rection of induction; 4. Of varying the investigation ac 

 cording to the nature of the subject; 5. Of the prerogative 

 natures with respect to investigation, or of what should be 

 the first or last objects of our research; 6. Of the limits of 

 investigation, or a synopsis of all natures that exist in the 

 universe; 7. Of the application to practical purposes, or of 

 what relates to man; 8. Of the preparations for investiga 

 tion; 9. And lastly, of the ascending and descending scale 

 of axioms. 38 



XXII. Among the prerogative instances we will first 

 mention solitary instances. Solitary instances are those 

 which exhibit the required nature in subjects that have 

 nothing in common with any other subject than the nature 

 in question, or which do not exhibit the required nature in 

 subjects resembling others in every respect except that 

 of the nature in question; for these instances manifestly 

 remove prolixity, and accelerate and confirm exclusion, so 

 that a few of them are of as much avail as many. 



For instance, let the inquiry be the nature of color. 

 Prisms, crystalline gems, which yield colors not only inter 

 nally but on the wall, dews, etc., are solitary instances; for 



that such a classification is much of the same nature as if, having to teach the 

 art of building, we were to describe tools with reference to the amount and place 

 of the work which they must do, instead of pointing: out their construction and 

 nse; as if we were to inform the pupil that we must have tools for lifting a 

 stone up, tools for moving it sidewise, tools for laying it square, and tools for 

 cementing it firmly. The means are thus lost in the end, and we reap the fruits 

 of unmethodical arrangement in the confusion of cross division. In addition, 

 all the instances are leavened with the error of confounding the laws with the 

 causes of phenomena, and we are urged to adopt the fundamental error of 

 seeking therein the universal agents, or general causes of phenomena, without 

 ascending the gradual steps of intermediate laws. Ed. 



38 Of these nine general heads no more than the first is prosecuted by the 

 author. 



