[ 200 ] 



tliat the attempt is not impradlicable. In his Novum Organum 

 he cautions us againft being difcouraged by the fiiilurc of 

 phyfical experiments, obferving that a negative fometimes brings 

 more light than an affirmative inflance. The errors and per- 

 plexities of metaphyficians are the negative tvjlances of meta- 

 phyfics. They . inform us in what cafes the powers of the 

 hiiman I'nind have failed, and perhaps "it rnay appear that this 

 attempt to afcertain the Umits of thdfe powei-s derives from 

 them an experimental confirmation. 



To enquli'e into ffie original, certainty and extent of human 

 knowledge, together with the grounds and degrees of belief, 

 opinion and affent, was the purpofe of the celebrated Locke, 

 and in the profecution of it ht has made many valuable obfer- 

 vations ; but it cannot be furpriring that in his early efTay 

 there fhould be fome deficiencies which a century of philo- 

 fophic refearches might enable tis to fupply. His general plkh 

 feems to be right, but fome parts of it appear to be erroneous, 

 and others to want the diftindlnefs neceffary for their appli- 

 cation. In the fecond clafs of probable propofitions he has 

 placed the manner of operation in moji parts of the 'uoorks of nature 

 (book iv. chap. i6) ; and yet I beheve upon examination it AVill 

 be found that the manner of operalirm does not entfer into any 

 of the probabilities which he has mentioned as examples, and 

 that it lies beyond the reach even of cdrijedlure. Hfe hrfs 

 been deficient in not giving with fufficient diflindlnefs a 

 general defcription of all thofe propofitions in which demon- 



ftrative 



