INTRODUCTION. - 17 



that " no more causes are to be admitted as existing in nature, than are true, 

 and sufficient for explaining the phenomena to be considered:" the second, 

 " therefore effects of the same kind are to be attributed, as far as is possible, 

 to the same causes:" thirdly, " those qualities of bodies which cannot be in- 

 creased nor diminished, and which are found in all bodies within the reach of 

 our experiments, are to be considered as general qualities of all bodies exist- 

 ing:" fourthly, "in experimental philosophy, propositions collected by induc- 

 tion from phenomena, are to be esteemed cither accurately or veiy nearly true, 

 notwithstanding any contrary hypotheses, until other phenomena occur, by 

 which they may either be corrected or confuted." 



As an illustration of the remark, that these axioms, though strictly true 

 are of little real utility in assisting our investigations, I shall give an instance 

 from the subject of electricity. Supposing that we wish to determine, whe- 

 ther or no the electric fluid has weight ; we are to inquire whether or no gra- 

 vitation is one of those properties which are described in the third rule, and 

 whether that rule will authorise us to apply it to the electric fluid, as one of 

 those qualities of bodies, which cannot be increased nor diminished, which 

 are found in all bodies within the reach of our experiments, and which are 

 therefore to be considered as general qualities of all bodies existing. Now 

 it appears to be in the first place uncertain whether or no the increase and di- 

 minution of gravity, from a change of distance, is strictly compatible with the 

 terms of the definition ; and in the second place, we are equally at a loss to 

 decide, whether or no the electric fluid can with propriety be called a body, 

 for it appears in some respects to be wholly different from tangible matter, 

 while it has other qualities in common with it. Such are the difficulties of 

 laying down general laws on so comprehensive a scale, that we shall find it 

 more secure to be contented to proceed gradually by closer inductions in par- 

 ticular cases. We shall however seldom be much embarrassed in the choice 

 of a mode of argumentation. The laws of motion, Avhich will be the first 

 immediate subjects of discussion, have indeed sometimes been referred to 

 experimental evidence ; but we shall be able to deduce them all in a satis- 

 factory manner, by means of our general axiom, from reasonings purely ma- 

 thematical, which, wherever they are applicable, are unquestionably prefer- 

 able to the imperfect evidence of the senses, employed in experimental inves- 

 tigations. 



VOL. I. D 



