EMPIRICAL LAWS. 41 



3. Now this last-mentioned element in the resolution 

 of a derivative law, the element which is not a law of causa 

 tion, hut a collocation of causes, cannot itself be reduced to 

 any law. There is (as formerly remarked*) no uniformity, 

 no norma, principle, or rule, perceivable in the distribution 

 of the primeval natural agents through the universe. The 

 different substances composing the earth, the powers that 

 pervade the universe, stand in no constant relation to one 

 another. One substance is more abundant than others, one 

 power acts through a larger extent of space than others, with 

 out any pervading analogy that we can discover. We not 

 only do not know of any reason why the sun s attraction and 

 the force in the direction of the tangent coexist in the exact 

 proportion they do, but we can trace no coincidence between it 

 and the proportions in which any other elementary powers in 

 the universe are intermingled. The utmost disorder is appa 

 rent in the combination of the causes ; which is consistent with 

 the most regular order in their effects ; for when each agent 

 carries on its own operations according to an uniform law, 

 even the most capricious combination of agencies will gene 

 rate a regularity of some sort ; as we see in the kaleidoscope, 

 where any casual arrangement of coloured bits of glass 

 produces by the laws of reflection a beautiful regularity in the 

 effect. 



4. In the above considerations lies the justification of 

 the limited degree of reliance which scientific inquirers are 

 accustomed to place in empirical laws. 



A derivative law which results wholly from the operation 

 of some one cause, will be as universally true as the laws of 

 the cause itself; that is, it will always be true except where 

 some one of those effects of the cause, on which the derivative 

 law depends, is defeated by a counteracting cause. But when 

 the derivative law results not from different effects of one 

 cause, but from effects of several causes, we cannot be certain 

 that it will be true under any variation in the mode of coexis- 



* Supra, book iii. ch. v. 7. 



