11 



results themselves may frequently be modified (or, ap- 

 parently, even reversed), from counter influences of 

 divers kinds. This touches not, however, the existence 

 of the law; and the effect is not the less specifically 

 dependent on its o'wn peculiar cause, because those 

 "counter influences" prevail, — and because different 

 effects may chance, therefore, to be occasionally brought 

 about by causes which may possibly seem to be identical. 

 We should, rather, bear in miad that the agents which 

 operate ia moulding the outward contour of organic 

 beiags are various, and capable inter se of permutations 

 innumerable; so that it is only on a broad scale that 

 parallel results can be looked for in creatures severally 

 exposed to the action of elements, which are liable to be 

 differently compounded from what may primd facie 

 appear to be the case : and that, consequently, where 

 opposite phsenomena are displayed under circumstances 

 seemingly coincident, our first object should be [not to 

 regard the phsenomena as indicative, that no constant 

 result can be anticipated from causes which are similar, 

 but), to inquire whether the circumstances in question 

 are really coincident or not, — seeing that some counter- 

 acting stimulus may have been, here or there, unex- 

 pectedly at work, which shall enable us, so soon as it is 

 detected, to account for the discrepancy. 



It is by this process alone that we can hope to make 

 real use of analogy, without abusing it : for whilst there 

 is danger, on the one hand, of needlessly rejecting the 

 argument which it suggests to us, through opposite 



