132 Mr Glover on Forms of Induction. 



of that logical fiction of which so much is said by Dr Whately 

 and others. Thus, to give an example which may perhaps il- 

 lustrate an oversight of writers on induction, we quote the fol- 

 lowing from Dr Brown : — " If, by the term general law," he 

 says, " be meant the agreement in some common circumstances 

 of a number of events observed, there can be no question but 

 the view is a just one, and that what we have already found 

 in a number of events, may be applicable to that number of 

 events ; in the same manner, as, after having combined in the 

 term animal the circumstances in which a dog, a horse, and a 

 sheep agree, we cannot err in applying the term animal to a dog, 

 ahorse, or a sheep. But the only particulars to which we can, 

 in this case, with perfect confidence apply the term animal, are 

 the very particulars before observed by us." * Now, here we 

 perfectly acquiesce with the argument of Dr Brown : but it is 

 sufficiently obvious, from the examples already given, that his 

 observations do not apply to all cases of inductive inference. 

 Logicians generally give examples of inductive generalization, 

 which do not shew the occasional and frequent power possessed 

 by facts of exhibiting the pi'operties of their class in a distinct 

 point of view, as in the above instance of Dr Brown. But 

 let the example produced by him be contrasted with the one 

 given by us of the mode in which a correct notion of the con- 

 nexions and relations of some properties in all animals could 

 be attained, from an investigation of them in a few, and it will 

 appear that his example does not give a fair view of the entire 

 character of the inductive process. 



It may be inferred, therefore, that some distinctions should 

 be drawn as to the methods of inductive procedure ; and we shall 

 now attempt to classify, and define, at least the more prominent 

 modifications. These seem to be nearly as follows : — 1. There 

 is a form of procedure, in which, in order that a law may be 

 expressed with logical precision, or possess physical certainty, 

 it is absolutely necessary for every individual instance included 

 in the original statement to be examined. Thus, in systematic 

 Botany, when the external conformation of a class of plants has 



• Brown's Lectures, vol. 1. sect. 8. p. 176. 



