IN RELATION TO ORGANISED BEINGS. 33 



sameness so far as these are concerned. Generalisation is no special 

 power of tlie mind, but attending to, and if useful, naming what is 

 common in several objects or complex ideas : classification is syste- 

 matic generalisation, by the mind desiring to know the resemblances 

 and differences of certain groups of objects which engage its attention. 

 With many kinds of objects all that is needful or useful is easily 

 attained, and if even there are resemblances in different points which 

 allow of different combinations of the same objects, yet which method 

 we select may be either unimportant, or may depend on the special 

 purpose we have in view at the moment, as in a large library, where 

 arrangement of the books according to their subjects would best 

 assist the general student, but for some special purposes resemblance 

 in language, in the size of the volume, or in the mode of binding 

 might be employed, and each of these might be a good classification 

 in reference to the purpose of him who thus combined them, all being 

 founded upon actual resemblances ; nor could any one of them be 

 justly said to be more natural than the other, each proceeding on 

 one definite character and suiting the convenience of him who uses 

 it. When first the study of organised nature was commenced, all 

 that was attempted was to collect together the various descriptive 

 notices of objects observed, as they occurred in different authors, and 

 put them in a form to be conveniently referred to, and for this pur- 

 pose an alphabetical arrangement of the names employed would be 

 first thought of, as enabling any who heard a name to look 

 what had been said of it and by whom it had been employed ; 

 this plan, however, could not long afford satisfaction. The vast- 

 ness of the subject makes it necessary to reduce the objects into 

 large classes, by means of their most general resemblances, and 

 then to break these up into more manageable groups, each marked 

 by some common character, and the attempt once made, subdivision 

 would be carried on with a view to the grand object of enabling the 

 observer, by following out the points of resemblance from the more 

 general to the more particular, to find for himself the name assigned 

 to the object before him, and thus become acquainted with whatever 

 was known of its history. Nothing of this kind can be accomplished 

 by means of an alphabetical catalogue, in using which we must know 

 the name in order to refer to the information, so that any system 

 enabling us to trace an object to its place must be accounted a grand 

 improvement — indeed it required much experience, and long continued 



Vol. XI. c 



