274 FIRST PRINCIPLES OF NATURAL CLASSIFICATION. 



as we have before explained *, may be accounted for in 

 two ways, either by our imperfect acquaintance with 

 the productions of nature, or by the extinction of those 

 animals which would render such groups perfect. 



{335.) Having now laid before the reader a few of 

 those facts which serve to verify the general truth 

 of the five propositions with which this division of our 

 volume commenced, we must revert to a subject in- 

 tim.ately connected with the definite character of natural 

 groups, and of which they are^ in fact, composed ; we 

 mean species and varieties — those individuals, in short, 

 which constitute the assemblages in question, and whose 

 variation leads to a knowledge of all higher combin- 

 ations. We alluded to a theoretic belief, even now com- 

 mon among naturalists, that species are the only absolute 

 divisions of nature. So far, however, from such being 

 the fact, we believe that the truth consists in this posi- 

 tion being reversed ; in other words, that if there are 

 any absolute natural divisions, they are to be found in 

 the different gradations of groups and types here pointed 

 out, but that in numberless cases it is utterly impossible 

 to discriminate species from varieties ; species, in short, 

 being, to human apprehension, the most indeterminate 

 of all the links in the chain of being. This opinion is 

 borne out by the sentiments of one whose peculiar line 

 of study renders him, on this subject, one of the highest 

 authorities in this country, f Setting aside, however, 

 those exceptions which give rise to these opinions, and. 

 where the discrimination of species from varieties is 

 impossible, we shall now proceed to describe those pe- 

 culiarities which generally constitute a species ; and we 

 do this fully, because we think the subject has not re- 

 ceived that attention, in introductory works, which it 

 merits. 



{336.) A species, in popular language, may be de- 

 fined as " a natural object, whose differences from those 

 most nearly related to it are, as far as observation has 



* See Preliminary Discourse on Nat. Hist. p. 21.'3. 



-(■ J. F. Stephens's Catalogue of British Insects, preface, p. xvi. 



