24 



The very imperfection of this table will constitute its principal 

 utility, because, instead of acknowledging variety as a suitable 

 definition of any particular part or state, the differences of which 

 in respective classes, entomologists have been accustomed to 

 consider characteristic, we find authors labouring to confine a 

 group by what they would wish to consider good and solid cha- 

 racters, which characters they often at last leave so comprehen- 

 sive, as not only to include the class which they had originally 

 intended to define, but also a majority of those other classes 

 which they had supposed previously disposed of. If, in reply, 

 my reader should tell me that my seventh class was somewhat 

 of this too comprehensive kind, I should simply reply that I 

 intended it to be so ; and if my reader happen to know a better, 

 he can interline it in his copy. A space would then be occupied, 

 which has hitherto in all such definitions been really, although 

 not verbally, vacant. 



It is hard to break through the trammels of habit ; it is hard 

 to give up what one has for a long time taken for granted ; it is 

 hard to relinquish favourite schemes, however untenable : an inno- 

 vator, however, is bound to deliberate well and coolly, — is bound 

 to try all the various schemes before him with the test of reason. 

 If the entomologist do this he will find his positive knowledge 

 much less than he expected, — he will perceive that he is book-wise 

 and fact-foolish ; if, therefore, he would wish to arrive at truth, 

 he must strip himself of his borrowed garments and all the theo- 

 retical dogmas he may have, however incautiously, imbibed, and 

 trust entirely to what he has discovered himself, or what has been 

 discovered by those who had no theory to support but truth, — no 

 end to answer but^ amusement ; for your theoretical writers, if 

 they meet with a fact which militates against a favourite theory, 

 will too often suppress it entirely, and on the same principle are 

 ever anxious to magnify to an unnatural size, any slight, and 

 often imaginary, circumstance, which they consider may tell in 

 their favour. Among theories that have been thus established 

 on very weak and insufficient foundations are all dichotomous 

 divisions, especially those in which one group is defined as 

 possessing and the other as wanting any fixed and peculiar 

 character ; a definition, by the by, applicable to nearly all 



