49^ - ANALYSIS. 



upon the erroneous basis of preconceived characters for 

 division. So that, because his philosophical idea of a ge- 

 nus was perhaps natural, he considered his genera, as he 

 has instituted them, to be the same. Yet if this great na- ' 

 turalist, instead of commencing with any rule from the an- 

 tennae or other solitary organs, had placed together in a 

 group all those insects which agreed in a majority of cha- 

 racters and habit, — if he had then endeavoured to discover 

 that character which was common to them all, he would 

 probably have had a group which might have been called 

 a natural genus without much error. Far from leading 

 him to indulge any idea of absolute division, the ^' nottz 

 aherrantes" would then have rendered such species as- they 

 distinguished the most valuable of all, as tending to point out 

 to his notice the neighbouring genus. And this Linnasus 

 appears indirectly to have perceived ; for when he says that 

 it is no argument against the validity of a genus, that some 

 species should gradually quit its type, there seems reason 

 to believe that his theoretical notion of genera may, with- 

 out any inconsistency, be reconciled with his maxim, that 

 no saltus exists in nature, which is positively contradicted 

 by his genera, as they are instituted. 



An example was given in the appendix to the former 

 Essay, of a genus containing five types of form, the pro- 

 gression of which returns into itself. This genus, Phanaus, 

 was there proved to be distinguished by a peculiarity of 

 geography, as well as of construction and appearance, 

 while the genus Scarabmis, as there developed, was con- 

 sidered to be probably artificial, hke almost every other 

 that has hitherto been instituted in Entomology. Now to 

 show, if possible, by analysis, that there were grounds for 

 this suspicion ; to show that Phanaus is by no means a 



