594 TAXONOMY. 



\ 



sometimes exactly alike in truly different genera ; but the stability of 

 genera depends in many instances much more upon the judicious 

 balancing of all the parts of the body, and their differences, in which 

 case it is only the above mentioned happy tact that can securely 

 guide the observer. For, however fixed the number of the joints in 

 the antennae, for example., may be in certain families, yet instances 

 occur in which they are subjected to much variety. The genus Cimbex 

 is a case in point, the antennae of which consists sometimes of six, 

 sometimes of seven joints, of which sometimes one, or two, or three of 

 them are swollen into a knob. If we compare with this the number 

 of the joints in the antennae in the whole family of the saw-flies, we 

 shall speedily perceive that genera formed merely from the number of 

 those joints cannot absolutely be considered as natural. This number 

 is still more variable in the genus Forficula, in which almost every 

 species has a different number, and which likewise is not even uniform 

 in the individuals of the same species ; and yet Leach has formed dis- 

 tinct genera founded on these differences. 



331. 



The sum of the characters adopted for the definition of the genus forms 

 the generic character, which is either natural when deduced from all 

 the organs, or artificial when it merely refers to the characters admitted 

 as the basis of the classification. This again constitutes the essential 

 character when it merely cites the distinguishing marks of the genus. 

 The generic description refers neither to the natural nor to the artificial 

 generic character, but presents the entire form, even to its most minute 

 divarications ; it is, as it were, a figure in words, whereas the characters 

 depict in words particular organs only. 



III. IDEA OP THE SUPERIOR GROUPS. 



332. 



It is from characters of greater generality, especially from resem- 

 blances in form, or the similar structure of certain parts, as the feet, 

 wings, &c., which yield no generic characters, that the genera group 

 themselves into superior divisions. This grouping must follow the 

 principles adopted in characterising genera, if no violence is to be done 

 to nature. We must here also strive fur an equality of value in the 

 groups formed. 



