120 



Antenna approximate. Antenna distantes. 



Tarsi didactyli. a. Tarsi didactyli. 



Chennium. Haraotus. 



Ceophyllus. Faronus. 



Cedius. 



Tniesiphorus. 



Cteuistes. 



Tyrus. 



Phamisus. 



Tarsi sesquidactyli. b. Tarsi sesquidactyli. 



Metopias. Batrisus. 



Tarsi monodactyli. c. Tarsi rnonodactyli. 



Pselaphus. Bryaxis. 



Tychus. Eupsenius. 



Claviger. Artlrraius. 



Adranes. Bythinus. 



Articerus (?) Rhexius. 



Euplectus. 



the tarsal claws, which are the bases of Aube's tables, is that the character selected by 

 me is entirely independent of particular stages of development. After the light which 

 has been thrown, by embryological researches, on the true signification of allied ani- 

 mals, scarcely any naturalist will be inclined to deny the existence of series in nature, 

 which, starting from a common origin, of low organization, can be traced in different 

 directions to very diverse forms. The particular members of each series therefore 

 stand in a linear relation to each other, and mark the progressive material evolution 

 of the intellectual idea, previously existing in the mind of the Creator, until the per- 

 fection of this idea is attained in the most highly organized member of each group. 

 Gradually, too, as the primary idea is evolved, others seem to be brought out into 

 greater prominence ; so that the intermediate forms of a small group may be in direct 

 relation with the lowest forms of a group standing higher in the same series. It is 

 therefore obvious, that if any group be defined by a character variable in the develop- 

 ment of similar animals, that group is ipso facto not natural ; because it indicates not 

 a series, but a stage of development. However, it may be itself a particular stage of 

 development in a group of higher value ; but in this case the definition of the entire 

 series will also be added to the definition of the group, and the latter will not be a dis- 

 tinctly limited collection of forms, but only a transitus from lower to higher species. 

 The primary principle of division of any natural group must therefore be independent 

 of any development which takes place after the animal has assumed a definite, pecu- 

 liar, recognizable form. Now here Aube's principles fail, because the number of the 

 articulations of the antennae and the tarsal claws varies from the larva to the imago, 

 and therefore indicates stages of development. Hence, if the group were composed 

 of smaller series, manifesting affinities in different directions, his arrangement would 

 not render them evident, and they would remain rather matters of inference than of 

 demonstration. It is of course very difficult to detect these characters which are in- 

 dependent of development, because they are usually obscured by physiological develop- 

 ments, required by the habits of the animal, and the part it has to play in tlie economy 



