38 MODERN CLASSIFICATION OF INSECTS. 



Fabricius added a great number of genera, and distributed the order 

 into ten divisons according to the form of the antennas. The Tarsal 

 System of Olivier has, however, been almost universally adopted, chiefly 

 in consequence of Latreille having employed it in his numerous 

 works. It would appear, however, from the observations of the last- 

 named author, that he has always doubted whether its rigid adoption 

 might not be inconsistent with nature. Thus, in several of his most 

 valuable works, and especially in the Genera Cricstaceorum, &c., we 

 find him questioning the series of Coleoptera founded thereon ; adding, 

 " Coleoptera palpis sex, robustiora, SLgiWora., primates ordinis*; rhyn- 

 cophora e contra instrumentorum cibariorum exiguitate, ore ros- 

 trato, debiliores, Mammalia edentata hie quodam modo aemulantes. 

 Articulorum tarsorum progressio numerica decrescens in methodo 

 naturali non admittenda." (Vol. i. p. 172.). In his Considerations 

 Generales, he has entered more fully into this subject, discussing the 

 distributions resulting from the employment of the antennae, tarsi, and 

 trophi, as leading characters, again giving the preference to the latter, 

 but admitting his inability to propose a natural system, and therefore 

 adopting the tarsal system. 



Against this system it may be advanced, that by rigidly employing 

 the numerical gi*adations of the tarsal joints, the real affinities of cer- 

 tain insects (such as the Pselaphidae and Clambus, above mentioned) 

 are disturbed. It appears to me, however, that these ought rather 

 to be regarded as exceptions, such as constantly oppose the adoption 

 of every rule, especially when founded upon a single character. I 

 do not, therefore, with Mr. MacLeay f^ regard these instances as 

 affording in themselves a sufficient ground for overthrowing the tarsal 

 system. But another and more forcible objection has been raised by 

 Mr. MacLeay, namely, that the Tetramei'ous Beetles are, in fact, Pen- 

 tamerous, and that the Trimerous are Tetramerous ; each of these 

 groups possessing a minute joint, observable between the two lobes of 

 the supposed penultimate joint. It appears to me, however, that this 

 objection can only be maintained against the names of these groups, 

 and not against the groups themselves, which, as it seems to me, are 



* Mr. Kirby more recently, on account of their perfection and svmmetrv of 

 general structure, the lightness and velocity of their motions, and brilliancy of 

 colours, has regarded the Cicindelida? as the analogues of the Lion amongst quadru- 

 peds, and as the typical and most perfect group of Coleopterous insects, 



f Linn. Trans, vol. 15. 



