[ 494 ] 74 



habits of all the other orders. The herbivorous habits of the Orthop- 

 tera, the carnivorous habits of the Hemiptera, the aquatic habits of the 

 Neuroptera, the honey-eating and the parasitic habits of the Hynienop- 

 tera, the leaf-eating- habits of the larvae of the Lepidoptera, and the 

 putrivorous habits of the Diptera — all find their parallel in the ranks 

 of the omnivorous Coleoptera. 



SUGGESTIONS TO AID IN CLASSIFYING THE COLEOPTERA. 



The student will perceive that the primary division of Coleopterous 

 insects is based upon the number of joints m their feet, or tarsi. This 

 character, though apparently of trivial importance, is found to furnish 

 an index to a more natural classification than can be established upon 

 any other single character. In almost all beetles of considerable size, 

 that is, more than quarter of an inch in length, this character is very 

 uniform, or, in other words, the number of tarsal joints in the insects of 

 any one section or family, is remarkably unexceptional. It also has the 

 advantage, in insects of this size, of being easily determined, if not by 

 the naked eye, at least by the aid of a simple lens. 



But the insects which are necessarily the most dilficult to examine 

 and classify are the very small ones, and here the character founded 

 upon the number of tarsal joints not only becomes more difficult to de- 

 termine, but more exceptional, and therefore of less value. We there- 

 fore give the following suggestions to aid the inexperienced student in 

 cases of this kind. 



It will be seen that the whole order of beetles may be di\ided into 

 four sections : — First, the pentamera, or those having five joints in all 

 their tarsi ; second, the heteromera, or those having five joints in the 

 anterior and middle tarsi, and four joints in the hindermost; third, the 

 tetramera, or those having four joints in all the tarsi ; and foiu'th, the 

 trimera, or those having three joints in all the tarsi. 



The last section contains strictly but two families, the Coccinellidop 

 and the Endomychid*, though another family, the ErotyUda?, having 

 four joints in their tarsi, are usually classed with them. The species in 

 this section are so comparatively few in number and usually so well 

 marked by their other characters that the student will seldom have any 

 difficulty in distinguishing them. 



The second and third sections are very nearly unexceptional in the 

 number of the tarsal joints. The i)rincipal difficulty occurs, therefore, 

 with respect to numerous families containing very small species in the 

 first or pentamerous section ; and these are mostly limited to what are 

 known as scavenger beetles, both the club-horned tribe, fCIavicornes ;J 

 and the short- winged tribe, [Brachelytra.) In the mmute species of both 

 of these tribes the number of tarsal joints is very irregular, one of them 



