SERRICORXIA. 



131 



insects, with a fine but dense pubescence ; they are sluggish in their 

 movements and live in galleries which they excavate in soft mud 

 near pools and small lakes, or on the nnrgins of muddy streams, 

 the ramifications of these burrows being very conspicuous. The 

 species have been supposed to be carnivorous in their habits, but 

 this has not been proved and it is believed that they eat the mud 

 into which they burrow. They have, in very many cases, a strong 

 family resemblance, and it is often exceedingly hard to determine 

 the species. 



The larvae are cylindrical, but the thoracic segments are con- 

 siderably broader than the abdominal segments and are furnished 

 with chitinous plates ; the abdomen is subparallel-sided, gradually 

 narrowed before the apex, and ends in a small projecting pseudo- 

 pod ; there are no cerci ; the antennae are rudimentary and the legs 

 short and rather stout ; the whole surface is thickly clothed with 

 fine pubescence, with long outstanding hairs at the sides ; the 

 stigmata are differently arranged from thosk of the larvae of the 

 Dryopidje with which they have been associated, but from which 

 they differ in several important points. 



Division 3. SERRICORNIA. 



For the sake of convenience the name Serricorxia has been 

 retained for the families included in f his section, but the antennae 

 vary very greatly, being filiform in Cantnaris (Telephones) and 

 clavate in Corynetes and certain other Clerio.e ; the gradual 

 transition, however, is very remarkable, the Caxtharix.e soon 

 branching off into flabellate or even plumose forms, while the 

 irregularly clavate-serrate antennae of Aulicus and Dorcatoma lead 

 naturally to Corynetes. The purely serrate antennas are found among 

 the Elaterid.e and their allies, but among these we find strongly 

 developed pectinate forms as in Corymbites. The Dasctllid.e ought 

 perhaps to be referred to the Byriihid.e and Dryoptd.e and classed 

 with them as one family, but the remainder of the families form a 

 distinct, although somewhat heterogeneous group, which ought 

 probably, as we have said before, to be placed before the Clayi- 

 corjSita in the phylogenetic series. The larvae are excessively 

 variable, both in form and habitat. 



Key to the Families. 



I. Prosternum not prolonged behind the 

 anterior coxae (except slightly in certain 



DASCILLIDiE). 



i. Tarsi 5-jointed. 



1. First ventral segment not elongate. 

 A. Onychium large and hairy ; pos- 

 terior coxae sulcate ; antennae 



usually flabellate in the male . . Rhipiceridae, p. 134. 



k2 



