ORIGIN OF THE COLEOPTERA. 
301 
Staphylinidce . — Maxilla with a 1-jointed inner lobe (Xantholinua), or 
the mala broad and setose as in the succeeding families (Platystethus 
and especially Bledius); maxillary palpi 3- and 4 jointed. 
The Staphylinid type of maxilla is simply a modification of the Cara- 
bid, with a tendency to degeneration in the lower genera (Bledius, etc). 
Many larv® in this family are carnivorous. 
ElaleruloH — Maxilla with a 2-jointed lobe or mala; the maxillary 
palpus 4jointed. Antenme 4-jointed, bifurcate as in Carabid larvae; 
mandibles toothed. The maxillae of Elater and Athous are free. While 
generally supposed to be vegetable- eaters (as Agriotes), those larvae 
which live under the bark of trees in mines made by Lougicorn and 
other borers have been shown by Ratzeburg, Dufour and Perris to be in 
part carnivorous, living on Dipterous and Lougicorn larvae, as well as on 
the excrementitious vegetable matter filling the burrows. Perris (Insectes 
du Pin maritime , p. 190) has poinied out the close resemblance of the 
mouth-parts of this family to those of the larval Carabidae. 
In the Scarabaeidffi, Buprestidae, and all the lower families of Coleo- 
ptera, the maxillae are of a rather simpler type than in the foregoing fam- 
ilies; the maxillary lobe, or mala, being simple and more or less fringed 
with stiff hairs. In the Scarabaeidae (Osmoderma), and in Pyrochroa, 
which is carnivorous, the mouth-parts are as complicated as in any; 
but in the Buprestidae and Chrysomelid® they are less developed, 
while they are most rudimentary in form and size in the wood-boring 
weevils and Scolytids ; the antennae and second maxillae and legs also 
share in the degradation of structure cousequeut ou the burrowing lig- 
nivorous habits of the larvae. 
But it is in the so-called hypermetamorphosis of the Meloidae, that of 
the blister beetle (Epicauta) as well as Hornia having been fully de- 
scribed and illustrated by Professor Riley in the First Report of the 
United States Entomological Commission (p. 297-302, PI. IV), that we 
have a clew to the probable origin of the different types of Coleopterous 
larvaB. The metamorphosis of the oil beetle (Melee) originally dis- 
covered by Siebold and Newport and also Fabre, is described in 
different entomological manuals. 149 In brief, the larvae of Meloii when 
hatched are very minute, active, six-legged, slender-bodied creatures, 
parasitic ou wild bees ; as the legs end in three claws the iusects in this 
stage are called “ triungulins.” These larvae attached to the bees are thus 
carried into the nests of the latter, where they feed ou the bee-larvae and 
bee-bread. Ou becoming fully fed, instead of transforming directly into 
the pupa state, they assume a second (coarctate) larval form, entirely 
unlike the first, the body being cylindrical and motionless, with long 
legs; they then attain a third larval stage, the head small and the body 
thick, cylindrical and footless; after this they assuiue a true pupa stage, 
and finally become beetles. 
Professor Riley has traced the hypermetamorphosis of t he blister 
>« See the writer’s “ Guide to the Study of luseots, ” pp. 477—479, figs. 447-451. 
