46 ILLINOIS BIOLOGICAL MONOGRAPHS [46 



are almost exactly like these parts of such genera of the Mordelloidea as 

 Epicauta, Cephaloon, and Macratria. The corpotentorium of Arthro- 

 macra and Epicauta is, however, lacking. The structure of the various 

 parts of the head-capsule of the representatives of the Bostrichidae varies 

 considerably. The epicranial arms are complete in Bostrichus, Lyctus, 

 Plesiocis, and Sphindus. With the exception of Sphindus these families 

 show a distinct ventral migration of the antennariae and pretentorinae. 

 In contrast, Ptinus and Sitodrepa do not possess complete arms, and the 

 antennariae and pretentorinae are located more dorsad, markedly so in 

 the former. On the ventral surface, there is considerable variation in 

 the position of the metatentorinae. The tentorium also varies a great 

 deal. This is probably not a very homogeneous group. Sphindus seems 

 to be structurally more similar to Sphaerius than any other genus studied. 



The Scarabaeoidea (Figs. 112-119, 260-267, 405-413, 549-556) is a 

 very homogeneous group, characterized by a heavy chitinization of the 

 head-wall, degeneration of the epicranial suture, ventral migration of the 

 antennariae and the pretentorinae, and a complete gula, except in Pseudo- 

 lucanus, in which the gular sutures extend no more than half the distance 

 between the occipital foramen and the submentum. In the development 

 of the \entral surface and perhaps the tentorium, the Scarabaeoidea 

 should be classed with the Adephaga and related groups. 



The Cerambycoidea (Figs. 120-135, 268-283, 414-429, 562-572) is 

 also a very homogeneous group, characterized by a weak chitinization of 

 the head-wall, a distinct dorsal position of the antennariae and the pre- 

 tentorinae, and a more or less well developed epicranial suture. The 

 gular sutures are generally short. The tentorium is delicate and mem- 

 branous throughout. 



The members of the Rhynchophora (Figs. 136-148, 284-296, 430-442, 

 573-585) may or may not possess a snout. They may or may not possess 

 confluent gular sutures. When they do, the condition is the same as in other 

 Coleoptera, except that the invaginations are apt to be more extensive. 

 They may or may not possess a labrum. There is no definite character of 

 the head-capsule by which the Rhynchophora as a group can be separated 

 from other Coleoptera. The affinities of this group are not at all clear. 



The condition of the dorsal surface throughout the groups is most 

 variable, and can be relied upon but little to furnish evidence of the degree 

 of relationship. On the ventral surface the condition is much more 

 stable, and probably much more reliable in indicating affinities. The 

 tentorium is less variable than the dorsal surface, and less stable than 

 the ventral surface. Weighing the evidence presented in this study, two 

 large groups of most of the families can be made. The Lymexyloidea, 

 the Cupesoidea, the Mordelloidea, the Elateroidea, the Dryopoidea, the 

 Dascilloidea, the Byrrhoidea, the Cucujoidea, the Tenebrionoidea, 



