LARVAL FORMS OF COLEOPTERA 



Paired ambiilatorial ampullae absent ; carclo with extensive 



sclerome Cleridae-Clerinae ( Opilo, Trich- 



odes, Thanasimus, Enoclerus, 

 Placoptenis) (pi. 95 A, I, J, 

 M, Q, T, U) 



15. Ocelli five ; more than two of abdominal segments with dorsal 



ambiilatorial ampullae Cleridae-Enopliinae (Neichnea, 



Phyllohaenvs, Chariessa, 

 Cregya) (pi. 95 0) 

 Ocelli four ; only two abdominal segments with ambulatorial 

 ampullae Cleridae-Tarsosteninae (pi. 95 



16. Antenna with sensory appendix absent ; ventral mouthparts ap- 



parently protracted ; ventral surface of head apparently 

 formed bj' fusion of the gular region and the cranial capsule ; 

 hypostomata rod-shaped and diverging posteriorly from the 



fossae for the mandibles Catogenidae^^ (pi. 33 I, J, L, 



_ M, 0) 

 Antennae with sensory appendix present, dilated, and as long as 

 distal and median joints together ; ventral mouthparts re- 

 tracted ; gular area distinct, not fused with head capsule ; 

 hypostomata not rod-shaped, and not diverging, but curved 



toward each other Bofhrideridae^^ (pi. 44 F-W) 



"^ The family Catogenidae, composed of the genera Scalidia and 

 Catogenus, is placed by most authors, and probably correctly, near 

 genera as Laemophloeus and Hemipeplus in the series Cucujoidea. 

 The characters defining the larvae as belonging to the series 

 Cleroidea are probably not fundamental but result from adaptation 

 to a parasitic life. The straight, pointed, rod-shaped h^^postomata 

 which diverge from the' mandibular fossae in a posterior direction 

 are similar to the ones found in Laemophloeus, Hemipeplus, and the 

 Phalacridae, and this similarity indicates strongly that the large 

 subfacial region between the rod-shaped hypostomata in the 

 Catogenidae is homologous with the region between the rod-shaped 

 hypostomata in the above-mentioned cucujoid larvae and partic- 

 ularly with the one in the Phalacridae. Thus the ventral surface 

 of the head in the Catogenidae is probably formed by a fusion of 

 the ventral sides of the cranial capsule, a pair of cardines, the 

 submental region between them, and a gular area posteriorly. The 

 family Catogenidae has therefore been tabulated also on page 35 in 

 the series Cucujoidea. 



"° The relationship of this family is problematic. It has also 

 been tabulated near the Colydiidae on page 40 in the series Cucu- 

 joidea where it most likely has its proper systematic place. How- 

 ever, the larvae are quite different from the larvae of the typical 

 Colydiidae. Their head structures, at least, suggest cleroid affin- 



57 



