LARVAL FORMS OF COLEOPTERA 



Two ocelli on eacli side (second ocellus small or even absent in 

 species of Horiinae) ; ninth abominal segment carrying 

 terminally one pair of small bristles, or no bristles; body 

 either fairly elongate with posteriorly attenuate abdomen, 

 or more often, navicular. (Fifth larval instar enveloped in 

 exuvia of the two preceding instars; sixth instar as well as 

 pupa inside of case formed by the exuvia of the three pre- 

 ceding larval instars) 4 



4. Spiracles of eighth abdominal segment placed normally 



Meloidae-Horiinae'- 

 Spiracles of eighth abdominal segment placed terminally on 

 hook- or wartlike j)rojections (spw. pi. 96 N, 0) 



Meloidae-Nemognathinae (pi. 96 

 L, N, 0, R, S) 



5. One ocellus on each side of head ; spiracle of first abdominal 



segment very large and placed on a laterally projecting, flat 

 lobe (pi. 97 B and D) ; claw falciform, with a single bristle 

 at base ; labial palpi vestigial and wart-shaped'^ 



Tetraonycidae (Based on Tet- 

 raonyx quadyiinaculafus) (pi. 

 97 A-D) 

 Several ocelli placed closely together on each side of head; 

 sjDiracle of first abdominal segment normal and not placed 

 on a projecting lateral lobe ; claw rather short, at base with 

 a large pulvillus (pon. pi. 97 F) and a very short, conical, 

 thick seta; labial palpus absent. (Median dorsal suture 

 absent on all body segments). (Either with the first larval 

 instar carried by wasps and in all the feeding stages devour- 

 ing wasp larvae in their cells, or parasitic on cockroaches) 



Rhipi'phoridae'^ (pi. 97 E-J) 

 '" According to the imagines, the Horniiae are usually placed as 

 a tribe of the subfamily Zonitinae. 



" According to the imagines, Tetraonyx is usually placed in the 

 subfamily Lyttinae. According to the first instars reared from 

 eggs laid by Tetraonyx quadrwiaculafus Fab., this larva forms an 

 intermediate type between the larvae of the family Meloidae and 

 the family Rhiijiphoridae but differs so distinctly from both of 

 them that the creation of a new family, Tetraonycidae, is deemed 

 advisable. 



'^ The first instar of Bhyzostylops, described by F. Silvestri 

 (Descrizione di un nuovo genere di Rhipiphoridae, Redia, vol. Ill, 

 1906, pp. 315-324, one plate) looks superficially like a Stenus larva, 

 has long legs, said to be tipped by two extremely minute claws, 

 several ocelli on each side of head, and is heavily bristled. It is 

 considered by Silvestri as a very aberrant rhipiphorid larva, in- 

 termediate between the first instar of Rhijiidius and the degenerate 

 first instar of the Strepsiptera. Probably the Strepsiptera are to 

 be classified in the Coleoptera and close to the Rhipiphoridae. 



59 



