Introduction 3 



families, with the rank of the Pseudomorphidse unknown. The sub- 

 families indicated in the Carabida^ are Lebiina^; Dromiinse, 

 Odacanthinse ; Driptinse, Nebriinag; Loricerinae ; Carabinse, Cychrinse, 

 Chla^niinae ; Licinin?e, Bembidiinae, Sphodrinae, Broscincp, Dyschiriinae, 

 Scaritinse, Elaphrinae; Patrobinae, Pterostichina^ Amarinae, 

 Harpalinac, Glyptinae. The semicolons in this enmiieration separate 

 the related groups of subfamilies. The Trechini appear to be sep- 

 arable from Bembidiini, on their larval characters, only by 

 differences of tribal rank. 



As to the sequence of the suborders Dr. Boving considers that 

 it would "be logical to place the Adephaga third, as the more 

 modern of the three," "but for practical purposes it appears more 

 advisable to rank the suborders in the commonly accepted sequence." 

 The well developed tarsus of adephagous larvae, carrying one or 

 two distinct and movable claws, may be a primitive character, as 

 regarded by us in 1920; and the great similarity in this respect be- 

 tween the larvae of the Caraboidea and those of certain neuropterous 

 larvae is interesting and possibly significant. However there are 

 several polyphagous larvae, specifically mentioned by Dr. Boving, 

 in which the tarsungulus is "divided by a faint suture into a proximal 

 and distal portion which possibly correspond respectively to tarsus 

 and claw." In Dr. Boving 's definition of the suborders by larval 

 characters, the Adephaga are placed second, and the resemblance 

 of their legs and claws to those of the Archostemata seems to us to 

 indicate this as their natural sequence. 



PODYPHAGA 



The greatest difficulties in the classification have been in this 

 suborder, and in the differences between the opinions expressed by 

 Boving, Forbes, and Tillyard, they are still apparent. The removal 

 of Cupesidae and Micromalthidae to Archostemata, and of Rhysodidae 

 to Adephaga, simplifies the classification of the Polyphaga; and 

 fortunately the isolation of the Scarabaeoidea, of the so-called 

 Phytophaga, and of the Hydrophiloidea and Staphylinoidea is not 

 questioned ; though, as to the latter two, the composition is somewhat 

 altered. Believing that the larval characters afford the safest 

 guidance, not however to the exclusion of other considerations, we 

 7-eview the superfamilies in the sequence adopted by Dr. Boving. 

 Staphylinoidea. The component families, based on larval charac- 

 ters, would be Limnebiidae, Hydroscaphidae, Leptinid^, Ptiliidae, 

 Anisotomidae ; Platypsyllidae, Scaphidiida? ; Silphidae, Staphylinidae, 

 Pselaphidae, Scydmaenidae. The families also included, though their 

 larvae are imperfectly known or entirely unknown, are Clambidae, 

 Brathinidae, Sphaeritidae, Sphaeriidae, and Micropeplidae. The differ- 

 ences between this and existing classification is the removal from 

 Hydrophilidae of the aberrant genera Limnebius, Ochthebius, and 

 Hydraena; and the suppression of Silphoidea as a superfamily. 



