Makarov K., 2008. Larval chaetotaxy in the genus Rhysodes Dalman, 1823 and the position of Rhysodidae within Adephaga (Coleoptera). in Back to the Roots and Back to the Future. Towards a New Synthesis amongst Taxonomic, Ecological and Biogeographical Approaches in Carabidology Proceedings of the XIII European Carabidologists Meeting, Blagoevgrad, August
20-24, 2007, pp. 101-123.
Subfamily Rhysodinae within the family Carabidae?
Since the grounds for the incorporation of the Rhysodidae into the Carabidae mainly lay in similarities shown by adaptive features (e.g. Bell, 1970; Hlavac, 1975), the position of Rhysodidae within this family was repeatedly revised. In particular, the presence of shared characters was revealed, such as disjunct middle coxal cavities both in Rhysodidae and a number of basal carabid groups (Paussini, Cicindelini, Loricerini, Elaphrini, Scaritini and so on). Th e viewpoint of Erwin & Sims (1984) and Erwin (1985) was particularly meaningful in suggesting similarities between Rhysodidae and some basal representatives of Pterostichini, Morionini and, especially, Psydrini. Erwin considered this group as a disjunct supertribe which, together with Psydritae and Trechitae, form the subfamily Psydrinae of a separate division, Psydriformes. Any further discussion of the rank and position of Rhysodidae in this context is only possible if the following question is answered positively: Do Rhysodidae really belong in the family Carabidae?
In general, the larval structure in Rhysodidae, including chaetotaxy, is indeed similar in many ways to that of Carabidae. Th is is refl ected in the same groundplan of the chaetome (setae and sensilla as homologized above being the proof ), antennae (4-segmented with a developed sensorium on antennomere 3) and legs (moveably articulated tibia and tarsus, a developed pretarsus with a claw). However, none of these characters can be treated as synapomorphies of Rhysodidae and Carabidae. Thus, the structure of the antennae and legs is shared by most of the Adephaga, whereas among the securely homologized setae there are many (e.g. PA1-PA2-PA3-PAa and EP1-EP2 complexes, etc.) that are also known in a wide range of beetle families, both Adephaga and Polyphaga. At the same time, some of the larval features of Rhysodes can securely be regarded as high-level autapomorphies, since none of them shows any analogs amongst Carabidae while some are even unique to the Coleoptera as a whole.
This primarily concerns the structure of the labiomaxillar complex (Figs 14-15) characterized by general consolidation. Th us, in the maxilla, the fused galea and lacinia are merged both with the stipes apex and palpifer, whereas all the appendages of the labium are fused to the mentum. Th e labiomaxillar complex is enlarged, fully covering both the oral opening and the ventral surface of the mandibles, with a particular porose structure formed in the narrow fi ssures between the labium and the maxillae. Th is structure can be
suggested to provide fl uid food, possibly myxomycetes, to be sucked in due to capillary powers. Carabidae do show some cases of feeding on fl uid or fungal food (Mormolycini), but no such structures appear (Lieftinck & Wiebes, 1968). Furthermore, liquid food consumption in Carabidae is always accompanied by the formation of pharyngeal pump musculature, whereas in Omoglymmius larvae these muscles are relatively poorly developed (Beutel, 1990b). Larval mandibles in Rhysodidae probably do not take part in feeding, because, when closed, they stay isolated from the antebuccal cavity through a long lobe of the epipharynx. So the mouthparts of Rhysodidae, certainly being homologs of those in Carabidae, show nonetheless no functional similarities to any of the mouthpart types occurring in Carabidae. Obviously, some analogs can only be traced to the larval mouthparts of the some mycetophagous Staphylinidae (Leschen & Beutel, 2001).
The next highly specifi c feachure of Rhysodidae larvae is the total absence of digitiform sensilla in the sensory complex of mouthpart appendages (Figs 14-17). Th is type sensilla are known in all studied Carabidae larvae, including such disjunct specialists as Paussini, Cicindelini, Mormolycini (Lieftinck & Wiebes, 1968), Peleciini (Liebherr & Ball, 1990), Brachinini, Pseudomorphini (Erwin, 1981), etc. At the same time, they are lacking in most of the Hydradephaga (Noteridae, Dytiscidae, Gyrinidae, Haliplidae), but occur in Trachypachidae. As far as possible to judge, the presence or absence of digitiform sensilla is not related to a xylobiotic way of life. In any case, sensilla of similar types are met with in wood-dwelling larvae of Archostemata (Grebennikov, 2004) and numerous Polyphaga families.
Finally, there are some more special peculiarities to diff er the larvae of Rhysodidae and Carabidae. Th ese concern the retention in Rhysodidae of only a single pair of setae in the gular area, the original chaetotaxy of the frontale (see above), the total reduction
of internal poststernites of the abdomen. Therefore, a whole complex of larval traits emphasizes a suffi ciently high degree of singularity of Rhysodidae to prevent them from being incorporated within Carabidae.
It is noteworthy that most of the students who based their results on adult characters (Bell & Bell, 1962; Hlavac, 1975; Bils, 1976; Bell, 1978; Baehr, 1979; Beutel, 1990, 1992a, 1995, 1998; Liebherr & Will, 1998) considered the rhysodines as only a taxon subordinate to Carabidae, whereas those who investigated the larvae (Bцving, 1929; Bцving & Craighead, 1930; Burakowski, 1975; Beutel, 1992b; Arndt, 1998) invariably arrived at the opposite conclusions.