FNZ 6 - Hydraenidae (Insecta: Coleoptera) - Systematics
Ordish, RG 1984. Hydraenidae (Insecta: Coleoptera). Fauna of New Zealand 6, 64 pages.
( ISSN 0111-5383 (print), ; no. 06. ISBN 0-477-06747-6 (print), ). Published 12 Nov 1984
Variously referred to as Hydraenidae (from "Hydraenaires") or Limnebiidae (from "Limnebiaires"), both names of Mulsant, 1844, the family Hydraenidae d'Orchymont, 1919 has often been regarded as a member of the superfamily Hydrophiloidea with strong hydrophilid affinities. It was in this way that Crowson (1967) viewed it in his Natural classification of the families of Coleoptera, where he cited the form and function of the antennae, the aquatic adaptations of the adult, the presence of a cephalic egg-burster on the first-instar larva, and the anchorage of eggs with silk as evidence of its hydrophilid affinity. There are, however, features linking the Hydraenidae with the Staphylinoidea - the form of the wing venation, larval mouthparts, aedeagus, and possibly the female genitalia. This apparent conflict of evidence is best resolved by demonstrating which characters result from convergent evolution, which are primitive, and which are derivative.
Orchymontia has for some time been considered a very primitive hydraenid genus having a nine-segmented antenna with a two- segmented club and no cupule. A review of the New Zealand hydraenid fauna as a whole can make a small contribution to phylogeny by documenting all the elements of a primitive fauna (until now known from only two species), since Orchymontia, Homalaena, and Podaena all appear to have various combinations of primitive characters. This would obviate the need to place too much reliance on the characters of Orchymontia spinipennis alone.
Members of the family Hydraenidae are usually defined as follows. Small beetles, flattened rather than convex; antennae usually with a five-segmented pubescent club, which in turn is often separated from (usually three) glabrous segments by a cupule; wing venation staphylinoid; tarsi without a distinct bisetose empodium between the claws.
Larvae with normal galea and lacinia; palpiger normal and distinct; mandibles stout, with a basal molar; urogomphi two- segmented, well developed; tenth abdominal segment developed as a pygopod, usually with a pair of downward-projecting hooks. Boving & Craighead (1930) have figured the larva of Ochthebius impressus Marsh and Limnebius papposus Mulsant from Denmark, and Samuelson (1964) has described and figured the larva of Meropathus campbellensis (Brookes) from Campbell Island, where a definite association of larva and adult can be established. This is not readily accomplished with the New Zealand mainland aquatic fauna because several species may live in one stream. Towns (1978) has, however, been able to publish two drawings of unidentified aquatic species. Features of a typical adult hydraenid that are visible in ventral aspect are illustrated in Figure l.
Most New Zealand hydraenids live in fast-flowing water, and have accordingly developed a range of sexually dimorphic characters which help prevent the male from being dislodged during copulation. These modifications of the legs and maxillary palps provide a convenient means of recognising species. They have no bearing on phylogeny, however; for insight into this, antennal segmentation (Figures 18-25) and the form of the prothorax and genitalia must be considered.
In primitive hydraenids the antennae are eleven-segmented, have a two-segmented pubescent club, but lack a cupule. The number of segments has progressively reduced to nine as in Orchymontia, Meropathus, and Hydraena. The antenna of Meropathus has been described as eight-segmented, but closer examination reveals that two reduced segments lie between club and pedicel, just as in Hydraena, bringing the total to nine. This arrangement is best viewed from slide mounts. If, as Jeannel (1940) suggests, divisions beyond the pedicel result from fragmentation of a former shaft, musculature will not facilitate interpretation because the muscle attachments typical of a many-segmented structure will not be evident.
The four-segmented maxillary palps (Figures 26-53) are an aid to identification in the male, because in a number of species the third (penultimate) segment is expanded and grooved ventrally. If the groove is shallow and not sharply defined, it is best examined dry.
The form of the pronotum (Figures 2-6) is constant within a genus. Diagnostic features are the outline when viewed from above, including the degree of constriction at the anterior and posterior angles, and the depressions in their vicinity.
Elytra may be uniform in colour or have contrasting pale yellow pigment on the humeral area. They may be rounded in outline or nearly parallel-sided. The rows of punctures (striae) are usually quite distinct, though not in Hydraena, and in some species there is a raised ridge (carina) on the seventh interstria. On the ventral surface, at about the mid-point of the outer edge, there is a broad-based triangular patch of short spines reminiscent of the stridulatory area of some hydrophilids (see Figures 9-11). They seem to occur throughout the New Zealand hydraenids, male and female. Certainly they are differently structured in Hydraena when compared with Orchymontia, but scanning electron micrographs indicate that they are not diagnostic at species level nor even between allied genera.
In the aquatic genera of New Zealand, hind wings may be twice as long as the elytra and fully developed in outline, but lack all but some basal venation representing the costal, median, and anal veins (Figure 7). In most specimens the wings are uniformly reduced, however (Figure 8). Fully winged examples of some species have not been seen, but this may be due to the smallness of the samples. Whether full size or reduced, the wing is armed with evenly spaced spines over both the dorsal and ventral surface. In Meropathus there is not even a vestigial wing.
Modifications of the legs are most often found in the tibiae of the male, which on some or all of the legs may be distally curved, flattened, variously notched, or armed with spines and long setae (Figures 54-79). No modification of the tarsus has been found, but there are instances of enlarged femora and of trochanters that bear spines. Only in Hydraena is the procoxal cavity enclosed behind by the prosternal process and the pleura. The metacoxal sensillum, reported by Perkins (1980) to be a possible water-current detector, is present in New Zealand aquatic species. The line drawings af appendages at the end of this contribution have been prepared from slide-mounted material, in some instances supplemented by details from scanning electron micrographs.
The metasternum is usually heavily sculptured and dull (pruinose), at least in part. A shiny area on the disc, where present, may be interrupted by a median pruinose groove (see Figure 1). Similarly, the abdominal ventrites are at least partly pruinose. The glabrous swellings of the metasternum seen in New Zealand species are not reliable characters, and possibly result from abrasion.
In the male (Figures 80-108) the lack of demarcation between the basal piece and the median lobe of the aedeagus has led to various interpretations. Perkins (1980) has provided good grounds, based on musculature, for regarding the main tube as the basal piece and the structures distal to it as the median lobe. In New Zealand's primitive genera Podaena, Homalaena, and Orchymontia the aedeagus (aedeagophore) is approximately symmetrical, with the parameres articulated to its base. In Hydraena the median lobe becomes particularly ornate and asymmetrical and the articulation of the parameres varies in position.
In Meropathus the parameres are absent, and what presumably originated as an eversible sac is now permanently everted as a flagellum, with the ostium still identifiable at its base. Except in some species of Orchymontia with the distal end of the aedeagus membranous, the male genitalia serve to identify species. They would be indispensable for this purpose in Meropathus were it not for the geographic isolation of the species. This seems to be true also of Hydraena in a country such as Australia, where it is well represented (Zwick 1977).
The sclerotised parts of the female genitalia (Figures 109-125) are of some help in systematics, at both specific and generic level. The distinct tenth abdominal segment (Figure 113) is to a variable degree extendable, and is presumably extruded during egg laying. Because the spermathecal duct opens into the pouch-like vagina, and the spermatheca is embedded in muscle tissue, the spermathecal duct shifts its position when segment 10 is extended, and the spermatheca may be repositioned to smaller extent. For descriptive purposes, therefore, all drawings have been made from specimens with segment 10 unextended.
Typically, the spermathecal duct widens as it enters the vagina. This duct may be as long as the spermatheca or up to three times this length, in which case it is completely looped. More frequently it doubles back on itself or is gently curved (Figure 113). It widens, usually abruptly, into the spermatheca, which is more heavily sclerotised and often is pigmented anteriorly. The spermatheca terminates in a spherical or near-spherical sac with a much greater capacity than the rest of the organ. Because sperm have been observed within, the sac is regarded as a part of the spermatheca rather than an accessory gland. In two species the sac itself is slightly doubled back (Figures 100 and 113). Among species of Orchymontia the spermatheca tends to be very uniform, though in the type species, O. spinipennis, it is curved (Figure 125). In other genera too it is very predictable in form, with only slight interspecific variation in either the spermatheca or its duct.