Arctiidae

The definition of the noctuoid families has become more precise in recent decades, though much work remains to be done (reviewed in Holloway, Bradley & Carter, 1987). The Arctiidae are defined principally by the presence of a tymbal organ on the metepisternum, and a prespiracular counter-tympanal hood. Vein Sc of the hindwing tends to be basally swollen.

Another potential arctiid apomorphy is the presence of a pair of glands, possibly pheromonal, dorsally and anteriorly between the ovipositor lobes. These are frequently branched, usually in a regular dichotomous manner (e.g. Figs. 41, 44, 102, 109, 111, 130, 136, 140, 144). They have not been noted in Lymantriidae, Noctuidae or Notodontidae. Their widespread occurrence in Lithosiinae needs confirmation; they tend to be short and broad in the few taxa examined. This character is used to reinstate the Syntominae within the Arctiidae following their exclusion as a distinct family by Minet (1986).

The Syntominae share with the Ctenuchinae, Euchromiinae and Thyretinae reduction and/or fusion of hindwing veins, particularly the total fusion of Sc with Rs, a feature not seen in other hymenopteron mimicking families of similar build such as Sesiidae and Zygaenidae and therefore more likely to be a synapomorphy than a homoplasy. A tendency towards this state is also seen in some phaegopterine Arctiinae (A. Watson, pers. comm.).

Within the Arctiidae can be recognised a series of conditions of the branching of the radial branches anteriorly and basal migration posteriorly, as illustrated below, using the formula discussed in Part 3 of this series (semicolons separate venation systems or veins that are separately rooted on the cell, and brackets indicate order of branching):

• R1; R2; (R3 (R4, R5))or ((R3, R4) R5) Most Lithosiinae
• R1; (R2 ((R3, R4) R5)) Arctiini, most S. American Euchromiinae
• R1; (R2 (R, R)) Ctenucha(R1 (R2 ((R3, R4) R5))) Euchromia, Thyretes (R1 ((R2 (R3, R4)) R5)) Most Thyretinae and Syntominae

In most cases M1 is connate with, or separate from (all Syntominae), this radial branching system, though sometimes included in it in the Lithosiinae. In many Lithosiinae R1 fuses with Sc. The above series is ordered from primitive to derived according to Brock (1971): R5 becomes progressively more ‘split back’. This again places the groups with Sc and Rs fused in the hindwing as derived amongst the Arctiidae.

No definitive feature has yet been recognised for the Arctiinae, and it is possible that they are paraphyletic relative to the group with fusion of hindwing veins Sc and Rs. Some tribal groupings within the Arctiinae can be recognised, but again there is no really satisfactory tribal classification.

The arctiid larva (Holloway, Bradley & Carter, 1987) usually has well-developed verrucae ringing each segment and bearing plumose, sometimes long setae, giving the larva a hairy appearance. The head lacks secondary setae, though these are present in the Lymantriidae, probable sister-group to the Arctiidae. The crochets of the prolegs are uniordinal, arranged in a heteroideous mesoseries (homoideous in Lithosiinae, Syntominae and Lymantriidae). The prolegs have the ends of the plantae drawn out. The mesothorax generally has two verrucae above the spiracles, though there is only one in the Ctenuchinae, Euchromiinae and Syntominae (see also Forbes, 1960: 13); the two verrucae are horizontal in Lithosiinae and vertical in the Arctiinae (Forbes).

Records of host-plants for arctiids indicate that a large number of Lithosiinae are lichen, algal or moss browsers, as are a few Syntominae. Most other arctiids tend towards polyphagy with some concentration on weedy, herbaceous plant taxa, though the Neotropical phaegopterines are mainly tree defoliators (A. Watson, pers. comm.). Amongst the arctiines polyphagy is noted for many of the Arctiini (e.g. Spilosoma Curtis). The Utetheisa Hübner*/Argina* Hübner group is to some extent specialist on Leguminosae, particularly Crotalaria, with a lineage of the former divergent on Boraginaceae. The genus Nyctemera Hübner has mostly been recorded from the Compositae.

The Bornean arctiid fauna shows a high degree of endemism amongst the Syntominae, Arctiini and the genus Nyctemera. In many instances taxa have only been recorded from a single mountain within Borneo, especially G. Kinabalu, and one Amata Fabricius species has three montane races that are separated from each other by only a few kilometres.

Several genera or species groups are more or less restricted to Sundaland: Trichaetoides Gen. n.; a section of Caeneressa Obraztsov; Auriculoceryx Gen. n.; sections of Spilosoma; Tinoliodes Wileman, shared with the Philippines. Streptophlebia Hampson is restricted to northern Borneo and the Philippines, reflecting the distribution of the butterfly genus Ptychandra Felder. The genus Aethalida Walker is mainly Wallacean (Philippines, Sulawesi, Moluccas) but has one Bornean species.

Several species are only known from Pulo Laut, an island to the south of Borneo, close offshore. Most are also found in the Lesser Sunda islands and sometimes Java. Examples are Amata exapta Swinhoe, Nyctemera pagenstecheri Pagenstecher and two Neochera species. Such species may favour areas where there is a definite dry season.