Abstract

The Australian Anthracini are revised. In all, 28 new species are described, bringing the total fauna to 34 species. The previously described species of Anthrax Scopoli – A. maculatus Macquart, A. incomptus Walker, A. confluensisRoberts, A. lepidiotus Roberts and A. proconcisus Hardy – are diagnosed and the following eight new species of Anthrax are described: A. argentia, A. asciculus, A. clinatus, A. crenatus, A. dolabratus, A. funestus, A. opacus and A. torulus. This taxonomic study reveals a group of at least 20 cryptic species previously included in collections under the name Anthrax angularis Thomson. A new genus, Thraxan, is erected to contain this cryptic group of species and the following 20 new species are described: T. acutus, T. abditus, T. caligneus, T. cinctus, T. cornuatus, T. depressus, T. echinatus, T. ebenus, T. emicatus, T. hamulus, T. luteus, T. misatulus, T. nodus, T. norrisi, T. obstipus, T. patielus, T. planus, T. prolatus, T. simulatusand T. spiculus. Many of these cryptic species have been collected sympatrically, hill topping together in eastern Australia. A key is provided to the species of Anthrax and Thraxan, genitalia drawings are presented for most species and distribution maps of all species are presented. A cladistic analysis of the species of Anthrax and Thraxan is also presented. A total of 26 of the species is compared for 125 synapomorphies in 39 adult morphological characters. Three species-groups were found: Thraxan, and two species-groups within Anthrax, the A. proconcisus species-group and the A. maculatusspecies-group.

Previous authors divided Anthrax into species-groups on the basis of wing patterns, but found that these species-groups were not confirmed when other characters were taken into consideration. We studied the congruence of seven different character sets within the clade comprising Anthrax and Thraxan – antennae, venation, wing patterns, vestiture, genitalia, male genitalia and female genitalia – using several incongruence indices. Significance of incongruence was measured using a randomisation procedure. Results of these studies indicate that the wing-pattern character set is significantly incongruent with the other morphological data. These quantitative cladistic results explain the difficulty previous authors experienced in finding suites of characters to support species-groups in Anthrax on the basis of wing patterns. A relationship is found between the level of incongruence and the distance over which mate-recognition signals operate.