Molecular phylogeny of the tribe Candalidini (Lepidoptera: Lycaenidae): systematics, diversification and evolutionary history

Abstract

The butterfly tribe Candalidini is geographically restricted to Australia and mainland New Guinea and its adjacent islands. With 60 species and subspecies, it represents a large radiation of Papilionoidea in the Australian region. Although the species-level taxonomy is relatively well understood, the number of genera is uncertain, varying from two to eight. We reconstructed the phylogeny of the Candalidini based on a 13-locus hybrid enrichment probe set (12.8 Kbp: COI, Thiolase, CAD, CAT, DDC, EF1-a, GAPDH, HCL, IDH, MDH, RPS2, RPS5, Wingless), including all previously recognized genera and 76% (28/37) of the species-level diversity of the tribe. Maximum likelihood analysis recovered the Candalidini as a strongly supported monophyletic group. In conjunction with morphological characters, the phylogeny provided a robust framework for a revised classification in which we recognize four genera, 37 species and 23 subspecies. The genus Nesolycaena Waterhouse & R.E. Turner is considered in synonymy with Candalides Hübner, and four other genera are not recognized, namely, Holochila C. Felder, Adaluma Tindale, Zetona Waterhouse and Microscena Tite. Of the four valid genera, the absimilis group (23 species) is placed in the newly described genus Eirmocides Braby, Espeland & Müller gen. nov. (type species Candalides consimilis Waterhouse). The erinus group (six species) is assigned to Erina Swainson, which is reinstated. Chrysophanus cyprotus Olliff is assigned to Cyprotides Tite, which is also reinstated as a monotypic genus. The remaining seven species are placed in Candalides sensu stricto. Overall, we propose 47 new nomenclatural changes at the species and subspecies levels, including the synonymy of Holochila biaka Tite as Eirmocides tringa biaka (Tite) syn. nov. et comb. nov. and recognition of Candalides hyacinthinus gilesi M.R. Williams & Bollam as a distinct species Erina gilesi M.R. Williams & Bollam stat. rev. et comb. nov. A dated phylogeny using Bayesian inference in BEAST2 and biogeographical and habitat analyses based on the DEC model in BioGeoBEARS indicated that the ancestor of the Candalidini most likely evolved in rainforest habitats

Immature stages of the butterfly Erina gilesi (M.R. Williams and Bollam, 2001) (Lepidoptera: Lycaenidae): description and comparative morphology of Erina species

“…Edwards & Kerr, 1978) and E. erina. Comparative differences between the immature stages of E. gilesi (Figures 5-22) and E. hyacinthina (Figures 23-37) support the species level status of the taxon recently proposed by Braby et al (2020). Although the egg, first instar larva and pupa are indistinguishable between the two species, they diverge in the late instar larvae, with at least two character states unique to E. gilesi: (1) body with dorsolateral line white and conspicuous; and (2) abdominal segments 1 and 6 each with reddish dorsal patches, when present, simple and not raised into projections.…”

“…The immature stages of E. gilesi, however, have not been compared, with only the egg and early instar larvae (instars I and II) so far described (Williams and Bollam 2001). Thus, comparative morphology of the immature stages may provide additional evidence in support of the two species hypothesis proposed by Braby et al (2020).…”

“…Erina gilesi belongs to a monophyletic group that includes five other species of Erina. According to a recent molecular phylogeny of the Candalidini, phylogenetic relationships of the six species of Erina are as follows: delospila + (acasta + ((gilesi + hyacinthina) + (geminus + erina))) (Braby et al 2020). Thus, E. gilesi is most closely related to E. hyacinthina, and this pair is sister to E. geminus (E.D.…”

“…The butterfly genus Erina Swainson, 1833 contains six species, most of which are endemic to Australia (Braby et al 2020). They occur in a variety of temperate and tropical woodlands and heathy woodlands where the larvae feed exclusively on Cassytha (Lauraceae) (Table 1).…”

“…The taxon was originally described by Williams and Bollam (2001) as a subspecies of Erina hyacinthina (Semper, 1879), but Williams and Williams (2006) suggested its status may warrant investigation after both E. gilesi and E. hyacinthina were found together on Mondurup Peak, Stirling Range, with no evidence of hybridisation. Subsequently, Braby et al (2020) proposed that E. gilesi be treated as a distinct species based on fundamental differences in genitalic morphology and adult phenotype (15 characters), phylogenetic pattern according to molecular data and level of pairwise divergence according to mtDNA (2.8% for COI), and its narrowly sympatric distribution with E. hyacinthina. The immature stages of E. gilesi, however, have not been compared, with only the egg and early instar larvae (instars I and II) so far described (Williams and Bollam 2001).…”