Hemiceratoides hieroglyphica ( Saalmüller, 1891 )

Hemiceratoides hieroglyphica ( Saalmüller, 1891) View in CoL

( Fig. 7A, B View Figure 7 )

Hemiceras hieroglyphica Saalmüller, 1891 . Lepidopteren von Madagascar 2: 405, pl. 11, fig. 208.

Type material: ♂ holotypus, by monotypy, in SNMF (not traced by A.Z., but identity not in doubt). Type locality (verbatim): Madagascar … N.-B. (=Nossi-Bé, Nosy Be).

Diagnosis: Hemiceratoides hieroglyphica is a relatively large species that can immediately be distinguished from its congeners by the concolorous pale-orange hindwing, which is deprived of any fuscous markings (cf. also: Saalmüller 1891, Gaede 1940 in 1939– 40: pl. 39, row b). Fore- and hindwings are also the broadest within the genus, the former being of rather homogeneous pale beige colour with a more or less emphasized orange tinge and weakly expressed pattern elements, the latter with well-rounded margin before tornus, and so comparatively broad as to make the forewing appear less externally protruding than in other orange-hindwinged congeners. The antenna of male shows long paired pectinations along the basal two-thirds of the flagellum, much as in H. vadoni Viette, 1976 and H. ornithopotis sp. nov.. The latter is a large-sized species with present albeit weakly expressed fuscous markings on the hindwing, which, therefore, most closely resembles H. hieroglyphica , but besides its allopatric distribution, it can easily be distinguished by conspicuous differences in the genitalia, as detailed below in the description of the new species. In the male genitalia of H. hieroglyphica , the sacculi bear along their internal edges two big lobes, these are, as usual, strongly asymmetric between the valvae, and more developed on the right valva, where a very broadly spatulate lobe oriented internally is followed by a transverse thumb-like one; there is no evident separation or constriction between the saccular parts and cuculli in the valval outlines; the distal section of valva terminates with three finger-like processes radiating at approximately 70–90° from each other, one long from the costal angle followed by a more external slightly longer and stouter one, while the third one is short, incurved, and has been interpreted here to be of saccular origin (see above under ‘Morphological remarks’: 4). The mastigojuxta is the shortest (reaching base of uncus) and stoutest in the genus, the distal section certainly being not flexible but in shape of a stiff, internally curved stick that terminates into clubbed blade-like end bearing on its outer edge a series of stout spines ( Fig. 9A View Figure 9 ). The phallus is immediately distinguishable from that of the other Madagascan endemic as it bears the typical auricolate carinal plate shared with species from the African mainland ( Fig. 10A View Figure 10 ). In the female genitalia, there is only a unique, very broad, fan-shaped midventral plate overhanging ostium bursae, such a plate being slightly incised midventrally at the posterior margin and broadly connected laterally with tergum A8 ( Fig. 11A View Figure 11 ) (sclerotizations on sternum A7 could not be studied).

Distribution: Endemic to Madagascar but only known from low to middle elevation (15–640 m) west of the highlands (Viette 1976; MNHN collection data: N = 29). Adults have been noted on the wing from early November to mid-February ( Hilgartner et al. 2007; MNHN collection data: N = 29). Often recorded from the African mainland, e.g. Liberia (Zaspel 2008), Nigeria ( Hampson 1902), Ghana ( Forsyth 1966), Malawi (Zaspel 2008, Zaspel et al. 2011), and South Africa ( Hampson 1902, Staude et al. 2016); these are all quotations that we ascribe to the other species described in this paper.

Molecular resources: BIN, BOLD:ACM6895. DNA barcoded specimens: Majunga, NW Madagascar, 25029-240120-MA; Nosy Be, KLM_Lep_01342 (C. Wieser).

Remarks: This species has been recorded as frequenting the eyes of Newtonia brunneicauda ( Vangidae ) (two episodes witnessed) and Copsychus albospecularis ( Muscicapidae ) (one episode) by Hilgartner et al. (2007). Pictures provided by these authors clearly show this species on the vangid host and one voucher specimen (a male) is also available, but the moth on the muscicapid species cannot be surely identified from the illustration. However, the fact that the observations were done in the same area and period reasonably allow us to exclude that H. vadoni was involved, as the latter is a second Madagascan endemic considered to be allopatric with respect to H. hieroglyphica in Madagascar (Viette 1976) (but see distribution of vadoni below).