Haemonides candida, (Houlbert, 1917)

Lamas, Gerardo, 2017, A review of the genus Haemonides Hübner, [1819] (Lepidoptera: Castniidae), Zootaxa 4320 (2), pp. 245-271 : 252-253

candida ( Houlbert, 1917)

Castnia candida Houlbert, 1917 : Études de Lépidoptérologie comparée 13: 77–78. (figs. 1, 9–10).

Type material. Described from a single female collected by Marc Hüe de Mathan (1876–1908) during the first half of 1887. This specimen is the holotype by monotypy, at the time of description it was in the collection of Charles Oberthür, it is now in BMNH.

Type locality. “ Pérou, Moyobamba”. Moyobamba is the capital of San Martín department in northern Peru; it is situated at 6°3’S, 76°58’W.

Taxonomic status. A valid species.

Distribution. This appears to be a very rare species with most known specimens confined to Peru (but see below). As well as the type from San Martín department, we are aware of specimens from Amazonas, Huánuco and Loreto departments (fig. 42).

Discussion. This species only appears to be known from females, the male is unknown to science. Apart from the holotype, there are six specimens in RW, one from Mallqui, Huánuco, three from Iquitos, Loreto and two from Pebas, Loreto. In MB there are three Peruvian specimens, although the provenance of all three is in doubt: one is from Rodríguez de Mendoza, Amazonas, but Mendoza is at 1500m, so in all probability the specimen was obtained much lower and further east, in San Martín, possibly collected by B. Calderón or one of his associates; one is from Mallqui and Büche (pers. comm.) says that he doesn’t necessarily trust the Mallqui data which came from dealers’ material; the final one is from Iquitos but the phenotype of this specimen is the Huallaga Valley one, and completely different from the other known Loreto specimens.

There appear to be consistent differences between all of the known specimens from along the Amazon River in Loreto, and those from the Huallaga Valley, further west in Peru, which were presumably collected at higher elevations. The Loreto specimens consistently have a much shorter radial black band on the dorsal forewing surface and the forewing marginal band has larger white patches in it. This could potentially justify subspecific separation but due to the lack of material, and the complete absence of males, the status quo is maintained herein.

With the above proviso H. candida appears to be phenotypically fairly stable and quite distinct from H. cronis . We have seen no specimens which are intermediate between the two species other than one specimen in MB which was collected in Juanjui (200–500 m), San Martín, Peru; even though it has slight similarities with H. candida it is clearly a variant of H. cronis .

In MCLB there is a specimen putatively collected in “Santa Cruz da Serra, Rio de Janeiro, Brazil” (fig. 11) which appears to be an extreme variant of this species (or possibly of H. cronis ). The hindwing pattern looks like candida and the forewing pattern is vaguely similar, however the anal margin, the apex and the radial band of the forewing have much more extensive black in them, in particular the radial band is much thicker and reaches the costa, this goes against one of the defining features of candida . Also, the wingshape is long and much more similar to cronis than candida . The geographic separation is unlikely as this would be the only specimen known from outside Peru. Rio de Janeiro has been heavily collected in the past so it is surprising that no other Haemonides has been reported from the state. The specimen probably has false locality data, although there is a "Santa Cruz da Serra" in Petrópolis, Rio de Janeiro. Finding more specimens from the same region might shed better light on this enigma.

In the past there has been some confusion over the name with some people (particularly dealers) using the name to refer to H. cronis , although the two are very easily distinguishable.

Material examined. All of the above mentioned specimens were examined for this study.