Description of two new Homoeocera Felder from Guatemala and Mexico ( Lepidoptera , Arctiidae , Arctiinae , Euchromiini )

Two new species of Homoeocera Felder from Central America are described from high-altitude, humid localities in Mexico and Guatemala, Homoeocera georginas sp. n. and H. papalo sp. n. Both are compared to the closely related Homoeocera gigantea Druce. Illustrations of adults and male genitalia of the three species are provided. Th e generic placement within the genus Homoeocera is discussed.


Introduction
Th e general classifi cation of the Noctuoidea superfamily followed throughout this work is that of Mitchell et al. (2006) established on nuclear genes rather than the less traditionnal and more controversial classifi cation proposed by Lafontaine and Fibiger (2006).At the tribe level the classifi cation used follows the recent works of Kitching and Rawlins (1998) and Jacobson and Weller (2002).
In 2007, I had the opportunity to collect in Guatemala in very well preserved localities which gave a slight idea of the richness of this country around the end of the 19 th century when the large work "Biologia Centrali-Americana" was written (Druce 1881(Druce -1890a)).Even though Guatemala has been largely explored since 1850, it is obvious that many new species of Lepidoptera are continually awaiting description.
During this trip, we spent two nights at Fuentes Georginas, Quetzaltenango, a high altitude (2460 m) locality with cold and very wet weather.We were surprised to fi nd two diff erent species of Homoeocera Felder in abundance: one larger species with wide black wing margins and clear transparent wings, and a second, smaller species with narrow black margins and yellowish transparent wings.Moreover, the smaller species displays conspicuous yellow intersegmental abdominal lines whereas the large one displays bright blue lines (Fig. 1).
Th e genus Homoeocera was created by Felder in 1874 for a single species, H. crassa Felder, 1874 (op. cit., Pl. 102, fi g. 26) and for the moment contains between 8 and 24 species depending of the accepted limits of the genus (Draudt 1915; Global Lepidoptera Names Index website; Global Diversity Information facility website).Th e two species found in Guatemala are very similar to Homoeocera gigantea Druce, 1884, described in the fi rst volume of "Biologia Centrali-Americana" (Druce 1884: 54, plate VIII fi g. 5) based on a single female specimen from Costa Rica (Van Patten leg.).Later, in the second volume, Druce added two more localities: Guatemala from a specimen in Staudinger collection (Conradt leg.) and Guatemala City (Rodriguez leg.) (Druce 1897 p. 341).Th e holotype is in the general collection in BMNH in London (specimen examined).It is interesting to note that in addition to the type in the BMNH, there is a male from Guatemala bearing the label: Guatemala, Santa Maria, 5500 ', Sept, 1919-266.Following this study it was clear that the two entities found in Guatemala represent two quite distinct species, the larger one having genitalia similar to Homoeocera gigantea from Costa Rica.Moreover it was also found that specimens from Oaxaca State in Mexico, also at high altitude, represent a new species diff erent from the two species found in Guatemala.Th ese two new species are described in the present work.

Methods and materials
Adult genitalia were prepared by boiling abdomens 15 minutes with 2 pellets of potash in 5 ml of water.After rinsing with water and then alcohol, genitalia were photographed in a natural position suspended in 95% alcohol.Types and museum specimens were mounted in Euparal, and remaining specimens were stored in glycerol in Eppendorf tubes.Photos were taken with a CoolPix 4500 Nikkon camera attached to a trinocular Nikkon stereomicroscope SMZ-10A.
In order to check the exact distribution of each species, specimens were dissected from as many localities as possible.Specimens from the collection of Xavier Lesieur were included (Honduras and Mexico) and specimens originating from Mexico, Guatemala, Honduras and Costa Rica were dissected.
Th ree specimens each of H. gigantea and H. georginas were sequenced and analyzed using the barcode fragment of the mtDNA COI gene (BOLD: Barcode of Life Data Systems).DNA was extracted from legs of dried specimens in the collection of the author, then amplifi ed and sequenced at the "Canadian Centre for DNA Barcoding" (CCDB) in Guelph, Ontario.Extraction, amplifi cation and sequencing protocols can be found on the CCDB website (http://www.dnabarcoding.ca/pa/ge/research/protocols) and are also described in Vaglia et al. (2008) Description.Head, collar, tegulae, thorax and antennae deep black.Abdomen above entirely black with bright blue intersegmental lines, diffi cult to see on dried specimens Forewing length 21-24 mm (n = 13).Forewings transparent and entirely bordered by a thin black line (around 0.3 mm) with a small indentation on vein CuA2.Base of the wings largely black.Black line at the end of the cell reaching the costa.Hindwings reduced, entirely transparent, with termen bordered with a thin black line.Border wider along anal angle and with a small indentation just above the anal angle.Th e immediate basal area black.
Th e underside is almost identical to the upperside, palpi are entirely red; hindwings with a milky white iridescence just inside the anal border; all legs are crimson red except the tarsi which are dark-brown.Th ere is a white depression at the base of the ventral abdomen in males.Distribution and biology.As far as is known, H. gigantea is distributed from Guatemala to Costa Rica, generally at high altitude in very humid biotopes.It is not presently known from Nicaragua possibly because collecting at high altitude is diffi cult due to the topography of the country.Th e biology is unknown.
Remarks.Specimens from Guatemala and Honduras exhibit wider black wing borders compared to Costa Rican specimens, but the genitalia are similar.DNA barcode divergence between H. gigantea and H. georginas samples is approximately 3.8%.Etymology.Th e name is a reference to the type locality: Fuentes Georginas, a popular spot with hot springs high in the mountains on the Pacifi c slope.

Homoeocera georginas
Diagnosis.Similar to gigantea but less robust species with a yellow cast on forewings and very characteristic deep crimson red legs.By comparison H. gigantea is a larger and more robust species well characterized by the wider black margin of the fore-and hindwings.Th ere is no yellowish cast on the forewings, the black indentation in space 2 of the hindwings is more pronounced, the abdomen upperside has a bluish hue with blue intersegmental lines (very diffi cult to see on dried specimens), the tip on the antennae below are not white, palpi are red not black.
In H. gigantea, the uncus is shorter and wider, the lateral protuberances are longer, wider and rounded (tapered in georginas) in lateral view, cylindrical in ventral view.Th e valvae are asymmetrical, the left one being slightly longer than the right one which reaches the base of uncus.Th e tegumen is a little wider, the concavity more pronounced and always without saccus.In the aedeagus, the ventral process is less than half the length of the upper process (greater than half in georginas), narrower but more sclerotized (almost black), and the apex of the ventral process is abruptly cut, not tapered as in georginas.
Description.Female unknown.Head, collar, tegulae, thorax and antennae deep black.Abdomen above entirely black with yellow intersegmental lines, diffi cult to see on dried specimens.Forewing length 18-20 mm (n = 10).Forewings transparent with a slight yellowish cast and entirely bordered by a thin black line (about 0.5 mm) with a small indentation on vein CuA2.Base of the wings largely black.Black subrectangular spot at the end of the cell reaching the costa.Hindwings reduced, entirely transparent, with termen bordered with a thin black line but without the yellowish cast.Border wider along anal angle and with a small indentation just above the anal angle.Th e immediate basal area black.Th e underside is almost identical to the upperside, except the tip of the antennae which is white, palpi are entirely black, there is a milky white iridescence just inside the anal border on the hindwings and all legs are deep crimson red except the claws which are dark-brown.
Male Genitalia (Fig. 3g-i): Uncus long, pointed and curved downward, like a bird beak with a protuberance on each side.In lateral view these protuberances are very slender with a total length less than half the length of uncus.In ventral view, these protuberances have a rounded and almost cylindrical appearance.Valvae are symmetrical slighly curved inward and slightly spatulate at tip, reaching the middle of uncus.Tegumen slightly concave and without saccus.Adaegus very characteristic, divided into two distinct parts: an upper one, long, thin, slightly curved upwards and bearing a small smooth vesica and a lower part slightly longer than half the upper one, wider and strongly sclerotized, corrugated and with a bevelled extremity.
Distribution and biology.For the moment, known only for a restricted area in Guatemala, on the Pacifi c slope, at high altitude.Biology is unknown.
Remarks.Although sample size of barcoded specimens is small, the relatively large genetic divergence between gigantea and georginas, at about 3.8%, fully support the recognition of these taxa as separate species.Etymology.Th e name is a reference to the type locality, Concepcion Papalo, a small city in north east Oaxaca state.Th e name Papalo means butterfl y.

Homoeocera papalo
Diagnosis.Similar to gigantea but a less robust species with elongated forewings and orange red legs.H. papalo sp.n. is intermediate in size between gigantea and georginas, but the forewings are obviously narrower and elongated.Th ere is no yellowish cast on the forewings as in georginas, however the underside tip of antennae is white as in georginas.Palpi and legs are orange not deep crimson red as in georginas and gigantea.Th e genitalia of H. papalo are also very distinct from the two other species.Here, the lateral protuberances of the uncus have a very diff erent shape, the symmetrical valvae are very short, not reaching the base of uncus.Th e adaegus has two long lobes of similar length, whereas the lobes are very obviously of diff erent lengths in gigantea and georginas.
Description.Female unknown.Head, collar, tegulae, thorax and antennae deep black.Th e collar displays a metallic bright blue cast.Abdomen deep black upperside with bright line at the end of segments.Forewing length 20-21 mm (n = 2).Forewings elongated, transparent, entirely bordered by a black line wider than in H. georginas sp.n. specially at apex where it is enlarged and presence of a small indentation on vein CuA2.Base of wings is largely black, the subrectangular cellular spot at the end of the cell is large and conspicuous.Hindwings transparent, black border wide except on costa with an indentation on vein CuA2.Th e extreme base black.Underside similar to upperside, tip of the antennae white, palpi and legs orange.A milky white iridescence just inside the anal border on the hindwings.
Male Genitalia (Fig. 3 j-l).Uncus conical, stout and pointed, with a protuberance on each side.In lateral view these protuberances are very slender with a total length exceeding half the length of uncus.In ventral view, these protuberances are semi-circular not unlike "ears".Valvae are symmetrical, slighly curved inward and spatulate at tip, not reaching the base of uncus.Tegumen rounded and without saccus.Adaegus very characteristic, divided into two distinct parts: a dorsal long, thin, rectilinear process with the apex slightly curved downward and bearing a small smooth vesica and a ventral thin, pointed process reaching about ¾ length of the dorsal one.
Distribution and biology.Currently known only from a very restricted area in Oaxaca state in Mexico, at high altitude.Th e biology is unknown.

Discussion
Th e three species discussed here constitute a very homogeneous group in both habitus and genitalia.Th is study adds data to the BOLD project, which aims to build a DNA barcode database of all macrolepidoptera (Hebert et al. 2003), and a campaign targeting the neotropical Arctiidae has been initiated.To date, more than 2200 sequences accounting for about 750 species (i.e., more than 10% of currently described neotropical Arctiidae) from at least 16 countries have been analyzed.
Th e main problem in Homoeocera seems to be the correct generic assignment of this group of species.When he described gigantea, Druce included this species within the genus Gymnelia Walker near torquata Druce, which he had described sometime earlier (Druce 1883).In 1898, Hampson transferred gigantea to Homoeocera Felder whereas in 1914 he transferred torquata to Dassysphinx Felder.What is clear from dissection of the generotype of Homoeocera (i.e.crassa Felder) is that the three above species are not congeneric with crassa.Furthermore, based on the habitus, torquata seems misplaced within the genus Dassysphinx, but together with the three species treated here, tarsipuncta Schaus, garleppi Rothschild and ockendeni Rothschild (all three presently included within Dassysphinx) seem to constitute a rather homogeneous group.A correct generic assignment of all these species requires a complete revision of the genus Homoeocera, which is far beyond the scope of this work.It seems prudent at the moment to maintain the status quo and assign the two new species to the genus Homoeocera.