A new genus , Atlanteuptychia gen . nov . , for Euptychia ernestina ( Lepidoptera : Nymphalidae : Satyrinae )

Atlanteuptychia Freitas, Barbosa & Mielke, gen. nov. is proposed for Euptychia ernestina Weymer, 1911 and illustrated. This taxon lacks the posterior projection of the tegumen, a synapomorphy of Euptychia Hubner, 1818, and does not share morphological synapomorphies with Cyllopsis R. Felder, 1869 and Paramacera Butler, 1868, two Central American genera apparently closely related to Euptychia ernestina, based on molecular data. This evidence supports the proposition of a new genus endemic to the Atlantic Forest, A. ernestina stat. nov.

omicroscope equipped with a micrometric scale and a drawing tube.Photographs of the male and female genitalia were taken using a Zeiss Discovery V20 Stereomicroscope.The following abbreviations were used: (FW) forewing, (HW) hind wing, (D) dorsal, (V) ventral.
Type species.Euptychia ernestina Weymer, 1911 Diagnosis.Eyes naked, dark brown with a light brown basal stripe.Palpus approximately 2.0 times longer than head height, covered with short, light yellow and long dark hairs (Fig. 3).Antenna 8.0-9.0 mm long, extending to mid-costa, with 29 to 32 segments; shaft dark brown with the internal margin bearing white dots; club composed of 10-11 segments, well developed, 3-4 of which are yellow in their middle section (Fig. 4).Legs covered with dark brown scales on femur and light yellow scales on tibia and tarsus.
Female genitalia : Papillae anales long and hairy, adorned with few setae, ductus bursae short and slightly sclerotized; corpus bursae ellipsoidal with two long sclerotized parallel signa running along the entire length of corpus bursae.
Etymology.The generic name refers in part to the Atlantic region of Brazil, to which the taxon is endemic, in addition to its relationship to other genera in Euptychiina.
Remarks.Among the genera with similar wing color patterns, Atlanteuptychia gen.nov.can be easily distinguished from Euptychia Hübner, 1818 by the absence of the posterior projection of the tegumen and by the presence of a well-developed gnathos in the male genitalia, and by the absence of a recurrent vein in the FW discal cell.The new genus differs from Cissia Doubleday, 1848 by the short saccus and gnathos (both usually longer in most Cissia); and from Carminda Dias, 1998 by the slightly wavy margin of the HW (strongly wavy in Carminda), and by the shape of the uncus in dorsal view (laterally extended in Carminda); differs from all species of the above three genera by the presence of a conspicuous androconial tuft of long hairs on the male HWD in space CuA 2 -2A, and by having the last two abdominal segments dorsally covered with distinct black scales.
Redescription.Male (Figs 1-6).Body entirely dark brown.Last two abdominal segments dorsally covered with distinct black scales (Figs 5 and 6).FW length 16.0-19.0mm (n = 5); HW length 15.0-16.0mm (n = 5).Dorsal ground color of both wings brown with few markings; dark sub marginal stripe especially conspicuous in HW; well-defined androconia comprising long, dark hair-like scales in space CuA 2 -2A on the HWD (Figs 1 and 5).Ventral color of both wings brown with several markings as following: FWV crossed by two weakly marked dark brown lines, first line extended from Sc, crossing discal cell to half the space CuA 2 -2A one third from base; second line, broader, extended from costa to half the space CuA 2 -2A two-thirds from the wing base; dark brown zigzagging sub marginal line and a brown regular marginal line extending from costa to 2A; apical region light, covered with cream scales, with well-developed dark ocellus in space R 5 -M 1 complete, outlined by yellow ring, and with silver pupil; two to three additional irregular weakly marked silver ocelli without the black areas and with incomplete yellow outline apparent in most examined individuals are usually present in space M 1 -M 2 and M 2 -M 3 .HWV with dark brown region in basal area near insertion of wings; crossed by two dark brown irregular lines from costa to anal margin delimiting lighter area, first line more regular, one third from wing base; second lines irregular in shape and thickness, two-thirds from wing base; margin light brown, with dark brown irregular zigzag sub marginal line and brown regular marginal line extending from costa to 2A; series of five ocelli present in spaces Rs-M 1 (ocellus 1), M 1 -M 2 (2), M 2 -M 3 (3), M 3 -CuA 1 (4), CuA 1 -CuA 2 (5); ocelli 1, 2 and 5 complete, black, outlined by yellow ring, and with silver pupil; ocelli 3 and 4 irregular in shape, silver, without black areas and with incomplete yellow outline.Wing venation, palpus and foreleg of male are illustrated in Figs 13,15,and 16,respectively.Female.FW length 19.0 mm (n = 2).HW length 15.0-16.0mm (n = 2).Except by the absence of patch of dark scales Habitat and geographic distribution.Based on all available information, A. ernestina is widely distributed in 21 sites in the Brazilian states of Santa Catarina, Paraná, São Paulo, Rio de Janeiro, Minas Gerais and Espírito Santo, mostly from well preserved areas of wet forests in the Atlantic Forest Biome (see Table I and Fig. 19).The species is more common at 800 to 1100 m, but it is also present at the sea level on the foothills of coastal mountains in southeastern Brazil.Based on the present records, the species is considered rare, and has been seldom observed and/or collected.However, a recent study with baited traps (using banana and sugarcane juice fermented for 48 hours) in Serra do Japi (JPS and AVLF, unpub.data), suggests that it is more common than previously thought.In this locality, during 15 months of study, 12 adults were recorded (9 males and 3 females), 11 of which in the canopy, and only one in the understory.In previous fieldwork in the same site, only 4 individuals had been collected over 30 collecting days through 20 years (AVLF AND KEITH S. BROWN JR, unpub.data).These observations suggest that A. ernestina may be a canopy-dwelling species, and adults are collected on rare occasions when they descend to the understory.Females are usually collected when flying low in the understory, with a behavior that resembles host-plant searching, but additional observations are needed to confirm this.There is no additional information about behavior of either males or females, and host plants are unknown.

DISCUSSION
Although having a small wingspan and a wing pattern resembling that of some dark Euptychia (e.g., Euptychia enyo Butler, 1867, and Euptychia picea Butler, 1867), the removal of A. ernestina from Euptychia is supported by morphological (present study) and molecular data (PEÑA et al. 2010).The male genitalia lacks the posterior projection of the tegumen, a structure considered a synapomorphy of Euptychia (FREITAS et al. 2012).Additionally, A. ernestina differs from all known Euptychia by the presence of a well-developed gnathos in the male genitalia (absent in Euptychia spp.), and by the absence of a recurrent vein in the FW (present in Euptychia spp.) (FORSTER 1964, PULIDO et al. 2011, FREITAS et al. 2012).
Based on molecular data, A. ernestina is not related to any Euptychia species, but instead, it is the sister group of a clade   Two characters of Atlanteuptychia, the conspicuous androconial tuft on the male HWD, and the distinct black scales on the abdomen, are both present in some Chloreuptychia Forster, 1964, namely Chloreuptychia catharina (Staudinger, [1886]), Chloreuptychia hewitsonii (Butler, 1867) and Chloreuptychia tolumnia (Cramer, 1777).However, in Chloreuptychia the androconial tuft is shorter and located in space 2A-3A, while in A. ernestina it is longer and located in space CuA 2 -2A; the distinct patch of black scales is located in the mid-ventral abdomen in Chloreuptychia, whereas in A. ernestina it is located in the last two dorsal segments.
Based on morphology only, no other euptychiine genus shows clear affinities with Atlanteuptychia, including Cyllopsis and Paramacera (the two related genera based on molecular data, see above) (AVLF and M. A. MARÍN unpub. data).This is quite significant from a biogeographical perspective, especially concerning the wide disjunction of Cyllopsis + Paramacera with Atlanteuptychia; while the first two genera are distributed in Central and North America, Atlanteuptychia is endemic to the Atlantic Forest.
The proposal of the new genus, Atlanteuptychia, is supported by the absence of synapomorphies uniting A. ernestina with any other known Euptychiina, including the genera

Cyllopsis
and Paramacera which were considered closely related in a recent molecular phylogenetic analysis (PEÑA et al. 2010).

Table I .
Data for studied individuals of Euptychia ernestina from seven Museum collections (see text for museum acronyms).KS Brown = unpublished record by Keith S. Brown Jr.
* Genitalia examined; 1 exact site of collecting unknown; 2 specimens collected with baited traps.