Male (n = 14, Figs 22–23, 28–29, 32, 34–35, 40–43, 68 part) – holotype forewing length = 16.5 mm. Forewing triangular, rounded at apex and tornus, costal and outer margins convex, inner margin almost straight, mildly concave mediad, two discal cell veins bulged at bases, vein 2A thickened basad. Hindwing rounded, almost circular. Wings dorsally dark-brown with sparse olive-beige overscaling and two darker-brown terminal lines. Wings ventrally pale-brown, paler towards inner margin of forewing, with extensive beige overscaling, particularly along veins in distal part in some specimens; submedial and postmedial dark-brown lines and dark-brown end-of-cell streak (smaller on hindwing) between them; forewing postmedial line bent basad near costa in many specimens; hindwing postmedial line almost straight near costa, rarely convex basad and typically convex distad posterior of M₃ (between the two small eyespots in the middle, closer to posterior eyespot); two terminal dark-brown evenly curved marginal lines, dark-brown sinuous submarginal line, and row of submarginal eyespots basad of the sinuous line and posteriad of outer discal line, largest eyespots black-centered and pupiled with pale-blue scales: on forewing, largest eyespot in cell M₁-M₂, eyespot in cell R₅-M₁ black-centered in some specimens; on hindwing, largest eyespots in cells Cu₁-Cu₂ and M₁-M₂, a smaller one in cell Cu₂-1A+2A, even smaller, but still black-centered and pale-blue pupilled in cell Rs-M₁, and two smallest, usually without black, but in some specimens pale-blue pupilled eyespots in cells M₂-M₃ and M₃-Cu₁. Fringes monochrome, a little paler than the ground color of wings. Head, palpi, thorax and abdomen dark-brown above, paler and mostly beige beneath. Antennae dark-brown above with pale scales at segments, orange-brown at the club, beneath beige basad, orange-brown in distal half. Legs brown with beige scales. Male genitalia (n = 14: 12 dissected, 2 inspected in situ, Figs 60c, f, i, l, 61a, 62n) – typical for the genus, smaller and darker in color (more sclerotized) than those of Hermeuptychia sosybius. Tegumen dome-like, rounded at margins. Uncus leaf-shaped in dorsal view, angled to the sides, roof-like, convex distally but almost flat basally in lateral view, without thin, membranous carina in basal half; apex of uncus pointed, not truncated. Gnathos arms thin, wide apart, divergent, about the same length as uncus. Valvae narrow, elongated with thin cuculli extending past gnathos not farther than a third of their length; cucullus more rounded at apex, usually with a couple of small teeth; cucullus ventrally with inner medial bulge. Saccus about the same length as cucullus, narrow. Aedeagus elongated, almost straight, only slightly and evenly curved, not bent, broader and shorter compared to Hermeuptychia sosybius, with a smaller, about as long as wide phallobase. Female (n = 8, Figs 23–27, 30–31, 33, 68 part) – similar to male in facies, with slightly more rounded wings and dorsally paler in color. Female genitalia (n = 8, Fig. 64i–p) with antrum darker in color and smaller than that of Hermeuptychia sosybius. Ostium bursae ellipsoidal, its ventral margin longer than dorsal margin. Antrum narrower anteriad, almost triangular in ventral view, somewhat kidney-shaped in lateral view, mostly symmetric. Ductus and corpus bursae each in length similar to antrum; corpus bursae with two signa, spines in a signum broad, leaf-shaped, usually shingled in two rows.

Genbank accession KJ025595, 658 base pairs:

AACTTTATATTTTATTTTTGGTATTTGAGCAGGAATAATTGGTACATCATTAAGTTTAATTATCCGAATAGAATTAGGTAATCCAGGATTTTTAATTGGAGATGACCAAATTTATAATACTATTGTTACAGCTCATGCTTTTATTATAATTTTTTTTATAGTAATACCCATTATAATTGGAGGATTTGGTAATTGACTTGTCCCTTTAATATTAGGAGCTCCTGATATAGCTTTCCCACGTATAAATAATATAAGATTTTGATTATTACCCCCATCTTTAATTTTATTAATTTCTAGTAGTATTGTAGAAAATGGAAGTGGGACAGGATGAACAGTTTACCCCCCCCTCTCATCTAATATTGCTCATAGAGGTTCTTCAGTAGATTTAACAATTTTTTCACTTCATTTAGCTGGAATTTCTTCAATCTTAGGAGCTATTAATTTTATTACAACAATTATTAACATACGAATCAATAATATATCTTATGATCAAATACCTTTATTTATTTGAGCTGTAGGAATTACAGCTCTTCTTTTACTTCTTTCATTACCTGTTTTAGCAGGAGCTATTACTATACTTCTTACTGATCGAAATTTAAATACATCATTTTTTGATCCTGCAGGAGGAGGAGATCCTATTTTATATCAACATTTATTT

In addition to the holotype, barcodes and ID tags were obtained for 19 paratypes (15 full-length barcodes and 4 ID tags, see Table 1, GenBank accessions: KJ025588–KJ025607, except KJ025595, which is the holotype). Full length barcodes revealed five haplotypes differing from each other by just 1 to 3 base pairs (less than 0.5%). The haplotype of the holotype was more frequently observed (Fig. 66b) and other four haplotypes were confined to a single specimen in the sample.

Holotype: ♂, has the following four rectangular labels: white printed - || USA: TEXAS: Fort Bend Co. | Brazos Bend State Park, | Hale Lake, 29.3801°, -95.5847°| 17-Aug-2013 Grishin N.V. ||; white printed - || DNA extraction | NVG-1560 | 2013-09-05 ||; white printed - || Genitalia vial # | NVG130927-14 | Prep. N. V. Grishin ||; red printed - || HOLOTYPE ♂ | Hermeuptychia | intricata Grishin ||. The holotype is illustrated in Figs 22–23, 60c, f, i, l, & 68 (first image), and the Genbank accession for its DNA COI barcode sequence is KJ025595. Upon publication, the holotype will be deposited in the National Museum of Natural History, Smithsonian Institution, Washington, DC (USNM). Paratypes: 13 ♂♂ and 8 ♀♀, all from USA. Of these, 2 ♂♂ and 5 ♀♀ with the same data as the holotype; and 3 ♂♂ (DNA vouchers: NVG-1541, NVG-1548, & NVG-1551) from 2.5 km to the east, i.e. USA: Texas: Fort Bend Co., Brazos Bend State Park, Horseshoe Lake trail, latitude 29°22'54.96", longitude -95°36'41.06", elevation 15 m, 17-Aug-2013, leg. N. V. Grishin. Sexes and GenBank accessions|DNA voucher numbers|genitalia codes (na if not available) for these paratypes (the same format is used below for others) are: ♂ KJ025588|NVG-1541|NVG131003-03, ♂ KJ025589|NVG-1548|na, ♂ KJ025590|NVG-1551|na, ♀ KJ025591|NVG-1554|NVG130927-07, ♂ KJ025592|NVG-1555|NVG131003-04, ♂ KJ025593|NVG-1556|NVG131003-05, ♀ KJ025594|NVG-1558|NVG130927-08, ♀ KJ025596|NVG-1563|NVG130927-11, ♀ KJ025597|NVG-1565|NVG130927-12, ♀ na|na|NVG131003-10. All but one of these paratypes are illustrated in Figs 24, 25, 68 (above the line). 1 ♂ Texas: Brazoria Co., Bar-X Ranch, Rd. 971N, 29.13252, -95.58340, 7 m, 4-Mar-2000, leg. Nick V. Grishin, KJ025599|NVG-1631|NVG131017-08 (Figs 28–29, 62n). 1 ♀ Texas: San Jacinto Co., Sam Houston National Forest, USF217 @ Big Creek, 58 m, 12-Apr-1998, leg. Nick V. Grishin, KJ025598|NVG-1629|NVG131017-06 (Figs 30–31). 1 ♂ South Carolina: Charleston Co., McClellanville, Wedge Plantation, 6-Apr-1970, leg. D. C. Ferguson, KJ025600|13385G07|NVG131102-38 (Fig. 34). 1 ♀ South Carolina: Clarendon Co., 9-Aug-1898, KJ025604|13385G08|NVG131102-39 (Fig. 33). 1 ♀ ibid., Aug-1910, KJ025605|13385G09|NVG131102-40 (Figs 26–27). 1 ♂ ibid., Aug-1910, KJ025606|13385G11|NVG131102-42 (Fig. 32). 1 ♂ Florida: “Putnam Co | Shell Bluff Landing”, 29-Sep-1985, George Balogh, KJ025602|13385H02|NVG131102-45 (Fig. 40). 1 ♂ Florida: Alachua Co., Gainesville, 12-Mar-1983, leg. Scott W. Gross, KJ025601|13385H01|NVG131102-44 (Fig. 41). 1 ♂ Louisiana: Jefferson Parish, Harahan, 28-Jun-1944, W. D. Field, KJ025603|13386A03|NVG131102-57 (Fig. 43). 1 ♂ Louisiana: Jackson Parish, Jonesboro, na|13386A05|NVG131102-59 (Fig. 42). 1 ♂ “Flatbush LI” (specimen curated in the USNM among Hermeuptychia from Louisiana), collected prior to 1941, G. P. Engelhardt Coll., KJ025607|13386A02|NVG131102-56 (Fig. 35).

USA: Texas: Fort Bend Co., Brazos Bend State Park, near Hale Lake, latitude 29°22'48.27", longitude −95°35'05.02", elevation 16 m. This locality is by a wooded, partly open, lowland hiking trail (near and along the park paved road) from a parking lot towards the Big Creek, north of the Hale Lake.

The name refers to the difficulty in recognizing this very distinct species and its intricate ventral wing patterns. The name is an adjective.

Generally, this is a species of eastern US coastal plains and is currently documented from Texas, Louisiana, Florida, and South Carolina (Fig. 67). It is expected to be more widely distributed in the region and the exact boundaries of the range remain to be investigated. For instance, photographs of live individuals from Alabama: Bibb Co., Blue Girth Creek, 08-VIII-2004 & 18-VI-2005 by Vitaly Charny (Warren et al. 2013, specimens not collected, excluded from the type series) exhibit characters more consistent with Hermeuptychia intricata than with Hermeuptychia sosybius (see discussion below). Furthermore, it is difficult to interpret the locality label for the last listed paratype other than “Flatbush Long Island” [New York, Kings Co.]. However, Hermeuptychia has not been recorded that far north–northernmost records are from southern New Jersey and southern Pennsylvania (Opler et al. 2013)–therefore this specimen might have been mislabeled. Nevertheless, searches for this species in the coastal New York/New Jersey area might be interesting to probe its northern distribution limits. An additional specimen (not examined, excluded from the type series) from Costa Rica: Puntarenas Province, GenBank accession AY508548 (Murray and Prowell 2005) has DNA sequence with only 1 bp difference (over 435 base pair C-terminal segment of the barcode) from the USA Hermeuptychia intricata barcodes. Unless this sequence is a contamination, it is possible that the Costa Rican specimen is Hermeuptychia intricata, which may be ranging southwards at least to Costa Rica. It is apparent, however, that Hermeuptychia intricata is either more restricted in distribution and local, or significantly less common than Hermeuptychia sosybius, because several dozen available barcode sequences of Hermeuptychia specimens from different parts of the range in east US (NC, TN, FL, LA, OK and TX, see Fig. 66b) clearly group with Hermeuptychia sosybius, and a sample of 177 genitalically inspected Hermeuptychia specimens from 13 US states (MD, VA, SC, GA, TN, AR, AL, KY, MS, LA, TX & FL) in the USNM yielded only 8 Hermeuptychia intricata (less than 5%). We hope that a timely description of this species within a few months after its initial discovery will stimulate further studies of this interesting cryptic-in-facies butterfly, which, however, can be easily distinguished from its more common congener by genitalia (Figs 60a, c, d, f, g, i, j, l, 61a, c, 62n–z2 & 64a–p) and DNA barcodes (Fig. 66). All known Hermeuptychia sosybius records should be scrutinized in search for Hermeuptychia intricata.

In wing pattern, the new species is very similar to Hermeuptychia sosybius. We were not able to find solid diagnostic characters for the new species, and only hypothetical field marks could be suggested (see discussion). However, it could be easily identified by many distinctive characters of genitalia.

Males of the new species possess: (1) smaller and more robust and darker genital capsule, even in males with larger body size (Fig. 60c)–genitalia of Hermeuptychia sosybius from various parts of the range are larger and look “wider” and are paler (Fig. 60a); (2) narrower and apically pointed uncus (Fig. 60c, f)–uncus of Hermeuptychia sosybius is wider and appears truncated at the apex in dorsal or ventral views (Fig. 60a, d); (3) uncus that is more angled to the sides along the dorsal “rim”, thus appearing “higher” in lateral view (Fig. 60l), but flatter basally due to the lack of prominent carina, vs. a dorsally flatter uncus in distal half, with a well-developed thin, membranous carina in basal half in Hermeuptychia sosybius (Fig. 60j); (4) shorter and stouter cucullus, which projects for less than a third of its length farther than the distal ends of gnathos arms (lateral view, Fig. 60i, l)–cucullus in Hermeuptychia sosybius is more gracile, narrower and longer, it projects for close to half of its length farther than the distal end of gnathos (lateral view, Fig. 60g, j); (5) cucullus more rounded at the apex, usually with a couple of barely defined, very small apical teeth, vs. three to five (mostly four) larger teeth in Hermeuptychia sosybius; (6) interior surface of cucullus ventrally with a more prominent bulge, best seen in ventral view (Fig. 60f vs. 60d); (7) more stout, thicker and shorter penis, best seen in ventral view (Fig. 60f)–penis is more gracile, narrower and longer, especially near the distal end, in Hermeuptychia sosybius (Fig. 60d); (8) shorter phallobase, which is about as long as wide (Fig. 60i, l), vs. phallobase that is much longer than wide in Hermeuptychia sosybius (Fig. 60g, j); (9) smaller and narrower saccus (Fig. 60f), vs. larger and wider one in Hermeuptychia sosybius (Fig. 60d); (10) more obtuse angle formed by the tegumen and vinculum in lateral view (Fig. 60l), vs. typically more acute angle in Hermeuptychia sosybius (Fig. 60j).

Females of the new species possess: (I) narrower ostium bursae and smaller, darker antrum (Fig. 64i, j)–ostium bursae and antrum are larger and antrum is paler in color in Hermeuptychia sosybius (Fig. 64a, b); (II) ventral margin of ostium bursae that extends farther back than its dorsal margin (Fig. 64k, l)–dorsal margin extends posterior of ventral margin in Hermeuptychia sosybius (Fig. 64a, b); (III) antrum that is narrower anteriad, almost triangular in ventral view and symmetric (Fig. 64k, m), vs. rounder, cup-like, slightly asymmetric to the left antrum in Hermeuptychia sosybius (Fig. 64e); (IV) more bent antrum, kidney-shaped in lateral view (Fig. 64n), than that of Hermeuptychia sosybius (Fig. 64j); (V) signa composed of wider, more flattened and rounder spines, mostly in two rows, vs. narrower spines in three to five irregular rows in Hermeuptychia sosybius.

Characters (2) and (3) in males (more pointed apex of uncus and uncus more angled to the sides from the central “rim”) seem to be the easiest to examine without full dissection by brushing the scales off the abdomen tip, even in dry specimens (Fig. 62a, c). Identification of dry females might be more problematic due to abdomen shriveling, however, in freshly caught individuals, ostium bursae and antrum can be easily exposed by squeezing the abdomen in distal third, and the character (II) becomes observable (relative position of ostium bursae margins). Due to these very significant and easily observed differences in genitalia, identification in the field immediately after capture is expected to be straightforward, however, more work remains to be done to discover diagnostic wing pattern characters.

DNA barcodes, consistently with genitalia, set the new species far apart from sympatric Hermeuptychia sosybius, and the difference is about 3.5%, which is significantly higher than “a clear threshold for intra- and interspecific mean distances around 2%”, as quoted from the recent comprehensive analysis of Hermeuptychia (Seraphim et al. 2014).

While the discovery of this second (and new) Hermeuptychia species in eastern USA was very unexpected to us, the next finding is less surprising, although also interesting. Our analysis of DNA barcodes of Texas Hermeuptychia revealed that populations from the lower Rio Grande Valley region of Texas (Webb, Zapata, Starr, Hidalgo, and Cameron Counties) form a tight cluster differing by at least 2% from closely clustered barcodes (divergence average 0.09%, standard deviation 0.19%, maximum below 1%) of over 50 Hermeuptychia sosybius specimens across its range from North Carolina to Texas (south to Uvalde, Comal, Guadalupe and Brazoria Counties, Figs 66–67). These south Texas (and northeast Mexico) Hermeuptychia populations are phenotypically characterized by smaller and more uniformly sized eyespots and more undulated brown lines. This butterfly has been called “Hermeuptychia hermes” in some of the recent literature that advocates the presence of two Hermeuptychia species in the US (Miller and Brown 1981, brief comment in Neck 1996, Pelham 2008, Warren et al. 2013). However, DNA barcodes clearly and confidently group these populations with Hermeuptychia sosybius (Fig. 66a, bootstrap support above 80%, about 2% sequence difference), and Hermeuptychia hermes sequences are more than 4% different from either of these [Fig. 66a and Seraphim et al. (2014)]. According to DNA barcodes, Hermeuptychia hermes – type locality Brazil: Rio de Janeiro – is in a different species group and clusters with Hermeuptychia maimoune (A. Butler, 1870) rather than with Hermeuptychia sosybius (Fig. 66a). Analysis of male genitalia agrees with this conclusion. Indeed, genitalia of south Texas specimens are clearly from the morphogroup 4 (i.e. Hermeuptychia sosybius) possessing all the characters specified by Seraphim et al. (2014) and are very different from those of Hermeuptychia hermes [see Forster (1964) and Seraphim et al. (2014) for illustrations]. Most obviously, Hermeuptychia hermes has much longer saccus compared to shorter and more constricted in the middle valvae. Nevertheless, in addition to at least 2% different barcodes, south Texas morphogroup 4 populations differ from eastern Hermeuptychia sosybius in facies to the extent that researchers have been treating them as a species distinct from Hermeuptychia sosybius (Miller and Brown 1981, Pelham 2008, Warren et al. 2013). Our analysis agrees with this conclusion. Furthermore, we find subtle, but quantifiable, differences in male genitalia between Hermeuptychia sosybius and south Texas Hermeuptychia populations. Evidence presented above suggests that the name Hermeuptychia hermes should not be applied to them. Since currently there are no named species in the Hermeuptychia sosybius group [i.e., molecular group G and morphogroup 4 of Seraphim et al. (2014)] other than Hermeuptychia sosybius, and south Texas populations fall confidently in the Hermeuptychia sosybius group (Fig. 66a), they represent an unnamed species that is described here.

Male (n = 56, Figs 48–49, 52–56, 58–59) – holotype forewing length = 16 mm. Forewing triangular, rounded at apex and tornus, costal and outer margins convex, inner margin almost straight, mildly concave mediad, two discal cell veins budged at bases, vein 2A thickened basad. Hindwing rounded, almost circular. Wings dorsally dark-brown with sparse olive-beige overscaling and two darker-brown terminal lines. Wings ventrally pale-brown, paler towards inner margin of forewing, with extensive beige overscaling, particularly along veins in distal part in some specimens; submedial and postmedial darker- to rusty- and olive-brown lines and end-of-cell streak (smaller on hindwing) between them; hindwing postmedial line more undulate that in Hermeuptychia sosybius, with a stronger bend in M₁-M₂ cell; two terminal dark-brown evenly curved marginal lines, dark-brown sinuous submarginal line, more undulate than in Hermeuptychia sosybius, barely touching the eyespot in cell Cu₁-Cu₂, and row of submarginal eyespots basad of the sinuous line and posteriad of postmedial line, eyespots frequently reduced in size and are more uniformly sized than in Hermeuptychia sosybius; usually largest eyespots black-centered and pupiled with pale-blue scales: on forewing, eyespots about the same size, frequently larger posteriad, but eyespot in cell M₁-M₂ (usually not the largest in size) and eyespot in cell R₅-M₁ (in some specimens) black-centered (more eyespots black centered in some specimens); on hindwing, largest eyespots in cells M₁-M₂ and Cu₁-Cu₂, a smaller one in cell Cu₂-1A+2A, even smaller, but still black-centered and pale-blue pupilled in cell Rs-M₁, and two smallest, usually without black, but in some specimens pale-blue pupilled eyespots in cells M₂-M₃ and M₃-Cu₁. Fringes monochrome, a little paler than the ground color of wings. Head, palpi, thorax and abdomen dark-brown above, paler and mostly beige beneath. Antennae dark-brown above with pale scales at segments, orange-brown at the club, beneath beige basad, orange-brown in distal half. Legs brown with beige scales. Male genitalia (n = 19, Figs 60b, e, h, k, 61b, 62a–m, 63 part) – typical for the genus, very similar to those of Hermeuptychia sosybius. Tegumen dome-like, rounded at margins. Uncus leaf-shaped in dorsal view, almost flat distally but convex basally in lateral view, with a well-developed thin, membranous carina in basal half; apex of uncus appears truncated in dorsal view and sides usually less concave than in Hermeuptychia sosybius. Gnathos arms thin, wide apart, divergent, about the same length as uncus. Valvae narrow, but typically broader than in Hermeuptychia sosybius, elongated with thin cuculli extending past gnathos usually farther than a quarter of their length; cucullus usually with four apical teeth; cucullus ventrally with inner medial bulge. Saccus about the same length as cucullus, narrow. Aedeagus elongated, bent around its middle, with a medium length phallobase. Female (n = 45, Figs 50–51, 57) – similar to male in facies, with slightly more rounded wings and dorsally paler in color. Female genitalia (n = 9, Fig. 64q–z) as in Hermeuptychia sosybius, with pale, yellowish, weakly sclerotized and broad, rounder anteriad, cup-like antrum slightly asymmetric to the left. Ostium bursae ellipsoidal, its ventral margin shorter or equal to dorsal margin. Ductus and corpus bursae each in length similar to antrum; corpus bursae with two signa, spines in a signum narrow, leaf-shaped, placed in three to five irregular rows.

Full length DNA barcodes were obtained for 19 paratypes (GenBank accessions: KJ025569–KJ025587). The most common haplotype present in 17 sequences (including all 5 barcoded siblings of the holotype) is exemplified by the voucher NVG-1603, Genbank accession KJ025569, 658 base pairs:

AACTTTATATTTTATTTTTGGTATTTGAGCAGGAATAATTGGAACATCATTAAGTTTAATTATTCGAATAGAGTTAGGTAATCCAGGATTTTTAATTGGAGATGACCAAATTTATAACACTATTGTTACAGCCCATGCTTTTATTATAATTTTTTTTATAGTAATACCTATTATAATTGGAGGATTTGGTAATTGACTTATTCCTTTAATATTAGGAGCTCCTGATATAGCTTTCCCACGTATAAATAATATAAGATTTTGATTATTACCCCCATCTTTAATTTTATTAATTTCTAGTAGTATTGTAGAAAATGGAAGTGGAACAGGATGAACTGTTTACCCCCCTCTTTCATCTAATATTGCCCATAGAGGTTCTTCAGTAGATTTAGCAATTTTTTCTCTTCATTTAGCTGGAATTTCATCAATTTTAGGAGCCATTAATTTTATTACAACAATTATTAATATACGAATTAATAATATATCTTATGATCAAATACCTTTATTTATTTGAGCTGTAGGAATTACAGCTCTTCTTTTACTTCTCTCATTACCTGTTTTAGCAGGAGCTATTACCATACTTCTTACTGATCGAAATTTAAATACATCATTTTTTGACCCTGCAGGAGGAGGAGATCCTATTTTATATCAACATTTATTT

The 2 remaining sequences were identical to each other (Fig. 66b) and differed from the sequence shown above by a single base pair (0.15%). Barcode from the oldest and westernmost specimen (TX: Laredo, 15-Apr-1949) was additionally verified with both DNA ID tags as described in Materials and methods section and confirmed to be this species.

Holotype: ♂, has the following two rectangular labels: white printed - || USA: TEXAS: Cameron Co. | E of Brownsville, ex ovum | ex ♀ collected 18-Jan-2003 | ecl. 12-Mar-2003 Grishin N.V. ||; red printed - || HOLOTYPE ♂ | Hermeuptychia | hermybius Grishin ||. The holotype is illustrated in Figs 48–49. Upon publication, the holotype will be deposited in the National Museum of Natural History, Smithsonian Institution, Washington, DC (USNM). Paratypes: 55 ♂♂ and 45 ♀♀, from USA: Texas, unless indicated otherwise. Of these, 9 ♂♂ and 12 ♀♀ are siblings of the holotype read from ova, with the same data, their sexes, eclosion dates and GenBank accessions|DNA voucher numbers|genitalia codes (where available, and in this format for other paratypes) are: 1 ♀ 8-Mar-2003; 1 ♂ 9-Mar-2003, KJ025572|NVG-1610|NVG131017-02 (Fig. 62b); 2 ♂♂ and 1 ♀ 9-Mar-2003; 1 ♂ and 1 ♀ 10-Mar-2003; 1 ♂ and 1 ♀ 11-Mar-2003; 3 ♂♂ 12-Mar-2003; 1 ♀ 14-Mar-2003, KJ025573|NVG-1611|NVG131017-03 (Fig. 64s–t); 1 ♀ 15-Mar-2003; 1 ♀ 16-Mar-2003, KJ025574|NVG-1612|NVG131017-04 (Fig. 64u–v); 1 ♀ 17-Mar-2003, KJ025569|NVG-1603|NVG130927-17 (Fig. 64q–r); 2 ♀♀ 17-Mar-2003; 1 ♀ 21-Mar-2003; 1 ♂ 30-Mar-2003, KJ025571|NVG-1609|NVG131017-01 (Fig. 62a); 1 ♀ 2-Apr-2003 (Figs 50–51). Other paratypes are: 1 ♂ ibid., collected on wing 18-Jan-2003, KJ025570|NVG-1607|NVG130927-18 (Figs 54–55, 60b, e, h, k). 1 ♀ Cameron Co., E of Brownsville, 19-Oct-1997, leg. N. V. Grishin, KJ025575|NVG-1628|NVG131017-05. 1 ♂ Cameron Co., Brownsville, {10-13}-Mar-1979, leg. T. Friedlander, NVG140104-01 [TAMU] (Fig. 62c). 1 ♂ (06-Jun-2007) 1 ♀ (07-Jun-2007) Cameron Co., Los Fresnos, Ted Hunt & Loop Rd., leg. William R. Dempwolf. 4 ♀♀ Hidalgo Co., 1.5 air mi SE of Relampago, Rio Rico Rd., 26.07, -97.891, 21 m, 13-Jun-2013, leg. W. R. Dempwolf; 2 ♂♂ ibid., 19-Oct-2013, KJ025577|NVG-1698|NVG131229-04 (Fig. 62d) and KJ025578|NVG-1699|NVG131229-05 (Figs 56, 62e); 1 ♀ ibid., 19-Oct-2013, KJ025576|NVG-1695|NVG131229-03 (Fig. 64w–x); 3 ♂♂ 4 ♀♀ ibid., 19-Oct-2013; 2 ♂♂ 4 ♀♀ ibid., 21-Oct-2013; 3 ♂♂ ibid., 24-Oct-2013. 1 ♀ TX: Starr Co., Rio Grande City, Fort Ringgold, 26.3707, -98.8064, 45 m, 12-Nov-2010, leg. W. R. Dempwolf; 1 ♀ ibid., 13-Jun-2013; 1 ♂ ibid., 20-Oct-2013, KJ025580|NVG-1714|NVG131229-07 (Fig. 62f); 1 ♀ ibid., 20-Oct-2013, KJ025579|NVG-1712|NVG131229-06; 2 ♂♂ ibid., 20-Oct-2013; 1 ♂ ibid., 23-Oct-2013; 2 ♂♂ 1 ♀ ibid., 9-Nov-2013. 2 ♂♂ Starr Co., Roma, S of Roma International Bridge, 26.4035, -99.0175, 50 m, 20-Oct-2013, leg. W. R. Dempwolf, KJ025581|NVG-1726|NVG131229-08 (Fig. 62g) and KJ025582|NVG-1727|NVG131229-09 (Fig. 62h); 8 ♂♂ 7 ♀♀ ibid., 20-Oct-2013. 1 ♀ Starr Co., Roma Creek, Hwy 650/Hwy 83, 29-Oct-2007, leg. W. R. Dempwolf. 2 ♀♀ Starr Co., 0.5 mi S of Fronton, 26.399, -99.085, 50 m, 20-Oct-2013, leg. W. R. Dempwolf, KJ025583|NVG-1735|NVG131229-10 and KJ025584|NVG-1737|NVG131229-11 (Figs 57, 64y–z); 7 ♂♂ 3 ♀♀ ibid., 20-Oct-2013. 1 ♂ Starr Co., Salineno @ Rio Grande, 26.51463, -99.11633, 53 m, 23-Oct-2013, leg. W. R. Dempwolf, KJ025585|NVG-1747|NVG131229-12 (Fig. 62i). 1 ♂ Zapata Co., San Ygnacio @ Rio Grande, 92 m, 7-Oct-2007, leg. N. V. Grishin, KJ025586|NVG-1635|NVG131017-12 (Figs 52, 61b, 62j). 1 ♂ Webb Co., Laredo, 15-Apr-1949, leg. E. L. Todd KJ025587|13385H10|NVG131102-53 [USNM] (Figs 53, 62k). 1 ♂ Mexico: Tamaulipas: Rt. 101 at Rio Corona, 1-Jan-1980, leg. P. W. Kovarik & D. S. Bogar, NVG140104-04 [TAMU]. 1 ♂ Mexico: Tamaulipas: El Canindo, nr. Ejido San José, 7.5 km W Gómez Farías, 1400 m, {19-21}-Jul-1994, leg. C. Cate & T. Riley, NVG140104-67 [TAMU]. 2 ♂♂ Mexico: Tamaulipas: Ciudad Mante, Los Arcos Ct., 19-Dec-1973, leg. R. O. & C. A. Kendall, NVG140104-22 and NVG130104-23 [TAMU] (Figs 58, 62m); 1 ♂ ibid., 28-Jan-1995, ex larva, foodplant Panicum maximus Jacq., NVG140104-24 [TAMU]. 1 ♂ Mexico: Tamaulipas: Quintero cave [22.6333, -99.0333], 7-Jan-1974, leg. R. O. & C. A. Kendall, NVG130104-24 [TAMU] (Figs 59, 62l). 1 ♂ 1 ♀ Mexico: San Luis Potosí: El Salto Falls, 30-Dec-1979, leg. P. W. Kovarik & D. S. Bogar, NVG140104-03 and NVG140104-02 [TAMU].

USA: Texas: Cameron County, east of Brownsville. It is a shaded area covered in Guinea grass (Panicum maximus), situated near a ravine and overgrown with taller trees.

The name is a fusion of two words: herm[es] beginning and [sos]ybius ending. It symbolizes that this species traditionally and previously regarded as Hermeuptychia hermes is phylogenetically closer to Hermeuptychia sosybius, and yet is distinct from it. The resulting word is unique and currently unknown to internet search engines, which is expected to ease its searches. The name is a noun in apposition.

This species is currently recorded from the lower Rio Grande Valley region of Texas along the Rio Grande from Laredo to the Gulf coast (Webb, Zapata, Starr, Hidalgo, and Cameron Counties, Fig. 67) and in neighboring Mexico (Tamaulipas, San Luis Potosí).

In wing pattern, the new species is most similar to Hermeuptychia sosybius, but typically can be differentiated from it by: (a) eyespots that are not only smaller, but also more uniform in size, i.e. out of 5 forewing eyespots, 4 (except the one near costa) are usually about the same size, and the eyespot that is black-ringed in most specimens (second from costa) is typically not the largest (this eyespot is frequently the largest in Hermeuptychia sosybius), but the next-to-last eyespot (4^th from the costa) is usually the largest one; (b) more undulate postmedial line on ventral hindwing, that frequently strongly bulges basad by the largest eyespot near apex (in cell M₁-M₂); (c) more undulate submarginal sinuous line, which on ventral hindwing barely touches the largest eyespot near the tornus (in cell Cu₁-Cu₂, second eyespot from tornus, indicated in Fig. 57)–this line is usually fully merged with this eyespot border for some distance in Hermeuptychia sosybius. Wing-based identification is not absolute due to extensive pattern variation in both species.

In male genitalia, the new species is also closest to Hermeuptychia sosybius and should be attributed to the same morphogroup 4 of Seraphim et al. (2014). It differs from Hermeuptychia sosybius in the following trends (Figs 60–61): (1) uncus is less convex and narrower on the sides in dorsal (or ventral) view, with a broader truncated apex, the width at the apex is usually more than 2/3 of the width at the narrowest point near the base (Figs 60b, e, 61b); (2) valva is typically “higher” in lateral view (dorso-ventral direction), more square at the base (Fig. 60k) and is less extended (Fig. 60h); (3) aedeagus is somewhat broader and is frequently bent near its middle, with a medium length phallobase (Fig. 60e); (4) usually more obtuse angle formed by the tegumen and vinculum in lateral view (Fig. 60k). These characters are quite subtle, and as illustrated in Fig. 62 (compare panels a–m with panels o–z2) are subject to significant variation. In contrast, distinction of Hermeuptychia intricata (Fig. 62n for comparison) is always definitive and clear-cut. To evaluate the confidence of Hermeuptychia hermybius identification by male genitalia and to test the ability to differentiate this new species from Hermeuptychia sosybius by objective criteria, we resorted to morphometric analysis (Fig. 63). For simplicity, we have chosen to exploit only two trends listed above: (1) shape of uncus in dorsal view and (2) shape of valva base in lateral view. The shape of uncus was measured by the ratio of width at the apex (a) to the width at the narrowest point near the base (b), and by the ratio of the distance from apex to the widest point in cross-section (c) to the distance from apex to the narrowest point near the base (d). We noticed that both of these ratios tend to be smaller in Hermeuptychia sosybius. Instead of applying PCA or other similar data-driven technique, which may be biased by the data at hand (i.e. the resulting transformation would change with the dataset used), we combined these measurements in a data-independent transformation. We used a weighted sum of the two ratios, with the weight of the second ratio arbitrarily set to half the weight of the first one: a/b+0.5c/d, since the ratio of widths (first ratio) seemed to tell the species apart better than the ratio of lengths (second ratio). The shape of the valva base in lateral view was quantified by the ratio of length of the dorsal “window” (less sclerotized, membranous and flat segment along dorsal side near the base) to the height of the valva at the distal end of the “window”. These variables were measured and computed on a diverse sample of 27 genitalia illustrated in Fig. 62. The resulting plot (Fig. 63 on the right) separated the two species. Therefore these simple measurements could be used to tell between these two cryptic Hermeuptychia species by male genitalia. However, we were not able to find characters in female genitalia to differentiate the new species from Hermeuptychia sosybius.

Finally, the most confident identification is provided by DNA barcode sequences (Fig. 66) that show little variation within each species (most sequences are identical across the range, maximum difference below 1% in Hermeuptychia sosybius), but reveal a definitive 2% hiatus between central and south Texas populations (Figs 66–67). We selected all positions that were invariant in the barcode sample of each species but different between the two species as characters to differentiate Hermeuptychia hermybius from Hermeuptychia sosybius. The resulting 11 positions are listed in the format “k X (not Y)”, where k is a sequential number of the position (numbering is from 1 to 658 for the barcode sequence shown above as a reference), X is a nucleotide in Hermeuptychia hermybius barcodes and Y is a nucleotide in Hermeuptychia sosybius barcodes: 64 T (not C), 73 G (not A), 82 T (not C), 118 C (not T), 133 C (not T), 235 C (not T), 238 A (not G), 364 C (not T), 436 C (not T), 526 A (not T), 616 C (not T). These positions distinguish the two species; however, some of the positions are expected to show variation when a larger sample of sequence is accumulated.

The holotype of the new species, along with 21 paratypes are specimen reared in the lab from ova obtained from a captive female. All life history stages are illustrated in Fig. 70, and could be compared to the images of Hermeuptychia sosybius life history (Fig. 69). Immature stages of both species are very similar and without larger sample it is difficult to derive solid conclusions about the differences. Nevertheless, the following observations were made. Natural foodplants seems to be Panicum maximus (Guinea grass) per R. O. Kendall & C. A. Kendall, who reared caterpillars found on this grass in Mexico: Tamaulipas [TAMU collection]. This plant is also common in the lower Rio Grande Valley and is ubiquitously present where Hermeuptychia adults were encountered. Caterpillars hatched from eggs in captivity readily accepted Cynodon dactylon (L.) Pers. (Bermuda grass) and were successfully reared on it. Both Hermeuptychia sosybius and Hermeuptychia hermybius caterpillars go through four instars prior to pupation, and the first instar has black head capsule (Figs 69b–d, 70b–c). In subsequent instars, head capsule is green and round, without horns and projections (Figs 69e–m, 70d–m). Caterpillars of both species typically rest below leaves on loosely made silk pads, frequently in pairs, when two caterpillars face each other “head-to-head” (Figs 69h, 70c). When disturbed, caterpillars first curl into a C head-to-tail while legs being attached to the leaf (Figs 69f, j, 70e), then to a full O, head-to-legs (Fig. 70g). White dorsolateral spots in ultimate instar seem to be more pronounced in Hermeuptychia hermybius than in Hermeuptychia sosybius (compare Fig. 70g–k with 69k). Pupae of Hermeuptychia hermybius were stronger patterned with brown on the sides (Fig. 70o) than those of Hermeuptychia sosybius from two distant-from-each-other Texas localities (Fig. 69n–o), and some Hermeuptychia hermybius pupae were brown in color (Fig. 70n).