Notes on the genus Celotes, with the description of a new species from Mexico (Lepidoptera: Hesperiidae: Pyrginae: Pyrgini) Author Warren, Andrew D. Author Steinhauser, Stephen R. Author Hernández-Mejía, Claudia Author Grishin, Nick V. text Zootaxa 2008 1926 27 40 journal article 10.5281/zenodo.184811 dea75d02-e7dd-4a10-8b72-29687f16215e 1175-5326 184811 Celotes spurcus , new species A. Warren, Steinhauser, Hernández-Mejía & Grishin Figs. 3–6 Description. Male ( Fig. 3 : 1, 3–4, Fig. 4 a,b): FW length = 14.1 mm ( holotype ), 13.3 mm ± 2.0 mm (n=25). Forewing width 0.6 times length, with a prominent costal fold reaching Sc. Termen undulate, concave between veins; anal margin prominently curved, w-shaped. Ground color brown, each cell except 2A, with areas of dark-brown, almost black scales. Three opaque, white, narrow, median macules from mid-costa caudad, located in Sc, R1 and discal cells (DC), with dark scaling prominent along veins separating the macules. Macules in Sc and R1 usually extend from vein to vein, macule in discal cell situated in its anterior half and typically the narrowest, while macule in Sc is the broadest. Shape of macules variable, most frequently rectangular, triangular and trapezoidal for the Sc, R1, and DC macules, respectively. Apical cluster of three opaque white macules in R3, R4 and R5 cells, smaller than median macules, R3 macule largest, R4 smallest, in some specimens dot-like; macule R4 offset basally and the cluster of the three macules is crescent-shaped. Finally, two postmedian opaque white macules in cells M3 and CuA1. Macule M3 occupies very basal area of cell between M3 and CuA1 and has a triangular shape similar to R1 macule. Macule CuA1 located at base of M3 macule and oddly-shaped, varying from rectangular, to trapezoid, triangular and crescent-shaped; the two macules form a roughly C-shaped arrangement. Each of these eight opaque macules surrounded by darker, almost black scales. In addition to surrounding the macules, similarly colored dark scales form four dark patches in the median area of CuA2 cell, two cephalad and two caudad, the cephalic pair sometimes fused in a dumbbell-shaped streak; a patch at base of CuA1 cell; two patches, just caudad of apical macules, in median area of M1 and M2, the latter patch being smaller and basad of the former; last, and smallest patch situated in R2 and in some specimens occupies entire cell. Dark brown scales (intermediate in color to the dark, almost black, scales described above) and brown background are formed into streaks in the marginal area; two streaks located in each cell between veins reaching the margin, except in 2A. Streaks start from the margin, in each cell from each vein, and are directed basad and towards each other in a cell, converging in postmedial area; streaks around vein M2 and caudad of CuA2 not well developed and lacking in paler specimens (in particular the M2 streaks), creating a paler appearance of the marginal region of M1 and M2 cells. Similar intermediate-color scales line up the basal area of the forewing. Ivory-colored scales, paler than the background, cover the areas distad, between, and proximad two anterior dark patches in CuA2 cell, distad to the cluster of apical white macules, distad discal cell, and form streaks from the margin in the middle of each cell, in between dark streaks. The amount of ivory-colored scaling varies between individuals. Finally, somewhat yellower and longer ivory-colored scales cover the basal area of the wing (overscaling). Hindwing termen undulate, concave between veins. Pattern similar to forewing, in particular posterior to Rs vein. Opaque macules in cells Rs, M2 and CuA1 homologous to macules on forewing. Main differences from forewing include absence of median row of three macules (no macules in Sc+R1 and discal cells), more basal location of dark patches in M1 and M2 (in line with the opaque macules), separation of CuA1 macule into two small macules in some specimens, and more extensive ivory-scaled areas, present between CuA1 and CuA2 dark patches. Cell Sc+R1 with a median dark patch, median brown streak and ivory-colored area between. Pattern is largely repeated on ventral surface of wings , but background is lighter, yellowish. Cells CuA2 and 2A mostly covered in whitish, ivory-colored scales between the darker patches and streaks, which are composed of much paler scales, as the dorsal background. Dark scales, matching those of dark patches on dorsal surface, are present basad of opaque macules, frequently reaching the base of each cell, and distad of the macules, reaching the costa in cells Sc and R1, and forming an edge to each macule in other cells; also filling basal half of cell R2. Hindwing ventral pattern more similar to that of dorsal surface than on forewing, except prominent white scaling at wing base, and areas of white scales in basal half of DC. Fringes on both wings light-gray, checkered dark, with darker scales at vein termini, dark-scaled intervals about 1/3–1/2 length of light-scaled intervals. Head dark-brown, white between antennae and dorsad of mid-eyes. Caudal portion of collar black, with white scales cephalad, extending behind eyes. Eyes black. Palpi brown with white scales present between the segments dorsad, and white ventrad. Antennae about half length of costa, reaching end of Sc, black dorsad, with prominent white scales between segments, paler ventrad, in particular around the club and apiculus, with many white scales; club about half of shaft length, apiculus pale brown ventrad, poorly differentiated from club, nudum of 11 segments in holotype (varies from 10 (n = 1) to 12 (n = 7) among 25 specimens counted). Thorax and abdomen dark-brown with light overscaling dorsad, white and ivory ventrad. Dorsal surface of metathoracic pouch ( Fig. 4 a,b) covered with flat, rounded shingle-like scales, whitish to cream-colored, linear scales present proximally, cream-colored. Legs light brown to ivory-colored, externally scaled whitish. Forecoxae white, foretibiae with slender pale brown epiphyses reaching tarsus; midtibiae smooth with single pair of spurs, hindtibiae with two pairs. Hindlegs each with two tufts of long, setiform scales, associated with metathoracic pouch, as follows; dark gray scales originating at proximal end of femur, just over 1 mm in length, extending to distal end of femur; pale brown scales, almost 2 mm in length, originating at proximal end of tibia, extending beyond distal end of tibia to overlap proximal margin of tarsi. FIGURE 3. Specimens of Celotes . 1–4 C. spurcus , all from Mexico: Querétaro: Mpio. Peñamiller: 4.2 rd. mi (6.8 km) W Peñamiller on camino Peñamiller-Boquillas; 5–8 C. nessus and 9–12 C. limpia ; d and v denote dorsal and ventral views, respectively. 1. Holotype ♂, 3-Sep-2007; 2. Allotype Ψ, 2 -Sep-2007; 3. Paratype ♂, paler specimen 2-Sep-2007; 4. Paratype ♂, smaller specimen 26-Mar-2001; 5. ♂ TX: Val Verde Co. 7 mi (11.3 km) E Del Rio, el (ex larva) 23-Dec- 1966, larval foodplant Abutilon fruticosum , leg. R.O. & C.A. Kendall; 6. Topotype Ψ, TX: Bexar Co. San Antonio, NE Preserve, el 18-Sep-1960, larval foodplant Abutilon incanum , leg. R.O. & C.A. Kendall; 7. ♂ AZ: Pima Co. Molino Basin, 31-Mar-1978, leg. J.P. Brock; 8. Ψ AZ: Pima Co. Molino Basin, el 22-Jun-1990, larval foodplant Ayenia compacta , emerged 26-Jul-1990, leg. J.P. Brock; 9. Paratype ♂ TX: Presidio Co. SH170 ca. 10-15 mi (16.1–24.1 km) SE Redford, el 31-Aug-1966, larval foodplant Abutilon malacum , leg. R.O. & C.A. Kendall; 10. Paratype Ψ TX: Jeff Davis Co. Musquiz Cyn SH118, el 14-Apr-1967, had larval diapause, larval foodplant Allowissadula holosericea , leg. R.O. & C.A. Kendall; 11. ♂, 12. Ψ, both TX: Brewster Co. Big Bend National Park, E of Chisos Basin Campground, el 16-Sep- 2005, larval foodplant Sphaeralcea angustifolia , leg. N.V. Grishin. FIGURE 4. Metathoracic pouch of Celotes males. Abdomen removed, dorso-posterior view. a , b ) C. spurcus ; c ) C. nessus ; d ) C. limpia ; top row ( a , b , c , d ) is the entire view of the two “formations”; bottom row (denoted by prime, a’ , b’ , c’ , d’ ) is the left “formation” magnified, of the specimen shown in the same column above. a , b ) Paratypes, MX: Querétaro: Mpio. Peñamiller: sandy gulch 4.2 rd. mi (6.8 km) W Peñamiller on camino Peñamiller-Boquillas, 3-Sep-2007, leg. A.D. Warren; c ) TX: El Paso Co. Franklin Mountains State Park, Cottonwood Spring Trail, 16-Apr-2007, leg. N.V. Grishin; d ) Topotype, TX: Jeff Davis Co. Davis Mountains State Park, Indian Lodge, 14-Apr-2007, leg. N.V. Grishin. The scale is the same in each row, bottom row is magnified 2.25× compared to the top row. Genitalia ( Fig. 5 a–j): Tegumen short as in C. nessus , not projecting prominently cephalad as in C. limpia . Uncus divided as deeply as C. nessus , but arms more divergent. Gnathos and saccus as in both C. nessus and C. limpia . Valvae similar to C. nessus , broad, but the long caudo-dorsal projection, which is as in C. limpia but thinner, more curved, extends further dorsad and widely separated from the harpe, rather than overlapping as in C. limpia . Harpe similar to C. nessus , narrows terminally, not as broad and robust as in C. limpia . The ampulla of C. spurcus protrudes further distad and dorsad than in C. limpia , thus slightly overlaps the harpe rather than being separated as in C. limpia . Penis long and slender with single tooth on left side near distal end, longer than C. limpia , but shorter than C. nessus ; ratio of penis length to length of right valva from cephalic end to caudal end of harpe is 1.38 in C. spurcus , 1.51 in C. nessus and 1.18 in C. limpia . Juxta simple, Ushaped sclerotized yoke, transtilla very weakly sclerotized, bearing long, distally directed hairs. Female ( Fig. 3 : 2): FW length = 14.1 mm (allotype) (n = 1). Forewing and hindwing pattern, dorsal and ventral , as in male, but lacking extensive pale overscaling at bases of wings above, presumably due to flightworn condition; wings slightly broader than in male. Female differs from male in lacking secondary sexual characters (forewing costal fold, metathoracic pouch, and tufts of hair-like scales on hindleg). Genitalia ( Fig. 6 a–e): Sterigma in ventral view about as long as wide, strongly narrowing caudad, terminally about 1/2 its basal width. Posterior half of sterigma with sides almost parallel to the body axis; anterior half trapezoidal, sides at an angle of about 30° with body axis. Caudal margin of lamella postvaginalis concave in middle, with asymmetric sides and irregular margin, heavily sclerotized terminal section bean-shaped, about 1/3 of sterigma length. Ostium bursae wide, as in C. nessus . Type material. Holotype male bearing the following labels: printed white label / MEXICO : QUERÉTARO: / Mpio. Peñamiller: sandy gulch / 4.2 rd. mi ( 6.8 km ) W Peñamiller on / camino Peñamiller- Boquillas / 21°04’53’’N 99°50’33’’W / 1465m , 3-IX-2007 / Andrew D. Warren /. Hand printed red label / HOLOTYPE male / Celotes spurcus / A. Warren, Steinhauser / C. Hernández-Mejía & Grishin/. Allotype female is from the same locality as the holotype male, 2-IX-2007 . There are 91 male paratypes : 39 males same locality and date as holotype , Andrew D. Warren and John Kemner; 32 males from same locality as holotype , 2-IX-2007 , Andrew D. Warren and John Kemner; 12 males same locality as holotype , 26-III-2001 , Andrew D. Warren, Thomas W. Ortenburger and Jose Luis Salinas-Gutierrez; 5 males , MEXICO : QUERÉTARO: Mpio. Cadereyta de Montes: sandy gulch SE Río Mezquitillo, just NE of Hwy. 120 vic. Puente Mezquitillo, 9.2 rd. mi ( 14.8 km ) N Vizarrón de Montes, 1460m , 20°56’23’’N 99°44’13’’W , 1-IX- 2007 , Andrew D. Warren and John Kemner; 3 males , MEXICO : SAN LUIS POTOSÍ, between Cárdenas and Ciudad del Maíz, 8-VII-1988 , John Kemner. The holotype and various paratypes are deposited in the Museo de Zoología “Alfonso L. Herrera”, Facultad de Ciencias, Universidad Nacional Autónoma de México , Mexico City. Additional paratypes are deposited in the McGuire Center for Lepidoptera and Biodiversity, Florida Museum of Natural History, University of Florida, and other collections. Type Locality ( Fig. 2 a,b). MEXICO : QUERÉTARO: Mpio. Peñamiller: 4.2 rd. mi ( 6.8 km ) W Peñamiller on camino Peñamiller-Boquillas, 1465m , 21°04’53’’N 99°50’33’’W . The site where the holotype and most paratypes were collected is a dry sandy wash in desert thornscrub habitat. Prominent plants in the habitat include various cacti, ocotillo ( Fouquieria Kunth ), Celtis L., and several species of malvaceous plants, possibly Abutilon . Adult males are most frequently encountered in sandy or rocky areas in the bottom of dry gullies, and males visit moist patches of sand and gravel. Males are typically nervous, and generally remain perched only for a few short moments ( Fig. 2 c–f). Distribution and phenology. Celotes spurcus is currently known from desert thornscrub habitats in central Querétaro State, Mexico , north to east-central San Luis Potosí State, Mexico ( Fig. 1 ). The species is not likely to occur far to the south of central Querétaro, since apparently appropriate habitats are eliminated by the rise in elevation of the Eje Neovolcánico of central Mexico . The northern distributional limits of C. spurcus remain conjectural, and will be determined only through future fieldwork. Burns (1974) reported C. limpia from southeastern Coahuila State, Mexico , from a single male taken 25 miles north of Saltillo. Considering the existence of C. spurcus , this specimen requires re-examination. If this specimen was determined only by dry examination of the valvae (by removing a few anal scales; Burns (1974) did not provide details), confusion with (the then unknown) C. spurcus cannot yet be ruled out. Biogeography. While Celotes is seemingly adapted to dry and seasonally hot conditions, the genus is characterized by an austral Nearctic distribution (sensu Halffter 1987 ). The Sonoran and Chihuahuan deserts appear to represent the center of Celotes distribution, although the genus extends onto the southern Great Plains of North America in New Mexico , Texas and Oklahoma ( Burns 1974 , Stanford & Opler 1993 ), and into the Sierra Madre Occidental and Sierra Madre Oriental in Mexico , almost to the Eje Neovolcánico of southern Mexico ( Fig. 1 ). Celotes spurcus occupies a region at the transition between the eastern edge of the Mexican Altiplano and the Sierra Madre Oriental. Its habitat can be considered an ancient southeastern extension of the Chihuahuan Desert, with faunal and floral influences extending south to the Tehuacán-Cuicatlán area of endemism in Puebla-Oaxaca ( Espinosa et al. 2006 ). Larval foodplants and early stages. Intensive searches for immature stages of C. spurcus were not conducted, but several malvaceous plants at the type locality were identified as potential larval foodplants, such as Abutilon . As a result, life history details for C. spurcus remain unknown, but should be similar to those of C. nessus and C. limpia (briefly described in the introduction). Etymology. Burns (1974) named C. limpia for the locality, Limpia Canyon, where the holotype was taken. Since “ limpia ” translates from Spanish as “clean”, and limpia is indeed the palest and most contrastingly-marked Celotes , it seemed appropriate to name this generally darker taxon “ spurcus ”, Latin for “dirty.” Diagnosis and discussion. Superficially, Celotes spurcus ( Fig. 3 : 1–4) looks very much like C. nessus ( Fig. 3 : 5–8) and C. limpia ( Fig. 3 : 9–12), but averages slightly larger and darker than both, with somewhat smaller opaque white wing macules. Burns (1974) was unable to identify wing pattern characters to separate C. limpia from C. nessus . Similarly, we were unable to discern reliable superficial wing markings that could be used to identify C. spurcus , largely due to similarity in wing patterns between all Celotes species, and because of great individual variation in the size and shape of macules, patches and streaks on all three species. Therefore, C. spurcus becomes yet another pyrgine that can be reliably determined only through genitalic examination ( e.g., see Steinhauser 1989 , Burns 2000 ). However, male genitalic differences between species of Celotes are so profound and so numerous that no doubt remains in the specific distinction of C. spurcus . Males of C. spurcus are immediately separated from the other two Celotes species by the distinctive long curved process (= projection, = prong, = style) from the ampulla, which can be seen by brushing away some anal scales from the abdomen ( Fig. 5 a,d,i). This process is particularly well developed on C. spurcus , claw-like, and armed with several terminal teeth, at least some teeth being longer than wide ( Fig. 5 d). Although C. limpia has a homologous process, it is less-developed and less curved ( Fig. 5 m). More precisely, the following features of male genitalia separate C. spurcus from C. limpia : 1) The process on the valva is more curved, C-shaped, in C. spurcus , and this in lateral view does not overlay the harpe; process is armed with larger terminal teeth. A homologous process in C. limpia is straighter, smaller, in lateral view (unless intentionally twisted and moved away) typically overlays the harpe and passes distad from it, teeth are less developed. 2) The valva is broader in C. spurcus than in C. limpia , and is strongly concave cephalad of the process ( Fig. 5 d,i). Valva is flatter cephalad of the process in C. limpia , and is narrower ( Fig. 5 m). Width of valva (measured at the widest section, which is around the middle of the valva) is less than 0.5 times its length (not including the harpe, from the base to the end of ampulla, at which point harpe protrudes caudad) in C. limpia , and larger than 0.5 in C. spurcus . 3) The ampulla of C. spurcus protrudes farther distad and dorsad than in C. limpia , thus almost or slightly overlapping the harpe, rather than being far apart as in C. limpia . In other words, the ampulla looks “cut” at around 30 degrees in C. limpia , and it appears “cut” almost vertically in C. spurcus . Due to this feature, the origin of the process in C. spurcus is well ventrad of the dorsal margin of valva, while the process originates very close to the dorsal margin in C. limpia . 4) The harpe narrows gradually towards the terminus in C. spurcus , whereas the harpe of C. limpia is more robust and follows a nearly constant width almost until the very terminus, where it rounds off; thus being broader just near the terminus than the harpe of C. spurcus . 5) The tegumen is short in C. spurcus ( Fig. 5 e,j), as in C. nessus , not projecting prominently cephalad as in C. limpia . 6) The uncus is narrower than the tegumen in C. spurcus ( Fig. 5 e,j), whereas the uncus is broad, about the same width as the tegumen in C. limpia ( Fig. 5 n). 7) The penis is longer in C. spurcus than in C. limpia ; ratio of penis length to length of right valva from cephalic end to caudal end of harpe is about 1.4 in C. spurcus , and about 1.2 in C. limpia . Features 2 and 4 may be the easiest to observe when differentiating between long-pronged species of Celotes , do not require full genitalic preparation, and may be visible after brushing some scales off the distal tip of the abdomen. The following features of the male genitalia separate C. spurcus from C. nessus : 1) The process on the valva is very short and spike-like in C. nessus , projecting dorsad, with no more than four terminal teeth ( Fig. 5 k). The process does not curve caudad terminally in any significant way. 2) The harpe is better developed in C. spurcus , further separated from the ampulla; its section that is directed cephalad is longer and less concave at the distal margin ( Fig. 5 d,i). The harpe in C. nessus is smaller, almost touching the ampulla by its cephalicpointing section, which is less clearly separated from the caudal-directed section, and is shorter and somewhat concave at the distal margin near the terminus ( Fig. 5 e,j). 3)The penis is usually slightly shorter in C. spurcus than in C. nessus ; the ratio of penis length to length of right valva from cephalic end to caudal end of harpe is about 1.4 in C. spurcus , and about 1.5 in C. nessus . Interestingly, the list of prominent differences between C. nessus and C. spurcus is shorter than that between C. limpia and C. spurcus , despite the presence of a long valval process in both C. limpia and C. spurcus . In fact, it appears that C. spurcus could essentially be considered a “long-pronged” version of C. nessus , however, this pronounced feature itself should be enough to warrant species-level status. Such long-prong/ short-prong species pairs are well known in other genera, for instance, Staphylus Godman & Salvin ( S . hayhurstii (W. H. Edwards) vs. S. mazans (Reakirt)) and Systasea W. H. Edwards ( S . pulverulenta (R. Felder) vs. S. zampa (W. H. Edwards)) ; see Steinhauser (1989) and Miller (1970) , respectively. Although it is premature to reach a definitive conclusion from a single specimen, it is not clear at present whether it is possible to distinguish females of C. spurcus from those of C. nessus using genitalic characters. Females of C. limpia can be easily separated from the other two species by: 1) the sterigma appears more square-shaped ventrally in C. limpia and more trapezoidal in the other two species ( Fig. 6 g); and 2) the ostium bursae is narrower in C. limpia than in the other two species. A potential difference between C. spurcus and C. nessus that needs to be substantiated through the collection and genitalic examination of many more females of C. spurcus , is the shape of sterigma, which narrows caudad more strongly in the single C. spurcus female ( Fig. 6 e) than that in C. nessus ( Fig. 6 f,f’), and the section of the lamella postvaginalis with the margins more or less parallel to the body axis (caudal section) is about half the length of the sterigma in C. spurcus , but is only about one third the length in C. nessus . We have illustrated dry as well as traditional wet mounts of female Celotes genitalia in order to facilitate in situ genitalic examination in the lab or field. Additional support for the species-level status of all three Celotes taxa is provided by features of the metathoracic pouch of males ( Fig. 4 ). As discussed by Burns (1974) , in C. limpia , the metathoracic pouch is dorsally covered with very thin, linear, pale scales ( Fig. 4 d); in C. nessus these scales are round and gray, with linear scales, if present, restricted to the proximal area ( Fig. 5 c). Although C. spurcus shares the lack of linear scales (except proximally) and presence of shingle-like, round scales with C. nessus , these scales are light and cream-colored (not gray as in C. nessus ), which in conjunction with the darker aspect of C. spurcus (compared to C. nessus ) is unexpected and appears to be indicative of a species-level difference from C. nessus ( Fig. 5 a,b). The color of the pouch scales in C. spurcus is more like that seen in C. limpia than in C. nessus . In summary, characters of the genitalia (long prong, shape of the harpe) and metathoracic pouch of C. spurcus are strong indicators of specific distinction between C. spurcus and C. nessus and leave no doubt that possibly allopatric populations of C. spurcus constitute a separate species-level taxon.