Athis miastagma (Dyar, 1925), stat. rest.

(Figs. 1C–1D)

Castnia miastagma Dyar, 1925: 19.

Material Examined: COLIMA: 1♂, Tropical deciduous forest, La Salada, Colima, Mexico, 300m, 15/Jun/2015, Leg. B. López (RW); 1♂, Tropical deciduous forest, La Salada, Colima, Mexico, 300m, July 2016, Leg. B. López (RW); 1♂, Tropical deciduous forest, La Salada, Colima, Mexico, 300m, 20/Jun/2015, Leg. B. López (RW); 1♀, Tropical deciduous forest, La Salada, Colima, Mexico, 300m, 12/Jun/2015, Leg. B. López (RW); 1♀, Tropical deciduous forest, La Salada, Colima, Mexico, 300m, 20/Jun/2015, Leg. B. López (RW); 1♂, La Salada, Colima, 29 Junio 2015, Leg. B. López (JJGD); 1♀, La Salada, Colima, 10 Julio 2015, Leg. B. López (JJGD); GUERRERO: 1♂, Type, Guerrero, Mexico, May, R. Müller, Cat. No. 27905, (mentioned in Dyar 1925 as a female); 1♂, CNIABM, Castnia hectiae [sic], No. 3523, país Mexico, miastagma, Guerrero, 3523, Coll. R. Müller (MHNCM); 1♂, Acahuizotla, Gro [Guerrero], VIII-[19]57, T. Escalante (MGCL); 1♂, Guerrero, Mexico, 4-2014 (DC) [this specimen is phenotypically similar to the Jalisco-Colima population, so the collecting data might be incorrect]; JALISCO: 1♂, Tolimán, Jalisco, Mexico, Tuxcacuesco, Jalisco. 1–9 julio 2013, 800 mts, Coll. B. López G. (BLG); 3♂♂, Jalisco, Mexico, 2014, Coll. B. López G. (BLG); 2♂♂, Tuxcacuesco, Jalisco, Mexico, 2014, Coll. B. López Godínez (BLG); 1♂, Tuxcacuesco, Jalisco, Mexico, 800m, 8/Jul/2014, leg. B. López (RW); 1♂, Tuxcacuesco, Jalisco, Mexico, June 2013, leg. B. López (RW); 1♂, Tuxcacuesco, Jalisco, Mexico, 820m, Jul/2013, leg. B. López (RW); 1♀, Tuxcacuesco, Jalisco, Mexico, 800m, 15/Jul/2009, leg. B. López (RW); 1♂, Mexique, Tuxcacuesco, 820m, Jalisco VII/2013 (DC); 2♂♂, Tuxcacuesco, Jalisco, 06 Julio 2017, Leg. B. López (JJGD); MICHOACÁN: 1♀, Putja, Michoacán, XII-[19]49, T. Escalante (MGCL) [after gathering and requesting information from many resources and collectors, we have been unable to identify this locality, therefore we consider it doubtful]; MORELOS: 1♂, Huitzilac, La Pera, 10/May/1967, col. R. G. de la Maza E. (CDM); 1♀, Tlaltizapán, Tlaltizapán, col. A. Díaz F. (IBUNAM); 1♀, Tlaltizapán, Temilpa, col. A. Díaz F. (IBUNAM).

Diagnosis of adults (phenotypes). The forewing of A. hechtiae is gray, with the ground color darker basally (Figs. 1A–1B). A creamy-white band goes from the base of the forewing, over CU to the bifurcation of CU1–CU2, filling the beginning of the interspace of CU1–CU2, running laterally and obliquely toward C in the sub-apical area of the wing. Before reaching C, the band ends in the interspace R5-M1. As an extension of the creamy-white band, an elongated, small, creamy-white spot is sometimes present in the interspace between R4–R 5 in the apical region. Hindwing orange, almost brick-red in mid cell and along the anal angle (tornus); a strongly bent dark to almost black band at end of discal cell, a submarginal row of pear-shaped spots and the margin narrowly black, the black running inward along the veins. Females are larger and usually paler than males, with orange tones more noticeable in the forewings dorsally (Fig. 1B). The female abdomen is orange and large whereas that of the male is gray and shorter. Wing expanse: Males: 51–67mm, Females: 72–85mm. Forewing length: Males: 28–38mm, Females: 41–51mm.

In the case of A. miastagma, the forewing ground color is gray (as in A. hechtiae) but darker, diluting into lighter brown sub-marginally towards the anal angle (Figs. 1C–1D). As in hechtiae, a creamy-white band goes from the base of the forewing, over CU to the bifurcation of CU1–CU2, filling the beginning of the interspace of CU1–CU2, to run laterally and obliquely toward C in the sub-apical area of the wing. However, in all specimens we have examined, an extension of the oblique section of the band runs backward toward the dorsum (anal margin) without touching it (Figs. 1C–1D). A largish creamy-white cuneiform spot is placed in interspace R4–R5 (Figs. 1C–1D). A dark, almost black, oval dot can be noticed outside the bifurcation of the creamy white band in the forewings. This dot is not found in hechtiae. Hindwing orange with a large dark brown almost black spot with an orange center in the discal cell. The black areas of the hindwings are very diffused. They also have a submarginal row of small slightly elongated orange spots. The females differ from the males in the same way as hechtiae. Wing expanse: Males: 75–80mm, Females: 80–100mm; Fore wing length: Males: 40–44mm, Females: 48–58mm.

Specimens of hechtiae are usually smaller than those of miastagma, although both taxa share similar wing shapes. The overall coloration tends to be darker in miastagma than in hechtiae. The creamy-white spot in the space between R4–R 5 in the apical region is always present and highly noticeable in miastagma, but it is not always present or is much smaller in hechtiae. The creamy-white branch that runs towards the tornus opposite the oblique band that runs towards C, is always present in miastagma but never in hechtiae. The area surrounded by this band and the one coming from the base of the wing to the discal cell is dark brown, almost black, in miastagma. This region, even though it is slightly darker than the rest of the forewing in hechtiae, is never as dark as in miastagma. Hindwings in hechtiae are mostly orange with the presence of an extradiscal band of orange spots that are easily seen, while those in miastagma are very dark and both postmedian and extradiscal bands made of dark spots are present.

Male genitalia (Fig. 2). The male genitalia of both taxa are close (especially the aedeagus), but slight differences can be easily identified as follows. Uncus simple, fused into a blunt apex, which is almost abruptly curved in hechtiae (Figs. 2A–2C) but gradually curved in miastagma (Figs. 2D–2F). Gnathos is moderately sclerotized in both species, fused anteriad, and highly excavate posteriad, especially in miastagma. Subscaphium lightly sclerotized anteriad but almost membranous posteriad in both taxa. Cucullus and valvae are short and almost quadrate in hechtiae but long and almost rounded in miastagma. Sacculus weakly developed in both taxa; saccus short, blunt, recurved ventrad in hechtiae, wider and not as recurved ventrad in miastagma. Aedeagus curved, twice the length of the coecum, moderately sclerotized, and recurved anteroventrally in both taxa (Figs. 2A–2F).

Species behavior. Athis hechtiae flies from late May to early August, from around 10:30–11:00h to 14:00– 15:00h on sunny days, with temperatures varying from 22 to 35°C. They fly later, around 12:00h, on cloudy days. They may also fly on cold days or under a light drizzle, but never on rainy days. When temperatures become warmer around midday (28–35°C at 12:00–12:30h) they become more active and fly faster, visiting plants. Males are most active from 12:30 to 14:00h, and females generally fly about an hour later. They tend to fly through open spaces where vegetation is low and Hechtia plants are abundant. Males tend to fly fast, in an “up and down” movement, reaching heights up to 1.5–2.5m above the ground and vegetation. Females have a similar flight pattern, but they are slower than males.

All individuals will perch on twigs or leaves of bromeliads or Agave plants once the temperature cools down or the day becomes cloudy. They typically perch with their forewings almost covering the hindwings, in a stegopterous position like many Castniidae (Miller 1986; Ríos & González 2011; Vinciguerra et al. 2011; González et al. 2019).

Males of A. hechtiae are territorial and easily disturbed by other males or potential predators. Males are frequently seen chasing other males and can then fly as high as 15–20m above the ground.

Athis miastagma differs behaviorally from A. hechtiae. Males begin to fly around 11 am on sunny days (one of the authors, BLG, has observed them flying on cloudy days only once), and activity ceases at 3 pm. Like most known Mexican castniids, they are territorial and fly at a height of 2–4 m, however, they often fly over stone walls up to 20m high, since their host plant is found throughout the area. They usually perch on dried leaves of bromeliads (rarely on green leaves) or some other plant. Compared to other Mexican castniids, the males of A. miastagma are not very active and remain perched most of the time (up to 1 hour); only if another male or female passes by does the perched male become active and fly. They only fly in sunny areas.

As in A. hechtiae, the females of A. miastagma fly later, between 1:30 pm and 3 pm. Females are rarely seen, mostly flying slowly among the bromeliads, possibly looking for suitable ones on which to lay their eggs. They are slow and ‘heavy’ fliers. They attract a lot of attention from the males, nearly always bringing a group of 3 or 4 males chasing them. Interestingly, when the males chase a female, they don’t fight, they just chase her. Like the males, females are very local and do not usually fly away from their bromeliad ‘colonies’, mostly flying around those found near rock walls of up to 20m.

Genetic analyses. Genetic distances ranged overall from 0.0000% to 0.1573% (K2P) and 0.1417% (p) (Tables 2). The average genetic distances of A. hechtiae to A. miastagma were 0.040 8 (K2P, Table 2.1) and 0.039 6 (p, Table 2.2) while intraspecific distances of A. hechtiae was 0.000 0 and intraspecific distances of A. miastagma was 0.000 5 (Fig. 3).

......continued on the next page

TABLE 2.1. (Continued)

......continued on the next page

TABLE 2.1. (Continued)

......continued on the next page

TABLE 2.2. (Continued)

......continued on the next page

TABLE 2.2. (Continued)

The alignment of the COI barcode region with 18 species contained 658 bp including 156 distinct patterns, 107 parsimony-informative, 64 singleton sites, and 486 constant sites. We further manually inspected the COI alignment for diagnostic characters, so-called character-based DNA barcodes (DeSalle et al. 2005; Rach et al. 2008). We only analysed pure diagnostic characters for A. hechtiae and A. miastagma. Pure diagnostic characters consisted of all the nucleotide positions of individuals from A. hechtiae that differed from those in all individuals of A. miastagma. Table 3 lists a total of 25 identified character-based DNA barcodes for A. hechtiae and A. miastagma.

The Maximum Likelihood (ML) analyses as implemented in IQ-TREE and using GTR+F+G4 as the best-fit model according to Bayesian information criterion (BIC) showed a clear separation of A. hechtiae (98% bootstrap support) and A. miastagma (97% bootstrap support) (Fig. 4).