Mondeguina, a new genus for Apatetris mediterranella Nel & Varenne, 2012, with description of a new species from Portugal (Lepidoptera, Gelechiidae)

Following the discovery of a new species on the western coast of Portugal which is closely related to Apatetris mediterranella Nel & Varenne, 2012, the generic placement of both species is considered in relation to other genera within the Apatetris complex resulting in the description of a new genus, Mondeguina.


Introduction
Apatetris mediterranella Nel & Varenne, 2012 was described from the coasts of France, Corsica and Italy. The species was observed to have distinctive characters, particularly the long antennae and unusually slender hindwings. These characters had baffled the authors for some time as they considered at one stage that the species might belong to Gracillariidae or even to Neomariania Mariani, 1943 in the Stathmopodidae. Eventually it was recognised as belonging to Gelechiidae. In spite of the unusual characters for this family, the species was placed in the genus Apatetris Staudinger, 1880. The authors considered the possibility of placing it in a new genus, but were advised not to do so because Apatetris and related genera remain unrevised on a global scale.
In 2017, J.R. found a small moth species near the Portuguese coast which he identified as possible A. mediterranella. Subsequently M.C. examined specimens and considered that this could represent an undescribed species. Meanwhile a DNA barcode was obtained within Portugal-based In-BIO Barcoding Initiative from a previously unidentified specimen collected in 2015 by J.R. which showed a 5.5% pairwise divergence from A. mediterranella, indicating that this was likely a related but different species. Detailed examination of male and female genitalia confirmed this hypothesis.
In this paper we describe a new genus, Mondeguina to accommodate Apatetris mediterranella and the new species from Portugal described below and we discuss the reasons for excluding them from the genus Apatetris.

Molecular data
All samples amplified the COI barcoding fragment. The final COI dataset consisted of 29 sequences (658 bp long) (Fig. 1). The most appropriate model for the COI dataset was GTR+G+I. Tree topologies from ML approach exhibit high bootstrap values (≥0.95%) to all species (Fig. 1).
Within the dataset the maximum pairwise divergence within genus was obtained between Epidola nuraghella Hartig, 1939 and Epidola stigma Staudinger, 1859 (p-distance = 9%) and the minimum pairwise divergence between species (4%) was observed between Dactylotula kinkerella (Snellen, 1876) from Germany and Denmark and D. kinkerella from southern countries that belong in fact to D. altithermella (Walsingham, 1903) (Table 2, see discussion). The specimens of the Mondeguina species exhibited 5% pairwise sequence divergence and 10% or more from other genera, namely Apatetris and Catatinagma. Description. Wingspan 9-12 mm. Head with face flat and vertex smooth with appressed scales. Haustellum rudimentary. Labial palp segment 2 porrect, slightly thickened with scales which extend beyond base of segment 3, segment 3 upturned, about half as long as and thinner than segment 2, obtuse at apex. Antenna about nine-tenths length of forewing, scape elliptic, bearing pecten of about 12 hair-like scales, flagellum filiform. Forewing elongate, five times as long as wide, costa slightly curved in middle, apex acute but appearing rounded due to long scales around apex, darktipped scales give whole wing a speckled appearance, indistinct markings formed by groups of scales with longer dark tips; fringe of larger scales with blacker tips, not organised into a fringe line, dorsal fringe of hair-like scales. Hindwing slender, half as wide as forewing, widest at one-quarter, tapering most strongly between half and three-fifths, apical two-fifths very slender, narrowly acute at apex; fringes whitish, as long as wing width on costal side near base, longer towards apex and up to four times wing width on dorsal side. Venation (figured by Nel and Varenne 2012) of forewing with cell long, R 4 and R 5 stalked, R 5 to apex, M 1 and M 2 stalked, M 3 absent; hindwing with reduced venation, Rs free and unbranched to apex, M 1 absent, M 2 reduced to a fold, M 3 , Cu 1 and Cu 2 curved, A 1+2 very short. Legs with unequal pair of spurs on mid-tibia, and two unequal pairs on hind tibia.
Male genitalia. (Figs 4-6). Uncus inserted on posterior side of tegumen, consisting of a pair of stout lobes; gnathos absent; tegumen wider than long; valva just longer than its width at base with broad sacculus reaching beyond middle, apex slightly expanded on ventral side forming a group of short sclerotised pegs; saccus short, wide and rounded, widest in middle; aedeagus tapering from base or a little above to two-thirds, apical third narrower sometimes ending in a slight hook.
Female genitalia. (Figs 7, 8). Papillae anales almost spherical but somewhat flattened on their adjacent faces, with a transverse fold on the dorsal side; posterior apophysis less than length of papilla analis; segment VIII about twice as wide as long, dorsal side rectangular, ventral side largely excavated, anterior apophysis short, half as long as posterior apophysis or less; ostium on anterior margin of segment VIII; ductus bursae unsclerotised, corpus bursae elliptic, without signum.
Diagnosis. Mondeguina gen. nov. is characterised by antenna nearly as long as forewing, scape with pecten; forewing venation (figured by Nel and Varenne 2012) M 1 and M 2 stalked, M 3 absent; hindwing slender, only half as wide as forewing, not bilobed, with dorsal fringes up to four times as long as wing width, venation reduced, with M 1 absent and M 2 weakly marked as a fold. Male genitalia with contiguous pair of large uncus lobes, gnathos absent, tegumen wider than long. Female genitalia without signum. These characters combined clearly separate the new genus from Apatetris Staudinger, 1880 in the wide sense, in which the antennae, although with pecten, are about twothirds length of forewing; forewing with M 1 , M 2 and M 3 separate; hindwing more than half as wide as forewing, bilobed with a pointed apex separated by an emargination from a shorter more rounded lobe, fringes mainly about twice as long as wing width; uncus not consisting of two separate lobes; two signa (rarely more) present in most species, although absent in Apatetris sensu stricto. Janse (1951) analysed the Apatetris group in an even wider sense, including some species that lack the diagnostic characters of Apatetris (antennal pecten and bilobed hindwing) and described new South African genera (Filisignella, Curvisignella and Anapatetris) which resemble Mondeguina in having hindwings not bilobed, but they all differ substantially in wing venation from Mondeguina.

Etymology.
Mondeguina is named after the Mondego River, the longest river that runs its entire course within Portugal. The type locality for the new species described from Portugal (below) is the Ilha da Morraceira which lies in the estuary of the River Mondego.
The Description. Wingspan 9-12 mm (Figs 2, 3). Head white, scales towards lateral margins tipped fuscous. Labial palp white with scales tipped fuscous on outer and ventral sides and lower part of inner side, segment 3 white. Antennal scape white with scales tipped fuscous, flagellum white, more or less ringed light fuscous in proximal part but markings increasingly obscure distally. Thorax white, scales tipped dark fuscous towards sides and on tegula. Forewing white, scales darktipped giving whole wing a dotted appearance, very indistinct markings formed by groups of scales with longer dark tips, particularly in mid-line of wing with a spot at two-fifths and a short streak around four-fifths, a small spot near dorsum at three-fifths; fringe of larger scales with blacker tips, not organised into a fringe line, dorsal fringe of hair-like scales; underside dull pale yellow more or less overlaid grey, particularly towards costa and apex. Hindwing whitish; fringes whitish, as long as wing width on costal side near base, longer towards apex and up to four times wing width on dorsal side. Legs white, scales tipped fuscous. Abdomen whitish yellow.
Variation. Ground colour may be white or whitish, the dark tips of the scales on head, scape, thorax, forewing and legs vary from fuscous to nearly black with blackest scales forming diffuse markings together with those at end of wing, the diffuse markings can be very obscure and may not be symmetrical on the two wings.
Male genitalia. (Figs 4, 6). Uncus a pair of stout lobes, twice as long as wide, tapering in distal half to truncate apex; tegumen approximately rectangular, wider than long, nearly as long as uncus lobes, anterior and posterior margins without significant bulges; valva as long as its width at base with broad sacculus reaching beyond middle and evenly curved distally, posterior portion of valva parallel-sided, apex rounded on costal side, slightly expanded on ventral side forming a group of about seven short sclerotised pegs; saccus deepest in middle where it is about one-third as long as uncus lobes; aedeagus tapering from base or a little above, to two-thirds, apical third narrower, sometimes ending in a slight hook.
Female genitalia. (Fig. 7). Posterior apophysis two-fifths length of papilla analis; segment VIII with ventral side largely excavated, anterior margin bulging craniad around the ostium, anterior apophysis short, half as long as posterior apophysis; ostium about one-quarter width of segment VIII. Apparent variation. Male genitalia of Mondeguina are small, three-dimensional and difficult to open in conventional presentation, whether the genitalia are unrolled or not. Some structures remain difficult to see clearly or to interpret. For this reason, we have not described transtilla lobes. We also found apparent variation in shape of valva, depth of tegumen and saccus depending both on the presentation of the various parts in the preparation and on the person performing the dissection. This is exemplified by comparison of Figure 4 with Figure 6, which are preparations of male M. atlanticella performed by different persons.
Diagnosis. M. atlanticella and M. mediterranella are very similar externally, although average wingspan is slightly greater in M. atlanticella (9-12 mm against 8-10 mm). M. atlanticella differs from M. mediterranella in male and female genitalia (Figs 4-8). In male genitalia, M. atlanticella has tegumen about twice as wide as long, without obvious bulge on anterior margin, and saccus deeper, two-fifths length of uncus lobe, while M. mediterranella has tegumen four times as wide as long, with length measured through the median bulge, and less deep saccus, about one-quarter length of uncus lobe. In the female of M. atlanticella, posterior apophysis is two-fifths length of papilla analis and ostium is about one-quarter width of segment VIII, while in M. mediterranella posterior apophysis is two-thirds length of papilla analis and ostium is two-fifths of the width of segment VIII.
Biology. Moths have been collected at light in late June and July and once in September. Although some came directly to the light, the majority were collected at rest on grasses a few metres away from the light. Larvae have not been found, but they may be leaf-miners on Elymus athericus (Link) Kerguélen.
Ilha da Morraceira lies in the tidal estuary of the Rio Mondego close to the mouth of the river. It consists of a mosaic of abandoned and active salinas (salt-pans) (Figs 9, 10). The vegetation is diverse, with few shrubs, but many different herbaceous plants on the sides of the access roads, on the banks separating the individual salinas and in abandoned ones. The lower parts of the separating banks are dominated by Halimione portulacoides (L.) Aellen and Juncus maritimus Lam., but the grass Elymus athericus (Link) Kerguélen is also abundant, and may be the food-plant of M. atlanticella. The other site on the western coast of Portugal, Lagoa de São José is a small lake behind sand dunes which is gradually silting up. It is surrounded by a forest of Pinus pinaster Aiton. The sandy soil around the lake has a variety of shrubs: Erica L. spp., Cistus L. spp., Halimium (Dunal) Spach spp., and Corema album (L.) D. Don with scattered larger shrubs such as Morella faya (Aiton) Wilbur, Rhamnus alaternus L., Arbutus unedo L. and Salix repens L. Herbaceous plants include Juncus spp., Ammophila arenaria (L.) Link, Elymus L. sp., Silene littorea Brot., Iberis procumbens Lange, Seseli tortuosum L. and Asphodelus L. spp.
The third site, 2 km north of Ludo is 3.5 kilometres inland from the south coast of Portugal. The trapping site was on a track separating a low sandy hill with Pinus pinaster and a rich understorey of shrubs and herbs on one side and a small floodplain meadow bordering the Ribeira de São Lourenço on the other side.  Distribution. M. atlanticella is currently known from two sites about 22 kilometres apart, close to the coast of Beira Litoral in central Portugal and a single site in Algarve, north-west of Faro airport (Fig. 11). It can be expected to be found more extensively along the Portuguese coast.
Etymology. The epithet atlanticella refers to the proximity to the Atlantic Ocean of the known sites and contrasts with the Mediterranean coast distribution of M. mediterranella (Fig. 11).
Discussion. Recent authors discussing the problems of the Apatetris group (Sakamaki 2000;Junnilainen and Nupponen 2010;Nel and Varenne 2012) have emphasised the lack of a global revision of the group, although Janse (1951) had been able to characterise several of the genera involved, largely based on wing venation, wing shape and the structure of the labial palpi. Classification of gelechiid genera in the second half of the 20 th century relied heavily on male genitalia and this has been a significant impediment to further revision of the Apatetris group because the type species of the genus, Apatetris mirabella Staudinger, 1880 was described from a single female from the province of Amasya in the Black Sea region of Turkey. Staudinger stated it to be male, but Janse (1951) showed it to be female. It has apparently never been collected again. The main features of the genus given by Staudinger were the antennal scape with pecten and the bilobed hindwing, with the costal lobe pointed and the dorsal lobe rounded, these lobes separated by a notch. Janse (1951) after studying material from Europe, Asia, Africa, North and Central America constructed a key to some of the genera of the Apatetris group and re-diagnosed the genus, adding the additional characters of a rudimentary haustellum and maxillary palps, the labial palpus shorter and less ascending than usual for Gelechiidae. Sakamaki (2000) recognised the high level of diversity within Apatetris sensu lato and the need for a world-wide revision of the genus. His contribution was to describe two new Japanese species in the genus which he considered would help to advance knowledge of the group. One of these species, A. elymicola is close to A. kinkerella (Snellen, 1876), which is now placed in Dactylotula (e.g. Karsholt and Nielsen 2013). The other, A. elaeagnella, while belonging to Apatetrini, appears to be a misfit in Apatetris: female genitalia have a belt of small signa whereas other species have a pair of signa or none and the labial palpus is unique in being reduced to a single segment. Junnilainen and Nupponen (2010) recognised two different groups within Eurasian Apatetris, referring them to the existing genus names Dactylotula and Catatinagma. These differ in several characters of which the most useful are forewing markings, valva structure and type of signum. Dactylotula has forewing without spots, short, broad valva and plate-shaped signa; Catatinagma has forewing with spots, valva almost separate from sacculus and brush-shaped signa. Figure 1 indicates a more complex situation, even if only European species of Apatetris sensu lato are considered. Four groups are evident: 1. Catatinagma; 2. Mondeguina; 3. An unnamed group including specimens from Italy and the Balkan countries; 4. Dactylotula.
This separation appears valid for Europe and the nearer parts of Asia but relationships are likely to be more complex if further taxa from the rest of the world are considered. In Figure 1 and Table  1 three species of the Australasian genus Epiphthora are included, but one is well separated from the other two. This suggests that the genus is likely to include paraphyletic elements.
The four European groups in Figure 1, derived from available DNA barcodes, require a few additional comments. In the course of preparation of this paper we also examined Portuguese specimens identified as Apatetris agenjoi Gozmány, 1954 (including INV06257). Junnilainen and Nupponen (2010) had omitted this species from consideration "because it is known only from Spain and does not coincide well with any of the genera treated here". Nevertheless, male genitalia and wing markings of this species place it in Catatinagma, and it also shares the head structure of Catatinagma, which has a truncated conical projection on the frons. Furthermore, genetic evidence provided by the DNA barcode fragments corroborates that A. agenjoi is more closely related to Catatinagma trivittellum Rebel, 1903 andC. kraterella Junnilainen &Nupponen, 2010 than to any other species of Apatetris sensu lato (Fig. 1, Table 2). Although Gozmány (1954) neither described nor figured the genitalia, Ole Karsholt (pers. comm.) informs us that he has seen the type of A. agenjoi and that it belongs to the species flying in the Iberian Peninsula. With this assurance we are able to make the new combination: Catatinagma agenjoi (Gozmány, 1954), comb. nov. Gozmány, 1954. Ann. Hist.-nat. Mus. Nat. Hung. 5: 282. Type locality. Spain, Murcia, Alberca. We have not had the opportunity to examine the unnamed 'Apatetris' from the Balkan countries, and consider it to be outside the scope of this paper. Available sequences from BOLD indicate two Dactylotula species (Table 1), yet all have been referred to D. kinkerella. From the photographs on BOLD, it is clear that those from the Czech Republic, France and Spain actually belong to D. altithermella (Walsingham, 1903).

Basionym. Apatetris agenjoi
Notes. Against the background outlined above, it may be considered unwise of us to describe a new genus based on a species already placed in Apatetris. We justify our decision by pointing out that M. mediterranella should never have been described in Apatetris as the shape of the hindwings is quite different from that which defines the genus. In addition, the unusually long antennae are not mentioned for any species in the Apatetris group, indeed Staudinger (loc. cit.) gave the antenna length as a little over half forewing length; forewing venation with M 1 and M 2 stalked appears to be unique in the Apatetris group; the absence of a signum in the female genitalia is also different from most Apatetris sensu lato species, although according to Janse (1951), A. mirabella has no signum. In our view the two species considered in this paper, A. mediterranella and A. atlanticella have no place in Apatetris as defined by either Staudinger or Janse, nor in any of the related genera and therefore require a new genus. Genetic evidence provided by the DNA barcode fragments sequenced corroborates their distinctness from other Apatetris sensu lato available in the BOLD database, either European or from other continents (Fig. 1).