Dyscophellus doriscus (Hewitson, 1867)

Zhang, Jing, Cong, Qian, Shen, Jinhui & Grishin, Nick V., 2022, Taxonomic changes suggested by the genomic analysis of Hesperiidae (Lepidoptera), Insecta Mundi 2022 (921), pp. 1-135 : 18-19

Dyscophellus doriscus (Hewitson, 1867) , reinstated status

Eudamus doriscus Hewitson, 1867 (type locality Brazil: Rio de Janeiro) has been treated as a subspecies of Dyscophellus porcius (C. Felder and R. Felder, 1862) (type locality “upper Rio Negro”) from Southeast Brazil since Evans (1952). Genomic analysis reveals notable separation between the two taxa with Fst/Gmin statistics on Z chromosome-encoded protein of 0.32/0.03 ( Cong et al. 2019a). These numbers suggest genetic diversification and limited gene exchange between the two taxa. Phenotypically they differ by the number of hyaline spots in males ( Evans 1952). Therefore, we propose species-level status for Dyscophellus doriscus (Hewitson, 1867) , reinstated status. Our genomic analysis included two syntypes of Netrocoryne coecutiens Herrich-Schäffer, 1869 from Brazil: Rio de Janeiro in the ZMHB (NVG-15031G01 and G02), which are D. doriscus , and specimens from Venezuela, Peru and Bolivia for D. porcius . Despite the notable genetic diversification in nuclear genomes, COI barcodes of these species differ by only 0.9% (6 bp), albeit consistently without much variation within each species.

Telegonus diophorus Möschler, 1883 is a junior objective synonym of Bungalotis corentinus (Plötz, 1882) , reinstated status

Telegonus corentinus Plötz, 1882 (type locality Suriname), whose drawing (No. 1333) according to Godman (1907: 151) was missing from the original set made by Plötz, has been since Mabille ( Mabille 1903) treated as a junior subjective synonym of Papilio midas Cramer, 1775 (type locality Suriname). However, Bungalotis midas specimens do not agree with the original description of T. corentinus . T. corentinus was described in a key to Hesperiidae species ( Plötz 1882c), and was the next species to B. midas , both unified by the following characters, as translated from German original: “Without hyaline spots. Forewing basad in cell 1 unspotted. Rust-yellow, hindwing above from vein 7 to the costa brown. Tornus somewhat pointed.” And also for T. corentinus : “Hindtibiae with very long hairs.” The lack of a brown spot doublet at the basal third of forewing cell CuA 2 -1A+2A excludes Dyscophellus Godman and Salvin, 1893 . The lack of hyaline spots combined with rusty-yellow (not brown) color excludes all other related genera except Bungalotis E. Watson, 1893 . Furthermore, according to Evans (1952: 137), “densely fringed” tibiae are characteristic of Bungalotis . Therefore, it is most probable that T. corentinus indeed belongs to Bungalotis .

Plötz’s key clearly spells out the differences between Bungalotis corentinus and B. midas . First, in B. corentinus : “Upper side almost without markings, the most noticeable is a brown spot in the middle cell of the hindwing.” In contrast, for B. midas we have: “Upper side of all wings with a brown spot in the middle and an unequal, curvy cross-band against the margin.” While we do find poorly marked specimens of B. midas , they are not common and they tend to have forewing spots more prominent that the discal cell spot on hindwing. Second, in B. corentinus : “Underside brownish, forewing only with a row of small brown spots against the margin, hindwing with a larger central spot and a circle of smaller ones, almost all white-centered.” In B. midas : “On the underside, … the markings of the forewing are as above, the hindwing has two transverse bands consisting of large square spots and a smaller one in cell 7 next to the base.” The description of B. midas is quite accurate. However, in B. midas males we inspected, the hindwing brown spots do not have white centers, as stated by Plötz for B. corentinus . Third, forewing length of B. corentinus is 28 mm, which is on the lower side for B. midas with the forewing length 30 mm as given by Evans (1952). Thus, if B. corentinus is indeed B. midas , as currently assumed ( Mielke 2005), it would have been one of the smallest, poorest-marked specimens with white-centered small ventral hindwing spots. Out of dozens B. midas we have seen, none matched this description. Therefore, B. corentinus is not likely to be B. midas .

Next, we attempted to locate syntypes of B. corentinus . We searched carefully all Hesperiidae drawers in the ZMHB collection, including the supplemental drawers that may contain additional syntype specimens not currently labeled as types. We also searched Hesperiidae holdings in the ZSMC that contain a number of Plötz type specimens. In these collections, the specimen that comes closest to the original description of T. corentinus is the holotype of Telegonus diophorus Möschler, 1883 (NVG-15031G10) in the ZMHB, also from Suriname. It is not likely that this specimen was a syntype of T. corentinus , because the most prominent spot on dorsal hindwing is the one closest to the dark costal area, not the discal cell spot as mentioned in the description (see above). Out of all specimens we have seen, the specimen that matches the Plötz description best is the specimen (NVG-15026B10) identified as Bungalotis gagarini Mielke, 1967 (type locality Brazil: Goiás) by Austin (2008) and illustrated in his figures 24 and 25. However, this specimen is from Brazil: Rondonia, not Suriname. We sequenced both of these specimens, and they are apparently conspecific ( Fig. 1 ). Furthermore, we sequenced another specimen from Rondonia (NVG-15026B11, also an excellent match to the original description of B. corentinus ) with genitalia GTA #1617 illustrated by Austin (2008: Fig. 89) as B. gagarini , and an old specimen from the Schaus collection in the USNM from French Guiana (NVG-17104D08) identified as B. diophorus . All these specimens cluster tightly together in the tree ( Fig. 1 ) and their COI barcodes show only a couple of base pair difference among them, suggesting that they are all conspecific, and are B. diophorus , because the B. diophorus holotype is among them. While we leave the question about possible synonymy of B gagarini and B. diophorus for future studies pending genomic sequencing of B. gagarini holotype, we use this opportunity to objectively define the taxonomic identity of B. corentinus by neotype designation. Here, N.V.G. designates the holotype of Telegonus diophorus Möschler, 1883 as the neotype of Telegonus corentinus Plötz, 1882 , making the former a junior objective synonym of the latter. It is the only species known to us that is a perfect match to the original description of T. corentinus .

We believe that there is an exceptional need to designate this neotype, not only because the name B. corentinus has been misapplied and its current treatment is inconsistent with its original description thus creating a source for future instability of names, but also because of an opportunity to correct the following long-standing confusion between orthographically similar names. The two names currently in use are Bungalotis diophorus (Möschler, 1883) and Dyscophellus diaphorus (Mabille and Boullet, 1912) . Their species epithets differ by only one letter: o vs. a. Their males are quite similar in appearance, most notably distinguished by a doublet of dark spots towards the base of forewing cell CuA 2 -1A+2A, absent in B. diophorus and present in D. diaphorus . A mnemonic to remember: o means no spots; o fused with l to form a, where l stands for the vertical doublet of spots, means spots. This spot doublet character was also mentioned in the key by Plötz (1882c). This similarity in names and appearance is a source of many confusions. We are taking this opportunity given by the misidentification B. corentinus that allows us to put the confusion behind and set the record straight about the true identity of B. corentinus , a name proposed earlier than B. diophorus .

Our neotype of B. corentinus satisfies all requirements set forth by ICZN Article 75.3, namely: 75.3.1. It is designated to clarify the taxonomic identity of Telegonus corentinus Plötz, 1882 , which has been inconsistent with its original description; 75.3.2. The characters for the taxon have been given in its original description by Plötz (1882c: 78) (some are discussed above), and also by Evans (1952: 138) as those for B. diophorus (keys out to D.1.2.); 75.3.3. The neotype specimen is also the holotype of Telegonus diophorus Möschler, 1883 , with the following labels: || Surinam | Prb. | Wd. | 79 | || Type. | Verh. z-b. Ges: Wien. | 1882. p.322. || Diophorus | Möschl. || Origin || Coll. | Staudinger || Coll. Möschl. || Diophorus | Möschl. || GEN.PREP., | MIELKE | 1996 || [barcode image] http://coll.mfn-berlin.de/u/ | 940b51 || DNA sample ID: | NVG-15031G10 | c/o Nick V. Grishin ||; 75.3.4. Our search for the syntypes is described above, it was not successful, and we consider that the specimens composing the type series of T. corentinus are lost; 75.3.5. As detailed above, the neotype is consistent with the original description, more, it apparently is the only currently known species that matches the original description; 75.3.6. The neotype is from Suriname according to its label, which is the type locality of B. corentinus ; 75.3.7. The neotype is in the collection of the Museum für Naturkunde, Berlin, Germany (ZMHB).

Figures

Figure 1. Genomic tree of Eudaminae. The tree is constructed from protein-coding regions. The tree is rooted with Pterourus glaucus (Linnaeus, 1758) (NVG-1670), not shown. Statistical support values are shown on nodes. For each specimen, its name adopted in this work is given first, and a previously used name is listed in square brackets (if different), supplemented with the DNA sample number, type status (see Table 2 caption for abbreviations) and general locality. See Table S1 in the Supplemental file for additional data about these specimens. Synonyms are given in parentheses preceded by “=”, and in addition by “‡” for unavailable names. The type status refers to this synonym, if the synonym name is provided. Clades corresponding to new genera, subgenera and species are colored in red, orange and green, respectively, and the names of new taxa are highlighted accordingly. Names of selected tribes, subtribes, genera, and subgenera are labeled at their clades. Clades corresponding to some genera are highlighted in yellow to compare their genetic diversifications (length of the highlight in horizontal dimension) with each other. Groups of similar genetic diversification would ideally correspond to genera. The