Eutocus rogan ( Evans, 1955 ), 2023

Zhang, Jing, Dolibaina, Diego R., Cong, Qian, Shen, Jinhui, Song, Leina, Mielke, Carlos G. C., Casagrande, Mirna M., Mielke, Olaf H. H. & Grishin, Nick V., 2023, Taxonomic notes on Neotropical Hesperiidae (Lepidoptera), Zootaxa 5271 (1), pp. 91-114 : 105

Eutocus rogan ( Evans, 1955) , Eutocus brasilia (Carneiro, O. Mielke & Casagrande, 2015) , and Eutocus fosca ( Evans, 1955) , new combinations

Phylogenetic analysis of the relatives of Artines Godman, 1901 (type species Thracides aepitus Geyer, 1832 ) reveals that Ginungagapus brasilia Carneiro, O. Mielke & Casagrande, 2015 and Artines fosca Evans, 1955 ( Fig. 6 magenta) are not monophyletic with the type species of their genera and instead are sisters to species of Eutocus Godman, 1901 (type species Eutocus phthia Godman, 1901 , a junior subjective synonym of Apaustus facilis Pl ̂tz, 1884) ( Fig. 6 red). The genitalia morphology and wing pattern of A. fosca already suggested its placement out of Artines , probably in Eutocus ( Medeiros et al. 2019; Medeiros & Dolibaina, per. comm.). Carneiro et al. (2015) recovered G. brasilia and G. rogan ( Evans, 1955) , so far known from a female and males, respectively, nested deeply within Ginungagapus , however, both species share genitalic characters with Eutocus , possibly being erroneously combined to Ginungagapus due to convergent wing pattern and unsampled key taxa ( Eutocus was not included in that study). Therefore, we propose Eutocus brasilia (Carneiro, O. Mielke & Casagrande, 2015) , comb. nov., Eutocus rogan ( Evans, 1955) , comb. nov. (which is phenotypically similar to G. brasilia and has COI barcode, GenBank HM375842, closest to it: 6.8%, 45 bp difference), and Eutocus fosca ( Evans, 1955) , comb. nov. for the time-being, because previous generic placement of these species was clearly incorrect, and it is better to improve it here, while looking for better solutions. However, the nuclear genome tree illustrates that these species ( Fig. 6 magenta) are more differentiated genetically from the core of Eutocus species ( Fig. 6 red), which is also reflected in their morphological differences. In Eutocus sensu stricto palpi are long and thin, but in E. brasilia and E. rogan palpi are short and conical.

The two alternatives to the proposed treatment would be to erect new and (nearly) monotypic genera for them (a splitting solution) or unify the entire clade that is sister to Tarmia Lindsey, 1925 (type species Tarmia monastica Lindsey, 1925 ) ( Fig. 6 ) into a single genus Eutocus and regard its current constituents as subgenera (a lumping solution). Even more, Tarmia and Lattus Grishin, 2022 (type species Eutocus arabupuana Bell, 1932 ) could be included as subgenera in Eutocus (a super-lumping solution) as discussed by Zhang et al. (2022), and the Artines group would consist of two genera: Artines and Eutocus . Each of these alternatives has certain pros and cons. The splitting solution will result in morphologically compact genera, which may be aesthetically pleasing, but the number of genera may increase unnecessarily, adding not particularly distinct monotypic genera to the classification and putting the burden on the name users beyond Hesperiidae specialists. The lumping and super-lumping solutions would group into a single genus all relatives that were a challenge to divide into genera by morphology (as revealed by previous classification mistakes) and may be easier for the users of the names but will create a large and morphologically diverse genus that may be difficult for Hesperiidae specialists to accept. Therefore, a compromise was chosen here: we corrected obvious mistakes and restored monophyly of Artines and Ginungagapus with minimal adjustment of the current classification. However, the best solution will reveal itself after a comprehensive genomic and morphological analysis of the entire group.

Figures

FIGURE 6. Nuclear genome tree of selected Moncina: Psoralis and Artines relatives. See Fig. 4 for notations. Note extreme heterogeneity of evolutionary rates in different lineages, compared to mostly homogenous rates in Fig. 4 tree. Some genera are labeled at their clades. Taxa discussed in the text are shown in different colors: Mucia (blue, with Mucia rusta, comb. nov. labeled in dark blue), Rhomba (cyan, with Rhomba mirnae, comb. nov. labeled in purple), Eprius (green, with Eprius planus, comb. nov. and Eprius obrepta, stat. nov. labeled in orange and olive, respectively), Panca (violet, with Panca fiedleri, comb. nov. labeled in brown), Eutocus (red, with Eutocus brasilia, comb. nov. and Eutocus fosca, comb. nov. labeled in magenta), and Lattus (gray, with Lattus minor, comb. nov. labeled in dark green). Green arrows point from the clade of the genus where a species was placed previously (name in square brackets) to the clade of the species that is being transferred between genera.