Lerodea dysaules Godman, 1900

Zhang, Jing, Cong, Qian, Shen, Jinhui, Song, Leina & Grishin, Nick V., 2023, Butterfly classification and species discovery using genomics, The Taxonomic Report of the International Lepidoptera Survey 11 (3), pp. 1-94 : 85-86

Lerodea dysaules Godman, 1900 is a valid species distinct from Lerodea arabus (W. H. Edwards, 1882)

Originally proposed as a species and treated (with some reservations) as such by Evans (1955), Lerodea dysaules Godman, 1900 (type locality Mexico: Guerrero, Venta de Zopilote) was later synonymized with Lerodea arabus (W. H. Edwards, 1882) (type locality USA: Arizona, Pima Co., Sabino Canyon), because no appreciable differences in genitalia were observed, and wing pattern differences given by Evans (1955) do not hold up in a larger series of specimens (Warren and Mielke 2005). Genomic sequencing of L. arabus from across its range reveals a deep split into two clades ( Fig. 54a, b), unexpected from phenotypic assessment. The two clades are most strongly differentiated genetically with Fst / Gmin statistics 0.70/0.00 and COI barcode difference of 8.2% (54 bp), typical for species from different subgenera and remarkably large for close relatives. Therefore, the two clades represent two distinct species.

The distribution of L. arabus as recorded on iNaturalist (2023) is disjoint, showing a cluster of observations in southeastern Arizona and the major range from Sinaloa – Durango – Coahuila – Nuevo Leon – south Texas southwards ( Fig. 54c). These two disjoint areas in the distribution correspond perfectly to the two distinct species revealed by genomic sequencing (plus specimens from both clades in Sonora, Mexico where no iNaturalist observations existed) ( Fig. 54). Sequenced specimens from near the type locality of L. arabus provide the name for the southeastern Arizona and Sonora (except southernmost) cluster ( Fig. 54 blue). The type locality of Lerodea dysaules falls well within the major range, and the sequenced specimen from Guerrero belongs to the second species along with other specimens across the major range ( Fig. 54 red), providing an available name for this species. Therefore, we reinstate Lerodea dysaules Godman, 1900 , stat. rest. as a species-level taxon. In conclusion, L. arabus is the northwestern species, and L. dysaules is the southeastern species. The two species may meet in southern Sonora, Mexico, and the westernmost record of L. dysaules is from Alamos, Sonora (sequenced as NVG-21056A02) ( Fig. 54 red). Curiously, genomic sequencing suggests that populations from Mexico Baja California Sur are L. arabus ( Fig. 54).

As for the phenotypic assessment, both species are variable in wing patterns, and if identification is possible, it should be done within each pattern form, comparing specimens with similar patterns between species. Generally, ventral overscaling in L. dysaules is more mottled in appearance, especially extending towards the apex but turning towards mid-costa and is bordered by more diffuse small pale spots (if the spots are present at all), these spots are also diffuse and frequently connected into a band even if the central brown patch is absent. In L. arabus , pale ventral overscaling is more uniform, without apparent mottling, the hindwing brown patch (if expressed) is larger and frequently extending towards the apex (at least the area between the patch and the apex is darker than the central submarginal area), giving the patch a more triangular appearance; if postdiscal small pale spots are expressed, their edges are better defined, and the spots are more contrasty on the background, better separated from each other, especially in specimens lacking the patch, and the spot in cell Sc+R1-RS may be particularly prominent. To aid in the identification of these species, we provide COI barcodes of specimens from near the type localities: Lerodea arabus from USA: AZ, Pima Co. [LACM], sample NVG-22094F10, GenBank OR578721: AACTTTATATTTTATTTTTGGAATTTGAGCAGGAATATTAGGAACATCATTAAGTCTTTTAATTCGAACAGAATTAGGTAATCCTGGATCTTTAATTGGAGATGATCAAATTTACAATACT ATTGTGACAGCTCACGCTTTTATTATAATTTTTTTCATGGTAATACCTATTATAATTGGTGGATTTGGTAATTGATTAGTTCCTCTAATATTAGGTGCCCCAGATATAGCTTTCCCACGAA TAAATAATATAAGATTTTGAATACTGCCACCTTCCTTAATATTATTAATTTCAAGTAGAATTGTAGAAAATGGTGCAGGAACAGGTTGAACTGTATATCCTCCTCTTTCTTCTAATATTGC CCACCAAGGAGCCTCAGTTGACTTAGCGATTTTTTCATTACATTTAGCAGGTATTTCATCTATTTTAGGAGCTATTAACTTTATTACTACTATTATTAATATACGAATTAAAAATTTATCA TTTGATCAAATACCTTTATTTGTTTGATCAGTAGGAATTACAGCATTATTATTACTTTTATCTCTACCTGTTTTAGCAGGTGCTATTACCATACTTTTAACTGACCGAAATTTAAACACTT CATTTTTTGATCCTGCTGGAGGAGGAGATCCTATTTTATATCAACATTTATTT

Lerodea dysaules from Mexico: Guerrero [MGCL], sample NVG-21085D09, GenBank OR578722: AACTTTATACTTTATTTTTGGAATTTGAGCAGGAATATTAGGAACATCTTTAAGTCTTTTAATTCGAACAGAATTAGGCAATCCTGGATCTTTAATTGGAGATGATCAAATTTATAATACT ATTGTAACAGCCCATGCCTTTATTATAATTTTTTTTATAGTTATGCCTATTATAATTGGAGGATTTGGTAATTGACTAGTTCCTTTAATATTAGGAGCACCTGACATAGCATTCCCACGAA TAAATAATATAAGATTTTGAATACTACCACCTTCTCTAATATTATTAATCTCAAGTAGAATTGTAGAAAATGGTGCAGGTACAGGTTGAACAGTATATCCCCCTCTTTCATCTAATATTGC ACATCAAGGAGCCTCAGTTGACCTTGCAATTTTTTCTCTTCATTTAGCTGGTATTTCATCCATTTTAGGAGCTATTAATTTTATTACTACTATTATTAACATACGAATTAAAAATTTATCA TTCGATCAAATACCTTTATTTGTCTGATCTGTAGGAATTACAGCATTATTATTACTTTTATCTTTACCTGTCTTAGCAGGAGCTATTACTATACTTTTAACCGATCGAAACCTTAATACTT CATTCTTTGATCCTGCTGGAGGAGGAGATCCTATTTTATATCAACATTTATTT

[No genus] osibius Draudt, 1924 is an unavailable name

The name “osibius” was published by Draudt (1921 –1924) below a specimen illustration on the plate 113B entitled “AGRIAS-ERYNNIS”, row c image 4 from the left, out of 7. No other mention of the name in the genus Agrias , and the species illustrated last (i.e., bottom right: row c image [7], current name Hesperia colorado (Scudder, 1874)) was placed by Draudt (1923b) in the genus Erynnis . The species illustrated next to last (row c, images [5] and [6], current name Turesis complanula ( Herrich-Schäffer, 1869) , misidentified as “ lucasi ”) was placed by Draudt (1923a) in the genus Turesis . Only two genera are mentioned in the title of this plate ( Agrias and Erynnis ), but at least three are illustrated. Therefore, it remains unclear which genus “osibius” was placed in ( Agrias , Turesis [not mentioned on the plate], Erynnis , or some other genus). Mielke (1993) listed “osibius” (as “osybius”) in combination with the genus Turesis Godman, 1901 (type species Hesperia lucas Fabricius, 1793 ) and treated it as a valid species ( Mielke 2004; Mielke 2005). However, the name “osibius” by Draudt is unavailable because it was not proposed “in unambiguous combination with a generic name” as demonstrated above (fails ICZN Code Art. 11.9.3) and therefore cannot be used as a valid name for any species. Finally, we were not able to unambiguously determine the identity of the specimen illustrated by Draudt (1921 –1924) (among others, it might have been Rhinthon Godman, 1900 or Niconiades Hübner, [1821] ), but it does not seem to belong to Turesis due to details in its wing pattern, such as a postdiscal arc of increasing in size pale spots on ventral hindwing, that are not characteristic of Turesis .