A new species of Lactura is described from Texas: Lactura rubritegula sp. n. Identity of the new species can be reliably determined by both larval and adult characters, CO1 haplotypes, and its late-spring period of flight activity. Male genitalic features overlap with those of L. basistriga (Barnes & McDunnough, 1913), whereas female structures differ markedly between the pair. The new Sideroxylon-feeding species, rare in collections, is found principally in limestone areas in the vicinity San Antonio, Texas, westward through the southern Hill Country. We illustrate the adult and larval stages and male and female genitalia, review available DNA barcode data that support the recognition of the new Lactura, and briefly characterize its life history.

Sapotaceae , gum bully, Sideroxylon , DNA barcodes, CO1, Sideroxylon celastrinum , tropical burnet moth

The Lacturidae (Tropical Burnet Moths) are a family of zygaenoid lepidopterans that inhabit tropical and subtropical regions of the world. Lacturid adults of North and Central America generally have white to gray forewings marked with antemedial and postmedial rows of red or black spots, and uniformly pink-red hindwings. Species delimitation in the New World has been hampered because this general description applies equally well to nearly all of the ca. 17 North and Central American species (Heppner and Duckworth 1983, Heppner 1984). And to further complicate taxonomic efforts, species such as L. subfervens (Walker, 1854) are exceedingly variable in color pattern with forms that overlap those of other members of the genus, including the new species described below.

The most recent checklist of Lepidoptera for the United States and Canada recognizes six species of Lactura (Heppner and Duckworth 1983). However, much confusion surrounds the taxonomic identity and validity of some of these species, and especially, the way these names have been applied in institutional collections, literature, genetic databases, and on-line identification resources. The chronic misidentification and misapplication of names within this genus, and the inadequate knowledge of the associated life histories, have made it heretofore difficult to recognize the new North American species described in this work, despite it having salient external features that allow identification of both its adult and larval stages. CO1 barcode data from authoritatively identified adults and larval collections helped us to unmask the existing tangle of misidentifications in collections and internet resources and recognize the new taxon.

Here we describe a new species of Lactura from the limestone areas and riparian corridors of south-central Texas’s Hill Country. The new taxon is believed to be a specialist on Sideroxylon lanuginosum (Sapotaceae), a host it shares with both L. pupula (Hübner, [1831]) and L. subfervens at the type locality. We describe and illustrate the larval and adult stages of the new species, illustrate the male and female genitalia, and provide a brief account of its biology and distribution.

Adults were obtained by light trapping with UV and mercury-vapor lights. Larvae (of all US species) were collected from Sideroxylon (Sapotaceae). Preserved larval specimens were reared from ova deposited by gravid females collected at light traps. Most of the type specimens were collected by Delmar Cain from the type locality in Boerne, Kendall Co., Texas. Additional paratypes were collected by Ed Knudson (Uvalde Co., Harris Co., and Kerr Co.) and Ann Hendrickson (Edwards Co.). Ova acquired from the type locality in Kendall Co., were sent to Berry Nall in Falcon Heights, Texas (Rio Grande Valley), where they were reared on Sideroxylon celastrinum. The adult description of L. rubritegula is based on 20 pinned specimens; the larval description is based on two ex ova cohorts and one wild larva from the type locality. Forty-two genitalic slides of Lactura were examined: 29 on loan from the National Museum of Natural History (USNM) and 13 prepared by Tony Thomas for this study. Eight genitalic slides of L. basistriga and three of L. rubritegula were examined. Over the course of this study we examined the Nearctic Lactura holdings of the National Museum of Natural History (USNM) (Washington DC) (including primary types); Mississippi Entomological Museum (MEM) (Mississippi State, MS); Texas A&M University (TAMUIC) (College Station, TX), University of Connecticut (UCMS) (Storrs, CT), and the personal collections of Vernon A. Brou, Jr. (Albita Springs, LA); Edward C. Knudson (ECK) (Houston, TX); and James R. McDermott (JRM) (College Station, TX). CO1 barcodes for North American Lactura were compiled from holdings in the following institutions and personal collections: Canadian National Collection of Insects, Arachnids, and Nematodes, ECK, JRM, MEM, TAMUIC, UCMS, and USNM. We had access to 111 North American Lactura COI barcode submissions: for each of these we examined the associated voucher image. A neighbor-joining tree, using the default Kimura 2-P model, was generated by the Barcodes of Life Project (BOLD) (www.boldsystems.org) (Ratnasingham and Hebert 2007). DNA extraction, PCR amplification, and CO1 barcode sequencing were performed at the Canadian Centre for DNA Barcoding (Centre for Biodiversity Genomics – University of Guelph) using their standard protocol (Wilson 2012). Barcode sequences for the holotype and paratypes of L. rubritegula were deposited in GenBank. SimpleMappr (www.simplemappr.net) was used to generate the geographic distribution point map (Shorthouse 2010). Type material has been deposited in the USNM, TAMUIC, and UCMS.

So far as known, Lactura rubritegula has a curiously restricted range relative to that of its hostplant, Sideroxylon lanuginosum (gum bully)—the most widely distributed Sideroxylon in the United States. Gum bully’s range extends from Arizona to southern Kansas, Missouri, and Illinois, eastward to Georgia and Florida, and southward into Mexico. The moth is known principally from the southern Hill Country in the vicinity of San Antonio, westward and southward. Ed Knudson took a single specimen in west Houston (Spring Valley), which would suggest that the moth has simply been overlooked over parts of its range. We did not find additional examples in his collection, nor did we find any in the USNM collection, which includes Andre Blanchard’s extensive Texas collection.

We had access to barcode data for 111 North American Lactura specimens representing approximately 15 species-level taxa. The species cluster that includes L. rubritegula is shown in Fig. 10, henceforth we refer to this group as the basitriga complex. Barcode data for the group includes group includes basistriga (n=3; southern Texas), near basistriga (n=3; Oaxaca, Mexico), and the new species (n=2; Boerne, Texas). L. rubritegula shows an (uncorrected) divergence of 3.6% to its nearest neighbor in the basistriga group (from Oaxaca) and a 4.8% divergence from its nearest Texas neighbor, nominate basistriga.

Figure 10. 

North American COI barcode tree for Lactura basistriga complex available as of 6 October 2017. Included for each taxon is the BOLD process ID, region, and sequence length.

Taxonomically, Lactura has proven itself to be a challenging genus to some of North America’s top systematic microlepidopterists, past and present, including August Busck, Harrison Dyar, and Ronald Hodges. Larval phenotypes and COI barcodes guided our understanding of this group by unmasking a tangle of misidentifications and segregating otherwise cryptic lineages. Based on our assessment of the North American fauna informed by larvae, life history data, and CO1 barcodes, the identifications in institutional collections, BOLD, and GenBank are riddled with misidentifications—e.g., 46% of the North American specimens of Lactura in BOLD are misidentified. Much of the confusion traces to concepts of L. subfervens, a highly variable taxon whose phenotypes overlap with those of L. basistriga, L. rubritegula, L. psammitis (Zeller, 1872) (TAM unpubl. data). Of the 52 specimens of L. subfervens and in BOLD, 65% of these appear to be misidentified.

We have attempted to rear two ex ova clutches of Lactura rubritegula from the type locality. Females captured on 25 April 2015 and 13 June 2017 were reared successfully by Berry Nall on Sideroxylon celastrinum in Falcon Heights, Texas. Pupae from the first cohort yielded an adult the following spring that is part of the paratype series. However, a wild larva, presumably an antepenultimate instar, found feeding on S. lanuginosum at the type locality in Boerne, Texas, on 13 June 2017 by Delmar Cain, straggled and died even when sleeved on the same branch from which it came, perhaps because the Sideroxylon lanuginosum foliage was too aged by mid-June to be acceptable. Whereas many Texas moths from arid and semi-arid habitats fly in almost any month when there has been appreciable rainfall, we only have late spring captures for L. rubritegula.

In contrast to the phenotypically uniform and confusing adults, the larvae of North American Lactura can be reliably identified to species. Larval features even unambiguously differentiate regional segregates within the two most widespread North American species, L. pupula and L. subfervens. Likewise barcodes disambiguate the North America species in that they align with our larval, life history, and biogeographic data. The confused nature of barcode data in BOLD appears to be entirely due to the misapplication of names by those making the initial submissions. We are currently collecting data for eight nuclear genes and preparing additional genitalic dissections for the North American fauna, and will await the outcome of these studies before making formal synonymies, producing adult and larval keys, and emending current BOLD determinations for the Lactura that occur north of Mexico.

Perhaps not surprisingly, for nearly every North American Lactura species there appears to be a closely related Mexican sister taxon, and in some case a species cluster in BOLD. There is much taxonomic work to be done, especially southward into Central and South America. We suspect the Nearctic fauna will continue to yield cryptic species and otherwise new species for decades, and that molecular data will play a critical role in getting this genus of look-a-likes properly partitioned into valid evolutionary units.

We would like to thank the National Museum of Natural History, Smithsonian Institution for their generous contribution of genetic data; Dr. Scott E. Miller collected much of the DNA barcode data that formed the scaffolding that allowed for the genetic confirmation of our new taxon. Specimen loans were provided by Edward C. Knudson (Texas Lepidoptera Survey), James R. McDermott, Richard L. Brown (MEM), and Scott E. Miller (USNM). Anne Hendrickson sent photographs and collected specimens of Lactura from her ranch in Camp Wood, Texas. A special thanks to Delmar Cain who helped with many aspects of this effort: hosting DLW, collecting many of the specimens in the type series, providing phenological data for the three species of Lactura on his property, egging females, and more. We are much indebted to Berry Nall for his exceptional rearing and photographic assistance—both of the larval images in this work are his. We also thank Tony Thomas for preparing several genitalic dissections, and Marc Epstein and Jean-Francois Landry for their helpful suggestions that improved an earlier version of this manuscript. Jim Paz and Seth Patterson assisted with visits to the Sabal Palm Sanctuary in Brownsville Texas where comparative material of L. basistriga was secured. DLW is supported by USFS Co-op Agreement 14-CA-11420004-138 and an award from the Richard P. Garmany Fund (Hartford Foundation).