Antennae are generally not very feathery, even in males.[2] They lack tympanal organs, but members of the group Choerocampini have hearing organs on their heads.[2] They have a frenulum and retinaculum to join hindwings and forewings.[2] The thorax, abdomen, and wings are densely covered in scales. Some sphingids have a rudimentary proboscis, but most have a very long one,[2] which is used to feed on nectar from flowers. Most are crepuscular or nocturnal, but some species fly during the day.[5] Both males and females are relatively long-lived (10 to 30 days).[5] Prior to flight, most species shiver their flight muscles to warm them up, and, during flight, body temperatures may surpass 40 ÂC (104 ÂF).[5]
Sphingid larvae tend to be specific feeders, rather than generalists.[5] Compared to similarly sized saturniids, sphingids eat soft young leaves of host plants with small toxic molecules, and chew and mash the food into very small bits.[11] Some species can tolerate quite high concentrations of specific toxins. Tobacco hornworms (Manduca sexta) detoxify and rapidly excrete nicotine, as do several other related sphinx moths in the subfamilies Sphinginae and Macroglossinae, but members of the Smerinthinae that were tested are susceptible.[12] The species that are able to tolerate the toxin do not sequester it in their tissues; 98% was excreted. However, other species, such as Hyles euphorbiae and Daphnis nerii, do sequester toxins from their hosts, but do not pass them on to the adult stage.[5]
Most adults feed on nectar, although a few tropical species feed on eye secretions, and the death's-head hawkmoths steal honey from bees.[5] Night-flying sphingids tend to prefer pale flowers with long corolla tubes and a sweet odor, a pollination syndrome known as "sphingophily".[3] Some species are quite general in visitations, while others are very specific, with the plant only being successfully pollinated by a particular species of moth.[3] Orchids frequently have such specific relations with hawk moths and very long corolla tubes. The comet orchid (Angraecum sesquipedale), a rare Malagasy flower with its nectar stored at the bottom of a 30-centimetre-long (12 in) tube, was described in 1822 by Louis-Marie Aubert du Petit-Thouars, and later, Charles Darwin famously predicted there must be some specialized moth to feed from it:
The predicted sphingid was discovered 21 years later and described as a subspecies of the one African species studied by Wallace: Xanthopan morganii praedicta,[16] for which, the subspecific name praedicta ("the predicted one") was given. The Madagascan individuals had a pink, rather than white, breast and abdomen and a black apical line on the forewing, broader than in mainland specimens. Molecular clock models using either rate- or fossil-based calibrations imply that the Madagascan subspecies X. morgani praedicta and the African subspecies morgani diverged 7.4 Â 2.8 Mya (million years ago), which overlaps the divergence of A. sesquipedale from its sister, A. sororium, namely 7.5 Â 5.2 Mya.[17] Since both these orchids have extremely long spurs, longspurs likely existed before that and were exploited by long-tongued moths similar to Xanthopan morganii praedicta. The long geological separation of subspecies morgani and praedicta matches their morphological differences in the color of the breast and abdomen.
When a specimen belongs to a species not yet represented in DNA barcode reference libraries there is disagreement over the effectiveness of using sequence comparisons to assign the query accurately to a higher taxon. Library completeness and the assignment criteria used have been proposed as critical factors affecting the accuracy of such assignments but have not been thoroughly investigated. We explored the accuracy of assignments to genus, tribe and subfamily in the Sphingidae, using the almost complete global DNA barcode reference library (1095 species) available for this family. Costa Rican sphingids (118 species), a well-documented, diverse subset of the family, with each of the tribes and subfamilies represented were used as queries. We simulated libraries with different levels of completeness (10-100% of the available species), and recorded assignments (positive or ambiguous) and their accuracy (true or false) under six criteria.
Our results suggest that when using a strict tree-based criterion for higher taxon assignment with DNA barcodes, the likelihood of assigning a query a genus name incorrectly is very low, if a genus name is provided it has a high likelihood of being accurate, and if no genus match is available the query can nevertheless be assigned to a subfamily with high accuracy regardless of library completeness. DNA barcoding often correctly assigned sphingid moths to higher taxa when species matches were unavailable, suggesting that barcode reference libraries can be useful for higher taxon assignments long before they achieve complete species coverage.
In this study, we test the ability of DNA barcodes to enable accurate higher taxon assignments. Specifically we ask: If species coverage in the DNA barcode library is incomplete, can the barcode from a sphingid species not represented in the library be assigned to the genus it belongs to, or, recognised as being from a sphingid genus missing from the library? Likewise, can the barcode from a sphingid genus not represented in the library be accurately assigned at the tribe and subfamily level? We address these questions using the moth family Sphingidae because a comprehensive global reference barcode library is available (86% of known species [15]) containing relatively stable and well-studied taxa (Figure 1A). This enables us to assemble sub-libraries with a wide range of different species completeness and also provides a robust taxonomic framework against which to judge assignment accuracy. We evaluated assignment accuracy using concordance with the current classification of Sphingidae [16] while recognising that morphologically derived taxonomy represents falsifiable hypotheses. Consequently, we also examined the assignments a posteriori in light of a more recent phylogenetic study of the family [17]. Sphingidae is the target of a global barcoding campaign [15] and shows high success for species-level barcode identifications (Figure 1B).
Taxonomy and DNA barcoding of Sphingidae. A). A tree representing the taxonomy of Sphingidae showing the three subfamilies and eight tribes currently recognised (based on the classification and relationships in Kitching and Cadiou [16]) and used for the purposes of evaluating assignment success in our experiments. There are 202 currently recognised genera. B) DNA barcoding of Sphingidae. The graphs show the divergences in DNA barcodes within sphingid species and between sphingid species within the same genus. This is based on the publicly available sphingid DNA barcodes on BOLD. A "barcode gap" between the intra and interspecific divergences indicates the ease of species assignment based on DNA barcodes and a "best match" type distance criterion.
Using barcode records assembled as part of the global barcoding campaign on Sphingidae [15], we selected one barcode from each species to act as a reference barcode for that taxon. Reference barcodes were available for 1088 of the 1270 described species listed in Kitching and Cadiou [16] and for an additional seven Costa Rican species described or revalidated since 2000 (= 1095 sphingid species). Barcode sequences were selected to maximize length and quality and ranged from 267-658 bp, with 77% being 658 bp and 93% > 600 bp. The sample comprised 200 genera with all the currently recognised tribes and subfamilies (Figure 1A) represented. Three saturniid barcodes (Arsenura drucei, Lonomia electra, Periga cluacina) were also included as this family represents the putative sister family to the Sphingidae [35] taking the full reference library to 1098 barcodes (see additional file 1: Full reference library).
Barcodes from 118 sphingid species collected in Area de Conservacion Guanacaste, northwestern Costa Rica, were used as query barcodes (see additional file 2: Query dataset). DNA was extracted following automated protocols [36] and the DNA barcode amplified and sequenced [37]. These Costa Rican sphingids comprised a well-documented [38, 39], diverse subset of the family, with each of the tribes and subfamilies represented among 29 genera. All the queries were correctly assigned to species when using the full reference library and a "best match" assignment criterion.
This study was funded by grants from NSERC and Genome Canada through the Ontario Genomics Institute to PDNH. Acquisition of the 118 species of query sphingids was supported by USA NSF grants DEB 0072730 and 0515699 to DHJ and the staff of ACG. The sphingid reference library includes contributions by Sam Adams, Philippe Annoyer, Patrick Basquin, Robert Beck, Alex Borisenko, Ron Brechlin, Philippe Darge, Ulf Eitschberger, Yves Estradel, Axel Hausmann, John Janovec, Jean-FranÃois Landry, Tomas Melichar, Scott J. Miller, JoÃl Minet, Kim Mitter, Jacques Pierre, Chris Schmidt, James Tuttle, Thierry Vaglia, Evgeny Zakharov; many taxa were only accessible within the collections of the following institutions: the Australian National Insect Collection (Canberra), The Smithsonian Institution (Washington), the MusÃum National d'Histoire Naturelle (Paris), the Canadian National Collections of Insects and Arachnids (Ottawa), the Bavarian State Collection of Zoology (Munich). JJW would like to thank the staff and visitors to BIO and the Hanner lab for helpful discussions and encouragement over the course of this study. Massimiliano Virgilio and an anonymous reviewer significantly improved the manuscript through their comments.
38c6e68cf9