Corresponding author: Nick V. Grishin (
Academic editor: Carlos Peña
Shiraiwa K, Cong Q, Grishin NV (2014) A new
Swallowtails (
A multifaceted pattern of speciation resulted in 7 or 8 species of tiger swallowtails (Pterourus glaucus group) in North America (
The Giant Swallowtails (Heraclides cresphontes group sensu Fig.
In 2005, KS collected several specimens of
Specimens used in this study were collected in the field under the permit #08-02Rev from Texas Parks and Wildlife Department to NVG, or examined in the following collections: Texas A&M University Collection, College Station, TX, USA (TAMU); University of Texas at Austin Insect Collection, Austin, TX, USA (TMMC); National Museum of Natural History, Smithsonian Institution, Washington, DC, USA (USNM); Colorado State University Collection, Fort Collins, CO, USA (CSUC); The Field Museum of Natural History, Chicago, IL, USA (FMNH); American Museum of Natural History, New York, NY, USA (AMNH); McGuire Center for
Legs (cut with scissors into tiny pieces in lysis buffer), crumbs and pieces of muscle tissue from the thorax of dissected specimens (plucked from the abdomen attachment site), or a distal part of an abdomen (dropped into lysis buffer, and after overnight incubation at 56°C transferred into 10% KOH for genitalia dissection) were used to extract genomic DNA with Macherey-Nagel (MN) NucleoSpin® tissue kit following the manufacturers protocol. The lysis buffer volume was scaled down to 70 μl for legs and volumes of subsequent reagents were proportionally reduced. Genomic DNA was eluted in a total volume of 40-100 μl MN BE buffer (concentration of DNA as measured by Promega QuantiFluor® dsDNA System was from near 0 to 20 ng/μl, mostly around 1 ng/μl, depending on specimen age and storage conditions) and was stored at -20°C.
PCR was performed using Invitrogen AmpliTaq Gold 360 master mix in a 20 μl total volume containing less than 1 ng of temfig DNA (usually 0.5–1 μl of DNA extract) and 0.5 μM of each primer. For legs from freshly collected specimens or those preserved in alcohol, the following primers were used to obtain the complete barcode: LepF: 5’-TGTAAAACGACGGCCAGTATTCAACCAATCATAAAGATATTGG-3’ and LepR: 5’-CAGGAAACAGCTATGACCTAAACTTCTGGATGTCCAAAAAATCA-3’. For older specimens (up to 1960) the following pairs of primers were used: swtl-COIF (forward, 5’-TTATTCAACAAATCATAAAGATATCGGA-3’) – swtl-mCOIR (reverse, 5’-GTTCCKGCYCCATTTTCTAC-3’), or sCOIF (forward, 5’-ATTCAACCAATCATAAAGATATTGG-3’) – smCOIR (reverse, 5’-CCTGTTCCAGCTCCATTTTC-3’), and swtl-mCOIF (forward, 5’-GACTTTTACCCCCTTCTCTAACTC-3’) – swtl-COIR (reverse, 5’-AAAATATAAACTTCAGGATGTCCAAA-3’), to amplify the barcode in two overlapping segments.
The barcodes of even older specimens (1900–1960) were amplified in four overlapping segments with the following four pairs of
For some old specimens (e.g., 1870–1960), amplification of longer DNA segments failed. To obtain their sequences for identification, we developed
The PCR reaction was cleaned up by enzymatic digestion for the whole barcode amplifications, ID tag amplification, and sequences amplified in more than 2 segments, with 4 μl Shrimp Alkaline Phosphatase (20 U/μl) and 1 ul Exonuclease I (1 U/μl) from New England Biolabs. For sequences obtained in two segments, due to the frequent presence of primer dimers and other short non-specific PCR products, Agencourt Ampure XP beads or Invitrogen E-Gel® EX Agarose Gels (followed by Zymo gel DNA recovery kit) were used to select the DNA products of expected length. Sequences were obtained using the M13 primers (for amplification from LepF and LepR primers): 5’-TGTAAAACGACGGCCAGT-3’ or 5’-CAGGAAACAGCTATGACC-3’ or with primers used in PCR. Sanger sequencing was performed with Applied Biosystems Big Dye Terminator 3.1 kit on ABI capillary instrument in the DNA Sequencing Core Facility of the McDermott Center at UT Southwestern. The resulting sequence traces were proofread in FinchTV <
As a result, we obtained complete or partial DNA barcode sequences from 395
Additional DNA sequences for analysis were downloaded from GenBank <
Comparison of
Specimens from the eastern North America, from Canada to Florida and central Texas, USA belong to the eastern group. Specimens from other parts of the range from central Texas to California, USA and southwards to Panama belong to the southwestern group. In central Texas, both groups are present. Genitalic differences between groups correlate with the differences in facies (Fig.
Correlation between genitalic and facies differences and geographic distribution suggests that we are dealing with two distinct evolutionary lineages diversified sufficiently to be treated as two taxonomic units. However, these units are mostly allopatric and overlap over narrow range in central Texas. Moreover, in the range of overlap, we see specimens with intermediate characters. Therefore, it was not clear whether to regard these taxonomic units as subspecies, or suggest that the divergence between them is sufficient for biological species.
To address this question, we determined COI mitochondrial DNA barcode sequences for 249
To put this number in perspective and compare it with divergence in other
However, it is conceivable that one of these two barcodes might have evolved not within this species, but could have been introgressed from a different, albeit closely related, species. If that were the case, it is possible that the two
Next, in a quest for the names to apply to the eastern and southwestern species, we analyze names proposed for
Cramer illustrated two specimens, a female in dorsal and ventral aspects (
Since no Giant Swallowtails are known from Guadeloupe (
We took the following steps to search for the
Similar challenges with Cramer type localities and syntypes were encountered by other researchers, so negative results were not surprising. For instance,
As a final resort, we consulted the original drawings by Gerrit Wartenaar Lambertz made for Cramer and used as prototypes for the published engravings. Presently, the drawings are in the library of the Natural History Museum, London (BMNH). One of the drawings for the Volume 2, “A” on the fig #134, reproduced here as Fig.
The above lectotype designation resolves the problem between
The two lectotypes designated above unambiguously distinguish
This neotype designation satisfies all seven provisions of the ICZN code (Art. 75.3.1.–7.) as follows. The neotype is designated to clarify the attribution of the name
Eight names have been considered synonyms of
We consider
The name
The remaining six names are infrasubspecific according to the Articles 45.5. & 45.6. of the ICZN Code, in agreement with
However, the second one,
Male (n=95, Figs
Female (n=28, Figs
Neck pattern, male genitalia, and morphometrics.
Female genitalia.
Facies differences between
Variation in male genitalia. Left lateral view of genital ring (uncus, brachium, dorsolateral sclerite, tegumen, vinculum and saccus) is shown, valvae, aedeagus and last tergum with pseuduncus are removed.
Genbank accession
AACATTATATTTTATTTTTGGAATTTGAGCAAGAATACTAGGAACTTCTCTTAGTTTACTAATTCGTACTGAATTAGGCA CCCCCGGCTCATTAATTGGAGATGATCAAATTTATAATACTATTGTTACAGCTCATGCTTTTATTATAATTTTTTTTATAG TTATACCTATTATAATTGGAGGATTTGGAAATTGATTAATTCCATTAATATTAGGAGCCCCTGATATAGCTTTTCCTCGTA TAAATAATATAAGATTTTGACTTTTACCCCCTTCTCTAACTCTCCTAATTTCAAGAATAATTGTAGAAAATGGGGCAGGAA CTGGATGAACTGTTTACCCTCCTCTTTCCTCTAATATTGCCCATGGAAGAAGATCAGTAGATTTAGTTATCTTTTCTTTAC ATTTAGCTGGTATTTCCTCAATTCTTGGAGCAATTAATTTTATTACTACAATTATTAATATACGAATTAATAGAATATCTT TTGATCAAATACCTTTATTTGTTTGAGCCGTAGGAATTACAGCTTTATTATTACTTTTATCTTTACCTGTTTTAGCAGGAG CTATTACTATACTTTTAACTGATCGAAATTTAAATACTTCATTTTTTGACCCTGCTGGAGGAGGAGATCCAATTTTATACC AACATTTATTT
In addition to the holotype, barcodes and ID tags were obtained for 110 paratypes: 93 full-length barcodes (658 to 664 bp), 3 partial barcodes (443 bp) and 14 ID tags (64 bp), see Suppl. material
35 specimens from Texas (mostly central) possessed DNA barcodes of
USA: Texas: Duval Co., Benavides, CR306, 1.8 mi west of SH339, latitude 27°36'27", longitude −98°26'29.4", elevation 124 m. This locality is at the sharp bend of the County Road 306, where several shrubs of Colima (
The species is named to honor the wife of the first author. Pronounced as ’roo(as in rue)-mee(as in meek)-koh(as in cod). The stress is on the first syllable. The name is a noun in apposition.
In the 1960s,
Female genitalia are very variable in both species (Fig.
(Figs
Larva eats egg shell upon hatching (Fig.
Pupa, 26–36mm in length (Fig.
In south Texas (e.g., near the type locality in Duval County),
COI DNA barcode distances within
COI DNA barcode trees. Trees of representative sequences of
COI DNA-barcodes. Relationships between
Localities of
Life history: eggs and 1st instar caterpillars.
Life history: 2nd and 3rd instar caterpillars.
Life history: 4th and 5th instar caterpillars.
Life history: 5th instar caterpillars and prepupae.
Life history: pupae.
Foodplants most commonly used by
Speculations about origins of the
Species in the genus
Analysis of the thoas group revealed that the Cuban taxon is very distant from the rest, showing more than 5% difference in COI barcodes, a difference much larger than the divergence within the cresphontes group falling within 3.5% (Figs
In agreement with
The machaonides group consists of two species:
To summarize the nomenclature of the H. cresphontes group, we provide a synonymic list of its species. Name combination from the original description is used for each synonym (= subjective synonyms; =† objective synonyms; =‡ unavailable names) and for species is given after “|”. Format of the data: reference to the description | category of a primary type (
Genus
Verz. bekannt. Schmett. (2): 83-84. Type species:
Subgenus
=†
Zool. Illustr. (2)3(26): pl. 121, unnumbered text. Type species:
cresphontes species group
Uitl. Kapellen 2(14): 106-107, pl. 165 f. A ♀ D&V, pl. 166 f. B ♂ D (LT) |
=†
Verz. bek. Schmett. (2): 83 (replacement name for
=‡
Entomol. Z. 22(23): 92 |
=‡
Z. wiss. InsektBiol. 7(5/6): 159 |
=‡
Soc. Ent. 33(12): 47; referred to Seitz (1907) Gross-Schmett. Erde 5: pl. 7 f. a [2] |
=
Bull. Brooklyn Ent. Soc. 14(1): 3, f. 2 ♂ D (HT) |
=‡
Can. Ent. 65(8): 171 |
=
Proc. Penn. Acad. Sci. 19: 38-39 |
ZooKeys 468: 85–135 |
=‡
An. Inst. Biol. Univ. Méx. 11(2): 633-634 |
Novit. Zool. 13(3): 561-562, no. 67 |
Novit. Zool. 13(3): 556, no. 66a |
The Giant Swallowtail
We follow
First, divergence within each of the three genera is already very significant, reaching 10% sequence difference in the COI DNA barcode. In recent work on
Second, each of the three
Third, the most important utility about using the three genera instead of
In contrast to genus, species is a more objective biological category. A number of species concepts has been proposed (
While hybridization experiments followed by fitness measurements in hybrids and backcrosses may be decisive in delineating species boundaries, phenotypic differences and genetic divergence are used as more practical criteria. If two populations of the same species have spent significant time in isolation, mutations randomly accumulating in them are likely to cause incompatibilities upon interbreeding, leading to speciation. Some mutations may also cause phenotypic effects, allowing researchers to recognize species by morphological characters. Gene regions rich in neural mutations, such as the COI barcode, are used as yardsticks to estimate divergence between populations. Larger divergence between populations indicates higher chance of speciation. No universal threshold for divergence to mean speciation is possible. Recently formed species may have identical DNA barcodes. High barcode variability within population may lead to conspecific individuals with large barcode differences. To derive sensible conclusions, comparison of barcode variation within and between populations is necessary. Since similar evolutionary mechanisms frequently occur in related organisms, evaluation of barcode variability across the genus is desirable. Finally, correlation between DNA differences and morphological differences is most effective for delineation of species.
In many animals, allopatric populations of the same species characterized by measurable morphological differences, such as those in shapes and colors, are frequently named as subspecies. Typically, subspecies diverged in morphology very recently. Therefore, differences between their DNA barcodes are small compared to those between species. Some of these subspecies are on a path to speciation. Given longer time, and thus more mutations accumulating in the DNA barcode, reproductive incompatibility between these populations will arise. Random extinctions of various populations prune phylogenetic tree and lead to formation of discrete clades that form various clusters. Comparative analysis of these clades and clusters suggests taxonomic hypotheses.
We applied these ideas to selected Neotropical representatives of the tribe
DNA barcodes of
Distribution ranges of
Ultimately, there is no proof, but a hypothesis—or prediction—that we think has a better chance of standing the test of time. Given all the information we assembled, our bet is on the species (and not subspecies) status of
As a summary, we observe three levels of differentiation at and near the species level. First, there are clusters of populations with small genetic differences between them (mostly within 1% in COI barcodes, sometimes no difference at all), but certain geographic differences in wing patters. These populations could be defined as subspecies. Next, there are groups with larger genetic differences (typically above 2% in COI barcodes, but could be less), frequently characterized by measurable differences in genitalia. These groups could be called species. Finally, several mostly allopatric species characterized by closely related phenotypes form very distinct genotypic groups (usually more than 5% in COI barcodes) could be termed a superspecies. All these levels are seen in
The 3% difference in DNA barcodes of
In our medium-scale barcoding study we didn’t see any significant invasion of
“Giant Swallowtail” is one of the species used in butterfly release ceremonies across the US. USDA lists
Qian Cong is a Howard Hughes Medical Institute International Student Research fellow. We thank Texas Parks and Wildlife Department (Natural Resources Program Director David H. Riskind) for the permit #08-02Rev making research based on material collected in Texas State Parks possible. We are grateful to Edward G. Riley (Texas A & M University insect collection, College Station, TX), Brian Harris, Robert K. Robbins, and John M. Burns (National Museum of Natural History, Smithsonian Institution, Washington DC), James R. Reddell (University of Texas at Austin Insect Collection, Austin, TX), Paul A. Opler (Colorado State University Collection, Fort Collins, CO), Rebekah Shuman Baquiran (The Field Museum of Natural History, Chicago, IL), David Grimaldi and Lesley Thayer (American Museum of Natural History, New York, NY), Andrew D. Warren (McGuire Center for
Supplementary Table 1
specimen data.
Data for specimens with DNA sequences determined in this study.