PROCEEDINGS of the ENTOMOLOGICAL SOCIETY of WASHINGTON

OF MANY SIMILAR SPECIES IN THE NEOTROPICAL GENUS PORPHYROGENES (LEPIDOPTERA: HESPERIIDAE), A NEW ONE, REPEATEDLY REARED IN COSTA RICA, IS RELATIVELY DISTINCT

JOHN M. BURNS,DANIEL H. JANZEN, AND WINNIE HALLWACHS PROC. ENTOMOL. SOC. WASH. 112(1), 2010, pp. 32–42

OF MANY SIMILAR SPECIES IN THE NEOTROPICAL GENUS PORPHYROGENES (LEPIDOPTERA: HESPERIIDAE), A NEW ONE, REPEATEDLY REARED IN COSTA RICA, IS RELATIVELY DISTINCT

JOHN M. BURNS,DANIEL H. JANZEN, AND WINNIE HALLWACHS

(JMB) Department of Entomology, National Museum of Natural History, Smithsonian Institution, P.O. Box 37012, MRC 127, room E-515, Washington, DC 20013-7012 USA (e-mail: [email protected]); (DHJ, WH) Department of Biology, University of Pennsylvania, Philadelphia, PA 19104 USA (e-mail: DHJ djanzen@ sas.upenn.edu, WH [email protected])

Abstract.—Associating sexes in many species of the Neotropical hesperiid genus Porphyrogenes has been difficult, erroneous, or impossible, owing to their sexual dimorphism. Despite the extreme sexual dimorphism of Porphyrogenes peterwegei Burns, new species, full description of this rainforest skipper from Area de Conservacio´n Guanacaste (ACG), northwestern Costa Rica, is no problem. The available sample amounts to nearly 100 reared adults, almost evenly divided between the sexes. Of these, 67 have been DNA barcoded and found to vary minimally in their COI haplotypes, with one haplotype predominating. Caterpillars and pupae consistently go through a distinctive set of developmental changes in their color patterns. Foodplant choice is conservative: of 227 immatures found in the wild, 94% were eating woody vines of the genus Machaerium, especially M. seemannii (Fabaceae), whereas 6% were feeding on the quite unrelated species Dichapetalum morenoi (Dichapetalaceae)—a peculiar pattern of larval foodplant selection seen in another skipper and two butterfly species in ACG. Caterpillars of P. peterwegei are occasionally attacked by a host-specific parasitoid (an undescribed tachinid of the genus Siphosturmia), which has not otherwise been encountered in 20,642 tachinid attacks on the caterpillars of .3,000 species of Lepidoptera in ACG. Morphologically, P. peterwegei differs significantly from its many congeners in both facies and genitalia. Key Words: Genitalia (male and female), sexual dimorphism, caterpillars, pupae, foodplants (Machaerium), DNA barcodes, tachinid parasitoid DOI: 10.4289.0013-8797.112.1.32

Speciation is kind to taxonomists ally valuable to taxonomists when, when, within a bevy of similar related among similar species, sexual dimor- species, an undescribed one differs phism is so great that associating the visibly from the rest. Multiple rearings sexes has often been incorrect or and DNA barcodes become exception- impossible. A recent detailed treatment of Porphyrogenes (Austin and Mielke * Accepted by Robert R. Kula 2008) recognizes 26 species (including VOLUME 112, NUMBER 1 33 the one named here), as well as three 2009) and have been deposited in the previously named but unassociated National Museum of Natural History females and five unnamed unassociated (USNM), Smithsonian Institution, females (which may or may not repre- Washington, District of Columbia, sent additional species). Owing, at least USA (see Type material). (In addition, in part, to their late-afternoon, crepus- seven reared adults preserved in EtOH cular, and perhaps nocturnal behavior, have been deposited in the Monell Cryo adults of these sizeable Neotropical Collection for Molecular and Microbial skippers have always been rare in Research at the American Museum of collections. Indeed, six of the nine Natural History, New York, New York, new species of Porphyrogenes in Austin USA; three have been deposited in the and Mielke (2008) are described from a Lepidoptera Tree of Life collection at single specimen, which is always male. the University of Maryland, College Porphyrogenes is known from most Park, USA.) The massive rearing pro- of Central and South America. Specific gram in Costa Rica is described in records extend from Belize (J. A. Shuey Burns and Janzen (2001), Janzen and pers. comm.), Honduras, Nicaragua, Hallwachs (2009), and Janzen et al. Costa Rica, and Panama, to Colombia, (2009). Four wild-caught adults of Ecuador, Peru, Bolivia, Paraguay, and Porphyrogenes have been borrowed northern Argentina, and all mainland from the Instituto Nacional de Biodi- countries to the east, plus Trinidad versidad (INBio), Santo Domingo de (Austin and Mielke 2008). About Heredia, Costa Rica. three-quarters of the species occur in In the course of rearing enormous South America and one-third in Central numbers of Lepidoptera and their par- America(thesefiguresincludetwo asitoids in ACG, new species have been species attributed to both regions and discovered at a rate far greater than that exclude one species from no one knows atwhichtheyhavebeenformally where). Although it is not surprising named and described. Temporary des- that 16 of the 26 species have been ignations are needed for referring to found in Brazil (Austin and Mielke undescribed species. When such species 2008), there are, at the very least, 12 are thought to be very closely related to species (two undescribed) in just the one that is already named, that name is limited (and ecologically less diverse) used, but it is followed by a code with area of French Guiana (B. Hermier in letters that denote the worker who litt.). What with the similarity of many detects two or more species instead of species, their excessive sexual dimor- just one and with numbers (01, 02, 03, phism, and a lack of material, the etc.) that arbitrarily distinguish these geographic distribution of many de- species—for example, Genus species scribed species is still nebulous, and XYZ01. When an undescribed species the number of undescribed species is can be placed to genus but no further, conjectural. its designation is simply Genus XYZ01. Forewing length of adults was mea- MATERIALS AND METHODS sured from base of costa to apex with a Adult specimens used in this study vernier caliper calibrated to tenths of a have been reared from caterpillars millimeter. After each measurement collected during an ongoing bioinven- was recorded, the distance between tory of Lepidoptera in Area de Con- caliper prongs was significantly reduced servacio´n Guanacaste (ACG), north- before the next specimen was mea- western Costa Rica (Janzen et al. sured. The measuring sequence was a 34 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON

Figs. 1–4. Reared paratype male, voucher code 04-SRNP-4400 (above) and holotype female (below) of Porphyrogenes peterwegei from ACG, Costa Rica, in dorsal (left) and ventral (right) view; pins and pinholes artificially removed; scale bar ¼ 1cm. random mix of males and females, and (except for two species) are spotless. all but two specimens were measured The female of P. peterwegei is unique in on the same day. Nudum segments of having only one spot—an outsized one the antennal club were counted with the in the center of the forewing (in cell aid of a stereomicroscope at a magni- CuA1-CuA2) (Figs. 3–4). Although fication of 403. After treatment with females of a few other species have a 10% KOH, genitalia were dissected similar large central spot, they also have with the aid of a stereomicroscope and a small distal spot (in cell M3-CuA1) thereafter were kept free in glycerin. (see figures in Austin and Mielke 2008). For details on the handling, comparison, ThemaleofP. peterwegei shares and storage of unmounted genitalia, see whitish wing fringes (Figs. 1–2) with Burns (1997). For information on meth- just one other species, P. stupa Evans, ods used in DNA barcoding, see which is known from a single dataless Hajibabaei et al. (2006). male whose ground color is virtually black with a purplish sheen, instead of RESULTS Porphyrogenes peterwegei Burns, dark brown as in P. peterwegei, and new species whose hind wing is broad, rounded, and lacking a tornal lobe, while that of P. Diagnosis.—Females of Porphyro- peterwegei is narrow with a prominent genes always have one or more fore- tornal lobe. wing spots (joined, in some species, to Description.—Size: Females average form a transverse band), whereas males larger than males. Forewing length in VOLUME 112, NUMBER 1 35 mm: Male: Mean ¼ 22.93, SD ¼ 0.982, (though variably) paler than ground n ¼ 45, Range 21.0–24.7. Female: Mean color, especially above tornal lobe. ¼ 25.77, SD ¼ 1.378, n ¼ 51, Range Hind wing postmedian band of darker 23.0–28.5. NB: Owing to the unnatural spots more evident than in male, and inferior conditions of rearing, especially ventrally. More or less con- reared adults average smaller than wild spicuous pale gray, submarginal to ones; the degree of dwarfing varies, marginal strip (interrupted by darker especially with respect to the stage of veins) distal to ventral hind wing the caterpillar when found. The three spotband and contrasting with it. Sim- ilar marginal strip of still paler gray on males and four females obtained from ventral forewing. All of ventral fore- wild-caught pupae of this species are wing space between vein 2A and inner near or at the upper end of the size margin extra pale. Legs dull orange. range, given above, for each sex. Secondary sex characters: Male: Number of segments in nudum of Forewing with costal fold. Forewing antennal club: The sexes do not differ. vein 2A bowed anteriad and, along Mean ¼ 29.96, SD ¼ 1.259, n ¼ 76, highest reaches of its arc, swollen and Range 27–34. The number of nudum ventrally devoid of scales; vein 2A, segments is not correlated with length from swollen stretch to base of wing, of the forewing. bisecting ventral speculum comprising Wing shape (Figs. 1–4): Sexual variously modified scales. On dorsal dimorphism pronounced. Wings of hind wing, two conspicuous, adjacent, male narrow and apically pointed; those distally light-brown and proximally tan of female broad and apically rounded. hair tufts arising near base of cell In both sexes, hind wing tornus con- ScþR1-Rs and near base of discal cell, spicuously lobed. and associated with large speculum Facies: Sexual dimorphism extreme. extending from discal cell to costa. Male (Figs. 1–2): Body and wings dark Both sexes: Ventral hind wing with brown both dorsally and ventrally, but deep, pale-scaled groove (most promi- dorsally conspicuously overscaled with nent proximally) in cell 2A-3A. dull orange on body and on about Genitalia: Male (Figs. 5–6): In proximal half of forewing and proximal lateral view, body of valva (i.e., anterior two-thirds of hind wing. Fringe along two-thirds) low; distal one-third of outer margin of both wings narrowly valva (¼ harpe) robust (i.e., not notably edged off-white. Hind wing with faintly narrower than adjacent part of body) suggested postmedian band of small, and abruptly bent dorsad, nearly or darker brown spots (more detectable quite forming right angle with body, ventrally than dorsally). Legs dull then terminating in large, sharp, ante- orange. Female (Figs. 3–4): Body and rodorsal point and bearing some preter- wings grayish brown dorsally and minal coarse teeth. Cornuti long, essen- especially ventrally, with one huge, tially straight, delicate spines elliptical, hyaline spot in middle of (numbering 31, 31, 32, 32, 34 [see forewing in cell CuA1-CuA2. Dorsal Fig. 6], and 36 in six individuals); in overscaling faint greenish gray; pur- everted vesica, rather evenly and widely plish sheen distal to this overscaling, spaced, except for some relatively primarily on forewing (visible when clustered proximal ones. Tegumen viewed at certain angles). Edge of stocky; its delicate posterior prongs fringe along outer margin not distinc- uneven in diameter, about as in Figs. tively off-white (as in male), but often 5–6 (or, in extreme case, so uneven as 36 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON

Figs. 5–6. Genitalia of paratype male of Porphyrogenes peterwegei (genitalia dissection code X- 5235, voucher code 01-SRNP-5734) from ACG, Costa Rica; scale bar ¼ 1.0 mm. 5, Tegumen and uncus in dorsal view. 6, Complete genitalia, with vesica everted, in left lateral view. to suggest linked sausages). Uncus, in shaped, and prominently hirsute around dorsal view, a wide U, evenly curved at V. Membrane of posterior 40% of bottom. Female (Figs. 7–8): Well-scler- corpus bursae with limited, longitudinal otized sterigma surrounding midventral, sclerotization suggesting trough or ca- paler, usually membranous area at, and noe. posterior to, ostium bursae. Central part Ecological distribution: Known from of lamella antevaginalis with conspicu- ACG rainforest habitats between 95 m ous posterior edge forming sill of and 760 m. Present but scarce in ostium bursae. Wrinkled membrane intergrade between ACG rain forest ventral to sterigma with exaggerated and dry forest. wrinkling laterally and sclerotized plate Larval foodplants (Table 1): Diet midventrally; sclerotized plate same very restricted. Total of 213 caterpillars width as central part of lamella ante- found feeding on seven species of vaginalis (and directly beneath it) but Machaerium (Fabaceae, Fabales)— extending farther posteriad. Posterior mainly M. seemannii Benth. and M. edge of lamella postvaginalis thickened, acuminatum Kunth—but 14 caterpillars darkly sclerotized, midventrally V- found feeding on Dichapetalum mor- VOLUME 112, NUMBER 1 37

Figs. 7–8. Genitalia of paratype female of Porphyrogenes peterwegei (genitalia dissection code X-5238, voucher code 02-SRNP-6579) from ACG, Costa Rica; scale bar ¼ 1.0 mm. 7, Ovipositor lobes, sterigma, and bursa copulatrix in ventral view. 8, The same, plus part of ductus seminalis, in right lateral view. 38 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON

Figs. 9–20. Caterpillars, foodplant, and pupae of Porphyrogenes peterwegei from ACG, Costa Rica, in dorsal (left) and left lateral (right) view. 9–10, Second instar. 11–12, Third instar. 13–14, Fourth instar. 15–16, Fifth (last larval) instar. 17, Pupal shelter on Machaerium seemannii. 18, Pupa in frontal view within opened shelter. 19–20, Pupa. VOLUME 112, NUMBER 1 39

Table 1. Larval foodplants of Porphyrogenes ATATAGCATTCCCCCGAATAAATA peterwegei in Area de Conservacio´n Guanacaste, ATATAAGATTTTGATTATTACCCC northwestern Costa Rica, and number of wild CTTCTTTAACTCTTTTAATTTCAAG caterpillars encountered by the bioinventory feeding on each species of plant. AAGAATTGTAGAAAATGGTGCAG GTACAGGTTGAACTGTATACCCC Number of Family Species caterpillars CCTTTATCATCAAATATTGCTCATC AAGGATCTTCTGTTGATTTAGC Fabaceae Machaerium 26 AATCTTTTCTCTACATCTTGCAGG acuminatum Machaerium 1 TATTTCTTCAATTTTAGGAGCTATT falciforme AATTTTATTACTACAATTATTAATAT Machaerium 3 ACGAATTAACAATTTATCTTTTGA floribundum CCAAATACCTCTATTTGTTTGAGC Machaerium kegelii 1 AGTAGGAATTACAGCTTTATTATTA Machaerium 3 salvadorense TTATTATCATTACCAGTTTTAGCTG Machaerium 177 GTGCAATTACTATACTTTTAACTGA seemannii TCGAAATTTAAATACTTCATTTTTT Machaerium 16556 2 GATCCTGCTGGAGGAGGTGATCCT ATTTTATC. Dichapetalaceae Dichapetalum 14 Paratypes: morenoi 45 males, 50 females, ACG, Costa Rica (USNM). For more detail, see Janzen and Hallwachs enoi Prance (Dichapetalaceae, Malpigh- (2009). iales). Etymology.—This skipper is named Developmental times: In the rainforest in honor of Peter Wege of Grand environment, the duration of larval Rapids, Michigan, whose extraordinary development (first through fifth instar) generosity has (1) significantly in- ranges from 90 to 100 days. Pupal creased the size of ACG and perma- development lasts 16 to 30 days. There nently secured thousands of hectares of is no evidence of pupal dormancy. P. peterwegei rainforest habitat, (2) Type material.—Holotype: female stimulated conservation overall by sup- (Figs. 3–4), voucher code 03-SRNP- porting the founding of the Costa Rican 6133, Rio Blanco Abajo, Sector San Ministry of Environment and Energy in Cristobal, Area de Conservacio´n Gua- 1986–1990, and (3) advanced the mis- nacaste, Alajuela, Costa Rica, 500 m, sion and deeds of DNA barcoding. latitude 10.90037, longitude 85.37254 DISCUSSION (USNM), collector Carolina Cano. Yel- low label reads ‘‘LEGS AWAY/FOR DNA barcodes.—In all, 67 speci- DNA.’’ DNA barcode: AACCTTATATT mens of P. peterwegei have been TTATTTTTGGAATCTGAGCAGGAA barcoded. Individual variation in their TAGTAGGAACTTCTCTAAGATTAC barcodes is minimal, and the DNA TAATTCGAACTGAATTAGGAACC sequence of the holotype is, by far, the CCCGGATCTTTAATTGGAGATGAC commonest. The fact that both males CAAATTTATAATACTATTGTTACA and females have identical COI haplo- GCTCATGCTTTTATTATAATTTT types is, of itself, sound evidence that, CTTTATAGTTATACCTATTATAAT despite disparate outward appearances, TGGAGGTTTTGGAAATTGACTAG they belong to one and the same TTCCTTTAATACTGGGAGCTCCTG species. 40 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON

Natural history.—Independent evi- rium are vines is probably irrelevant. dence from numerous rearings makes Because P. peterwegei caterpillars have the same point. Caterpillars with the occasionally been found feeding on four same foodplants, the same behavior, of the viny species of Machaerium in and the same color patterns that change the intergrade between ACG rain and the same way in the course of larval and dry forest, they may eventually be pupal development (Figs. 9–16, 18–20) discovered eating M. biovulatum at the have produced nearly equal numbers wet margin of its dry-forest distribution. (52 males, 60 females) of the two sexes, Regardless of that possibility, one wonders why P. peterwegei does not which look so different from each other exploit M. acuminatum and M. kegelii as adults (Figs. 1–4). more fully by spreading into ACG dry Although the bioinventory of ACG forest—at least temporarily, during the Lepidoptera and their parasitoids began six-month rainy season. in 1978, it was not until 2000 that an Machaerium is in the family Faba- adult of P. peterwegei was successfully ceae in the order Fabales, and Dicha- reared. (One caterpillar had been found petalum is in the family Dichapetala- in 1995 and another in 1998; both died ceae in the order Malpighiales. of disease.) This lag was due, in large Although these orders are broadly part, to the inventory’s early emphasis related phylogenetically, each relates on dry forest habitats—but also to the more closely to a few other orders caterpillars’ construction of unusually (Stevens 2001); each comprises such a cryptic shelters. The flimsy shelter of a great diversity of taxa that the extreme- sizable caterpillar is hard to spot ly limited foodplant selection of P. because, in most cases, its occupant peterwegei may well stem from one or has lightly silked together one or two more similar or identical chemical cues leaves whose small and numerous having arisen independently in Machae- leaflets are close-set and more or less rium and Dichapetalum. In the ACG overlapping anyway (Fig. 17). Pupation bioinventory, P. peterwegei shares an takes place within the larval shelter emerging pattern—with Astraptes eno- (Fig. 18). trus (Stoll) (Hesperiidae), Morpho am- Why P. peterwegei is limited to rain athonte Deyrolle, and M. helenor forest is not entirely clear. Its known (Cramer) (Nymphalidae)—in which a larval foodplants (Table 1) are seven species not only specializes on an array species of woody vines in the legumi- of closely related foodplants (in this nous genus Machaerium (especially M. case, species of the genus Machaerium) seemannii), plus, infrequently, the quite but also eats one to three species in the unrelated species Dichapetalum more- genus Dichapetalum. noi, all of which grow in ACG rain In ACG, the caterpillar of P. peter- forest. On the one hand, neither Ma- wegei shares its primary foodplant, M. chaerium nor Dichapetalum occurs in seemannii, with the caterpillars of Tele- ACG cloud forest. On the other hand, miades nicomedes (Mo¨schler) (Hesper- two of the rainforest foodplants (M. iidae), Morpho amathonte, M. helenor, acuminatum and M. kegelii Meisn.) are and M. granadensis C. Felder and R. present in ACG dry forest as well, along Felder (Nymphalidae), Hapigia repan- with a common, but unused, dry-forest dens Schaus DHJ02 and Antaea lichyi congeneric tree (M. biovulatum M. Franclemont (Notodontidae), Oxidercia Micheli). That M. biovulatum is a tree toxea (Stoll) DHJ01 (Noctuidae), and and all other ACG species of Machae- four species of Stenomatinae (Janzen VOLUME 112, NUMBER 1 41 and Hallwachs 2009). None of these to eat chiefly at night, they are usually caterpillars are common enough to seen when exploring diurnal vertebrate compete significantly with P. peterwe- predators open their shelters—and then, gei, and none share parasitoids with P. if moldy or moldy-looking, are proba- peterwegei. bly bypassed. Presumably, the pale Like most species of ACG caterpil- pupae (Figs. 18–20) in larval shelters lars, the caterpillar of P. peterwegei is are likewise mimetically defended attacked by parasitoids that are host- against visually-orienting vertebrate specific. Of the 227 wild-caught cater- predators. pillars found to date, one was attacked No free-living adults of Porphyro- by an extreme generalist tachinid fly genes have been encountered in ACG (Patelloa xanthura (Wulp) DHJ01, during 40 years of random insect voucher code 02-SRNP-33798); 11 observationbyJanzenandHall- were attacked by a medium-sized, wachs—no doubt a consequence of undescribed tachinid (Siphosturmia these skippers’ late-day to at least Wood02) that apparently attacks no late-twilight period of activity (and other species of ACG caterpillar (out possibly to some activity far above- of 20,642 tachinid attacks of ACG ground in the canopy, although such caterpillars of more than 3,000 species behavior has never been documented). of Lepidoptera). A lone male specimen dated January The body of well grown caterpillars 1984 likely came from diverse lepidop- of P. peterwegei is ‘‘moldy white,’’ with teran material taken in a light trap. a lateral series of black-bordered Taxonomic tangentia.—The known ‘‘stacked windows’’ (Figs. 13–16)—a range of P. peterwegei is moderately color pattern that is more or less increased by a male in the INBio apparent in Bungalotis quadratum collection with the following label data: (Sepp), B. diophorus (Mo¨schler), B. Estacio´n Quebrada Bonita, R. B. Car- midas (Cramer), Nicephellus nicepho- rus (Hewitson), and Ocyba calathana ara, Prov. Puntarenas, Costa Rica, 50 m, (Hewitson) (Hesperiidae). Porphyro- agos 1993, R. N. Guzma´n. Identifica- genes peterwegei differs from these five tion of this worn, wild-caught specimen species in, e.g., its choice of foodplants, included genitalic examination (J. M. the number of windows in the side Burns genitalia dissection X-6353). stack, and details of color and pattern of In a brief account of P. peterwegei a prominently striped head (see images under the heading ‘‘Porphyrogenes un- in Janzen and Hallwachs 2009). Al- described species,’’ Austin and Mielke though these five species, which are (2008) refer to the ‘‘genitalia (Fig. 152) closely related, are also related to of an apparent male of this species’’ Porphyrogenes, it is not yet known if from Finca Hamadryas, Ciudad Colon, the larval similarities reflect a common Costa Rica, 15 September 1998, dissec- origin or parallel evolution (or both). tion GTA # 13917. The genitalia in Caterpillars with their seemingly figure 152 are not those of P. peterwegei ostentatious white color are probably (they may be those of yet another mimicking caterpillars killed by white undescribed species of Porphyrogenes), moldy fungi, which, in turn, defend and figures 83 and 84 in Austin and their food by producing secondary Mielke (2008) show dorsal and ventral compounds highly toxic to vertebrates facies of this misdetermined individual. (e.g., Janzen 1977). Because ACG Caterpillars of P. spina Austin and skipper caterpillars leave their shelters Mielke (described from two adult males 42 PROCEEDINGS OF THE ENTOMOLOGICAL SOCIETY OF WASHINGTON from Panama) have been found twice in Dry Forest Conservation Fund and ACG ACG rain forest: once feeding on M. to DHJ. seemannii, the favored foodplant of P. peterwegei, within a few meters of LITERATURE CITED caterpillars of that species; once on M. Austin, G. T. and O. H. H. Mielke. 2008. cobanense Donn. Sm. (see images of P. Hesperiidae of Rondoˆnia, Brazil: Porphyr- spina in Janzen and Hallwachs 2009). ogenes Watson (Lepidoptera: Pyrginae: Eu- These are the only two species of damini), with descriptions of new species Porphyrogenes known from ACG. from Central and South America. Insecta However, Austin and Mielke (2008) Mundi 0044: 1–56. record two more species, P. omphale Burns, J. M. 1997. Presidential address 1996: On (Butler) and P. speciosus Austin and the beauties, uses, variation, and handling of Mielke, from Costa Rica; another two, genitalia. Journal of the Lepidopterists’ Society 51: 1–8. P. glavia Evans and P. spoda Evans, Burns, J. M. and D. H. Janzen. 2001. Biodiversity from Panama; and one, P. sula (Wil- of pyrrhopygine skipper butterflies (Hesper- liams and Bell), from Honduras. More- iidae) in the Area de Conservacio´n Guana- over, their records of P. zohra (Mo¨sch- caste, Costa Rica. Journal of the Lepidopter- ler) include Honduras, Nicaragua, and ists’ Society 55: 15–43. Venezuela. The occurrence of this last Hajibabaei, M., D. H. Janzen, J. M. Burns, W. taxon in Costa Rica is no longer implied Hallwachs, and P. D. N. Hebert. 2006. DNA barcodes distinguish species of tropical but real, thanks to three males in INBio Lepidoptera. Proceedings of the National (J. M. Burns genitalia dissections X- Academy of Sciences USA 103: 968–971. 5032, X-5240, and X-6354). In light of Janzen, D. H. 1977. Why fruits rot, seeds mold, all these records and the scarcity of and meat spoils. American Naturalist 111: wild-caught adults, there are probably 691–713. more than two species of Porphyro- Janzen, D. H. and W. Hallwachs. 2009. Event- genes in ACG. based database of caterpillars, their host plants, and their parasitoids in Area de Conservacio´n Guanacaste, northwestern Cos- ACKNOWLEDGMENTS ta Rica. http://janzen.sas.upenn.edu Our thanks to Donald Harvey for Janzen, D. H., W. Hallwachs, P. Blandin, J. M. dissecting genitalia, Young Sohn for Burns, J. Cadiou, I. Chacon, T. Dapkey, A. R. Deans, M. E. Epstein, B. Espinoza, J. G. drawing two of them (Figs. 5–8), Karie Franclemont, W. A. Haber, M. Hajibabaei, J. Darrow for photographing adults (Figs. P. W. Hall, P. D. N. Hebert, I. D. Gauld, D. J. 1–4) and numbering plates, Mehrdad Harvey,A.Hausmann,I.Kitching,D. Hajibabaei for DNA barcoding, Bernard Lafontaine, J. Landry, C. Lemaire, J. Y. Hermier for sharing his extensive Miller, J. S. Miller, L. Miller, S. E. Miller, J. knowledge of Porphyrogenes,Sarah Montero, E. Munroe, S. Rab Green, S. Burns for helping with various tasks, Ratnasingham, J. E. Rawlins, R. K. Robbins, J. J. Rodriguez, R. Rougerie, M. J. Sharkey, ACG parataxonomists for finding and M. A. Smith, M. A. Solis, J. B. Sullivan, P. rearing the caterpillars and parasitoids, Thiaucourt, D. B. Wahl, S. J. Weller, J. B. and John Shuey and an anonymous Whitfield, K. R. Willmott, D. M. Wood, N. E. reviewer for useful information. This Woodley, and J. J. Wilson. 2009. Integration study was supported by the National of DNA barcoding into an ongoing inventory Museum of Natural History Small of complex tropical biodiversity. Molecular Grants Program to JMB; and by U.S. Ecology Resources 9(Supplement 1): 1–26. Stevens, P. F. 2001. Angiosperm Phylogeny National Science Foundation grants Website. Version 9, June 2008 [and more or BSR 9024770 and DEB 9306296, less continuously updated since]. http://www. 9400829, 9705072, 0072730, and mobot.org/MOBOT/research/APweb/ [ac- 0515699, and grants from Guanacaste cessed 2009].