2009 SOUTHEASTERN NATURALIST 8(4):739–745 Phylogenetic Analysis of Anisota (Insecta: Lepidoptera: Saturniidae) Based on Scolus Size and Structure of Mature Larvae Joel T. Burke1 and Richard S. Peigler2,* Abstract - A phylogenetic analysis of 13 species of the genus Anisota was carried out using morphological data from mature larvae. Based on quantitative traits, length, and structure of scoli, 18 characters from 3 thoracic and 4 abdominal segments were characterized and coded. The phylogenetic analysis using the larval scoli yielded results in conjunction with previously proposed taxonomic relationships. For instance, the eastern North American species A. consularis, A. fuscosa, and A. manitobensis were found to share a close relationship. Similarly, A. discolor, A. virginiensis, and A. pellucida, three other eastern North American species, were found to be most closely related. However, discrepencies between expected and actual results were also observed: A. stigma and A. leucostygma fell out as sister species to all other Anisota species. This disassociation of the two species from their nearest relatives is believed to be due to the longer branched scoli found on both species and the designated outgroup, Citheronia regalis. Introduction Caterpillars of the genus Anisota, invariably called oakworms, are commonly encountered in forests, parks, and yards throughout eastern and central North America (Marshall 2006), especially in the southeast, where they can cause defoliation of oaks (Quercus spp.). Ten species of Anisota are known from the eastern half of North America. At least six additional species of Anisota range in the North American Southwest and throughout much of Mexico, where they are to be found with the rich diversity of oaks in the foothills and mountains of those regions (Peigler and Wolfe 2004). The adults are generally reddish brown or tan moths, the females of which are commonly attracted to lights, but the males of most species are diurnal fl iers, in which cases they are effective mimics of bees and wasps when in fl ight. Adults of most of the species of Anisota were shown in color by Ferguson (1971), Riotte and Peigler (1981), Lemaire (1988), and d’Abrera (1995). Aside from several species in Mexico and the southwestern United States, the Anisota in southern Canada and the eastern half of the United States have long been thought to fall into three distinct (presumably monophyletic) species groups. The common names Spiny Oakworm (A. stigma), Orangestriped Oakworm (A. senatoria), and Pinkstriped Oakworm (A. virginiensis) are 1Department of Marine Science and Environmental Studies, University of San Diego, 5998 Alcala Park, San Diego, CA 92110. 2Department of Biology, University of the Incarnate Word, 4301 Broadway, San Antonio, TX 78209-6397. *Corresponding author – peigler@uiwtx.edu. 740 Southeastern Naturalist Vol. 8, No. 4 widely applied to the representatives in the north, and all of the allied species in the south that were later discovered and named fall into one of these three groupings. Peigler and Wolfe (2004) hypothesized that these three intrageneric groups originated and diverged in Mexico, before each dispersed into North America. The Orangestriped Oakworm is the most notable of the defoliators, with documented outbreaks in Connecticut, Missouri (Riotte and Peigler 1981 and references cited therein), Michigan, New York, New Jersey, Pennsylvania, and Virginia (Coffelt and Schultz 1993a,b; Coffelt et al. 1993). The closely allied A. peigleri (Yellowstriped Oakworm) sometimes causes severe defoliation from the Piedmont of the Carolinas to northern Florida (Serrano 2001, Serrano and Foltz 2003). Color photographs of oakworms were published by Riotte and Peigler (1981), Laplante (1985), Lemaire (1988), Bouseman and Sternburg (2002), and Wagner (2005). These caterpillars are most readily recognized by: two prominent scoli located on the second thoracic segment; a granular, hairless integument; and longitudinal stripes (color dependent on species). External characters on caterpillars are routinely used to elucidate relationships among closely related species of Lepidoptera (Stehr 1987, Wagner 2005). The purpose of this study was to build and analyze a phylogenetic tree of the genus Anisota, based on quantitative traits, specifically scolus size and structure on mature larvae. Methods In our study, we examined mature larvae of 13 of the 16 known species of the genus (Table 1). We made field trips to eastern Texas to collect caterpillars of Anisota along highways in forests. These localities were near Huntsville (Walker County) and near Crockett (Houston County). Larvae of three species were collected, namely A. fuscosa, A. discolor, and A. senatoria, and brought back to San Antonio and reared to maturity in the lab Table 1. Anisota species, distribution, and collection sites included in the present study. Species Distribution Collection site A. assimilis (Druce) Mexico: Sierra Madre Occidental Chihuahua A. consularis Dyar Southeastern USA: coastal plain Tangipahoa Parish, LA A. discolor Ferguson Eastern Texas; Oklahoma Houston County, TX A. dissimilis (Boisduval) Mexico, widespread Guerrero A. fuscosa Ferguson Eastern Texas; western Louisiana Houston County, TX A. leucostygma (Boisduval) Mexico: Oaxaca; Tamaulipas Tamaulipas A. manitobensis McDunnough Southern Manitoba; Wisconsin Manitoba A. oslari Rothschild Southwestern USA; Sonora Santa Cruz County, AZ A. peigleri Riotte Southeastern USA Pickens County, SC A. pellucida (J.E. Smith) Southeastern USA Marion County, FL A. senatoria (J.E. Smith) Eastern North America Trempeleau County, WI A. stigma (Fabricius) Eastern North America Greenville County, SC A. virginiensis (Drury) Eastern North America Manitoba Citheronia regalis (Fabricius) Eastern USA Walker County, TX 2009 J.T. Burke and R.S. Peigler 741 on local oaks. Species of oaks that were favored in the field (i.e., selected by ovipositing females in nature) in eastern Texas were Quercus nigra L. (Water Oak), Q. falcata Michaux (Southern Red Oak), and Q. marilandica Muenchh. (Blackjack Oak), although larvae of Anisota were also observed by us on other oaks. Two to five mature larvae were examined for each species in Table 1 to assess and record the character states in Table 2. Caterpillars preserved in 70% isopropyl alcohol were examined under a dissecting microscope, and characters were coded. The 18 characters selected consisted of the following thoracic (T1, T2–T3) and abdominal (A1, A6, A8, A9) scoli: dorsal, subdorsal, lateral, subventral. As is normally the case (Stehr 1987:295), scoli of segments T2 and T3 are similar, A1 is similar to A2, and A6 is similar to A3–A5 and A7. Therefore, we did not tabulate characters for abdominal segments A2 through A5 and A7. Each character was assigned one of seven character states (1 = more than 12 mm long; 2 = 2–4 mm long, branched or forked; 3 = 2–4 mm long, unbranched; 4 = 1–2 mm long, unbranched; 5 = less than 1 mm long; 6 = fl attened, yet discernible; and 7 = absent, or not discernible), with no attempt to assign polarity, and given in Table 2. Characters 1–18 were as follows: 1 = T1 dorsal, 2 = T1 subdorsal, 3 = T1 lateral, 4 = T1 subventral, 5 = T2–T3 dorsal, 6 = T2–T3 subdorsal, 7 = A6 dorsal, 8 = A6 subdorsal, 9 = A6 lateral, 10 = A6 subventral, 11 = A8 dorsal, 12 = A8 subventral, 13 = A8 lateral, 14 = A8 subdorsal, 15 = A9 dorsal, 16 = A9 subdorsal, 17 = A9 lateral, and 18 = A9 subventral. After tabulation, we deleted the scoli on A1 because we found the character states to be identical in all 14 species, so those data were not informative and would thus make no difference in the resulting cladograms. Assigning polarity and lumping character states into smaller numbers early in the study yielded pairings of unrelated taxa and therefore erroneous trees. MacClade 4 (version 4.06, Maddison and Maddison 2003) translated Table 2. Character states for scoli observed in mature larvae of all 14 species. Character states Species 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 A. dissimilis 6 6 7 7 4 4 3 4 4 7 3 3 4 5 2 3 6 5 A. assimilis 6 6 5 5 4 5 4 5 5 5 3 5 5 5 3 5 5 5 A. oslari 6 6 6 5 5 5 4 5 4 5 4 5 4 5 2 4 7 5 A. consularis 6 6 6 5 2 5 4 5 4 5 2 4 4 5 2 4 7 5 A. fuscosa 5 5 5 5 2 4 2 5 4 7 2 5 4 5 2 4 5 5 A. manitobensis 5 5 6 6 5 4 4 5 5 5 4 5 5 5 4 5 5 5 A. peigleri 5 5 5 5 2 4 4 5 4 5 2 5 4 5 2 5 5 5 A. senatoria 6 6 6 5 5 5 5 5 5 5 4 5 5 5 4 5 5 5 A. virginiensis 6 6 6 5 5 5 5 5 5 5 4 5 5 5 2 5 5 5 A. pellucida 5 5 5 5 5 5 5 5 5 5 5 5 5 5 4 5 5 5 A. discolor 6 6 6 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 A. stigma 6 6 6 5 5 5 5 5 5 5 5 5 5 5 4 5 5 5 A. leucostygma 6 6 6 5 5 5 5 5 5 5 5 5 5 5 5 5 7 5 C. regalis 5 5 5 5 1 5 5 5 7 5 5 5 5 7 5 5 5 7 742 Southeastern Naturalist Vol. 8, No. 4 the scolus characters conceptually into the data set used for phylogenetic analysis. PAUP* 4.0b10 (Swofford 2002) was used to run maximum parsimony analysis of the data using unordered, equally weighted character transformations. The large number of taxa used in the study required a heuristic tree search in order to find the most parsimonious trees. An initial max tree setting of 10,000 using tree-bisection-reconnection branch swapping with 100 replicates of random taxon-addition sequences was performed, and only the best trees were kept (Hall 2004). Citheronia regalis (Regal Moth) was defined as the outgroup, because it is considered to be a basal representative of the subfamily Ceratocampinae (Lemaire 1988). Results and Discussion Of the 317,348 rearrangements found in the heuristic search, the nine most parsimonious trees were retained. The strict consensus of the nine most parsimonious (MP) trees is presented in Figure 1. Most of the nine trees did not show polytomies, although two polytomies are quite conspicuous in the strict consensus tree. In the other trees, A. consularis was usually grouped with A. manitobensis and A. fuscosa; all previous authors placed these in the “stigma group.” Anisota discolor was also grouped with the pair A. virginiensis and A. pellucida; previous authors united these “pinkstriped oakworms” as well. The close relationship of A. dissimilis, A. assimilis, and A. oslari has also been pointed out by Riotte and Peigler (1981) and Lemaire (1988). Therefore, the cladistic analysis of the larval scoli was generally concordant with previous proposed relationships, although these were unfortunately obscured by polytomies in the strict consensus tree. Figure 1. Cladogram showing hypothetical phylogeny of 13 species of Anisota based on the strict consensus of the nine most parsimonious trees. Tree length = 46, consistency index = 0.674, and retention index = 0.681. 2009 J.T. Burke and R.S. Peigler 743 In the strict consensus tree and all of the nine most parsimonious trees, A. leucostygma sorted out as the sister-group to all others in the genus, and A. stigma as the sister-group to all other Anisota except for A. leucostygma. We believe that the longer and branched scoli seen in Anisota stigma and A. leucostygma are plesiomorphic traits that caused these two species to fall outside of their respective nearest relatives. We do not believe that they are sister-species to all other Anisota in the tree. The outgroup species, Citheronia regalis, has long, branched scoli, and these data (correctly, we believe) apparently caused the computer program to recognize this plesiomorphy. This study can be viewed in terms of character evolution of scoli in Anisota. There is a trend toward reduction of scoli in at least three lineages within this genus. Speed and Ruxton (2007) stated that “When aposematic displays evolve, they cause reduced investment in costly spines …” Larvae of Anisota fit that model, where members of the orangestriped and pinkstriped groups are the most colorful (aposematic) yet have the smallest scoli. The high tannin content in late-season oak leaves may render these oakworms less palatable or less nutritious, since tannins may be sequestered in their midguts. Tannins bind with proteins leading to reduction in assimilation of nitrogen in insects, and are for the same reasons sometimes toxic to vertebrates (Feeny 1970). The different degrees of aposematism and their corresponding development of armature on the various Anisota could explain the above incongruous trees and why it appears that these characters (spines) are not evolving in whatever framework the polarity was constructed. The similar appearance of the moths and mature caterpillars of Anisota consularis and A. fuscosa leads us to believe that these two species are each other’s nearest relatives, and our morphological observations support this hypothesis. Ferguson (1971), who described A. fuscosa as a subspecies of A. stigma, even misidentified a female of A. consularis that he figured (plate 5, fig. 5) as “A. stigma stigma - Floridian form resembling subspecies fuscosa.” In September 2003, we were sent mature larvae of A. consularis from Tangipahoa Parish, LA, by Donald Henne. These appeared to us to be virtually identical to larvae of A. fuscosa that we collected near Ratcliff, Houston County, TX the same month, and we suspected that the material from Louisiana was probably A. fuscosa. However, we were able to find minute differences in the scoli (Table 2), and when adults emerged in summer 2004, the diurnal males from Louisiana were obviously A. consularis, but the larger nocturnal males from Texas were those of A. fuscosa. Females of these and other species (stigma, peigleri, senatoria) are sometimes difficult to distinguish, and wing color of all of them varies from pale tan to darker reddish brown, with various amounts of purplish shading and dark fl ecks. However, females of A. consularis probably show the greatest color variation of any species in the genus, ranging from very dark purplish brown to pale tan. Extremes of these forms were shown in color by Kimball (1965, plate 3, figs. 24, 25). The close relationship between A. fuscosa and A. consularis also argues against placing A. fuscosa as a subspecies or synonym of 744 Southeastern Naturalist Vol. 8, No. 4 A. stigma, as was done by Ferguson (1971), Lemaire (1988), and d’Abrera (1995). Whether the ranges of A. stigma and A. fuscosa meet or even overlap in western or central Louisiana remains unknown. In conclusion, the combination of molecular, morphological, and behavioral data is the best approach when conducting phylogenetic work. However, the results above reinforce the importance and power of using morphological characters in the construction of phylogenetic trees. Acknowledgments We thank Charles Mitter and Jerome Regier (both at University of Maryland) for their support and for providing technical training to J.T. Burke. This study was supported by the National Science Foundation, including an REU grant to J.T. Burke. Caterpillars of species that were difficult to obtain were given to us by Donald C. Henne (Louisiana State University), the late Roy O. Kendall, and Kirby L. Wolfe. Robert A. Miranda assisted with collecting in the field and phylogenetic analyses in the lab. Chris Peters (University of San Diego) gave critical comments on the manuscript. Literature Cited d’Abrera, B. 1995. Saturniidae Mundi: Saturniid Moths of the World, Part 1. Automeris Press, Keltern, Germany; Hill House, Melbourne, Australia and London, UK. 177 pp. Bouseman, J. K., and J.G. Sternburg. 2002. Field Guide to Silkmoths of Illinois. Illinois Natural History Survey, Champaign, IL. x + 97 pp. Coffelt, M.A., and P.B. Schultz. 1993a. Larval parasitism of Orangestriped Oakworm (Lepidoptera: Saturniidae) in the urban shade tree environment. Biological Control 3:127–134. Coffelt, M.A., and P.B. Schultz. 1993b. Quantification of an aesthetic injury level and threshold for an urban pest management program against Orangestriped Oakworm (Lepidoptera: Saturniidae). Journal of Economic Entomology 86:1512–1515. Coffelt, M.A., P.B. Schultz, and D.D. Wolf. 1993. Impact of late-season Orangestriped Oakworm (Lepidoptera: Saturniidae) defoliation on oak growth and vigor. Environmental Entomology 22:1318–1324. Feeny, P. 1970. Seasonal changes in oak leaf tannins and nutrition as a cause of spring feeding by winter moth caterpillars. Ecology 51:565–581. Ferguson, D.C. 1971. Bombycoidea: Saturniidae (part). Pp. 1–153, pls. 1–11, In R.B. Dominick et al. (Eds.). The Moths of America North of Mexico, fasc. 20.2A. E.W. Classey and R.B.D. Publishers, London, UK. Hall, B.G. 2004. Phylogenetic Trees Made Easy: A How-to Manual, 2nd Edition. Sinauer Associates, Sunderland, MA. 221 pp. Kimball, C.P. 1965. The Lepidoptera of Florida: An Annotated Checklist. Florida Department of Agriculture, Gainesville, FL. 363 pp. Laplante, J.-P. 1985. Papillons et Chenilles du Québec et de l’Est du Canada. France- Amérique, Montréal, PQ, Canada. 280 pp. Lemaire, C. 1988. Les Saturniidae Américains … The Saturniidae of America ... Los Saturniidae Americanos (= Attacidae): Ceratocampinae. Museo Nacional de Costa Rica, San José, Costa Rica. 480 pp., 64 pls. Maddison, D.R., and W.P. Maddison. 2003. MacClade 4: Analysis of phylogeny and character evolution. Version 4.06. Sinauer Associates, Sunderland, MA. 2009 J.T. Burke and R.S. Peigler 745 Marshall, S.A. 2006. Insects: Their Natural History and Diversity, with a Photographic Guide to Insects of Eastern North America. Firefl y Books, Buffalo, NY. 718 pp. Peigler, R.S., and K.L. Wolfe. 2004. Rediscovery of Anisota leucostygma (Boisduval, 1872), description of its immature stages, and notes on the genus Anisota Hübner, [1820] in Mexico (Lepidoptera: Saturniidae). Revista de Lepidopterología (Madrid) 32:5–12. Riotte, J.C.E., and R.S. Peigler. 1981. A revision of the American genus Anisota (Saturniidae). The Journal of Research on the Lepidoptera 19:101–180. Serrano, D. 2001. Biology, ecology, behavior, and natural history of the Yellowstriped Oakworm, Anisota peigleri (Lepidoptera: Saturniidae). Unpublished M.Sc. Thesis. University of Florida, Gainesville, FL. iv + 48 pp. Serrano, D., and J.L. Foltz. 2003. Natural history of Anisota peigleri (Lepidoptera: Saturniidae) in Gainesville, Florida. Florida Entomologist 86:217–218. Speed, M.P., and G.D. Ruxton. 2007. Warning displays in spiny animals: One (more) evolutionary route to aposematism. Evolution 59:2499–2508 Stehr, F.W. 1987. Immature Insects. Kendall/Hunt, Dubuque, IA. 754 pp. Swofford, D.L. 2002. PAUP*, 4.0, beta version. Sinauer Associates, Sunderland, MA. Wagner, D.L. 2005. Caterpillars of Eastern North America: A Guide to Identification and Natural History. Princeton University Press, Princeton, NJ. 512 pp.