2009 SOUTHEASTERN NATURALIST 8(4):739–745
Phylogenetic Analysis of Anisota (Insecta: Lepidoptera:
Saturniidae) Based on Scolus Size and Structure of Mature
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.
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 – email@example.com.
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
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.
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.
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.
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.
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
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
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
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.
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,  in Mexico (Lepidoptera: Saturniidae). Revista de Lepidopterología
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.