Revision and Phylogeny of the
Limacodid-Group Families,
with Evolutionary Studies on
Slug Caterpillars
(Lepidoptera: Zygaenoidea)
MARC E. EPSTEIN
I
W9\
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY ? NUMBER 582
SERIES PUBLICATIONS OF THE SMITHSONIAN INSTITUTION
Emphasis upon publication as a means of "diffusing knowledge" was expressed by the first
Secretary of the Smithsonian. In his formal plan for the institution, Joseph Henry outlined a
program that included the following statement: "It is proposed to publish a series of reports,
giving an account of the new discoveries in science, and of the changes made from year to year
in all branches of knowledge." This theme of basic research has been adhered to through the
years by thousands of titles issued in series publications under the Smithsonian imprint,
commencing with Smithsonian Contributions to Knowledge in 1848 and continuing with the
following active series:
Smithsonian Contributions to Anthropology
Smithsonian Contributions to Botany
Smithsonian Contributions to the Earth Sciences
Smithsonian Contributions to the Marine Sciences
Smithsonian Contributions to Paleobiology
Smithsonian Contributions to Zoology
Smithsonian Folklife Studies
Smithsonian Studies in Air and Space
Smithsonian Studies in History and Technology
In these series, the Institution publishes small papers and full-scale monographs that report
the research and collections of its various museums and bureaux or of professional colleagues
in the world of science and scholarship. The publications are distributed by mailing lists to
libraries, universities, and similar institutions throughout the world.
Papers or monographs submitted for series publication are received by the Smithsonian
Institution Press, subject to its own review for format and style, only through departments of the
various Smithsonian museums or bureaux, where the manuscripts are given substantive review.
Press requirements for manuscript and art preparation are outlined on the inside back cover.
I. Michael Heyman
Secretary
Smithsonian Institution
S M I T H S O N I A N C O N T R I B U T I O N S T O Z O O L O G Y ? N U M B E R 5 8 2
Revision and Phylogeny of the
Limacodid-Group Families,
with Evolutionary Studies on
Slug Caterpillars
(Lepidoptera: Zygaenoidea)
Marc E. Epstein
SMITHSONIAN INSTITUTION PRESS
Washington, D.C.
19%
A B S T R A C T
Epstein, Marc E. Revision and Phylogeny of the Limacodid-Group Families, with
Evolutionary Studies on Slug Caterpillars (Lepidoptera: Zygaenoidea). Smithsonian Contribu-
tions to Zoology, number 582, 102 pages, 409 figures, 7 tables, 1996.?The limacodid group,
composed of Limacodidae (+ Chrysopolomidae), Dalceridae, Megalopygidae, Aididae, and
Somabrachyidae, is revised, and diagnoses, redescriptions, and keys to larval and adult stages
are provided for each family. Comparative morphology and ontogeny of the larval stage are
emphasized. Immature stages, including first instars, of a number of taxa are figured for the first
time. New character data for adults and immatures are presented, including the presence of larval
crochets on abdominal segments A2-A7 and A10 in species of Megalopygidae and
Limacodidae. Computer-assisted phylogenetic analyses of limacodid-group families using
separate character-data sets from adult and immature stages produces congruent cladograms of
equal parsimony, with Zygaenidae selected as the outgroup. Megalopygidae (Megalopygi-
nae + Trosiinae) is found to be the most basal family among the limacodid group, and Aididae,
previously a subfamily of Megalopygidae, is the sister group to the Limacodidae + Dalceridae
clade. Chrysopolomidae is placed as a subfamily of Limacodidae based on new character
evidence from adult and immature stages. Relationships of Somabrachyidae, Epipyropidae, and
Cyclotornidae also are discussed. Previous interpretations of character homologies for
abdominal proiegs in the larvae are revised based on the phylogenetic analysis, resulting in a
novel hypothesis concerning the evolutionary transformation from proiegs to a slug-like ventral
surface. New representations of homology for larval body setae and for juxta and valva of male
genitalia also are given. Behavioral observations of immatures and adults of species in the
limacodid group are presented, along with a review of their biology.
OFFICIAL PUBLICATION DATE is handstamped in a limited number of initial copies and is
recorded in the Institution's annual report, Smithsonian Year. SERIES COVER DESIGN: The coral
Montastrea cavernosa (Linnaeus).
Library of Congress Cataloging-in-Publication Data
Epstein, Marc E.
Revision and phylogeny of the Limacodid-group families, with evolutionary studies on slug caterpillars (Lepidop-
tera : Zygaenoidea) / Marc E. Epstein.
P. cm. ? (Smithsonian contributions to zoology ; no. 582)
Includes bibliographical references (p. 99).
1. Lepidoptera?Classification. I. Title. II. Series.
QL1.S54 no. 582 [QL545] 591 s?dc20 [595.78M] 96-16840
? The paper used in this publication meets the minimum requirements of the American
National Standard for Permanence of Paper for Printed Library Materials Z39.48 1984.
Contents
Page
Introduction 1
Acknowledgments 3
Material and Methods 3
Abbreviations 4
History of Classification of the Limacodid Group 5
Biology 6
Larvae 6
Cocoons 8
Adults 10
Eggs 10
Morphology 10
Adults 13
Larvae 25
Pupae 56
Phylogenetic Analysis 61
Data Analysis 61
Results and Discussion 66
Keys to Limacodid-Group Families and Allies 70
Key to Adults 70
Key to Larvae 71
Limacodid-Group Families and Allies 72
MEGALOPYGIDAE Herrich-Schaffer, 1855 72
AlDlDAE Hopp, 1927, new status 75
DALCERIDAE Dyar, 1898 79
LIMACODIDAE Duponchel, 1844 81
CHRYSOPOLOMINAE Aurivillius, 1895, new status 84
SOMABRACHYIDAE Hampson, 1920 85
EPIPYROPIDAE Dyar, 1903, and CYCLOTORNIDAE Meyrick, 1912 86
Ontogeny and Function of Limacodid Spinnerets 87
Homology and Evolution of Larval Body Setae 91
DALCERIDAE 91
LIMACODIDAE 91
Evolution and Origin 93
Evolution of the Ventral Abdominal Surface 94
Prolegs and Crochets 94
Homology of Suckers 95
Function and Adaptation 96
Summary 97
Literature Cited 99
111
Dedication
This contribution is dedicated to the memories of my father, Joseph Samuel Epstein, and
fellow lepidopterists F. Martin Brown and B. Adrienne Venables.
Revision and Phylogeny of the
Limacodid-Group Families,
with Evolutionary Studies on
Slug Caterpillars
(Lepidoptera: Zygaenoidea)
Marc E. Epstein
Introduction
Limacodidae (=Eucleidae, Cochlidiidae) is the richest
(- 1000 species) and most widespread moth family among a
monophyletic group of five families in the Zygaenoidea, which
I term the "limacodid group." Distributed worldwide, although
primarily in the tropics, the family Limacodidae is related to the
New World limacodid-group families Dalceridae (84 spp.)
(Miller, 1994), Aididae (=Aidinae, Megalopygidae, 6 spp.),
and Megalopygidae (~ 230 spp.) (Hopp, 1934; Becker, 1995).
Two Old World relatives are Chrysopolomidae (35 spp.) and
Somabrachyidae (~ 10 spp.). Restricted to Africa and Madagas-
car (Hering, 1937), Chrysopolomidae is placed herein as a
subfamily of Limacodidae. The Somabrachyidae, often re-
garded as constituting a subfamily of Megalopygidae, are
found in Africa and in Mediterranean Europe (Geertsema, pers.
comm.; Freina and Witt, 1990).
The limacodid group is characterized by external-feeding
larvae that have retractile heads and either prolegs on
abdominal segments A2-A7 or a flexible, slug-like ventral
Marc E. Epstein, Department of Entomology, National Museum of
Natural History, Smithsonian Institution, Washington, D.C. 20560.
Review Chairman: John M. Burns, National Museum of Natural
History, Smithsonian Institution, Washington, D.C. 20560.
Reviewers: Jeremy D. Holloway, CA.B. International Institute of
Entomology, 56 Queen's Gate, London, SW7 5JR, England; James S.
Miller, Department of Entomology, American Museum of Natural
History, New York, New York 10024-5192; John E. Rawlins, Section of
Invertebrate Zoology, Carnegie Museum of Natural History, Pitts-
burgh, Pennsylvania 15213-4080; Frederick W. Stehr, Department of
Entomology, Michigan State University, East Lansing, Michigan
48824-1115.
surface. There is a wide diversity of larval forms in the
limacodid group (Figures 1-6), although it is best known for
spiny larvae that cause dermatitis. The Limacodidae, Mega-
lopygidae, and Dalceridae are economic pests of palms and
other plantation crops in the tropics (Genty et al., 1978; Cock et
al., 1987).
Relationships among families in the limacodid group and
their allies are poorly understood. In large part this is due to an
incomplete knowledge of immature stages. Furthermore, there
has been little detailed morphological study of immature stages
or of adults.
Fortunately the fog is lifting. During the past decade there
have been studies on the limacodid-group families Limacodi-
dae (Holloway, 1986; Holloway et al., 1987; Epstein, 1988)
and Dalceridae (Stehr and McFarland, 1987; Miller, 1994) as
well as on others associated with them, such as parasitic
families Epipyropidae and Cyclotomidae (Davis, 1987; Com-
mon, 1990); however, comparative morphology from these
studies has been needed in order to gain a better understanding
of the interfamilial relationships. I base my phylogenetic
(= cladistic) analysis of the limacodid group on the compara-
tive morphological data given herein and in other recent studies
cited above. Although I do not treat characters of the
Zygaenidae extensively, I examine them in phylogenetic
context of the limacodid group.
Dyar (1895a, 1899b) believed ventral suckers in limacodid
larvae to be derived from fusion of the membranous pads found
in megalopygids. Hinton (1955:516), however, indicated that
there was no convincing explanation for the "loss of abdominal
prolegs" in Limacodidae. I introduce an alternative evolution-
ary hypothesis to explain the formation of suckers on the
ventral surface of limacodids. This hypothesis is based on
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
; ; 6
FIGURES 1 -6.?Habitus of live larvae (late instars) in the lirnacodid group: 1, two species of Megalopyge
(Megalopyginae), Pakitza, Peru (photo by K. Sandved); 2, Norape cretata (Trosiinae), USA, Maryland (photo
by D. Ferguson); 3, Dalcerides ingenita (Dalceridae), USA, Arizona (photo by F.W. Stehr); 4, Aidos amanda
(Aididae), Venezuela (photo by K. Sandved); 5, Chrysopoloma similis (Chrysopolominae), northern Transvaal,
South Africa (photo by N. Duke); 6, Acharia sp. (Limacodidae), Venezuela (photo by K. Sandved).
evidence of homology and on the results of phylogenetic
analysis of the limacodid group. Other character homologies
that I reinterpret include the larval setae on the thorax and
abdomen, and the juxta and valva of the male genitalia.
NUMBER 582
ACKNOWLEDGMENTS
The majority of this research was conducted during
postdoctoral study in the Department of Entomology, National
Museum of Natural History, Smithsonian Institution, and was
funded by the Office of Fellowships and Grants, Smithsonian
Institution. Field work was funded through the International
Environmental Science Program and the BIOLAT Program,
both of the Smithsonian Institution.
I thank my principal postdoctoral advisor, Don R. Davis, and
other active advisors Vitor O. Becker (Empresa Brasileira de
Pesquisa Agropecu&ria (EMBRAPA)), William E. Miller
(University of Minnesota), Scott E. Miller (Bishop
Museum), John E. Rawlins (Carnegie Museum), Jerome C.
Regier (University of Maryland), and Richard C. Froeschner
(Smithsonian Insitution) for their encouragement and support.
Hendrik Geertsema (University of Stellenbosch) provided me
with important information on Somabrachyidae. I wish to
acknowledge Allan Watson, David Carter, Malcom Scoble, and
Martin Honey, for their assistance during my visit to the
Natural History Museum, London. During my postdoctoral
fellowship, I had fruitful research discussions with fellow post-
and predoctoral students David Adamski, John W. Brown,
Greg Courtney, Chris Dietrich, Jon K. Gelhaus, Gustavo
Hormiga, Patricia Gentili, the late B. Adrienne Venables, and
Susan J. Weller. I also benefited from discussions with Douglas
C. Ferguson, Thomas J. Henry, Ronald W. Hodges, Robert W.
Poole, and M. Alma Solis (Systematic Entomology Lab, United
States Department of Agriculture (USDA)); Mary F. Mick-
evich (National Oceanic and Atmospheric Administration);
John M. Bums, the late J.F. Gates Clarke, Don Harvey, Wayne
Mathis, Ronald McGinley, Michael G. Pogue, and Robert K.
Robbins (Department of Entomology, Smithsonian Institu-
tion); Pamela M. Henson (Smithsonian Archives); and Timo-
thy P. Friedlander and Robert Leclerc (Center for Agricultural
Biotechnology, University of Maryland, College Park). D.R.
Davis, S.E. Miller, WE. Miller, and J.C. Regier commented on
presubmission versions of the manuscript. I wish to thank my
editor at Smithsonian Institution Press, Cheryl Roesel, for
making a number of important suggestions to the manuscript.
Barbara Mann (University of Toledo) and John Dodge
(Annandale, Virginia) also gave me helpful editorial advice.
A number of people were instrumental in field aspects of this
project. I wish to thank Tom?s Blohm for permission to
conduct field work at Hato Masaguaral, Venezuela, and R.
Rudran (National Zoological Park, Smithsonian Institution) for
administrative assistance that made my visits possible. Those
acknowledged for field assistance at Hato Masaguaral include
G. Agoramoorthy (Smithsonian Institution), Roger Blahnik
(University of Minnesota), Manuel F. Rodriguez and M.
Eugenia Rodriguez (Hato Masaguaral), and John Wilterding III
(Michigan State University).
I appreciate the assistance of Smithsonian photographers
Kjell B. Sandved, Victor Krantz, Laurie Minor-Penland, Carl
Hansen, and Chip Clark. V. Krantz and Jane Beck provided
contact prints. I also thank D.C. Ferguson, Noel McFarland
(Ash Canyon, Arizona), S.E. Miller, and Frederick W. Stehr
(Michigan State University) for contributing photographs.
Elaine R.S. Hodges, Gustavo Hormiga, Vichai Malikul, and
Young Sohn (Smithsonian Institution), and Linda Lawrence
(USDA) all made helpful suggestions on the preparation
of the figures. Susann Braden, Walt Brown, Peter Viola, Brian
Kahn, and Victoria Godwit (Smithsonian Institution) assisted
with scanning electron microscopy. Jerry Louton (Smithsonian
Institution) provided computer assistance. John Dodge assisted
in preparing the plates, and Birger Neuhaus (Smithsonian
Institution) helped me to get started in the darkroom. Last but
not least, I wish to thank my wife Joan F. Epstein, whose
patience and support for my research were essential to the
completion of this project.
Specimens used in this study were borrowed from The
Natural History Museum, London (formerly British Museum
(Natural History)) (BMNH); Carnegie Museum of Natural
History (CMNH); and National Museum of Natural History,
Smithsonian Institution, Washington, D.C. (collections of the
former United States National Museum (USNM)). A number of
colleagues provided specimens essential to this project,
including James Adams (Dalton College, Georgia), J.M. Burns,
Neville Duke (Swaziland), Phillipe Genty (Industrial Agraria
La Palma South America (INDUPALMA)), Hendrik
Geertsema, Dale Habeck (University of Florida), Daniel H.
Janzen (University of Pennsylvania), Candy Feller (Smith-
sonian Environmental Research Center), Noel McFarland,
Claus Naumann (Alexander Koenig Zoological Research
Institute and Zoological Museum, Bonn, Germany) (through
J.A. Scott), Steven C. Passoa (Animal Plant Health Inspection
Service, USDA (APHIS)), J.E. Rawlins, James A. Scott
(Lakewood, Colorado), F.W. Stehr, and David L. Wagner
(University of Connecticut).
MATERIAL AND METHODS
Specimens examined for this work are listed in Table 1, with
additional limacodid material reported in Epstein (1988),
Epstein and Miller (1990), and Epstein and Becker (1994).
Standard dissection and wing clearing procedures for adult
specimens were used (Clarke, 1941). Drawings were made with
the assistance of a camera lucida. Specimens studied by
scanning electron microscopy (SEM) were glued to a stub
using water-soluble household glue and were sputter-coated
with gold-paladium in the SEM Laboratory, Smithsonian
Institution. Immature stages examined by SEM were stored in
70% ethanol, put through a dehydration series to absolute
ethanol, and critical-point dried. Pupal and larval exuvia,
usually stored dry, were removed from cocoons and glued to a
round coverslip, which was then taped to the SEM stub. Dry
larval exuvia sometimes were softened with trisodium phos-
phate, stored in 70% ethanol, and later critical-point dried.
Micrographs were taken using Cambridge Stereoscan 100 and
Hitachi 570 scanning electron microscopes. Observations of
larval behavior (functional morphology) were assisted by the
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
TABLE 1.?Specimens examined by SEM or dissected for character data, comparative morphology, or
ontogenetic information. Immature stages: e = egg, I = first instar, L = post-first instar. Lex = larval exuvia,
P = pupa, Pex = pupal exuvia, C = cocoon. Adults: M = male genitalia, F = female genitalia, H = adult
mouthparts (SEM), T = tarsi (SEM), WB = whole body. (Collection data is given in parentheses.)
ZYGAENIDAE
Harrisina americana (Gu6rin-M6neville): (USA: Virginia, USNM) L;
(USNM) e
Harrisina metallica Stretch: (Dyar, USNM) L
Levuana iridescens Baker (Fiji: Clarke, USNM) P
Zygaena trifolii (Esper): (USNM) L; (Naumann) 1
EPIPYROPIDAE
Fulgoraecia exigua (Hy. Edwards): (USA: Maryland, Epstein, USNM) P
MEGALOPYGINAE
Megalopyge crispata (Packard): (USA: Maryland, Epstein, USNM) 1, L, P
Megalopyge nr. melaina: (Peru: Epstein, USNM) L, P, Pex
Megalopyge lanata Stoll: (Ecuador: USNM) L; (Ecuador: CMNH) L, P;
(Panama: USNM) F
Megalopyge sp.: (Venezuela: Epstein, USNM) WB, e, F
Podalia sp.: (Venezuela: Epstein, USNM) WB, M
Megalopyge sp. (Venezuela: BMNH) 1
TROSIINAE
Mesocia pusilla (Stoll): (Colombia: Genty, USNM) L
Norape corporalis Major (Venezuela: Epstein, USNM) WB, M
AIDIDAE
Aidosamanda Stoll: (Venezuela: Epstein, USNM) F, M, WB, e, L, C; (Brazil:
USNM) F
Aidos sp.: (Venezuela: BMNH, 2 specimens) L
Aidos sp.: (Brazil: BMNH. 2 specimens) Lex, Pex, C
Aidos yamouna (Dognin): (Peru: Sanders, USNM) Pex
Brachycodilla carmen (Schaus): (Brazil: USNM) WB, M
SOMABRACHYIDAE
Psycharium sp.: (South Africa: USNM. 2 specimens) L
Psycharium sp.: (South Africa: Cape Province, Geertsema, 2 specimens) e, 1
Somabrachys aegrotus Klug: (Morocco: CMNH) M
Somabrachys sp.: (Algeria: BMNH) L
DALCERIDAE
Acraga coa (Schaus): (Mexico: USNM) F
Acraga infusa complex: (Colombia: Genty, USNM) e, L, P
Dalcera abrasa H-S: (Venezuela: USNM) T
Dalcerides ingenita (Hy. Edwards): (USA: Arizona, McFarland) e, 1, L, P, C
Dalcerina tijucana (Schaus): (Brazil: USNM) (e, L)
dalcerid sp.: (Ecuador: CMNH) I
LlMACODIDAE
Acharia (= Sibine) sp.: (Venezuela: Epstein, USNM) 1, L, P, A
Apoda biguttata (Packard): 1
Belippa horrida Walker: (Taiwan: Rawlins, CMNH, 88-229) 1, F
Belippa laleana Moore: (China, USNM) L
Crothaema sp.: (South Africa: northern Transvaal, Duke, USNM) Lex, P, C
Crothaema sp.: (Malawi: Rawlins, CMNH, 88-636) 1, F
Euclea delphinii (Boisduval): (USNM) H
Euprosterna sp.: (Venezuela: Epstein, USNM) I, L, P
Heureles sp.: (Dominican Republic: Rawlins, CMNH) 1
Monoleuca semifascia (Walker): (USNM) e; (USA: Florida, Habeck) P
Natada subpectinaia Dyar: (Colombia: Genty, USNM) L
Pantoctaenia prasina (Butler): (Malawi: Rawlins, CMNH, 88-634) e, 1, F
Pantoctaenia gemmans Felder: (Swaziland: Duke, USNM. 2 specimens) L
Parasa chloris (Moore): (USNM) H
Phobetron hipparchia (Cramer): (Venezuela: Epstein, USNM) I
Phobetron pithecium Abbott and Smith: (USA: Virginia, J. Burns, USNM) C,
Pex; (USA: North Carolina, USNM) WB, M
Prolimacodes badia Hiibner: (USA: Connecticut, Wagner, USNM) 1
Semyra coarctata complex: (Venezuela: Epstein, USNM) e, 1, L, P, M, F
Talima postica Walker: (Venezuela: Epstein, USNM) 1, L, P
CHRYSOPOLOMINAE
Achrocerides theorini (Aurivillius): (Cameroon: USNM) F
Slrigivenifera venata (Aurivillius) (Uganda: USNM) T
Chrysopoloma similis Aurivillius: (South Africa: northern Transvaal, Duke)
Lex, Pex, C
use of 8 mm micro- and macrovideo and the use of a 16 mm
movie camera fitted with a macro lens. Locomotion and
silk-spinning behaviors were observed on a glass substrate and
were video recorded from beneath by using an inverted
compound microscope.
Larval chaetotaxy used herein largely follows Stehr (1987b)
and Tremewan (1985). The examination of first instars was a
major part of this study because few first instars have been
figured since Dyar's contributions in the 1890s and in 1907,
and all of his first instar specimens now are lost. Specimens
were obtained from the eggs of females collected at lights in the
field. Later instars were collected or reared using basic methods
described by Dyar (1899a). First-instar limacodids in many
genera are figured herein for the first time. These include
African Crothaema and Pantoctaenia (Rawlins, CMNH);
Neotropical Semyra, Talima, Euprosterna, Phobetron hip-
parchia (Epstein, USNM), and Heuretes (Rawlins, CMNH);
and Asian Belippa (Rawlins, CMNH). Also figured for the first
time are first instars and pupae of Dalceridae; male and female
genitalia of Aididae; and larvae, pupae, and female genitalia of
Chrysopolominae (sensu nova). Larval and pupal descriptions
of Chrysopoloma similis are based on exuviae, photographs
(Figure 5), and sketches, all provided by N. Duke.
ABBREVIATIONS
Following is a list of abbreviations used in the text and
figures:
1A-3A anal veins
A1 -A 10 abdominal segments
A A anterior apophysis
AC accessory cell
Aed aedeagus
AG accessory gland
An antenna
AP posterior apophysis
AS axial seta
AT anal tube
CB corpus bursae
Cr crochets
Crm cremaster
CuA. , anterior cubital veins
NUMBER 582
CuP posterior cubital vein
Cx coxa
D dorsal group (setae or derivatves D1-D2)
DB ductus bursae
DS ductus seminalis
En epicranial notch
Epm mesepimeron
Fe femur
FS frass-flipping seta
FW forewing
Ga galea
Gn gnathos
HW hindwing
Jx juxta
L lateral group (setae or derivatives LI -L3)
Lbr labrum
Lg leg
LL lateral lobe
LP labial palpus
M,_3 medial veins
Md mandible
ML mesal lobe
MSD subdorsal proprioceptor group (setae or derivatives MSD1-
MSD2)
MV ventral proprioceptor seta
MT Malpighian tubules
Mx maxilla
MXD proprioceptor seta
MxE maxillary extension
MxP maxillary palpus
OB ostium bursae
Ov oviduct
PA papillae anales
PI proleg
PP paraproct seta
Rl sensillum
R,_s radial veins
Rs radial sector vein
S1-S2 stemmatal setae
Sa saccus
Sc subcostal vein
SD subdorsal group (setae or derivatives SD1-SD3)
SG sebaceous gland
Si silk
SiP silk pore
So socius
Sp spiracle
SpS spiracular sensillum
Spt spinneret
St spermatheca
Stl-St6 stemmata 1-6
Stm stemmata
SV subventral group (setae or derivatives SV1 -SV4)
Tl prothorax
T2 mesothorax
T3 metathorax
Ta tarsus
Te tegumen
Ti tibia
Tr transtilla
Un uncus
V ventral group (seta or derivatives VI)
Va valva
Vi vinculum
XD X-dorsal group (setae or derivatives XD1-XD2)
History of Classification of the Limacodid Group
This section focuses on the placement of the limacodid-
group families within the classification of Lepidoptera from the
1890s to the present. Additional review of limacodid-group
families, as well as of Epipyropidae and Cyclotornidae, is
found under each family description below. Family-group
names or groupings are given in the current sense. Names in the
sense of the authors below are given in Table 2.
Packard (1894) believed the Megalopygidae to be general-
ized "bombycines," and he believed the Limacodidae to have
"originated from the Saturniidae or forms allied to them"
(Packard, 1893:83). During the same period, Dyar (1894)
and Chapman (1893) placed Limacodidae, Megaiopygidae, and
Zygaenidae in "Microlepidoptera."
Epstein and Henson (1992) provide a historical review of
Dyar's interests in both the Limacodidae and the higher
classification of Lepidoptera. Dyar (1894), on the basis of
larval setae, assigned Limacodidae, Megalopygidae, Zy-
gaenidae (in part), and Procridinae (Zygaenidae) to "Anthrocer-
ina." The Pterophoridae were temporarily added (Dyar, 1895a)
but were later placed in uncertain status (Dyar, 1895b).
Ultimately, Dyar combined the "Cossina" and Anthrocerina
into the superfamily "Tineides" (Dyar, 1896a). Both Cossina
and Tineides were known at the time as "Microlepidoptera."
Chapman (1893, 1894) lumped the Zygaenidae with Lima-
codidae and Micropterygidae into the pupa group "Incomple-
tae." The two families Limacodidae and Micropterygidae were
thought by Chapman (1894) to share homologous setal and
proleg arrangements; however, these views did not gain
acceptance (Dyar and Morton, 1895; Packard, 1895a:797;
Hinton, 1955).
By 1895, Packard concurred with the idea of a close
relationship between the Limacodidae and Megalopygidae, but
he placed them below Tineina, apart from Zygaenidae
(Packard, 1895a, 1895b). The Dalceridae were not included
in these groupings because they were not recognized as
separate from the Limacodidae or Megalopygidae until Dyar
erected the family in 1898.
Fracker's (1915) Zygaenoidea included Chalcosiinae and
Procridinae, Epipyropidae, Dalceridae, Megalopygidae, and
TABLE 2.?Family- and subfamily-group names of Zygaenoidea in the current
sense and in the sense of the cited authors.
Author
Packard (1894)
Dyar (1894)
Chapman (1894)
Fracker(1915)
Current sense
Megalopygidae
Limacodidae
Limacodidae
Zygaenidae (in part)
Procridinae (Zyg.)
Limacodidae
Micropterygidae
Limacodidae
Chalcosiinae (Zyg.)
Procridinae (Zyg.)
Old sense
Lagoidae
Cochliopodidae
Eucleidae
Anthroceridae
Pyromorphidae
Cochliopodidae
Eriocephalidae
Cochlidiidae
Chalcodidae
Pyromorphidae
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
Limacodidae. This classification has been expanded by modern
workers to include all of the current subfamilies of Zygaenidae,
Cyclotornidae, Somabrachyidae, Chrysopolomidae, and Heter-
ogynidae (Common, 1975,1990; Kuznetsov and Stekol'nikov,
1981; Minet, 1986; Scoble, 1992).
Except Zygaenidae, Brock (1971) placed most families
previously in Zygaenoidea in the Cossoidea, based primarily on
character systems of the adult thorax and forewings. Brock
shifted Epipyropidae and Heterogynidae to Tineoidea.
Heppner (1984) retained Megalopygidae, Somabrachyidae,
and Heterogynidae in Zygaenoidea, but he placed Epipyropi-
dae, Cyclotornidae, and the remainder of the limacodid group
in the Cossoidea. He did not provide justification, however, for
this arrangement.
Brock's Cossoidea, except for the position of Epipyropidae
and Heterogynidae, has been followed in recent catalogs (e.g.,
Fletcher and Nye, 1982) and geographic checklists and has
been followed by the Zoological Record (Zoological Society of
London, 1992-1993). It has been criticized, however, both for
the weakness of its adult characters and because of the strength
of immature-stage characters as evidence for Zygaenoidea in
the earlier sense (Common, 1975; Kuznetsov and Stekol'nikov,
1981; Minet, 1986). For example, Kuznetsov and Stekol'nikov
(1981) suggested that the musculature of the male genital ia
does not provide any evidence of synapomorphies between the
Cossidae and the limacodid group, and Minet (1986) provided
a list of zygaenoid synapomorphies based mostly on immature
stages.
Currently, even Brock no longer subscribes to the arrange-
ment in his earlier work. He included the "Eucleoidea" and
Zygaenoidea in the Zygaeniformia, apart from the Cossiformia
(Brock, 1990). I follow herein the superfamily grouping of
Common (1990) and Minet (1986).
Biology
The biology of the Limacodidae was reviewed by Holloway
(1986) and by Cock et al. (1987), both of whom emphasized the
Old World literature. Either host plant records or natural
enemies of Neotropical members of the limacodid group were
given by Lima (1945), Silva et al. (1968), Biezanko et al.
(1974), Miller (1994; Dalceridae only), and Epstein (1995;
Aididae only).
LARVAE.?Limacodid-group larvae are frequently polyph-
agous and feed on old or tough leaves, as illustrated by larvae
of temperate species that feed on oaks from July through
September (Dyar, 1899a:241) when tannins and secondary
compounds are at peak concentrations (Feeny, 1970). Larvae of
limacodids and dalcerids reportedly feed on smooth leaves of
trees and shrubs (Dyar, 1899a; Miller, 1994). In early instar
larvae of the limacodid group, leaf feeding is characterized by
the consumption of mesophyll from one surface (Figure 9). Not
until the third or fourth instar do larvae feed on the entire leaf.
Middle to late instars feed with their heads retracted beneath the
thorax, while clasping the leaf edge by using flaps on each side
of the prothorax (Figure 10). This probably makes it more
difficult for a predator viewing the larva from below to detect
the motion of mouthparts. Limacodid, dalcerid, and aidid
larvae are further concealed by the absence of verrucae or
urticating setae on the prothorax.
A general review of the medical literature on the urticating
setae of Lepidoptera included those of limacodids and
megalopygids (Delgado Quiroz, 1978). Pain-producing hista-
mines have been reported in venomous setae of limacodid
species and in a species of zygaenid (Itokawa et al., 1985).
Megalopygid verrucae bear urticating setae as well as plumose
setae. Long plumose setae have been considered to be merely
ornamental (Figure 1) (El-Mallakh et al., 1986); however, these
"harmless" setae spread the toxin released from the sharp
spines beneath.
When disturbed, some smooth limacodid larvae secrete
droplets from pores on the dorsum (Figure 8). This has been
reported in Prolimacodes badia (as Limacodes, see Patton,
1891:43; as Eulimacodes, see Dyar, 1896a: 173, 177), and it
occurs in the related Semyra coarctata complex (Figure 8).
Although their chemistry is unknown, the droplets often smell
like the host plant (Patton, 1891).
Larvae of the Dalceridae have gelatin-covered setae. This
sticky coating has been shown to be a deterrent to ants, and it
is regenerated following removal (Epstein et al., 1994). The
production and chemistry of this integumental coating is
presently under study in collaboration with T. Eisner and S.
Smedley (Cornell University).
Unique aspects of locomotion in caterpillars of the limacodid
group can be observed laterally or from below as a larva crawls
on glass (Figures 11, 12). The ventral abdominal segments in
limacodids and dalcerids move in fluid waves, or peristalsis.
This is facilitated by the extreme flexibility of the ventral
cuticle, reduction in proleg size, and by having additional
segments in complete contact with the substrate (e.g., segments
A1-A2, A7-A9). The presence of suckers, in varying degrees
of development, on abdominal segments A1-A8 also may
increase the adhesion of the ventrum. Capable of forward or
reverse locomotion, a forward wave begins when the venter of
abdominal segment A9 contacts the substrate, because the A10
proleg is raised during locomotion. As a wave progresses, the
portion retracted from the substrate is followed by a lateral
bulge until it reaches the first abdominal segment. When a larva
angles its head and thorax to turn, an oblique wave follows. On
narrow surfaces, such as leaf edges and stems, larvae can clasp
longitudinally from each side of the venter or can clasp from
anterior to posterior.
Hinton (1955:516) suggested that in limacodid larvae "a
liquid is secreted over the [ventral] cuticle." In limacodid and
dalcerid caterpillars, I find that the liquid, at least in part, is
semifluid silk. The apparently sticky silk is dabbed on the
substrate, sometimes in a figure eight, with the brush-like
spinneret (Figures 152-156, 158-162). Although the produc-
tion of a thin ribbon of silk by Apoda limacodes Hufnagel (as
Limacodes testudo) was reported by Chapman (1894), semi-
NUMBER 582
FIGURES 7-13.?Larval biology in Semyra coarctata complex (Limacodidae) (7, 8, 10-13, photos by C. Clark):
7, lateral view; 8, "defensive" droplets secreted on the dorsum (arrow); 9, early instar feeding; 10, late instar
feeding on leaf edge (note mouthparts covered by prothorax); 11-13, ventral surface viewed from below, through
glass (note Malpighian tubules); 11, 12, during locomotion (arrow points to leading edge of wave); 13, resting
prior to molt (note opaque ventnim and spinneret tucked under flexible cuticle). (MT = Malpighian tubules.)
fluid silk has not been previously associated with the fluid on
the ventral surface or with ventral surface stickiness. Using
scanning electron microscopy I have found no secretory pores
on the ventrum that could produce the fluid (Figures 265-270),
8 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
as was suggested by Holloway (1986:51). I also have observed
that brush-like spinnerets in limacodids and in dalcerids
(Figures 152-156) clean off debris that stick to the ventrum,
enabling the caterpillars to adhere to substrates more effi-
ciently.
Locomotion in larval aidids (e.g., Aidos amanda) is similar
to that in limacodids and dalcerids, except the ventrum is much
less flexible (pers. obs.). The motion of the megalopygid
ventrum on smooth surfaces is less fluid than it is in
limacodids, dalcerids, or aidids because the membranous pads
on the prolegs, rather than the entire ventrum, contact the
substrate (Figures 252-257) (pers. obs.; Dyar, 1899b). The
presence of these pads appears to be unique in the limacodid
group.
Aidids, megalopygids, and somabrachyids (e.g., Psychar-
ium) all issue normal silk when set on glass substrates (pers.
obs.).
COCOONS.?The onset of cocoon construction in limacodid-
group species is marked by final frass pellets that are frequently
a chalky white. The release of fluid from pores on the dorsum
of the prepupae helps to mat setae for a more compact shape;
simultaneously, it reduces the size of the prepupa (to nearly
one-half in limacodids (McNaulty, 1967)). A change in color
and loss of pigment pattern also occur on the dorsum in
limacodids (McNaulty, 1967) and dalcerids (e.g., Dalcerides
ingenita, pers. obs.; F.W. Stehr, pers. comm.). The ventral
cuticle in the two families becomes more opaque prior to
molting or cocoon construction (Figure 13). Temperate species
FIGURES 14-19.?Cocoons of the limacodid group (14-16, photos by L. Minor-Penland; 17, 18, photos by V.
Krantz): 14, Megalopyge nr. melaina (Megalopyginae), Peru; 15, Aidos sp. (Aididae), Brazil (trap-door exit
below); 16, Talima postica (Limacodidae), Venezuela; 17, Aidos sp., Brazil; 18, Dalcerides ingenita
(Dalceridae); 19, detail of dalcerid cocoon silk (photo by S.E. Miller).
NUMBER 582
FIGURES 20-25.?Cocoon of Semyra coarctata complex (Limacodidae) (scale length in parentheses): 20,
cocoon with pupal exuvia (photo by C. Hansen); 21, cocoon and lid (2 mm); 22, cross section of cocoon where
lid has been detached (50 u,m); 23, dried excretion on outside of cocoon from Malpighian tubules (50 |im); 24,
silk on inside of lid (200 u.m); 25, silk inside cocoon (100 \un).
of limacodids and megalopygids diapause in the cocoon as
prepupae during the winter (pers. obs.).
With the exception of the Dalceridae, cocoons in the
limacodid group have trap-door exits (Figure 15) or lids
(Figures 16, 21). Lids are opened by the pupa (Schremmer,
1990), which pushes nearly completely out of the cocoon prior
to eclosion (Figures 17, 20) (further description is given in the
pupal morphology section, below).
Megalopygid cocoons (Figure 14) vary in shape depending
on the amount of plumose larval setae incorporated into the
cocoon. Those with highly plumose larvae tend to be softer,
with the trap-door hidden by the outer mesh of silk, whereas
other megalopygids have a harder, more leathery cocoon.
Aidid cocoons are spun between leaves and have one or two
pairs of small round holes in the cocoon wall (Figures 15, 17).
These holes are in the outer mesh only and seem to mimic
parasitoid exit holes (Hopp, 1930, 1934; Forbes, 1942:3%;
Epstein, 1995).
Spun between leaves, dalcerid cocoons (Figures 18,19) have
an outer portion of diffuse silk and an oval inner chamber.
Gelatinous warts on the larva are sloughed off in the inner
chamber during its construction (e.g., Dalcerides ingenita,
10 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
pers. obs.). The pupae extend out one side of the chamber just
prior to eclosion (pers. obs.; Dyar, 1925).
Similar to those found in Zygaenidae, limacodid cocoons
(Figures 16, 20) are usually harder and more ovoid than
cocoons in the other families. The circular lid of the limacodid
cocoon, unlike the trap door in Megalopygidae and Aididae
cocoons, is not visible prior to eclosion. In fact, the prepupa
does not construct a hatch with a preformed lid but, rather,
spins a circular region of weakness at one end (pers. obs.; Dyar
and Morton, 1895:150; Common, 1990). In some limacodid
genera, prepupae spray a white muddy liquid, excreted by the
Malpighian tubules, from the anus once the mesh of the cocoon
is formed. The liquid is spread around by the ventral surface of
the prepupa and later dries into crystaline structures on the
outside of the cocoon (Figure 23), forming layers between the
outside and the inner silk wall (Figure 22) (pers. obs.; Ishii et
al., 1984). Ishii et al. (1984) found that cocoons in Momma
flavescens Walker contain 35% ash, with calcium oxalate
crystals from the Malpighian tubules predominant, and 34%
protein, made up of dense silk mesh and other sclerotized
proteins that they presumed to be from salivary secretions. Ishii
et al. attributed the hardness of the cocoon to the proteins rather
than to the ash; however, the hardness of cocoons of a
limacodid in the Semyra coarctata complex (Figures 20-25)
may be due more to ash, as indicated by the crystals both on the
external surface and in the layers seen in cross section of the
cocoon (Figures 22, 23).
Prepupae in some limacodid genera incorporate deciduous
spines (e.g., Acharia (= Sibine)) or tubercles (e.g., Phobetron)
into the cocoon.
ADULTS.?The biology of the adult stage in limacodid-group
species is poorly known. Adults have a tentiform posture, with
wings held below the dorsal plane of the body (Figures 26,
28-30). Dalcerids and limacodids often position their heavily
pilose front and middle legs in front of the head (Figures 29,
32). Male Limacodidae exhibit several unusual postures
(Figures 31-36), which apparently do not occur in other
limacodid-group families. These postures may aid in crypsis
and provide visual cues for mating, or they may be important in
receiving or broadcasting pheromones. Some prop themselves
up on their hindlegs and wings, with their bowed-out forelegs
resting on their midlegs (Semyra coarctata complex) (pers.
obs.) or with forelegs raised (Perola spp., Figures 31, 34) (K.B.
Sandved, pers. comm.). Males of Scopelodes hang upside down
in a similar posture, except the long labial palpi protrude rather
than the forelegs (Figures 35,36) (K.B. Sandved, pers. comm.).
It also is common for male limacodids to curl the abdomen
cephalad above their bodies (Figure 30).
Adult feeding in the limacodid group has not been reported
until recently. In the laboratory, Podalia bolivari (Heylaerts)
will feed on a solution of water and honey absorbed on cotton
(Miller et al., 1995). I found that some limacodid species drink
liquid in the laboratory: a female Apoda biguttata lowered its
head into a sugar-water droplet containing 1 % methylene blue;
later dissected, it had a stained esophagus. Perhaps feeding
occurs in most of the limacodid-group families, as galeae have
been found in all families except Aididae and Somabrachyidae.
EGGS.?Within the limacodid group, only the Limacodidae
are known to have scale-like eggs (i.e., dorsoventrally
flattened), laid either individually or in overlapping clusters
(Figures 39, 40). A few limacodid species cover their eggs
with scales (e.g., Monoleuca semifascia and Pseudonapaea
trigona (Turner)) (Dyar, 1914; N. McFarland, pers. comm.).
Limacodid eggs are similar to those of some zygaenids,
including Lactura (Peterson, 1967; Common, 1990) and
Zygaena (Tremewan, 1985). Preliminary observations on the
chorion in the Semyra coarctata complex (Limacodidae),
however, suggest it is much thinner than reported for Zygaena
trifolii (Zygaenidae) (Fehrenbach, 1995), an indication that its
chorion is perhaps nonlamellar (J. Regier, pers. comm.). The
eggs of the zygaenid Harrisina americana (Procridinae) are
ovoid and are laid in nonoverlapping groups (Peterson, 1967),
similar to those of megalopygids and dalcerids.
Megalopygid and dalcerid eggs are covered during oviposi-
tion. Megalopygid eggs are blanketed with piliform scales from
a pouch on the end of the abdomen (Figure 37). Dalcerid eggs
are coated with a fast-drying fluid secreted by accessory glands
(Miller, 1993); a number of flattened scales from the female are
attached (Figure 38).
Morphology
This section reviews the literature on the morphologies of
adults and of immature stages in the limacodid-group families
and contributes new findings as well. Comparative information
on the families Zygaenidae, Epipyropidae, and Cyclotornidae
is presented when it is pertinent to homology. New World
limacodid generic complexes cited herein and their constituent
genera are listed in Table 3.
TABLE 3.?New World limacodid generic complexes cited herein and their
constituent genera. Nomenclature for Neotropical genera follows Becker and
Epstein, 1995.
Generic complex Genera included
Apoda Apoda, Heterogenea, Lithacodes, Packardia, Tortri-
cidia
Phobetron Alarodia, Euphobetron (in part), Heuretes, Isochaetes,
Leucophobetron, Microphobetron, Phobetron,
Vipsophobetron
Prolimacodes Dichromapterix, Euphobetron (in part), Prolimacodes,
Semyra, Tanadema, Venadicodia
Perola Blazia, Epiperola, Paleophobetron, Perola
Natada Euprosterna, Isa, Narosopsis, Natada, Platyprosterna
Parasa Acharia (= Sibine), Adoneta, Euclea, Monoleuca,
Paraclea, Parasa, Talima, Zaparasa
Vipsania Pseudovipsania, Ulamia, Vispsania
Miresa Miresa (near Parasa and Vipsania complexes)
Unplaced Hepialopsis (in Phobetron or Prolimacodes complex),
Cryptophobetron
NUMBER 582 11
29
FIGURES 26-30.?Postures in adults (26-29, photos by K. Sandved): 26, Megalopyge sp. (Megalopyginae).
Costa Rica; 27, closeup of forewing of Megalopyge salebrosa (Clemens). Mexico; 28, Norape sp. (Trosiinae).
Colombia; 29, Minacraga nr. disconitens (Dalceridae), Venezuela (det. S.E. Miller); 30, Pseudanapaea trigona
(Limacodidae), Australia (photo by N. McFarland).
12
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 31-36.?Postures in adult male Limacodidae (photos by K. Sandved): 31, Perola sp., Amazonas; 32,
Euclea distrahens Dyar, Costa Rica; 33, Euphobetron cypris (Grote), Costa Rica; 34, Perola sp., Brazil, Belem;
35, 36, Scopelodes sp.. New Guinea. (Lg = leg, LP = labial palpus.)
NUMBER 582 13
FIGURES 37-40.?Eggs (scale length in parentheses): 37, Megalopyge sp. (Megalopyginae), Venezuela (500
um); 38, Dalcerina tijucana (Dalceridae), Brazil (500 um); 39, Pseudanapaea trigona, Australia (Limacodidae)
(photo by N. McFarland); 40, egg cluster of Euclea delphinii (Limacodidae), USA, Maryland.
ADULTS
Members of the limacodid group are frequently woolly and
have stout bodies and bipectinate male antennae (sometimes
also bipectinate in females). They are small to medium large,
with forewing lengths (base to apex) ranging from 4 mm to
- 4 0 mm. This general habitus, along with the often spiny
larvae, led early workers to place them in the "Bombyces."
The wings of limacodid-group species exhibit a diverse array
of ground colors, including green (in limacodids), white,
salmon, orange, reds, yellows, grey, and brown. Forewings in
megalopygids (Figures 43, 45, 46), aidids (Figure 48), and
limacodids (Figures 51, 52) commonly have distinct patterns,
such as medial crescents or postmedial bands or spots.
Dalcerids (Figure 49) tend to be less patterned, often having
only vague "water marks." Hindwings in the group tend to be
less distinctly patterned and may match or contrast the ground
color of the forewings.
Sexual dimorphism of the wings also occurs. The most
extreme cases are found in species that have smaller, partially
clear-winged males. Examples include Phobetron species
(Limacodidae) and Podalia bolivari (Megalopygidae) (Miller
etal., 1995).
Scales on the wings and body range from piliform to dentate.
Dentate scales are found in dalcerids (Miller, 1994) and
limacodids. These are deeply divided in megalopygids (Khalaf,
1984), producing a woolly appearance (Figure 27). Spatulate
scales occur in aidids (Dyar, 1895c).
HEAD.?Although conspicuous ocelli are found in the
Zygaenidae (Common, 1990), they are absent throughout the
limacodid group. Jordan (1928) and Hopp (1934) described the
presence of chaetosemata in the Megalopygidae and Aididae.
In Psycharium (Somabrachyidae), the chaetosema forms "a
belt extending from side to side," similar to that of Anomoeotes
(Phaudinae, Zygaenidae) (Jordan, 1928:135). Chaetosemata
are absent in the Limacodidae (Common, 1970), Dalceridae
(Miller, 1994), Epipyropidae, and Cyclotomidae (Common,
1990).
FIGURES 41-46 (above).?Adult specimens of the limacodid group and other
Zygaenoidea (male unless otherwise indicated) (photos by V. Krantz)
(forewing length in parentheses): 41, Pyromorpha dyari (Jordan) (Zygaenidae),
USA, Arizona (11 mm); 42, Fulgoroides exigua (Epipyropidae), USA,
Maryland (5 mm); 43, Trosia sp. (Trosiinae) Colombia (19 mm); 44, Norape
cretata (Trosiinae), USA, Maryland (13 mm); 45, 46, Megalopyge salebrosa
(Megalopyginae), Mexico: 45, male (23 mm); 46, female (30 mm).
FIGURES 47-52 (below).?Adult specimens of the limacodid group (photos by
V. Krantz) (forewing length in parentheses): 47, Somabrachys sp. (Som-
abrachyidae), Algeria (12 mm); 48, Aidos amanda (Aididae), Venezuela (17
mm); 49, Dalcera abrasa (Dalceridae), Venezuela (18 mm); 50, Strigivenifera
venata (Chrysopolominae), Kenya (23 mm); 51, Semyra coarctata complex
(Limacodidae), Venezuela (9 mm); 52, Euprosterna elaea (Druce)
(Limacodidae), Venezuela (9 mm).
NUMBER 582 15
54a
56b
56a
FIGURES 53-56.?Labial palpi (male unless otherwise indicated): 53a, Norape corporalis (Trosiinae),
Venezuela; 53b, Podalia sp. (Megalopyginae), Venezuela; 54a, Aidos amanda (Aididae), Venezuela; 546,
Brachycodilla carmen, Brazil; 55, Acraga coa (Dalceridae), Mexico; 56a, Vipsania rosabella (Limacodidae)
female, Costa Rica; 566, Phobetron pithecium (Limacodidae), USA, North Carolina
Clubbed male antennae, common in the Zygaenidae, do not
occur in the limacodid group. Male antennae are bipectinate for
their entire length in most of the Megalopygidae (except
Zyzypyge; Hopp, 1934) (Figures 43-45), Somabrachyidae
(Figure 47), and Dalceridae (Figure 49) but are bipectinate
only in the basal two-thirds to one-third in Aididae (Figure 48).
A complete range of these bipectinate conditions occurs in
male Limacodidae (Janse, 1964), and there are unipectinate and
filiform antennae as well.
Antennae are usually sexually dimorphic in the limacodid
group, with pectinations being shorter or absent in the female
(compare Figures 45, 46), although monomorphism can occur.
Examples of broadly bipectinate antennae in both sexes include
some species in the Perola complex (Limacodidae) and Trosia
species (Megalopygidae). Species in the Apoda complex
(Limacodidae) have filiform antennae in both males and
females.
The labial palpus in the limacodid group varies in length and
in the number of segments (Figures 53-56). It can be well
developed in the Limacodidae, where it is usually three-
16 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
_ , \
?
MxP
A
FIGURES 57-59.?Galeae (haustellum) and maxillary palpi in male Dalceridae and Limacodidae (scale length in
parentheses). 51a-c Dalcera abrasa (Dalceridae) (courtesy of S.E. Miller): 57a, head, labial palpi, and location
of galeae and maxillary palpi (arrow) (0.33 mm); 51b, galeae and maxillary palpi in a (see arrow) (120 |im); 57c.
detail of galea and maxillary palpus in a (see arrow) (43 \un). 58, 59, galeae and maxillary palpus in
Limacodidae: 58, Parasa chloris (30 |im); 59, Euclea delphinii (holes in maxilla are from missing scales) (30
Urn). (Ga = galea, LP = labial palpus, MxP = maxillary palpus, Rl = sensillum.)
segmented (the second segment often more than twice the
length of the other segments), upturned, and commonly
reaching the vertex or beyond. The third segment, although
frequently short (Figure 56b), is quite long in Scopelodes
(Figures 35, 36) and Vipsania (Figure 56a). The Dalceridae
lack segment three of the labial palpus (Figures 55, 57a)
(Miller, 1994). The Megalopygidae can have two short
segments (Figure 53a) (Philpott, 1926), as in Zygaenidae and
Epipyropidae, or only one segment (Figure 53/?). The Aididae,
with either two or three segments (Figures 5Aa,b), generally
have a longer labial palpus than do the Megalopygidae.
In contrast to the well-developed proboscis in the Zy-
gaenidae, the galeae, or haustellum, in limacodid-group species
are often vestigial or absent (Forbes, 1923; Common, 1990). In
both the Limacodidae and Megalopygidae, galeae may be
present or absent, sometimes within the same generic complex.
Some of the Limacodidae have a narrow, coiled haustellum
equal in length to the labial palpus, whereas in others the
haustellum is the length of the first segment of the labial palpus
(Figure 58) or shorter (Figure 59). Limacodid galeae are
relatively smooth compared with those found in dalcerids, and
they have Rl sensilla in the apical portion (sensilla homology
follows Miller, 1991). Galeae of trosiine megalopygids (e.g.,
Trosia, Mesocia) are visible at low magnification of a
dissecting microscope, whereas in megalopygines they are
absent (e.g., Megalopyge crispata, see Philpott, 1926). Dal-
cerids have short galeae with dense spinules at the base
(Figures 57a-c). The apical portion of the galea with Rl
NUMBER 582 17
sensilla is incorrectly identified as the maxillary palpus by
Miller (1994). Galeae are absent in aidids and somabrachyids.
The Limacodidae can have a maxillary palpus of up to three
segments (Philpott, 1926), although it also can be vestigial or
absent. In Euclea delphinii the maxillary palpus is highly
reduced rather than absent (Figure 59) (Epstein, 1988) as
previously reported by Philpott (1926). In megalopygids it is
either one segment and visible (e.g., Trosia, Mesocia) or absent
(M. crispata, see Philpott, 1926). The maxillary palpus is
vestigial in dalcerids (Figures 51a-c) (= galea in Miller, 1994)
and is absent in aidids.
THORAX.?Hopp (1934:1071) reported a "deep longitudinal
median fold" in the anterior portion of the mesonotum of
Megalopygidae (+Aididae), and Jordan (1928) noted addi-
tional lateral depressions (Figures 60, 61). Aidids have long
medial and lateral depressions (Figure 61). Deep medial folds
were also reported in the Zygaenidae (Jordan, 1928). These
folds are not present in the Limacodidae or Dalceridae (Figures
62, 63), although carinate ridges in the former may be vestiges
of lateral folds.
The degree of sclerotization of the mesepimeron appears to
differ among families. The mesepimeron is more narrowly
60 61
62 63
FIGURES 60-63.?Mesonotum in dorsal view: 60, Podalia sp. (Megalopyginae), Venezuela; 61, Aidos amanda
(Aididae), Venezuela; 62, Dalcerides ingenita (Dalceridae), USA, Arizona; 63, Acharia sp. (Limacodidae),
Venezuela.
18
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
64 65
67
66
FIGURES 64-67.?Mesepimeron in lateral view, anterior to left: 64, Phobetron pithecium (Limacodidae); 65,
Acraga coa (Dalceridae); 66, Aidos amanda (Aididae); 67, Podalia sp. (Megalopyginae). (Epm = mesepimeron.)
sclerotized in the Limacodidae and Dalceridae (Figures 64,65)
than it is in the Aididae and Megalopygidae (Figures 66, 67).
Shepard (1930) noted similarities between the precoxal and
pleura! sutures in megalopygids and procridines (Zygaenidae).
No tympanum is present in the Zygaenoidea.
Venation: Forewing R 2 5 or R3_5 are usually stalked
beyond the discal cell in members of the limacodid group. In
contrast, those of zygaenids are usually not stalked or only one
pair (e.g., R3 and R4) is stalked. Each limacodid-group family
has a distinct forewing venation pattern. Both the Mega-
lopygidae (Figures 68, 70) and Dalceridae (Figures 72, 74)
have R4 and Rs stalked; the Dalceridae also have R2 and R3
stalked, although fusions can occur in either stalk (Figure 74)
(Miller, 1994). The Aididae (Figure 69) and Limacodidae
(Figures 71, 73) have R3 stalked with R4, although aidids are
unique in having R. and R2 stalked independently as well
NUMBER 582 19
R2
Rs
Mi
M2
M3
CuAi
CuA2
71
FIGURES 68-71.?Wing venation: 68, Norape corporalis (Trosiinae), Venezuela; 69, Brachycodilla carmen
(Aididae), Brazil; 70, Podalia bolivari (Megalopyginae); 71, Pantoctaenia (Limacodidae) (after Janse, 1964).
(lA-3A = anal veins, Cu A, _2 = anterior cubital veins, CuP = posterior cubital vein, M, .,= medial veins,
R 1 5 = radial veins, Rs = radial sector vein, Sc = subcostal vein.)
20
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
73
R2+3
FIGURES 72-75.?Wing venation: 72, forewing of Minacraga disconitens (Dalceridae), French Guiana; 73,
Acharia stimulea (Clemens) (Limacodidae); 74, forewing of Dalcerides sofia (Dalceridae), Mexico; 75,
hindwing of Dakerina tijucana (Dalceridae), Brazil. (AC = accessory cell, M, = medial vein, R 2 3 = radial veins,
Rs = radial sector vein.)
(Figure 69) (Dyar, 1895). In addition, aidids and limacodids
have R3 and R4 stalked off R5, but the Limacodidae
(+ Chrysopolomidae) may also have the two veins stalked off
R2 (or independent) or stalked between R2 and R5. Some
limacodids have an R5 that is split back from the radial sector
and arises from the discal cell (Brock, 1971; Holloway, 1986).
Jordan (1916:355) noted the presence of only four R (= Sc) in
Somabrachys (Somabrachyidae).
Medial veins in the forewing discal cell can be either present
or absent (Comstock, 1918). Jordan (1928) and Hopp (1934)
reported their absence in Megalopygidae (+ Aididae); however,
this is not always the case, as illustrated by Norape and Podalia
(Figures 68, 70), and by Brachycodilla (Aididae) (Figure 69).
The Dalceridae (Figure 74), Epipyropidae, Chrysopolominae
(sensu nova) (Hering, 1937), and Zygaenidae (Common, 1990)
sometimes have a radial vein (chorda) in the discal cell that
forms an areole or accessory cell. Forewing anal veins are fused
from near the base in the Dalceridae (1A + 2A) (Figure 72)
(Miller, 1994) and Limacodidae (Figure 73). In the Aididae
(Figure 69) and Megalopygidae (Comstock, 1893), 2A may
diverge from 1 A.
The frenulum occurs throughout the limacodid group, but it
is lost in the Chrysopolominae (sensu nova) (Hering, 1937),
three genera of Dalcerinae (Miller, 1994), and in some
Megalopyginae (e.g., Megalopyge basalis (Walker)). Within
Megalopyginae, the retinaculum is lost in Megalopyge (even in
species that retain the frenulum) but is present in Podalia
(Hopp, 1934).
NUMBER 582 21
Hindwing Sc + R, anastomoses with Rs for almost the entire
length of the discal cell in the Trosiinae (Figure 68), female
Megalopyginae (Hopp, 1934), Somabrachyidae (H. Geertsema,
pers. comm.), and Zygaenidae (Jordan, 1928). Sc + R, is free
from Rs or slightly fused near its base in the Aididae (Figure
69) (Dyar, 1895c), some Dalceridae (Figure 75) (Miller, 1994),
male Megalopyginae (Figure 70), and Limacodidae (Figures
71, 73). In the limacodid-group families Dalceridae, Mega-
lopygidae, Aididae, and in some Limacodidae (e.g., Apoda
complex, Pantoctaenia) (Comstock, 1893), hindwing Rs and
M, are separated by a perpendicular or oblique cross vein at the
end of the discal cell (Figures 68-71, 75). Limacodidae is
unique, however, in having Rs and M, stalked beyond the
discal cell (Figure 73) in most of the genera.
Legs: The epiphysis on the foretibia is present in some of
the Zygaenidae (e.g., Zygaeninae and Lactura) (Common,
1990), Heterogynidae (Scoble, 1992), and Somabrachyidae
(e.g., Somabrachys, see Jordan, 1916; Hampson, 1920;
Geertsema, pers. comm.) but is absent in the remainder of the
limacodid group. The tibial spur formula for the Mega-
lopygidae and Aididae is 0-2-2, whereas in Limacodidae and
Zygaenidae it is either 0-2-2 or 0-2-4 (Janse, 1964; Common,
1990), and in Cyclotornidae it is 0-2-4 (Common, 1990). Tibial
spurs are absent in the Dalceridae (Miller, 1994) and
Epipyropidae (Common, 1990).
The fifth tarsomere is sexually dimorphic in most species of
the limacodid group. There are dense clusters of sensilla
trichodea on the ventral female tarsomere, usually with a
concurrent reduction of scales (Figures 76-84); however,
neither trait is found in males (Figures 87, 88). In female
Norape (Trosiinae), the tarsomere has a narrow, medial group
of scales with sensilla on each side (Figures 76, 77), whereas
female Megalopyge have more medial scales (Figure 79). The
tarsomere lacks scales among the sensilla in female aidids
(Figure 82) and limacodids (+ Chrysopolomidae). Female
dalcerids and some female limacodids (e.g., Pantoctaenia and
Crothaema) have fewer sensilla, which are found in a narrow,
distal band, and more numerous scales (Figures 85,86,91-95).
Other Limacodidae have relatively short sensilla recessed in an
ovate or quadrate pad, which is truncate on the distal end
(Figures 96-103).
Female dalcerids have a large spur on each of the third
through the fifth tarsomeres, of equivalent length to the
surrounding scales (Figures 85, 89, 90).
ABDOMEN.?The Zygaenoidea have a tortricid-type sternum
II (Kyrki, 1983). Spinose tergites have been reported in the
Megalopygidae (+Aidinae) (Hopp, 1934) and in Boisduval-
odes. The latter was recently placed in Limacodidae (Viette,
1980); however, it shares a number of potential adult apo-
morphies with the Somabrachyidae (Geertsema, pers. comm.).
MALE GENITALIA.?The variable nature of the male genita-
lia throughout the limacodid group makes characters associated
with them difficult to describe and homologize. External male
genitalia have been described in the Megalopygidae (Eyer,
1924; Hopp, 1927, 1934), Limacodidae (Janse, 1964; Hol-
loway 1986, 1987), Dalceridae (Eyer, 1924; Miller, 1994),
Epipyropidae (Jordan, 1928; Heinrich, 1931; Common, 1990),
and Cyclotornidae (Common, 1990). The muscles of the male
genitalia in the Limacodidae and Zygaenidae were discussed by
Kuznetsov and Stekol'nikov (1981), and those in Podalia
albescens Schaus (Megalopygidae) by Oiticica Filho (1946).
The uncus in the Megalopygidae is narrow and hooked
(Figure 351), as in the Zygaenidae (Alberti, 1954). In the
Aididae and Limacodidae, the uncus is larger and broader at its
base where it connects to the tegumen (Figures 352, 356, 357),
as it is in Somabrachys (Somabrachyidae) (Jordan, 1916). In
the Dalceridae the uncus is highly reduced and has socii
(Figures 354, 355); its function is perhaps replaced by the
gnathos (Miller, 1994). The vinculum is much wider in the
Megalopygidae and Aididae (Figures 351, 352) than it is in the
Dalceridae and Limacodidae (Figures 354, 356). The saccus is
well developed in the Dalceridae (Figures 354, 355) and
Limacodidae (Figure 356), in dalcerids frequently being
one-fourth or more the length of the genitalia (Miller, 1994).
The gnathos was considered by Common (1975) to be
independently derived in different groups of Lepidoptera, even
within superfamilies. It is simple and lance-shaped in most of
the Limacodidae (Figures 356, 357) (Janse, 1964; Holloway,
1986), whereas in the Dalceridae it is sometimes more divided
and elaborate (Miller, 1994). Other Zygaenoidea purportedly
with a gnathos include the Epipyropidae (Heinrich, 1931) and
Cyclotornidae (Common, 1990). The gnathos is absent in the
Aididae, Somabrachyidae (Jordan, 1916; Geertsema, pers.
comm.), and most of the Megalopygidae; Hopp (1927, fig. 12)
reports a gnathos in the Trosiinae (e.g., Microcladia pygmaea).
Species of Chrysopolominae (sensu nova) have paired transtil-
lar lobes located in a position similar to that of the gnathos
(Hering, 1937).
The paired valvae are symmetrical, wide at the base, and
entire in the Aididae (Figure 352), Somabrachyidae (Jordan,
1916; Geertsema, pers. comm.), and most Limacodidae
(Figures 356, 357), similar to those of Procridinae (Zy-
gaenidae) (Tarmann, 1984). In some limacodids, the valvae are
apically emarginate or are deeply split (e.g., Heuretes, see
Epstein and Miller, 1990, figs. 6, 8). The valvae are extremely
reduced in dalcerids (Figures 354, 355) (Miller, 1994). Eyer
(1924) and Hopp (1927, 1930) considered the valva in megalo-
pygids to be divided into a dorsal harpe (= cucullus of Eyer)
and a ventral sacculus. I treat the dorsal lobe as homologous to
the valva, and the ventral lobe as homologous to the medial
juxtal lobes found in the Aididae (compare Figures 351 and
352; see "Megalopygidae" for a discussion of the homology of
these lobes). Processes on the valva are rare, but they do occur
in the Limacodidae, sometimes in conjunction with what may
be androconial pouches (e.g.. New World Natada).
The region between the valva and the aedeagus presents a
number of homology problems (Miller, 1994). Various lobes
and processes can occur either dorsal or ventral to the aedeagus
22
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
pIsi
82
FIGURES 76-84.?Female 5th tarsomere and pretarsus of Megalopygidae and Aididae (scale length in paren-
theses). 76-78. Norape sp. (Trosiinae): 76, ventral view of 4th and 5th tarsomere (200 nm); 77, 5th tarsomere
(100 nm); 78, sensilla trichodea (20 nm). 79-81, Megalopyge crispata (Megalopyginae): 79, 5th tarsomere (200
Urn); 80, sensilla trichodea (20 Jim); 81, sensilla trichodea (10 \un). 82-84, Aidos amanda (Aididae): 82, 5th
tarsomere and pretarsus (200 urn); 83, sensilla trichodea (50 ^m); 84, sensilla trichodea (20 Jim).
NUMBER 582 23
?PP.;
11 ill
]
M
mmfJmm
A
FIGURES 85-95.?Fifth tarsomere and pretarsus of Dalcendae and primitive Limacodidae (scale length in
parentheses). 85-92, Dalcera abrasa (Dalcendae): 85, female pretarsus and 5th tarsomere (upper arrow points to
location of sensilla tnchodea; lower arrow points to tarsal spur) (200 urn); 86, sensilla tnchodea on distal portion
of 5th tarsomere in 85 (see upper arrow) (38 urn); 87, male pretarsus and 5th tarsomere (arrow points to distal
portion of 5th tarsomere) (200 urn); 88, detail of male pretarsus and distal portion of 5th tarsomere in 87 (see
arrow; note absence of sensilla trichodea) (50 |im); 89, tarsal spur on female 5th tarsomere in 85 (see lower
arrow) (43 nm); 90, detail of female tarsal spur (10 um); 91, female sensilla trichodea (10 |im); 92, detail of
female sensilla trichodea (4 um). 93, 94, female Pantoctaenia prasina (Limacodidae): 93, pretarsus and 5th
tarsomere (arrow points to location of sensilla trichodea) (176 \im); 94, detail of sensilla trichodea in 93 (see
arrow) (38 urn). 95, pretarsus and 5th tarsomere of female Crothaema sp. (Limacodidae) (150 urn).
24
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
^ - ? ,
mi
gypgg
FIGURES 96-103.?Female 5th tarsomere and pretarsus of Chrysopolominae and Limacodidae (scale length in
parentheses). 96-98. Strigivenifera venata: 96, pretarsus and 4th and 5th tarsomeres (500 |xm); 97,5th tarsomere
(200 ^ ) ; 98. detail of sensilla trichodea (20 Jim). 99-101, Semyra coarctata complex (Limacodidae): 99,
pretarsus and 5th tarsomere (200 um); 100, 5th tarsomere (100 urn); 101, sensilla trichodea (note spinules) (20
Urn). 102. 103, Apoda biguttata (Limacodidae): 102, pretarsus and 5th tarsomere (150 \xm); 103, 5th tarsomere
(note long spinules) (50 um).
NUMBER 582 25
and the anellus surrounding it. Dorsal in the Limacodidae
(+ Chrysopolomidae), the transtilla can be simple, a membra-
nous hood, or setose processes (Figure 356), and it is
sometimes broadly connected to the anellus. Ventral to the
aedeagus, the juxta can be composed of a simple plate, paired
processes arising from a plate (Limacodidae; Figure 357), or it
can be entirely lobes (Aididae; Figures 352, 353) or processes
(Dalceridae; Figures 354, 355).
The aedeagus generally lies parallel to the axis of the
abdomen or is only slightly curved; however, in the Limacodi-
dae it is often strongly bent or curved at the base (Holloway,
1986). The Megalopygidae and Zygaenidae (e.g., Procridinae,
see Tarmann, 1984) have a broad aedeagus with one or several
large nondeciduous comuti. The Dalceridae (Miller, 1994) and
Limacodidae have smaller cornuti in patches or have cornuti
absent. In the Limacodidae, the apex of the aedeagus exhibits a
wide diversity of shapes and processes.
FEMALE GENITALIA.?Most of the families in the limacodid
group have open, disk-shaped papillae anales (ovipositor
lobes). In the Limacodidae (+ Chrysopolomidae), the shape of
the lateral margins varies from convex (Figure 362) to concave
(Figure 363). The Aididae and Dalceridae have the papillae
divided equally into dorsal and ventral lobes (Figures 347,
349), although the dorsal lobes are medially fused in aidids.
The ovipositor lobes of megalopygids are medially ap-
pressed and laterally flattened (Figures 345, 346) (Holloway,
1986). This is perhaps due to the presence of a dense clump of
deciduous piliform scales near the end of the abdomen. Used to
cover oviposited eggs, the scales are attached to a membranous
A7, which is positioned posteriorly and extends to the middle
of A8 (Figure 346).
Other external lobes may be located anteriorly on A8 near
the papillae anales, either dorsally or ventrally. These may be
sclerotized with intersegmental pouches or may be membra-
nous. Small membranous digitate or pedunculate lateral lobes
are present on the eighth segment in many limacodids (Figure
362) (Holloway, 1986; Epstein, 1988) and in chrysopolomines
(Figures 363, 364).
Both anterior and posterior apophyses are present in the
Megalopygidae (Figure 346), Aididae (Figure 348), and
Limacodidae (+Chrysopolomidae) (Figures 362-364). Ante-
rior apophyses are absent in the Dalceridae (Figure 350)
(Miller, 1994).
The bursa copulatrix and its components are quite variable
within the limacodid group. This variability includes the length
and shape of the corpus and ductus bursae relative to each other
and to the spermatheca. In megalopygids the bursa copulatrix
may be shorter than segment A8 (Figure 346), whereas a coiled
ductus bursae may be either present or absent in limacodids
(Holloway, 1986). In the limacodid group, signa have been
reported or observed only in Limacodidae thus far.
In addition to sebaceous glands, large "secondary accessory
glands" (Petersen, 1900) occur in the Zygaenoidea (Miller,
1993). The term "Petersen's glands" has been applied to these
glands in Zygaena and in other zygaenine genera (Bode and
Naumann, 1988; Naumann, 1988). Naumann (1988) pointed
out structural and histological differences between the glands
of the Zygaeninae and the Procridinae, although he acknowl-
edged possible homology.
Secondary accessory glands occur in other zygaenoid groups
including the Dalceridae (Figure 350) (Miller, 1993), Aididae
(Figure 348) (Epstein, 1988), Epipyropidae, and Cyclotomidae
(Common, 1990) (see further discussion under each family,
below). The glands in dalcerids consist of two independent,
pitted, sclerotized, and convoluted lobes. Occupying much of
the internal space of A8, each lobe has an external duct leading
to the cleft between the upper and lower portions of the papillae
anales (Miller, 1993). The glands in aidids differ by being less
sclerotized, being broadly connected at the median, and in not
having external ducts. Examination of the Limacodidae
(Figures 362-364) (see also Naumann, 1988) and Mega-
lopygidae (Figure 346) has not revealed secondary accessory
glands in either family.
LARVAE
At rest or in motion, larvae of the limacodid group are
characterized by having a greater proportion of their ventral
surface area in contact with the substrate than zygaenid larvae
do. Reasons for this include (1) the presence of prolegs on
segments A2-A7; (2) prolegs that are short relative to the
remainder of the ventral surface; (3) the narrowness of the
ventral portions of nonproleg segments A8 and A9 (described
for megalopygids by Stehr, 1987a); (4) a reduction in the size
of the thoracic legs; and (5) the absence of numerous long
secondary setae over the lateral and ventral surfaces (excluding
megalopygids and somabrachyids).
The larval dorsum exhibits a diversity of conditions in the
limacodid group (Figures 1-6). Many species have rows of
verrucae, tubercles (scoli), or warts, although these characters
are not always homologous (discussed under "Homology and
Evolution of Larval Body Setae," below) (see definitions of
setal derivatives under "Dorsal and Lateral Body Surfaces,"
below). In megalopygids and aidids, the cuticle on the thorax
and abdomen is shagreened or spinulose throughout except for
the membranous pads on megalopygid prolegs (see "Ventral
Surface of Body," below). The dorsum and ventrum are mostly
smooth in dalcerids and in some limacodids (Figure 7),
excluding the anal proleg and the region surrounding the
abdominal spiracles.
Limacodid larvae have been termed heteromorphic (= hav-
ing two or more forms) because the first instar usually differs
from the remaining instars in types of setae or in degree of
spinosity (Common, 1990). This is exemplified by species
having late instars that are smooth-backed and nonurticating.
These species have first instars with bisetose tubercles or warts
that are lost in subsequent instars (e.g., Semyra coarctata
complex; compare Figure 170 with Figures 7-9) (Dyar,
1899a). I refer to such change in limacodids simply as
26 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
"ontogenetic" because it is not accompanied by major shifts in
biology, unlike the parasitic zygaenoid families Epipyropidae
and Cyclotornidae, which have hetero- or hypomorphic larvae
(Davis, 1987; Common, 1990). Other examples of ontogenetic
change include the spinneret in the Limacodidae and the
appearance of crochets in post-first instar Dalceridae (Stehr and
McFarland, 1985) and Pantoctaenia (Limacodidae) (described
in "Limacodidae," below).
HEAD.?The dorsum of the head is difficult to observe in
preserved larvae because the head is often retracted (Figures
104, 106). Thus, it is best described either in newly hatched
larvae, which have a fully extended head (Figures 105,107), or
by removing the head of preserved specimens. The frontoclyp-
eus is indistinct and the epicranial notch may be raised, perhaps
to keep the head retracted (Figure 105). Spinules at the base of
the head may also serve this function (Figures 107, 108).
Long antennae (Figures 111-116) and a tight arrangement
of the stemmata in the Zygaenoidea may be related to the
retractile head (Stehr, 1987b; Common, 1990); however, there
appears to be variation in stemmata arrangement in both the
Zygaenidae and the limacodid group. Stemma 5 is far removed
from stemma 4 and is proximate to the antenna in the
Somabrachyidae (Figure 118), Megalopygidae (Figure 119),
and Aididae (Figures 120, 124). The same is true of Harrisina
americana (GueYin-M6neville) (Zygaenidae: Procridinae)
(Stehr, 1987c), but in Zygaena trifolii (Zygaenidae: Zygaen-
inae) stemma 5 is not disjunct (Figure 117). The Aididae have
a gap of approximately the width of a stemma between
stemmata 1 and 2 (Figures 120, 124). In the Limacodidae
(+ Chrysopolomidae) there is a similar-width gap between
stemmata 4 and 5 (Figures 125-128), which is somewhat
wider in the Dalceridae (Figure 122).
The Zygaenidae, Megalopygidae, Aididae, and Psycharium
(Somabrachyidae) each have two S setae, SI and S2, encircled
by stemmata (Figures 117-121). One specimen of Aidos sp.
was found to have only one stemmatal seta pit on the left side
(Figure 124), whereas it had two on the right side (Figure 120).
The Limacodidae (+Chrysopolomidae) and Dalceridae have
only one S seta, probably SI because it is close to stemmata 3
and 4 (Figures 122,123,125-128) (based on location given by
Stehr, 1987b).
Epipharyngeal spinules in Lepidoptera are usually most
abundant beneath the labrum (Figure 130), with few on the
margins. This is the case in megalopygids and aidids (Figures
129-133) and in zygaenids. In many limacodids, multiple
hair-like spinules occur on the anterior and lateral margins of
the labrum as well (Figures 138-141). A similar condition has
been reported in gracilleriids. tineids and psychids (Davis et al.,
1991), and in phycitine pyralids (Leidy and Neunzig, 1989).
The Dalceridae (Figure 134) and Crothaema (Limacodidae)
have the anterior spinules exclusively on the anteromedial
portion of the labrum. Larger and fewer anterior spinules are
found in the Aididae (Figure 129) and Megalopygidae (Figure
133). In megalopygids these may be curved under the labrum
(Figure 131).
The spinneret in the Zygaenidae, Megalopygidae, and
Somabrachyidae is of the tubular type found in most
Lepidoptera (Figures 142-147). Aidids have a spinneret with
an irregularly shaped tip (Figures 148-151), which may assist
in constructing the unusual holes in their cocoons (Figures 15,
17) (Epstein, 1995). Spinnerets in dalcerids (Figures 152-156)
and limacodids (Figures 158-162) are usually brush-like, with
the apex wider than the base. This general shape was described
for Apoda (Limacodidae) by Chapman (1894) and Christensen
(1950). The dalcerid spinneret shows little ontogenetic change
except that in late instars the silk pore, not visible in early
instars, appears as a longitudinal slit, and the brush-like anterior
margin becomes deeply and evenly cleft (Figure 154) (e.g.,
Acraga infusa complex and Dalcerina tijucana). Spinnerets in
the Limacodidae, on the other hand, undergo extensive
ontogenetic change (see examples and discussion in "Ontogeny
and Function of Limacodid Spinnerets").
The ventral surface of the spinneret in the limacodid group is
divided into three longitudinal bands (Figures 150, 163), with
the median part overlying the silk gland channel (the dalcerid
silk gland is shown in Figure 157). This occurs whether the
spinneret is tubular or brush-like above and also is typical of
zygaenids (Figure 143) and most other Lepidoptera (Forbes,
1910).
DORSAL AND LATERAL BODY SURFACES.?Body setae in
the limacodid group occur in a wide variety of derivative
conditions in addition to the common tactile hair-like condi-
tion. These setae may remain unchanged or may undergo
changes to derivative forms ontogenetically. Setae are referred
to as primary if they appear to correspond to the chaetotaxy
found in most Lepidoptera during all instars and are referred to
as secondary if they do not (Stehr, 1987b). Examples of setal
derivatives herein include tubercles, scoli, verrucae, cuticular
warts, and buttons. Tubercles refer to elongate, fleshy
(= dilated) setae, often bearing several secondary setae in early
instars, and if they persist in late instars they are hairy without
urticating setae. Scoli are elongate, fleshy setae that bear
urticating setae (= spines). Verrucae are short, wart-like
structures that are covered with urticating setae and/or plumose
setae. The term tubercle will often be used rather than scolus or
verruca when referring to early instars of species with spiny
larvae because it cannot be determined whether secondary setae
are urticating at this stage. Cuticular warts are naked verrucae
or verrucae bearing primary or a few secondary setae. Buttons
appear to be tubercles that have undergone reduction. Fungi-
form setae are tiny dome-shaped setae usually found on the
ventral surface but which also occur laterad near spiracles.
Gelatinous warts refer to the covering that is secreted by a
primary seta rather than to the seta itself.
In most limacodid-group families, the prothoracic shield has
only primary setae, allowing it to be somewhat retractile into
the mesothorax. Megalopygids differ in having a prothoracic
shield with verrucae from the first instar onward, making
retraction of the prothorax difficult. These verrucae are formed
NUMBER 582 27
106 1
1
"
r
 Mi
FIGURES 104-110.?Retractile head and prothorax in 1st instar Limacodidae (scale length in parentheses): 104,
Semyra coarctata complex, frontal view (100 ^m); 105, Phobetron hipparchia, frontolateral aspect (note
epicranial notch above) (100 |im); 106, Pantoctaenia prasina, dorsal view of head partially retracted from
prothorax (200 \un). 107, 108, Phobetron hipparchia: 107, dorsal view of fully retracted head (arrow points to
epicranial notch) (200 fim); 108, detail of epicranial notch (note spinules at base of head above) (100 \im). 109,
lateroventral view of Semyra coarctata complex (100 Jim); 110, ventral view of Pantoctaenia prasina (50 \im).
(An = antenna. En = epicranial notch, MxP = maxillary palpus, Spt = spinneret, Stm = stemmata.)
28
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 111-116.?Larval head (scale length in parentheses): 111, Chrysopoloma similis (Chrysopolominae),
laterofrontal view of ultimate instar exuvia (500 |jmn); 112, Acraga infusa complex (Dalceridae), - 4th instar (1
mm): 113, Semyra coarctata complex (Limacodidae), 3rd instar (200 Jim); 114, Pantoctaenia prasina
(Limacodidae), 1st instar (100 urn); 115, Megalopyge crispata (Megalopyginae), post-third instar (500 nm);
116, Mesocia pusilla (Trosiinae), post-first instar (200 nm). (An = antenna, Lbr = labrum, Lg = Ieg,
Md = mandible, Mx = maxilla, Spt = spinneret.)
NUMBER 582 29
T 1
/
ip^si
S2
117
?
=
? * . a
I \ f
121
4/
122
^ i
> -, st4 I
? v '- .. ? r
FIGURES 117-124.?Stemmata (on left side of head unless otherwise indicated) (scale length in parentheses):
117, Zygaena Irifolii (Zygaenidae) (20 \un); 118, Psycharium sp. (Somabrachyidae) (50 u.m); 119, Megalopyge
crispata (Megalopyginae) (200 \im). 120, 121, Aidos sp. (Aididae), exuvia from ultimate instar: 120, right side,
has SI and S2 setae (note distance between stemmata 1 and 2) (ISO u.m); 121, detail of stemmata 1-4 viewed
from more frontal perspective (86 \im). 122, 123, Acraga infusa complex (Dalceridae): 122, stemma 5 disjunct
from stemma 4 (136 urn), 123, detail of stemmata 2, 3 (75 urn). 124, Aidos sp. (Aididae), arrow points to only
1 stemmatal seta pit, same individual as in 120 (176 \xm). (S1-S2 = stemmatal setae, Stl-St6 = stemmata 1-6.)
30 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 125-128 (above).?Stemmataof Chrysopolominae and Limacodidae (scale length in parentheses): 125,
Chrysopoloma similis (100 urn); 126, Semyra coarctata complex (20 urn); 127, Prolimacodes budia, detail of
stemmata2-4(15jim); 128, Pantoctaenia prasina (20 \im). (SI =stemmatal seta.)
from contiguous D and SD rows, whereas L and SV rows are
independent (Figure 166). Late instar Zygaenidae and Som-
abrachyidae frequently have a similar condition, although in
several examples, such as Zygaena trifolii (Zygaenidae) and
Psycharium (Somabrachyidae), the first instars have primary
setae (Figures 164, 165) that undergo ontogenetic change to
form verrucae.
First instar dalcerids and limacodids have Tl spiracles with
two closely associated L setae, LI and L2 (Figures 181-186).
Dalcerids sometimes have three L setae in later instars (e.g.,
Acraga infusa complex) (Figures 180, 187). Megalopygids
(Stehr, 1987a) and aidids (first instar not observed) have Tl
spiracles without proximate, hair-like setae (Figure 338).
Setae on T2, T3, and the abdomen in the limacodid group
occur in a number of different forms. Megalopygids and many
zygaenids have D, SD, L, and SV setae as verrucae (Stehr,
1987a, 1987c); however, first instar Megalopyge and perhaps
all megalopygids differ from first instar Zygaena trifolii in the
presence of verrucae rather than primary setae. In late instar
megalopygids, the D, SD, and L contain both urticating and
plumose secondary setae (Figure 169), whereas the SV
contains only plumose setae. Late instar Somabrachys larvae
have urticating and hair-like setae in all positions. Spiny
limacodids, dalcerids, and aidids have very different setal
modifications or coverings in the D and SD positions compared
to the primary tactile setae in the L and, in the first two families,
the SV positions.
In dalcerids all D and SD setae on the thoracic and
abdominal segments, except Tl and A10, have a translucent,
gelatinous covering (Figures 3, 180). On the same segments,
limacodid first instars have rows of warts or tubercles (Figures
170-176, 178, 179, 181) in the D and SD positions. These
often change form in later instars. For example, species with
smooth larvae and simple setae in late instars have first instars
with long-branched or unbranched tubercles (Figures 178, 179,
195-203) or warts, with or without setal remnants (Figures
188-192).
Limacodid first instars in spiny species have from three to
seven setae per tubercle (Figures 172, 174); these tubercles
become either scoli or verrucae in later instars (see more
detailed figures below in "Homology and Evolution of Larval
Body Setae"). Many species in the Parasa complex have
deciduous patches of defensive setae located at the bases of D
scoli on A8 and A9 (Dyar, 1899a). These are also incorporated
into the silk of the cocoon.
Limacodid species of the Phobetron complex (Figures
198-203) have tubercles different from species with smooth
larvae or species with spiny larvae. First instars have long
tubercles similar to those found in smooth larvae (e.g., Apoda
complex), but the tubercles are retained rather than converted
into simple, hair-like setae as in later instars of smooth larvae.
In several respects tubercles of late instars in the Phobetron
complex differ from scoli in spiny larvae. They are covered
with branched nonstinging setae, sometimes with urticating
setae at the base (e.g., Heuretes, Epstein and Miller, 1990).
Tubercles of Phobetron species are completely detachable in
later instars and are incorporated into the cocoon (Dyar,
1896a: 184).
Tubercles of first instar limacodids are everted in the egg and
at hatching (Figures 196, 201, 202), telescoping outward soon
after (Figures 195, 203). This has been described for Apoda
limacodes (Hufnagel) (as Limacodes testudo (Denis and
NUMBER 582 31
wft HI
FIGURES 129-133.?Larval labrum of A ididae and Megalopygidae (scale length in parentheses). 129,130, A idos
sp. (Aididae), ultimate instar exuvia: 129, dorsum (100 (im); 130, ventral aspect, showing spinules (100 \un).
131, 132, Megaiopyge crispata (Megalopyginae): 131, detail of left side, arrow points to spines that are curved
under (50 (Am); 132, labrum and mandibles (200 \im). 133, Mesocia pusilla (Trosiinae) (200 urn).
Schiffermiiler) in Chapman, 1894:346; Christensen, 1950), as
well as for other species in the Apoda and Phobetron
complexes (Dyar and Morton, 1895:152; Dyar, 1899c:204).
Zygaenids and megalopygids have L and SV verrucae on the
abdomen. In post-first instar Zygaena trifolii, Tremewan
(1985) refers to verrucae on the outer portion of the proleg base
as L3, with LI and L2 below the spiracle in that order. This
interpretation differs from Stehr (1987a, 1987c), who defines
verrucae on the outer portion of the proleg base in mega-
lopygids and zygaenids as SV setae. I follow Stehr's system
32
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 134-141.?Larval labrum of Dalceridae, Chrysopolominae, and Limacodidae (scale length in
parentheses): 134, Dalcerides ingenita (Dalceridae), post-first instar (20 \un); 135, Acraga infusa complex
(Dalceridae), late instar (100 um); 136, Panloctaenia prasina (Limacodidae), 1st instar (20 um); 137,
Chrysopoloma similis (Chrysopolominae), prepupal exuvia (200 um); 138, Phobetron hipparchia (Limacodi-
dae), 1st instar (20 um); 139, Talima postica (Limacodidae), 1st instar (20 um). 140, 141, Semyra coarctata
complex (Limacodidae): 140, 3rd instar, (100 um); 141, detail of 140 (see arrrow) (20 um).
NUMBER 582 33
FIGURES 142-151.?Larval spinneret in Zygaenidae, Somabrachyidae, Megalopygidae, and Aididae (scale
length in parentheses). 142, 143, Zygaena trifolii (Zygaenidae): 142, dorsal view (17.6 urn); 143, lateroventral
view (7.S |im). 144, Psycharium sp. (Somabrachyidae), lateroventral view (8.6 \un); 145, Mesocia pusilla
(Trosiinae), dorsal view (50 |xm). 146, 147, Megalopyge crispata (Megalopyginae): 146, dorsal view (50 \un);
147, lateral view (50 |im). 148-151, A/dos sp. (Aididae), exuvia from ultimate instar. 148, detail of tip and silk
pore, dorsal view (20 \un); 149, dorsal view (100 (xm); 150, frontoventral view (50 (Am); 151, frontolateral view
(50 Jim). (LP = labial palpus.)
because there appear to be L3 verrucae anterodorsal to the
proleg base in the Megalopygidae (e.g., Megalopyge lanata,
Figure 406). In aidid larvae there is an L3 wart in the same
position, with the two L setae proximoventral to the spiracles
being LI and L2 (Figure 407). L3 warts may form part of the
flexible ventral surface found in the limacodid + dalcerid clade
(see further discussion in "Homology of Suckers," below).
Limacodids and dalcerids (Figures 204-214) and aidids
have mostly primary L and SV setae on the abdomen. In
Pantoctaenia (Limacodidae) and aidids, the abdominal LI and
L2 setae are hair-like and are borne on one (Figure 213) or two
warts (Figure 407) proximoventral to the spiracles. In some
limacodids, the dorsal abdominal LI seta is hair-like, whereas
the ventral L2 seta is fungiform (Figure 205). Other L setae in
limacodids are spatulate (Figure 204), long (Figure 207), or
shagreened (Figures 199, 212). In first instar Crothaema
(Limacodidae), the abdominal LI setae are similar in appear-
ance to warts in the D and SD rows (Figure 214).
34
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 152-157.?Larval spinneret in Dalceridae (scale length in parentheses). 152-155, Acraga infusa
complex, late instar. 152, frontal view (100 \un)\ 153, ventral view (200 Jim); 154, dorsal view (100 um); 155,
detail of anterior margin (20 \im). 156, dalcerid sp. 1 st instar, Ecuador (20 \im) (arrow points to ribbon-like silk);
157 silk gland from Acraga infusa complex, late instar (2 mm). (LP = labial palpus, SiP = silk pore.)
NUMBER 582 35
FIGURES 158-163.?Larval spinneret in Limacodidae and Chrysopolominae (1st instar unless otherwise
indicated; scale length in parentheses). 158, 159, Pantoctaenia gemmans: 158, dorsal view (13.6 urn). 159,
ventral view (20 ^m). 160. Prolimacodes badia, dorsal view (20 u,m); 161. Heuretes sp., dorsal view (12 \un);
162, Natada subpectinata, frontodorsal view, 3rd or 4th instar (50 \im); 163, Chrysopoloma similis, ventral view,
from exuvia of ultimate instar (100 urn). (LP = labial palpus.)
36
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 164-169 (above).?Larval setae and verrucae of Zygaenidae,
Megalopygidae, and Somabrachyidae (1st instar unless otherwise indicated;
scale length in parentheses). 164-166. prothoracic setae: 164, Zyagaena trifolii
(136 urn); 165. Psycharium sp. (Somabrachyidae) (ISO urn); 166. Mesocia
pustlla (Trosiinae), post-third instar (1 mm). 167, Psycharium sp., urticating
setae in verrucae. dorsal row (top of photo is anterior) (200 um). 168, 169,
Mesocia pusilla. post-third instar, SD verrucae: 168, thorax and abdomen
(anterior to left) (231 um); 169, detail of abdominal verruca (arrow points to
urticating seta among plumose setae) (SO um). (D, L, SD, SV = dorsal-,
lateral-, subdorsal-, and subventral-group verrucae, Sp = spiracle,
SpS = spiracular sensillum.)
FIGURES 170-176 (opposite page).?Larval habitus of 1st instar Limacodidae
(scale length in parentheses): 170, Semyra coarctata complex (500 |im); 171,
Belippa horrida (200 um); 172, Euprosterna sp. (200 \un)\ 173, Phobetron
hipparchia (500 Jim); 174, Talimapostica (200 Jim); 175, Crothaema sp. (200
Um); 176, Pantoctaenia prasina (200 Jim).
NUMBER 582
37
38 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
A2 A3 A4
A8
177
D
178
D
179
FIGURES 177-179.?Body setae in Dalceridae and 1st instar Limacodidae (semidiagrammatic): 177, Acraga
infusa complex (Dalceridae), post-third instar (wart-like covering removed); 178, Tortricidia paltida
Herrich-Schaffer (after Dyar, 1896b; L setae added); 179, Semyra coarctata complex. (Al-A9 = abdominal
segments 1-9, D, L, SD = dorsal-, lateral-, and subdorsal-group setae, Sp = spiracle, Tl, T2, T3 = pro-, meso-,
and metathorax.)
NUMBER 582 39
A2-8
180
D1
SD1
MSD1
/
o
D2
SD2
L1
SV
o
O
T2 A2-8
181
FIGURES 180, 181.?Maps of body setae in late instar Dalceridae and 1st instar Limacodidae: 180, Dalceridae;
181, Limacodidae (Tl =Semyra coarctata complex; others = Pantoctaenia prasina). (A2-A8 = abdominal
segments 2-8, Cr = crochets, D, L, MSD, SD, SV, V, XD = dorsal-, lateral-, subdorsal-proprioceptor-,
subdorsal-, subventral-, ventral, and X-dorsal-group setae, MXD = proprioceptor seta, SP = spiracle, Tl, T2,
T3 = pro-, meso- and metathorax.)
40 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
-D1
XD1- XD2\ SD2
MGi RES 18^-is/.?Larval protnoracic setae in umacoaiaae ana uaicenaae (1st instar unless otherwise
indicated; scale length in parentheses). 182, 183, Semyra coarctata complex: 182, prothoracic setae (100 nm);
183, L setae and spiracle (20 \un). 184, Phobetron hipparchia (50 \im); 185, Crothaema sp. (20 urn); 186,
dalcerid sp.. Ecuador (20 urn); 187, Acraga infusa complex (Dalceridae), late instar (note 3 L setae) (200 nm).
(D, L, SD, XD = dorsal-, lateral-, subdorsal-, and X-dorsal-group setae, Sp = spiracle.)
NUMBER 582 41
/
- ' ' ' "V
. ? ? ; ? " ' . " ? ' . ? . ? '.'? * \ * ! ?:>'''.'? JE>
Ik
i
193
? *
. ? "
ir
~ " ' ? ?
 : \ ^ '
-? VW^94
FIGURES 188-194.?Cuticular and gelatinous warts in 1st instars of Limacodidae and Dalcendae (scale length
in parentheses). 188-190, Belippa horrida: 188, D and SD warts (100 u.m); 189, detail of reduced seta on A3
wart (10 Jim); 190, lateral view of setose wart on thorax (20 \im). 191, 192, Pantoctaenia prasina: 191, D and
SD warts on A1-A7 (anterior to right) (100 Jim); 192, detail of wart (20 (im). 193, 194, dalcerid sp., Ecuador:
193, D and SD rows A6-A9 (anterior to left) (100 um); 194, detail of gelatinous wart (50 Jim).
42 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
i m
^ JP
= 195
?x , ? ? ? -
\ ; - . - .
X'
196
FIGURES 195-203.?Long tubercles in 1st instar Limacodidae (scale length in parentheses). 195-197, Apoda
Nguttata: 195, anterolateral view of D and SD tubercles (100 u.m); 1%, apex of SD tubercle prior to extension
(5 urn); 197, apex of D tubercle (6 \un). 198-200, Heuretes sp.: 198, D tubercles (86 u.m); 199, D and SD
tubercles (SD are short, lateral setae to left) (43 um); 200, SD tubercle (3.8 urn). 201-203, Phobetron
hipparchia: 201, detail of partially extended tubercle (20 u.m); 202, SD tubercles partially extended (50 ^m); 203,
tubercles fully extended (100 um). (L, SD = lateral-, subdorsal-group setae or tubercles.)
NUMBER 582 43
\ x^
?
1
MSD2? ^ .
^? 209
? ^
FIGURES 204-209.?Abdominal setae proximate to spiracles in 1st instar Limacodidae and Dalcendae (scale
length in parentheses): 204, Semyra coarctata complex (20 (im); 205, Talima postica (20 i^m); 206, Talima
postica, arrow points to "dorsal" spiracle on Al between scoli on T3 (left) and A2 (100 urn); 207, Phobetron
hipparchia (100 ^m). 208,209, dalcerid sp., Ecuador (anterior to left): 208, spiracular region (50 urn); 209, detail
of setae and spiracle in 208 (see arrow) (20 |xm). (L, MSD = lateral-, subdorsal-proprioceptor-group setae,
Sp = spiracle.)
44 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 210-214.?Abdominal setae proximate to spiracles in 1st instar Limarodidat (scale length in
parentheses): 210, Euprosterna sp. (100 urn); 211. Apoda biguttata (15 \un): 212. Hturetes sp. (90 iimt. 213.
Pantoctaenia prasina (75 nm); 214, Crothaema sp. (38 Jim). (L, MSD. SD * lateral-. MJMOCWJ propn
and subdorsal-group setae, Sp = spiracle.)
In dalcerids and in some limacodids, the LI setae on T2 and
T3 resemble D and SD setae or tubercles (Figures 177-181).
SV setae on T2 and T3 occur dorsal to the legs in dalcerids
(Figure 180) and in some limacodids (Figure 181). These
are fungiform in first instar Pantoctaenia (Limacodidae) (see
description of abdominal SV setae under "Ventral Surface of
Body," below).
Dalcerids have a pair of proprioceptor setae on T2, MSD1
and MSD2, anteroventral to and near the gelatinous SD setae
(Figure 180). There are three small setae proximate to each
abdominal spiracle in dalcerids (Figures 180. 208. 209). I
interpret the seta anterodorsal to the spiracle as MSD2
(fungiform in first instars) and the two posterior setae, one
above the other, as LI and L2. In zygaenids. Tremewan (1985)
designated the MSD2 seta as SD2; however. MSD2 may be
more appropriate for dalcerids because they often have two
gelatinous D and SD setae in addition to the small seu (Figure
180). Singh and Goel (1987). and Miller (1991). noted that
setae referred to as SD2 in noctuids and notodontids should be
termed MSD2. Limacodids with two D and SD warts on the
NUMBER 582 45
abdomen appear to lack MSD setae on the abdomen (e.g.,
Pantoctaema. Figures 181. 213); however, fungiform setae,
proximodorsal to the spiracle, may be the MSD in Apoda
biguttata (Figure 211).
Spiracles can vary in position; the Megalopygidae and
Dakeridae. like Zygaenidae. have spiracles in roughly die same
position on all abdominal segments, whereas in the Aididae and
in some Limacodidae. die First abdominal spiracle is located
more dorsally. supplanting the SD tubercle (Figure 206) (Dyar,
I899a:236).
Trosiines and megalopygines are unique in having either a
digitate or a bulbous tensillum associated with each spiracle
(Figures 168. 328-334). This has been referred to as a gland
by Packard (1894) (see discussion of homology and function in
"Megalopygidae" section, below).
VENTRAL SURFACE OF BODY.?The ventral surface of
limacodid-group larvae is characterized either by the presence
of prolegs on A2-A7. or by a slug-like surface bearing suckers
(Figures 213-222. 223. 224). Thoracic legs are frequently
reduced in the Zygaenotdea. especially in Limacodidae,
Dakendae. bpipyroptdae. and Cyclotornidae. The tiny thoracic
legs in first iratar dakerids and limacodids are roughly the size
of the maxilla (Figures 112. 113).
The pretarsus of the thoracic legs has a similar shape in
aidids. megalopygids, and somabrachyids (Figures 225-230).
The claw projects horizontally from where it meets the round
pretarsal base, and the claw apex extends past the base. In
dakerids and limacodids on die other hand, die claw is more
curved and projects more vertically (Figures 231-233); as a
result, die pretarsal base usually extends past die claw apex
(Figures 231-233. 234-239). In aidids. megalopygids, and
somabrachyids there is a globular seta on die inner margin of
die pretarsus. at die base of die claw (Figures 225-230). Setae
in mis location have been referred to as axial setae in the
Tineidae (Davis and Pena, 1990). In dakerids and limacodids,
die axial seta is hair-like (Figures 231.232) or peg-like (Figure
233).
Thorack kg setae, proximal to die pretarsus, can be either
spatulate or hair-like and can either reach beyond the claw or be
highly reduced (Figures 225-233). Spatulate setae appear in
Tint or early instar megalopygids and in Psycharium (Figures
229. 230) and become more hair-like in later instars (Figure
228). In the Limacodidae the setae can be long and hair-like,
extending past die claw, be reduced to fungiform either
throughout the kg or on the coxae only (Figures 237-239). or
be somewhat in-between as capitate setae (Figure 234. 273).
The VI seta of the thorax is hair-like in megalopygids (Figure
242). whereas it is fungiform in dalcerids (Figure 243) and in
limacodids.
The cutick texture of die sternum and of the coxae of the
thorack kgs is shagreened in megalopygids and aidids but is
smooth in dalcerids and limacodids (compared in Figures
240-245). There is a remnant of shagreened cuticle on the
anteroventral portion of Tl in dakerids (Figure 243).
Prolegs in zygaenids (Figures 246-248) are more elongate
and often have fewer crochets than do those of megalopygids,
aidids. or somabrachyids. The Megalopygidae are widely
reported to possess prolegs on A2 and A7 mat lack crochets
(Figures 254, 257). This condition also occurs in die
Somabrachyidae (Figures 249, 250) (H. Geertsema, pen.
comm.) and Aididae (Figure 223); however, some mega-
lopygids have crochets on all prolegs (e.g., Mesocia pusilld)
(Figures 216. 252. 253).
Megalopygid prolegs are unique among the Zygaenoidea in
having membranous pads anterolateral to die crochets (Figures
252-257). The pads are absent on A10 and can be roughly
equal in size on other segments or can be relatively large on die
prolegs of A2 or A7 when crochets are absent Similar pads
occur in larvae of the Lycaenidae (Downey, 1987) and
Riodinidae (Harvey, 1987), but in these families die pad
interrupts die band of crochets.
The presence of abdominal suckers in the Limacodidae
(Chapman. 1894; Dyar, 1899b) and Dakeridae is unique in
external-feeding lepidopteran larvae. Genera in diese families
have up to eight abdominal suckers, although die suckers on Al
and A8 are often feebly developed or absent (Figures 219.222,
261, 266) (see "Homology of Suckers" for discussion).
Late-instar larvae in die limacodid group exhibit a wide
range in die number and a diversity of configurations of the
uniordinal crochets on the A2-A7 prolegs. Late instar
megalopygids have 20-70 crochets in linear or nonlinear rows,
either divided by a distinct gap (e.g.. Nor apt) or undivided
(Figures 254, 256). Larval Aididae have 50-70 crochets that
form a nearly continuous V-shaped row (Figures 340-344).
Late instars of Acraga infusa (Dakeridae) have up to 17
crochets (Figure 264), five more than in Dalcerides ingenita
(Stehr and Me Far!and. 1987). A Megalopyge sp. (Colombia,
BMNH) has 12-14 crochets in the first instar (Figures 258,
259), an unusually large crochet number compared to early
stage Zygaena trifolii and Psycharium sp., which each have
four (Figures 248, 251). In the Somabrachyidae, late-instar
larvae have crochets arranged in two separate rows. A
Somabrachys sp. has up to 28 crochets (12 anterior, 16
posterior), and a Psycharium sp. has 18 crochets (five anterior,
13 posterior).
Larvae of die Dakeridae lack crochets in early instars
(Figures 224, 262) but develop them on the A3-A6 prolegs
by 4th instar (Figures 261, 263, 264). Additional crochets are
sometimes added on A2 and A7, although fewer in number, in
late instars (Stehr and McFarland. 1985, 1987). Crochets are
reported as absent in the Limacodidae (Figures 222, 265-270)
(Holloway, 1986; Stehr, 1987d); however, I found them to be
present in Pantoctaema gemmans (Figures 359, 360). Pantoc-
taenia develops crochets on A2-A7 and AI0 (in roughly equal
numbers) after the first instar, although the precise instar is not
known.
FIGURES 215-222.?Larval habitus in ventral view (post-first instar unless otherwise indicated): 215, Aidos sp.
(Aididae) (photo by C. Hansen); 216, Mesocia pus ilia (Trosiinae); 217, Psycharium sp. (Somabrachyidae) (1st
instar); 218, Megalopyge lanata (Megalopyginae) (photo by L. Minor-Penland); 219, Pantoctaenia gemmans
(Limacodidae) (photo by C. Hansen); 220, Semyra coarctata complex (Limacodidae) (1st instar); 221,
Euprosterna sp. (Limacodidae) (1st instar); 222, Acharia sp. (Limacodidae) (photo by L. Minor-Penland).
NUMBER 582 47
A10
224
FIGURES 223. 224.?Detail of larval ventrum: 223. Aidos sp. (Aididae).
Venezuela (late instar) (photo by L. Minor-Penland); 224, dalcerid sp., Ecuador
(1st instar). (Al-A 10 = abdominal segments 1-10.)
48 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 225-233.?Larval thoracic legs and pretarsal claw in the limacodid group (scale length in parentheses).
225-227, Aidos sp. (Aididae), from exuvia of ultimate instar 225, thoracic legs (0.38 mm); 226, detail of 225
(120 urn); 227, detail of claw (75 urn). 228, Mesocia pusilla (Trosiinae) (50 urn); 229, Megalopyge sp.
(Megalopyginae). 1st instar (20 urn); 230, Psycharium (Somabrachyidae), 1st instar (15 |im); 231, Acraga
infusa complex (Dalceridae). middle instar (50 |im); 232, Semyra coarctata complex (Limacodidae), 3rd instar
(20 Jim); 233, Apoda biguttata (Limacodidae), 1st instar (4.3 |xm). (AS = axial seta.)
NUMBER 582 49
1
235
? - v
-.' . _ . '
a
? ' | T a y ^ ; \/
' Tik-" J
? / i
Fe
-
j / Cx j
i
A 237
FIGURES 234-239.?Larval thoracic legs in Limacodidae and Dalceridae (scale length in parentheses): 234,
Semyra coarctata complex, 3rd instar (SO um); 235, Acraga infusa complex (Dalceridae), middle instar (200
|im); 236, Pantoctaenia, legs on T1-T3,1st instar (50 Jim); 237, Phobetron hipparchia, 1st instar (20 um); 238,
Talima posiica, 1st instar (10 Jim); 239, Natada subpectinata, post-first instar (20 Jim). (Cx = coxa, Fe = femur,
Ta = tarsus, Ti = tibia.)
50
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
? ?
?
? - ^
ii
-
1
240
;
i? /
 x - ^
FIGURES 240-245.?Comparison of thoracic ventral skin texture and setae between middle instar Mega-
iopygidae (left) and Dakeridae (right) (scale length in parentheses). 240, 242, 244, Mesocia pusilla
(Megaiopygidae): 240, entire legs, T1-T3, and shagreened cuticle at base (200 ^m); 242, base of 1 leg (arrow
points to homologous VI seta found in Acraga, to right) (200 Jim); 244, detail of 1 leg (200 Jim). 241,243,245,
Acraga infusa complex (Dalceridae): 241, entire legs, Tl and T2 (200 urn); 243, base of 1 leg (arrow points to
homologous VI seta, found to left; note shagreened cuticle above) (200 urn); 245, detail of 1 leg (100 \un).
(Cx = coxa.)
NUMBER 582 51
ik /ilm
?
FIGURES 246-251.?Prolegs in 1st instar Zygaenidae and Somabrachyidae (scale length in parentheses).
246-248, Zygaena trifolii: 246, lateral view of prolegs A3-A6 (136 nm); 247, lateral view of A4 (30 urn); 248,
proleg A5, viewed from below (20 (im). 249-251, Psycharium sp. (Somabrachyidae): 249, proleg pairs on A2
(above) and A3 (75 \un); 250, detail of proleg A2 (50 urn); 251, detail of proleg A4 (27 Jim). (A3,
A6 = abdominal segments 3,6, L = lateral-group setae.)
52 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
f
FIGURES 252-257.?Prolegs in Megalopygidae (scale length in parentheses). 252, 253, Mesocia pusilla
(Trosiinae): 252. prolegs A2 and A3 (arrow points to additional membranous pad on SV verruca) (1 mm); 253,
detail of A2 proleg (200 um). 254-257. Megalopyge crispata (Megalopyginae): 254, prolegs A2 and A3 (arrow
points to normal SV verrucae) (500 u.m); 255, proleg A3, lateral aspect (200 um); 256, detail of A3 proleg (200
um >. 257, detail of A2 proleg (200 y,m).
NUMBER 582 53
FIGURES 258-260.?Prolegs of 1st instar Megalopygidae (Megalopyge sp.) (scale length in parentheses): 258,
prolegs A2-A4 (120 urn); 259, detail of A4 proleg (50 u,m); 260, detail of A4 subventral setae located at top of
259 (12 urn).
- -
-
V1 -
? - V
FIGURES 261-264.?Crochets and ventral abdominal surface in Dalceridae (anterior at top, late instar unless
otherwise indicated; scale length in parentheses): 261, crochets on segments A3-A5, Dalcerina tijucana (photo
by L. Minor-Penland); 262, detail of ventral surface and fungiform setae, 1 st instar dalcerid species, Ecuador
(meson to left) (20 \im). 263, 264, Acraga infusa complex: 263, crochets in relation to homologous setae in 262
(meson to left) (200 urn); 264, detail of crochets (100 urn). (Cr = crochets, SV, V = subventral-. ventral-group
fungiform setae.)
54 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 265-270.?Ventral abdominal surface of Limacodidae (anterior at top, 1st instar unless otherwise
indicated; scale length in parentheses): 265, Belippa horrida (200 u.m). 266,267, Semyra coarctata complex, 4th
instar (200 Jim): 266, suckers on A2-A4 (500 Jim); 267, detail of 266 (see arrow), arrows point to fungiform
setae homologous to those of Dalceridae in 262, 263 (200 um). 268, 269, Phobetron hipparchia: 268, ventrally
extended segments A1 - A5 (150 \im); 269, detail of 268 (see arrow), arrows point to fungiform setae SV1, SV2,
VI (60 Jim). 270, Pantoctaenia prasina, posterior segments (note shagreened texture of A10 at bottom vs.
smooth segments above) (86 urn).
The crochets on the A10 proleg are a linear undivided
mesoseries in the Megalopygidae (Figures 278, 284), Aididae
(Figure 223), and Zygaenidae, although they are semicircular
in first instar Psycharium and Zygaena trifolii (Figures 281,
283). Crochets on A10 are absent in dalcerids and Iimacodids
except in post-first instar Pantoctaenia (Figure 360).
NUMBER 582 55
FIGURES 271-275.?Modified larval setae on ventral surface and thoracic legs in Dalceridae and Limacodidae
(scale length in parentheses): 271, Acraga infusa complex (Dalceridae), lateral view of fungiform seta (5 |im).
272, 273, Semyra coarctata complex: 272, abdominal seta (20 nm); 273, capitate seta on leg (5 fim). 274,
Acharia stimulea, abdominal seta (10 \im); 275, Acraga infusa complex, abdominal seta (20 \im).
First instar Zygaena trifolii (Figure 247), Psycharium
(Figure 249-251), and megalopygids (Figure 260) have two
spatulate SV setae on the anterior portion of the prolegs. These
are hair-like in later instars except in Psycharium, in which they
become spines and increase in number.
Fungiform S V and V setae are present on the ventral portions
of the abdomen on A1-A9 in larval limacodids and dalcerids
(Figures 224, 271, 272, 274, 275). The most dorsal of these
fungiform setae is SV1, with SV2 and VI located towards the
meson (Figures 180, 181, 262, 263). SV and V setae were
previously unreported in limacodids and were considered
absent in dalcerids (Stehr and McFarland, 1987); however, they
are present from first to final instar in each family, and crochets
develop between SV2 and VI setae in later instar dalcerids and
in Pantoctaenia (Figures 262, 263).
Stehr (1987a:455) reported that the dorsum of A10, or anal
plate, in megalopygid caterpillars is reduced to "a small lobe"
that is "tucked beneath" the dorsum of A9. Viewed from above,
this gives A9 the appearance of being "the terminal segment."
Adding to this impression is the location of the A10 proleg,
directly below the dorsum of A9. The ventral portion of A9 is
very narrow and forms a functional segment with A8 above and
below. In effect, the ventrum of segment A9 and of A10 has a
functional counterpart on the dorsum one segment removed to
the anterior. Although reported only in Megalopygidae, this
condition occurs throughout the limacodid group but to a lesser
56 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
SD1
276
277
FIGURES 276, 277.?Chaetotaxy of anal segment (A 10) in Dalceridae and Limacodidae (late instars): 276,
Acraga infusa complex (Dalceridae); 277, Phobetron hipparchia. (D, L, SD, SV, V = dorsal-, lateral-, subdorsal-.
subventral-, and ventral-group setae, FS = frass-flipping setae, LL = lateral lobe, ML = mesal lobe, PP = para-
proct seta.)
degree in the Somabrachyidae. The anal plate in most of the
limacodid group bears either pairs only of primary D and SD
setae (Figures 276, 277) or a narrow row that includes
secondary setae, compared to an anal plate with verrucae in late
instar Zygaenidae and Somabrachyidae.
An anal fork or anal comb used for frass removal is located
between the anal plate and the anus. It is present in the
Megalopygidae (Figure 278) and Somabrachyidae (Jordan,
1909) (Figures 279, 280, 282) and is present or absent (Figure
283) in the Zygaenidae and Aididae. I use the term anal fork for
a single medial process with a emarginate dorsal border (Figure
278) and the term anal comb for a row of stout spines (Figures
279,280). Neither is present in the Limacodidae and Dalceridae
(Figures 276, 277, 285-287, 293-296), although nonho-
mologous setae of similar function sometimes occur below the
anus (see next paragraph).
Prolegs on A10 in the Dalceridae (Stehr and McFarland,
1987) and Limacodidae consist of two small spinulose lobes
(Figures 276, 277). The lateral lobe has five or six stout
setae on each side, and the smaller mesal lobe has two. In
middle to late instars, the outer portion of the lateral lobe
consists of three L setae and either one secondary or one SV
seta; the one or two setae near the meson are either paraproct
setae or SV1 setae (Stehr, 1987b). The two setae of the mesal
lobe represent either SV4 and VI setae anterior to the proleg
base in other Lepidoptera (Stehr, 1987b) or any of the
SV1-SV3 setae (Figure 276). In either case this would be a
reduction from the condition in Zygaena trifolii (Tremewan,
1985). Wider, peg-like setae on the dorsomesal margin of the
lateral lobe are used for flipping frass (Stehr and McFarland,
1987). These setae occur in dalcerids (Figures 286, 289) and
limacodids (Figures 277, 287, 288, 290), although they can be
absent in either family (Figures 276, 293, 295, 296). Crochets
in late-instar Pantoctaenia (Limacodidae) develop between the
lateral and mesal lobes of A10 (Figures 360, 361).
PUPAE
Several pupal character states in the Zygaenidae and in the
limacodid group have led past workers to consider them
primitive. These include (1) thin cuticle; (2) moveable
abdominal segments A3-A6; (3) eclosion outside the cocoon;
and (4) visible Al spiracles. The last, however, may be a
synapomorphy for the Zygaenoidea because it does not occur in
lower ditrysian superfamilies, such as Tineoidea (D.R. Davis,
pers. comm.). The presence of an exposed spiracle on Al was
noted by Mosher (1916) as a character state shared by the
Zygaenidae (as Pyromorphidae), Limacodidae (as Eucleidae),
and Megalopygidae. The spiracle also is visible in the
Cyclotornidae (Common, 1990), whereas it is either hidden
(Common, 1990) or visible in the Epipyropidae (discussed
further in the section on Epipyropidae, below).
NUMBER 582 57
FIGURES 278-284.?Anal segment (AlO) in Megalopygidae, Zygaenidae, and Somabrachyidae (scale length in
parentheses): 278, anal fork and prolegs of Megalopyge crispata (Megalopyginae), late instar (500 nm).
279-282, Psycharium sp. (Somabrachyidae), 1st instar: 279, anal comb (upper arrow) and prolegs (lower arrow)
(111 urn); 280, detail of anal comb, arrow points to middle prong (40 jim); 281, detail of anal proleg (75 Jim);
282, detail of middle prong of anal comb in 280 (see arrow) (3 \un). 283, anal prolegs and plate of Zygaena trifolii
(Zygaenidae) (60 ^m); 284, anal proleg of Mesocia pusilla (Trosiinae) (100 ^m). (L, SV = lateral-,
subventral-group setae, PP = paraproct seta.)
58 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
mm
?L. - -
31^
287,
....
' V ^ 1
FIGURES 285-290.?Anal segment (A 10) in lst instar Limacodidae and Dalceridae (scale length in parentheses):
285. posteroventral aspect of Phobetron hipparchia (200 urn). 286, 287, posterolateral aspect of anal plate and
vestigial proleg: 286. Dalcerides ingenita (Dalceridae) (100 ^m); 287, Semyra coarctata complex (100 Jim).
288, 289, detail of frass-flipping setae: 288, Phobetron hipparchia, detail of 285 (20 urn); 289, D. ingenita
(Dalceridae) (20 \xm). 290, ventral aspect of Semyra coarctata complex (50 urn).
NUMBER 582 59
FIGURES 291-296.?Anal segment (A 10) in 1st instar Limacodidae and Dalceridae (scale length in parentheses).
291, 292, ventral view of anal proleg of Phobetron hipparchia: 291, contrast between smooth A9 and spinulose
A10 (SO urn); 292, detail of spinulose A10 in 291 (27 \im). 293, lateral view of anal proleg and plate of
Pantoctaenia gemmans (50 urn). 294-296, posterior view of anal segment: 294, Heuretes sp. (27 urn); 295, P.
gemmans (50 Jim); 296, Euprosterna sp. (50 nm).
60 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
297 298
FIGURES 297, 298.?Pupa of Dalceridae {Acraga infusa complex) (scale = 5 mm): 297, ventral aspect; 298,
lateral aspect
Mi net (1986) considered an A2 spiracle that is covered by
the hindwing to be a synapomorphy for the Zygaenoidea. In
limacodid (Common, 1990) and megalopygid pupae, however,
spiracles on both Al and A2 are visible. Only the A2 spiracle
is visible in dalcerid pupae (Figures 297, 298). Pupal spiracles
in the Megalopygidae (Figure 302) are unique in having lobes
(Mosher, 1916) and are similar in shape and position to those
found in the larvae (Figures 328-334), but they occur on fewer
segments (e.g., Megalopyge lanata, A1-A6 in females, Al
only in males).
The pupal eyepiece in the limacodid group has a lateral
extension called the sculptured flange (Chapman, 1894:349;
Mosher, 1916) (Figures 303, 307). Reported as a potential
synapomorphy for Limacodidae + Megalopygidae (Holloway,
1986), it also occurs in the Dalceridae (Figure 303) (Epstein,
1988) and Aididae. The sculptured flange, located over the T2
spiracle, moves up and down during respiratory ventilation in
megalopygid pupae (Mosher, 1916). Following eclosion, this
structure either remains attached (Figures 303, 315, 317) or
dehisces (Figures 304, 312-314, 316, 318, 319) from the head
capsule.
As is the adult proboscis, the pupal proboscis in the
limacodid group is weakly developed compared to the
Zygaenidae. A lateral lobe connected to the maxilla, found in
some lineages of Limacodidae (+ Chrysopolomidae) (Figures
316, 317), has been termed the "maxillary palpus" or "eye
collar" (Chapman, 1893:115), "lateral process connected with
it [the maxillary palpus]" (Packard, 1895a:797), and the
NUMBER 582 61
"lateral prolongation [of the maxilla]" (Mosher, 1916:43). I
refer to it as the maxillary extension. This structure may
function to hold the free front legs of "incomplete" pupae in
place at the femur-tibia joint, along with the sculptured eye
flange and antenna. Loss of the maxillary extension is likely to
have occurred independently within several lineages of New
World Limacodidae with spiny larvae (Figures 318, 319). A
similar maxillary structure occurs in the Nepticuloidea (Chap-
man, 1893), including Opostegidae (Davis, 1989).
The pupal frons, which commonly projects as a rounded
bump (Figure 320), may assist in forcing open the cocoon in
megalopygids. In dalcerids, the frons can have up to six (in two
groups of three) closely appressed setae (e.g., Dalcerides
ingenita) (Figure 325). The frons may be rough and shagreened
in limacodids, perhaps to assist with weakening or freeing the
lid (e.g., Phobetron pithecium, Limacodidae) (Figures 322-
324). The nonmovable pupal mandibles in species of the
limacodid group are usually distinguishable from the fronto-
clypeus (Figure 321).
Stiff abdominal setae on the pupal dorsum, directed
posteriorly (Figure 311), may prevent pupae in the limacodid
group from back-sliding while projecting from the cocoon prior
to eclosion. Cremaster hooks may be either present (Figures
308-310) or absent.
Phylogenetic Analysis
DATA ANALYSIS.?Reliance on characters of either adults
(Brock, 1971) or immatures (Common, 1975; Minet, 1986)
for the limacodid group and its allies has led to different
theories concerning membership of the Cossoidea and
Zygaenoidea (see "History of Classification of the Limacodid
Group," above). Therefore, independent data sets of adult- and
immature-stage characters for phylogenetic analyses were
used. Following the initial analyses, the data sets were
combined to compare the influences of each. The outgroup
method was used to polarize the characters (Maddison et al.,
1984).
FIGURES 299-302.?Pupa of Megalopygidae (Megalopyge lanata) (299-301 photos by L. Minor-Penland): 299,
ventral female (length = 27 mm); 300, dorsal male (arrow points to longitudinal depression on mesonotum,
similar to adults); 301, ventral male (length = 22 mm); 302, detail of female abdominal spiracles and sensilla,
anterior below (sensilla similar to larvae) (scale = 1 mm) (photo by C. Hansen). (SpS = spiracular sensillum,
Sp = spiracle.)
62 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 303-311.?Detail of pupae in Limacodidae and Dalceridae (scale length in parentheses): 303,
sculptured eyepiece in eclosed pupa of Dalcerides ingenita (Dalceridae) (200 Jim). 304-308, Semyra coarctata
complex (Limacodidae): 304, eclosed pupa minus eye and sculptured portion (500 \im); 305, uneclosed pupa
with sculptured eyepiece (arrow points to fracture line) (500 urn); 306, head and wings of uneclosed pupa (2
mm); 307, detail of antenna and sculptured eye (200 urn); 308, cremaster (50 urn). 309-311, D. ingenita: 309,
posterior end of abdomen, ventral view (0.6 mm); 310, cremaster, ventral view (86 \xm); 311, dorsum of A8
(above) and A9 (0.38 mm).
NUMBER 582 63
317
? .^H
V
FIGURES 312-319.?Pupal head (scale length in parentheses): 312, Aidos yamouna (Aididae) (1 mm); 313,
Megalopyge nr. melaina (Megalopyginae) (1 mm); 314, Semyra coarctata complex (Limacodidae) (2 mm); 315,
Dalcerides ingenita (Dalceridae) (500 |im); 316, Acharia sp. (Limacodidae) (1 mm); 317, Chrysopoloma similis
(Chrysopolominae) (2 mm); 318, Euprosterna sp. (Limacodidae) (1 mm); 319, Monoleuca semifascia
(Limacodidae) (1 mm). (LP = labial palpus, Md = mandible, Mx = maxilla, MxE = maxillary extension.)
64 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 320-325.?Detail of frontal portion of pupal head (scale length in parentheses): 320, Megalopyge nr.
melaina (Megalopyginae) (500 urn). 321-324, Phobetron pithecium (Limacodidae): 321, frons and mandibles
(500 Jim); 322. dorsal ridge of frons, frontal view (200 urn); 323, detail of 322 (see arrow) (50 Jim); 324,
frontolateral view (200 urn); 325, setae on frons in Dakerides ingenita (Dalceridae) (231 \un).
(Md = mandible.)
NUMBER 582 65
Hennig86 (Farris, 1988) was used for parsimony analyses of
the two independent data sets. The adult- and immature-stage
data had 18 (11 multistate) and 25 characters (8 multistate),
respectively (Tables 4, 5). Larval characters included early and
late instars. The total number of steps of the character state
transformations for each cladogram was measured as the length
(L). The fit of characters to the cladograms was measured by
the consistency index (CI) (Kluge and Farris, 1969) and by the
retention index (RI) (Farris, 1989). Multistate character states
were ordered into morphoclines from the zygaenid outgroup
(Mickevich and Weller, 1990), and the analysis was
performed using the ordered default in Hennig86 (the unor-
dered option was also used for comparison) (Farris, 1988).
The Zygaenidae were chosen as the outgroup because they
have the largest number of potentially synapomorphic charac-
ter states with the limacodid-group families. These include the
visible first abdominal spiracle of the pupa, uniordinal crochets
TABLE 4.?Taxonomic characters and codes used in phylogenetic analysis
(0 = plesiomorphic condition, 1-3 = apomorphic conditions).
CHARACTERS OF ADULTS
Head
1. Labial Palpi: (0) 2 segments, not reaching vertex; (1)3 segments, porrect
nearly equal length or distal segment xh length of middle segment; (2)
2 segments, porrect, reaching vertex.
2. Chaetosema: (0) present; (1) absent
3. Galea: (0) very long, much > head width; (1) short, often < head width; (2)
absent.
4. Ocellus: (0) present; (1) absent.
Thorax
5. Forewing radial veins, RUy (0) independent; (1) R, and R2 branched; (2)
R2 and R3 branched.
6. Forewing radial veins, R3_5: (0) independent or R3 and R4 branched from
end of discal cell; (1) R, and R4 branched off vein beyond discal cell; (2)
R4 and R5 branched.
7. Hindwing discal cell and Sc + R,: (0) Sc + R, anastomoses with Rs for
length of cell (at least midway); (1) Sc + R, anastomoses with Rs for
length of cell in female only (Megalopyginae); (2) Sc + R, branches
from Rs at base, both sexes (in Aididae, R, anastomoses with Rs for
length of cell, anastomoses with Sc beyond cell).
8. Mesepimeron: (0) broadly sclerotized; (1) narrowly sclerotized.
9. Mesonotum: (0) anterior simple; (1) anteromedial groove.
10. Tibial spurs: (0) 1 pair on mid- and 2 pair on hindleg; (1) 1 pair on mid- and
hindleg; (2) absent
11. Ventral surface of 5th tarsomere: (0) monomorphic; (1) dimorphic, sensilla
trichodea only on female and restricted to distal portion; (2) dimorphic,
sensilla trichodea only on female and intermixed with scales; (3)
dimorphic, without scales, usually as a recessed pad.
Male genitalia
12. Uncus: (0) broad and gently tapering; (1) narrow, sclerotized hook; (2)
reduced or absent.
13. Gnathos: (0) absent; (1) present
14. Juxta: (0) paired processes, sometimes arising from plate; (1) simple.
15. Valvae: (0) wide relative to aedeagus and juxta; (1) digitate and narrow,
dorsal; (2) reduced or absent
Female genitalia
16. Papillae anales (ovipositor lobes): (0) small, lobes flat and open,
perpendicular to abdomen, enure; (1) lobes appressed; (2) large, flat and
open, moderately or deeply cleft; (3) large, disk shaped, sometimes
emarginate.
17. Dense scale pouch on female 7th segment: (0) absent; (1) present
18. Large secondary accessory glands: (0) present, empty into oviduct; (1)
absent; (2) present external ducts.
CHARACTERS OF IMMATURES
Larva
19. Stemmata: (0) stemma 5 disjunct ventrally, removed > 1.5 x width of 1
stemma from stemma 4; (1) stemma 5 closer to stemma 4, - width of 1
stemma from stemma 4.
20. Stemmatal setae on head: (0) SI and S2 present; (1) only SI present
21. Labrum (anterior margin): (0) normal spines; (1) dense spines in medial
region; (2) dense spines over entire margin.
22. Spinneret (labium): (0) tubular; (1) narrow opening, laterally expanded,
brush-like.
23. Thoracic legs: (0) easily visible, length > width of labrum; (1) small to
minute, <, width of labrum.
24. Axial seta between hook and base of pretarsal claw: (0) short and rounded;
(1) conical or hair-like, sometimes reaching hook.
25. Abdominal segments with prolegs or suckers: (0) A3-A6 and A10 with
prolegs; (1) A2-A7 and A10 with prolegs; (2) A2-A7 and A10 with
vestigial prolegs (A1-A8, with or without suckers).
26. Abdominal segments A2-A7 with crochets in final instar (0) crochets on
A3-A6 (prolegs present or absent on A2 and A7; segments with
prolegs without membranous pad); (1) crochets on A2-A7 (crochets
often absent; A2-A7 prolegs with membranous pads or A2-A7 with
vestigial prolegs and slug-like ventrum); (2) crochets absent
27. Abdominal segment A10 with crochets in final instar. (0) present; (1)
absent
28. Presence of crochets in first two instars: (0) present; (1) absent
29. Length of ventrum of segments A8 and A9 relative to dorsum: (0) normal;
(1) narrow, - length of 1 segment on dorsum.
30. texture of ventral surface at base of prolegs A1-A9 and thoracic legs: (0)
spinulose; (1) smooth.
31. Distinct membranous pads on prolegs A2-A7: (0) absent; (1) present
32. Frass flipper on anal prolegs: (0) absent; (1) present
33. Anal fork: (0) present; (1) absent.
34. Dorsum of A10 (anal plate): (0) normal; (1) reduced, below dorsum of A9.
35. V setae on thorax, V and SV setae on abdomen: (0) hair-like; (1)
fungiform.
36. L setae on abdomen: (0) a number of secondary setae (verrucae); (1 )2
primary setae below spiracle; (2) 2 primary setae in line with spiracle.
37. D and SD setae: (0) plumose and spiny setae together, (1) spiny setae or
tactile setae after the first instar, (2) primary setae with smooth,
gelatinous covering.
38. Digitate spiracle sensilla: (0) absent; (1) on Tl, A1-A7.
Pupa
39. Eyepiece: (0) simple; (1) sculptured flange.
40. Labial palpus: (0) width much < width of tibia; (1) - equal to width of tibia.
41. Maxilla: (0) long proboscis, reaching length of forewing costa; (1) lobes,
much shorter than forewing costa; (2) additional lobe, parallel and <
adjoining labial palpus.
42. Cocoon prior to eclosion: (0) ellipsoid, with trap-door end truncate, lid
often visible; (1) ovoid, no visible lid; (2) diffuse with inner chamber.
Egg
43. Shape: (0) flat, scale-like, thick chorion; (1) ovoid; (2) flat, scale-like, thin
chorion.
66 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
TABLE 5.?Character matrix of taxa used in phylogenetic analysis of adults and immatures. Character numbers
correspond to the list in Table 4 and Figure 326 (0 = plesiomorphic, 1 -3 = apomorphic). Numbers in superscript
refer to homoplastic alternate character states found in the taxon.
Taxon
Zygaenidae
Megalopyginae
Trosiinae
Aididae
Dalceridae
Pantoct/Croth
Limacodidae
1
0
0
0
12
2
1
1
2
0
0
0
0
1
1
1
3 4 5
0 (
2
1
2
1
1
1
) 0
0
0
1
2
0
0
6
0
2
2
1
2
1
1
Characters of Adults
7
0
1
0
2
2
2
2
8
0
0
0
0
1
1
1
9
0'
1
1
1
0
0
0
10
0
1
1
1
2
0
0
11
0
2
2
3
1
1
3
12
1
1
1
0
2
0
0
13
0
0
o1
0
1
1
1
14
1?
0
0
0
0
1
0'
15
0
1
1
0
2
0
0
16
0
1
1
2
2
3
3
17
0
1
1
0
0
0
0
18
0
1
1
0
2
1
1
Taxon
Characters of Immatures
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43
Zygaenidae 01 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0'
Megalopyginae
Trosiinae
Aididae
Dalceridae
Pantoct/Croth
Limacodidae
1
1
1
1
0
0
0
0
0
1
1
1
0
0
0
1
1?
2
0
0
0
1
1
1
0
0
0
1
1
1
0
0
0
1
1
1
1
1
1
2
2
2
1?
ln
0
1
1
2
0
0
0
1
0
1
0
0
0
1
1
1
0
0
0
1
1
1
1
1
0
0
0
0
0
0
0
1?
0
1?
0
0
0"
1
1
1
0
0
0
1
1
1
0
0
1
2
1
1
0
0
1
2
1
I2
1
1
0
0
0
0
1
1
1
1
2
2
0 1
0 1
0 1
2 1
0 1
1 2
in mesoseries (rare in Apoditrysia), arrangement of setae or
verrucae, and retractile head of the larva. Although adult
synapomorphies are more problematic, the presence of secon-
dary accessory glands in females is a potential synapomorphy.
The Cossidae were not used as an outgroup because little
evidence was found for synapomorphies between adult or
immature stages with the limacodid group; however, character
states of cossids and other families in Apoditrysia were useful
in determining character codes for adult Zygaenidae.
Ingroup terminal taxa represented families Limacodidae
(minus Pantoctaenia and Crothaema) and Dalceridae and
previous megalopygid subfamilies Megalopyginae, Trosiinae,
and Aidinae. Both Trosiinae and Megalopyginae, although
having nearly redundant characters (excluding autapomor-
phies), were included to test the position of "Aidinae."
"Chrysopolomidae" was not included because preliminary
evidence suggested synapomorphies with limacodid lineages
(see Chrysopolominae section, below). Epipyropidae, Cyclo-
tornidae, and Heterogynidae, although considered to be within
Zygaenoidea (Common, 1990), were not included in the
analysis because, at present, no synapomorphies could be
found with the limacodid group at a level below Zygaenoidea.
Furthermore, they could be members of clades presently within
Zygaenidae that were not included in the analysis. Somabra-
chyidae was not included due to incomplete character data.
Pantoctaenia and Crothaema (Limacodidae) were combined
and included in a second analysis, with the adult- and
immature-stage data sets analyzed separately as well as
combined, to test the monophyly of the Limacodidae. These
two genera exhibit potentially plesiomorphic character states
also found in Dalceridae (e.g., late instar Pantoctaenia
gemmans has crochets). Crothaema was combined with
Pantoctaenia because these genera share a number of traits not
found in other limacodids. Furthermore, pupal and cocoon
character states from Crothaema were used to supplement
missing data for Pantoctaenia.
Character data assigned to the limacodid-group taxa are from
specimens in Table 1 and from the following: Hopp (1927,
1930) for Megalopygidae; Miller (1994) for Dalceridae; Janse
(1964), Holloway (1986), Holloway et al. (1987), and Epstein
(1988) for Limacodidae. Adult character states assigned to the
zygaenid outgroup were from subfamilies Zygaeninae and
Procridinae and were taken from Alberti (1954), Naumann
(1977), Tarmann (1984), and specimens in the USNM
collection. Larval zygaenid characters were from examination
of USNM and Naumann specimens and from descriptions by
Dyar and Morton (1895), Forbes (1910, 1923), Fracker(1915),
Tremewan (1985), and Stehr (1987c). Pupal zygaenid charac-
ters were from examination of USNM specimens and from
Mosher (1916), whereas egg characters were from USNM
specimens and from Peterson (1967), Common (1990), and
Fehrenbach (1995).
RESULTS AND DISCUSSION.?The first set of analyses using
independent character data from immature and adult stages
resulted in the same most-parsimonious cladogram (Figure
326, minus Pantoctaenia/Crothaema). The megalopygid sub-
families Trosiinae and Megalopyginae were monophyletic and
arose from the base of the limacodid group. Aididae formed a
monophyletic group with Limacodidae + Dalceridae. Synapo-
morphies supporting this hypothesis are listed in Table 6 and
are shown in Figure 326. Tree length (L), consistency index
(CI), and retention index (RI), from each analysis, are
summarized in Table 7.
Although the results of the first set of analyses produced
NUMBER 582 67
Characters of
Immature
Stages
FIGURE 326.?Cladogram of limacodid group based on parsimony analysis of
adult and immature-stage characters, with Zygaenidae as outgroup. Shaded
rectangles represent character codes for synapomorphies; multiple-state
characters with lower numbers are plesiomorphic compared to those with
higher numbers. Character numbers and states correspond to descriptions in
Table 4 and matrix in Table 5. Numbers 1-18 and 19-43 refer to adult and
immature-stage data, respectively. Taxon "Pant/Croth" represents a combina-
tion of characters found in Pantoctaenia and Crothaema (Limacodidae).
cladograms congruent between immature and adult stages, the
inclusion of Pantoctaenia/Crothaema produced two clado-
grams. The adult and combined charater data placed Dalceridae
as the sister group to Limacodidae + Pantoctaenia/Crothaema,
whereas characters of the immatures showed Pantoctaenia/
Crothaema to be the sister group of Dalceridae + Limacodidae.
The different cladogram based on the character data of
immature stages is explainable by two "synapomorphies" for
Limacodidae + Dalceridae, each related to the anal proleg: (1)
the presence of frass-flipping setae below the anus (character
32), and (2) the absence of crochets in mature larvae (character
27). These synapomorphies may be flawed, as indicated by the
results of the adult and combined data sets. In Pantoctaenia the
absence of frass-flipping setae may result from a secondary loss
rather than from plesiomorphy (as treated here). This is
suggested by the absence of these setae in a number of
limacodids (Figure 296) and in some dalcerids (Figure 276).
Furthermore, crochets on A10 may have been lost independ-
ently in both limacodids and dalcerids rather than the absence
being a synapomorphy, because throughout the limacodid
group similar losses appear to have occurred on A2 and A7. In
order to test the influence of the two characters, I recoded
character 32 in Pantoctaenia/Crothaema as a loss, rather than
as a plesiomorphy, and I recoded character 27 as a convergent
loss in limacodids and dalcerids. I used linear coding for
character 27 to allow transformation of each loss of A10
crochets from the plesiomorphic condition (Mickevich and
Weller, 1990). This resulted in congruence with the adult data
set (L = 38, CI = 0.92, RI = 0.90).
Transitional character states (morphoclines) between zy-
gaenids and limacodids, as noted by Dyar (1894) and Fracker
(1915), are more fully understood with new or recently
discovered character data for megalopygids, aidids, dalcerids,
and limacodids. These morphoclines and the complexity of
larval synapomorphies provide strong support for unifying
these families with external-feeding larvae into one superfa-
mily (e.g., Zygaenoidea sensu Common) rather than separating
them into different superfamilies, as proposed by Brock (1971)
and Heppner (1984) (see "History of Classification of the
Limacodid Group," above).
Many transitional larval characters on the cladogram (Figure
326) involve reductions, except for the addition of two pairs of
abdominal prolegs with crochets in the limacodid-group
hypothetical ancestor (characters 25,26). Graphical representa-
tion of the proleg transformation series (Figure 327) illustrates
reduction in relative proleg length from the ancestral zygaenid
condition. In the limacodid-group ancestor, these transitions
involve the shortening and widening of prolegs, reduction of
the anal plate (character 34), and compression of the ventral
portion of segments A8 and A9 (character 29). A change in the
L and SV rows from multiple secondary setae to primary setae
in the aidid + (limacodid + dalcerid) clade can be interpreted as
a reduction (Figure 327) or as a change in the developmental
clock in which primary setae do not transform or progress into
verrucae (with multiple secondary setae) (see further discussion
under "Homology and Evolution of Larval Body Setae,"
below). Reductions in the limacodid + dalcerid hypothetical
ancestor include (1) thoracic leg size (character 23); (2) size of
the VI seta at the base of thoracic legs and of V and SV setae
on the abdomen (character 35); and (3) proleg size and crochet
size and number. Shagreened cuticle has apparently been lost
68 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
TABLE 6.?Apomorphic characters in the limacodid group.
Taxon Immature Adult
LIMACODIDAE + DALCERIDAE
AIDIDAE + (LIMACODIDAE + DALCERIDAE)
MEGALOPYGIDAE + (AIDIDAE +
(LIMACODIDAE + DALCERIDAE))
LIMACODIDAE*
DALCERIDAE
AIDIDAE
Synapomorphies
1. fungiform setae (SV and V) on
ventral surface
2. crochets absent in 1st 2-3 instars
3. prolegs highly reduced on A2-A7
4. brush-like spinneret
5. reduced thoracic legs
6. stemmata with S1 seta only
7. anal fork absent
8. elongate axial setae
9. frass-flipping setae
10. ventral surface smooth
1. plumose setae absent
2. primary L setae on abdomen
1. abdominal prolegs on A2-A7,
with crochets
2. length of A8-A9 reduced be-
neath
3. anal plate reduced, without ver-
rucae
4. prolegs short
5. pupal maxilla shorter than labial
palpus
6. sculpted flange on pupal eyepiece
1. chaetosemata absent
2. reduced sclerite on mesepimeron
3. gnathos present
1. papillae anales open and flat,
cleft
2. labial palpus porrect, reaching
vertex or beyond
3. hindwing Sc + R, branches from
base of discal cell (both sexes)
1. ocelli absent
2. short galeae
3. female 5th tarsomere with dense
sensilla trichodea beneath
4. branched forewing R, and R4 or
R4 and R5 stalked beyond dis-
cal cell
Autapomorphies
1. crochets absent, all instars
2. flat, thin eggs
3. extended pupal maxilla, contigu-
ous with labial palpus
4. hard, ovoid cocoons, lid invisible
when uneclosed
1. gelatin-covered setae, all instars
2. cocoon without lid, with ovoid
inner chamber
3. L setae on abdomen above or in
line with spiracles
1. spinneret with amorphous tip
2. cocoons with 1 or 2 pairs of holes
in outer mesh
1. female legs with recessed pad of
sensilla trichodea underneath
2. disk-shaped papillae anales, en-
tire
1. uncus highly reduced
2. valvae reduced or fused
3. female accessory glands with
external ducts
4. forewing R2 and R3 stalked
1. forewing R, and Rj stalked
2. male antennae bipectinate in
basal two-thirds
MEGALOPYGIDAE 1. digitate sensilla proximal to all
spiracles
2. prolegs A2-A7 with large,
membranous pads
1. dense pack of pilose scales on
end of female abdomen
2. valvae reduced to small digitate
lobes
3. forewing R2-R4 pectinate off R5
* Characters in Umacodidae (+ Chrysopolomidae) do not include those of Pantoctaenia and Crothaema (except autapomorphy 3 of
immature and autapomorphy 2 of adult).
NUMBER 582 69
TABLE 7.?Tree length (L), consistency index (CI), and retention index (RI) for
phylogenetic analyses. The second group of analyses included the additional
terminal taxon Pantoctaenia/Crothaema.
Life stage
Adults
Immatures
Combined data
Unordered
L
42
34
76
70
First group of
analyses
CI
0.73
0.91
0.81
0.88
RI
0.54
0.85
0.68
0.75
Second group of
analyses
L
44
37
82
75
CI
0.70
0.86
0.76
0.84
RI
0.59
0.85
0.71
0.79
on ventral A1-A9 in the limacodid + dalcerid ancestor
(character 30). Reductions within Limacodidae (excluding
Pantoctaenia) include the complete loss of crochets and the
reduction in size of thoracic leg setae.
Genitalia do not follow smooth transitions as in characters of
the immature stages, perhaps due to homology problems. The
gnathos in the limacodid + dalcerid clade (character 13) is of
questionable homology with the structure found in the
Tineoidea and the lower Ditrysia because it has been lost in the
Zygaenidae (Kuznetsov and Stekol'nikov, 1981; Common,
1990), Megalopygidae, and Aididae. A gnathos-like structure
may have been derived, perhaps from transtillar lobes like those
found in the Chrysopolominae, as suggested in zygaenids by
Kuznetsov and Stekol'nikov (1981). Although not coded as
having the same character condition, independent reductions of
the male valvae in megalopygids and dalcerids may be related
to specialized aspects of the female ovipositor (character 15).
The dense clump of piliform scales on female megalopygids
(character 17) and the external ducts for sticky secretions in
dalcerids (character 18) could make lateral clasping by typical
valvae more difficult. The absence of large accessory glands in
female limacodids and megalopygids (character 18) may be
Zygaenidae Megalopygidae Aididae
considered convergent, secondary losses. This is supported by
the presence of these glands in other members of the limacodid
group and in the Zygaenoidea; the glands occur in dalcerids,
aidids, zygaenids, cyclotornids, and epipyropids.
There is less homoplasy in the two analyses of immature-
stage characters (CI = 0.91 and 0.89) compared to the two
analyses of adult characters (CI = 0.73 and 0.70). Examples of
homoplasy in the larval characters include the absence of
crochets on segments A2 and A7 in aidids (character 26). This
is likely to be a secondary loss rather than the plesiomorphic
condition found in zygaenids (see further discussion under
"Prolegs and Crochets," below). The close proximity of
stemma 5 to stemma 4 in limacodids (+ Pantoctaenia) may be
a reversal to the zygaenid condition (character 19); however,
some confusion exists due to variation of this character state
within zygaenids (see "Morphology," above).
Examples of homoplasy in the adult characters, in addition to
those given above, include the number of spurs on the hindtibia
(character 10). The configuration 0-2-4 occurs in zygaenids and
does not appear in my cladogram until limacodids, indicating a
gain from the 0-2-2 condition in megalopygids and aidids. The
absence of tibial spurs in the Dalceridae is hypothesized to have
been from a common ancestor with limacodids that had 0-2-2
or 0-2-4. Limacodid genera with the largest number of
Dalceridae
Limacodid + Dalcerid
Ancestor
FIGURE 327.?Character transformation of lateral and ventral surfaces in limacodid-group larvae, based
on results of the phylogenetic analysis. The hypothetical ancestor of Limacodidae + Dalceridae is similar
to limacodids considered to be primitive (e.g., Pantoctaenia). The limacodid figured represents a member
of a group with spiny larvae, considered to be a more derived lineage. (D, L, SD, SV, V = dorsal-, lateral-,
subdorsal-, subventral-, and ventral-group setae or derivatives.)
Limacodidae
70 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
plesiomorphies in the larval stage (e.g., Pantoctaenia) all have
the 0-2-4 condition. The branching of forewing R4 and R5
(character 6) in dalcerids appears to be convergent with the
condition found in megalopygids.
Alexander (1990), for nomadine bees, and Miller (1991), for
notodontid moths, also found more homoplasy in cladograms
generated using characters of adults compared to those
generated from characters of immature stages. Underlying
causes were believed to include problems of homology in adult
genitalia (Miller, 1991) and slower evolution and the conserva-
tion of novel features in larval forms (Alexander, 1990). The
occurrence of larval transitional forms in my cladogram may be
a byproduct of slow evolution, as referred to by Alexander
(1990). In contrast to Miller (1991), I found the homoplasy in
adults to be equally distributed in the three body regions. In
several examples, I suspect that homoplasy resulted from loss
of characters that were found in ancestral taxa. This is based on
the presence of these ancestral states in both lower and higher
lineages on the cladogram (Figure 326). Examples include the
absence of secondary accessory glands in megalopygids and
limacodids (character 18) and the absence of galeae in
megalopygines and aidids (character 3). Similar examples in
the larval stage include characters 27 and 32, discussed above,
and the absence of crochets on A2 and A7 in aidids and in a
number of megalopygids and dalcerids (character 26) (see
"Evolution of the Ventral Abdominal Surface," below).
Keys to Limacodid -Group Families and Allies
The following keys to adults and larvae primarily discrimi-
nate between members of the limacodid group and their allies,
which together consitute the Zygaenoidea (sensu Common,
1970). Adults in the limacodid group can be separated from
similar Cossidae, Psychidae, and other lower Ditrysia by the
absence of the stem of medial veins within the discal cell of the
hindwing. They can be separated from the Lasiocampidae,
Lymantriidae, and other higher Ditrysia by the presence of a
strong CuP in the forewing and the presence of veinlets in the
forewing discal cell. Larval Lycaenidae and Riodinidae can be
easily confused with Zygaenoidea; lycaenids often have
retractile heads, and both groups have a variety of dorsal setae
and have short prolegs on A3-A6 and A10, giving them a
slug-like appearance. They can often be separated from
Zygaenoidea by the mesoseries being divided by a membra-
nous pad. Larval Olona (Limacodidae) are included in the key
based solely on figures and descriptions in the literature (e.g.,
Holloway et al., 1987) because I have been unable to examine
them.
Key to Adults
1. Ocelli present; commonly diurnal 2
Ocelli absent; commonly nocturnal 4
2. Forewing R with veins stalked from end of discal cell; females winged
ZYGAENIDAE (in part)
Forewing R unstalked from discal cell; females winged or apterous 3
3. Wings smokey colored, sparsely scaled and nearly equal in size (males), or wings
absent (females); proboscis reduced or absent; Europe and Africa
HETEROGYNIDAE
Wings often brightly colored, wings present in both sexes; proboscis usually long;
Cosmopolitan ZYGAENIDAE (in part)
4. Forewing R unstalked or only R4 and R5 stalked 5
Forewing R stalked in various combinations or [rarely] wings absent; if R4 and R5
stalked without R3 from discal cell, R2 and R3 stalked or fused [Figures 72, 74]
6
5. Forewing triangulate, larger than hindwing; tibial spurs absent; small and dark with
bipectinate antennae [Figure 42]; mostly pantropical EPIPYROPIDAE
Wings rounded, nearly equal in size; tibial spurs 0-2-4; small with filiform
antennae; Australia CYCLOTORNIDAE
6. Head with sculptured process or less developed wrinkles on frons; chaetosemata
present; epiphysis present or absent; Africa or Mediterranean [Figure 47] [wings
absent in female Sombrachys] SOMABRACHYIDAE
Head without process or wrinkles on frons; chaetosemata present or absent;
epiphysis absent 7
7. Forewing R4 and R5 stalked (if fused, R2 and R3 fused) 8
Forewing R3 and R4 stalked (if independent, branch off R2) 9
NUMBER 582
8. Forewing R2 and R3 independent [Figures 68, 70]; hindwing Sc + R, fused with Rs
along discal cell (all females, males in part) [Figure 68]; labial palpi short and
hidden by surrounding scales [Figures 53a,b]', chaetosemata present; tibial spurs
present (0-2-2); females with dense pack of woolly scales on end of abdomen . .
MEGALOPYGIDAE
Forewing R2 and R3 stalked or fused [Figures 72, 74]; hindwing Sc + R,
independent of Rs along discal cell except at base [Figure 75]; labial palpi curved
upward and easily visible [Figures 55, 57a]; chaetosemata absent; tibial spurs
absent; females with normal scales on end of abdomen DALCERIDAE
9. Forewing R} and R2 stalked [Figure 69]; chaetosemata present; Neotropics . . . .
AIDIDAE
Forewing R, and R2 independent; chaetosemata absent 10
10. Frenulum and gnathos absent; hindwing large, nearly equal in size to forewing
[Figure 50]; forewing R3 and R4 stalked or independent, branch off R2; male
antenna bipectinate to apex; Africa and M a d a g a s c a r . . . . CHRYSOPOLOMINAE
Frenulum and gnathos present; hindwing usually smaller than forewing [Figures
51, 52]; forewing R3 and R4 stalked (unless fused), branch off R5, R2, or
in-between [Figures 71, 73]; male antenna bipectinate to apex, filiform, or many
conditions in between; Cosmopolitan LIMACODIDAE
Key to Larvae
1. Dorsum with spiny and/or hairy verrucae, scoli, or tubercles or with smooth,
nondeciduous warts 2
Dorsum without verrucae, rarely with primary setae covered with gelatinous warts
that are deciduous and translucent 8
2. Prolegs on segments A3-A6 and A10 3
Prolegs on segments A2-A7 and A10, or ventral surface slug-like 5
3. Dorsum with smooth warts. ZYGAENIDAE (in part)
Lactura, see Common, 1990
Dorsum with hairy or spiny verrucae 4
4. Verrucae with spiny setae ZYGAENIDAE (in part)
Verrucae without spiny setae; Africa and Mediterranean . . . . HETEROGYNIDAE
5. Dorsum of T2, T3, and A1-A9 with spinose vemicae or scoli, or hairy tubercles; L
row with 2 primary, often hair-like setae; slug-like ventral surface frequently with
visible suckers (A1-A8) [Figure 222] LIMACODIDAE (in part)
Dorsum of T2, T3, and A1-A9 with spinose verrucae; L row with verrucae or
smooth warts; prolegs distinct on A2-A7 and A10 [Figures 215-218, 223] . . 6
6. Each spiracle with 1 digitate sensillum [Figures 328-334]; anterior portion of
prolegs with membranous pads and 2 stout SV setae [Figures 256,257]; D and SD
verrucae often bearing both plumose and spinose setae [Figures 166, 169]; L and
SV rows with verrucae; New World MEGALOPYGIDAE
Each spiracle without digitate sensillum; prolegs without membranous pads; D and
SD verrucae bearing primarily spinose setae [Figures 167, 336, 337]; L and SV
rows with or without numerous setae 7
7. Scale-like setae dorsal to spiracles or oval pouch filled with minute hairs; L and SV
verrucae with hair-like and spinose setae (late instars); D and SD verrucae with
spinose [Figure 167] or plumose setae; anal comb present [Figure 280]; Africa
and southern Mediterranean SOMABRACHYIDAE
No scale-like setae or oval pouch dorsal to spiracles; L and SV rows without
verrucae, L row with single primary setae on 2 horizontal warts below spiracle
[Figure 339]; D and SD verrucae with spinose setae [Figures 336, 337]; small
anal fork present or absent; Neotropics AIDIDAE
72 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
8. Prolegs or crochets present on segments A3-A6 and A10; crochets forming a
penellipse or mesoseries; dorsum sometimes with waxy excretions; gelatinous
warts absent 9
Prolegs reduced, ventral surface slug-like, with or without suckers or crochets;
dorsum without waxy excretions; gelatinous warts present or absent 10
9. Crochets in a penellipse; Cosmopolitan EPIPYROPIDAE
Crochets in a mesoseries; Australia CYCLOTORNIDAE
10. Dorsum smooth or with spinules 11
Dorsum with gelatinous tubercles 12
11. Long black primary setae with white apex on dorsum; dorsum spinulate [Figure 5];
Africa and Madagascar CHRYSOPOLOMINAE
Short primary setae on dorsum; dorsum smooth or spinulate [Figure 7];
Cosmopolitan LiMACODlDAE (in part)
12. Pigmented pattern on dorsum beneath gelatinous warts; Neotropics and south-
western United States DALCERIDAE
Without pigmented pattern on dorsum beneath gelatinous warts; Southeast Asia
Olona (Limacodidae)
Limacodid-Group Families and Allies
This section presents adult diagnoses for each limacodid-
group family. Redescriptions for poorly known adults and for
immature stages are included. Also presented under each
family are discussions of relationships and taxonomic position,
as well as reinterpretation of character polarity and homology
based on evidence to date. Family status for Aididae reflects the
results of my phylogenetic analysis, and Chrysopolomidae is
placed as a subfamily of Limacodidae based on morphological
evidence presented herein. Often included as a subfamily of
Megalopygidae, Somabrachyidae is put at the family level for
reasons stated below. Discussion of Epipyropidae and Cy-
clotornidae does not imply membership in the limacodid group.
MEGALOPYGIDAE Herrich-Schaffer, 1855
(MEGALOPYGINAE and TROSIINAE)
ADULT DIAGNOSIS (Figures 43-46).?Megalopygidae dis-
tinguished from other New World limacodid-group families by
forewing R,_5 forming pectinate branch from radial sector
(Figures 68, 70); hindwing Sc + R, anastomosed with Rs for
entire length of discal cell (except male Megalopyginae)
(Figure 68); wing scales piliform or cleft, often tripartite;
labial palpi short, one- or two-segmented, often hidden by
surrounding scales (Figure 53a,b); female abdomen with dense
tuft of piliform scales on A7 (Figure 46); and female genitalia
with papillae anales laterally flattened and medially appressed
(Figures 345, 346). Separated from Aididae by male antennae
bipectinate to apex (Figures 43-45) and by fifth tarsomere of
female with sensilla trichodea beneath intermixed with scales
(Figures 76-81); female antennae bipectinate as in males or
serrate. Male genitalia differing from Aididae and Limacodidae
by short, narrow uncus and small digitate valvae dorsal to juxtal
lobes (Figure 351). Mesonotum with longitudinal depressions
(Figure 60) and mesepimeron broadly sclerotized (Figure 67).
Anterior and posterior apophyses of female well developed
(Figures 345, 346).
LARVA (Figures 1, 2, 328-334).?Head: Stemma 5
ventrally disjunct by greater than 1 stemma width from others,
stemma 4 closer to stemma 6 than to stemma 5 (Figure 119); S1
and S2 setae present (Figure 119); labrum without small
epipharyngeal spines on anterior margin (Figures 131-133);
spinneret tubular and symmetrical, with dorsal opening near
apex (Figures 145-147).
Dorsum: Prothoracic shield with D and SD verrucae
combined and with 1 L and 2 distinct SV verrucae below
(Figure 166); T2, T3, and A1-A9 with D, SD, and L
verrucae of long plumose setae and short, frequently hidden,
urticating setae (Figure 169) and with L3 and SV verrucae with
plumose setae; T2 with Dl and D2 verrucae contiguous and
partially distinct, whereas other D verrucae completely fused;
digitate sensilla proximate to each spiracle (Figures 328-334).
Ventrum: Thoracic leg with short, axial seta beneath claw
(Figures 228, 229); abdominal prolegs with membranous pad
present on segments A2-A7 (Figures 216, 218) and 2 broad
setae anterior to and near pad (Figures 252-257). A2 and A7
with or without crochets; crochets on A2-A7 uniordinal in
V-shaped mesoseries, divided into anterior and posterior
portions or entire (Figure 256); anal proleg (A 10) with
undivided mesoseries (Figures 278, 284); anal fork with
emarginate tip dorsal to anus, below D and SD setae on A10
(Figure 278).
PUPA (Figures 299-302, 313, 320).?Labium of nearly
equal width throughout, not tapered toward tip, projecting only
NUMBER 582 73
FIGURES 328-334.?Spiracular sensilla in Megalopygidae (scale length in parentheses). 328-330, Megalopyge
crispata: 328, Tl and T2 (anterior to left), arrow points to Tl spiracle and sensilium (1 mm); 329, Tl spiracular
sensillum and L vemicae (anterior to right) (200 fim); 330, Tl spiracle and sensillum, detail of 328 (see arrow)
(100 urn). 331, 332, Mesocia pusilla: 331, A1 sensillum (anterior to right) (60 urn); 332, detail of sensillum (3
Jim). 333, 334, Megalopyge sp., 1st instar (BMNH): 333, prothoracic spiracle and sensillum below (23.1 Jim);
334, spiracular sensillum, L vemicae, SV setae, and proleg on 3rd abominal segment (50 nm). (Cr = crochets, L,
SV = lateral-, subventral-group setae, SpS = spiracular sensillum.)
74 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
short distance below lateral portion of maxilla (compare with
Limacodidae, Dalceridae, and Aididae). Mesonotum with
longitudinal depression as in adults (Figure 300). Spiracles on
A1-A6 with lobes similar to digitate sensilla of larva in
females (Figure 302); lobes only distinct on Al in males.
Unopened cocoons (Figure 14) often with visible escape hatch
(compare with Limacodidae).
DISCUSSION.?Character states of adult Megalopygidae are
given by Hopp (1927, 1934). He divided Megalopygidae into
three subfamilies (Megalopyginae, Trosiinae, and Aidinae) and
suggested that trosiines were the most generalized mega-
lopygids based on hindwing venation and male genitalia
(Hopp, 1934).
Although there is strong evidence to support a monophyletic
Megalopyginae + Trosiinae (Figure 326, Table 6), Hopp's
generic assignments for these subfamilies require future work.
A sister-group relationship between "Aidinae" and Trosiinae or
Megalopyginae is not supported by synapomorphies. Further-
more, to be monophyletic with either megalopygid subfamily
the synapomorphies between Aididae and Limacodi-
dae + Dalceridae (Figure 326, Table 6) would have to have
been independently derived in "Aidinae." If new information or
a new interpretation of character coding were to support placing
Aididae back in the Megalopygidae, perhaps a more plausible
hypothesis would be Aidinae as the sister group to Tro-
siinae + Megalopyginae.
Interpreting the homology of male genitalia in the Mega-
lopygidae (Figure 351) is facilitated by character states of male
genitalia in the Aididae (Figures 352, 353). Eyer (1924) and
Hopp (1927, 1930) divided the "valva" found in megalopygi-
nes and trosiines into a dorsal harpe (= cucullus in Eyer) and
ventral sacculus. Neither of them described the male genitalia
in aidids, although Hopp (1934:1073) discusses the greater
structural modification in trosiines compared to aidines or
megalopygines. Birket-Smith (1974) referred to a lobe mesal of
the valva and lateral to the juxta in the Cossidae as the valvella.
This structure may be homologous with the juxtal lobes, or
processes, found in aidids (Figures 352, 353) and dalcerids
(Figures 354, 355). Applying this concept to the Mega-
lopygidae, the dorsal lobe is simply the valva, and the lower
lobe is the juxtal lobe (Figure 351). The condition found in
megalopygids could have been derived from the one in aidids
by reduction of the broad valva to a digitate structure, with the
juxta remaining as the ventral lobe (further discussed in
Dalceridae section, below).
In my phylogenetic analysis, I treat the fusion of hindwing
Sc with Rs along the discal cell as plesiomorphic in the
Megalopygidae, based on the condition in the Zygaenidae. This
was, in fact, the interpretation used by Hopp (1934); according
to him, the character went from complete anastomosis
(Trosiinae) (Figure 68) to having only the basal portion
fused (Megalopyginae) (Figure 70) to having the vein free for
the entire discal cell ("Aidinae") (Figure 69). Because the free
condition is plesiomorphic for Lepidoptera, however, the
polarity of this character is uncertain. It is possible that the
fused condition in megalopygids evolved independently of
zygaenids, and the condition in Aididae is plesiomorphic.
Traditionally, trosiine larvae (Figure 2) are regarded as being
less pilose than their megalopygine relatives (Figure 1).
Neither Mesocia (Trosiinae) nor Megalopyge lanata (Mega-
lopyginae), however, fits this characterization; the former is
densely pilose, whereas the latter exhibits verrucae in distinct
clumps without pilose setae. One new character of the adults
may have utility in separating the two subfamilies. The ventral
surface of the fifth tarsomere in female trosiines (Figures 76,
77) appears to have fewer scales versus sensilla trichodea than
in female megalopygines (Figure 79).
Fracker (1915:97) described stemma 5 as being close to
stemma 4 in megalopygine species Megalopyge (=Lagoa)
crispata but distant from stemma 4 in M. opercularis (J.E.
Smith); however, stemma 5 appears to be distant for M.
crispata as well (Figure 119). Furthermore, Fracker only
mentioned trosiine species Norape (= Carama) cretata (Grote)
as having equal distance between stemmata 4 and 1 and
between stemmata 4 and 5; this condition also occurs in M.
crispata.
Homology and function of the spiracular sensilla (Figures
328-334) in megalopygids are not well understood. Packard
(1894) indicated that homology with primary setae was
unlikely; however, the close proximity of SD2 and L setae to
the spiracles in dalcerids and in some limacodids on A1-A8
suggests otherwise. According to Packard's (1894:289) Neo-
Lamarckian view, the sensilla evolved from everted glands or
osmetena, which through disuse lost both the ability to retract
and secrete "malodorous fluid." He also thought they might be
"pleuropodia, or homologues of the temporary embryonic
abdominal legs of lower insects."
The spiracular sensilla appear to have a defensive function.
When disturbed in the vicinity of the sensilla, the larva will
move both SD and L rows of the urticating verrucae towards
the sensilla/spiracles in what appears to be a defensive posture;
to my knowledge, this was first reported in a megalopygid,
misidentified as Aidos amanda, by Hoffmann (1932). Similar
structures occur in close association with spiracles of pupae
(e.g., Megalopyge lanata) (Figure 302) (Mosher, 1916), as well
as in some adult females.
Crochets equivalent in size and number to those on A3-A6
may occur on prolegs A2 and A7, contrary to previous
descriptions of megalopygid larvae (Packard, 1894; Dyar,
1899b; Hopp, 1934). Unfortunately, I have been able to
associate only one species possessing these "extra" crochet
segments, Mesocia pusilla (Trosiinae) from Colombia, with
adult vouchers. I examined other undetermined larvae sharing
this configuration, including another very distinct trosiine from
Veracruz, Mexico (USNM), two presumably megalopygine
larvae from Argentina (BMNH), and other specimens from
Colombia and Ecuador (USNM).
The size of the membranous pads varies within individuals.
NUMBER 582 75
They can be either uniform between prolegs or much larger on
A2 and A7, especially in species that lack crochets on these two
segments (Figure 218). Larvae of Mesocia pusilla and the
trosiine from Veracruz, Mexico, have a further modification. In
these, the SV row of verrucae is modified into a membranous
pad or suction disk, similar to that found on the proleg (Figures
216, 252). This increases the surface area for contact with the
substrate and may be of particular utility on smooth surfaces (as
discussed in "Function and Adaptation," below).
AlDlDAE Hopp, 1927, new status
ADULT DIAGNOSIS (Figure 48).?Only member of lima-
codid group with forewing R, and R2 stalked alone near costal
margin (Figure 69). Distinguished from Megalopygidae by
forewing R3 and R4 stalked from R5 (Figure 69); papillae
anales flat and open (Figure 347); without dense pouch of
hair-like scales on end of female abdomen; male genitalia with
uncus and valvae broad (Figures 352, 353); sensilla trichodea
on ventral fifth tarsomere of females without scales intermixed;
male antennae bipectinate in basal portion; female antennae
filiform; labial palpus porrect and usually three-segmented,
longer and less hidden by scales (Figures 54a,b). Distinguished
from Limacodidae and Dalceridae by presence of chaetose-
mata; absence of a gnathos (males) (Figures 352, 353);
presence of longitudinal depressions on mesonotum (Figure
61); broadly sclerotized mesepimeron (Figure 66). Separated
from Limacodidae alone by deeply cleft papillae anales and
large secondary accessory glands (females) (Figures 347, 348).
ADULT.?Head: Male antenna bipectinate from one-half
to two-thirds distance from base (Figure 48); labial palpus
either two- or three-segmented (Figures 54a,b), porrect,
reaching as high as dorsum of eye; galeae and maxillary palpi
absent.
Thorax: Forewing cell with stem of medial vein from near
base, branched in a V near distal end; R, and R2 stalked in distal
one-third of wing beyond discal cell and near costa; R3 and R4
stalked from R5 (Figure 69) (Dyar, 1895c, fig. 21); scales on
forewing and legs often spatulate; mid- and hindlegs with 1
short pair of tibial spurs; ventral sensilla trichodea on fifth
tarsomere of female without scales interspersed medially
(Figures 82-84).
Male genitalia (Figures 352, 353): Uncus gradually taper-
ing from tegumen, apex downcurved, gnathos absent; aedeagus
slightly curved, narrowed at tip; valvae broad, entire, and
somewhat triangulate with rounded apex; juxtal process
(valvella) on each side of aedeagus with upturned, clawed end;
vinculum over one-half width of tegumen, saccus weakly
developed.
Female genitalia (Figures 347, 348): Papillae anales
flattened to posterior, somewhat divided into dorsal and ventral
regions, with dorsal portion fused medially; both anterior and
posterior apophyses longer than A8; corpus bursae oval and
membranous; spermatheca connected with short ductus semi-
nalis near ductus bursae; 2 large sebaceous glands fused at base
where connected to common oviduct; secondary accessory
glands with 2 medially connected and weakly sclerotized lobes,
occupying large proportion of A8; connection of secondary
glands to common oviduct unclear, although without external
ducts present in Dalceridae.
LARVA (Figures 4, 215, 223, 335-344, 407).?Head: SI
and S2 present, stemmata as in Megalopygidae except
stemmata 1 and 2 have wider gap between them (Figures 120,
121); labrum with stout spines on anterior margin (Figures 129,
130); spinneret tubular, with amorphous tip (Figures 148-
151).
Dorsum: Green, without pigment pattern; plumose setae
absent, short secondary setae sparse; verrucae with thick, black
urticating setae, 5-11 per cluster in D rows A2-A9 and SD
rows A2-A7, and fewer urticating setae in D and SD rows on
T2; urticating spines sunken and appressed to a common point
when in repose (Figure 336), with D row directed ventrad and
SD row directed dorsad; spines splayed into rosettes when
protruded (Figure 337); spiracle on Al midway between
position of other spiracles and SD row of spines; LI and L2
setae tactile, borne on 2 warts, equidistant and proximoventral
to each abdominal spiracle; LI and L2 separated by longitudi-
nal ridge and sulcus from row of ventral warts (L3?) found
between convex lateral portion of proleg bases (Figures 339,
407).
Ventrum: Length of thoracic legs greater than width of
labrum and with unmodified setae (Figures 225-227); prolegs
on A2-A7 and A10 as in megalopygids, although without
membranous pads or plumose setae (Figures 215, 223);
crochets in V-shaped row (50-70 in number) on prolegs
A3-A6 (Figures 340-344), linear on A10, and absent on A2
and A7 (Figure 223); anal fork reduced or absent.
PUPA (Figure 312).?Similar to dalcerids except mandibles
nearly absent and fractures occur along eye suture during
eclosion; frons often with projecting hump; spiracular sensilla
absent. Cocoon spun between leaves, with preformed escape
hatch, and with 1 or 2 pairs of small, round holes in outer silk
mesh, giving appearance of emergence holes of parasitic
Hymenoptera (Figures 15, 17).
DISCUSSION.?Aidos amanda (Stoll) was "provisionally"
placed in Megalopygidae by Dyar (1895c :243-244; as Brachy-
codiori), although he was "in doubt whether to refer it to
Eucleidae [= Limacodidae]." For the most part, Aidinae has
remained in Megalopygidae ever since, although Forbes (1923)
considered it a family. Characters of adult Aididae are given by
Dyar (1895c), Hopp (1934-1935), and Holloway (1986). The
larva (lateral and ventral aspects), cocoon, and pupa were
figured and described by Dewitz (1878) based on descriptions
and specimens from Gollmer. Descriptions of Aidos larvae by
Hopp (1934) were less detailed. The cocoons also were
described by Klug (1836), Hopp (1930, 1934), and Forbes
(1942). Cocoon construction, life history, and host plant
information for the Aididae are given in Epstein (1995). Larval
76 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 335-339.?Morphology of the larval stage in Aididae (scale length in parentheses) (336,338 photos by
L. Minor-Penland; 337, 339 photos by C. Hansen): 335, Aidos amanda, ventral aspect of live larva, note silk on
ventrum (photo by K. Sandved). 336-339, Aidos sp.: 336, detail of SD verruca, spines in repose (1 mm); 337,
detail of verruca, partially opened (1 mm); 338, detail of prothoracic spiracle (1 mm); 339, lateroventral view,
anterior to left (arrow points to longitudinal ridge and sulcus) (5 mm).
and egg stages described for the Aididae by Hoffmann (1932)
(as Aidos amanda) were based on a misidentified mega-
lopygine (Epstein 1995).
Aidids have a number of character states not found in either
the Megalopygidae or Limacodidae + Dalceridae, some of
which are autapomorphic (Table 6) or transitional, and others
are plesiomorphic. Placement of "Aidinae" in Megalopygidae
would be based on symplesiomorphies with respect to the
limacodid group. Examples include the presence of chaetose-
mata on the adult head and the presence of larval prolegs on A2
and A7. Character states that the Aididae do not share with
megalopygids are outlined in the diagnosis above. Holloway
(1986) suggested that Aidos (Aidinae) either belonged within
Limacodidae or was closer to Limacodidae than to Mega-
lopygidae. He based this on characters of the forewing veins
(R3 and R4 stalked), male genitalia, and female ovipositor
lobes. The radial veins and several aspects of the male genitalia,
however, are possibly plesiomorphic with respect to the
limacodid group, and female genitalic characters suggest a
close relationship with Dalceridae (compare Figures 347-350).
NUMBER 582 77
Urticating spines in aidid larvae are borne on a verruca or
tubercle and are apparent only when in full defensive display as
a rosette (Figure 337). In repose they are somewhat sunken, as
noted by Hopp (1934), and appear as if a single spine (Figure
336). A similar condition occurs in Limacodidae in the Parasa
chloris complex, in New World Natada (D.H. Janzen, pers.
comm.), and in Australian Doratifera. A possible relationship
between Somabrachys and aidids (as Aidinae) was suggested
by Hopp (1934), based on the sunken urticating setae in the
larvae; however, the setae in aidids are much more similar to
those found in limacodids. The "sunken spines" in Somabra-
chys are actually spiracle-shaped pouches above the spiracles,
which contain deciduous pilose material (see Somabrachyidae
section, below).
Another larval character found both in the Aididae and in
some genera of Limacodidae is the loss of the SD tubercle on
Al, with the spiracle located dorsally.
It remains unclear whether the female glands in aidids
(Figure 348) are homologous with those found in other
Zygaenoidea. Their bilobed appearance most closely resembles
the glands of dalcerids (Figure 350) (Miller, 1993) and
zygaenids (Naumann, 1988). The female glands in epipyropids
and cyclotomids are composed of one part rather than two
(Common, 1990). The aidid glands may produce a sticky
secretion that coats the eggs, as reported for dalcerids (Miller,
1993). This is suggested by the observation that eggs are
sometimes found stuck to the outside of the ovipositor in
preserved specimens.
344
FIGURES 340-344.?Prolegs and crochets in larval Aididae (scale length in parentheses). 340-342, Aidos sp.
from larval exuvia in cocoon: 340, crochets on 2 adjoining prolegs (0.3 mm); 341, detail of crochets (107 \ua);
Ml, detail of 1 crochet (27 |im). 343, proleg on A3, upper left corner is anterior (1 mm); 344, A4 proleg, Aidos
amanda (meson to left, 1 mm).
78
SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
345 PA 346
347 348
349 350
FIGURES 345-350.?Female genitalia in Megalopygidae, Aididae, and Dalceridae. 345, 346, Megalopyge lanata
(Megalopyginae), Panama: 345, frontal view; 346, lateral view. 347, 348, Aidos amanda (Aididae), Brazil: 347,
frontal view; 348, lateral view. 349, Acraga coa (Dalceridae), Mexico, frontal view; 350, Dalcerides bicolor
(Dalceridae), lateral view (after Miller, 1994, fig. 252). (AA = anterior apophysis, AG = accessory gland,
AP = posterior apophysis. AT = anal tube, A7 = abdominal segment 7. CB = corpus bursae, DB = ductus bursae,
OB = ostium bursae, PA = papillae anales, SG = sebaceous gland, St = spermatheca.)
79
352 Te Un 353
Aed
354 355 356
Aed-A
357
FIGURES 351-357.?Male genitalia of the limacodid group: 351, Norape corporate (Trosiinae), Venezuela,
lateral view. 352, 353, Brachycodilla carmen (Aididae), Brazil (left valva removed): 352, lateral view; 353,
frontal view. 354,355, Minacraga aenea (Dalceridae) (after Miller, 1994, fig. 227): 354, lateral view; 355, ventral
view. 356, 357, Perola (Limacodidae): 356, P. producta, Costa Rica, lateral view (left valva removed); 357, P.
murina, Brazil, frontolateral view. (Aed = aedeagus, Gn = gnathos. Jx=juxta. Sa = saccus. So = socius,
Te = tegumen, Tr = transtilla, Un = uncus, Va = valva, Vi = vinculum.)
DALCERIDAE Dyar, 1898
ADULT DIAGNOSIS (Figure 49).?Forewing R2 and R3
stalked or occasionally fused (Figures 72, 74), unique in
limacodid group; R4 and R5 stalked or fused; wings orange,
yellow, or white, sometimes with shiny markings. Chaetose-
mata absent; labial palpus two-segmented, upturned, reaching
vertex (Figures 55, 57a); galea and maxillary palpus vestigial,
galea basally spinulose (Figures 57a-c). Antennae short, about
one-fourth length of forewing costa, bipectinate throughout in
male, often with scales on apex (Figure 49). Mesepimeron
(Figure 65) narrowly sclerotized compared to megalopygids
80 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
and aidids. Tibial spurs absent, tarsomeres with large spurs on
ventral surface (Figure 89); fifth tarsomere in female covered
with scales beneath and with a narrow distal band of sensilla
trichodea (Figures 85, 86). Male genitalia with uncus and
valvae reduced; socii and gnathos present; juxtal arms often
present; saccus narrow, extending cephalad (Figures 354-
355). Female genitalia with papillae anales deeply cleft;
accessory glands large, with external ducts (Figures 349, 350).
LARVA (Figures 3, 177, 180).?Head: Stemma 5 disjunct
ventrally, more distant than width of 1 stemma from stemma 4;
only SI seta present, located proximate to stemmata 1-3
(Figures 122, 123); labrum with dense anterior bristles on
mesal portion only (Figures 134,135). Spinneret wider at apex
than at base (Figures 152-156); in post-first instars, a
brush-like comb along apical margin, with deeply divided
bristles (Figures 152-155) and with dorsal surface reduced
medially, revealing a narrow, longitudinal silk pore (Figure
154).
Dorsum: Prothoracic shield translucent, with tactile pri-
mary setae present; XD, D, SD, and L all hair-like at middle to
late instars (the cuticle was too distorted to observe setae,
except L in first instars); 2 (first instar) or 3 (later instars) L
setae below spiracle (Figures 186, 187). D and SD setal rows
consist of thick primary setae covered with gelatinous warts on
T2, T3, and A1-A9; L seta on T2 and T3 covered with
gelatinous wart (Figures 177, 180); often with brightly colored
pigmented patterns on cuticle beneath gelatinous warts. T2
with Dl and D2, and SD1 and SD2, each in horizontal rows,
with LI below; tactile setae MSD1 and MSD2 anterior to SD
setae. T3 with setae as on T2 except groups vertical, 1 SD
present, and MSD setae absent. A1-A8 with Dl, D2, and SD1
setae (also with SD2 in late instars); A9 with 2 D and 1 SD
setae. Three tactile setae present below gelatinous setae on T2
and T3, perhaps representing 2 SV setae and 1 MV seta. Lateral
abdominal region between gelatinous setae and ventrum
spinulate; spiracles closer to ventrum than in Limacodidae; 2
hair-like L setae posterior to each spiracle, and 1 MSD2 seta
anterodorsal to spiracles fungiform (first instars) (Figures 208,
209) or hair-like (later instars) (Figures 177, 180).
Ventrum: Thoracic legs small, about equal in length to
width of labrum (Figure 112); axial seta inside pretarsal claw
hair-like (Figure 231); VI seta mesal of thoracic legs fungiform
(Figure 243). Ventral surface of thorax and A1-A9 with cuticle
translucent and flexible throughout, except narrow band of
spinulose cuticle on anterior portion of Tl (Figure 243);
shallow suckers on A1-A8 in later instars (Figures 261, 408),
with flexible cuticle lateral to suckers as in limacodids. Ventral
fungiform setae in horizontal row of 3, with SV1, SV2, and VI
from lateral margin to meson (Figure 262). Crochets (Figures
261,264) on A3-A6 small, few in number (up to 17), arranged
in a crescent between SV2 and VI setae (Figure 263); absent in
early instars (Figures 224, 262). Additional sets of crochets
sometimes added in late instars on A2 and A7 (e.g., Dalcerides
ingenita). Anal proleg (A 10) vestigial, spinulose, and without
crochets; with 8 primary setae, 6 on lateral lobe and 2 on mesal
lobe (Figure 276); frass-flipping setae either absent or present
on dorsomesal portion near the anus (Figures 286, 289).
PUPA (Figures 297, 298).?Eye with sculptured flange that
remains connected to eyepiece at eclosion (Figure 303). Labial
palpi one-quarter length of head, maxilla not reaching eye
flange (Figure 315). Four (2 groups of 2) or 6 (2 groups of 3)
setae present on frons (Figure 325). Cremaster hooks present
(Figures 309, 310); rows of spinules present on anterodorsal
portion of segments A2-A6 (Figure 311).
DISCUSSION.?Adult characters of the Dalceridae were
given by Miller (1994). The larva has been described by Dyar
(1925), Hopp (1928), and Stehr and McFarland (1985, 1987).
Dalceridae has been previously reported as being closely allied
to Epipyropidae (Jordan, 1928), Limacodidae (Dyar, 1925),
and Megalopygidae (Minet, 1986). Since achieving family
status, Dalceridae has been cited no fewer than six times as a
subfamily of Limacodidae (Miller, 1994), including citations
by Janse (1925), Hopp (1928), and Zerny and Beier (1936).
Forbes (1942:392-393) considered the larva, cocoon, and
forewing accessory cell of dalcerids to be primitive for
Zygaenoidea, suggesting that dalcerids were a lineage separate
from both limacodids and megalopygids. The division of
Dalceridae into the subfamilies Dalcerinae and Acraginae by
Orfila (1961) was confirmed by Miller (1994).
Monophyly of the Limacodidae and Dalceridae is supported
by a suite of larval synapomorphies (Table 6). New character
information for primitive limacodid genera, such as Pantoc-
taenia and Crothaema, may eventually support either the
inclusion of Dalceridae within Limacodidae or an expanded
concept of Dalceridae. Autapomorphies (Table 6) support the
Dalceridae as a monophyletic group, although modifications in
male genitalia and in forewing radial veins, and the loss of
tibial spurs, could have been derived within Limacodidae. The
Dalceridae, however, do have putative plesiomorphies in the
immature stages that are not observed in the Limacodidae
(+ Chrysopolomidae, Pantoctaenia, Crothaema), including (1)
a disjunct stemma 5 in the larva (Figure 122); (2) long setae on
the larval thoracic legs (Figure 231); (3) a pupal labial palpus
without contiguous maxillary lobes (Figure 315); and (4) setae
on the pupal frons (Figure 325). Although present in the
Zygaenidae (e.g., Onceropyga anelia Turner, see Common,
1990), the setae on the pupal frons have not been observed
elsewhere in the limacodid group. Female characters shared
with aidids include cleft papillae anales and a large female
accessory gland (compare Figures 347, 348 with 349, 350).
These may be considered synapomorphies, although the
accessory gland may also be plesiomorphic if it is homologous
with the gland found in Zygaenidae.
I concur with Forbes (1942) and Miller (1994) that the
forewing accessory cell (areole) is primitive (plesiomorphic) in
Dalceridae (Figure 74). It appears to have been independently
lost or fused in the subfamily Dalcerinae (Figure 72) (Miller,
1994) and in other members of the limacodid group. The
NUMBER 582 81
posterior vein of the areole is traditionally considered to be a
radial vein (Miller, 1994), with the other solitary cell vein
being medial. Among terminal taxa of the cladogram presented
(Figure 326), the radial vein in the discal cell has not been
found in the Aididae or Megalopygidae. This radial vein occurs
in some Chrysopolominae (sensu nova) (Ectropinae of Hering,
1937) and in other zygaenoids including the Epipyropidae and
some Zygaenidae (Common, 1990).
Based on the phylogenetic analysis and new character
information presented herein, I disagree with Forbes's (1942)
interpretation that the larva and cocoon of the Dalceridae are
primitive within the limacodid group. I consider the inner
chamber of dalcerid cocoons to be autapomorphic in the
Zygaenoidea, supported by its unique function in the removal
of the gelatinous material from the prepupa (see "Biology,"
above). It is unclear which aspects of the larva Forbes
considered to be primitive, although my phylogenetic evidence
does not support his contention. I discuss the polarity of
dalcerid larval characters in the body setae and the prolegs
sections, below.
Electron micrographs of early instars of Dalcerides ingenita
and of a dalcerid species from Ecuador (CMNH) (Figures 224,
262) did not reveal crochets. This confirms reports that crochets
do not occur in early-instar dalcerids (Stehr and McFarland,
1985, 1987). Crochets, reported on segments A2 and A7 in late
instars of D. ingenita and Acraga coa (Stehr and McFarland,
1985), have not been found in late instars of other dalcerid
species examined thus far. These include larval specimens in
the Acraga infusa complex from Colombia (Genty collection),
D. tijucana, and several undetermined specimens from Mexico,
Brazil, and Peru.
The homology of male genitalic structures in Dalceridae
proposed by Miller (1994) is supported by observations of
genitalic structures in the Megalopygidae, Aididae, and
Limacodidae. In Dalceridae, Minacraga (Figure 354, 355) and
Dalcera species have "juxtal processes" and "valvae" in similar
position and appearance to those found in the Megalopygidae
(Figure 351) and Aididae (Figures 352, 353). The gnathos in
the Dalceridae appears to be homologous with the structure in
the Limacodidae (Figures 356, 357).
LIMACODIDAE Duponchel, 1844
ADULT DIAGNOSIS (Figures 51, 52).?Forewing R3 and R4
stalked (Figures 71,73), rarely fused, branching from R2, R5, or
in-between; R, and R2 separate, not stalked as in Aididae. Only
limacodid-group family with hindwing Rs and M, stalked from
near the end of the discal cell or beyond (Figure 73) (found in
majority of species). Chaetosemata absent; labial palpus
three-segmented, often relatively long and upturned (Figures
56a,b); galea present and smooth compared to Dalceridae
(Figures 58, 59), or absent; maxillary palpus commonly
one-segmented (Figures 58, 59). Male antennae ranging from
bipectinate to filiform. Hindtibia with 1 or 2 pairs of spurs; fifth
tarsomere of female usually with dense mat of ventral sensilla
trichodea on recessed pad without scales interspersed (Figures
99-103). Male genitalia with well-developed gnathos, uncus,
and valvae (Figures 356, 357). Female genitalia usually with
large, flat ovipositer lobes, disk shaped or emarginate on lateral
margin, not deeply cleft as in dalcerids and aidids (Figure 362);
digitate and membranous lateral lobes often present on A8,
dorsal to anterior apophyses (Figure 362); corpus bursae with
signum present or absent.
LARVA (Figures 6, 7).?Head: Stemmata 1-4 nearly
evenly spaced; stemma 5 slightly more removed from stemma
4 than stemmata 1-4 are from each other, stemma 6 most
isolated, equidistant to stemmata 1 or 5; only S1 seta present,
proximate to stemmata 1-4 (Figures 126-128). Labrum often
with a large number of bristles covering anterior margin
(Figures 138-141), more extensive than in related families.
Spinneret shape variable, usually widest at apex (Figures
158-162, 367-370), although sometimes tapering distally
(post-first instar) (Figures 371-384) or narrow and more
tubular (first instar) (Figures 389-392); dorsal surface with
(Figure 385) or without longitudinal striae (Figure 367).
Dor sum (first instar): D and SD setal groups as tubercles
or warts on T2, T3, and A1-A9 (Figures 170-176, 181, 393).
L setae paired on Tl, proximate to spiracle (Figure 183-185);
L on T2 and on T3 as an individual tubercle (Figure 178, 179,
181), fused with SD row, or absent.
Dorsum (post-first instar): Larva with spiny scoli, hairy
tubercles, or verrucae (Figures 402, 405), or larva relatively
smooth, without tubercles, and with simple setae; rarely with
gelatinous warts similar to those of Dalceridae. L setae ventral
to spiracles on abdomen (Figures 204-207, 210-214), with
ventral seta L2 sometimes modified to fungiform (Figure 205)
and dorsal LI rarely spatulate (Figure 204); L setae on
relatively stiff longitudinal ridge (= flange of Holloway, 1986)
bordering flexible ventral surface.
Ventrum: Pretarsal claw on thoracic legs with axial seta
directly under hooked portion, seta hair-like and somewhat
closer to tip than to base (Figure 234) as in Dalceridae, or
reduced (Figure 233) or vestigial (Figure 239). Paired setae on
each side of thoracic legs shorter than in Dalceridae (compare
Figures 231 and 232, 233); tips on longest setae blunt or
capitate (Figure 273); setae longest on pretarsus, varying from
roughly equal to length of pretarsal claw (Figures 232, 234) to
reduced and often fungiform (Figures 237-239). Crochets
absent in first instars (Figures 266-270, 348), present only in
late instar Pantoctaenia on A2-A7 and A10 (Figures 358-
360). Ventral cuticle flexible and translucent, as in dalcerids,
with oval suckers on segments A1-A8, usually more defined in
later instars (Figure 222), and with flexible region between
lateral margin of suckers and flange below spiracles (Figure
409). Anal proleg as in dalcerids except crochets present in late
instar Pantoctaenia; usually spinulate and lobate with 5 or 6
primary setae on lateral lobe and 2 primary setae on mesal lobe
(Figure 277); frass-flipping setae occurring singly or in small
82 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
groups (Figures 277, 288, 290), or absent (Figures 295, 296).
Anal plate with Dl, D2, SD1, and SD2 setae present; anal comb
absent (Figure 277).
PUPA (Figures 314, 316, 318, 319).?Labial palpus long,
often one-half length of head, with parallel maxillary lobe
adjoining it laterally (Figures 316, 318). Eyepiece breaks at
suture during eclosion (Figures 304, 305) (compare Chryso-
polominae and Dalceridae). Maxilla with (Figure 316) or
without (Figure 318) a lateral extension; when present
extension holds foreleg beneath at femur-tibia junction.
Cremaster hooks present (Figure 308) or absent.
DISCUSSION.?Relationships with other family groups are
reviewed in discussions of the Dalceridae and Aididae above.
The Limacodidae have been divided into suprageneric groups
although these have not been widely accepted, due at least in
part to the broad geographic range of this family. Taxa in
limacodid generic complexes mentioned below are given in
Table 3.
Dyar's (1899a) classification of New York limacodids was
based on larval characters. He considered primitive limacodids
to have three rows of tubercles on thoracic segments in the first
instars, more closely matching the megalopygid and zygaenid
(= pyromorphid) condition. Dyar regarded the three-rowed
Phobetron complex ("Tropic hairy" larvae) as both the oldest
group and the group that gave rise to the two-rowed Parasa and
Natada complexes ("Tropic spined" larvae). Groups with three
rows of tubercles in first instars and without spiny tubercles in
late instars, Prolimacodes ("Tropic smooth" larvae) and Apoda
("Temperate smooth" larvae) complexes, constituted the other
main branch.
Tutt (1899) and Forbes (1923) gave tribal names to Dyar's
lineages. Crothaema was given monotypic subfamily status by
Hering (1955), but this was not accepted by Janse (1964).
Holloway (1986) suggested possible groupings based on the
signum of the corpus bursae in females.
Characters of Pantoctaenia and Crothaema, plesiomorphic
based on the phylogenetic evidence presented above, suggest
that these are the most primitive iimacodids examined. The
presence of crochets on A2-A7 and on A10 in Pantoctaenia
gemmans (Figures 358-360) is a condition previously found
only in some megalopygids. As in dalcerids, these crochets do
not appear until after the first instar; this follows the reasonable
assumption that the first instar of P. gemmans is like that of P.
prasina (i.e., lacks crochets). The close relationship between
these two species is supported by adult features given by Janse
(1964). The flattened end of the cocoon in Crothaema suggests
a hatch, a condition not reported in other limacodids (cocoons
in Pantoctaenia have not yet been examined). Another
symplesiomorphy of PantoctaenialCrothaema + Dalceridae is
the condition of the fifth tarsomere in females, where the
sensilla trichodea are confined primarily to the distal portion
(Figures 93-95). The ovoid egg in Pantoctaenia also
appears to be plesiomorphic, resembling that of megalopygids
and dalcerids rather than the flat, scale-like eggs of limacodids.
Some larval plesiomorphies found in Pantoctaenia and
Crothaema occur in other genera as well. These include the
presence of Dl, D2, SD1, SD2, and LI setae (tubercles) on T2
in the Apoda and Prolimacodes complexes (Figures 178, 179),
as well as the presence of two D and one SD tubercles on the
abdomen in these complexes and in Belippa (Figure 188)
(discussed further in "Homology and Evolution of Larval Body
Setae," below). Also potentially plesiomorphic is the smooth
anterior margin of the labrum (Figure 136), which occurs in
Apoda, Pantoctaenia, and Chrysopoloma (see below); as in
dalcerids, epipharyngeal spinules of Crothaema are visible
from above only in the medial region.
Comstock (1918) considered the presence of a crossvein
connecting Rs and M, at the end of the hindwing discal cell
(e.g., Apoda) to be the primitive condition for limacodids
(Figure 71), and he considered the stalked condition of
Rs + M, (e.g., Adoneta) to be derived (Figure 73). This is
supported by the presence of the crossvein in genera with
primitive larvae, such as Pantoctaenia and Crothaema (see
Janse, 1964), the Apoda complex (Packardia, see Comstock,
1918), Pseudonapaea (Common, 1990), and Demonarosa
(Holloway, 1986). The presence of a bisignate corpus bursae
in Pantoctaenia and Crothaema (Janse, 1964) suggests that
other character states of the signum (Holloway, 1986) may be
derived from this condition.
The digitate maxillary extensions of the pupa represent a
putative synapomorphy for a lineage comprising some species
in the Apoda complex (Chapman, 1893; Packard, 1895a) and
Chrysopoloma (Figure 317), as well as complexes with spiny
larvae, such as the Parasa complex (Figure 316) (Packard,
1895a; Mosher, 1916) and Natada complex. These maxillary
extensions do not occur in plesiomorphic groups, such as
Crothaema, the Prolimacodes complex (Figure 314) (Mosher,
1916), or the Phobetron complex. This is contrary to Dyar's
genealogy (1899a), in which the genera with spiny larvae form
a lineage with the Phobetron complex, and the genera with
smooth larvae form the other major lineage. Independent losses
of maxillary extensions appear to have occurred in two lineages
of spiny limacodids, the Natada complex (e.g., Euprosterna,
Figure 318) and the Parasa complex (e.g., Monoleuca, Figure
319).
Because of their position, the homology of limacodid juxtal
lobes (Figure 357) or processes is less certain than the
homology of those of aidids, dalcerids, and megalopygids.
Juxtal homologies, however, may be easier to defend in
limacodid lineages that have species with processes in both
transtillar and juxtal regions (e.g., Perola complex, Figures
356, 357).
Greater species richness in the Limacodidae appears to be
correlated with greater morphological diversity compared to
other families in the limacodid group. This diversity of
characters is probably due in large part to radiation into a
greater number of lineages, which of course increases with the
inclusion of Chrysopolomidae. The discovery of suites of
NUMBER 582 83
!
'j?
361
FIGURES 358-361.?Crochets in Pantoctaenia (Limacodidae) (scale length in parentheses). 358-360,
Pantoctaenia gemmans, ultimate instar: 358, ventral view, arrows point to suckers on Al and A8 (5 mm) (photo
by C. Hansen); 359, crochets mesal of sucker on A7, A10 proleg below (1 mm); 360, crochets on A10 (1 mm)
(359, 360, photos by L. Minor-Penland). 361, Pantoctaenia prasina, 1st instar (arrow points to position on A10
where crochets are expressed in late instar P. gemmans (50 nm). (Cr = crochets.)
plesiomorphies in Pantoctaenia illustrates that a number of challenge in arriving at a phylogenetic hypothesis and higher
character states once believed to be the ground plan of classification of Limacodidae will be to adequately sample the
Limacodidae have been derived within the family. The future diversity within its lineages.
84 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
CHRYSOPOLOMINAE Aurivillius, 1895, new status
ADULT DIAGNOSIS (Figure 50).?Forewing R3 and R4
sometimes unstalked, branching from R2. Male antennae
bipectinate to tip. Reduced sclerites in mesepimeron as in other
Limacodidae and in Dalceridae. Distinguished from other
limacodids by the absence of gnathos (males) and frenulum.
Female with lateral lobes on A8, as in Limacodidae, and
concave lateral margins of papillae anales (Figures 363, 364).
Female fifth tarsomere with recessed pad similar to but more
ovoid than that of other Limacodidae (Figures 96-98).
LARVA (Figure 5).?Head: One SI seta on head (Figures
111, 125), as in limacodids and dalcerids; stemmata arrange-
ment similar to limacodids. Labrum without multiple spines on
anterior margin commonly found in Limacodidae (Figure 137).
Dor sum: Spinulate, green or pink, with long, black,
hair-like setae having white tips, 1 seta in D row per segment,
and 2 L setae below spiracles; length of each seta over one-half
body width. Spinules evenly distributed over dorsum, longer
proximate to spiracles.
Ventrum: Crochets absent; thoracic legs as in Limacodi-
dae.
PUPA (Figure 317).?Maxillary lobe parallel to and adjoin-
ing labial palpus, lateral extension of maxilla adjacent to eye,
363
362 PA
364
FIGURES 362-364.?Female genitalia of Limacodidae and Chrysopolominae: 362. Perola producta (Limacodi-
dae), Costa Rica, frontal view. 363. 364. Achrocerides theorini (Chrysopolominae). Cameroon: 363, frontal view;
364. viewed from behind. (AA = anterior apophysis, AP = posteriorapophysis, CB = corpus bursae, DB = ductus
bursae. DS = ductus seminalis, LL = lateral lobe. OB = ostium bursae, Ov = common oviduct, PA = papillae
anales, SG = sebaceous gland, St = spermatheca.)
NUMBER 582 85
both digitate; mandibles very small. Sculptured eyepiece intact
on exuvia after eclosion, as in Dalceridae. Cocoon smooth,
oval, and hard, with circular lid, as in Limacodidae; constructed
just below ground surface (N. Duke, pers. comm.).
DISCUSSION.?Adult characters of the Chrysopolominae
(sensu nova) were described by Hering (1937, 1961). Pinhey
(1975:50-51) described the larva as "like Limacodidae
but. . . have no spines." He described the larva of Chrysopoloma
varia flaviceps as "bright pink, short and squat, resem-
bling ... jujubes... feeds on Gymnosporia." N. Duke (pers.
comm.) found a green fourth instar of Chrysopoloma similis
(Figure 5) on Maytenus heterophylla (Celastraceae) near
Pietersburg, northern Transvaal, South Africa (January, 1988).
Subfamily placement should be viewed as preliminary
because definitive status must await a higher classification for
the Limacodidae. Chrysopolominae (sensu nova) has been
considered to be closely allied with Limacodidae, a member of
the same superfamily (Zerny and Beyer, 1936), or unrelated
(Bomer, 1939). Likely synapomorphies with Limacodidae
or with lineages presently in the family include (1) absence of
larval crochets (except Pantoctaenia); (2) stemma 5 proximate
to stemma 4 (Figures 111, 125); (3) a hard, oval cocoon; (4)
lateral extension of the pupal maxilla (Figure 317); (5) portion
of pupal maxilla contiguous with labial palpus; (6) lateral lobes
on adult female A8 (Figures 363, 364); and (7) pads with
sensilla trichodea on the fifth tarsomere in females (Figures
96-98).
The synapomorphies above occur in limacodid genera with
smooth larvae, including the Apoda complex, and in spiny
larvae, including the Parasa and Natada complexes. Synapo-
morphies 4-7 are not found in limacodid genera presumed to
be primitive (e.g., Pantoctaenia, Crothaema) or in the
Dalceridae; synapomorphies 4 and 6 are absent in the
Phobetron complex, and synapomorphy 4 is absent in the
Prolimacodes complex. Thus, "smooth-larva" genera in the
Apoda complex appear to have closer affinity to chrysopolo-
mines than do other genera with smooth larvae (e.g.,
Prolimacodes complex, Crothaema). Potential synapomor-
phies with Limacodidae + Dalceridae include the presence of
SI setae only in larvae (Figure 125) and the narrow, sclerotized
band on the adult mesepimeron.
The absence of the frenulum and gnathos in chrysopolo-
mines (Hering, 1937) appears to be autapomorphic with respect
to limacodids. To my knowledge, the only plesiomorphic
character in chrysopolomines compared to limacodids is the
presence of a radial vein, forming an elongate accessory cell,
within the forewing discal cell. The presence of a smooth larval
anterior labral margin (Figure 137), rather than one with
multiple spines, also occurs in Pantoctaenia and Apoda.
Brock (1971:53) hypothesized that the forewing radial vein
arrangement of R3 and R4 branched from R2 in chrysopolomids
and in some limacodids is "advanced." This is supported by my
cladogram and by the condition in "primitive" limacodids. His
one example, Strigivenifera (= Chrysopoloma) venata, which
has R3 and R4 unstalked, differs from the stalked condition in
Chrysopolomides nivea (Aurivillius) and Chrysopoloma rudis
(Walker), figured in Hering (1937, figs. 1,2). Further study will
be required to determine which condition of the two veins
represents the ground plan of Chrysopolominae.
SOMABRACHYIDAE Hampson, 1920
ADULT DIAGNOSIS (Figure 47).?Forewing R,_5 with 2
veins fused; arise at upper distal part of discal cell with middle
2 veins stalked; females winged (Psycharium) or apterous
(Somabrachys). Hindwing Sc + R, fused with discal cell.
Chaetosemata form distinct band, more developed than in
megalopygids and aidids. Sculptured process (Somabrachys) or
less developed wrinkles (Psycharium) present on frons;
epiphysis present (Somabrachys) or absent (Psycharium).
LARVA.?Head: Stemma 5 disjunct, similar to Mega-
lopygidae and Aididae; 2 S setae present (Figure 118);
spinneret (Figure 144) and labrum as in Megalopygidae.
Dorsum (early instar Psycharium): D verrucae on T2, T3,
and A1-A9, with urticating setae only (Figure 167); solitary
seta (SD) proximodorsal to spiracle; L and SV verrucae with
tactile setae.
Dorsum (late instar Psycharium): Tl and A10 with D and
SD verrucae continuous and covered with entirely tactile setae;
T2 and A9 with small pair of D verrucae with short urticating
setae and few tactile setae; T3 and A1-A8 with 1 large D
verruca on dorsum, consisting of only urticating setae in
rosette. A1-A8 SD verrucae reduced to few urticating and
tactile setae anterior and proximate to spiracles; A9 SD verruca
either absent or merged with L verruca; indistinct patches of
flat, scale-like setae on anterior margin of T3 and ventral to D
verrucae on A1-A8, with smaller group posterior to D setae on
T2; hair-like setae ventral of D verrucae, extending to L
verrucae along posterior margin of T3 and A1-A8. Dense
patch of tiny detachable spinules on Al, between spiracle and
D verruca, not easily distinguished from pigmented region in
same position on other segments. L verrucae proximoventral to
spiracle with short urticating setae and long tactile setae; SV
verrucae on lateral portion of proleg bases and lateral to
thoracic legs with only tactile setae.
Dorsum (late instar Somabrachys): All verrucae, T1-T3,
and A1-A9 with urticating and tactile setae; T2 and T3 with
Dl, D2, SD, L, and SV verrucae; A1-A8 each with 1 D and 1
SD verruca and a shallow, recessed pouch, dorsoposterior to
spiracle and SD verruca, lined with thin, detachable hair-like
scales (detachable scales also line spiracles); A1-A9 with 2 L
verrucae and 1 SV verruca.
Ventrum: Thoracic legs large in relation to prolegs,
especially in Psycharium (Figure 217). Two spatulate setae
near claw on thoracic legs (Figure 230) (early instar Pychar-
ium). Prolegs without membranous pads, segments A2 and A7
86 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
without crochets (Figures 249,250); mesoseries in later instars
divided into 2 parts, anterior set with fewer crochets. Early
instar prolegs with 2 spatulate setae on anterior margin
(Psycharium) (Figures 249-251). Late instar prolegs with up
to 7 stout, spinose SV setae anterolateral to crochets (Psychar-
ium) or with normal hair-like setae lateral to crochets
(Sombrachys). Anal plate equal to A9 above, with numerous
tactile setae. Anal comb present in Psycharium (Figures 279,
280, 282) and Somabrachys (Jordan, 1909).
DISCUSSION.?The relationship between Somabrachys and
Psycharium, as well as the relationships between these genera
and the limacodid group, are still not well understood;
however, revisionary work on Somabrachyidae is now in
progress (H. Geertsema, pers. comm.).
The placement of Somabrachyidae in Megalopygidae is
problematic because most characters shared between the two
groups, such as larval prolegs on A2-A7, are shared with the
entire limacodid group. Other examples include (1) a disjunct
stemma 5, which occurs in all the families except Limacodidae,
although to a lesser degree in Dalceridae; and (2) spatulate
ventral setae on the tarsus and prolegs, found in first-instar
Psycharium (Figures 230, 250), megalopygids (Figure 260),
and zygaenids (Figure 247) (Stehr, 1987c). The presence of
deciduous scales on the end of the female abdomen is a
potential synapomorphy for megalopygids and Psycharium (H.
Geertsema, pers. comm.), although this trait also occurs in
some zygaenids (Common, 1990).
Jordan (1909) noted the variation in shape and number of
prongs (three or four) in the anal fork of presumably late instar
Somabrachys collected at the same locality in Algeria. An early
instar Psycharium sp. has five prongs, and late instar
Psycharium have three and five prongs (from two USNM
specimens of perhaps different species accidentally imported
from unknown locations in South Africa (USNM)).
Minet (1986) considered the divided crochet mesoseries to
be a symplesiomorphy between Somabrachyidae and the
limacodid group. This condition, however, found in both
Psycharium and Somabrachys, is not universally present in the
Megalopygidae (Figures 253, 256) or Aididae (Figures 340,
343, 344). Although one group of the divided series may have
been lost in some megalopygids and in dalcerids and
crochet-bearing limacodids, it is also possible that the ancestral
condition was undivided, as occurs in aidids or zygaenids.
Larval characters shared between the Somabrachyidae and
Aididae do not provide support for common ancestry (see
Aididae section, above). Prolegs without membranous pads,
and the absence of spiracular sensilla in both families, appear to
be plesiomorphic. The absence of crochets on A2 and A7
occurs frequently within megalopygids and dalcerids, as well.
One potential synapomorphy would occur if the "missing"
forewing radial vein in Somabrachyidae resulted from fusion
between R, and R2, because the two are stalked in Aididae.
Jordan (1916:355) suggested that R2 (=Sc2) was absent. The
loss, however, may have resulted from fusion of other stalked
radial veins, including the most common in the limacodid
group: R3 and R4 or R4 and R5.
Based on the information presented, the best evidence for the
membership of Somabrachyidae in the limacodid group is the
presence of larval prolegs on A2 and A7. Somabrachyidae as
the basal lineage of the limacodid group is suggested by
putative plesiomorphies that are not found in the remaining
families, including the large anal plate, anal comb, and
relatively large thoracic legs in the larvae and the epiphysis in
the adults (Somabrachys). Clearly, more information on
Somabrachys and Psycharium is needed before the position of
Somabrachyidae can be understood.
EPIPYROPIDAE Dyar, 1903,
and CYCLOTORNIDAE Meyrick, 1912
DISCUSSION.?The cosmopolitan family Epipyropidae (Fig-
ure 42) has traditionally been difficult to place. This may be
due to generalized wing venation, absence of mouthparts in the
adults, highly reduced male genitalia, and hypermetamorphic
larval stages with a penellipse of crochets. Adult characters
have led to placement in or near Tineoidea (Forbes 1923;
Turner, 1946; Brock, 1971); however, interpretations of larval
and adult habitus have suggested placement in Zygaenoidea.
The larval stages and biology of the Epipyropidae have been
described and reviewed by Davis (1987) and Common (1990).
The pupa was described by Heinrich (1931, pi. 2) and Common
(1990).
The Australian family Cyclotornidae has many similarities
with Epipyropidae, including hypermetamorphic and parasitic
larvae (Davis, 1987; Common, 1990). The two families also
share distinctive genitalic features. Females in both groups
possess large secondary accessory glands that appear to
perform a similar function. In dissecting a freshly killed female
epipyropid, Fulgoraecia exigua, I observed that the large
accessory gland was filled with a black fluid. Ovulated eggs in
the common oviduct were white, whereas the eggs that had
been laid were coated with the black secretion. A similar black
fluid was reported in the Cyclotornidae (Common, 1990). Male
genitalia in both families have reduced valvae and a paired
gnathos (Common, 1990).
One difference between the Epipyropidae and Cyclotornidae
is the crochet arrangement on the larval prolegs. In epipyropids
the crochets on A3-A6 form a penellipse, whereas in
cyclotornids they are in a mesoseries (Common, 1990). Indeed,
the penellipse may be considered a plesiomorphic condition
because it is found in lower ditrysian superfamilies; however,
unless epipyropids are at the base of the Zygaenoidea, I suggest
this to be a convergent condition. This is supported by the
NUMBER 582 87
i
L
365
P i - .
p^r
^ %
FIGURES 365, 366.?Pupal head of Epipyropidae (Fulgoraecia exigua) (scale length in parentheses): 365, frontal
view of labial palpi (200 urn); 366, frontoventral view, arrow points to lobe beneath eye (200 Jim). (LP = labial
palpus, Md = mandible, Mx = maxilla.)
presence of "retainer spines" on the crochets, which along with
the penellipse appear to be a specialization for attachment to
homopteran hosts (Davis, 1987).
Common (1990) reported another difference between epipy-
ropids and cyclotomids: the spiracles on Al of the pupa are
hidden in Heteropsyche (Epipyropidae) but are visible in
cyclotomids and in other Zygaenoidea. In the North American
epipyropid species, Fulgoraecia exigua, however, the first
abdominal spiracle is indeed visible. This character thus
appears to vary within the family, much the way a visible
second spiracle does in the Zygaenidae (Common, 1990).
The pupal eyepiece in epipyropids (e.g., F. exigua) appears
to lack the sculptured flange, which is synapomorphic for the
limacodid group. A sculptured lobe (Figure 366), however,
more mesal and below the eye, may be a homologous vestige.
Another small lobe, contiguous with the labial palpus but
lateral to it, may be a vestige of the maxilla (Figure 365). This
would contradict Common (1990), who reported absence of the
pupal maxilla in the Epipyropidae.
It is unlikely that Epipyropidae is the sister group to
Dalceridae as suggested by Jordan (1928). Most of the
characters shared between the two families are plesiomorphic;
furthermore, a number of larval synapomorphies for Dalceri-
dae + Limacodidae do not occur in Epipyropidae (Table 6).
Potential synapomorphies between the Epipyropidae and
Cyclotomidae, including their parasitic habits, eggs, and
female accessory glands, suggest that they form a mono-
phyletic group. The pupal spiracle and female glands of
epipyropids and cyclotomids are potential synapomorphies
with other Zygaenoidea. A problem for future work will be
determining the phylogenetic position of the two families
within the superfamily. The absence of prolegs and crochets on
A2 and A7 suggest placement outside of the limacodid group;
however, more precise relationships are difficult to establish
due to the specialized characters of parasitic larvae and to poor
knowledge of subfamily relationships in Zygaenidae.
Ontogeny and Function of Limacodid Spinnerets
In this section, I describe the ontogeny and morphology of
the spinneret in several limacodid species and discuss its
homology and function. The spinneret of the Semyra coarctata
complex, as in many limacodids, undergoes major ontogenetic
change (Figures 367-384). First and second instars exhibit the
common "fishtail" form (Figures 367-370). In the third instar,
spinneret shape changes, with the base becoming wider than the
tip. The upper lip, emarginate in the middle, is now shorter than
the lower lip (Figure 371), and the lower lip has a spongy
appearance (Figures 374, 375). The lateral portion, with
numerous fissures leading from the dorsal margin, forms a soft
flange (Figure 371-373). At the base of the spinneret are two
elliptical lobes (Figures 371,372). In the fourth instar, the most
obvious change in the spinneret concerns the apex; the upper
lip is subdivided into paired, smooth triangulate lobes with
brush-like tips (Figures 376-379). The dorsum of the lower lip
also is brush-like (Figure 379), and the lateral flange appears to
be more spongy (Figures 380, 381). In the ultimate instar, the
apex has a rag-mop appearance (Figures 382-384). Other
88 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 367-375.?Larval spinneret ontogeny of 1st through 3rd instar Semyra coarctata complex
(Limacodidae) (scale length in parentheses). 367-369, 1st instar: 367, dorsum (20 nm); 368, ventrum (20
\im); 369, detail of dorsum (10 urn). 370, 2nd instar, dorsum (30 ^m). 371-375, 3rd instar: 371, dorsum (100
Urn); 372, frontoventral view (75 \un); 373, lateroventral view (100 urn); 374, detail of upper and lower lips of
apex, dorsal view (20 |im); 375, detail of dorsal and ventral surfaces of apex, and silk pore (15 urn).
species exhibit a less dramatic ontogenetic change (e.g., Talima
postica) (Figures 385-388).
The plesiomorphic state for the limacodid spinneret is to be
apically wide. Often this condition is present only in early
instars, as described above. First instars of a few species have
parallel-sided spinnerets (e.g., Crothaema, Phobetron, Apoda)
(Figures 389-392), similar to the condition found in mega-
lopygids and many other lepidopteran larvae (Figures 142-
147). Because of this similarity, parallel-sided spinnerets may
seem to be the primitive condition for Limacodidae; however,
the presence of apically-wide spinnerets in closely related
genera of limacodids and in dalcerids suggests that the
parallel-sided spinneret is derived. This is further supported by
the presence of a rugose portion anterior to the opening, a
condition that occurs in the wide spinneret of first instar
Pantoctaenia (Figures 158, 159) but not in megalopygids or
NUMBER 582 89
ill
^
FIGURES 376-384.?Larval spinneret ontogeny of 4th and ultimate Semyra coarctata complex (Limacodidae)
(scale length in parentheses). 376-381, 4th instar: 376, dorsum (100 Jim); 377, detail of dorsum (50 urn); 378,
frontolateral view (75 fim); 379, detail of apex (20 ^m); 380, detail of dorsal margin of lateral flange (10 urn);
381, detail of lateral flange (6 \un). 382-384, ultimate instar, from exuvia: 382, dorsum (75 urn); 383, apex (27
Urn); 384, detail of apex (12 Jim).
90 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 385-388.?Ontogeny of larval spinneret of Talima postica (Limacodidae) (scale length in parentheses).
385, 386, early instar: 385, dorsum (20 fun); 386, ventrum (20 \un). 387, 388, ultimate instar, from exuvia: 387,
dorsum (50 (irn); 388, detail of anterolateral margin (20 \xm).
zygaenids. By the later instars, the shape of the once
parallel-sided spinneret in Phobetron becomes more triangu-
late, much as in Semyra.
The diversity of spinneret form and function has been
documented in several families of Lepidoptera. Ripley (1923)
noted the correlation between brush-like spinnerets and the
production of subterranean pupal cells in noctuids. MacKay
(1963, 1964) reviewed possible relationships between spin-
neret form and function in Tortricidae and other Lepidoptera.
Ontogenetic change in the spinnerets in limacodids was noted
by Chapman (1893) and in notodontids by Miller (1991).
MacKay (1964) proposed that variation in silk-pore shape
and the presence of apical fringes in noctuids (Ripley, 1923)
must produce variation in silk shape and viscosity and in
resulting silk function. Ripley (1923) interpreted "a very thin
ribbon" of silk in limacodids (Chapman, 1894:347) to indicate
that silk is issued in different physical states. The "narrowing"
of the spinneret apex in middle-instar limacodids suggests the
issuing of a more thread-like silk; however, this does not appear
to be the case. Some of the cocoon silk applied by the narrowed
spinneret in Semyra is ribbon-like (Figure 25). Furthermore,
the apex of dalcerid spinnerets remains wider than the base in
all instars, yet the larvae spin ribbon-like silk in first instars
(Figure 156) and cover the outer portion of their cocoon with
thread-like silk (Figures 18, 19).
The assumption that the size of the silk pore or the shape of
the spinneret's anterior margin solely determines the type of
silk ignores other potentially important aspects. In later instars
of Semyra, the brush-like lateral margins may change the form
of the silk during application to the substrate. Spinneret shape
may also relate to non-silk functions, such as brushing off
debris on the ventral surface of the thorax and abdomen, and the
NUMBER 582 91
FIGURES 389-392.?Parallel-sided larval spinnerets in 1st instar Limacodidae (dorsal view) (scale length in
parentheses): 389, Crothaema sp. (6 Jim). 390, 391, Phobetron hipparchia: 390, overall view, without silk (10
Urn); 391, detail of tip and silk pore, with thread-like silk (6 urn). 392, Apoda biguttata (7.5 Jim). (Si = silk.)
spreading of liquid Malpighian tubule secretions during cocoon
construction (see "Biology," above).
Homology and Evolution of Larval Body Setae
DALCERIDAE.?Setae on the dorsum of T2 and T3 and on
segments A1-A9 in dalcerid caterpillars are covered with a
gelatinous material, giving them a wart-like shape. Removal of
this material reveals setae similar to primary setae found in
most families of Lepidoptera (Figure 177). First instars in
nonspiny lineages of Limacodidae have similar setal configura-
tions (Figures 178, 179) but commonly have some degree of
fusion or secondary branching (Figures 393, 395, 397).
There has not been a definitive homology for the body setae
in dalcerids. The gelatin-covered setae have been incorrectly
interpreted as smooth homologues of the spiny verrucae found
in megalopygids, zygaenids, and many limacodids. Most
abdominal segments in megalopygids and zygaenids have two
verrucae, the D and SD, above the spiracles, and two below, the
L and SV. Hopp (1928:285) mistakenly homologized the two D
setae in dalcerids with D and SD verrucae, and the SD1 setae
with L verrucae. The true L abdominal row in dalcerids consists
of tactile setae proximate to the spiracles (Figures 180, 208,
209). Verrucae on T2 and T3 of megalopygids roughly match
dalcerid setae in number and position, differing in the presence
of only one SD group on T2 and T3 and one D group on T3.
Stehr and McFarland (1987) considered the gelatinous setae in
dalcerids to be secondary.
Dyar's (1925) homology concept for dalcerid setae is
somewhat unclear. His setal terminology for Acraga coa is
similar to that used in earlier work on the Limacodidae.
According to Dyar (1925:45), the two D rows on segments
A2-A6 are "subdorsal" and "lateral," corresponding to the two
rows of verrucae or scoli dorsal to the spiracles in spiny
limacodids. The third "suprastigmatal" row was probably
viewed as an extra row. Dyar regarded the presence of only two
rows, as found in megalopygids and zygaenids, to be the
ancestral condition. In keeping with earlier larval descriptions,
Dyar did not attempt to serially homologize dalcerid setae on
the thorax with those on the abdomen (see next section).
LIMACODIDAE.?Limacodids are typically characterized as
having either spined or smooth caterpillars. Early instars of
smooth taxa often have long tubercles or Y-shaped setae in the
D and SD positions (Figures 393, 397); however, Belippa
(Figure 171), Crothaema (Figure 175), and Pantoctaenia
(Figure 176) all have short cuticular warts. These warts
occur in similar positions and bear a superficial resemblance to
the gelatinous warts that cover dalcerid setae. Olona spp. are
the only known limacodid larvae with gelatinous, dalcerid-like
warts in late instars (Holloway, et al., 1987). This genus clearly
belongs in the Limacodidae based on male genital ia and on the
cocoon. Although I have not examined larval specimens of
Olona, it appears from figures in Holloway et al. (1987) and in
Piepers and Snellen (1900) that the larval warts are homolo-
gous with those found in dalcerids. I suspect that the L row of
Holloway et al. (1987) is in fact an SD row as in the dalcerids.
Holloway et al. (1987) noted that the warts do not correspond
to the scoli of limacodids.
Other first instar limacodids have setal arrangements similar
to dalcerids and the limacodid genera mentioned in the
previous paragraph, although there is often some degree of
fusion. Examples include the Apoda, Phobetron, and Prolima-
codes complexes (Table 3). Dyar (1896a: 179) considered the
"Y-shaped setae" (=bisetose tubercles) found in first instar
Apoda to be derived from the fusion of Dl and D2 (= Dyar's ia
and ib) and SD 1 and SD2 (= iia and iib). This hypothesis is
supported by the presence of unfused pairs of D and SD setae
on T2 and T3 (Figure 178) in the Apoda complex. Caterpillars
of some species in this group (e.g., Apoda biguttata) have one
arm of the Y shortened, producing nearly a simple tubercle
92 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
FIGURES 393-400.?Larval tubercles of 1st instar Semyra and Prolimacodes (Limacodidae) (scale length in
parentheses). 393-397, Semyra coarctata complex: 393, anterolateral aspect (200 nm); 394, SD button and
spiracle on A4 (10 u,m); 395, detail of SD tubercle on T2 (10 urn); 396, SD and L tubercles and buttons on T3
(50 urn); 397, detail of SD2 tubercle on T3 (20 urn). 398-400, Prolimacodes badia: 398, anterolateral aspect
(tubercles only partially everted) (86 urn); 399, SD and L buttons on T3 (compare SD2 with 396) (38 urn); 400,
detail of SD2 button (8.6 urn). (L, SD = lateral-, subdorsal-group tubercles or buttons.)
(Figure 195). This type of fusion and reduction was
considered by Dyar (1896a) to explain evolution of tubercle
shape in Phobetron.
Serial homology problems are inherent in Dyar's system for
body setae. Dyar (1895a) considered the L (= iii) and SV (= iv
and v) rows on the thorax to be homologous with the SD and L
rows, respectively, on the abdomen. Dyar (1901) acknowl-
edged this mistake in a reply to criticism from O. Hofmann
(1898), but he chose not to make the necessary changes in
deference to his earlier work. An example of this problem can
be found in Dyar, 1896a. He reported that one of two pairs of
D bisetose tubercles on the abdomen in first instar Prolima-
codes (Limacodidae) represented the SD group (= Dyar's
lateral row), moved dorsad of its normal position. This
explanation, however, does not consider serial homology with
thoracic setae. There is a pair of bisetose D tubercles in the
NUMBER 582 93
same position on T2 and T3 (see Figure 179, for closely related
Semyra). Scanning electron microscopy in Semyra reveals an
additional setal row, reduced to buttons near the spiracles
(Figure 394). These setae, perhaps not visible to Dyar, likely
represent the SD group. Reduction from tubercles to buttons
can be seen when comparing the SD2 tubercle on T3 in Semyra
(Figures 396, 397) and Prolimacodes (Figures 399, 400).
First instar spiny limacodids have at least three setae per
tubercle (Figures 401,403; see also Figures 172,174), whereas
later instars have a maximum of four scoli per segment, (two on
each side) (Dyar, 1899a). In the Natada complex (e.g.,
Euprosterna), first instars have the three setae on the
abdominal tubercles (Figure 403), but there are seven on the
thoracic tubercles (Figure 404) (see also Dyar, 1899d, pi. 1).
Some of these setae are likely to be secondary, but others may
have evolved by fusion of tubercles. For example, three setae
on a D tubercle may result from fusion of two bisetose D
tubercles as found in the Semyra coarctata complex (Figures
179, 393). In contrast to the variable number of setae in the
Natada complex, first instars of the spiny Parasa complex have
three setae on every tubercle. These also differ in being more
fused at the base and by having sharper points (Figures 174,
206,401).
EVOLUTION AND ORIGIN.?The sister-group relationship
between limacodids and dalcerids sheds light on the evolution
of scoli in the Limacodidae. The presence of primary setae in
positions homologous to those found in dalcerids is plesiomor-
phic in the Limacodidae (Figures 180, 181). Often only
occurring in the first instar, these setae are in the form of warts
or tubercles. I suspect that the common ancestor of the
Limacodidae and Dalceridae had thick, unfused primary setae
or tubercles, similar to the setae beneath the gelatinous
secretions in the Dalceridae (Figure 327). Limacodids with
primitive body setae in first instars include Pantoctaenia,
Crothaema, and Belippa. Body setae in the Semyra and Apoda
complexes also are primitive but have some degree of fusion
(Figures 178, 179). No taxa with scoli in late instars have
been found to have the primitive setal arrangement This
FIGURES 401-405.?First and 3rd instar scoli of spiny Limacodidae (scale length in parentheses). 401-402,
Talima postica: 401, trisetose scolus of thorax, 1st instar (10 \im); 402, SD scolus on mesothorax, 2nd or 3rd
instar (50 ]im). 403-404, Euprosterna sp., 1st instar: 403, D scolus on abdomen (50 urn); 404, thoracic scolus
(50 \im). 405, Natada subpectinata, middle instar, SD abdominal scoli (200 \un).
94 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
contradicts the hypothesis that the limacodid ancestor was
spiny, based on a phylogeny with a megalopygid- or
zygaenid-like ancestor (Dyar, 1899a; Holloway, 1986).
The evolution of cuticular warts from thick primary setae can
be hypothesized in the following manner. Warts in first instar
limacodids (e.g., in Crothaema and Pantoctaenia) may
represent tubercles that have remained enclosed in their bases
and have fused in the embryo, rather than telescoping outward
upon hatching. "Buttons" found on warts of other primitive
limacodids may represent the distal end of tubercles that
extended just beyond the base, with the remainder undergoing
fusion and reduction in the embryo (e.g., Belippa, Figures
188-190). This is plausible because buttons are quite similar to
the tips of normal tubercles (Figure 395), and they occur in
homologous positions to the tubercles of related taxa that evert
after eclosion (compare SD2 button in Prolimacodes and
bisetose SD2 tubercle in Semyra; Figures 396, 397, 399, 400).
Partial fusion may explain the condition of SD tubercles in first
instar Heuretes and Alarodia (Figures 199, 200), which are
short in relation to the D and SD tubercles found in related
Phobetron spp. of the same stage (Figure 203).
Everted tubercles and warts in the limacodid embryo may
relate to the evolution of a flat, scale-like egg with a thin
chorion. Spiny verrucae, found in first instar megalopygids, are
possibly correlated with a thicker chorion because urticating
setae could more easily damage a thin chorion.
L or SV verrucae in zygaenids, megalopygids, and som-
abrachy ids serve a defensive function. The presence of primary
setae and the reduction of verrucae in the L and S V positions in
the aidid + (limacodid + dalcerid) clade appear related to a
more slug-like locomotion and to the greater contact of the
lateral and ventral surfaces with the substrate that results. The
link between setal reduction and locomotion can be seen in the
following two examples. When comparing the ontogeny of L
and SV setae in Zygaena trifolii during the first three instars, L
setae increase with each instar, whereas the three SV setae on
the proleg planta are unchanged (Tremewan, 1985). Clearly,
SV setae have more direct contact with the substrate than L
setae during locomotion. Dalcerids and spiny limacodids show
a similar trend when one compares the type of L setae between
the thorax and abdomen (Figures 177-181). In contact with the
substrate, the L setae on the abdomen are hair-like or
fungiform, whereas those on T2 and T3, farther removed from
the substrate, are defensive, gelatin-covered setae or scoli.
It has been suggested that the two L setae in limacodids
evolved by reduction of L verrucae as found in megalopygids
(Figure 406) and zygaenids (Dyar and Morton, 1895). I
propose that L setae in aidids (Figure 407) represent an
intermediate condition between L setae found in megalopygids
and limacodids + dalcerids. in which the setae-bearing warts
found in megalopygids are formed but do not develop dense
secondary setae. In the limacodid + dalcerid clade these warts
arc lost (or retained, e.g. Crothaema), whereas the primary
setae remain (see further discussion in "Homology of Suckers,"
below).
Evolution of the Ventral Abdominal Surface
Dyar (1894:212) and Stehr and McFarland (1985) consid-
ered the presence of prolegs on A2 and A7 to be derived from
an ancestral condition of prolegs on A3-A6 (Hinton, 1955;
Common, 1975). Dyar's contemporaries had several other
interpretations. Packard (1894) suggested that prolegs on A2
and A7 represent a persistent, ancient condition within
Lepidoptera. Chapman (1893:115) considered the "apod
character" of the ventral surface in limacodids to be the result
of "very recent descent from a footless mining larva," although
he later wrote that limacodid suckers were "homologous with
prolegs, and also with the eight pairs of abdominal legs of
Eriocephala [= Micropterygidae]" (Chapman, 1894:345). Dyar
(1895a, 1899b) considered limacodid "suckerdisks" to be
derived from the medial fusion of pads on megalopygid
prolegs. He further believed that suckers on Al and A8 were
not derived from prolegs, but he did not suggest their origin
(Dyar, 1899b).
PROLEGS AND CROCHETS.?A new interpretation of how the
ventral abdominal surface evolved on proleg-bearing segments
in the limacodid group, based on my phylogenetic evidence, is
as follows: (1) the limacodid-group ancestor gained additional
crochet-bearing prolegs on A2 and A7; (2) in the aidid ancestor,
dense plumose SV setae on proleg bases, as found in
megalopygids, were lost; (3) the limacodid + dalcerid ancestor
had reduction of proleg size, suckers formed from lateral proleg
bases on A2-A7, and crochets were not phenotypically
expressed until after the first instar (on A2-A7 and A10); (4)
post-first instar crochets were lost on A10 and at least once on
A2 and A7 in Dalceridae; and (5) post-first instar crochets were
lost in all lineages of Limacodidae examined except Pantoc-
taenia (present on A2-A7 and A10).
According to this hypothesis, crochets were present on A2
and A7 in ancestors of both megalopygids and dalcerids and
were independently lost in each lineage. An equal size and
number of crochets on each crochet-bearing segment in
Pantoctaenia may represent the ancestral condition for the
Limacodidae + Dalceridae clade. The smaller and less numer-
ous crochets on A2 and A7 in Dalcerides ingenita (Stehr and
McFarland, 1985, 1987) may represent the stage prior to their
complete loss on these segments in other dalcerids. Intermedi-
ate stages of crochet loss between the condition found in
Pantoctaenia and the condition in other limacodids have not
been observed. Crochets are absent on A2 and A7 in all Aididae
and Somabrachyidae so far examined. A sister-group relation-
ship of either family to megalopygids or of aidids to
limacodids + dalcerids suggests that the crochets were present
in the common ancestor.
The discovery of crochets on A2 and A7 in the Dalceridae by
Stehr and McFarland (1985) provided new evidence of their
relationship with the Megalopygidae. Previous to this finding,
megalopygids had been the only family of Lepidoptera
reported to have prolegs on these two segments. They proposed
an evolutionary progression for proleg development in the
NUMBER 582 95
Zygaenoidea that went from "an ancestor with equal-sized
prolegs with crochets on A2-A7, to the dalcerids with smaller
prolegs and reduced numbers of crochets on A2 and A7, to the
megalopygids with smaller prolegs and no crochets on A2 and
A7, to the limacodids with all prolegs reduced to suckers and
crochets totally absent" (Stehr and McFarland, 1987:461).
My phylogenetic analysis (Figure 326), which includes both
characters of immature stages and of adults, clearly rules out
Dalceridae as ancestral to Limacodidae + Megalopygidae;
however, this also is true if one exclusively examines prolegs
and crochets. The discovery of megalopygid larvae with
prolegs of equal size and with equal numbers of crochets on
A2-A7 (Figures 216, 252, 253) eliminates the need to evoke a
hypothetical ancestor with this condition. Furthermore, the
presence of crochets on A2-A7 and on A10 in Pantoctaenia
(Figures 358-360) suggests this as the ground-plan state for
limacodids. A similar degree of sucker development in
primitive limacodids (Figures 219, 266) and in dalcerids
(Figures 261) supports the occurrence of suckers in the
ancestor of Limacodidae + Dalceridae (see further discussion
in "Homology of Suckers" below).
I have interpreted the presence of crochets on A2 and A7 as
a synapomorphy for the limacodid group, although they are
more frequently absent than present in megalopygids and
dalcerids. Crochets on A2 and A7 have not been found thus far
in aidids or somabrachyids. Perhaps the presence or absence of
crochets on these segments is controlled by a gene that switches
expression on or off. If this were the case, it raises the question
whether prolegs and crochets on A2 and A7 are suppressed in
other groups of Lepidoptera.
The presence or absence of crochets in dalcerids and
limacodids can be viewed in terms of developmental timing, or
heterochrony. The delay of crochet development until after the
first instar in dalcerids (Figures 262, 263) and in Pantoctaenia
(Figures 360, 361), and subsequent addition of crochets on A2
and A7 in Dalcerides ingenita, are examples of postdisplace-
ment, a late start in development (Funk and Brooks, 1990). In
the context of heterochrony, the absence of crochets in known
limacodids other than Pantoctaenia may have resulted from
progenesis, an early stop in their development (Funk and
Brooks, 1990). Studies on the genetics of crochet expression
should be pursued in future molecular or developmental
research.
HOMOLOGY OF SUCKERS.?The lateral portion of the proleg
base on A2-A7, which bears SV verrucae in the Mega-
lopygidae, appears to be homologous with the suckers in the
limacodid + dalcerid clade. This is best seen when comparing
proleg bases, along with L and SV setal rows, in representative
megalopygids, aidids, dalcerids, and limacodids (Figures
406-409). The proleg base is laterally convex in megalopygids
and aidids (Figures 406, 407). Aidids have an intermediate
condition in which the dense setae are lost Through further
FIGURES 406-409.?Evolutionary hypothesis of proleg reduction and the formation of suckers in the limacodid
group (arrows point to lateral margin of proleg base and to lateral margin of suckers) (406, 408 photos by C.
Hansen; 407, 409 photos by L. Minor-Penland): 406, Megalopyge lanata (Megalopyginae); 407, Aidos sp.
(Aididae); 408, Dalcerina tijucana (Dalceridae); 409, Acharia sp. (Limacodidae). (Sp = spiracle, Cr = crochets,
L = lateral-group setae.)
96 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
reduction, the proleg base forms the lateral portion of the
suckers in dalcerids and limacodids (Figures 408, 409).
Suckers may have been derived on nonproleg-bearing
segments in the following manner. In megalopygids, Al and
A8 each have two S V verrucae, one lateral to the other, whereas
A9 has only one, in line with the more mesal member of A8.
Aidids have presumably homologous warts, although these do
not bear long secondary setae. It is these verrucae or warts,
rather than prolegs, on Al and A8 that likely form the lateral
margins of the suckers. Suckers on these two segments, often
poorly developed or absent, are well developed in late stage
Pantoctaenia (Figures 219, 358).
The longitudinal ridge and bordering depression mesad
between the two L warts and the proleg bases in aidids (Figure
339) may be the precursor of the subspiracular flange in
Limacodidae described by Holloway (1986). The region of
flexible cuticle, below the flange and lateral to the suckers in
limacodids and dalcerids, may have been derived in part from
the L3 verrucae or warts in megalopygids and aidids. The
presence of the L3 wart in the Aididae, without plumose setae,
is a plausible intermediate stage (Figure 407).
My evolutionary hypothesis of the ventral surface in
limacodids differs from previous interpretations. According to
Dyar (1895a, 1899b), limacodid suckers are derived from an
extension of the membranous pads found on megalopygid
prolegs. This is not supported, because membranous pads
appear to be uniquely derived in megalopygids and absent in
aidids, the group considered herein to be ancestral to
limacodids and dalcerids. Moreover, the lateral portions of
prolegs more closely resemble the lateral margins of the
suckers. The pads would have had to become greatly expanded
to achieve this configuration.
Character states in dalcerids and Pantoctaenia suggest that
both suckers and crochets were present in the lima-
codid + dalcerid ancestor. Previous interpretations proposed
that crochets were lost prior to (Holloway, 1986:52) or after the
formation of suckers in the limacodid ancestor (Dyar, 1899b).
FUNCTION AND ADAPTATION.?I hypothesize that extra
prolegs and a slug-like ventral surface benefit larvae, whether at
rest or in motion, by increasing adhesion to the substrate. In
families with discernable prolegs (e.g., megalopygids and
aidids), additional prolegs and the shortened ventral segments
on A8-A9 provide a broader adhesive zone on flat leaf surfaces
than is normally found in external-feeding caterpillars (Figures
215-218, 223). Extra prolegs may also increase the clasping
ability of a larva lengthwise on a leaf edge or stem. In spite of
reductions in proleg size and crochets, limacodid and dalcerid
larvae are adept at medially clasping narrow surfaces due to the
retention of ventral muscles and the flexibility of the ventral
surface.
Further, I suggest that the ventral surface and brush-like
spinneret of the caterpillars in the limacodid + dalcerid clade
represent apomorphic function (sensu Coddington, 1988) and
are an adaptation for existence on smooth leaf surfaces. Dyar
(1899a:241) observed that temperate limacodids "principally
frequent smooth, glabrous leaves of trees and shrubs." The
flexible and sticky ventral surface in the clade provides
maximum contact with the substrate. The function of this type
of ventrum can be seen when comparing the crawling abilities
of limacodids and dalcerids to that of megalopygids. The
slug-like larvae of the first two families have little problem
crawling on glass at any slope, including upside down, whereas
megalopygids easily become dislodged and can only make an
attempt to stay on the glass by spinning copious amounts of
silk. Limacodid and dalcerid larvae also have the advantage of
having a ventral surface that is sticky and in continuous motion.
They do not have to plant each segment individually as do
larvae with prolegs. The presence of suckers in limacodid and
dalcerid larvae may also contribute to this ability.
Larvae in the limacodid group feed on either leaf surface. In
megalopygids and aidids, the ability to feed while clinging
underneath leaf surfaces is presumably enhanced by the large
numbers of crochets and the spinning of large quantities of silk
but not necessarily by their extra prolegs, which commonly
lack crochets. Dalcerid caterpillars (e.g., Dalcerides ingenita)
are difficult to detect while they are feeding on leaf
undersurfaces (Stehr and McFarland, 1985). Limacodid cater-
pillars may feed on either leaf surface (e.g., Phobetron
pithecium) (Dyar, 1896a: 180) or may feed soley on undersur-
faces (e.g., ha textula (Herrich-Schaffer) (Dyar, 1896a: 187).
The presence of membranous pads in megalopygids may
help the larvae adhere to smooth surfaces. In conjunction with
prolegs, the pads can stick to glass while the crochets are turned
under (pers. obs.; Dyar, 1899b). Dyar (1899b) considered this
to be an adaptation for allowing polyphagous larvae to feed or
crawl on smooth or rough surfaces. Retractor muscles are likely
attached to these pads as evidenced by the presence of dense
muscle tissue attached to the ventral surface of megalopygid
larvae. This may produce a degree of suction, such as described
for plantae by Hinton (1955). Further elaboration involves the
modification of SV verrucae into an additional row of
membranous pads in Mesocia pus ilia (Figure 252). The
absence of crochets on A2 and A7 in the majority of
megalopygid and aidid larvae, presumably independent losses,
may have similar function to the pads.
It appears that Psycharium larvae, which lack the membra-
nous pads (Figures 217, 249), are specialized for clinging to
narrow leaf surfaces found on their narrow-leaf hosts, such as
Restionaceae and Pinus (H. Geertsema, pers comm.). When
placed on a smooth glass surface, early instar Psycharium
larvae clasp prolegs together rather than place them on the
substrate (pers. obs.).
There have been other suggestions regarding the adaptive
significance of the slug-like ventral surface. Packard
(1890:148) stated that the "slow gliding motion" of smooth,
nonspiny limacodid species "renders them less liable to be
NUMBER 582 97
observed by passing birds." There may indeed be some cryptic
advantage in lacking prolegs, because typical caterpillars can
be detected by their prolegs when crawling on leaf edges. The
locomotion in smooth species, however, surely did not evolve
independently from the more visible spiny limacodids as
suggested by Packard's Neo-Lamarckian philosophy (Packard,
1895b: 83-84).
The presence of larval crochets in a mesoseries may be
considered a synapomorphy for the Zygaenoidea because
potential sister groups in the lower Ditrysia have crochets in an
ellipse. Hinton (1955:502-503) hypothesized that larvae with
crochets in mesoseries are adapted for climbing. He based this
on the ability of a larva with this arrangement of crochets to
medially clasp twigs and on the appearance of this characteris-
tic in "all exclusively arboreal caterpillars" that are external
feeders. He suggested that this development evolved independ-
ently in higher (e.g., papilionids) and lower ditrysian groups
(e.g., zygaenids). A test of whether the presence of mesoseries
in the Zygaenoidea represents an adaptation (in a cladistic
sense) for climbing has potentially been made because possible
out groups (e.g., cossoids, sesioids, tortricoids) are not tree
climbers, and it is improbable that a mesoseries is plesiomor-
phic in the lower Ditrysia.
Future comparative studies on the energetics and biomecha-
nics of locomotion in the limacodid group are warranted. One
approach would be to measure the energetic cost of locomotion
by measuring oxygen consumption (Denny, 1980; Casey,
1991). Perhaps the cost of additional prolegs in megalopygids
could be estimated by comparing megalopygid oxygen
consumption with that of external-feeding caterpillars having
typical numbers of prolegs (e.g., gypsy moth larvae; Casey,
1991). It also would be worthwhile to determine if the cost of
locomotion in limacodid and dalcerid caterpillars is similar to
that of an analogue group, such as terrestrial slugs (Denny,
1980).
Summary
The phylogenetic analyses described herein, which include
separate and combined data sets of both adult and immature
stages, support a monophyletic Megalopygidae consisting of
Megalopyginae + Trosiinae at the base of the limacodid group.
Formerly a subfamily of Megalopygidae, Aididae is found to
be the sister group of the Limacodidae + Dalceridae clade. It is
clear that aidids do not form a clade with either megalopygid
subfamily, and contrary to previous assertions, they do not
belong within the Limacodidae. Chrysopolomidae is placed as
a subfamily within Limacodidae, based on synapomorphies
with apparently higher lineages of Limacodidae, but subfa-
milial status is provisional until a more detailed study of
limacodid classification can be undertaken. Further work is
needed on the Somabrachyidae to determine whether Som-
abrachys and Psycharium are sister genera or belong in
Megalopygidae. They share a number of characters with
megalopygids, but most can be argued to be symplesiomor-
phies in relation to the Megalopygidae or the Zygaenoidea.
Synapomorphies for the limacodid group include the
presence of abdominal segments A2 and A7 with crochets,
ventral reduction of segments A8 and A9, reduction of the anal
plate, and short prolegs in the larva; short maxilla and
sculptured flange on the eyepiece of the pupa; and absent ocelli,
short galeae, and dense sensilla trichodea on the fifth tarsomere
of females of the adult. The Limacodidae + Dalceridae clade is
strongly supported by synapomorphies including fungiform SV
and V setae, absence of crochets in early instars, highly reduced
prolegs and thoracic legs, a brush-like spinneret, absence of S2
seta and anal fork, elongate axial seta, presence of frass-
flipping setae, and smooth ventral surface in the larva and
absence of chaetosemata, reduced sclerite on the mesepimeron,
and presence of a gnathos in the adult.
A second phylogenetic analysis, using adult and immature
stages, is also performed to diagnose the placement of African
limacodid genera Pantoctaenia and Crothaema. Each genus is
found to have a number of primitive character states in the
immature stages, some shared with Dalceridae (e.g., crochets),
but not found in other Limacodidae. The results are inconclu-
sive, showing the genera to be either a primitive group of
Limacodidae (combined and adult data) or the sister group to
Limacodidae + Dalceridae (data of immatures). Congruent
cladograms result with Pantoctaenia and Crothaema as
primitive Limacodidae, however, by recoding two potentially
weak characters of the A10 prolegs, one as convergent losses of
crochets rather than as a synapomorphy, and the other as a loss
of frass-flipping setae rather than as a plesiomorphy.
The presence of exposed abdominal spiracles in the pupa and
the female secondary accessory glands in epipyropids and
cyclotornids lends support for placing these groups in
Zygaenoidea. A more distant relationship, however, between
Epipyropidae and the limacodid group (especially Dalceridae)
than previously thought is suggested based on a large number
of synapomorphies between Dalceridae and Limacodidae and
the presence of either crochets or prolegs on A2 and A7 in all
limacodid-group families but not in epipyropids.
The phylogeny of the limacodid group suggests new
interpretations of the evolution of the limacodid ventral
surface. Prolegs in the Limacodidae are often reported as being
absent or modified into suckers. It is proposed that suckers on
A2-A7 are homologous with the lateral portion of the proleg
bases, which bear SV verrucae in Megalopygidae, and it is
proposed that suckers on Al and A8 are derived from verrucae
or warts, as found in megalopygids and aidids, and are not
homologous with prolegs. Delayed appearance of larval
crochets until the middle or late instars of dalcerid species and
primitive limacodid species (e.g., Pantoctaenia) suggests that
the absence of crochets in most limacodids evolved from an
98 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
early termination in their expression during the larval stage
rather than during the embryonic stages.
The homology of several confusing character states in the
limacodid group is clarified. Removal of the gelatinous warts
on dalcerid larvae reveals primary setae that are homologous to
those found in early instars of limacodids with smooth later
instars, rather than homologous with verrucae found in
zygaenids and megalopygids as has been previously implied.
SV setae, unreported or considered absent in Limacodidae and
Dalceridae, occur as modified fungiform setae on the slug-like
ventrum. The "ventral valva" of the male genitalia in
Megalopygidae is reinterpreted as a juxtal lobe, based on the
condition of the juxta in Aididae and Dalceridae.
Behavioral observations of limacodid and dalcerid larvae
show that semifluid silk is the liquid observed on the ventral
surface, and brush-like spinnerets in the two families are used
to remove debris from the ventrum. Spinnerets in the
Limacodidae may undergo major changes in shape between
instars, whereas spinneret shape in the Dalceridae remains
unchanged.
Literature Cited
Alberti, B.
1954. Ober die stammesgeschichtliche Gliederung der Zygaenidaenebst
Revision einiger Gruppen (Insecta, Lepidoptera). Mitteilungen
Zoologischen Museum in Berlin, 30:115-481.
Alexander, B.
1990. A Cladistic Analysis of Nomadine Bees (Hymenoptera: Apoidea).
Systematic Entomology, 15:121 -152.
Aurivillius, C.
1895. Diagnosen neuer Lepidopteren aus Afrika. Entomologisk Tidskrift,
16:113-120.
Becker, V.O.
1995. Megalopygidae. In J.B. Heppner, editor, The Atlas of Neotropical
Lepidoptera; Checklist: Part 2, Hyblaeoidea, Pyraloidea, Tortricoi-
dea, 3:118-122. Gainesville, Florida: Association for Tropical
Lepidoptera, Inc.
Becker, V.O., and M.E. Epstein
1995. Limacodidae. In J.B. Heppner, editor, The Atlas of Neotropical
Lepidoptera; Checklist: Part 2, Hyblaeoidea, Pyraloidea, Tortricoi-
dea. 3:128-133. Gainesville, Florida: Association for Tropical
Lepidoptera, Inc.
Biezanko, CM., A. Ruffinelli, and D. Link
1974. Plantas y otras sustancias alimenticias de las orugas de los
Lepidopteros Uruguayos. Revista Centro Ciencias Rurais, 4:
107-148.
Birket-Smith, S.J.R.
1974. Morphology of the Male Genetalia of Lepidoptera, I: Ditrysia.
Entomologica Scandinavica, 5:1-22.
Bode, W., and CM. Naumann
1988. Little-known Accessory Glands in Female Zygaena Moths (Lepi-
doptera, Zygaenidae). Zoological Journal of the Linnean Society,
92:27-42.
Borner, C.
1939. Die Grundlagen meines Lepidopterensystems. VII International
Kongress fur Entomologie, 2:1372-1424, 51 figures.
Brock, J.P.
1971. A Contribution Towards an Understanding of the Morphology and
Phylogeny of the Ditrysian Lepidoptera. Journal of Natural History,
5:29-102.
1990. Origins and Phylogeny of Butterflies. In O. Kudma, editor,
Butterflies of Europe, 2: Introduction to Lepidopterology, pages
209-233. Wiesbaden: AULA-Verlag.
Casey, T.M.
1991. Energetics of Caterpillar Locomotion: Biomechanical Constraints of
a Hydraulic Skeleton. Science, 252:112-113.
Chapman, T.A.
1893. On Some Neglected Points in the Structure of the Pupae of
Heterocerous Lepidoptera, and Their Probable Value in Classifica-
tion; with Some Associated Observations on Larval Prolegs.
Transactions of the Entomological Society of London, 1893:
97-119.
1894. Some Notes on the Micro-Lepidoptera Whose Larvae are External
Feeders, and Chiefly on the Early Stages of Eriocephala calthella
(Zygaenidae, Limacodidae, Eriocephalidae). Transactions of the
Entomological Society of London, 1894:335-350.
Christensen, P.J.H.
1950. Studien iiber die Postembryonale Entwicklung bei Cochlidion
limacodes Hufn. (Fam. Cochlididae, Lepidoptera). Det Kongelige
Danske Videnskabernes Selskab, Biologiske Skrifter, 7:1-34, 9
figures.
Clarke, J.F.G.
1941. The Preparation of Slides of the Genitalia of Lepidoptera. Bulletin of
the Brooklyn Entomological Society, 36:149-161.
Cock, M.J.W., H.CJ. Godfrey, and J.D. Holloway, editors
1987. Slug and Nettle Caterpillars: The Biology, Taxonomy and Control of
the Limacodidae of Economic Importance on Palms in South-East
Asia. 270 pages, 36 plates. Wallingford: C.A.B. International.
Coddington, J.A.
1988. Cladistic Tests of Adaptational Hypotheses. Cladistics. 4:3-22.
Common, I.F.B.
1970. Lepidoptera (Moths and Butterflies). In The Insects of Australia: A
Textbook for Students and Research Workers, pages 765-866.
Melbourne: Melbourne University Press (Commonwealth Scientific
and Industrial Research Organization).
1975. Evolution and Classification of Lepidoptera. Annual Review of
Entomology, 20:183-203.
1990. Moths of Australia. 535 pages, 32 color plates. Melbourne:
Melbourne University Press.
Comstock, J.H.
1893. Evolution and Taxonomy. In The Wilder Quarter-Century Book,
pages 37-113, 33 figures, 2 plates. Ithaca, New York: Comstock
Publishing Company.
1918. The Wings of Insects. 430 pages, 10 plates. Ithaca, New York:
Comstock Publishing Company.
Davis, D.R.
1987. Epipyropidae. In F.W. Stehr, editor. Immature Insects, 1:458-460.
Dubuque, Iowa: Kendall/Hunt Publishing Company.
1989. Generic Revision of the Opostegidae, with a Synoptic Catalog of the
World's Species (Lepidoptera: Nepticuloidea). Smithsonian Contri-
butions to Zoology, 478: 97 pages, 320 figures.
Davis, D.R., R.C Kassulke, K.L.S. Harley, and J.D. Gilbert
1991. Systematics, Morphology, Biology, and Host Specificity of Neuros-
trota gunniella (Busck) (Lepidoptera: Gracillariidae), an Agent for
the Biological Control of Mimosa pigra L. Proceedings of the
Entomological Society of Washington. 93:16-44.
Davis, D.R., and J.E. Pefia
1990. Biology and Morphology of the Banana Moth, Opogona sacchari
(Bojer), and Its Introduction into Florida (Lepioptera: Tineidae).
Proceedings of the Entomological Society of Washington, 92:
593-618.
Delgado Quiroz, A.
1978. Venoms of Lepidoptera. In Sergio Bettini, editor. Arthropod
Venoms, pages 555-611. Berlin: Springer-Veriag.
Denny, M.
1980. Locomotion: The Cost of Gastropod Crawling. Science, 208:1288-
1290.
Dewitz, H.
1878. Entwickelung einiger Venezuelanischer Schmetterlinge nach Beo-
bachtungen von Gollmer. Archiv fur Naturgeschichte, 44:1-36,
plate 1.
Downey, J.
1987. Lycaenidae (Papilionoidea). In F.W. Stehr, editor. Immature Insects,
];443-445. Dubuque, Iowa: Kendall/Hunt Publishing Company.
Dyar, H.G.
1894. A Classification of Lepidopterous Larvae. Annals of the New York
Academy of Sciences, 8:194-232.
1895a. ("1894"). Additional Notes on the Classification of Lepidopterous
Larvae. Transactions of the New York Academy of Sciences, 14:
49-62. [Date on title page is 1894; actually published in 1895.]
99
100 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
1895b. A Combination of Two Classifications of Lepidoptera. Journal of
the New York Entomological Society. 3:17-26.
1893c. Notes on Some Southern Lepidoptera. The Canadian Entomologist,
27:242-247.
1896a. The Life-Histories of the New York Slug Caterpillars, iii-vi.
Journal of the New York Entomological Society. 4:167-190. plates
5-9.
1896b. On the Larvae of the Higher Bombyces (Agrotides Grote).
Proceedings of the Boston Society of Natural History. 27:127-147.
1898. Notes on Certain South American Cochlidiidae and Allied Families.
Journal of the New York Entomological Society, 6:231-239.
1899a. The Life-Histories of the New York Slug Caterpillars (Conclusion).
Journal of the New York Entomological Society, 7:234-253. plates
6-8.
1899b. Note on the Secondary Abdominal Legs in the Megalopygidae.
Journal of the New York Entomological Society. 7:69-70. figures
1-3. plate 2.
1899c. Life-History of a European Slug Caterpillar. Cochlidion avellana.
Journal of the New York Entomological Society. 7:202-208. plate 5.
1899d. The Life-Histories of the New York Slug Caterpillars. XVIII.
Journal of the New York Entomological Society, 7:61 -67, plate 1.
1901. A Century of Larval Descriptions. The Entomologist's Record,
13:37-41.
1903. A List of North American Lepidoptera and Key to the Literature of
This Order of Insects. Bulletin of the United States National
Museum, 52: 723 pages.
1914. The Life Histories of the New York Slug-Caterpillars. 20. Journal of
the New York Entomological Society. 22:223-229.
1925. A Note on the Larvae of the Dalceridae (Lepidoptera). Inset utor
Inscitiae Menstruus, 13:44-47.
Dyar. H.G., and E.L. Morton
1895. The Life-Histories of the New York Slug Caterpillars. 1. Journal of
the New York Entomological Society, 3:145-157, plate 6.
El-Mallakh, R.S.. D.L. Baumgartner, and N. Fares
1986. "Sting" of the Puss Caterpillar Megalopyge opercularis (Lepidop-
tera: Megalopygidae): First Report of Cases from Florida and
Review of Literature. Journal of the Florida Medical Association,
73:521-525.
Epstein, M.E.
1988. An Overview of Slug Caterpillar Moths (Lepidoptera: Limacodidae)
with Emphasis on Genera in the New World Parasa Group, xi + 149
pages. Doctoral thesis. Department of Entomology, University of
Minnesota, St. Paul, Minnesota.
1995. False-parasitized Cocoons and the Biology of Aididae (Lepidoptera:
Zygaenoidea). Proceedings of the Entomological Society of Wash-
ington, 97:750-756.
Epstein, M.E., and V.O. Becker
1994 ("1993")- Combinations and Synonymies in Neotropical Limacodi-
dae, Megalopygidae, Lasiocampidae and Arctiidae (Lepidoptera).
Revista Brasileira de Zoologia, 10:289-319. [Date on title page is
1993: actually published in 1994.]
Epstein, M.E., and P.M. Henson
1992. Digging for Dyar: The Man Behind the Myth. American Entomolo-
gist. 38:148-169.
Epstein, M.E., and S.E. Miller
1990. Systematics of the West Indian Moth Genus Heuretes Grote and
Robinson (Lepidoptera: Limacodidae). Proceedings of the Entomo-
logical Society of Washington. 94:705-715.
Epstein, M.E., S. Smedley, and T. Eisner
1994. Sticky Integumental Coating of a Dalcerid Caterpillar A Deterrent
to Ants. Journal of the Lepidopterists' Society. 48:381-386.
Eyer. J.R.
1924. The Comparative Morphology of the Male Genitalia of the Primitive
Lepidoptera. Annals of the Entomological Society of America.
17:275-328.
Farm. J.S.
1970. Methods for Computing Wagner Trees. Systematic Zoology.
19:83-92.
1988. Hennig86 Ver. 1.5. Port Jefferson Station. New York: James S.
Farris.
1989. The Retention Index and the Rcscalcd Consistency Index. Cladistics.
5:417-419.
Feeny. P.P.
1970. Seasonal Changes in Oak Leaf Tannins and Nutrients as a Cause of
Spring Feeding by Winter Moth Caterpillars. Ecology. 51:565-581.
Fehrenbach. H.
1995. Egg Shells of Lepidoplera?Fine Structure and Phylogenetic
Implications. Zoologist her Anzeiger, 234:19-41.
Fletcher. D.S.. and I.W.B. Nye
1982. In I.W.B. Nye. editor. The Generic Names of Moths of the World
Volume 4. xiv + 192 pages. London: British Museum (Natural
History).
Forbes. W.T.M.
1910. A Structural Study of Some Caterpillars. Annals of the Entomologi-
cal Society of America. 3:94-132.
1923. The Lepidoptera of New York and Neighboring Stales (part I ] .
Cornell University Agricultural Experiment Station Memoir. 371:
1-188.
1942. The Lepidoptera of Barro Colorado Island. Panama. No. 2. Bulletin
of the Museum of Comparative Zoology. 90:265-405.
Fracker. S.B.
1915. The Classification of Lepidoplerout Larvae. Illinois Biological
Monographs. 2:1-169.
Freina. JJ.de. and TJ. Witt
1990. Die Bombyces und Sphinges des Westpalatarktis: (Insecta. Lepidop-
tera). Volume 2. 134 pages. MUnchen: Forschung and Wisiemchaft
Funk. V.A.. and D.R. Brooks
1990. Phylogenetic Systematic* as the Basis of Comparative Biology.
Smithsonian Contributions to Botany. 73: 45 pages.
Genty. P.. R. Desmier. J.P. Morin. and C.A. Korytkowski
1978. Les ravaguers du palmier a huile en Amerique Latine. Oleagineux.
33:325-419.
Hampson, G.F.
1920. On New Genera and Species of Lepidoplera Phalaenae. with the
Characters of Two New Families. Novitates Zoologicae. 24:
253-282.
Harvey, D.J.
1987. Riodinidae (Papilionoidea). In F.W. Stehr, editor. Immature Insects.
1:443-445. Dubuque. Iowa: Kendall/Hunt Publishing Company.
Heinrich, C.
1931. Notes on and Descriptions of Some American Moths. Proceedings of
the United States National Museum. 79:1 -16.7 plates.
Heppner. J.B.
1977. The Status of the Glyphipterigidae and a Reassessment of
Relationships in Yponomeutid Families and Ditrysian Superfamil-
ies. Journal of the Lepidopterists' Society, 31:124-134.
1984. Outline of Classification of the Lepidoplera. In J.B. Heppner, editor.
Atlas of Neotropical Lepidoptera; Checklist: Part I. Micropterigoi-
dea-lmmoidea. pages xix-xxiii. The Hague: Dr. W. Junk Publish-
ers.
Hering. [E.] M.
1937. Revision der Chrysopolomidae (Lep.). Annals of the Transvaal
Museum. 17:233-257.
1955. Synopsis der afrikanischen Gattungen der Cochlidiidae (Lepidop-
tera). Transactions of the Royal Entomological Society of London.
107:209-225.
1961 ("1960"). Die erste Chrysopolomide von Madagascar [Lep.]. Bulletin
de la Societe Entomologique de France. 65:302-306. [Date on title
page is I960; actually published in I%1.]
NUMBER 582 101
HernchSchaffer.G.A.W.
1855. Systematische Bearbeitung der Schmetterlinge von Europa. Volume
6, part 68. pages 81-152. Regensburg.
Himon. HE.
1955. On the Structure. Function, and Distribution of the Prolegs of the
Panorpoidea. with a Criticism of the Berlese Imms Theory.
Transactions of the Royal Entomological Society of London,
106:455-540.
Hoffmann. F.
1932. Beitrage zur Naturgeschichte Brasilianischer Schmetterlinge.
Deutsche Eniomotogische Zeilschrifi. 1932:97-148.
Hofmann. O.
1898. I her die Anordnung der borstentragenden Warzen bei der Raupen
der Pterophoriden. Mustnerte Zeitschrift fur Emomologie. 3:
129-131.
Holloway. J.D.
1986. The Moths of Borneo: Key to Families; Families Cossidae.
Metarbelidae. Ratardidae. Dudgeoneidae. Epipyropidae and Lima
codidae. The Malayan Nature Journal. 40:1 -166.
Holloway. J.D.. M.J.W. Cock, and R. Desmier de Chenon
1987. Systematic Account of South-East Asian Pest Limacodidae. In
M J.W. Cock. H.C.J CiiHlfray. and JO. Holloway. editors. Slug and
Nettle Caterpillars The Biology, Taxonomy and Control of the
I inuhikluliif of Economic Importance on Palms in South-East Asia,
pages 15-117. Wallingford: C.A.B. International.
Holloway. J.D.. J.D. Bradley, and DJ. Carter
1987. Lepidoptcra. In C.R. Betts. editor. CIE Guides to Insects of
Important e to Man, 1: 262 pages. Wallingford: C.A.B. International.
Hopp. W.
1927. Die Megalopygiden-unterfamilie der Trosiinae (Lep. Megalopyg.).
Mitteilungen aus dem Zoologisches Museum in Berlin. 13:209-336.
186 figures.
1928. Beitrag zur Kenntis der Dalceriden. Deutsche Entomologische
Zeitschrift Iris. 42:283-287.
1930. Lepidoptera: Megalopygidae in Ergebnisse einer zoologischen
Sammelreise nach Brasilien. insbesondere in das Amazonasgebeit,
ausgefUhrl von Dr. H. Zerny. Annalen des Naturhtstorischen
Museums in Wien. 44:269-277.
1934-1935. Megalopygidae. In A. Seitz. editor. The Macrolepidoptera of
the World. 6:1071-1101. Stuttgart: Alfred Keman. [Plates published
in 1935.]
Ishii. S.. T. Inokuchi. J. Kanazawa, and C. Tomizawa
1984. Studies on the Cocoon of the Oriental Moth. Monema flavescens
(Lepidoptera: Limacodidae). 3: Structure and Composition of the
Cocoon in Relation to Hardness. Japan Journal of Applied
Entomology and Zoology. 28:269-273.
Itokawa, H.. R. Kano. T. Nakajima, and T. Yasuhara
1985. Examination of the Pain Producing Amines in the Venomous Hairs
or Spines of Japanese Urticating Caterpillars. Japan Journal of
Sanitary Zoology, 36:83-86.
Janse. A.J.T
1925. Key to the Families. Subfamilies and Tribes of the Order
Lepidoptera. with Special Reference to the South African Species.
South African Journal of Science. 22:318-345. plates IX-XI.
1964. Limacodidae. In The Moths of South Africa, 7: 136 pages.
Johannesburg: Cape & Transvaal Printers. Ltd.
Jordan. K.
1909. Note on the Larva of Somahrachys, a Genus of Moths. Novitates
Zoologicae. 16:331-332.
1916. On the Species of Somabrachys in the Tring Museum. Novitates
Zoologicae. 23:350-358, plates 7, 8.
1928. On Some Lepidoptera of Special Interest, with Remarks on
Morphology and Nomenclature. Novitates Zoologicae. 34:132-146.
Khalaf. K.T.
1984. The Identity of Wing Hairs in Megalopygidae. Journal of the
Lepidopterists' Society. 38:64.
Klug. G.R.
1836. Auszug aus einem Briefe des Hrn. A. Moritz. Caracas vom 27. Febr.
1836. Archiv fur Naturgeschichte, 2:302-306.
Kluge. A.G.. and J.S. Farris
1969. Quantitative Phyletics and the Evolution of Anurans. Systematic
Zoology, 18:1-32.
Kuznetsov, V.I.. and A.A. StekoTnikov
1981. [Functional Morphology of the Male Genitalia and the Phylogenetic
Relationships of Some Primitive Superfamilies of the Infraorder
Papilionomorpha (Lepidoptera: Sesioidea, Cossoidea. Zygaenoidea)
of the Asiatic Part of the U S S R ] [In Russian; translated for the
Smithsonian Institution, edited and distributed by the Bishop
Museum in 1988.] Trudy Zoologicheskogo Instituta Akademiya
Nauk SSSR. 92:38-73.
Kyrki.J.
1983. Adult Abdominal Sternum II in Ditrysian Tineoid Superfamilies?
Morphology and Phylogenetic Significance (Lepidoptera). Annals
Entomologica Fennica, 49:89-94,22 figures.
Leidy. N.A.. and H.H. Neunzig
1989. Taxonomic Study of the Larvae of Six Eastern North American
Dioryctria (Lepidoptera: Pyralidae: Phycitinae). Proceedings of the
Entomological Society of Washington. 91:325-341.
Lima, A.M. da Costa
1945. Insetos do Brasil. Volume 5. 379 pages. Rio de Janeiro: Imprensa
Nac tonal.
MacKay. M.R.
1963. Evolution and Adaptation of Larval Characters in the Tortricidae.
The Canadian Entomologist. 94:1321 -1344.
1964. The Relationship of Form and Function of Minute Characters of
Lepidopterous Larvae, and Its Importance in Life-History Studies.
The Canadian Entomologist, 96:991-1004.
Maddison. W.P., W.J. Donoghue, and D.R. Maddison
1984. Outgroup Analysis and Parsimony. Systematic Zoology, 33:83-103.
McNaulty. B.J.
1967. Outline Life Histories of Some West African Lepidoptera, Part II:
Limacodidae. Proceedings of the South London Entomological and
Natural History Society. 1967:1-12.
Meyrick. E.
1912. Supplement In F.P. Dodd. Some Remarkable Ant-Friend Lepidop-
tera. Transactions of the Entomological Society of London,
1911:589-590. plate 48.
Mickevich, M.F.. and S.J. Weller
1990. Evolutionary Character Analysis: Tracing Character Change on a
Cladogram. Cladistics, 6:137-170.
Miller, J.S.
1991. Cladistics and Classification of the Notodontidae (Lepidoptera:
Nociuoidea) Based on Larval and Adult Morphology. Bulletin of the
American Museum of Natural History, 204:1-230, 541 figures.
Miller. S.E.
1993. Unique Secondary "Accessory Glands" in the Female Genitalia of
Dalceridae (Lepidoptera). Annals of the Entomological Society of
America, 86:179-181.
1994. Systematics of the Neotropical Moth Family Dalceridae (Lepidop-
tera). Bulletin of the Museum of Comparative Zoology. 153:
301-495, 268 figures.
Miller. S.E., V.O. Becker, and R. Velez-Angel
1995. Podalia bolivari: A Highly Sexually Dimorphic Neotropical
Megalopygid Pest (Lepidoptera). Proceedings of the Entomological
Society of Washington. 97:117-122.
Minet. J.
1986. Ebauche dune classification moderne de l'ordre des Llpidopteres.
Alexanor. 14(7):291-313.
Mosher, E.
1916. A Classification of the Lepidoptera Based on Characters of the Pupa.
Bulletin of the Illinois State Laboratory of Natural History.
12:17-159.
102 SMITHSONIAN CONTRIBUTIONS TO ZOOLOGY
Naumann, CM.
1977. Stammesgeschichte und tiergeographische Beziehungen der
Zygaenini (Insecta, Lepidoptera, Zygaenidae). Mitteilungen Miinch-
ner Entomologischen Gesellscaft, 67:1-25.
1988. The Internal Female Genitalia of Some Zygaenidae (Insecta,
Lepidoptera): Their Morphology and Remarks on Their Phylogen-
etic Significance. Systematic Entomology, 13:85-99.
Oiticica Filho, J.
1946. Sobre a morfologia do penis em Lepidoptera. Boletim do Museu
National Rio Janeiro, Zoologia, 50:1-79.
Orfila, R.N.
1961. Las Dalceridae (Lep. Zygaenoidea) Argentinas. Revista de Investi-
gations Agricolas, Instituto de Patologia Vegetal, Institute Nacio-
nal de lecnologia Agropecuaria, Argentina, 15:249-264.
Packard, A.S.
1890. Insects Injurious to Forest and Shade Trees. Fifth Report of the
United States Entomological Commission. 957 pages, 40 plates.
Washington, D.C.: United States Department of Agriculture.
1893. The Life History of Certain Moths of the Family Cochliopodidae,
with Some Notes on Their Spines and Tubercles. Proceedings of the
American Philosophical Society, 31:83-108, 3 plates.
1894. A Study of the Transformations and Anatomy of Lagoa crispata, a
Bombycine Moth. Proceedings of the American Philosophical
Society, 32:275-292, plates 1-7.
1895a. On a New Classification of Lepidoptera. American Naturalist,
29:636-647; 788-803,16 figures.
1895b. Monograph of the Bombycine Moths of America North of Mexico,
Part 1, Family 1: Notodontidae. Memoirs of the National Academy of
Sciences, 7:1-291,49 plates.
Patton, W.H.
1891. Scent-Glands in the Larva of Limacodes. The Canadian Entomolo-
gist, 23:42-43.
Petersen, W.
1900. Beitrage zur Morphologie der Lepidopteren. Zapiski Imperatorskoi
Akademii Nauk VHIe series, 9:1-144,4 plates.
Peterson, A.
1967. Some Eggs of Moths from Several Families of Microlepidoptera.
The Florida Entomologist, 50:127-132.
Philpott, A.
1926. The Maxillae in the Lepidoptera. Transactions of the New Zealand
Institute, 57:721-746.
Piepers, M.C.. and P.C.T. Snellen
1900. Enumeration des Lepidopteres Heteroceres receuillis a Java.
Tijdschrift voor Entomologie, 43:12-106.
Pinhey, E.C.G.
1975. Moths of Southern Africa. 273 pages, 63 plates. Cape Town:
Tafelberg.
Ripley, L.B.
1923. The External Morphology and Postembryology of Noctuid Larvae.
Illinois Biological Monographs, 8:1 -85, 8 plates.
Schremmer, F.
1990. Das Schliipfen der Puppe des Asselspinners (Apoda limacodes
Hufn.) (Lepidoptera, Limacodidae). Zeitschrift der Arbeitsgemein-
shaft Osterreichischer Entomologen, 42:44-52.
Scoble, M.J.
1992. The Lepidoptera: Form, Function and Diversity. 404 pages, 321
figures. Oxford: Oxford University Press.
Shepard, H.H.
1930. The Pleural and Sternal Sclerites of the Lepidopterous Thorax.
Annals of the Entomological Society of America, 23:237-260.
Silva, A.G. D'Araujo e, C.R. Goncalves, D.M. Galvao, A.J.L. Goncalves, J.
Gomes, M. do Nascimento Silva, and L. de Simoni.
1968. Insetos, hospedeiros e inimigos naturais. In Quarto catdlogo dos
insetos que vivem nas plantas do Brasil, seus parasitos e
predadores, 2(1): 622 pages. Rio de Janeiro, Brazil: Laboratorio
Central de Patologia Vegetal Minist6rio da Agricultura.
Singh, G.P.. and S.C. Goel
1987. Body Chaetotaxy of Three Noctuid Caterpillars (Lepidoptera).
Deutsche Entomologische Zeitschrift, 34:373-381.
Stehr, F.W.
1987a. Megalopygidae. In F.W. Stehr, editor, Immature Insects, 1:454-456.
Dubuque, Iowa: Kendall/Hunt Publishing Company.
1987b. Lepidoptera. In F.W. Stehr, editor, Immature Insects, 1:288-305.
Dubuque, Iowa: Kendall/Hunt Publishing Company.
1987c. Zygaenidae. In F.W. Stehr, editor, Immature Insects, 1:453-454.
Dubuque, Iowa: Kendall/Hunt Publishing Company.
1987d. Limacodidae. In F.W. Stehr, editor, Immature Insects, 1:456-458.
Dubuque, Iowa: Kendall/Hunt Publishing Company.
Stehr, F.W, and N. McFarland
1985. Crochets on Abdominal Segments 2 and 7 of Dalcerid Caterpillars:
"Missing Link" or Anomaly? Bulletin of the Entomological Society
of America, 31:35-36.
1987. Dalceridae. In F.W. Stehr, editor. Immature Insects. 1:460-461,
Dubuque, Iowa: Kendall/Hunt Publishing Company.
Tarmann, G.
1984. Generische Revision der Amerikanischen Zygaenidae. Ento-
mofauna, supplement 2, volume 1 + 2: 176 pages, 438 figures. Linz.
Tremewan, W.G.
1985. Zygaenidae. In J. Heath and A.M. Emmet, editors. The Moths and
Butterflies of Great Britain and Ireland, 2: Cossidae-Heliodinidae,
pages 74-123. Colchester: Harley Books.
Turner, A.J.
1946. A Review of the Phylogeny and Classification of the Lepidoptera.
Proceedings of the Linnaen Society of New South Wales, 71:
303-338.
Tutt, J.W.
1899. A Natural History of the British Lepidoptera. Volume 1, 560 pages.
London: Sonnenschein and Co.
Viette, P.
1980. Insectes Lepidopteres Limacodidae. Faune de Madagascar, 53:
1-162.
Zerny, H., and M. Beier
1936. Lepidoptera = Schmetterlinge. In W. Kiikenthal and T. Krumbach,
editors, Handbuch der Zoologie: Insecta 2, pages 1554-1728.
Berlin: Walter de Gruyter and Co.
Zoological Society of London.
1992-1993. Insecta, Lepidoptera, Section 13D. Zoological Record,
129:249.
<r U.S. GOVERNMENT PRINTING OFFICE: 1996 405-261/20012


REQUIREMENTS FOR SMITHSONIAN SERIES PUBLICATION
Manuscripts intended for series publication receive substantive
review (conducted by their originating Smithsonian museums or
offices) and are submitted to the Smithsonian Institution Press with
Form SI-36, which must show the approval of the appropriate
authority designated by the sponsoring organizational unit. Requests
for special treatment?use of color, foldouts, case-bound covers,
etc.?require, on the same form, the added approval of the
sponsoring authority.
Review of manuscripts and art by the Press for requirements of
series format and style, completeness and clarity of copy, and
arrangement of all material, as outlined below, will govern, within the
judgment of the Press, acceptance or rejection of manuscripts and
art.
Copy must be prepared on typewriter or word processor,
double-spaced, on one side of standard white bond paper (not
erasable), with 11/4" margins, submitted as ribbon copy (not carbon
or xerox), in loose sheets (not stapled or bound), and accompanied
by original art. Minimum acceptable length is 30 pages.
Front matter (preceding the text) should include: title page with
only title and author and no other information; abstract page with
author, title, series, etc., following the established format; table of
contents with indents reflecting the hierarchy of heads in the paper;
also, foreword and/or preface, if appropriate.
First page of text should carry the title and author at the top of the
page; second page should have only the author's name and
professional mailing address, to be used as an unnumbered footnote
on the first page of printed text.
Center heads of whatever level should be typed with initial caps of
major words, with extra space above and below the head, but no
other preparation (such as all caps or underline, except for the
underline necessary for generic and specific epithets). Run-in
paragraph heads should use period/dashes or colons as necessary.
Tabulations within text (lists of data, often in parallel columns) can
be typed on the text page where they occur, but they should not
contain rules or numbered table captions.
Formal tables (numbered, with captions, boxheads, stubs, rules)
should be submitted as carefully typed, double-spaced copy
separate from the text; they will be typeset unless otherwise
requested. If camera-copy use is anticipated, do not draw rules on
manuscript copy.
Taxonomic keys in natural history papers should use the
aligned-couplet form for zoology and may use the multi-level indent
form for botany. If cross referencing is required between key and text,
do not include page references within the key, but number the
keyed-out taxa, using the same numbers with their corresponding
heads in the text.
Synonymy in zoology must use the short form (taxon, author,
yearpage), with full reference at the end of the paper under
"Literature Cited." For botany, the long form (taxon, author,
abbreviated journal or book title, volume, page, year, with no
reference in "Literature Cited") is optional.
Text-reference system (author, yearpage used within the text,
with full citation in "Literature Cited" at the end of the text) must be
used in place of bibliographic footnotes in all Contributions Series
and is strongly recommended in the Studies Series: "(Jones,
1910:122)" or "...Jones (1910:122)." If bibliographic footnotes are
required, use the short form (author, brief title, page) with the full
citation in the bibliography.
Footnotes, when few in number, whether annotative or biblio-
graphic, should be typed on separate sheets and inserted immedi-
ately after the text pages on which the references occur. Extensive
notes must be gathered together and placed at the end of the text in
a notes section.
Bibliography, depending upon use, is termed "Literature Cited,"
"References," or "Bibliography." Spell out titles of books, articles,
journals, and monographic series. For book and article titles use
sentence-style capitalization according to the rules of the language
employed (exception: capitalize all major words in English). For
journal and series titles, capitalize the initial word and all subsequent
words except articles, conjunctions, and prepositions. Transliterate
languages that use a non-Roman alphabet according to the Library
of Congress system. Underline (for italics) titles of journals and
series and titles of books that are not part of a series. Use the
parentheses/colon system for volume (number):pagination:
"10(2):5-9." For alignment and arrangement of elements, follow the
format of recent publications in the series for which the manuscript is
intended. Guidelines for preparing bibliography may be secured from
Series Section, SI Press.
Legends for illustrations must be submitted at the end of the
manuscript, with as many legends typed, double-spaced, to a page
as convenient.
Illustrations must be submitted as original art (not copies)
accompanying, but separate from, the manuscript. Guidelines for
preparing art may be secured from the Series Section, SI Press. All
types of illustrations (photographs, line drawings, maps, etc.) may be
intermixed throughout the printed text. They should be termed
Figures and should be numbered consecutively as they will appear
in the monograph. If several illustrations are treated as components
of a single composite figure, they should be designated by lowercase
italic letters on the illustration; also, in the legend and in text
references the italic letters (underlined in copy) should be used:
"Figure 9b." Illustrations that are intended to follow the printed text
may be termed Plates, and any components should be similarly
lettered and referenced: "Plate 9b." Keys to any symbols within an
illustation should appear on the art rather than in the legend.
Some points of style: Do not use periods after such abbrevia-
tions as "mm, ft, USNM, NNE." Spell out numbers "one" through
"nine" in expository text, but use digits in all other cases if possible.
Use of the metric system of measurement is preferable; where use of
the English system is unavoidable, supply metric equivalents in
parentheses. Use the decimal system for precise measurements and
relationships, common fractions for approximations. Use day/month/
year sequence for dates: "9 April 1976." For months in tabular listings
or data sections, use three-letter abbreviations with no periods: "Jan,
Mar, Jun," etc. Omit space between initials of a personal name: "J.B.
Jones."
Arrange and paginate sequentially every sheet of manuscript
in the following order: (1) title page, (2) abstract, (3) contents, (4)
foreword and/or preface, (5) text, (6) appendices, (7) notes section,
(8) glossary, (9) bibliography, (10) legends, (11) tables. Index copy
may be submitted at page proof stage, but plans for an index should
be indicated when the manuscript is submitted.
i