Contributions in science

IV

Number 476
16 December 1998

Contributions
in Science

Osteology and Phylogeny of the Cutlassfishes

( SCOMBROIDEI: TRICHIURIDAE)

F. Javier Gago

Natural History Museum of Los Angeles County

Serial
Publications
ol THE
Natural History
Museum oe
Los Angeles
County

Scientific

Publications

Committee

Robert J. Lavenberg, Deputy Director
for Research and Collections
John M. Harris, Committee Chairman
Brian V. Brown
Kenneth E. Campbell
Kirk Fitzhugh
Karen Wise

Robin A. Simpson, Managing Editor
K. Victoria Brown, Editorial Assistant

The scientific publications of the Natural History Museum
of Los Angeles County have been issued at irregular in-
tervals in three major series; the issues in each series are
numbered individually, and numbers run consecutively, re-
gardless of the subject matter.

# Contributions in Science, a miscellaneous series of tech-
nical papers describing original research in the life and
earth sciences.

# Science Bulletin, a miscellaneous series of monographs
describing original research in the life and earth sci-
ences. This series was discontinued in 1978 with the
issue of Numbers 29 and 30; monographs are now
published by the Museum in Contributions in Science.

# Science Series, long articles and collections of papers on
natural history topics.

Copies of the publications in these series are sold through
the Museum Book Shop. A catalog is available on request.

The Museum also publishes Technical Reports, a miscel-
laneous series containing information relative to scholarly
inquiry and collections but not reporting the results of
original research. Issue is authorized by the Museum’s Sci-
entific Publications Committee; however, manuscripts do
not receive anonymous peer review. Individual Technical
Reports may be obtained from the relevant Section of the
Museum.

Printed at Allen Press, Inc., Lawrence, Kansas
ISSN 0459-8113

Natural History Museum
of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

Osteology and Phylogeny of the Cutlassfishes

(ScOMBROIDEI: TrICHIURIDAE)

F. Javier Gago1

CONTENTS

ABSTRACT 1

INTRODUCTION 1

Historical Background 2

Materials and Methods 4

Results 6

DESCRIPTIVE OSTEOLOGY OF ADULTS 9

Opercular Series 10

Circumorbital Series 14

Jaws 18

Suspensorium 23

Hyoid Complex 26

Branchial Complex 31

Neurocranium 34

Pectoral Girdle 43

Pelvic Girdle 48

Axial Skeleton 52

Caudal Complex 60

OTOLITH MORPHOLOGY 65

DISCUSSION 66

ACKNOWLEDGMENTS 74

LITERATURE CITED 74

APPENDIX 77

ABSTRACT. This study describes the osteology and otolith morphology of the genera of the Trichiuridae.
Evolutionary relationships of the group are investigated based on a cladistic analysis of adult characters.
Evidence is provided for the monophyly of the trichiurids. The data support the following phyletic sequence
among the genera and clades: Aphanopus, Benthodesmus, Lepidopus caudatus-L. fitchi clade, Lepidopus
altifrons-Evoxymetopon clade, Assurger, Tentoriceps, Eupleurogr ammus, Leptur acanthus, and Trichiurus.
Osteological data suggest that Lepidopus is paraphyletic. In addition, the sister group relationship of a
gempylid clade Diplospinus-Paradiplospinus to the trichiurids is strongly supported. The most parsimo-
nious hypotheses of relationships indicate that the caudal fin has been lost only once during the evolution
of the trichiurids, whereas the pelvic fin appears to have disappeared or become reduced twice within the
group and independently in the outgroup Paradiplospinus. The data of this study are compared with
previous studies on trichiurid morphology and analyses of relationships among the scombroids. Prior stud-
ies, particularly those based on analysis of ontogenetic characters, support the results of the study reported
herein.

INTRODUCTION

The trichiurids, commonly known as cutlassfishes,
hairtails, frostfishes, scabbardfishes, or ribbonfish-
es, are benthopelagic predators inhabiting the con-

1. Research Associate, Natural History Museum of Los
Angeles County, Section of Vertebrates (Ichthyology), 900
Exposition Boulevard, Los Angeles, California 90007.

Contributions in Science, Number 476, pp. 1-79
Natural History Museum of Los Angeles County, 1998

tinental shelf and slope worldwide. Their habitats
in tropical and temperate regions range from estu-
aries to open water 2,000 m in depth. Adults are
generally identified by their extremely elongate, lat-
erally compressed bodies; a cluster of long, fang-
like teeth on the premaxillary symphysis; presence
of a single nostril on each side of the head; a lach-
rymal that covers most of the descending arms of
the maxilla and premaxilla; and reduction or ab-

sence of the caudal and pelvic fins in some genera.
Although not as important commercially as their
tuna and billfish relatives, some species of cutlass-
fishes constitute valuable fisheries in several areas
of the world, such as the East China Sea, the North
Indian Ocean, and the Mediterranean (Ye and Ro-
senberg, 1991; Nakamura and Parin, 1993).

Most authors separate the cutlassfishes and the
closely related snake mackerels into the Trichiuri-
dae and Gempylidae, respectively. Nakamura and
Parin (1993) included the Gempylidae and Tri-
chiuridae within the superfamily Trichiuroidea,
suborder Scombroidei.

The Trichiuridae comprises at least 35 species be-
longing to the following nine genera: Aphanopus
Lowe 1839, Assurger Whitley 1933, Benthodesmus
Goode and Bean 1882, Eupleurogrammus Gill
1862, Evoxymetopon Gill 1863, Lepidopus Goiian
1770, Lepturacanthus Fowler 1905, Tentoriceps
Whitley 1948, and Trichiurus Linnaeus 1758 (Nak-
amura and Parin, 1993; Parin, 1995). Evidence for
the monophyly of the trichiurids has been presented
previously (Collette et al., 1984; Johnson, 1986),
and most authors have concluded that these fishes
represent a highly derived branch of some group of
gempylids (Tucker, 1956; Matsubara and Iwai,
1958; Parin and Becker, 1972; Collette and Russo,
1986; Johnson, 1986; Potthoff et al., 1986; Naka-
mura and Parin, 1993; Carpenter et al., 1995). In
a cladistic classification of the scombroids, Johnson
(1986) placed the cutlassfishes (his subfamily Tri-
chiurinae) and the snake mackerels (his subfamilies
Gempylinae and Lepidocybiinae) as part of his
Gempylidae. Throughout this manuscript, the most
common usage of the names Trichiuridae and Gem-
pylidae (their limits according to the classification
presented by Nakamura and Parin, 1993) is main-
tained to avoid confusion when reviewing the lit-
erature. Thus, unless otherwise indicated, Trichiur-
idae (cutlassfishes) and Gempylidae (snake mack-
erels) refer to the Trichiurinae and Gempylinae plus
Lepidocybiinae of Johnson (1986), respectively.

In this study, a cladistic hypothesis of relation-
ships among the genera of the Trichiuridae ( sensu
Nakamura and Parin, 1993) is proposed, using a
comparative analysis of adult osteology, including
otoliths. The hypothesis constructed with adult
characters is compared to previous hypotheses of
relationships and, in particular, with that of Gago
(1997), which was based on larval characters.

HISTORICAL BACKGROUND

The trichiurids and gempylids were recognized as
scombroids by Cuvier (in Cuvier and Valenciennes,
1832). Later workers proposed variations on clas-
sifications of the trichiurids and considered the
gempylids and trichiurids to be closely related
(Swainson, 1839; Gunther, 1860; Gill, 1863; Ca-
pello, 1868; Goode and Bean, 1895; Boulenger,
1904; Goodrich, 1909).

Since the definition of the Scombroidei proposed

Trichiurus

Lepturacanthus

Tentoriceps

Assurger

Lepidopus

Evoxymetopon

Eupleurogrammus

Benthodesmus

Aphanopus

(Paradiplospinus)

Diplospinus

Nesiarchus

(Thyrsitoides)

Gempylus

Figure 1. General diagram of the hypothesis of relation-
ships between the trichiurid genera according to Tucker
(1956). In parentheses: Mimasea ( = Thyrsitoides ) was in-
cluded in Tucker’s (1956) study but is not part of this
analysis; Paradiplospinus was not recognized at the time
of Tucker’s (T956) study, but it is included in this analysis.

by Regan (1909), the trichiurids have consistently
been included in this suborder. Regan (1909) con-
sidered the Scombroidei to comprise the Gempyli-
dae, Istiophoridae, Luvaridae, Scombridae, Tri-
chiuridae, and Xiphiidae. The trichiurids and gem-
pylids were placed within his division Trichiurifor-
mes, characterized by: caudal fin rays not deeply
forked at the base; premaxilla beak-like and de-
tached from the nasals; mouth with a lateral cleft
and strong anterior canines; epiotic separated by
the supraoccipital; gill membranes free from the
isthmus; and pectoral fins located low on the body.

Starks (1911) and Gregory (1933) suggested a
close relationship between Gempylus Cuvier 1829
and the trichiurids. Matsubara and Iwai (1958)
suggested that the gempylids Gempylus and Mi-
masea { = Thyrsitoides ) Fowler 1929 are the most
closely related genera to the trichiurids and that
Gempylus approaches the “primitive trichiurid”
Diplospinus Maul 1948 in several characters.

Tucker (1956) presented the first modern com-
parative study of the trichiurids, including a more
thorough analysis of the gempylid-trichiurid rela-
tionships. Figure 1 shows a general diagram of the
hypothesis of Tucker (1956) based on his figure 26.
He divided the Trichiuridae into three subfamilies:
Aphanopodinae (Gill, 1863), including Aphanopus,
Benthodesmus, and Diplospinus ; Lepidopodinae
(Gill, 1863), including Assurger, Eupleurogram-
mus, Evoxymetopon, Lepidopus, and Tentoriceps ;

2 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

scombrids

billfishes

gempylids

trichiurids

Sphyraena

billfishes

- scombrids

trichiurids
other gempylids
Lepidocybium
Sphyraena

billfishes

scombrids

trichiurids

Diplospinus

Paradiplospinus

other gempylids

Tongaichthys

Lepidocybium

Sphyraena

Figure 2. General diagrams of the hypotheses of relationships among the scombroids based on morphological data and
following the work of: (A) Collette et al. (1984); (B) Collette and Russo (1986); (C) Johnson (1986); (D) strict consensus
tree from figures 5-8 of Carpenter et al. (1995).

Trichiurinae (Swainson, 1839), including Leptura-
canthus and Trichiurus. Tucker (1956) also consid-
ered the trichiurids as an offshoot of the gempylids,
and he suggested that the gempylids Gempylus,
Nesiarchus Johnson 1862, and Mimasea ( = Thyr -
sitoides ) are the most closely related to the trichiur-
ids. He hypothesized that Diplospinus and Nesiar-
chus represent the bridge connecting the two fam-
ilies, and he considered Diplospinus as the most
primitive trichiurid. However, Tucker (1956: 125)
noted that “whether the Trichiurinae crossed the
same bridge or by a parallel bridge further down-
stream is still debatable.” Andriashev (1960) de-
scribed Paradiplospinus antarcticus and considered
it the most primitive representative of the Aphan-
opodinae of Tucker (1956). Parin and Becker
(1970) recognized the Aphanopodinae of Tucker
(1956) as a natural group excluding Diplospinus
and Paradiplospinus. Parin and Becker (1972) re-
moved Diplospinus and Paradiplospinus from the
Trichiuridae and included them with the gempylids
based on the following characters: two external na-
res on each side; the number of dorsal-fin ptery-
giophores does not correspond to the number of
neural spines; a low number of caudal vertebrae;
and a gempylid-like larval morphology. They
placed Diplospinus and Paradiplospinus in the
Gempylidae but noted that these two genera occu-
py an intermediate position between the gempylids

and the trichiurids. Furthermore, they indicated
that Aphanopus is the most primitive genus within
the trichiurids.

Collette et al. (1984) placed the trichiurids as the
sister group of all other scombroids, except Scom-
brolabrax beterolepis Roule 1921, which they used
to root their cladogram (Fig. 2A). In their hypoth-
esis of relationships the gempylids appeared as a
paraphyletic group. Collette and Russo (1986) re-
evaluated the previous phylogeny and presented a
hypothesis of relationships in which the trichiurids
appeared to form a monophyletic group with the
gempylids Gempylus, Nealotus Johnson 1865, Nes-
iarchus, Promethichthys Gill 1893, Rexea Waite
1911, Thyrsitoides, and Tongaichthys Nakamura
and Fujii 1983 (Fig. 2B).

Johnson (1986) discussed the cladistic analysis of
Collette et al. (1984) and proposed an alternative
hypothesis of scombroid relationships (Fig. 2C). He
defined the monophyly of the trichiurids based on
nine meristic, osteological, external anatomy, and
larval morphology synapomorphies, including two
reversals. He recognized three gempylid subfami-
lies: Lepidocybiinae [including only Lepidocybium
flavobrunneum (Smith 1849)]; Gempylinae [Gem-
pylidae of Collette et al. (1984), excluding L. fla-
vobrunneum]-, and Trichiurinae [Trichiuridae of
Collette et al. (1984)]. In Johnson’s (1986) hypoth-
esis, Scombrolabrax Roule 1921 was placed as an

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 3

outgroup and Sphyraena Rose 1793 was included
within the Scombroidei as the sister group to all
other scombroids. Johnson (1986) considered his
Gempylidae as the sister group of all other scom-
broids, except Sphyraena. He also placed Lepido-
cybium Gill 1862 as the sister group of all the other
gempylids and trichiurids. He concluded that his
monophyletic Trichiurinae represented a highly
specialized branch of the Gempylidae with some
small group of his subfamily Gempylinae being its
sister group. Johnson (1986) noted the need for
more systematic work to resolve the precise rela-
tionships among gempylids and trichiurids.

Potthoff et al. (1986) described the development
of bone and cartilage in several scombroid groups.
They noted that the gempylids and the trichiurids
are very closely related and that the trichiurids rep-
resent a group derived from the gempylids.

The work of Block (1991), Block et al. (1993),
and Finnerty and Block (1995; Fig. 3) and data pre-
sented by Finnerty and Block at the 1994 meetings
of the American Society of Ichthyologists and Her-
petologists addressed the question of scombroid
phylogeny using molecular systematics. Although
all of their molecular data sets have consistently
supported the monophyly of the billfishes and their
separation from the gempylids, scombrids, and tri-
chiurids, the placement of Trichiurus lepturus Lin-
naeus 1758 and Gempylus serpens Cuvier 1829
(the only trichiurid and gempylid used in their anal-
ysis) is very unstable.

Carpenter et al. (1995) examined different hy-
potheses of scombroid relationships obtained by re-
analyzing the data of Johnson (1986), analyzing the
revised combined data sets of Collette et al. (1984)
and Johnson (1986), and reinterpreting the gill rak-
er character of Johnson (1986: character 44). A
summary diagram of Carpenter et al.’s (1995) hy-
pothesis is shown in Figure 2D. The different ana-
lyses based on the revision of Johnson’s (1986) data
set did not change his original hypothesis of rela-
tionships among the gempylids and the trichiurids.
In these revised analyses the trichiurids always ap-
peared as an offshoot of some group of the gem-
pylids (gempylines of Johnson, 1986), and Lepi-
docybium appeared as the sister group of all the
other gempylids and trichiurids. All of the resulting
cladograms from the analysis of Carpenter et al.’s
(1995) expanded data matrix, which included the
data of Collette et al. (1984) and Johnson (1986),
also agreed with the hypothesis of relationships
among the gempylids and trichiurids proposed by
Johnson (1986). The expanded analyses of Carpen-
ter et al. (1995), including their change in the cod-
ing of Johnson’s (1986) character 44, always re-
sulted in placing Diplospinus (which includes Para-
diplospinus in their data matrix) as the sister group
of the trichiurids. The relationships among the rest
of the gempylids, except Lepidocybium and Ton-
gaichthys, which always appear in that phyletic or-
der as the sister groups of the gempylids and tri-
chiurids, were unresolved in the unweighted anal-

ysis of Carpenter et al.’s (1995) expanded data ma-
trix (their figs. 5, 7). The placement of Gempylus
and Nesiarchus (two genera that have been previ-
ously proposed as being closely related to the tri-
chiurids) in Carpenter et al.’s (1995) weighted an-
alyses was variable. A strict consensus tree of their
142 most parsimonious trees obtained after succes-
sive character weighting of their expanded data ma-
trix (their fig. 6) placed Nesiarchus as the sister
group to an unresolved clade including the gem-
pylids Epinnula Poey 1854, Ruvettus Cocco 1829,
and Thyrsitoides. Gempylus appeared as part of a
trichotomy that includes Nealotus Johnson 1865
and a clade including Epinnula, Nesiarchus, Pro-
methichthys, Ruvettus, Thyrsitoides, and Thyrsi-
tops Gill 1862. In their weighted analysis based on
a different interpretation of a gill raker character
(their fig. 8), Nesiarchus appeared as the sister
group of a clade that includes Epinnula, Neoepin-
nula Matsubara and Iwai 1952, Ruvettus, and
Thyrsitoides. Gempylus appeared as part of a tri-
chotomy with Nealotus and a clade that includes
all of the six taxa above plus Promethichthys and
Thyrsitops. Carpenter et al. (1995) concluded that
a data set with more characters that vary within
the gempylids and trichiurids would be necessary
to resolve the relationships among these groups.

Gago (1997) analyzed the relationships among
the trichiurids based on a data matrix of ontoge-
netic characters. He included most gempylids as the
outgroups and rooted the resulting trees at Lepi-
docybium. The results of his analysis did not re-
solve the relationships among the gempylids. FIow-
ever, the ontogenetic data increased the support for
the monophyly of the trichiurids. Within the tri-
chiurids, he found that those genera lacking a well-
developed caudal fin complex constitute a clade.

MATERIALS AND METHODS
Comparative Material

The comparative material is listed in alphabetical order
according to the taxon and the institutions that provided
the specimens. The catalogue number is followed by the
number of specimens and the range (within parentheses
and in mm) of standard lengths (SL; when a caudal fin is
present) and total lengths (TL; taxa without a caudal fin).
A question mark indicates unknown or unavailable data.
All specimens listed are cleared and stained, except those
followed by the abbreviation “sk,” which indicates skele-
tonized material. Many radiographs and alcohol-pre-
served specimens as well as cleared and stained specimens
from uncatalogued collections were also studied, but are
not listed here. Institutional abbreviations follow Leviton
et al. (1985). Species names are those recognized by Nak-
amura and Parin (1993) and Parin (1995).

Gempylidae

Diplospinus multistriatus Maul 1948: LACM 45450-1 (1;

193 SL), 45604-2 (2; 178, -183 SL); USNM 194475

(1; -172 SL).

Gempylus serpens: LACM 34160-8 (1; 430 SL).

4 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

billfishes

Auxis

Euthynnus

Katsuwonus

Sarda

Thunnus

Scomber

Trichiurus

gempylids +

other

scombrids

Coryphaena

Sphyraena

Figure 3. General diagrams of the hypotheses of relationships among the scombroids based on the cytochrome b gene
data of Finnerty and Block (1995: figs. 2, 4, 7): (A) unweighted analysis of nucleotide substitution data; (B) weighted
analysis of nucleotide substitution data; (C) amino acid sequence data.

Nesiarchus nasutus Johnson 1862: USNM 236803 (1; 67
SL), 324038 (1; 276 SL).

Paradiplospinus antarcticus Andriashev 1960: LACM
10942 (1; 283 SL), 11325-22 (1; 198 SL), 11511 (1;
278 SL); USNM 208448 (1; 320 SL).

Trichiuridae

Apbanopus arigato Parin 1995: LACM 37113-1(1; 968
SL), 38240-1(1; 648 SL).

Apbanopus carbo Lowe 1839: AMS 1.25852004 (1; 418
SL).

Assurger anzac (Alexander 1916): SIO 63-229 (1; =750
SL).

Benthodesmus simonyi (Steindachner 1891): USNM
292768 (1; 537 SL).

Benthodesmus tenuis (Gunther 1877): LACM TC61-117
(1; 522 SL); SIO 82-43 (1; 369 SL).

Eupleurogrammus glossodon (Bleeker 1860): LACM
38131-15 (1; 356 TL), 38134-14 (1; 345 TL).

Evoxymetopon taeniatus Gill 1863: USNM 321690 (1;
190 SL).

Lepidopus altifrons Parin and Collette 1993: USNM
292765 (1; =390 SL), 317979 (1; =345 SL).

Lepidopus caudatus (Euphrasen 1788): AMS IA.7041 as
L. lex (1; ?)sk; USNM 268911 (1; =246 SL).

Lepidopus fitchi Rosenblatt and Wilson 1987: LACM
31683-1 (1; =239 SL), 32684-1 (1; 185 SL), 37102-1
(1; 685 SL)sk, 37102-2 (1; 785 SL)sk, 37103-1 (1;
=900 SL)sk, 38511-1 (1; 272 SL), 45602-1 (3; 198-
223 SL), 45855-1 (1; 275 SL); SIO 72-84 (2; 152-164
SL), 72-209 (1; 237 SL).

Lepturacantbus savala (Cuvier 1829): AMS IB. 1797 (1;
?)sk; LACM 38131-16 (1; 421 TL), 38134-15 (1; >275
TL), 38136-21 (1; 376 TL).

Tentoriceps cristatus (Klunzinger 1884): AMS 1.17805002

as Tentoriceps sp. nov. (1; 301 TL), 1.22830008 (1; 354
TL); LACM 44793-11 (1; 310 TL).

Trichiurus lepturus: AMS IB. 7447 as T. coxii Ramsay and
Ogilby 1887 (1; ?)sk; LACM 6945-11 (1; 340 TL),
37104-1 as T. nitens Garman 1899 (1; 488 TL)sk,
37906-22 (1; 452 TL)sk, 37955-1 (1; ?)sk, 37955-2 (1;
?)sk, 37955-3 (1; ?)sk, 37956-1 (1; ?)sk, 37957-1 as
Trichiurus sp. (1; ?)sk, 38117-100 as T. nitens (1; 1060
TL)sk, 38117-101 (1; 953 TL)sk, 33807-12 as Trichiu-
rus sp. (1; 586 TL)sk, 38130-19 (1; 288 TL); SIO 55-
58 (1; 281 TL).

I analyzed 336 sagittae from the Fitch otolith collection
at the Natural History Museum of Los Angeles County.
They included specimens of the following scombroid gen-
era: Apbanopus, Assurger, Benthodesmus, Lepidopus, and
Trichiurus (Trichiuridae); Diplospinus, Gempylus, Lepi-
docybium, Nealotus, Paradiplospinus, Promethichthys,
Rexea, and Ruvettus (Gempylidae); Euthynnus Liitken
1882 (in Jordan and Gilbert 1882), Grammatorcynus Gill
1862, Scomberomorus Lacepede 1801, and Thunnus
South 1845 (Scombridae); Istiophorus Lacepede 1801,
Makaira Lacepede 1802, and Tetrapturus Rafinesque
1810 (Istiophoridae); Xiphias Linnaeus 1758 (Xiphiidae).
Six sagittae of the trichiurid species Lepturacantbus savala
were obtained as a loan from the Australian Museum
(1.21955-013). In addition, a drawing of the medial face
of a left sagitta of the gempylid Nesiarchus nasutus (pro-
vided by D. Nolf, Institut Royal des Sciences Naturelles,
Belgium), plus several descriptions and drawings of sag-
ittae of other scombroid species (taken from the litera-
ture), were compared.

Comparative Analysis

Osteological characters were examined from radiographs,
cleared and stained specimens, and dry skeletal prepara-

Contributions in Science, Number 476

Gago: Trichiurid Phytogeny ■ 5

tions of adults. I followed the method of Potthoff (1984)
for clearing and staining. A Wild M-5 dissecting micro-
scope with a camera lucida was used for the preparation
of drawings. Representative drawings of most of the adult
osteological characters of each of the trichiurid genera are
included for comparative purposes throughout the text. In
most cases within the outgroups, I have only included os-
teological drawings of Diplospinus and Paradiplospinus
since detailed drawings of most gempylids can be found
in the work of Russo (1983). Osteological terminology
follows mainly the works of Collette and Chao (1975) and
Collette and Russo (1984), unless otherwise indicated.

The medial face of the sagittae was lightly rubbed with
graphite for enhancement of morphological features. The
sagittae were examined under a Wild M-5 dissecting mi-
croscope, and scanning electron microscopy was per-
formed on representative specimens. Terminology for the
otolith morphology follows Chaine and Duvergier (1934).

Phylogenetic Analysis

Polarity of the characters was determined by outgroup
comparison. Choosing the outgroup is a critical step in
cladistic analysis since character argumentation depends
heavily on this decision. For the trichiurids this step was
facilitated by the large amount of data available regarding
scombroid interrelationships. As previously indicated,
most workers have proposed a close relationship between
gempylids and trichiurids.

Although noncladistic, the works of Tucker (1956) and
Parin and Becker (1972) suggested that the gempylids
Diplospinus, Gempylus, Nesiarchus, and Paradiplospinus
are the best candidates for trichiurid outgroups. In addi-
tion, prior to the work of Parin and Becker (1972), Dip-
lospinus and Paradiplospinus were included in the Tri-
chiuridae. The consistent earlier placement of these two
genera within the Trichiuridae was based on numerous
similarities among these fishes and was interpreted by
Tucker (1956) as evidence of common ancestry.

Johnson (1986) concluded that the trichiurids represent
a highly derived offshoot of a paraphyletic Gempylidae
(his subfamily Gempylinae). The work of Russo (1983) is
particularly important because it provides a data matrix
of osteological characters for the species of Gempylidae.
Russo (1983) could not demonstrate the monophyly of the
Gempylidae but concluded that the family comprises six
groups. His most derived group was composed of the fol-
lowing genera in phyletic sequence: Thyrsitoides, Nesiar-
chus, Gempylus, and his Paradiplospinus-Diplospinus
clade. Although Paradiplospinus was not recognized at
the time of Tucker’s (1956) study, Diplospinus, Gempylus,
Nesiarchus, and Mimasea Kamohara 1936 ( = Thyrsito -
ides Fowler 1929) were placed as the basal branches of
his trichiuroid tree. Furthermore, the results of the work
of Carpenter et al. (1995) always placed the Diplospinus-
Paradiplospinus clade as the sister group to the trichiurids.

A preliminary survey of the adult osteology and otolith
morphology of the trichiurids and gempylids and its com-
parison with hypotheses and data sets presented by pre-
vious authors (Tucker, 1956; Parin and Becker, 1972; Rus-
so, 1983; Collette et al., 1984; Johnson, 1986; Carpenter
et al., 1995) indicate that Diplospinus and Paradiplospi-
nus are the closest sister groups to the trichiurids. Thus,
these two genera, plus Gempylus and Nesiarchus, are used
as the outgroups in this study.

Trees were rooted at a basal polytomy with Nesiarchus
and Gempylus, and following the conclusions of most au-
thors, Diplospinus and Paradiplospinus were used as a
clade representing the sister group to the trichiurids. For

those characters that appear to be heterogeneous among
the outgroups and in which the hypothesized state at the
outgroup node was equivocal (Maddison et al., 1984), the
plesiomorphic condition was determined by comparison
to the conditions in other gempylids and scombrids based
on information taken from the literature. I agree with the
conclusions of Russo (1983), Johnson (1986), and Car-
penter et al. (1995) and consider Lepidocybium as the
most basal gempylid. Thus, the plesiomorphic condition
is assumed to be that present in Lepidocybium . If the con-
dition in Lepidocybium is unknown, the plesiomorphic
condition is assumed to be that which is most common
among other gempylids or scombroids. Although “most
common” does not necessarily indicate plesiomorphy
(Maddison et al., 1984; Wiley et al., 1991), “the primitive
state of a character for a particular group is likely to be
present in many of the representatives of closely related
groups” (Kluge and Farris, 1969 p. 5). Maddison et al.
(1984) indicated that this form of outgroup methodology
can lead to cladograms that are not globally parsimoni-
ous. This type of character assessment is used in this study,
but more definite conclusions on the polarity of those
characters that appear to be equivocal must await the re-
sults of a study in progress that includes all gempylids and
trichiurids (F.J. Gago and J.L. Russo, unpublished data).
The distribution among other gempylids and scombroids
of the conditions of those characters that are equivocal is
discussed under each of the osteological sections where
these characters are described.

A data matrix of characters (Table 1) was constructed
using MacClade version 3 (Maddison and Maddison,

1992) and analyzed with PAUP version 3.1.1 (Swofford,

1993) . Character states were coded as numerals and “?,”
where 0 represents the plesiomorphic condition and “?”
missing or nonapplicable data

RESULTS

All multistate characters (Table 1) were consid-
ered unordered, and the trees were rooted at a basal
polytomy with Gempylus and Nesiarchus. An anal-
ysis using the branch-and-bound algorithm and
ACCTRAN transformations resulted in three
equally most parsimonious trees. These three trees
have 99 steps, a consistency index of 0.869, and a
rescaled consistency index of 0.809. All three trees
have the same topology as the tree of Figure 4, ex-
cept for the resolution between Evoxymetopon,
Lepidopus altifrons, L. caudatus, and L. fit chi. Fig-
ure 5 shows the only portion of the topology
among the most parsimonious trees that is variable.
The variation in the topology of these trees is the
result of different interpretations about the evolu-
tion of character 36. Figure 5 only includes char-
acter 36 since the interpretation of all the other
characters analyzed using ACCTRAN is identical
for all three trees at these variable nodes.

One tree places Lepidopus caudatus and L. fitchi
as a monophyletic group (node IVa) and Evoxy-
metopon and L. altifrons as an unresolved polyto-
my at node V (Fig. 5A). In this tree the derived
condition of character 36 is assumed to have
evolved independently in the clade uniting L. cau-
datus and L. fitchi (node IVa) and in the monophy-
letic group above node V. This tree topology also

6 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Table 1. Data Matrix of Adult Characters. 0 = plesiomorphic state; 1, 2, 3, 4 = apomorphic states; ? = missing data
or not applicable.

Characters

Taxa

10 20 30 40 50 60

Aphanopus

Assurger

Benthodesmus

Eupleurogrammus

Evoxymetopon

Eepidopus altifrons

L. caudatus

L. fitchi

Lepturacanthus

Tentoriceps

Trichiurus

Diplospinus

Gempylus

Nesiarchus

Paradiplospinus

100112000101000211010101001000000000100701021020110000000212111110
110112011001000211011121111001111101100101021121211001000212111010
12011200100100021101010100100011 010 0100101021020210000000212111100
120112010011011321010111101000121211110111121121311114121?????????
120112011001000211011121111001111100100101021021311001000212111???
120112011001000211011121111001111100100101021021311001000212111???
120112011001000211010101101001111101100101021021311001000212111010
120112011001000211010101101001111101100101021021311001000212111010
121112112011111321010101101000121210111??? ?2112 13 11112121??????011
120112010001001321011121111001121101110111121121311113111?????????
121112112011111321010101101000121210111????211213 1111212 1??????011
000011000000000200100001000110000000000100010000000000000102100100
000000000100000100000000000000200200010000000000000000000001000000
000000000000000000000000000000000000000000000010000000000000000000
000011000000000200100001000110000000000?00010000000000000102100100

assumes that character 36 reverted to the plesiom-
orphic condition at node IX.

A second tree places Evoxymetopon and Lepi-
dopus altifrons as a clade (node Va) and L. cau-
datus and L. fitchi as an unresolved polytomy at
node IV (Fig. 5B). Finally, the third tree places L.
caudatus and L. fitchi and Evoxymetopon and L.
altifrons in two separate clades, respectively (Fig.
5C, nodes IVa and Va). This last hypothesis of re-
lationships has a “zero length” branch (node IVa),
but it is included in the results of the analysis by
PAUP to indicate that there is potential support for
these monophyletic groups under some “most par-
simonious reconstructions” (Swofford, 1993). The
trees of Figure 5B and 5C assume that the derived
condition of character 36 evolved only once below
node IV and was lost independently in the clades
including Evoxymetopon and Eepidopus altifrons
(node Va) and Lepturacanthus and Trichiurus
(node IX), respectively. A different interpretation of
the third tree using DELTRAN assumes that the
derived condition of character 36 evolved indepen-
dently in the clade above node V and the clade of
node IVa. This alternative hypothesis also con-
cludes that character 36 reverted to the plesiom-
orphic condition in node IX and that the branch
leading to node Va is “zero length.”

The tree in Figure 4 is not considered to be the
final hypothesis of relationships. One must be
aware that there are three equally parsimonious hy-
potheses as described earlier in this section and the
tree of Figure 4 is just one of these hypotheses.
However, because the three hypotheses differ only
in the interpretation of a single character (character
36), the tree in Figure 4 serves as a good summary
of the hypotheses of relationships and character
evolution. The description that follows is based on
the results of the analysis using ACCTRAN, as pre-

sented in Figure 4. An analysis of the tree in Figure
4 using DELTRAN results in different interpreta-
tions of character transformations for 10 of the
characters utilized. For comparative purposes, Fig-
ure 6 shows the distribution of character states
when using DELTRAN.

A monophyletic group including the gempylid
genera Diplospinus and Paradiplospinus (node la)
is supported by three synapomorphies and appears
as the sister group of the trichiurids. The mono-
phyletic group including these two outgroup genera
plus the trichiurids (node I) is supported by 10 syn-
apomorphies, including only one homoplasy.

Aphanopus appears as the sister group to the rest
of the trichiurids at node II. A total of 21 synapo-
morphies, including only two homoplasies, support
the monophyly of the trichiurids (node II).

The monophyly of the group that includes all the
trichiurids except Aphanopus (node III) is support-
ed by six synapomorphies, including three homo-
plasies. A monophyletic group including all tri-
chiurids except Aphanopus and Benthodesmus
(node IV) is supported by 10 synapomorphies, in-
cluding four homoplasies.

As indicated earlier, in the tree of Figure 4 the
branch leading to the monophyletic group of Lep-
idopus caudatus and L. fitchi is interpreted as hav-
ing no support in the form of synapomorphies (zero
length). However, the results using DELTRAN in-
clude character 36 as a synapomorphy at this node
(Fig. 6).

The clade including Evoxymetopon and Lepi-
dopus altifrons appears as the sister group to the
monophyletic group that includes Assurger, Tento-
riceps, Eupleurogrammus, Lepturacanthus, and
Trichiurus, in that phyletic order. The monophyletic
group including these seven taxa (node V) is sup-

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 7

TL: 99
Cl: 0.869
RC: 0.809

1-1, 4-1
6-2, 12-1
17-1, 18-1
20-1, 22-1
27-1,37-1
42-1,44-2
45-1,47-2,
(49-1), 50-1
58-2, 59-1
62-1, 63-1
(65-1)

5-1, 6-1
16-2, 24-1
40-1, 44-1
58-1, 60-2
61-1, (64-1)

II

2-2, (9-1)
31-1, 32-1
(34-1), (49-2)

III

I

3-1, 7-1

(65-0)

(10-1)

(10-1), 16-1, 31-2,
(34-2), (38-1), 60-1

Trichiurus
Leptur acanthus
Eup leurogrammus
Tentoriceps
Assurger
Evoxymetopon
L. altifrons

L. caudatus
L. fitchi

Benthodesmus

Aphanopus

Diplospinus

Paradiplospinus

47-1

Gempylus

Nesiarchus

Figure 4. Hypothesis of relationships resulting from the branch-and-bound analysis of the data matrix. Character trans-
formations follow ACCTRAN. Homoplastic characters are enclosed within parentheses; character numbers are followed
by the state present at each respective node.

ported by three homoplasies (independent acquisi-
tions).

A clade including Evoxymetopon and Lepidopus
altifrons (node Va) is only supported by a reversal
in character 36 (Fig. 5B, C). An analysis using DEL-
TRAN includes this node, although it provides no
support for its monophyly (Fig. 6; node Va).

Node VI places Assurger as the sister group to

the clade formed by Tentoriceps, Eupleurogram-
mus, Leptur acanthus, and Trichiurus, but this is
supported by only a single synapomorphy. The sis-
ter group relationship between Tentoriceps and the
clade including Eupleurogrammus, Lepturacan-
thus, and Trichiurus (node VII) is supported by 15
synapomorphies, including two homoplasies.

Furthermore, Eupleurogrammus appears as the

8 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 5. General diagrams of the most parsimonious hypotheses obtained during the branch-and-bound analysis of the
adult data matrix (Appendix). Character transformations follow ACCTRAN. The direction of change between the states
of character 36 (Appendix) is indicated with an arrow at each of the nodes. Dashed lines above and below nodes IV and
V indicate portions of the hypotheses that are identical to those in the tree in Figure 4. Open circle = zero length branch.

sister group to Lepturacanthus and Trichiurus
(node VIII). The monophyletic group including
these three genera is supported by nine synapo-
morphies, including five homoplasies.

Lepturacanthus and Trichiurus appear as a
monophyletic group (node IX). Six synapomor-
phies, including two homoplasies, support the
monophyly of these two genera.

A tree with the topology presented by Tucker
(1956; Fig. 1) was created using MacClade and im-
ported into PAUP as a topological constraint for a
branch-and-bound search using the data matrix of
Table 1, but excluding Paradiplospinus and Thyr-
sitoides from the analysis. As mentioned earlier,
Tucker’s (1956) tree places Gempylus (plus Thyr-
sitoides, which was not included in this study), Nes-
iarchus, and Diplospinus ( Paradiplospinus was not
recognized at the time of Tucker’s 1956 study), in
that phyletic order, as the sister taxa of the trichiur-
ids. He proposed three major groups in his tree:
Aphanopus-Benthodesmus (his subfamily Aphano-
podinae minus Diplospinus ); Lepturacanthus-Tri-
chiurus (his subfamily Trichiurinae); Assurger-Eu-
pleurogrammus-Evoxymetopon-Lepidopus-Tento-
riceps (his subfamily Lepidopodinae). The branches
leading to these three groups appear as a trichoto-

my above Diplospinus. Within his monophyletic
Lepidopodinae he shows a trichotomy that includ-
ed the following groups: Assurger-Tentoriceps ; Eu-
pleurogrammus-Evoxymetopon ; Lepidopus. A
branch-and-bound search with all characters treat-
ed as unordered resulted in three equally most par-
simonious trees with a length of 134 steps and
consistency and rescaled consistency indexes of
0.672 and 0.468, respectively. In all three resulting
trees the clades Aphanopus-Benthodesmus and
Lepturacanthus-Trichiurus appear, in that phyletic
order, as the sister groups to Tucker’s Lepidopodi-
nae. The trees differ in the three possible resolu-
tions among the clades proposed by Tucker (1956)
within the Lepidopodinae.

DESCRIPTIVE OSTEOLOGY OF ADULTS

The characters for the phylogenetic analysis are in-
dicated with numbers that correspond to those in
the data matrix (Table 1), list of characters (Ap-
pendix), and phylogenies (Figs. 4-6). Character
numbers are followed by the state present at each
particular node. Character numbers in parentheses
represent homoplasies.

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 9

3-1, 7-1

(9-2),13-l

Trichiurus
Leptur acanthus
Eup leurogr animus
Tentoriceps
Assurger
Evoxymetopon
L. altifrons

L. caudatus
L. fitchi

Benthodesmus

Aphanopus

Diplospinus

Paradiplospinus

Gempylus

Nesiarchus

Figure 6. Hypothesis of relationships resulting from the branch-and-bound analysis of the data matrix. Character trans-
formations follow DELTRAN. Homoplastic characters are enclosed within parentheses; character numbers are followed
by the state present at each respective node. Those interpretations of character transformations that differ from the results
using ACCTRAN (Fig. 4) are enclosed within a rectangle.

OPERCULAR SERIES

The bones of the opercular series in both the tri-
chiurids and gempylids are poorly ossified. The
thickest, most strongly ossified areas, which are the

articular corners of the opercle, subopercle, and in-
teropercle, are spongy in appearance (Fig. 7).

Character 1 . In the trichiurids, the posterior and
ventral margins of the opercle, subopercle, and in-
teropercle (the interopercle to a lesser degree) are

10 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

strongly splintered or fimbriated (Fig. 7C-K). John-
son (1986: character 26) considered this condition
(including the ventral margins of the lachrymal) as
a synapomorphy of his Trichiurinae. In the out-
groups, the posterior and ventral margins of the
opercle, subopercle, and interopercle are mostly
complete (only the dorsal flap of the opercle and
the posterior corner of the subopercle may be
slightly splintered; Fig. 7A, B).

Opercle

The opercle of all gempylids and trichiurids is
quadrilateral and characterized by the presence of
a posterodorsal notch. Russo (1983: character 55)
indicated that the posterodorsal notch in all gem-
pylids, except Gempylus and Lepidocybium, is
deep and bordered by a wide dorsal flap. In con-
trast, Gempylus and Lepidocybium have shallow
opercular notches with a dorsal margin that tapers
posteriorly to a point. The trichiurids have a deep
dorsal notch and the dorsal flap above it is narrow-
er than that in most gempylid genera, but it is not
pointed.

Russo (1983: character 56) noted that all gem-
pylids, except Diplospinus, Gempylus, Lepidocy-
bium, Nealotus, Nesiarchus, Rexea, Thyrsites Cu-
vier (in Cuvier and Valenciennes 1832), and Thyr-
sitoides, bear a spinous ventral margin on the pos-
terodorsal notch of the opercle. All of the
trichiurids analyzed here have opercular notches
with ventral margins that are not pointed or spi-
nous. Some specimens of Paradiplospinus utilized
in this study have a small feeble spinous ventral
margin on the opercular notch, and one specimen
of Diplospinus multistriatus has a spinous opercu-
lar notch on only one side of the body. The pres-
ence or absence of a spinous ventral margin on the
opercular notch is variable within these two out-
group genera.

The opercle articulates with the posterior condyle
of the hyomandibula by an anterodorsal articular
head that houses an articular fossa. All gempylids,
except Lepidocybium, and the trichiurids share this
condition. At its base, the anterodorsal articular
head bears one to three struts or ridges that support
the main body of the opercle medially. The articular
head is short in all the genera studied, except Eu-
pleurogrammus, Leptur acanthus, Tentoriceps, and
Tricbiurus, in which it clearly projects from the an-
terodorsal corner of the opercle by way of an elon-
gate neck-like base. However, it is difficult to cat-
egorize this condition objectively because Assurger,
Evoxymetopon, and Lepidopus share an interme-
diate state in the elongation of the articular head.

Character 2. The outgroup genera and the tri-
chiurid Aphanopus have a lateral plate-like process
at the anterodorsal corner of the opercle that covers
all or most of the articular head (Fig. 8A). In all
the other trichiurids, except Assurger, the articular
head lacks a lateral plate, but it bears a well-devel-
oped elongate lateral process that is round in cross

section (Fig. 8C). Assurger also bears an elongate
lateral process, but it is flat in cross section (Fig.
8B). In Gempylus the plate-like process is dorsally
elongate, but the condition is not comparable with
that of Assurger. The flat process of Assurger cross-
es the articular head of the opercle at about its cen-
ter (as in those trichiurids with an elongate process
that is round in cross section), whereas the flat pro-
cess of Gempylus extends posterodorsally of the ar-
ticular head and appears to be a modification of
the plate covering the dorsal margin of the articular
head. The lateral process is longer in Assurger, Eu-
pleurogrammus, Lepidopus, Tentoriceps, and Tri-
chiurus, but the degree of elongation is difficult to
categorize objectively.

Subopercle

The subopercle of all the genera analyzed in this
study is flat and triangular. The dorsal margin of
the subopercle abuts the ventral margin of the op-
ercle medially. The subopercle of all trichiurids is
poorly ossified and has a strongly fimbriated or
splintered ventral margin, whereas the ventral mar-
gin of the subopercle in the outgroup genera is
mostly complete, and fimbriations, if present, are
extremely reduced and restricted to the posterodor-
sal corner of the bone.

Character 3. The fimbriations on the posteroven-
tral corner of the subopercle in the trichiurid genera
Lepturacanthus and Trichiurus are longer than the
preceding ventral ones. Thus, the contour formed
by the ventral margin of the subopercle and the
posteroventral corner of the opercle in these two
genera appears to be slightly concave (Fig. 71, K).
Tucker (1956) used this character to group these
two genera into his subfamily Trichiurinae. All the
other trichiurids and the outgroups have a subop-
ercle which, together with the opercle, forms a con-
vex ventral margin.

Character 4. The outgroups and trichiurids are
characterized by the presence of an anterodorsal
articular process on the subopercle. In the out-
groups, the articular process extends dorsally at a
right angle to the dorsal margin of the subopercle
and articulates mainly with the anteroventral cor-
ner of the opercle (Fig. 9A). A much smaller an-
terior projection on the articular process articu-
lates with the interopercle. All trichiurids possess
an articular process that extends anteriorly, artic-
ulating mainly with the posterodorsal corner of
the interopercle (Fig. 9B). The articular process is
short in Aphanopus, Benthodesmus, and Leptur-
acanthus, whereas it is longer and more pointed
in other trichiurids. The degree of elongation of
the anterodorsal process seems to vary in a con-
tinuous gradient from pointed to rounded and is
difficult to categorize.

Interopercle

The interopercle is triangular and poorly ossified.
The dorsal margin is characterized by a shallow lat-

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 1 1

Figure 7. A-E. Lateral view of the left opercular series: (A) Diplospinus multistriatus; (B) Paradiplospinus antarcticus;
(C) Aphanopus arigato ; (D) Assurger anzac ; (E) Benthodesmus tenuis.

eral fossa, which serves as a facet for the ventral
portion of the posterior wing of the preopercle.
Posteriorly, the margin of the interopercle overlaps
the anterior margin of the subopercle laterally. The
posterodorsal corner of the interopercle articulates
with the anterodorsal articular process of the sub-
opercle. Medially, the interopercle has a small an-
terodorsal fossa that articulates with the posterior

corner of the epihyal and the interhyal. The inter-
opercle is similar among the species studied and dif-
fers only slightly in its overall shape.

Preopercle

The preopercle is crescent-shaped with a longitu-
dinal lateral-line canal on the anterior margin and

12 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 7. F-K. Lateral view of the left opercular series: (F) Eupleurogrammus glossodon; (G) Evoxymetopon taeniatus ;
(H) Lepidopus fitchi ; (I) Leptur acanthus savala; (J) Tentoriceps cristatus; (K) Trichiurus lepturus.

a posterior strut on the dorsal half of the bone. The
canal and the strut converge in the shape of a “Y.”
The anterior longitudinal canal carries the preoper-
cular branch of the laterosensory canal system
(Coombs et ah, 1987), which it receives from a lat-
eral pore on the pterotic. This preopercular canal
exits the preopercle at the anteroventral tip to enter
the articular bone. As it passes along the preoper-
cle, the preopercular canal opens laterally through
sensory pores. The preopercle is the most heavily

ossified bone in the opercular series. The posterior
wing of the preopercle is less ossified than the cen-
tral axis, and it overlays the anterior margin of the
opercle and a shallow, dorsolateral fossa on the in-
teropercle.

The posterior margin of the preopercle of all
adult trichiurids is smooth and devoid of spines.
Russo (1983) noted the variability in the postero-
ventral margin of the preopercle of gempylid gen-
era. He indicated that the preopercular margin of

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny H 13

Figure 8. Lateral (top) and medial (bottom) views of the lateral process on the articular head of the left opercle: (A)
Diplospinus multistriatus ; (B) Assurger anzac ; (C) Trichiurus lepturus.

the gempylids can be smooth ( Diplospinus ) or ser-
rate ( Paradiplospinus ), have ventrally or dorsally
directed spines ( Gempylus and Nesiarchus ), or be
irregular in shape because of poor ossification. In
some cases, the presence or absence of spines is dif-
ficult to evaluate because their recognition depends
on the degree of ossification or development. For
example, Russo (1983) noted the presence of small
spines in Paradiplospinus, but concluded that they
could be interpreted as serrations because of their
poor development. Russo (1983) omitted this char-
acter from his analysis because of the difficulty in
its interpretation.

Character 5. The preopercle of the outgroups
Gempylus and Nesiarchus bears a convex postero-
dorsal margin. Russo (1983: character 48) identi-
fied the presence of a slightly concave posterodorsal
margin on the preopercle as a synapomorphy of his
gempylid clade comprising Diplospinus and Para-
diplospinus. All trichiurids examined in this study
share this feature. Because of the presence of a con-
vex posterodorsal margin in the preopercle of all
gempylids, except Diplospinus and Paradiplo-
spinus, I agree with Russo (1983) and consider the
presence of a concave posterodorsal margin as the
derived condition.

CIRCUMORBITAL SERIES

The circumorbital series is a group of poorly ossi-
fied, small bones that carry the infraorbital branch
of the lateral sensory canal system. Russo (1983)
indicated that the numbers of left circumorbital el-

ements present in the gempylids Diplospinus mul-
tistriatus, Gempylus serpens, Nesiarchus nasutus,
and Paradiplospinus gracilis (Brauer 1906) were
26, 29, 21, and 21, respectively. However, he noted
that the number of circumorbitals varied not only
within species but also between sides of the same
specimen. Jollie (1986) indicated that hypotheses of
homology concerning the infra- and postorbitals
(excluding the lachrymal and jugal) are impossible
because of the variation in the number of elements,
as well as the arbitrariness of their recognition.

Character 6. All of the circumorbital series of the
gempylids examined by Russo (1983: character 35)
are continuous, except those of Diplospinus and
Paradiplospinus, which have a short gap separating
the first infraorbitals from the posterior circumor-
bitals (Fig. 10A, B). Russo (1983) considered the
condition present in Diplospinus and Paradiplo-
spinus as a synapomorphy uniting these two gen-
era. A further derived condition, as explained by
Johnson (1986: character 27), is present in the tri-
chiurids, which have an extremely reduced circu-
morbital series (only the lachrymal and jugal are
present, Fig. 10C-K). Although this character ap-
pears equivocal at the outgroup node in this study,
I agree with Russo (1983) and Johnson (1986) in
their polarization of the character states.

Lachrymal

The lachrymal constitutes the largest element of the
circumorbital series. The ventral wing of the lach-
rymal is poorly ossified and membranous. The

14 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 9. Lateral view of the articular process of the left
subopercle: (A) Paradiplospinus antarcticus ; (B) Lepido-
pus altifrons.

poorly ossified ventral margin of this wing is fim-
briated or splintered in the trichiurids. The latter
character state was included by Johnson (1986) as
part of his character complex 26. 1 do not consider
this condition to be independent of that present in
some of the opercular series (character 1). All tri-
chiurids are characterized by a lachrymal with an
extremely large ventral wing that completely covers
the descending arms of the maxilla and premaxilla
and in some cases extends past the ventral margin
of the premaxilla. This character was recognized
and used by Regan (1909) to define his family Tri-
chiuridae. In the outgroups, the ventral wing does
not cover the descending arms of the maxilla and
premaxilla completely, and the ventral margin of
the wing is complete. Matsubara and Iwai (1952)
noted that the maxilla in Gempylus is largely hid-
den by the infraorbital membrane. However, vari-
ation in the extent of the ventral wing is continuous
and the character is not included in the analysis
since its categorization into objective states is dif-
ficult.

The ventral wing of the lachrymal is joined to a
dorsal, longitudinal lateral-line canal. A perpendic-
ular dorsal process of this longitudinal canal di-
vides it into an anterior and a posterior section and
forms the articular process that joins the lateral eth-
moid. All the outgroups and trichiurids are char-
acterized by articular processes that are elongate
and pointed.

Character 7. In the trichiurids Lepturacanthus
and Trichiurus, the ventral wing of the lachrymal
becomes separated from the longitudinal, dorsal
lateral-line canal at its anterior and posterior tips.
The fimbriations in the anterior and posterior mar-
gins of the ventral wing of Lepturacanthus and Tri-
chiurus extend vertically and are almost perpendic-
ular to the dorsal, longitudinal lateral-line canal.
Thus, Lepturacanthus and Trichiurus have a ven-
tral wing on the lachrymal that appears quadrilat-
eral (Fig. 101, K). The rest of the trichiurids and all
outgroups, except Gempylus, have a lachrymal in
which the anterior and posterior portions of the
ventral wing are connected to the anterior and pos-
terior tips of the dorsal, longitudinal lateral-line ca-
nal. In Gempylus and Tentoriceps, the anterior
margin of the ventral wing seems to become slightly
separated before its anterior tip. However, this con-
dition is not comparable to that in Lepturacanthus
and Trichiurus, where the anterior tip of the dorsal,
longitudinal lateral-line canal extends free of the
ventral wing for a longer distance. In the outgroups
and the rest of the trichiurids, the fimbriations in
the anterior and posterior portions of the ventral
wing of the lachrymal are not perpendicular to the
dorsal, longitudinal lateral-line canal, and the ven-
tral wing appears ovoid (Fig. 10A-H, J).

Character 8. The posterodorsal angle between
the articular process and the posterior section of the
dorsal, longitudinal lateral-line canal is strength-
ened by a plate-like ossification. This plate-like os-
sification extends posteriorly and terminates before
or above the posterior pore of the dorsal, longitu-
dinal lateral-line canal in the outgroups and the tri-
chiurids Aphanopus and Benthodesmus (Fig. 10A-
C, E). In the rest of the trichiurids this plate-like
ossification extends past the posterior pore of the
longitudinal, dorsal lateral-line canal and ends
above the jugal (Fig. 10D, F-K).

Jugal

The jugal is the next bone, posterior to the lach-
rymal. Russo (1983: character 33) indicated that in
all gempylids, except Lepidocyhium and Ruvettus,
the articulation between the lachrymal and the ju-
gal is weak and the two elements are not in contact.
Although the outgroup specimens analyzed in this
study, except Gempylus where the jugal and lach-
rymal are separated, have a weak articulation be-
tween the jugal and lachrymal, these elements re-
main in contact with each other. The contact be-
tween the first two infraorbitals seems to be com-
pletely lost only in the trichiurids and Gempylus.
The degree of separation between the lachrymal
and the jugal is difficult to interpret and appears to
be highly variable. Jollie (1986) considered the term
jugal as useful only in its positional reference. Prob-
lems of interpretation of the homology of this bone
may be, in part, a result of the probable separation
of free lateral-line canal units from the posterior
end of the lachrymal.

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 15

16 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 10. A-E. Lateral view of the left circumorbital series: (A) Diplospinus multistriatus ; (B) Paradiplospinus antarcticus; (C) Aphanopus arigato’, (D) Assurger anzac, (E) Bentbodesmus

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 17

Figure 10. F-K. Lateral view of the left circumorbital series: (F) Eupleurogrammus glossodon ; (G) Evoxymetopon taeniatus ; (H) Eepidopus fitchi ; (I) Lepturacanthus savala ; (J) Tentoriceps
cnstatus; (K) Tnchiurus lepturus. K F

The jugal in the gempylids Diplospinus , Nesiar-
chus, and Paradiplospinus and the trichiurids
Aphanopus carbo, Benthodesmus , Leptur acanthus,
and Trichiurus is a simple tube with ventral and
dorsal laminar ossifications that are extremely re-
duced (i.e., they do not extend along the entire
length of the bone) or absent. Gempylus and the
trichiurids Aphanopus arigato, Assurger, Eupleu-
rogrammus, Evoxymetopon, Lepidopus, and Ten-
tor iceps have a jugal with well-developed ventral
and dorsal laminar extensions that extend along the
whole or most of the length of the bone. However,
the condition is variable within genera (e.g.,
Aphanopus) and difficult to interpret since it de-
pends on the size and degree of ossification of the
specimens. The cleared and stained specimen of
Aphanopus arigato analyzed in this study had well-
developed laminar ossifications in the jugal, where-
as a smaller specimen of Aphanopus carbo lacked
the condition. Some specimens of Diplospinus and
Trichiurus have small laminar ossifications anteri-
orly. Other specimens seem to show the presence of
membranous dorsal and ventral extensions, which
are not yet ossified.

Postorbital Ossification

Character 9. Assurger, Benthodesmus, Evoxy-
metopon, Lepidopus, Leptur acanthus, and Trichiu-
rus have a paired, ossified element suspended in the
adipose tissue of the posterior margin of the orbit
(Fig. 11). The postorbital ossification in Leptura-
canthus and Trichiurus is large, thick, and strongly
ossified, whereas in Assurger, Benthodesmus, Evox-
ymetopon, and Lepidopus it is small, thin, and
poorly ossified. This element does not appear to
have a lateral-line canal. Laterally, it appears as a
crescent-shaped ossification, but upon close exam-
ination it is plate-like and extends medially as a
small postorbital shelf. The posterior face of the
postorbital ossification tends to be slightly convex,
whereas the anterior face is concave. In anterior
view, the postorbital ossification of Leptur acanthus
and Trichiurus appears triangular. The shape of this
element in Assurger, Benthodesmus, Evoxymeto-
pon, and Lepidopus is quite variable and dependent
on the size of the specimens. In Evoxymetopon and
Lepidopus the postorbital ossification bears dorsal
or ventral processes at the lateral or medial corners.
The outgroups and the rest of the trichiurids lack
postorbital ossifications. Johnson (1986), Senta
(1975), and James (1961) mentioned the presence
of these ossifications but did not attempt any hy-
potheses of homology. Senta (1975) and James
(1961) described these ossifications as the dermo-
sphenotic. Although the homology between the
conditions observed in Lepturacanthus and Tri-
chiurus and the rest of the trichiurids having a post-
orbital ossification is inconclusive, I include this
character complex in the data matrix as a multi-
state character.

JAWS

Johnson (1986) indicated that the scombroids are
characterized by jaws with teeth that are peripher-
ally ankylosed to the walls of a longitudinal crypt
(tooth attachment Type 1 of Fink, 1981). He also
indicated that tooth replacement within these lon-
gitudinal crypts occurs between the mature anky-
losed teeth.

Accurate counts of the actual numbers of mature,
ankylosed teeth in trichiurids are difficult because
the old teeth are rapidly shed or resorbed after re-
placement and are easily lost in preserved speci-
mens. There is no apparent systematic pattern of
replacement, and within a species, two specimens
of the same size may not correspond in their dental
formula or arrangement of replacement and mature
teeth. Soot-Ryen (1936) indicated that in Aphano-
pus minor Collett 1887 {=A. carbo), the number
and placement of teeth vary considerably between
individuals and so cannot be used as specific char-
acters.

The trichiurids, as well as all the other scom-
broids, have teeth of reticulate or cancellous ap-
pearance internally when viewed in glycerin (John-
son, 1986). A longitudinal crypt extends along the
dorsal margin of the dentary and the ventral mar-
gin of the premaxilla, serving as the base for a uni-
serial row of teeth in each of these bones. These
teeth are mediolaterally flattened and triangular,
with well-developed anterior and posterior cutting
edges, except in the dentary of Nesiarchus where
they are retrorse. Large specimens of Nesiarchus
and the trichiurids Lepturacanthus and Trichiurus
are distinct in that some of the anterior-most teeth
in the longitudinal series of the dentary are barbed.
The serial teeth of the trichiurids Lepturacanthus
and Trichiurus have well-developed barbs at their
points, whereas those of Nesiarchus have no barbs
or are extremely reduced in some specimens.

Most trichiurids and outgroups have a pair of
small fangs on the dentary (Fig. 12). These fangs
are barbed and large in Lepturacanthus and Tri-
chiurus. Some large specimens of Lepidopus cau-
datus and Lepidopus fitchi also bear reduced barbs
on the dentary fangs. The condition is also variable
within genera. For example, of the two species of
Eupleurogrammus, only E. glossodon has dentary
fangs (Nakamura and Parin, 1993).

A pair of small canine teeth on the premaxilla
project forward and are visible from a dorsal aspect
in Lepturacanthus. Eupleurogrammus, Tentoriceps,
and Trichiurus also have a pair of small, anteriorly
directed teeth on the premaxilla, but these are not
visible in dorsal view. Nakamura and Parin (1993)
noted the presence of two small canine teeth pro-
jecting forward on the premaxillary symphysis as a
character that differentiates Lepturacanthus and
Trichiurus. This character is dependent on the size
of the specimens, and it is quite variable within gen-
era.

All gempylids and trichiurids in this study had a

18 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 11. Anterior view of the left postorbital ossification: (A) Assurger anzac; (B) Bentbodesmus tenuis ; (C) Evoxy-
metopon taeniatus ; (D) Lepidopus caudatus ; (E) Lepturacanthus savala ; (F) Trichiurus lepturus.

cluster of up to six fangs at the tip of the premax-
illa. These teeth are not part of the longitudinal
crypt and are rounder in cross section, larger, and
stronger than those of the longitudinal series. Some
of the teeth in this cluster are depressible. The an-
terior fangs in Lepturacanthus and Trichiurus have
well-developed barbs. Some large specimens of Eu-
pleurogrammus, Lepidopus caudatus, and Nesiar-
chus may also bear reduced barbs on their premax-
illary fangs. The presence of barbs might represent
a more derived condition than that noted by Russo
(1983: character 37), where the gempylids Gem-
pylus, Nealotus, Nesiarchus, Promethichthys, Rex-
ea, and Thyrsitoides are characterized by having
fangs with an extended, flattened cutting edge on
the posterodistal surface. He considered that the
condition of simple pointed fangs found in Diplo-
spinus, Paradiplospinus, and the rest of the gem-
pylid genera represented the plesiomorphic state.
However, the presence of barbs on the premaxillary
fangs is variable within genera and species. For ex-
ample, Nakamura and Parin (1993) reported that
Trichiurus auriga Klunzinger 1884 and most spec-
imens of Eupleurogrammus lack barbs on the pre-
maxillary fangs.

Character 1 0. All of the species analyzed in this
study have serial teeth with smooth edges, except

for those in the outgroup Gempylus and the tri-
chiurid Aphanopus, which have serrate edges. The
premaxillary fangs of large specimens of Aphano-
pus and the outgroup Gempylus are also serrate.
Maul (1953) reported the presence of minute ser-
rations along the anterior margin of the premaxil-
lary canines of Benthodesmus simonyi. The speci-
men of B. simonyi analyzed in the present study
(USNM 292768) has only slight irregularities along
the margins of the fangs and serial teeth. These ir-
regularities are not comparable to the serrations of
Aphanopus or Gempylus. Russo (1983) noted that
among the gempylids only Gempylus and Thyrsites
have serrate edges on their fangs and serial teeth.
However, Gempylus has serrations in both the pos-
terior and anterior edges, whereas Thyrsites bears
serrations only on the anterior edge of the fangs
and the posterior edges of the serial teeth. Russo
(1983: character 38) considered the presence of ser-
rations on the premaxillary fangs as the apomorph-
ic condition. I agree with Russo (1983), but extend
the character to include the presence of serrations
on the anterior and posterior edges of the fangs and
the serial teeth as the apomorphic condition.

Lower Jaw

DENTARY. Posteriorly, the dentary is divided
into a ventral and a dorsal arm (process). A large

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 19

SUPRAMAXILLA

20 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 12. A-E. Lateral view of the left lower and upper jaw bones: (A) Diplospinus multistriatus; (B) Faradiplospinus antarcticus; (C) Apbanopus arigato; (D) Assurger anzac; (E)
Benthodesmus tenuis.

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny H 21

anteromedial fossa between the two arms accepts
the anterior process of the articular. The dentary is
sutured anteriorly with the tip of Meckel’s cartilage
at the mandibular symphysis.

Russo (1983) noted that the presence of a fleshy
conical process at the mandibular symphysis is al-
ways correlated with the occurrence of a similar
structure at the premaxillary symphysis, and he
considered it apomorphic among gempylids. How-
ever, there is some disagreement among authors
about the presence or absence of these conical pro-
cesses in the gempylids Diplospinus and Nealotus.
Russo (1983) reported them in Nealotus, whereas
Parin and Becker (1972) and Nakamura and Parin
(1993) did not. Russo (1983) also characterized
Diplospinus by the absence of such processes,
whereas Nakamura and Parin (1993) noted their
presence only at the premaxillary symphysis. The
confusion may be a result of the reduced condition
of these conical processes in some genera. The man-
dibular symphyses of all the trichiurids and the out-
groups analyzed in this study have an anteriorly
directed, external conical process (dermal process
of Nakamura and Parin, 1993). In the outgroups
Diplospinus and Paradiplospinus, the trichiurid
Evoxymetopon, and some species of Lepidopus,
this conical process is extremely reduced, and it is
easily overlooked, except in cleared and stained
specimens examined through transmitted light.
This conical process has been described by some
authors (Tucker, 1956; James, 1961; Parin and
Becker, 1972; Russo, 1983) as a cartilaginous pro-
jection, but it does not stain with alcian blue.

ARTICULAR. The articular forms the posterior
angle of the lower jaw and couples it to the pala-
toquadrate. Anteriorly, a large, triangular process
fits into the anteromedial fossa of the dentary. Be-
sides this anteromedial process, the articular also
bears a dorsal, a ventral, and a posterior process.
The dorsal and ventral processes are anteriorly di-
rected and shorter than the anteromedial triangular
process of this bone. Their tips abut or approach
the posterodorsal and posteroventral arms of the
dentary. The posterior process of the articular is
hook-shaped, and it bears a transverse dorsal artic-
ular fossa for articulation with the quadrate. The
ventral corner of this posterior process bears medial
and lateral depressions in which the angular fits.
The morphology of this bone is similar in all of the
species studied.

ANGULAR. The angular is a small hook-shaped
bone that fits on the posteroventral corner of the
articular. A small arm fits laterally on the postero-
ventral corner of the articular, whereas a longer
arm curves around and fits into a depression on the
medial side. In all the outgroups and the trichiurids
this medial arm extends anteriorly along the medial
face of the posteroventral corner of the articular as
an elongate, tubular process. The extent of the
elongation of the medial arm is variable among
genera. In some genera the medial arm extends dor-
sally, forming a wider, plate-like process that covers

most of the medial face of the posteroventral corner
of the articular. The shape of the medial arm of the
angular is quite variable and it is difficult to cate-
gorize into objective character states.

Upper Jaw

PREMAXILLA. All the trichiurids and the out-
groups have an external dermal process at the pre-
maxillary symphysis. As indicated earlier in the de-
scription of the dentary bone, this structure is con-
sidered together with the presence of a dermal pro-
cess at the mandibular symphysis. The premaxillary
dermal process tends to be smaller than that present
at the dentary. This dermal process in the trichiur-
ids is also smaller than that of the outgroups Dip-
lospinus and Paradiplospinus.

All scombroids are characterized by having a
nonprotrusible upper jaw, with the premaxilla
strongly attached to the maxilla and the ethmovo-
merine region of the neurocranium (Collette et al.,
1984: character 19; Johnson, 1986: character 9).

A small, ascending process extends dorsally from
the anterior tip of the premaxilla. The ascending
process abuts the rostral cartilage, which provides
a pivot point for the dorsoventral rotation of the
premaxilla (Johnson, 1986). The trichiurids are
characterized by the presence of a shortened, as-
cending process on the premaxilla. Russo (1983:
character 39) indicated that the gempylids also
share the presence of a short, ascending process of
the premaxilla.

Posterior to the ascending process, a small, dor-
sal process serves as the articular condyle for the
maxilla. The descending arm of the premaxilla ex-
tends posteriorly and medially to the maxilla. The
posterior-most margin of the premaxilla does not
extend past the descending arm of the maxilla. Rus-
so (1983: character 40) also found that all the gem-
pylids were characterized by a posterior premaxil-
lary arm that does not extend beyond the descend-
ing arm of the maxilla.

MAXILLA. Anteriorly, the maxilla has a strong,
rounded articular head that articulates with the pal-
atoquadrate and the ethmovomerine region of the
neurocranium. Ventrally, the articular head is also
divided by a notch that accepts the articular con-
dyle of the premaxilla.

Posterior to the articular head, the maxilla be-
comes narrow, forming a small dorsal depression
that accepts the maxillary process of the palatine.
The descending arm of the maxilla becomes wider
as it extends posteroventrally and completely cov-
ers the descending arm of the premaxilla. Dorsally,
the descending arm of the maxilla may expand into
a dorsal ridge that serves as an attachment point
for the adductor mandibulae. Russo (1983: char-
acter 42) indicated that all gempylids share the
presence of a well-developed dorsal ridge that arch-
es high above the dorsal margin of the maxilla. He
also noted that a more derived state was present in
the genus Gempylus, where the ridge is extremely

22 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

large and occupies about two thirds of the anterior
half of the maxilla. All the trichiurids, except
Aphanopus, Leptur acanthus, and Trichiurus, have
well-developed dorsal ridges. The presence of a
well-developed dorsal ridge on the maxilla is cor-
related with a dorsal notch formed between the
dorsal ridge and the posterior margin of the max-
illa. In those cases where the dorsal notch is pre-
sent, the posterior margin of the maxilla is dorso-
ventrally expanded. The extent of the dorsal ridge
of the maxilla is quite variable, and this character
cannot be easily categorized.

The posterodorsal corner of the maxilla accepts
the posterior half of the supramaxilla, which ex-
tends anteriorly and sometimes reaches the poste-
rior margin of the dorsal ridge. The posteroventral
comer of the maxilla is hook-shaped and expands
beyond the ventral margin of the descending arm
of the premaxilla.

SUPRAMAXILL-A. All the trichiurids, except
Eupleurogrammus, and outgroups, except Nesiar-
chus, have an extremely reduced, splint-like supra-
maxilla. Russo (1983: character 43) considered the
presence of a reduced splint-like supramaxilla as a
derived condition uniting the gempylids Diplospi-
nus, Gempyius , and Paradipiospinus. The trichiur-
id Eupleurogrammus and the outgroup Nesiarchus
are characterized by having a much wider, well-de-
veloped supramaxilla. However, it is difficult to cat-
egorize the variation in the size of the supramaxilla
into objective character states. In those trichiurids
and gempylids having a well-developed dorsal
notch on the descending arm of the maxilla ( As -
surger, Benthodesmus, Diplospinus, Eupleurogram-
mus, Evoxymetopon, Gempyius, Lepidopus, Nes-
iarchus, Paradipiospinus, and Tentoriceps ), the su-
pramaxilla originates laterally on a dorsal plate-like
expansion of the posterior margin of the maxilla.
In most cases, the supramaxilla extends anteriorly
over the dorsal notch of the maxilla and its anterior
half remains unattached. In Eupleurogrammus and
Lepidopus, the anterior margin of the supramaxilla
reaches and attaches to the dorsal ridge of the max-
illa. The central part of the body of the supramax-
illa remains free and unattached to the maxilla. The
trichiurids Aphanopus, Leptur acanthus, and Tri-
chiurus lack a well-developed dorsal notch, and the
supramaxilla simply extends along the dorsal mar-
gin of the maxilla to which it attaches.

SUSPENSORIUM

Collette and Russo (1984) separated the bones of
this series into the palatine and hyoid arches. The
palatine arch consists of the palatine, ectopterv-
goid, endopterygoid, and metapterygoid. The hyoid
arch includes the hyomandibula, symplectic, quad-
rate, and hyoid complex. I include the palatine and
hyoid arches as part of the suspensorium (Fig. 13)
but exclude the hyoid complex, which is discussed
separately.

Hyomandibula

The hyomandibula has a cruciform dorsal process
bearing three articular condyles. The anterior and
dorsal condyles articulate with the hyomandibular
fossa of the otic capsule (formed by the sphenotic,
pterotic, and prootic), whereas the posterior con-
dyle articulates with the articular fossa of the op-
ercle.

The hyomandibula bears an elongate ventral arm
with a prominent lateral ridge that is pointed at its
dorsal tip. The lateral ridge serves as the attach-
ment point to the anterior margin of the preopercle
and the posterior margin of the metapterygoid. The
angle between the anterior articular condyle and
the ventral arm extends as a small plate that is
fused medially and laterally to the dorsal comer of
the metapterygoid. The morphology of the hy-
omandibula is similar among the taxa analyzed in
this study.

Symplectic

The symplectic lies between the quadrate, metap-
terygoid, interhyal, and hyomandibula. It is tube-
like, with the ventral half fitting along the postero-
medial fossa of the quadrate. Anterodorsally, it has
a plate-like extension that abuts the posteroventral
comer of the metapterygoid medially. Posterodor-
sally, the symplectic of all trichiurids and the out-
groups bear a spine-like extension that is covered
by the posterodorsal process of the quadrate.

Quadrate

In the trichiurids, the quadrate joins the lower jaw
to the rest of the suspensorium. It is triangular and
bears a transverse mandibular condyle on the ven-
tral corner that fits on the articular fossa of the
articular bone. The anterior margin articulates with
the ventral arm of the ectopterygoid, whereas the
dorsal margin abuts the metapterygoid. Postero-
medially, it bears a depression that accepts the ven-
tral portion of the symplectic.

Character 1 1 . The posterior margin of the quad-
rate is strongly ossified and bears a posterodorsal
process. This process is elongate and extends well
past the ventral margin of the metapterygoid in all
outgroups and trichiurids, except Eupleurogram-
mus, Leptur acanthus, and Trichiurus. In Eupleu-
rogrammus, Lepturacanthus, and Trichiurus the
process is shorter and does not extend well past the
ventral margin of the metapterygoid. The strong
posterior margin serves as an attachment surface
for the antero ventral arm of the preopercle.

Metapterygoid

Anteromedially, the metapterygoid articulates with
the ectopterygoid and the endopterygoid. Medially
along the posteroventral comer, an extremely shal-
low fossa accepts the anterodorsal plate-like exten-
sion of the symplectic. The dorsal corner is usually
divided into a medial and a lateral extension that

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 23

Figure 13. A-E. Lateral view of the left suspensorium: (A) Diplospinus multistriatus; (B) Paradiplospinus antarcticus;
(C) Aphanopus arigato ; (D) Assurger anzac\ (E) Benthodesmus tenuis.

24 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 13. F-K. Lateral view of the left suspensorium: (F) Eupleurogrammus glossodon ; (G) Evoxymetopon taeniatus;
(H) Lepidopus fitchi ; (I) Lepturacanthus savala ; (J) Tentoriceps cristatus ; (K) Trichiurus lepturus.

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 25

are strongly fused to the plate-like process at the
angle between the anterior articular condyle and
the ventral arm of the hyomandibula. All trichiur-
ids and outgroups, except Diplospinus and Para-
diplospinus, have well-developed lateral and medial
processes on the metapterygoid. Russo (1983: char-
acter 45) noted that among the gempylids, only
Diplospinus, Lepidocybium, and Paradiplospinus
bear reduced lateral and medial processes on the
metapterygoid. He noted that in these genera the
dorsal margin of the metapterygoid appears to
come to a single point, giving the bone a triangular
appearance. All trichiurids, except some specimens
of Eupleurogrammus, have a lateral process that
comes to a point dorsally. However, the medial pro-
cess is well developed and flat on its dorsal margin.
The lateral shape of the metapterygoid is quite var-
iable and difficult to categorize. This potential char-
acter is excluded from the present analysis, al-
though it may prove to be useful in a future study
of the gempylids and trichiurids in which its vari-
ation can be assessed more extensively.

Ectopterygoid

The ectopterygoid is composed of three or four
arms (anterior, posteromedial, posterolateral, and
ventral) that join this bone to the endopterygoid,
palatine, quadrate, and sometimes metapterygoid.
The posterior margin of the ventral arm articulates
with the anterior margin of the quadrate. The an-
terior arm articulates with the lateral margin of the
endopterygoid and fits into the longitudinal fossa
of the palatine. Nesiarchus bears both a postero-
lateral and a posteromedial arm. These posterior
arms articulate with the lateral margin of the en-
dopterygoid and the anterodorsal corner of the
quadrate. The posteromedial arm of Nesiarchus is
much longer than the posterolateral arm, but it nev-
er reaches the anteroventral corner of the metap-
terygoid. In Diplospinus, Gempylus, Paradiplospi-
nus, and all trichiurids only the posteromedial arm
is present.

Character 12. In the outgroups, the posterome-
dial arm of the ectopterygoid articulates only with
the lateral margin of the endopterygoid and the an-
terodorsal corner of the quadrate. All trichiurids
are characterized by having a much longer postero-
medial arm that extends up to and articulates with
the anteroventral corner of the metapterygoid.

Endopterygoid

The lateral margin of the endopterygoid articulates
anteriorly with the ectopterygoid and palatine and
posteriorly with the metapterygoid. The endopter-
ygoid of Lepturacanthus has a tuberculous patch
anteriorly that bears a few small teeth. Although
the anterior portion of the endopterygoid is better
ossified than the posteromedial shelf, none of the
outgroups or the rest of the trichiurids bear any
teeth on this bone.

Palatine

The palatine attaches to the maxilla, the ethmoid,
and the lateral ethmoid. Posteriorly, it bears a lon-
gitudinal fossa that serves as the surface for artic-
ulation with the anterior arm of the ectopterygoid
and the lateral margin of the endopterygoid. An-
teriorly, a hooked maxillary process fits above the
dorsal depression that is present posterior to the
articular head of the maxilla.

Character 13. A small, medially directed shelf
that serves as an articulation point with the eth-
moid is evident in the trichiurids and outgroups.
Posterior to this small medial shelf, Lepturacanthus
and Trichiurus also bear a well-developed, medially
directed condyle at the dorsal corner between the
maxillary process and the main body of the pala-
tine. This condyle abuts the posterior margin of the
palatal process of the ethmoid and is visible in lat-
eral view.

Character 14. All of the outgroups and most of
the trichiurids are characterized by having only a
few teeth arranged uniserially and not covering the
whole length of the ventral margin of the palatine.
A few replacement teeth are usually present on the
medial face of the palatine above and between the
teeth in the main ventral series. Of the trichiurids,
only Eupleurogrammus, Lepturacanthus, and Tri-
chiurus bear large numbers of teeth arranged in
several rows and covering most of the length of the
ventral margin of the palatine. In Lepturacanthus
the condition is further modified and part of the
medial side of the palatine is covered by a patch of
small teeth. Maul (1953) noted that the palatine
teeth in Benthodesmus simonyi are covered by a
fleshy fold and that, in preserved specimens, the
fold closes up so tightly that the presence of pala-
tine teeth can be easily overlooked. The case is sim-
ilar for the preserved specimens of the other tri-
chiurid genera, which may explain previous reports
of the absence of palatine teeth in some taxa. Tuck-
er (1956) noted the difficulty in finding the palatine
teeth in trichiurids and suggested that reports on
their absence should not be taken too seriously.

HYOID COMPLEX

The hyoid complex is composed of a series of bones
that articulate with the suspensorium and support
the branchiostegal rays (Fig. 14).

Interhyal

The interhyal connects the hyomandibula to the
epihyal. It is a small, cylindrical bone that is usually
constricted in the middle and expanded at the tips.
The morphology of this bone is similar among the
taxa analyzed in this study.

Epihyal

The anterior margin of the epihyal is broad and
articulates with the posterior margin of the cera-
tohyal. This anterior articulation includes blocks of

26 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

cartilage at the dorsal and ventral corners and su-
turing by way of odontoid processes in the middle.
The posterior end of the epihyal narrows into an
apex that bears a dorsal articular fossa for articu-
lation with the interhyal. In the trichiurids, the
three posterior-most branchiostegal rays attach lat-
erally to the ventral margin of this bone. The mor-
phology of the epihyal is similar among the taxa
analyzed in this study.

Ceratohyal

The ceratohyal is a flattened bone that articulates
posteriorly with the epihyal via blocks of cartilage
and suturing by way of odontoid processes. The
three anterior-most branchiostegals articulate me-
dially on the ventral margin of this bone, whereas
the fourth branchiostegal articulates laterally at the
postero ventral corner. The hyoidean groove (Col-
lette and Russo, 1984) runs longitudinally along
the lateral face of the ceratohyal. A small slit on
the hyoidean groove (the beryciform foramen of
McAllister, 1968; or ceratohyal window of Collette
and Chao, 1975) is present in Eupleurogrammus,
Evoxymetopon, Leptur acanthus, Tentoriceps, and
the outgroups Diplospinus and Nesiarchus. This in-
terpretation differs from the observations of Russo
(1983), who indicated that Diplospinus and Nes-
iarchus lack a ceratohyal window. Collette and
Russo (1984) noted that the presence of a cerato-
hyal window within the Spanish mackerels ( Scorn -
beromorus : Scombridae) is quite variable.

Anteriorly, the ceratohyal articulates with the
dorsal and ventral hypohyals by way of a layer of
cartilage; the anteroventral corner of the ceratohyal
projects anteriorly to articulate with the postero-
ventral notch of the ventral hypohyal.

Character IS. In Eupleurogrammus, Leptur acan-
thus, Tentoriceps, and Trichiurus, the anterodorsal
corner of the ceratohyal is pointed and extends an-
teriorly, abutting the dorsal margin of a layer of
cartilage on the posterior margin of the dorsal hy-
pohyal. All other trichiurids and outgroups lack
this anterodorsal extension of the ceratohyal. In
Aphanopus and some specimens of Lepidopus fit-
chi, the anterodorsal corner extends only slightly
anteriorly, but it is not pointed and does not abut
the dorsal margin of the cartilage of the dorsal hy-
pohyal.

Dorsal Hypohyal

Anteriorly, the dorsal hypohyal articulates with the
ventral hypohyal by suturing with odontoid pro-
cesses, whereas posteriorly it articulates via a block
of cartilage. The posterior margin articulates with
the anterior margin of the ceratohyal by way of a
layer of cartilage. The anterodorsal corner bears a
medial projection that forms a symphysis with the
opposing dorsal hypohyal, the anterior tip of the
first basibranchial, and the posterior margin of the
glossohyal.

The dorsal hypohyal of the trichiurid Leptura-

canthus is unique in that it bears teeth along the
anterior half of its dorsal margin. These teeth are
not part of a dermal plate or patch, and they are
fused to the bony element. Eupleurogrammus bears
longitudinal patches of teeth on the dorsal margin
of the dorsal hypohyal. However, these tooth
patches are part of the epithelium covering the bone
and they are not fused to it.

Ventral Hypohyal

The ventral hypohyal is joined dorsally to the dor-
sal hypohyal. The posteroventral corner forms a
longitudinal notch where the elongate anteroven-
tral corner of the ceratohyal fits. The anteroventral
corner bears a ventral projection that serves as the
site for attachment of the ligaments coming from
the articular head of the urohyal. The morphology
of the ventral hypohyals is similar among the taxa
analyzed in this study.

Glossohyal

The glossohyal is a median bone supporting the
tongue. It is covered with flesh, and in the trichiurid
Eupleurogrammus, it is unique in that it bears two
elongate, dermal tooth patches dorsally. Evoxyme-
topon bears a few minute teeth on the lateral mar-
gins of the tongue. In both Eupleurogrammus and
Evoxymetopon, the tooth patches are part of the
flesh covering the glossohyal; they are not fused to
the bone. All other trichiurids and outgroups ana-
lyzed in this study lack glossohyal teeth. Posteriorly,
the glossohyal articulates with the anterodorsal cor-
ner of the dorsal hypohyals and the anterior margin
of the first basibranchial. The dorsal face of this
bone is flat or slightly concave in all the taxa ana-
lyzed in this study. Posteroventrally, the trichiurids
and the outgroups, except Gempylus, have two lat-
eral processes that converge anteriorly by way of a
longitudinal keel. Gempylus bears these postero-
ventral processes but lacks a longitudinal keel.

Character 16. Russo (1983: character 5 1 ) consid-
ered the presence of quadrilateral posteroventral
processes on the glossohyal as a derived condition
among gempylids. I agree with Russo (1983) and
consider the condition observed in Nesiarchus (and
most of the gempylids including Lepidocybium ), in
which the posteroventral processes do not appear
quadrilateral, as the plesiomorphic state. He sepa-
rated the condition observed in Gempylus as a dif-
ferent character. As indicated by Russo (1983: char-
acter 50), the posteroventral processes of the glos-
sohyal in Gempylus are wing-like and have the dis-
tal ends pointing posteriorly. In this study, the
conditions regarding the shape of the posteroven-
tral processes are combined into a single multistate
character. There are three derived conditions re-
garding the shape of the posteroventral processes
relative to the state found in Nesiarchus and most
of the gempylids: triangular, quadrilateral, and
wing-like with the distal ends pointing posteriorly.
Diplospinus, Paradiplospinus, and all trichiurids,

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 27

Figure 14. A-E. Lateral view of the left hyoid complex; the glossohyal is represented from left to right in ventral and
dorsal views, respectively: (A) Diplospinus multistriatus; (B) Paradiplospinus antarcticus; (C) Aphanopus arigato ; (D)
Assurger anzac; (E) Benthodesmus tenuis.

except Eupleurogr ammus, Leptur acanthus, Tento-
riceps, and Trichiurus, have ventral processes that
appear quadrilateral in ventral view. The trichiurids
Aphanopus and Benthodesmus bear posteroventral
processes that appear quadrilateral, although not as
pronounced as those in Diplospinus and Paradip-
lospinus. In some large specimens of Aphanopus
and Benthodesmus, the lateral corners of the pos-

teroventral processes are rounded. The glossohyal
of Eupleurogr ammus, Leptur acanthus, Tentori-
ceps, and Trichiurus bears triangular posteroventral
processes. Furthermore, the posteroventral process-
es in Gempylus and Nesiarchus are located more
anteriorly, leaving a well-developed posterior artic-
ular head in the glossohyal. In all trichiurids and
the outgroups Diplospinus and Paradiplospinus,

28 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 14. F-K. Lateral view of the left hyoid complex; the glossohyal is represented from left to right in ventral and
dorsal views, respectively: (F) Eupleurogrammus glossodon; (G) Evoxymetopon taeniatus; (H) Lepidopus fitchi; (I) Lep-
turacanthus savala ; (J) Tentoriceps cristatus; (K) Trichiurus lepturus.

the posteroventral processes are located farther
posteriorly and the glossohyal bears a more re-
duced posterior articular head.

Urohyal

The urohyal is a median bone with an anterior end
that is a thin rod bearing a small articular head.
This articular head is generally forked and is con-
nected to the ventral hypohyals by two strong lat-

eral ligaments. Posterior to the articular head, the
outgroup Paradiplospinus antarcticus bears a small
posteriorly directed dorsal process (the basibran-
chial attachment of Kusaka, 1974). This interpre-
tation differs from Russo’s (1983: character 54) ob-
servations, which indicated that such a dorsal pro-
cess is absent in all gempylids, except Epinnula,
Lepidocybium, Thyrsitoides, and Thyrsitops. All
trichiurids and the rest of the outgroups in this
study lack this dorsal process on the urohyal. Pos-

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 29

PHARYNGOBRANCHIALS p ^UPPER^

PHARYNGOBRANCHIALS TOOTH PLATE

30 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

teriorly, the urohyal is poorly ossified and laterally
compressed into a plate-like process that is attached
to the cleithrum and coracoid by the sternohyoi-
deus.

Branchiostegal Rays

Seven acinaciform branchiostegal rays (McAllister,
1968) are associated with each side of the hyoid
complex. The articular heads are spatulate and ex-
panded to a greater degree in the posterior-most
three rays. The three anterior rays are the shortest,
and they articulate medially with the ventral mar-
gin of the ceratohyal. The fourth branchiostegal ray
articulates laterally at the posteroventral comer of
the ceratohyal, whereas the three posterior rays ar-
ticulate laterally with the ventral margin of the epi-
hyal. This distribution of the branchiostegal rays
was interpreted by Russo (1983: character 53) as a
synapomorphy uniting all gempylids. Johnson
(1986: character 10) concluded that the articulation
of the fifth branchiostegal on the anteroventral cor-
ner of the epihyal represents a synapomorphy for
the Scombroidei.

BRANCHIAL COMPLEX

Only the branchial complex of Trichiurus lepturus
is drawn (Fig. 15) because most of the characters
differ only slightly in magnitude and can be easily
identified on Trichiurus. The first and second basi-
branchials are also drawn separately for all the gen-
era analyzed, except Gempylus and Nesiarchus
(Fig. 16). Collette et al. (1984: character 1) and
Johnson (1986: character 6) indicated that in the
scombroids the cartilaginous anterior tip of the sec-
ond epibranchial articulates with the second phar-
yngobranchial. A medial cartilaginous process of
the second epibranchial extends well beyond the
lateral margin of the third pharyngobranchial and
articulates with a small cartilaginous condyle at the
anterior tip of a longitudinal column that runs
along the dorsal surface of the third pharyngobran-
chial.

Scombroids are also characterized by the absence
of the fourth pharyngobranchial cartilage (Johnson,
1986: character 7). The fourth epibranchial artic-
ulates with a more extensive posterior cartilage of
the third pharyngobranchial. The extensive poste-
rior cartilaginous tip of the third pharyngobranchi-
al fits into the dorsal surface of the fourth pharyn-
geal tooth plate.

In addition, scombroids are characterized by the
presence of fourth pharyngeal tooth plates and by
having extremely elongate third pharyngobranchi-
als. The third pharyngobranchial bears a reduced

lateral shelf that has a straight medial margin
(Johnson, 1986: character 8).

Lower Branchial Apparatus

BASIBRANCHIALS. The basibranchials are me-
dian bones arranged in a longitudinal series. The
first basibranchial articulates anteriorly with the
posterior margin of the glossohyal and lies between
the left and right dorsal hypohyals. The second
serves as the point of articulation for the first hy-
pobranchial. In all outgroups and trichiurids, the
third basibranchial has a tubular central axis with
lateral shelves that cover the medial margins of the
third hypobranchial.

Character 17. All of the trichiurids are charac-
terized by the presence of a well-developed, knob-
like, anterior articular head on the first basibran-
chial. In addition, in all trichiurids, except Aphan-
opus, Assurger, Benthodesmus, Evoxymetopon,
and Lepidopus , the articular head has small dor-
solateral processes or wings. All outgroups have a
first basibranchial with a broad base that gradually
tapers to a point anteriorly.

Character 18. The second basibranchial in the
outgroups has an expanded posterior margin and
elongate anterior end. Russo (1983: character 62)
considered this condition as a synapomorphy unit-
ing the gempylids Diplospinus, Gempylus, Nesiar-
chus, and Paradiplospinus. In the trichiurids, the
second basibranchial bears two small, laterally
pointed processes posterior to the place of articu-
lation with the heads of the first hypobranchial.

HYPOBRANCHIALS. The hypobranchials are
three pairs of bones that form part of the lower
arms of the gill arches; their dorsal heads articulate
with the ceratobranchials. The ventral heads of the
first and second hypobranchials articulate with the
cartilaginous junction between the first and second
and second and third basibranchials, respectively.
The third hypobranchial is triangular and smaller
than the first and second hypobranchials, and it has
an elongate anterior process that extends iu,.d 'r the
lateral shelves of the third basibranchial.

The first hypobranchial has a medially curved ar-
ticular head that may bear a lateral or a medial
process, or both. Russo (1983: character 63) noted
the absence of an anterior (lateral) process in the
first hypobranchial of all gempylids, except Diplo-
spinus, Gempylus, Lepidocybium, and Tongaich-
thys. He considered the presence of an anterolateral
process to be plesiomorphic. All trichiurids, except
Aphanopus and Benthodesmus, and outgroups, ex-
cept Nesiarchus, have a medial and a lateral pro-
cess on the anterior articular head of the first hy-
pobranchial. The cleared and stained specimens of

Figure 15. Dorsal view of the lower branchial apparatus of Trichiurus lepturus (top), the left side does not include the
gill rakers or tooth patches; cartilaginous articulations striped. Right upper branchial apparatus of Trichiurus lepturus:
ventral view (center); dorsal view showing the cartilaginous articulations (bottom).

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 31

Figure 16. Dorsal views of the first basibranchial (top) and second basibranchial (bottom) of: (A) Diplospinus multis-
triatus; (B) Paradiplospinus antarcticus; (C) Aphanopus arigato; (D) Assurger anzac; (E) Benthodesmus tenuis ; (F) Eu-
pleurogrammus glossodon; (G) Evoxymetopon taeniatus; (H) Lepidopus fitchi ; (I) Lepturacanthus savala ; (J) Tentoriceps
cristatus ; (K) Trichiurus lepturus.

Benthodesmus tenuis and Nesiarchus analyzed in
this study lacked both lateral and medial processes
on the first hypobranchial. The specimens of
Aphanopus and B. simonyi, however, had only
well-developed medial processes. The medial and
lateral processes are well developed in the rest of
the trichiurids analyzed, where they clearly extend
far from the lateral and medial margins of the first

hypobranchial. Diplospinus and Paradiplospinus
also bear well-developed medial processes on the
first hypobranchial, but their lateral processes are
reduced and appear as rounded outgrowths that
may extend only slightly past the margin of the first
hypobranchial. This condition differs from the ob-
servations of Russo (1983) who noted that, where-
as Diplospinus bears a lateral process on the first

32 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

hypobranchial, Paradiplospinus lacks such a struc-
ture.

The second hypobranchial has a slightly curved
articular head and bears no distinct lateral or me-
dial processes in the outgroups, except Gempylus,
which seems to bear a reduced lateral process on
the second hypobranchial. All trichiurids are char-
acterized by the presence of distinct lateral and me-
dial processes in the second hypobranchial. These
processes are usually pointed and well defined, ex-
cept in Aphanopus and Bentbodesmus where they
may be reduced. The specimens of Bentbodesmus
tenuis have reduced medial and lateral processes,
whereas those of B. simonyi have well-developed
medial processes. The presence or absence of lateral
and medial processes in the first and second hypo-
branchials is quite variable among and within gen-
era, and its categorization into objective character
states is difficult.

The third hypobranchial is triangular, with the
tubular anterior end curving medially under the lat-
eral wings of the third basibranchial. In all trichiur-
ids and outgroups, except Apbanopus, Bentbodes-
mus, Gempylus, and Nesiarcbus, the posterolateral
corner of this bone usually bears pointed projec-
tions. This condition varies between specimens of
the same species, and it is dependent on the size of
the specimens available and the degree of ossifica-
tion of this bone in any given specimen. In some
species, the posteromedial corners of this paired
bone cover the posterior end of the third basibran-
chial.

CERATOBRANCHIALS. The ceratobranchials
are pairs of bones that form the lower arms of the
gill arches. These are the longest bones in the gill
arches, and they support most of the gill filaments
and rakers. The anteroventral heads of the first, sec-
ond, and third ceratobranchials articulate with the
first, second, and third hypobranchials, respectively.
The anteroventral head of the fourth ceratobran-
chial articulates with the third basibranchial. The
posterior tip of the fifth ceratobranchial has an
elongate cartilaginous cap that lies within the epi-
thelium of the branchial cavity. The anterior carti-
laginous tips of the paired elements of the fifth cer-
atobranchial may be fused at their tips and articu-
late with the complex of the fourth ceratobranchial
and third basibranchial by ligaments. The fifth cer-
atobranchial bears dorsal tooth plates. The mor-
phology of the fifth ceratobranchial is similar
among the taxa analyzed in this study.

Character 19. The outgroups and trichiurids
have straight first, second, and third ceratobran-
chials. Russo (1983: character 64) considered the
sigmoid shape of the fourth ceratobranchial to be
a synapomorphy uniting the gempylids Diplospinus
and Paradiplospinus. In the trichiurids and the rest
of the outgroups and gempylids, the anteroventral
head of the fourth ceratobranchial is twisted me-
dially, but the bone does not appear sigmoidal. Al-
though this character appears equivocal at the out-

group node, I agree with Russo (1983) and consider
the presence of a sigmoidal fourth ceratobranchial
as a synapomorphy uniting Diplospinus and Para-
diplospinus.

Upper Branchial Apparatus

EPIBRANCHIALS. The epibranchials are four
pairs of bones that form part of the upper arms of
the gill arches. The posteroventral ends of the epi-
branchials articulate with the posterodorsal ends of
the ceratobranchials. The anterodorsal head of the
first epibranchial articulates with the first pharyn-
gobranchial and bears an uncinate process that ar-
ticulates with the second pharyngobranchial by
way of an interarcual cartilage. The anterodorsal
head of the second epibranchial articulates not only
with the second pharyngobranchial but also with
the third pharyngobranchial via an elongate medial
cartilaginous process. This medially elongate carti-
laginous process joins a small articular condyle on
a longitudinal ridge of the dorsal face of the third
pharyngobranchial. Posterior to this cartilaginous
head, all gempylids and trichiurids, except Leptur-
acantbus and Tricbiurus, have a truncated dorso-
medial process that bears a ligamentous attachment
to the third epibranchial. Leptur acanthus and some
specimens of Tricbiurus are characterized by having
a second epibranchial in which the dorsomedial
process is pointed. The third epibranchial also
bears a dorsomedial process that connects it to the
fourth epibranchial and a cartilaginous knob at the
anterior tip that attaches it to the third pharyngo-
branchial. In Diplospinus and Paradiplospinus, a
shelf-like dorsomedial plate extends longitudinally
between the anterodorsal process and the posterior
end of the third epibranchial. The rest of the out-
groups and the trichiurids lack such a modification
or have a partial shelf that does not extend along
the entire posterior half of the third epibranchial.
However, the extent of this shelf is variable and
difficult to categorize into well-defined character
states. A dorsolateral process on the fourth epi-
branchial articulates with the third epibranchial.

PHARYNGOBRANCHIALS. The pharyngo-
branchials are three pairs of small bones that are
attached to the anterodorsal heads of the epibran-
chials. The first pharyngobranchial articulates dor-
sally with the prootic. In all trichiurids and out-
groups, the first pharyngobranchial is a small, tu-
bular, edentulous bone. In Eupleurogrammus, Lep-
turacanthus, and Tricbiurus, the first pharyngo-
branchial is strongly curved, whereas in the rest of
the trichiurids and the outgroups, the bone is
straight or slightly curved. The degree of curvature
on the first pharyngobranchial is highly variable
and difficult to categorize.

The second and third pharyngobranchials bear
well-developed tooth plates ventrally. The third
pharyngobranchial is the largest, and it bears a
small dorsolateral cartilaginous knob that articu-
lates with the cartilaginous tip of the second epi-

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 33

branchial. The posterior margin of the third phar-
yngobranchial bears a large cartilaginous cap that
articulates with the third and fourth epibranchials.

GILL RAKERS. Gill rakers tend to be more re-
duced and poorly ossified in the outgroups Diplo-
spinus, Nesiarchus, and Paradiplospinus. The bet-
ter ossified gill rakers of the trichiurids can be spi-
nous, tuberculous, or a combination of both, with
the spinous ones usually at the posterior end of the
hypobranchials. Trichiurids have two rows (lateral
and medial) of gill rakers on the first and second
hypobranchials, except Benthodesmus, which bears
a single row on the first hypobranchial. All out-
groups have a single row of gill rakers on the first
hypobranchial and two rows on the second hypo-
branchial, except Gempylus, which has single rows
on the first and second hypobranchials. In all tri-
chiurids and outgroups, except Nesiarchus, the lat-
eral rows of gill rakers on the first and second hy-
pobranchials extend anteriorly past the articular
head of the bone as part of the epithelial covering
of the branchial arches. The third hypobranchials
bear small tuberculous gill rakers, which are ex-
tremely reduced in Diplospinus and Paradiplospi-
nus.

The outgroups and trichiurids have first, second,
and third ceratobranchials bearing two longitudinal
(lateral and medial) series of gill rakers on the an-
terior margins. The lateral series of gill rakers are
spinescent, with spine length increasing toward the
posterodorsal end where these bones articulate with
the epibranchials. The medial series of gill rakers
tend to be smaller and more tuberculous in nature.
The fourth ceratobranchial may bear one or two
rows of gill rakers. The medial row is present only
in the largest specimens, and it is extremely reduced
and nonspinous.

All the epibranchials in most of the taxa analyzed
have either one or two series of gill rakers or tooth
plates along their ventral margin. In Diplospinus,
Nesiarchus, and Paradiplospinus, the fourth epi-
branchial is edentulous. The size and number of gill
rakers and tooth plates on the other epibranchial
bones are reduced in the outgroups Diplospinus
and Paradiplospinus. The trichiurids and the rest of
the outgroups have epibranchials bearing gill rakers
or tooth plates with stronger spination.

All trichiurids and Gempylus have two ventral
series of gill rakers on the third epibranchial. In
Diplospinus, Nesiarchus, and Paradiplospinus,
there is only a single series of reduced gill rakers on
this bone. Russo (1983: character 65) indicated
that all gempylids have two rows of tooth plates
on the third epibranchial, except Epinnula, Para-
diplospinus, and Thyrsitops, which bear one row,
and Ruvettus and Thy r sites, which bear no rows.
The observation of a single row of gill rakers on
the third epibranchial of Diplospinus differs from
the condition noted by Russo (1983). However, the
presence or absence of gill rakers on these bones is
difficult to evaluate since the gill rakers are ex-

tremely reduced and could easi±y be lost during
handling and dissection.

Matsubara and Iwai (1958) noted that Gempylus
differs from other gempylid genera in that the gill
raker at the angle of the first gill arch is small, tri-
angular, and exposed at its tip. The rest of the gem-
pylids have a gill raker at the angle of the arch that
is T-shaped, larger, and more exposed. In the dia-
grams of Nakamura and Parin (1993: figs. 26, 27),
it is evident that some trichiurids share, with Gem-
pylus and Paradiplospinus, the presence of a small
gill raker at the angle of the first gill arch. The gill
raker at the angle between the first ceratobranchial
and the first epibranchial in Aphanopus, Assurger,
and Benthodesmus has a longer spine with an ex-
panded tip. Tucker (1956) described the morphol-
ogy of the gill rakers in Benthodesmus tenuis, not-
ing that some of the gill rakers toward the angle of
the gill arches have one large barbed spine. How-
ever, the size and shape of the spines on the gill
rakers is quite variable between specimens. The gill
raker at the angle between the first epibranchial
and first ceratobranchial in Diplospinus and Para-
diplospinus is characterized by having a base that
bears a small root-like process extending toward
the articulation of the ceratobranchial and epibran-
chial bones. In Nesiarchus, the tips of the base in
this gill raker bend toward the first epibranchial-
first ceratobranchial junction, giving it the appear-
ance of a tri-rooted element. Gempylus is charac-
terized by having an angular gill raker with a tri-
angular base that lacks these root-like processes.
The angular gill raker in the trichiurids has a cir-
cular base bearing one larger spine surrounded by
smaller spinules, and it lacks the root-like process
observed in Diplospinus and Paradiplospinus. Al-
though the difference in morphology of the angular
gill raker between the trichiurids and the outgroups
is obvious, the condition at the outgroup node ap-
pears equivocal, and the character is uninformative
within the trichiurids. Thus, the morphological fea-
tures of the angular gill rakers are excluded from
this analysis, but they remain as a potentially in-
formative character, or characters, for a future
study in which both the trichiurids and gempylids
are analyzed together.

NEUROCRANIUM

In dorsal view, the neurocranium of the trichiurids
and outgroups is triangular, being narrow anteri-
orly and wider posteriorly (Fig. 17). The skulls of
most gempylids and all trichiurids are mainly char-
acterized by the elongation of some of their bones.
Posterodorsally, the neurocranium bears three
prominent ridges. Medially, a single supraoccipital
ridge is formed by the confluence of two frontal
ridges onto the supraoccipital. Lateral to the supra-
occipital ridge, a pair of epiotic ridges extends
through the parietal and the posterior edge of the
frontal. Lateral to the epiotic ridge, a pair of pter-

34 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 35

36 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 17. C-E. Lateral and dorsal views (top and bottom, respectively) of the neurocranium: (C) Aphanopus arigato; (D) Assurger anzac ; (E) Bentbodesmus tenuis .

I

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 37

Figure 17. F H. Lateral and dorsal views (top and bottom, respectively) of the neurocranium: (F) Eupleurogrammus glossodon; (G) Evoxymetopon taeniatus; (H) Lepidopus fitchi.

38 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 17. I-K. Lateral and dorsal views (top and bottom, respectively) of the neurocranium: (I) Lepturacanthus savala ; (J) Tentoriceps cristatus; (K) Trichiurus lepturus.

otic ridges extends through the pterotic and the
frontal.

Ethmoidal Region

NASAL. The nasal appears as a simple tube in
the outgroup Nesiarchus and the trichiurids Ben-
thodesmus and Lepidopus. Ossified, lateral laminar
extensions are present on the nasal and cover the
lateral palatal processes of the ethmoid in the out-
groups Gempylus and Paradiplospinus and the tri-
chiurids Assurger and Eupleurogr ammus. The rest
of the trichiurids and outgroups bear laminar ex-
tensions that do not cover the palatal processes of
the ethmoid. Eupleurogr animus, Evoxymetopon,
Gempylus, and Paradiplospinus are characterized
by lateral laminar extensions that extend along the
whole length of the nasal. In addition, Eupleuro-
grammus bears smaller medial extensions anteri-
orly. The lateral extensions are restricted to the pos-
terior half of the nasal in the outgroup Diplospinus
and the trichiurids Leptur acanthus, Tentoriceps,
and Trichiurus. This potential character is difficult
to evaluate because the detection of such lateral ex-
tensions depends on the degree of ossification of the
lateral membrane on the nasal. For example,
whereas some specimens of Diplospinus, Lepido-
pus, and Paradiplospinus seem to show the pres-
ence of well-developed, ossified lateral extensions,
others are characterized by their reduction or ab-
sence.

Character 20. The nasal is straight and runs par-
allel to the ethmoid and frontal in the outgroups.
The anterior head of the nasal is curved laterally in
all the trichiurids. The anterior head of the nasal in
one of the specimens of Nesiarchus appeared to be
slightly curved, but the condition is not comparable
to that of the trichiurids, in which the anterolateral
margin of the nasal appears concave in dorsal view
and clearly extends past the lateral margin of the
ethmoid.

ETHMOID. Anterodorsally, the ethmoid sup-
ports the nasal and bears two lateral processes, that
are attached to the palatine (the palatal processes
of Russo, 1983). Anteroventrally, the ethmoid ar-
ticulates with the dorsal edge of the vomer. Poste-
riorly, it abuts the lateral ethmoid and articulates
with the anterior margin of the frontal. The eth-
moid is elongate in all trichiurids and the out-
groups.

Character 21. In Assurger, Evoxymetopon, Lep-
idopus altifrons, and Tentoriceps, the ethmoid
bears two dorsally expanded ridges that extend
well above the dorsal margin of the nasal in lateral
view. The dorsal ridges on the ethmoid of Evoxy-
metopon and Tentoriceps are more elevated and
appear triangular in lateral view. Aphanopus, Eu-
pleurogr ammus, Leptur acanthus, and Trichiurus
are characterized by the presence of two longitu-
dinal ridges on the dorsal face of the ethmoid.
However, these ridges do not extend well above the
nasal in lateral view. Russo (1983: character 2) not-

ed that the presence of a medial ridge on the dorsal
surface of the ethmoid represented a synapomorphy
uniting the gempylid genera Diplospinus and Par-
adiplospinus. This ridge is reduced and does not
extend well above the nasal. The other outgroups
in this study, plus Benthodesmus, Lepidopus cau-
datus, and L. fit chi, lack a dorsal ridge on the eth-
moid.

LATERAL ETHMOID. Anteriorly, the lateral
ethmoid joins the vomer and the ethmoid. It at-
taches to the frontal dorsally and the parasphenoid
ventrally. The posterior margin of the lateral eth-
moid forms a lateral process that is divided into a
ventral and a dorsal articular head. The ventral ar-
ticular head attaches to the palatine, whereas the
dorsal articular head joins the dorsal articular pro-
cess of the lachrymal.

The articulations of the lachrymal and palatine
with the ethmoid are by way of cartilaginous knobs
that may or may not be separated by a bony ridge.
Russo (1983: character 8) considered the absence
of a bony ridge as the apomorphic condition
among some of the gempylids in his study. All of
the outgroups and the trichiurids share this derived
condition.

VOMER. The vomer articulates anterodorsally
with the ethmoid and posteriorly with the lateral
ethmoid and the parasphenoid. The trichiurids and
all outgroups have a vomer with an expanded an-
teroventral head that lacks teeth. In all trichiurids,
except Assurger and Eupleurogr ammus, the ante-
rior margin of the head of the vomer appears flat
in ventral view. Eupleurogr ammus and Gempylus
are similar in that the anterior margin of the ventral
head bears an anteriorly elongate process. Assurger
and Paradiplospinus tend to have a rounded ante-
rior margin on the head of the vomer, whereas Nes-
iarchus and some specimens of Diplospinus bear a
flat anterior margin. However, the condition seems
variable within species. For example, Russo (1983:
character 10) noted that the anterior margin of the
vomer in the gempylids Diplospinus, Gempylus,
Neoepinnula, Nesiarchus, Paradiplospinus, Ruvet-
tus, Thyrsites, and Thyrsitops is rounded in ventral
view. In this study, however, one specimen of Dip-
lospinus multistriatus has a rounded head on the
vomer; another has a flat anterior margin.

Anteriorly, the vomer has a process that extends
anterodorsally and attaches to the anterior tip of
the ethmoid. All the trichiurids and the outgroups,
except Aphanopus, have a vomer in which the an-
terior tip of this process terminates below or slight-
ly anterior to the ethmoid. In Aphanopus, this pro-
cess is extremely elongate and extends slightly past
the anterior margin of the rostral cartilage.

Orbital Region

FRONTAL. Anteriorly, the frontal attaches to
the nasal, ethmoid, and lateral ethmoid. Posteriorly,
it articulates with the pterosphenoid, sphenotic, pa-
rietal, and supraoccipital. Medially, the frontals are

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 39

joined to each other. A pineal foramen appears as
a gap in this medial junction and is located anterior
to the supraoccipital. The pineal foramen is not ev-
ident in some large specimens and those trichiurids
with a well-developed frontal crest. Russo (1983)
noted that there are variations in the size of the
pineal foramen among gempylids and indicated
that in some genera, such as his outgroup Poma-
tomus Lacepede 1802, the pineal foramen is not
present because the supraoccipital crest is carried
forward as a ridge onto the frontal. Ventrally, there
are short, poorly developed sheets of bone (the or-
bital lamellae of Collette and Russo, 1984) that
abut the lateral ethmoid anteriorly and the pteros-
phenoid posteriorly. Anterolaterally, the frontal
carries the supraorbital canal of the laterosensory
canal system, and posterolaterally it carries the an-
terior half of the otic canal (Coombs et al., 1987).
Johnson (1986: character 3') noted that, in the
scombroids, the frontosphenotic shelf is horizontal
and has a sharp edge. The supraorbital canal is sep-
arated from the dorsolateral margin of the orbit by
a large fossa that houses the dilatator operculi. The
infraorbital canal passes over the sharp edge of the
frontosphenotic shelf and joins the supraorbital ca-
nal medially.

Character 22. Johnson (1986: character 29) not-
ed that in the trichiurids the supra- and infraorbital
canals are joined by way of a bony tube that ex-
tends laterally from the supraorbital canal. This
condition is present in all the trichiurids studied.
The plesiomorphic condition observed in all other
scombroids is characterized by the presence of an
elongate dilatator operculi fossa that separates the
junction of the infra- and supraorbital canals,
which communicate through a dorsally or laterally
directed pore on the frontal ridge (Johnson, 1986).

Character 23. Two frontal ridges become conflu-
ent posteriorly, forming the supraoccipital ridge. In
Assurger, Evoxymetopon, and Tentoriceps, the
frontal ridges become confluent well anterior to the
supraoccipital (on the ethmoidal region) and form
an extremely well-developed frontal crest that ap-
pears as laterally flat sheets. Lepidopus altifrons is
unique among the species of this genus in having a
well-developed frontal that is similar to that of As-
surger, Evoxymetopon, and Tentoriceps. Tucker
(1957: 426), in his description of a specimen of
Evoxymetopon taeniatus (=L. altifrons ), noted
that the ridge-like elevation of the ethmofrontal re-
gion “is not so much an osseous elevation but an
outgrowth of soft tissue, normally increasing with
age as in numerous other percomorph fishes.” The
frontal crest present in the ethmofrontal region of
Assurger, Evoxymetopon, Lepidopus altifrons, and
Tentoriceps is flexible, but it represents an ossified
extension of the frontal ridges. In Eupleurogram-
mus, the frontal ridges become confluent anterior
to the supraoccipital and form a frontal crest that
is elevated above the interorbital space. These ridg-
es in Eupleurogrammus are laterally convex. In the
rest of the trichiurids and the outgroups, the frontal

ridges are not elevated as a frontal crest, and they
do not become confluent until they reach or are
close to the supraoccipital.

The frontal crest in Assurger, Evoxymetopon,
Lepidopus altifrons, and Tentoriceps extends above
the orbits making the interorbital space convex.
Lepidopus dubius Parin and Mikhailin 1981 is the
only species within the genus that is similar to L.
altifrons in the morphology of the frontal region.
However, in L. dubius, the frontal ridges are not as
elevated as in L. altifrons, do not extend far onto
the ethmoidal region, and become confluent closer
to the posterior margin of the orbits. The rest of
the trichiurids and the outgroups have an interor-
bital space that is concave or flattened. Parin and
Collette (1993) warned that the presence of a con-
vex interorbital space and a sagittal crest that ex-
tends onto the ethmoidal region are characters that
seem to change in a gradual manner in the series
L. manis Rosenblatt and Wilson 1987 and L. fitchi,
L. caudatus, and L. calcar Parin and Mikhailin
1982, L. dubius, and L. altifrons. Assurger, Evox-
ymetopon, and Tentoriceps could be added to this
series as a more derived condition.

PTEROSPHENOID. The pterosphenoid forms
the margins of the pterosphenotic window (Collette
and Chao, 1975). The pterosphenoid articulates
with the frontal dorsally and the basisphenoid ven-
trally. Ventrolaterally it joins the sphenotic and the
prootic. The morphology of this bone is similar
among the taxa analyzed in this study.

BASISPHENOID. The basisphenoid is a Y-
shaped median bone. The elongate ventral process,
or base, of the basisphenoid extends toward the
parasphenoid and bisects the entrance of the pos-
terior myodome (Russo, 1983). Dorsally, it joins
the posteroventral edges of the pterosphenoid,
whereas laterally it attaches to the prootic. The
morphology of this bone is similar among the taxa
analyzed in this study.

SCLEROTICS. The sclerotics are ossifications of
the cartilaginous or fibrous regions on the sclera of
the eyes. Nakamura and Yamaguchi (1991) found
that the 21 teleost species analyzed in their study
had at most two sclerotics occupying the anterior
and posterior poles of the scleral equator.

Character 24. No ossified sclerotics are present
in the trichiurids and the outgroups Diplospinus
and Paradiplospinus. However, the outgroups Nes-
iarchus and Gempylus have anterior and posterior
sclerotics on the scleral equator. I have found no
information in the literature about the presence of
these ossifications among the rest of the gempylids.
Nakamura and Yamaguchi (1991) included two
scombroids in their analysis: Thunnus thynnus
(Linnaeus 1758) and Trichiurus lepturus. They in-
dicated that Thunnus thynnus has sclerotics that
grow to form a complete ring around the sclera,
whereas Trichiurus lepturus lacked paired sclerot-
ics. Although they did not consider them scom-
broids, Nakamura and Yamaguchi (1991) also de-
scribed the sclerotics of a marlin and a barracuda.

40 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

They noted that both of these species have a pair
of sclerotics. Collette and Chao (1975) described
the presence of sclerotics in all the genera of the
tribe Sardini within the family Scombridae. Fur-
thermore, Collette and Russo (1984) also noted the
presence of sclerotic bones in the genera Acantho-
cybium Gill 1862, Grammatorcynus, and Scomber-
omorus. Although the character is variable within
the outgroups, I consider the absence of sclerotics
as the apomorphic condition since the presence of
these ossifications seems to be widely distributed
among the scombroids. The presence of sclerotics
in the sphyraenids might be a good indicator of the
plesiomorphic state of this condition because the
barracudas have been proposed as the sister group
to the rest of the scombroids (Collette and Russo,
1986; Johnson, 1986; Carpenter et al., 1995).

Otic Region

SUPRAOCCIPITAL. The supraoccipital is a me-
dian bone forming the posterodorsal corner of the
neurocranium. A supraoccipital crest, which origi-
nates posterior to the pineal foramen, is formed by
the confluence of two ridges from the frontal. The
supraoccipital articulates anteriorly with the fron-
tal, posteriorly with the exoccipital, and laterally
with the parietal and the epiotic.

Starks (1911) described a specimen of Trichiurus
lepturus in which the epiotic, frontal, parietal, and
supraoccipital were covered by a spongy, bony sub-
stance. James (1960) and Nakamura and Parin
(1993) reported that some specimens of T. lepturus
from waters around India show extreme ossifica-
tion (hyperostosis of Barnard, 1948) of the supra-
occipital bone. James (1960) indicated that, in
those specimens of T. lepturus with extreme hyper-
ostosis of the supraoccipital, the whole occipital re-
gion of the neurocranium may be covered by a
thickened bony mass, which may extend to the first
or second vertebral elements. He remarked that in
those specimens the preoccipital profile of the head
approaches the orbital region steeply. Some of the
specimens of T. lepturus analyzed in this study also
show the presence of hyperostosis on the supraoc-
cipital. No instances of hyperostosis on the supra-
occipital have been noted in any other trichiurid or
gempylid. Smith-Vaniz et al. (1995) listed the pres-
ence of hyperostotic bones on 92 species belonging
to 22 teleost families and concluded that hyperos-
tosis has arisen independently many times among
the teleosts.

Character 25. Russo (1983: character 20) noted
that the supraoccipital of Diplospinus, Nealotus,
Nesiarchus, and Paradiplospinus is a small thin
ridge, low on the cranium. He noted that all the
other gempylids in his study, except Lepidocybium
and Ruvettus, have a moderately high supraoccip-
ital crest, extending well above the epiotic ridges.
Russo (1983) described the supraoccipital crest of
Gempylus as being moderate to low in height. In
contrast to Russo’s (1983) observation, the speci-

men of Gempylus analyzed in this study has a su-
praoccipital crest that is higher than in the other
outgroups, but it runs parallel to the epiotic ridges.
The other outgroups, plus the trichiurids Aphano-
pus and Benthodesmus, are characterized by having
a supraoccipital crest that is reduced and runs close
and nearly parallel to the epiotic ridges. In lateral
view, the rest of the trichiurids have a higher su-
praoccipital crest that does not run parallel to the
epiotic ridges and that extends posterodorsally, di-
verging at an angle from the epiotic ridges. Thus,
the outgroups and the trichiurids Aphanopus and
Benthodesmus have a neurocranium with a dorsal
profile that appears flat in lateral view.

Character 26. In Assurger, Evoxymetopon, Lep-
idopus altifrons, and Tentoriceps, the highest point
of the supraoccipital crest ends above the orbits. In
all other trichiurid and outgroup genera, the high-
est point of the supraoccipital crest ends posterior
to the orbits and above the otic or occipital regions
of the neurocranium.

PARIETAL. The parietal articulates with the
frontal anteriorly and the epiotic posteriorly. Lat-
erally, the parietal joins the pterotic bone. A ridge
that originates on the frontal, lateral to the supra-
occipital ridge, continues across the parietal and
onto the epiotic. The morphology of this bone is
similar among the taxa analyzed in this study.

EPIOTIC. The epiotic articulates anteriorly with
the parietal and posteroventrally with the exoccip-
ital. It joins the pterotic laterally and the supraoc-
cipital medially. The epiotic is the posterior-most
bone forming the epiotic ridges, which also cross
the parietal and extend onto the frontal in the out-
groups and most trichiurids. In the trichiurids, the
epiotic ridges usually become confluent with the
pterotic ridges above the parietal or slightly ante-
rior to it on the frontal. The epiotic serves as the
site of attachment for the dorsal articular process
of the posttemporal, uniting the neurocranium and
the pectoral girdle. The posterior margin of the epi-
otic appears more pointed and posteriorly elongate
in dorsal view in Assurger, Benthodesmus, Gem-
pylus, and Nesiarchus and some specimens of Di-
plospinus, Lepidopus, and Trichiurus. In contrast,
the posterior margin of the epiotic does not become
extremely elongate posteriorly and appears to be
flat or rounded in the rest of the outgroups and
trichiurids. However, the shape of the epiotics is
quite variable among and within species. The con-
ditions described above are difficult to categorize
objectively and are dependent on the size of the
specimens available.

PTEROTIC. The pterotic articulates anteriorly
with the frontal and the sphenotic and posteriorly
with the exoccipital and the intercalar. It articulates
ventrally with the prootic and intercalar and me-
dially with the epiotic and the parietal. Ventrolat-
erally, the pterotic forms a fossa that serves as the
articular facet for the dorsal articular condyle of the
hyomandibula. Another fossa at its junction with
the sphenotic accepts part of the anterior articular

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 41

condyle of the hyomandibula. The posterior por-
tion of the otic branch of the laterosensory canal
system (Coombs et ah, 1987), the pterotic canal,
crosses the pterotic longitudinally. Russo (1983)
identified the complete enclosure of the canal as a
synapomorphy of the gempylids. The same condi-
tion characterizes the trichiurids, although the ca-
nal tends to be wider and less ossified in Aphano-
pus and Benthodesmus.

Posteriorly, the pterotic bears two pores of the
laterosensory canal system. The dorsal pore re-
ceives the postotic canal from the anteroventral
branch of the supratemporal, and the lateral pore
connects to the preopercular canal on the preoper-
cle (Coombs et ah, 1987). Russo (1983: character
23) noted that in all gempylids, except Diplospinus,
Lepidocybium, Nesiarchus, Paradiplospinus, and
Thyrsitoides, the dorsal pore extends as a separate
canal in a bony shelf away from the main pterotic
canal. In Gempylus, this shelf appears as a dorsal
ridge that runs throughout the whole length of the
pterotic canal and completely separates the dorsal
and lateral pores. In Aphanopus, Benthodesmus,
and the outgroups, except Gempylus, the dorsal
pore originates directly from the main pterotic ca-
nal, and the dorsal ridge separating the dorsal and
lateral pores is extremely reduced or absent. In the
rest of the trichiurids, the dorsal pore also origi-
nates directly from the main pterotic canal, but the
dorsal ridge separating the dorsal and lateral pores
is restricted to the posterior portion of the pterotic
canal. Assurger, Eupleurogrammus, Evoxymeto-
pon, Lepidopus, and Tentoriceps bear a well-de-
veloped, laterally flattened, porous wall separating
the dorsal and lateral pores. However, the extent of
the dorsal ridge, as well as the presence of a small
porous wall separating the dorsal and lateral pores,
is quite variable and difficult to categorize objec-
tively. This potential character is not included in
this analysis and must await a future study that
incorporates all gempylids and trichiurids

Character 27. In the outgroups, the posterior tip
of the pterotic terminates in front of the posterior
margin of the neurocranium. Trichiurids have a
longer pterotic with the posterior tip terminating
beyond the posterior margin of the neurocranium.
In Lepturacanthus and Trichiurus, the pterotic ex-
tends as far as the first vertebra and, in some spec-
imens, past the anterior margin of the second ver-
tebra. All specimens of Aphanopus, Benthodesmus,
Eupleurogrammus, Evoxymetopon, and Lepidopus
analyzed have a pterotic extending as far as the sec-
ond vertebra. Senta (1975) noted that in Tentori-
ceps, the pterotic processes are well developed and
extend beyond the posterior end of the supraoccip-
ital. In fact, the pterotic of Assurger and Tentori-
ceps extends as far as the third vertebral element.
However, the length of the pterotic among the tri-
chiurids is variable and dependent on the size of
the specimens. In this study I use this character as
a dichotomy in which the states are categorized ac-

cording to whether the pterotic extends past the
posterior margin of the neurocranium or not.

INTERCALAR. The intercalar serves as the
point of attachment for the ligament connecting the
anteroventral articular process of the posttemporal
to the neurocranium. Ventrally, it joins the exoccip-
ital and is overlapped by the anterolateral corner
of this bone. It is strongly fused to the pterotic dor-
sally, and it may articulate with the posterodorsal
corner of the prootic anteriorly.

In the trichiurids and outgroups analyzed, the in-
tercalar varied in shape from truncated to pointed.
In Assurger, the exoccipital and the intercalar form
a plate that appears pointed. Tentoriceps has an
intercalar that is pointed but separate from a pos-
teriorly directed projection on the exoccipital. Rus-
so (1983: character 27) also noted that among the
gempylids, Diplospinus, Lepidocybium, Neoepin-
nula, Paradiplospinus, Ruvettus, Thyrsites, Thyr-
sitops, and Tongaichthys have an intercalar with a
flat or slightly rounded posterior end. However, the
shape of the posterior margin on the intercalar is
highly variable and dependent on the size and de-
gree of ossification of the specimens. In the trichiur-
ids, except Assurger, and the outgroups, the inter-
calar is flat posteriorly and does not extend past
the posterior margin of the neurocranium. Assurger
is unique in that the intercalar is extremely long
and extends posteriorly to reach a position above
the second vertebra.

Character 28. Russo (1983: character 26) noted
that the intercalar on Diplospinus and Paradiplo-
spinus is confined to the ventral surface of the neu-
rocranium and is not visible in dorsal view (a
unique condition among gempylids). In Gempylus,
Nesiarchus, and the trichiurids, the intercalar is re-
duced but visible dorsally. Although the condition
at the outgroup node is equivocal, I follow the con-
clusion of Russo (1983) and consider the condition
present in all gempylids, except Diplospinus and
Paradiplospinus, to be plesiomorphic.

SPHENOTIC. The sphenotic joins the frontal
anteriorly forming the frontosphenotic shelf. Pos-
teriorly, the sphenotic joins the pterotic, whereas
anteromedially it joins the pterosphenoid. Ventral-
ly, the sphenotic articulates with the prootic. In dor-
sal view, the lateral margin of the sphenotic shelf
bears a laterally directed process that accepts the
anterior articular condyle of the hyomandibula.
The morphology of this bone is similar among the
taxa analyzed in this study.

PROOTIC. The prootic is a paired bone; each
unit articulates anteriorly with the basisphenoid
and with each other. The juncture of the two units
of the prootic along their ventral margins forms the
anterior portion of the posterior myodome (Collet-
te and Chao, 1975). The prootic joins the basioc-
cipital, exoccipital, and intercalar posteriorly. It ar-
ticulates dorsally with the pterotic and sphenotic
and ventrally with the parasphenoid. Allis (1903)
described the trigeminofacial chamber as the area
under the lateral arch of the prootic, which encases

42 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

the prootic foramen. All trichiurids and outgroups
bear a prootic arch forming a foramen that con-
nects the trigeminofacial chamber with the anterior
half of the myodome. The morphology of this bone
is similar among the taxa analyzed in this study.

Basicranial Region

EXOCCIPITAL. The exoccipital is a paired bone
partially forming the foramen magnum. It bears an
occipital condyle, which articulates with the atlas
vertebra. This bone articulates with the basioccip-
ital ventrally, the intercalar and supraoccipital dor-
sally, and the prootic anteriorly. Dorsally, it also
joins the pterotic and the epiotic. In most trichiur-
ids, the dorsomedial margins of the exoccipital do
not meet each other, leaving an opening that runs
from the posterior tip of the supraoccipital to the
dorsal margin of the foramen magnum. In some
specimens the dorsomedial margins bear cartilagi-
nous edges that abut each other. An anterior glos-
sopharyngeal foramen (Allis, 1903) is present close
to the prootic, whereas posterior to this a second
foramen is present for the vagus nerve (Russo,
1983).

The exoccipital ridge of the genus Assurger ap-
pears as an extremely elongate process. This elon-
gate process is formed by the posterior extension
of the intercalar and the exoccipital; its posterior
tip reaches above the second vertebral element. Al-
though it is possible that the extreme elongation of
the exoccipital is not independent of the elongation
of the intercalar, both conditions are treated sepa-
rately and considered autapomorphies for the
monotypic genus Assurger.

The exoccipital of Eupleurogrammus is unique in
that it has two small, laterally directed processes on
the dorsal face of the occipital condyle. The ventral
process bears a small and poorly ossified intermus-
cular bone (the cephalic intermuscular bone of Col-
lette and Chao, 1975). None of the outgroups, or
the rest of the trichiurids, has such a modification
on the exoccipital.

Character 29. Russo (1983: character 29) noted
that Diplospinus, Nealotus, and Paradiplospinus
are characterized by the presence of a ridge on the
exoccipital, which is short and does not reach the
vagus foramen. This ridge extends from the pter-
otic, crosses the intercalar, and reaches the exoccip-
ital. In the rest of the gempylids and the trichiurids,
the exoccipital ridge extends as a shelf-like struc-
ture over the intercalar, reaching the vagus foramen
on the exoccipital. Although the condition at the
outgroup node appears to be equivocal, I consider
the presence of an exoccipital ridge that reaches the
vagus foramen to be the plesiomorphic condition
because it is present in all gempylids, except Dip-
lospinus, Nealotus, and Paradiplospinus.

BASIOCCIPITAL. The basioccipital is a median
bone forming the posteroventral comer of the neu-
rocranium and the saccular bulla. Internally, it
forms the lateral and ventral walls of the posterior

myodome (Collette and Russo, 1984). The basioc-
cipital also forms the ventral edge of the foramen
magnum and bears a concave facet for articulation
with the atlas vertebra. It articulates anteriorly with
the prootic and the parasphenoid. Dorsally, it joins
the exoccipital. The morphology of this bone is
similar among the taxa analyzed in this study.

PARASPHENOID. The parasphenoid joins the
vomer anteriorly, and it articulates with the lateral
ethmoid on the posterior wall of the ethmoidal re-
gion. Posteriorly, it articulates with the basisphe-
noid, the prootic, and the basioccipital, in that or-
der. Its articulation with the prootic is by two dor-
solateral extensions that form part of the anterov-
entral wall of the posterior myodome (Collette and
Chao, 1975). The morphology of the parasphenoid
is similar among the trichiurids and outgroups.

PECTORAL GIRDLE

The pectoral girdle is formed by those bones that
support the pectoral fin rays and connect the pec-
toral fin to the neurocranium (Fig. 18).

Supratemporal

The supratemporal is positioned ventrolaterally to
the dorsal articular process of the posttemporal
(Fig. 19). The supratemporal of the trichiurids is
characterized by three branches that bear laterosen-
sory canals. The posterior branch forms the junc-
tion between the temporal canal in the posttem-
poral bone and the system of canals in the skull
(Coombs et al., 1987). The anteroventral branch
joins the temporal canal with the dorsal pore in the
pterotic bone. The anterodorsal branch carries the
supratemporal division of the temporal canal
(Coombs et al., 1987) onto the parietal region. All
the trichiurids and outgroups have a tripartite su-
pratemporal bone with an elongate anterodorsal
arm. However, the supratemporal of the specimens
of Tentoriceps examined have a poorly developed
partition between the two anterior canals. One
specimen of Tentoriceps has a simple, longitudinal,
tube-like supratemporal on the right side followed
anteriorly by another unbranched, longitudinal,
tube-like ossification. In contrast, the left side of the
supratemporal shows a tube with three pores, but
the anterodorsal branch is not extremely elongate.
Russo (1983) indicated that in the gempylids, the
elongate dorsal branch on the supratemporal rep-
resents the medial extrascapular, which has become
fused to the dorsal branch of the supratemporal.
The morphology of this bone is similar among the
taxa analyzed in this study.

Posttemporal

The posttemporal has three arms or articular pro-
cesses: the dorsal articular process attaches to the
dorsal surface of the epiotic; the anteroventral ar-
ticular process bears a ligamentous attachment to
the intercalar; the posteroventral articular process

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 43

POSTTEMPORAL

SUPRACLEITHRUM

POSTCLEITHRA
DORSAL
VENTRAL

CORACOID

Figure 18. A-E. Lateral view of the left pectoral girdle: (A) Diplospinus multistriatus; (B) Paradiplospinus antarcticus ;
(C) Aphanopus carbo ; (D) Assurger anzac ; (E) Benthodesmus tenuis.

44 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 18. F-K. Lateral view of the left pectoral girdle: (F) Eupleurogrammus glossodon; (G) Evoxymetopon taeniatus ;
(H) Lepidopus fitchi ; (I) Eepturacantbus savala; (J) Tentoriceps cristatus ; (K) Trichiurus lepturus.

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 45

POSTTEMPORAL

Figure 19. Lateral view of the left supratemporal, posttemporal, and the articular head of the supracleithrum: (A)
Gempylus serpens ; (B) Aphanopus carbo; (C) Assurger anzac; (D) Eupleu rogrammus glossodon.

overlaps, or is attached by ligaments, to the dorsal
articular head of the supracleithrum laterally. The
posterior margin of the posttemporal overlaps the
articular head of the supracleithrum laterally. This
bone carries the temporal canal of the laterosensory
canal system. It joins the canals in the supratem-
poral with the system of trunk canals in the body
(Coombs et al., 1987).

Character 30. Assurger, Evoxymetopon, Lepi-

dopus, and Tentoriceps are characterized by having
an extremely elongate dorsal articular process,
which extends past the anterior margin of the su-
pratemporal and is at least twice the length of that
bone (Fig. 19C). All the other trichiurids and the
outgroups have a dorsal articular process that is
short and terminates before or shortly after the an-
terior tip of the supratemporal (Fig. 19A, B, D).

Character 31. All the trichiurids, except Aphan-

46 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

opus, share the presence of a short posteroventral
process on the posttemporal that is in direct con-
tact, or in contact via ligamentous association, with
the articular head of the supracleithrum (Fig. 19C,
D). Gempylus bears a well-developed rod-like pos-
teroventral process, but this condition is not com-
parable to that described for most trichiurids (Fig.
19A). In the trichiurids, the process is flat in cross
section, further reduced in size, and more posteri-
orly directed and abuts the supracleithrum medially
or bears a ligamentous connection with its articular
head. In contrast, in Gempylus the posteroventral
process is round in cross section and well developed
and extends ventrally, and it is not in contact or
close association with the articular head of the su-
pracleithrum. Aphanopus, Diplospinus, Nesiar-
chus, and Paradiplospinus lack a posteroventral
process (Fig. 19B).

Character 32. Russo (1983: character 69) iden-
tified the presence of a thin, rod-like anteroventral
articular process as a synapomorphy of Diplospi-
nus, Gempylus, and Paradiplospinus. He concluded
that the presence of this process represents the ple-
siomorphic state characteristic of all the other gem-
pylids, including Nesiarchus. The difference in the
shape of the anteroventral articular process be-
tween Nesiarchus and the other outgroups (robust
versus thin and rod-like) is difficult to evaluate.
However, the anteroventral articular process of the
trichiurids, except Aphanopus, is clearly different.
It is elongate and well developed in the outgroups
plus Aphanopus, and it is reduced in the rest of the
trichiurids. In addition, the process originates at the
posterior corner of the posttemporal of the out-
groups plus Aphanopus (Fig. 19A, B), whereas in
the trichiurids it originates close to or on the an-
terior half of the canal portion of the posttemporal,
never at the posterior corner of the bone (Fig. 19C,
D). Within the trichiurids, Eupleurogrammus, Lep-
turacanthus, Tentoriceps, and Trichiurus have an
anteroventral process that originates separately
from the origin of the posteroventral process. In
contrast, Assurger, Benthodesmus, Evoxymetopon,
and Lepidopus bear a reduced anteroventral pro-
cess that originates from a common ventral ridge
with the posteroventral process.

Character 33. The posterodorsal corner of the
posttemporal in all the trichiurids, except Aphan-
opus and Benthodesmus, is expanded and plate-like
(Fig. 19C, D). In Aphanopus, Benthodesmus, and
the outgroups, the posterior end of the posttem-
poral is tube-like and does not have a plate-like
expansion (Fig. 19A, B).

Supracleithrum

The posterior half of the supracleithrum in all the
outgroups is expanded; the space between the mar-
gins of the supracleithrum is narrow anteriorly and
wide posteriorly, giving the bone an ovoid appear-
ance. All the trichiurids are characterized by having
a supracleithrum with a posterior half that is not

expanded and in which the margins are nearly par-
allel throughout its length. Russo (1983: character
71) noted that Gempylus is characterized by having
a supracleithrum with an elongate shape more sim-
ilar to that described for the trichiurids in this
study. He warned, however, that assignment of
shapes to discrete categories and their subsequent
polarization may be unwarranted. In addition, he
suggested that ontogenetic changes in the shape of
this bone weaken the argument for the categoriza-
tion of this character. I agree with Russo (1983)
that the categorization of this character is difficult,
and I do not include it in the data matrix of this
study.

Character 34. Russo (1983: character 72) noted
that the gempylids Diplospinus, Paradiplospinus,
Promethichthys, and Thyrsites have a posteriorly
expanded process on the head of the supracleith-
rum, which bears a canal and transmits the lateral
line to the posttemporal. The specimens of Nesiar-
chus examined in this study also have a canal bear-
ing a posteriorly expanded process, whereas those
of Gempylus and the trichiurids Eupleurogrammus,
Leptur acanthus, and Trichiurus lack a canal (Fig.
19 A, D). The rest of the trichiurids, except Aphan-
opus, bear a posteriorly expanded head on the su-
pracleithrum, with a semienclosed canal (Fig. 19C).
Aphanopus shares a condition similar to that of the
outgroups, except Gempylus, in which the articular
head of the supracleithrum is posteriorly expanded
and bears a completely enclosed canal (Fig. 19B).

Character 35. Eupleurogrammus, Lepturacan-
thus, and Trichiurus bear a lateral process on the
articular head of the supracleithrum that extends
dorsally (Fig. 19D). This dorsally directed lateral
process is well developed in Lepturacanthus and
Trichiurus, but extremely reduced in Eupleuro-
grammus. The lateral process is absent in all the
other trichiurids and the outgroups (Fig. 19A-C).

Character 36. The articular head of the supra-
cleithrum in Assurger, Eupleurogrammus, Lepido-
pus caudatus, L. fitchi, and Tentoriceps bears a
small anteroventral process (Fig. 19C, D). The rest
of the trichiurids and the outgroups lack this pro-
cess (Fig. 19A, B). However, although this condi-
tion is included as a character in the analysis, the
validity of this character is questionable because the
anteroventral process is quite variable in size within
the five taxa above.

Cleithrum

The main body of the cleithrum is formed by two
longitudinal shelves that meet along their anterior
margins and run parallel to each other. The scap-
ula, coracoid, and dorsal postcleithrum articulate
on the medial longitudinal shelf of the cleithrum.
The two units of the cleithrum meet at their anter-
oventral tips.

The medial shelf of the cleithrum extends poste-
riorly and overlaps the anterior margin of the scap-
ula laterally. This extension varies in shape from

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 47

round to triangular. Starks (1911) described a
thickening on the lower part of the cleithrum in an
Atlantic specimen of Trichiurus lepturus. He con-
cluded that the bone structure was not similar to
that of the hyperostosis of the supraoccipital.

Scapula

The scapula supports three of the four actinosts on
its posterior margin. A scapular foramen is present
and variable in size and shape within species. An
actinost process or posterodorsal facet accepts the
first fin ray of the pectoral fin. The morphology of
this bone is similar among the genera analyzed in
this study.

Coracoid

The coracoid is a paired bone that articulates dor-
sally with the scapula and anteriorly with the
cleithrum. It bears an elongate anteroventral pro-
cess, which attaches to the anteroventral tips of the
cleithrum. It supports the fourth actinost on its pos-
terodorsal corner.

Character 37. Posteroventrally, all trichiurids
have a coracoid with a well-developed plate bearing
a convex ventral margin that extends beyond the
posterior margin of the fourth actinost. The out-
groups lack a posteroventral plate, and they have a
flat ventral margin on the coracoid that ends before
the fourth actinost.

Actinosts

The actinosts are four pairs of bones that support
the bases of the pectoral fin rays. The dorsal-most
actinost is numbered as the first and is the smallest
of the series. The other three actinosts are each
greater in size, with the fourth being the largest.
The first three articulate with the posterior margin
of the scapula, whereas the fourth articulates with
the posterodorsal corner of the coracoid, or par-
tially between the coracoid and the scapula. The
morphology of these bones is similar among the
genera analyzed in this study.

Postcleithrum

The postcleithrum includes two pairs of elongate
bones. The first pair, or dorsal postcleithrum, at-
taches to the posteromedial surface of the cleithrum
above the scapula. The second pair, or ventral post-
cleithrum, attaches to the posterior tip of the dorsal
postcleithrum and extends posteroventrally into the
hypaxial musculature.

The trichiurids and the outgroups have a dorsal
postcleithrum that is short and crooked, or sig-
moid-shaped, and a ventral postcleithrum with a
broad lamellar articular head and a long, styliform
descending process. Gempylus is characterized by a
ventral postcleithrum that is longer than the height
of the pectoral girdle (i.e., from the dorsal margin
of the posttemporal to the ventral margin of the
cleithrum), whereas all the trichiurids and the rest

of the outgroups bear a shorter ventral postcleith-
rum.

Pectoral Fin Rays

Nakamura and Parin (1993) noted that Trichiurus
gangeticus Gupta 1966 bears serrations on the first
pectoral ray. They considered the first pectoral ray
in Eupleurogrammus, Leptur acanthus, and Tri-
chiurus to be a spine. However, cleared and stained
specimens show that the first ray of all the trichiur-
ids and the outgroups is bilaterally divided and that
it has an open base that embraces a cartilaginous
knob above the first actinost. These characteristics
identify this first pectoral element as a soft ray.

Character 38. The relative lengths of the pectoral
fin rays vary among genera and account for the dif-
ferences in shapes of the pectoral fin. In the out-
group genera Diplospinus and Paradiplospinus and
in the trichiurids Aphanopus, Assurger, Benthodes-
mus, Evoxymetopon, and Lepidopus, the posterior
rays are longer. In contrast, in the outgroup Gem-
pylus and in the trichiurids Eupleurogrammus,
Leptur acanthus, Tentoriceps, and Trichiurus, the
anterior rays are the longest. The specimens of Nes-
iarchus analyzed in this study have pectoral fins
with longer posterior rays. However, in the drawing
of an adult Nesiarchus presented by Nakamura and
Parin (1993: 35), the condition appears to be re-
versed. I have not been able to confirm the varia-
tion in this condition, and so I consider the state
present in the specimens used in my study as char-
acteristic of this genus.

PELVIC GIRDLE

The pelvic girdle is composed of the basipterygium,
which supports the fin rays (Fig. 20). Stiassny and
Moore (1992) divided the basipterygium into a cen-
tral part, anterior and posterior processes, and os-
sified wings. The central part bears cartilaginous
tips anteriorly and an articular surface for the fin
rays posteriorly. The posterior and anterior pro-
cesses extend posteriorly and anteroventrally from
the articular surface, respectively. The central part
might bear four wings of membranous origin: ex-
ternal dorsal, external ventral, internal, ventral.

All the taxa analyzed in this study, except Ben-
thodesmus, Evoxymetopon, Lepidopus, and Nes-
iarchus, have a posterior process that is extremely
elongate and extends well past the distal tip of the
pelvic scale or spine. In all of these taxa, the pos-
terior process is about the same length as the cen-
tral part. This process is greatly reduced in size in
Benthodesmus and Nesiarchus, and it does not ex-
tend past the tip of the external fin rays. The pos-
terior process is of about the same length as the
central part in Benthodesmus, whereas it is about
one-third the length of the central part in Nesiar-
chus. The posterior process of Evoxymetopon and
Lepidopus is elongate and about the same length
as the central part. In Evoxymetopon, L. caudatus,
and L. fitchi, it is about the same length as the

48 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

scale-like spine. In L. altifrons, the posterior pro-
cess is slightly larger than the scale-like spine. The
length of the posterior process relative to the cen-
tral part is quite variable and difficult to categorize
objectively.

In all the taxa analyzed, except Aphanopus, the
anterior process appears to be absent or extremely
reduced. The length of the external ventral wing
and the degree of thickness of its ventral margin
are variable. It is well ossified and thick in the out-
groups Diplospinus, Gempylus, and Paradiplospi-
nus and the trichiurids Benthodesmus tenuis, Eu-
pleurogr animus, and Tentoriceps. In contrast, it is
not well ossified in Assurger, Evoxymetopon, Lep-
idopus, and the outgroup Nesiarchus. The length
of the external ventral wing is quite variable and
varies from less than half of the length ( Diplospinus
and Paradiplospinus) to more than half and up to
about equal the length of the central part (all other
taxa analyzed). Some specimens of Tentoriceps
have an external ventral wing that extends only
half the length of the central part, whereas in others
it extends throughout the whole central part. The
length of the external ventral wing is also variable
among the species of Benthodesmus.

The degree of extension of the ventral margin of
the external ventral wing is extremely variable and
dependent on the degree of ossification and size of
the specimens. Potthoff (1980) reported that in
Coryphaena Linnaeus 1758, the structures of mem-
branous origin in the pelvic girdle (anterior process
and wings) develop last, after the ossification of the
central part. Thus, categorization of the extent and
degree of ossification of the external ventral wing
is difficult, and the character is not included in this
analysis.

Character 39. The outgroups are characterized
by the presence of a well-developed basipterygium.
In the trichiurids, the basipterygium has become ex-
tremely reduced or completely lost as it occurs in
Lepturacanthus and Trichiurus.

Although the basipterygium of Aphanopus is re-
duced to a small internal plate, some parts are dis-
cernible under the microscope. Anteriorly the bas-
ipterygium bears what appears to be an extremely
reduced central part. The central part appears as
two unfused, short processes that are slightly in-
clined dorsally. In Aphanopus and the outgroup
Paradiplospinus, the pelvic girdle bears a single
spine in the juveniles, but the spine becomes ex-
tremely reduced in the adult where only the internal
basipterygium remains.

Character 40. Diplospinus bears a pelvic fin with
an ornamented scale-like element and no soft rays.
The lateral margins of the scale-like element in Dip-
lospinus are serrate. However, although this con-
dition is described as different from that present in
the trichiurids with a pelvic fin (i.e., presence of an
unornamented scale-like element), one must be
careful because the largest specimen of Diplospinus
available in this study was only 193 mm SL. Nak-
amura and Parin (1993) reported that this species

is common up to 200 mm and that it reaches a
maximum of 330 mm SL. It is possible that larger
specimens of Diplospinus have lost the ornamen-
tation of the lateral margins of this scale-like ele-
ment. In contrast, Gempylus and Nesiarchus have
stronger, better developed pelvic girdles with one
spine and from three to four and one to five soft
rays, respectively. Eupleurogrammus and Tentori-
ceps have a reduced scale-like element and no soft
rays. Benthodesmus bears a scale-like element and
a single soft ray. Assurger, Evoxymetopon, and the
species of Lepidopus analyzed in this study have a
scale-like element and two soft rays. Lepturacan-
thus and Trichiurus lack the pelvic girdle and fin
elements.

Character 41. The position of the basipterygium
relative to the pectoral girdle has been used as a
taxonomic character for the identification of some
trichiurid taxa (Nakamura and Parin, 1993). Eu-
pleurogrammus and Tentoriceps are characterized
by an abdominal basipterygium that is located pos-
teriorly, well past the pectoral girdle, including the
tip of the ventral postcleithrum. The outgroup taxa
analyzed in this study are characterized by having
a basipterygium that is located under the pectoral
girdle. Aphanopus, Assurger, Benthodesmus, Evox-
ymetopon, and Lepidopus have a basipterygium
that is located completely or partly under the pec-
toral girdle (including the ventral postcleithrum).
The basipterygium of adult Aphanopus is extremely
reduced and located under the coracoid, anterior to
the actinosts. The pelvic girdle of Benthodesmus is
also reduced in size, and its position is variable
among species. Nakamura and Parin (1993) used
the position of the base of the pelvic-fin rays with
respect to the pectoral-fin rays as a character for
the identification of the different species of Bentho-
desmus. In the cleared and stained specimens of the
species of Benthodesmus analyzed in this study, the
basipterygium is always located under the pectoral
girdle anterior to the ventral postcleithrum. Assur-
ger, Evoxymetopon, and Lepidopus have a basip-
terygium in which the posterior process is partially
or completely located posterior to a vertical | from
the ventral tip of the ventral postcleithrum. How-
ever, in Assurger, Evoxymetopon, Lepidopus, and
those species of Benthodesmus not available for
this study, in which the base of the pelvic-fin rays
is posterior to the pectoral-fin base, the articular
facet of the basipterygium is always anterior to the
tip of the ventral postcleithrum.

Character 42. In all the outgroups the central
part of the basipterygium is dorsally inclined and
extends anteriorly between the cleithrum and cor-
acoid bones. Stiassny and Moore (1992) described
this condition among percomorphs and noted that
the central part of the pelvic girdle attaches to the
cleithrum or coracoid by ligaments. They also con-
cluded that among the acanthomorphs, this condi-
tion can be considered as apomorphic, with the ple-
siomorphic condition represented by a pelvic girdle
that is parallel to the ventral body wall. All the

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 49

50 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Assurger anzac, (E) Benthodesmus tenuis.

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 51

trichiurids, except Lepturacanthus and Trichiurus,
are characterized by having a basipterygium that is
parallel or nearly parallel to the central body wall.
The central part of the basipterygium in Aphano-
pus, Benthodesmus, Evoxymetopon, and Lepido-
pus may be slightly inclined dorsally. However, the
condition in these genera is not comparable to that
present in the outgroups, in which the central part
is extremely inclined and extends between the
cleithrum and coracoid.

Character 43. Eupleurogrammus and Tentori-
ceps are characterized by having a basipterygium
that is completely fused along its longitudinal axis.
In Benthodesmus, the posterior process appears to
be fused, whereas the central part is not. The re-
duced basipterygium of Aphanopus bears a central
part that appears as two unfused processes. In con-
trast, Assurger, Evoxymetopon, Lepidopus, and the
outgroups have a basipterygium that is not fused
along its entire longitudinal axis.

AXIAL SKELETON
Vertebral Column

In the outgroups and the trichiurids, except Ten-
toriceps, the first rib is present on the third precau-
dal centrum (Fig. 21). Tentoriceps is the only taxon
analyzed in which the first rib is on the second pre-
caudal centrum. The trichiurids, except some of the
specimens of Lepidopus fitchi, are characterized by
the presence of ribs in all the rest of the precaudal
centra. In the outgroups, some posterior precaudal
centra bear haemal ribs that articulate laterally
with a fully formed haemal arch. I found no evi-
dence of haemal ribs in the posterior precaudal cen-
tra of the trichiurids, except in Lepidopus fitchi, in
which the last two to four precaudal centra bear
haemal ribs. However, in some specimens of Lepi-
dopus fitchi, the last haemal ribs may be fused at
their tips.

The neural prezygapophyses are well developed
in the outgroups and the trichiurids. Those in Eu-
pleurogrammus, Lepturacanthus, and Trichiurus
are larger in size. In contrast, the neural postzyga-
pophyses are smaller than the prezygapophyses.
The haemal pre- and postzygapophyses are well de-
veloped in the caudal centra of all the outgroups
and trichiurids. In Eupleurogrammus, Lepturacan-
thus, Tentoriceps, and Trichiurus, the haemal pre-
and postzygapophyses are more vertically directed
when compared with the rest of the taxa analyzed.

Lateral apophyses (extending on the frontal
plane of the vertebral column) are present in the
posterior-most precaudal centra and most of the
caudal centra of the trichiurids. In the trichiurids
with a caudal fin, the last caudal centrum bearing
a lateral apophysis is that which still maintains an
articulation with a dorsal or anal pterygiophore, or
both. The proximal shafts of a few of the last anal
and dorsal pterygiophores have lost their direct ar-
ticulation with the corresponding neural and hae-
mal spines.

The specimens of Benthodesmus tenuis analyzed
in this study have a few posterior caudal vertebrae
that lack lateral apophyses, although they still
maintain an articulation with the corresponding
dorsal and anal pterygiophores. The last centrum
bearing an articulation with the corresponding
pterygiophores has lateral apophyses.

Those trichiurids lacking a caudal fin (i.e., Eu-
pleurogrammus, Lepturacanthus, Tentoriceps, and
Trichiurus) also have lateral apophyses. The lateral
apophyses are better developed along the anterior
half of their respective centra. As one proceeds pos-
teriorly, the length of the lateral apophyses increas-
es, and the length is gradually reduced again in the
posterior caudal vertebrae. The last lateral apoph-
ysis is on the last centrum maintaining an articu-
lation with the dorsal and anal pterygiophores. In
the posterior caudal centra, the lateral apophyses
extend anteriorly onto the preceding vertebral ele-
ment. These lateral apophyses are more elongate in
Eupleurogrammus, Lepturacanthus, and Trichiurus
than in any of the other taxa analyzed. In Eupleu-
rogrammus, the lateral apophyses become more
vertically oriented as one proceeds posteriorly. As
the lateral apophyses change orientation, the hae-
mal prezygapophyses become reduced in size and
appear to be replaced in position by the vertically
oriented lateral apophyses. The outgroups, except
Gempylus, lack or have extremely reduced lateral
apophyses. Gempylus bears reduced lateral apoph-
yses that do not extend past the anterior margin of
their corresponding centra, as in the trichiurids.

Character 44. A characteristic of the trichiurids
is the extreme elongation of their bodies. According
to Nakamura and Parin (1993), the total number
of vertebrae range from 57 to 64 in Diplospinus,
48 to 55 in Gempylus, 34 to 36 in Nesiarchus, 60
to 67 in Paradiplospinus, and 84 to 198 in the tri-
chiurids. Collette et al. (1984: character 29) and
Johnson (1986: character 16) utilized vertebral
counts as a multistate character series. Johnson
(1986) warned of the arbitrariness involved in cat-
egorizing the states of a meristic character. I agree
with the conclusions of Carpenter et al. (1995) and
consider their groupings among all the scombroids
as a better representation of the distribution of ver-
tebral numbers. Following the categorization of
Carpenter et al. (1995) for all scombroids (with a
slight modification to account for meristic data
from this study and the literature), I consider the
following character states with respect to the total
number of vertebrae: 30 to 55 as state 0 among the
gempylids (including Gempylus and Nesiarchus );
57 to 67 as state 1 among the gempylids (including
Diplospinus and Paradiplospinus ); 84 to 198 as
state 2 for the trichiurids. Although this multistate
character results in the assignment of an equivocal
state at the outgroup node, I consider the condition
present in the basal gempylids (i.e., gempylids mi-
nus Diplospinus and Paradiplospinus) as plesiom-
orphic.

Character 45. The first neural spine of all the

52 ■ Contributions in Science, Number 476

Gago: Trichiurid Phytogeny

Figure 21. Left view of the anterior elements of the axial skeleton: (A) Bentbodesmus tenuis ; (B) Trichiurus lepturus.

trichiurids analyzed is distally bifurcate, and the
proximal-middle radial of the first dorsal ptery-
giophore partially fits between its tips. The condi-
tion has been described in Trichiurus (Fig. 2 IB) by
Potthoff et al. (1986). The outgroups, except Gem-

pylus, have a first neural spine that is not bifurcate
at its tip. In the specimen of Gempylus analyzed in
this study, the first vertebral element bears an ex-
tremely reduced, distally bifurcate neural spine.
However, the condition in Gempylus is not com-

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 53

parable to that of the trichiurids. The unfused tips
might be part of the neural arch, which is never
completely closed during their ontogeny. The prox-
imal-middle radial of the first dorsal pterygiophore
does not fit between the unfused tips of the first
neural spine.

Character 46. In Leptur acanthus and Trichiurus,
the neural spines of centra two and three are ex-
panded and plate-like with their anterior and pos-
terior margins well ossified and bearing pointed tips
that give the neural spines a forked shape (Fig.
2 IB). Assurger, Eupleurogrammus, and Tentoriceps
share the same condition, but the expanded neural
spines are on centra two to four, two to five, and
two to six, respectively. All of the neural spines in
the outgroups and in the trichiurids Aphanopus
and Benthodesmus are simple and not expanded
(Fig. 21 A). Most of the specimens of Evoxymeto-
pon, Lepidopus altifrons, L. caudatus, and L. fit chi
analyzed bear slightly expanded anterior neural
spines (not including the first centrum), but they are
not distally forked as in Assurger, Eupleurogram-
mus, Leptur acanthus, Tentoriceps, and Trichiurus.

Intermuscular Bones

The intermusculars are segmental, serially homol-
ogous ossifications or ligaments in the myosepta of
teleosts (Patterson and Johnson, 1995) that can be
divided into myorhabdoi, epipleurals, epineurals,
and epicentrals. Myorhabdoi, which occur only in
a few teleosts, are not present in the trichiurids.
Patterson and Johnson (1995) noted that epipleur-
als are absent in all acanthomorphs, except Poly-
mixia Lowe 1838, Velifer Temminck and Schlegel
1850, and holocentrids. Epineurals appear as a se-
ries of bones or ligaments that develops in a ros-
trocaudal direction from the occipital region back.
In the plesiomorphic condition, epineurals develop
as outgrowths of the neural arches. In derived tel-
eosts, epineurals have lost the ossified continuity
with their respective neural arch and retain only a
ligamentous attachment (Patterson and Johnson,
1995).

Epineurals are present in all the outgroups and
the trichiurids, ranging from three to seven and ar-
ticulating directly with the first three to seven ver-
tebrae (Fig. 21). The first epineural originates on
the neural arch of the first vertebral element, where-
as the others originate farther ventrally on the head
of the rib or the parapophyses. The distal margin
of the anterior epineurals is forked in Tentoriceps
and simple and pointed in other trichiurids and out-
groups. However, Evoxymetopon and Lepidopus
bear slightly expanded epineurals with a well-ossi-
fied central axis that supports two lateral thin
plates.

In addition to the series of three to seven epineu-
rals that articulate directly with the vertebrae, the
outgroups are characterized by the continuation of
the ossified epineural series farther back into the
caudal region. Nesiarchus bears what appear to be

unattached epineurals from vertebrae 6 to 17 (the
vertebral numbers are in reference to the position
of the proximal tip of the epineural with respect to
the vertebrae). These epineurals are simple, rod-like
ossifications, with the exception of the epineurals
on vertebrae 9 and 10, which are slightly forked
proximally. In Paradiplospinus, the attached epi-
neurals on vertebrae 4 to 6 are forked proximally.
The anteromedial branch attaches to the corre-
sponding centrum, whereas the anteroventral
branch extends next to the preceding centrum. Dor-
sal to the seventh attached epineural there is a free,
rod-like epineural similar in orientation and shape
to the anteroventral branch and distal portion of
the preceding epineurals. Farther posteriorly, Par-
adiplospinus bears simple, unattached, rod-like epi-
neurals from vertebrae 8 to 59. 1 consider these ap-
parently unattached epineurals as homologous with
the anteroventral branches and distal portions of
those anterior epineurals that are proximally
forked. In Diplospinus, the unattached epineurals
extend between vertebrae 6 and 54, and the at-
tached epineurals of vertebrae 3 to 6 are forked.
Gempylus bears three attached epineurals that ar-
ticulate directly with the first three vertebrae (the
first articulates at the base of the neural arch). The
first two are simple and rod-like, but the third is
forked proximally (i.e., the anteromedial branch at-
taches to the corresponding centrum and the anter-
oventral branch extends next to centrum 2). Gem-
pylus is characterized by having the first unattached
epineural originating anterior to the attached epi-
neural of the first vertebra and medial to the su-
pratemporal. The second epineural of Gempylus is
also unattached; it is parallel to the distal portion
of the first attached epineural, with its proximal tip
lying above centrum 2. The next element of the se-
ries is represented by the anteroventral branch and
distal portion of the forked epineural attached to
centrum 3. Posterior to these few elements the rest
of the epineurals are unattached, simple, and rod-
like, with the series extending caudally to centrum
45. Although Diplospinus, Nesiarchus, and Para-
diplospinus have no unattached epineurals origi-
nating anterior to centra 3 to 7, a series of liga-
ments parallel to the ossified, unattached epineurals
extends anteriorly to these centra.

The epicentrals of teleosts develop in a rostro-
caudal direction, lie in the horizontal septum, and
are almost always simple rods (Patterson and John-
son, 1995). Diplospinus, Gempylus, and Paradip-
lospinus have a series of intermuscular bones that
extend into the hypaxial musculature. These bones
are simple, rod-like ossifications, and they are in
reverse orientation to their unattached epineural
counterparts. Their proximal tips appear to be un-
attached and lie laterally to their corresponding
centrum. I consider this series of intermusculars as
epicentrals. The proximal tips of the epicentrals
originate next to centra 4 to 54 in Diplospinus, 3
to 45 in Gempylus, and 6 to 59 in Paradiplospinus.
Although both of the epicentral series and most of

54 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

the epineurals appear to be unattached, it is possi-
ble that they have a ligamentous connection with
their corresponding centra. The proximal tips of
the unattached intermuscular bones in the out-
groups are extremely thin, and I could not deter-
mine the presence of a ligamentous attachment at
their proximal tips. Furthermore, the vertebral
ranges described above are based on a single or a
few representative specimens of each genus. Varia-
tion in these ranges was observed in those taxa for
which more than one specimen was available. In
this section, only the upper and lower limits of
these ranges are presented. Thus, these meristics
should not be taken as an absolute description of
the distribution of intermusculars in these fishes.

Character 47. Presence or absence of unattached,
ossified epineurals and epicentrals is considered as
a multistate character. Diplospinus, Gempylus, and
Paradiplospinus are characterized by the presence
of unattached epineural and epicentral series that
extend into the caudal region. Nesiarchus has a se-
ries of unattached epineurals that extends into the
caudal region but lacks an epicentral series. Tri-
chiurids bear a few attached epineurals, but they
lack the series of unattached epineurals and epicen-
trals.

Dorsal Fins

Tucker (1953) indicated that the presence of a con-
tinuous dorsal fin in most trichiurids clearly sets
them apart from Aphanopus, Benthodesmus, the
gempylids, and other scombroids.

Character 48. Aphanopus and Benthodesmus
have a notch on the dorsal-fin membrane that sep-
arates the two dorsal fins, whereas the rest of the
trichiurids have a continuous dorsal-fin membrane.
All gempylids have a well-developed notch that sep-
arates the soft and spinous portions of the dorsal
fin.

Character 49. In the gempylids the spinous dor-
sal fin bears from 19 to 39 spines and its base is
longer than that of the soft dorsal. Most trichiurids
have a spinous dorsal fin with a base that is shorter
than the soft dorsal. Within the trichiurids, Aphan-
opus has a spinous dorsal fin that bears from 38 to
45 spines and is only slightly shorter than the soft-
dorsal portion. Assurger and Benthodesmus are
both characterized by a spinous dorsal fin that
bears 31 to 46 spines and is less than half of the
length of the soft-dorsal portion. A further derived
condition, where the spinous dorsal fin is extremely
short with only three to 10 spines, could be hy-
pothesized for all trichiurids, except Aphanopus ,
Assurger, and Benthodesmus. Nakamura and Parin
(1993: 70) indicated that Aphanopus and Bentho-
desmus have from 38 to 45 and 31 to 46 dorsal-
fin spines, respectively. They also reported that As-
surger only has “a few weak anterior spines hardly
differing from the soft rays.” However, upon close
examination of radiographs and cleared and
stained specimens of Assurger I found that its dor-

sal fin is composed of 34 to 35 spinous rays. These
rays are true spines in that they are unsegmented,
median elements with a closed base bearing a cen-
tral foramen. In contrast, the soft rays are usually
segmented, bilaterally divided, and branched at
their tips. The spinous rays can also be identified
in the trichiurids and outgroups by the type of ar-
ticulation with their respective pterygiophores. The
bases of the spinous rays do not embrace the distal
radial (extra distal “x” radial of Johnson, 1986:
character 30).

Character 50. Johnson (1986) noted that the
morphology and development of pterygiophores of
the soft-dorsal and anal fins in the trichiurids is
similar to that of spinous pterygiophores. The pter-
ygiophores supporting spinous-dorsal rays in the
gempylids and the trichiurids are composed of
proximal-middle and distal radials, which articu-
late by extensive overlapping (Johnson, 1986: char-
acter 21). These proximal-middle and distal radials
have a concave dorsal face. In Gempylus the prox-
imal-middle radials are extremely concave dorsally
and bear well-developed pointed corners in their
middle portions. Furthermore, Johnson (1986:
character 22) indicated that in the gempylids, the
posterior facet of the spinous distal radials is con-
vex and acts as an articular condyle for the concave
ventral margin of the closed bases of the dorsal
spines. This condition is also shared by the spinous-
dorsal elements of the trichiurids.

Johnson (1986: character 30) noted that the ba-
ses of the soft rays in the dorsal and anal fins of
trichiurids embrace an extra distal “x” element that
develops, as a separate ossification, posterior to its
corresponding distal radial. The pterygiophores of
the soft-dorsal and anal fins in the outgroups uti-
lized in this study also have a separate radial that
is embraced by the bases of the soft rays. This el-
ement is similar in shape to the “x” radial described
by Johnson (1986), which is embraced by the open
bases of the soft rays of the trichiurids (Fig. 22).
The “x” element in the trichiurid soft rays bears an
anteroventrally projecting pedestal (anterodorsally
in the anal-fin pterygiophores) that ends in a car-
tilaginous knob that articulates with the concave
articular facet of the preceding distal radial. The
ux” radial also bears two laterally directed wings
posteriorly. In trichiurids, the “x” element is a sin-
gle ossification. In the outgroups, the distal radial,
which is embraced by the open base of the corre-
sponding soft ray, is not ossified along its medial
axis, but the posterolateral wings and the lateral
faces of the anterior pedestal are ossified. Johnson
(1986) considered the possibility that the “x” ele-
ment, embraced by the bases of the soft rays in the
trichiurids, is a neomorph. The ontogenetic data of
Gago (1997) indicate that the extra distal radial of
Johnson (1986) is not a neomorph. Ontogenetic se-
ries of the outgroup genera show that the trichiur-
ids and gempylids share a similar pattern of devel-
opment of their dorsal fin pterygiophores. The pres-
ence of only two radials (proximal-middle and dis-

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 55

Figure 22. Ventral and lateral views (top and bottom, re-
spectively; cartilage is not stippled) of the anal-fin ‘x’ ra-
dial of Johnson (1986) in adult: (A) Trichiurus lepturus,
288 mm TL; (B) Paradiplospinus antarcticus, 278 mm SL.

tal) supporting the soft rays of the dorsal fin in
gempylids is the result of the fusion of the proximal
and middle radials during development. In the tri-
chiurids, the proximal and middle radials do not
fuse during development and the adult retains three
separate radials. Thus, the extra distal radial of
Johnson (1986) appears to be homologous to the
distal radial of other scombroids.

Gunther (1860) and James (1960) reported hy-
perostosis of some of the dorsal proximal-middle
radials in some specimens of Lepidopus caudatus
and Trichiurus lepturus. James (1960) indicated
that of the four species of trichiurids occurring in
Indian waters (i.e., Eupleurogrammus glossodon,
E. muticus (Gray 1831), Lepturacanthus savala,
and T. lepturus) this condition was only present in
T. lepturus. Radiographs of the holotype of Evox-
ymetopon taeniatus also show hyperostosis of some
of the dorsal-fin proximal-middle radials. Smith-
Vaniz et al. (1995) only listed the presence of hy-
perostosis in one species of trichiurid. To this list I
add the report of hyperostosis on the dorsal-fin
pterygiophores of L. caudatus and the new record
of this condition in the holotype of E. taeniatus.
The first dorsal pterygiophore of all the outgroups
and the trichiurids bears a supernumerary spine
and a second spine that articulates with the poste-
rior facet of the distal radial.

Character 51. Johnson (1986: character 20) re-
ported an extreme plate-like expansion of the prox-
imal shaft of the first proximal-middle radial in the
trichiurids (Fig. 21) and gempylids, except Lepi-
docybium and Ruvettus, which bear only a mod-
erate expansion. In all trichiurids, except Aphano-
pus and Benthodesmus, this plate-like expansion of
the first dorsal pterygiophore extends anteriorly
above the occipital region of the neurocranium.
The outgroups plus Aphanopus and Benthodesmus
are characterized by a first proximal-middle radial

that does not extend onto the occipital region of
the neurocranium.

In Benthodesmus, several of the anterior proxi-
mal-middle radials following the first pterygiophore
also bear a plate-like extension on the anterior and
posterior margins of the proximal shaft. The pos-
terior margin of the plate-like extension is well os-
sified and, in conjunction with the obliquely ori-
ented proximal shaft, gives the bone the appearance
of an inverted “V.” The tubular shaft bears a car-
tilaginous tip and lies free in the corresponding in-
terneural space, whereas the posterior margin of
the plate-like extension articulates with the neural
spine of the subsequent vertebral element. As one
proceeds posteriorly, the plate-like ossifications be-
come smaller and the tubular margin shifts gradu-
ally and posteriorly until it comes in contact with
the neural spine of the subsequent vertebral ele-
ment.

All trichiurids, except Aphanopus and Bentho-
desmus, have a first dorsal proximal-middle radial
with an elongate proximal shaft that fits primarily
in the second interneural space. However, the an-
terior plate-like extension of the proximal shaft ex-
tends onto the occipital region and passes between
the bifurcate tip of the first neural spine. Aphano-
pus and Benthodesmus have a shorter proximal
shaft on the first dorsal proximal-middle radial that
does not extend very far into the second interneural
space. The anterior plate-like extension of the prox-
imal-middle shaft in these two genera partially fits
between the bifurcate tip of the first neural spine.
Diplospinus and Paradiplospinus also have a prox-
imal shaft on the first dorsal pterygiophore, which
sits above the first and second interneural spaces,
but the first neural spine is not bifurcate. The first
neural spine of Nesiarchus is fused. The proximal
shaft of the first dorsal pterygiophore is long in
both Gempylus and Nesiarchus. In Nesiarchus it
fits primarily in the second interneural space, but
in Gempylus it extends above both the first and the
second interneural spaces.

Character 52. In Eupleurogrammus, Lepturacan-
thus, Tentoriceps, and Trichiurus, a plate-like os-
sification at the anteroventral corner between the
proximal and distal portions of the proximal radi-
als bears a small foramen. This foramen is absent
or extremely reduced in the proximal radials of the
other trichiurids and the outgroups.

Anal Fin

The morphology of the anal-fin pterygiophores in
the outgroups and most of the trichiurids is iden-
tical to that described for the dorsal-fin pterygiop-
hores.

Character 53. The first anal pterygiophore in the
outgroups bears two well-developed supernumer-
ary spines (Fig. 23 A, B). Aphanopus, Assurger, Ben-
thodesmus, Evoxymetopon, and Lepidopus bear
two supernumerary spinous elements on the first
anal pterygiophore (Fig. 23C-E, G, H). The first

56 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

supernumerary spine in these taxa is extremely re-
duced in size. Nakamura and Parin (1993) and Sen-
ta (1975) described the presence of two spines in
the anal fin of Tentoriceps. However, in all the
cleared and stained and alcohol-preserved speci-
mens of Tentoriceps analyzed, I only found a single
scale-like supernumerary element (Fig. 23J). Eu-
pleurogrammus, Lepturacanthus, and Trichiurus
also have a single supernumerary element on the
first anal pterygiophore. In the outgroups, the third
fin-ray element associated with the first anal pter-
ygiophore appears spinous in nature. The base of
this element is completely or almost completely
fused, forming a central foramen that accepts the
posterior tip of the distal radial in Gempylus and
Nesiarchus. However, this fin-ray element is bilat-
erally divided. Diplospinus and Paradiplospinus
also have a third fin-ray element that is bilaterally
divided, but the distal radial of the first anal pter-
ygiophore does not pass through the central fora-
men. The distal radial of the first anal pterygiop-
hore, which is in close association to the third fin-
ray element, appears as a single ossification. The
third fin-ray element of the first anal pterygiophore
in the trichiurids is extremely reduced in size. In
some specimens only one or two extremely reduced
scale-like elements remain, and they appear to be
remnants of a soft ray.

The two supernumerary spines of the first anal
pterygiophore in Gempylus and Nesiarchus are
well developed and rounded in cross section, except
at their base, which embraces the articular facet of
the first proximal-middle radial. The first supernu-
merary spine is longer than the second in Nesiar-
chus, whereas the rest of the outgroups have a lon-
ger second supernumerary spine. The supernumer-
ary spines of the first anal pterygiophore of Dip-
lospinus and Paradiplospinus are well developed,
but V-shaped in cross section and bearing lateral
wings. The first supernumerary spine of Aphano-
pus, Assurger, Benthodesmus, Evoxymetopon, and
Lepidopus is extremely reduced. The second super-
numerary spine of Aphanopus is elongate, trian-
gular in ventral view, and slightly V-shaped in cross
section. The supernumerary elements of the first
anal pterygiophore in Assurger, Benthodesmus,
Evoxymetopon, and Lepidopus are modified and
scale-like. The second supernumerary element is
variable in shape, but it appears cardiform, or tri-
angular, in ventral view, with well-developed lateral
wings and V-shaped cross section.

The single scale-like supernumerary element of
the first anal pterygiophore in Tentoriceps is similar
to the second supernumerary elements of Assurger,
Benthodesmus, Evoxymetopon, and Lepidopus. In
Eupleurogrammus, the single supernumerary ele-
ment is scale-like and bears lateral wings, but it is
extremely reduced in size (Fig. 23F). The single su-
pernumerary element in Lepturacanthus and Tri-
chiurus is spinous in appearance (Fig. 231, K) but
triangular in cross section. This spine is much lon-
ger in Lepturacanthus than in Trichiurus, and this

feature was used by Nakamura and Parin (1993)
as a character to differentiate these two genera.

The morphology of the posterior supernumerary
element of the first anal-fin pterygiophore could be
treated as a multistate character. However, the con-
dition at the outgroup node is equivocal and the
character is variable and difficult to categorize ob-
jectively. This potential multistate character must
await a study that includes the gempylids and tri-
chiurids together.

Benthodesmus is unique in that the first anal
pterygiophore is abdominal in position, being lo-
cated under the precaudal vertebrae. In all the out-
groups and the rest of the trichiurids, the first anal
pterygiophore articulates with the haemal spine of
the first (or one of the first) caudal vertebrae.

The first anal pterygiophore of all the trichiurids,
except Benthodesmus, and of the outgroups bears
an elongate proximal shaft on the proximal-middle
radial that articulates with its corresponding hae-
mal spine. Benthodesmus elongatus and B. simonyi
lack a proximal shaft on the first pterygiophore.
However, the specimen of B. tenuis analyzed in this
study has two poorly ossified proximal shafts that
do not articulate with the vertebral column. The
presence or absence of a proximal shaft in the first
anal pterygiophore might prove to be a phyloge-
netically informative character within the genus
Benthodesmus. However, its incorporation into a
phylogenetic analysis must await the inclusion of
most or all of the species of this genus in future
studies. All of the trichiurids and the outgroups,
except Gempylus, have a short anterior process on
the first anal pterygiophore. Gempylus is unique in
that the anterior process is extremely elongate and
is almost equal in length to the proximal shaft.

Radiographs of the holotype of Evoxymetopon
taeniatus show hyperostosis of some of the anal-fin
proximal-middle radials.

Character 54. All the outgroups and the trichiur-
ids Aphanopus and Benthodesmus have true soft
rays (i.e., bilaterally divided, segmented, and dis-
tally branched) throughout the whole length of the
anal fin (Fig. 24A, B). In Assurger, Lepidopus, and
Evoxymetopon, most of the anal-fin rays are ex-
tremely reduced; only the last few rays are external,
can be identified as true soft rays, and are connect-
ed by a membrane. In Aphanopus and some species
of Benthodesmus, the anal fin is composed of true
external soft rays that are united by a membrane
along its entire length. However, one must be cau-
tious in the interpretation of this character because
its condition appears to be dependent on the size
of the specimens. In this study, the cleared and
stained specimens of Lepidopus have well-devel-
oped soft rays throughout the anal fin but only the
posterior ones are connected by a membrane. Larg-
er dry-skeletal preparations of this taxon show an
extreme reduction of these anterior fin rays. In Lep-
turacanthus and Trichiurus, the anal-fin soft rays
are reduced to spinule-like processes that are not
branched or segmented (Fig. 24E). These spinules

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 57

58 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 59

Figure 23. F K. Left view of the first and second anal-fin pterygiophores: (F) Eupleurogrammus glossodon ; (G) Evoxymetopon taeniatus ; (H) Lepidopus fitchi ; (I) Leptur acanthus savala;

are V-shaped in cross section and resemble the mor-
phology of the supernumerary spine of the first anal
pterygiophore.

Tentoriceps shares the same condition as Leptur-
acantbus and Trichiurus, but the rays are reduced
to minuscule scale-like elements that do not pene-
trate the skin externally (Fig. 24D). Eupleurogram-
mus shows another derived condition in which the
rays barely penetrate the skin and appear as small
fused knobs on the proximal-middle radials (Fig.
24C).

Character 55. The anal-fin pterygiophores of the
outgroups are characterized by the presence of
proximal-middle and distal radials (Fig. 24 A). Tri-
chiurids, except Eupleurogrammus, Lepturacan-
tbus, Tentoriceps, and Trichiurus, have an anal fin
in which the pterygiophores of the soft ray portion
are composed of proximal, middle, and distal ra-
dials (proximal-middle and extra distal of Johnson,
1986; Fig. 24B). Eupleurogrammus, Lepturacan-
thus, Tentoriceps, and Trichiurus are characterized
by the fusion of the radials as a single unit (Fig.
24C-D).

CAUDAL COMPLEX

The caudal complex is composed of a series of cau-
dal rays and supporting bones (Fig. 25).

Character 56. All the gempylids are characterized
by the presence of a well-developed caudal com-
plex. In some of the trichiurids, the caudal fin com-
plex has become reduced or completely lost (Fig.
26). Senta (1975) briefly described the presence of
an extremely reduced internal caudal complex in
Tentoriceps. The posterior tip of the body in Ten-
toriceps is more rounded in appearance than those
of Lepturacanthus and Trichiurus, which bear ex-
tremely pointed caudal tips. Internally, Tentoriceps
bears two rudimentary hypural plates on the last
vertebral element. From 6 to 14 rudimentary rays
are also present in the cleared and stained speci-
mens of Tentoriceps analyzed in this study. These
rays barely penetrate the skin and articulate mainly
on the margins of the reduced hypural plate and
the dorsal surface of the last vertebral element.
Adults of Eupleurogrammus, Lepturacanthus, and
Trichiurus are characterized by the complete ab-
sence of a caudal complex.

Character 57. All outgroups and those trichiurids
with a well-developed caudal skeleton bear an ul-
timate centrum that has undergone flexion and
forms a urostyle. In Eupleurogrammus, Leptura-
canthus, Tentoriceps, and Trichiurus, the last ver-
tebral centrum has not undergone flexion. In Lep-
turacanthus and Trichiurus, the posterior tip of the
body is pointed, and the last vertebral element
bears an extremely reduced neural process and a
haemal spine that is branched distally in some spec-
imens. The last vertebral element of Eupleurogram-
mus also bears an extremely reduced neural pro-
cess. However, the haemal spine of this last verte-

bral element is spatulate and supports the more
rounded tail of this genus.

The following descriptions refer only to those tri-
chiurids and outgroups that bear a well-developed
caudal complex, unless noted otherwise.

Caudal-fin Rays

The caudal rays are divided into principal and pro-
current rays. The principal caudal rays are usually
branched and more elongate and articulate with the
upper and lower hypural plates, including the par-
hypural (Dunn, 1983). The procurrent rays are usu-
ally unbranched and articulate with the neural or
haemal elements of the preural centra. Collette et
al. (1984) tabulated the meristic differences of cau-
dal fin rays among the scombroids. All scombroids
bear a 9 + 8 pattern of principal caudal rays on the
dorsal and ventral hypural plates, respectively.
However, a slight meristic difference appears in the
numbers of procurrent rays among the major scom-
broid groups. Trichiurids are characterized by hav-
ing 6 or 7 dorsal and ventral procurrent rays. All
gempylids, except Diplospinus and Paradiplospi-
nus, have between 8 and 11 procurrent rays dor-
sally and ventrally. Diplospinus and Paradiplospi-
nus are similar to the trichiurids in that they bear
4 or 5 and 5 or 6 procurrent rays dorsally and ven-
trally, respectively. Scombrids have between 10 and
17 dorsal and 10 and 18 ventral procurrent rays,
whereas billfishes bear between 8 and 13 and 11
and 13, respectively. Although there is a reduction
in the number of procurrent rays in the trichiurids,
the character is not used in the analysis because of
the subjectivity in the determination of possible
character states. Collette and Chao (1975) noted
that in trichiurids and gempylids the bases of the
caudal-fin rays extend only partly onto the hypural
plate. In contrast, scombrids are characterized by
having caudal-fin ray bases that cover the hypural
plates almost completely.

Epurals

The epurals are median bones located between the
neural spine of preural centrum 2 and the upper
hypural plate.

Character 58. All trichiurids have a single epural.
Diplospinus and Paradiplospinus are characterized
by the presence of two epurals, whereas Gempylus
and Nesiarchus have three. Russo (1983) noted
that all gempylids, except Diplospinus and Para-
diplospinus, have three epurals. Although the con-
dition for this character in this study appears to be
equivocal at the outgroup node, I consider the ple-
siomorphic condition to be that present in most
gempylids (i.e., presence of three epurals).

Uroneurals

Uroneurals are paired bones located anterodorsally
above the urostyle. The uroneurals represent the
modified neural arches of the first preural and ural
centra.

60 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

B

Figure 24. Left lateral view of two anal-fin pterygiophores and soft rays: (A) Paradiplospinus antarcticus; (B) Aphanopus
carbo-, (C) Eupleurogrammus glossodon; (D) Tentoriceps cristatus; (E) Trichiurus lepturus.

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny M 61

Figure 25. Left lateral view of the caudal-fin skeleton (fin rays not included; cartilaginous elements striped; CPNPU3,
postneural spine cartilage of preural centrum 3; CPEP, postepural cartilages; CPHY5, posthypural cartilage of hypural 5;
CMC, median caudal cartilage; CPHPU, parhypural and haemal spine cartilages of preural centra): (A) Paradiplospinus
antarcticus; (B) Assurger anzac.

All trichiurids and the outgroups Diplospinus,
Gempylus, and Paradiplospinus have a single uro-
neural. Nesiarchus has two uroneurals. Russo
(1983) considered the fusion of the uroneurals with
the urostyle in Diplospinus, Gempylus, and Para-
diplospinus to be a synapomorphy uniting these
three genera. Fujita (1990) reported that the uro-
neural of Diplospinus is fused to the urostyle and
the upper hypural plate but concluded that the uro-
style of Gempylus is autogenous. In this study, I
also note that the uroneurals of Paradiplospinus
and all the trichiurids with a well-developed caudal
fin complex are fused, except in Apbanopus and
Benthodesmus. In these two trichiurids, the uro-
neural is closely associated with the urostyle, but a
joint between these two bones is visible. The degree

of fusion of the uroneurals to the urostyle and up-
per hypural is excluded from this analysis because
the condition appears to be quite variable and de-
pendent on the size and degree of ossification of the
specimens and is difficult to interpret.

Preural Centra

The preural centra discussed in this study include
the three posterior-most vertebral elements, exclud-
ing the urostyle. They bear well-developed haemal
and neural arches and spines. These elements are
numbered from posterior to anterior, starting an-
teriorly to the urostyle, as preural centra 2, 3, and
4. In the trichiurids and outgroups, preural centra
2 and 3 bear well-developed haemal spines that

62 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

B

C

Figure 26. Left lateral view of the caudal tip (cartilaginous elements striped): (A) Eupleurogrammus glossodon; (B)
Tentoriceps cristatus ; (C) Trichiurus lepturus.

support the ventral procurrent rays. In addition,
preural centrum 3 of the trichiurids bears a well-
developed neural spine that supports some of the
dorsal procurrent rays.

In trichiurids, the haemal spines of preural ver-
tebrae 2 and 3 are fused to their respective centra.
The haemal spines of preural centra 2 and 3 in the
outgroups are usually autogenous. However, in a
large specimen of Paradiplospinus, the haemal
spine of the preural centrum 3 appeared to be
fused. In his phylogenetic analysis of the gempylids,
Russo (1983) noted that all gempylids, except Dip-

lospinus and Paradiplospinus, bear preural centra
with fused haemal spines. He considered the pres-
ence of an autogenous haemal spine on preural cen-
trum 2 to be a synapomorphy uniting these two
genera. The degree of fusion of the haemal spines
to their respective preural centra varies with size, is
dependent on the degree of ossification of the spec-
imens, and is difficult to interpret. In scombroids
preural centrum 2 lacks a neural spine (Russo,
1983).

Character 59. In trichiurids, preural centrum 4 is
characterized by the presence of shortened haemal

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 63

and neural spines which do not extend well past
the anterior margin of preural centrum 3. The out-
groups are characterized by the presence of haemal
and neural spines on preural centrum 4 that are
long and extend past, or as far as, the posterior mar-
gin of preural centrum 3.

Urostyle

The urostyle supports the parhypural and the dor-
sal and ventral hypural plates posteriorly and the
uroneural anterodorsally. The morphology of the
urostyle is similar among the taxa analyzed in this
study.

Parhypural

The parhypural is a median bone located ventral to
the urostyle and anteroventral to the ventral hy-
pural plate. The parhypural is considered to be the
modified haemal spine of the first preural centrum.
It bears the last haemal arch that is traversed by
the dorsal aorta. The parhypural bears a longitu-
dinal lateral shelf or parhypurapophysis. Fujita
(1990) noted that the parhypural is autogenous in
all the gempylid taxa that he analyzed, except Dip-
lospinus , where the parhypural is fused to the uro-
style and the ventral hypural plate. Gempylus and
Nesiarchus share the condition described by Fujita
(1990) for the rest of the gempylids. Russo (1983)
indicated that the parhypural of Gempylus is close-
ly associated with the ventral hypural plate, but a
distinct joint can be seen between them. Fujita
(1990: table 2-20) indicated that Benthodesmus
elongatus pacificus Parin and Becker 1970 (=B. pa-
cificus) is characterized by a parhypural that is
completely fused to the ventral hypural plate. How-
ever, in his illustrations, Fujita (1990: fig. 522) de-
picted a clear border between the parhypural and
the ventral hypural plate. All of the specimens of
Diplospinus and the trichiurids analyzed in this
study share the presence of a fused parhypural. A
small specimen of Paradiplospinus was character-
ized by the presence of an autogenous parhypural,
whereas larger specimens had fused parhypurals.
Russo (1983) also indicated that a joint is present
distally between the parhypural and the ventral hy-
pural plate of Paradiplospinus. The degree of fu-
sion between the parhypural and the ventral hy-
pural plate is variable among the gempylids and
dependent on the degree of ossification of the spec-
imens.

Hypurals

The hypurals are a series of five median bones that
are subtriangular in shape and articulate or are
fused to the posterior margin of the urostyle. The
hypurals are fused into two or more units or plates
that support the bases of the principal caudal-fin
rays.

Character 60. Russo (1983) considered the pat-
tern of fusion among the hypurals as two charac-

ters. In his character 79, Russo (1983) concluded
that the presence of separate third, fourth, and fifth
hypurals represents the plesiomorphic condition;
the fusions of the third and fourth or the third,
fourth, and fifth represent the apomorphic condi-
tions. Furthermore, in his character 81, Russo
(1983) considered the separation of the first and
second hypurals as the plesiomorphic condition and
their fusion as the apomorphic condition. All tri-
chiurids, plus the outgroups Diplospinus and Par-
adiplospinus, have a ventral and a dorsal hypural
plate formed by the fusion of the first and second
and the third, fourth, and fifth hypurals, respec-
tively. In Gempylus, the dorsal and ventral hypural
plates are formed by the fusions of the first and
second and the third and fourth hypurals, respec-
tively. The fifth hypural of Gempylus is closely as-
sociated with the fourth hypural and the urostyle,
but a distinct joint is present between these bones.
In Nesiarchus, all hypurals are separate. The degree
of fusion of these elements is dependent on the size
of the specimens. Although this character should be
interpreted with caution, I include it in the analysis
as a single multistate character that combines the
conditions of the ventral and dorsal hypural plates
described by Russo (1983). The condition of this
character at the outgroup node is equivocal. How-
ever, I agree with Russo’s (1983) analysis and con-
sider the plesiomorphic condition to be that in
which all the hypurals are free (i.e., the hypural
plates formula is I + II + III + IV + V). The apo-
morphic conditions are the presence of three hy-
pural plates (i.e., I— II + II— IV + V) or two hypural
plates (i.e., I-II + III-IV-V).

Character 61. The ventral and dorsal hypural
plates of the trichiurids and the gempylids Diplo-
spinus and Paradiplospinus are fused longitudinal-
ly, except distally at their corners. Thus, the pos-
terior margin of the hypural plates forms a notch
(hypural notch of Russo, 1983). Russo (1983: char-
acter 80) concluded that the presence of a small
hypural notch represents a synapomorphy uniting
Diplospinus and Paradiplospinus. The trichiurids
share this condition. The plesiomorphic condition
is that present in the rest of the gempylids, which
bear a large hypural notch.

Cartilaginous Elements

Fujita (1990) described the cartilaginous elements
of the caudal fin complex of several scombroids,
including the trichiurid Benthodesmus elongatus
pacificus ( =B . pacificus ) and the gempylids Dip-
lospinus multistriatus, Nealotus tripes Johnson
1865, Neoepinnula orientalis (Gilchrist and von
Bonde 1924), Nesiarchus nasutus, Promethichthys
prometheus (Cuvier in Cuvier and Valenciennes
1832), and Ruvettus pretiosus Cocco 1829.

All trichiurids and outgroups bear a single post-
hypural cartilage at the posterior tip of the fifth
hypural (CPHY5 of Fujita, 1990). Trichiurids and
the outgroups, except Diplospinus and Paradiplo-

64 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

spinus, have a single post-epural cartilaginous ele-
ment (CPEP of Fujita, 1990). In trichiurids this
post-epural cartilaginous element is associated with
the single epural bone of these taxa (CPEP1). In
contrast, the single post-epural element of the out-
groups, excluding Diplospinus and Paradiplospi-
nus, is associated with the third epural bone of
these taxa (CPEP3). Diplospinus and Paradiplo-
spinus are characterized by the presence of two sep-
arate post-epural cartilages (CPEP1 and 2 of Fujita,
1990) associated with the two epurals in their cau-
dal skeleton. A cleared and stained specimen of
Paradiplospinus antarcticus in this study shows a
single post-epural cartilage. However, upon close
examination through transmitted light, one can dis-
cern the presence of two darker blue cartilaginous
centers within this single unit.

All of the trichiurids bear three posthaemal spine
cartilaginous elements distally on the parhypural
and the haemal spines of preural centra 2 and 3
(CPHPU1, 2 and 3 of Fujita, 1990). Diplospinus,
Gempylus, and Paradiplospinus share this condi-
tion with the trichiurids. Fujita (1990) noted that
Nesiarchus and all of the other gempylids, except
Diplospinus, plus the three species of scombrids in
his study have two posthaemal spine cartilages.

Character 62. All of the trichiurids with a well-
developed caudal fin have a median caudal carti-
lage (CMC of Fujita, 1990). This single cartilagi-
nous block lies slightly posterior to the hypural
notch formed by the upper and lower hypural
plates. All the outgroups lack this element.

Character 63. All of the outgroups bear a carti-
laginous postneural spine element on the third
preural vertebra (CPNPU3 of Fujita, 1990). The
third postneural spine cartilage is absent in all tri-
chiurids.

OTOLITH MORPHOLOGY

Otolith features are depicted in Figure 27. The me-
dial face of the sagittae is characterized by the pres-
ence of a longitudinal groove, the sulcus, which is
usually divided into an anterior ostium and a pos-
terior cauda. The ostium usually opens anteriorly
through the excisura, which may or may not be
bordered by a rostrum and an antirostrum. If pre-
sent, a postcaudal trough marks the opening of the
cauda. The dorsal and ventral borders of the sulcus
are delimited by longitudinal ridges called the cris-
tae superior and inferior, respectively.

The outgroup genera Gempylus and Nesiarchus
(Fig. 28B, C) have very similar otolith morpholo-
gies. The sagittae have extremely elongate rostra
and well-defined antirostra; an extremely shallow
sulcus that opens both posteriorly and anteriorly;
reduced, nonoverhanging cristae superior and in-
ferior. The only difference between the sagittae of
these two genera seems to be the presence of a
notch in the posterior margin of the otolith and a
shorter antirostrum in Nesiarchus nasutus. How-
ever, the variability of these characters cannot be

evaluated since only a single drawing of the otolith
of Nesiarchus was available for comparison.

Diplospinus, Paradiplospinus, and all the tri-
chiurids analyzed in this study, except Lepidopus
fitchi (Fig. 281), have sagittae with a sulcus that
lacks a postcaudal trough. On the other hand, the
sagittae of Gempylus and Nesiarchus show a very
shallow sulcus that is open posteriorly. Other scom-
broids analyzed (e.g., basal gempylids and billfish-
es) have sagittae with a wide postcaudal trough or
a sulcus with a wide, fan-shaped posterior margin
(e.g., some scombrids). The latter group, even
though they have a continuous posterior margin,
could still be considered as open because of the ex-
treme fan shape of the posterior margin of the cau-
da. Frost (1928) described the sagittal otoliths of
Lepidopus caudatus as having a narrow postcaudal
trough. A description and drawing of a sagitta of
L. caudatus by Demestre et al. (1993) does not
show a postcaudal trough. The 20 sagittae of this
species analyzed in this study do not have a post-
caudal trough (Fig. 28H). If the sagittae of some
specimens of L. caudatus have a postcaudal trough,
it may be very uncommon. Fitch and Gotshall
(1972) included a photograph of a sagitta of Lep-
idopus xantusi Goode and Bean 1895 (=L. fitchi)
that clearly shows a posterior opening of the sulcus.
The same condition was found in most of the 42
sagittae of this species analyzed.

Anderson and Cailliet (1975) noted that the oto-
lith morphology of their specimen of Benthodesmus
elongatus pacificus ( =B . pacificus) differs from that
described by Fitch and Gotshall (1972) in the pres-
ence of a reduced antirostrum and a more rounded
ventral surface with the greatest height occurring
along the middle third of the longitudinal axis. Al-
though they acknowledge that the specimens had
been preserved in formalin prior to examination of
the sagittae, the morphology is clearly distinct. In
Anderson and Cailliet (1975; fig. 2) it appears that
the sulcus extends along the whole medial face and
opens both posteriorly and anteriorly. The well-de-
veloped sulcus and the elongate rostrum are in con-
trast with the sagitta of B. pacificus analyzed in this
study (Fig. 28G).

Reduced ornamentation of the ventral margins of
the otoliths is characteristic of Diplospinus, Gem-
pylus, Paradiplospinus, and all the trichiurids in
this study, except Lepidopus, Leptur acanthus, and
Trichiurus. Diplospinus seems to have some slight
ornamentation on the ventral margin, but its con-
dition is not comparable to the strongly serrate and
irregular ventral margins that characterize Nesiar-
chus, Lepidopus, Leptur acanthus, Trichiurus, and
most scombroids. However, the degree of ornamen-
tation seems to be quite variable within species. For
example, large sagittae of Gempylus appear to have
little ornamentation, whereas smaller sagittae are
heavily ornamented.

Diplospinus, Paradiplospinus, and all the tri-
chiurids have sagittae with reduced ornamentation
on their dorsal margins. All other scombroid genera

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 65

POSTCAUDAL

Figure 27. Generalized diagram of the medial face of a left sagitta from a hypothetical trichiuroid.

analyzed have sagittae with strong ornamentation
in the form of irregular, serrate dorsal margins.
Some of the otoliths of Diplospinus and Paradi-
plospintis have a slightly ornamented dorsal mar-
gin. However, this is not comparable to the ple-
siomorphic condition present in Gempylus and
Nesiarcbus. Again, the ornamentation appears to
be variable through ontogeny.

A possible autapomorphy for the genus Assurger
is the presence of a tapering posterior end that
turns abruptly downward forming a posteroventral
dome (Fig. 28F), which is absent or reduced in the
other genera studied. However, this possible char-
acter is not included in the data matrix because its
variability could not be determined since only three
otoliths (i.e., two individuals) of Assurger were
available.

Character 64. Diplospinus and Paradiplospinus
(Fig. 28A, D), plus the trichiurids Aphanopus and
Benthodesmus (Fig. 28E, G), have poorly devel-
oped rostra and antirostra. Fitch and Gotshall
(1972) described the sagittae of Aphanopus as lack-
ing a rostrum. Gempylus and Nesiarchus, the rest
of the trichiurids, and the other scombroid genera
analyzed in this study have well-developed rostra
and antirostra. Although the condition for this
character at the outgroup node of this study is
equivocal, the presence of well-developed rostra
and antirostra is considered to be the ancestral con-
dition since it is present in the rest of the scom-
broids analyzed outside of the trichiurids and the
outgroups Diplospinus and Paradiplospinus. This
character, however, is variable and should be inter-
preted with caution. Gempylus and Nesiarchus
have an extremely elongate rostrum, whereas As-
surger, Lepidopus, Leptur acanthus, and Trichiurus

possess a well-developed rostrum and antirostrum
that is not nearly as elongate.

Character 65. The presence of overhanging aris-
tae inferior and superior characterizes the genera
Aphanopus, Assurger, Lepidopus, Leptur acanthus,
and Trichiurus. All the other trichiurids and out-
groups lack the presence of overhanging cristae.

Character 66. Lepturacanthus and Trichiurus
(Fig. 28J, K) share the presence of a longitudinal
ridge on the ostium. The longitudinal division of
the ostium by a ridge was already noted by Frost
(1928) in the sagittae of Trichiurus. The otoliths of
Lepturacanthus have not been previously de-
scribed. The otolith morphology of Lepturacanthus
is extremely similar to that of Trichiurus. The rest
of the trichiurids and outgroups lack a longitudinal
ridge on the ostium.

DISCUSSION

Analysis of the data matrix supports some of the
previously proposed hypotheses of relationships
within the trichiuroid fishes. Among these hypoth-
eses are: the monophyly of the trichiurids; the
monophyly of Diplospinus and Paradiplospinus
and their sister group relationship to the trichiurids;
the close relationships of the trichiurids Leptura-
canthus and Trichiurus ; and the basal position
within the cutlassfishes of Aphanopus and Bentho-
desmus.

Johnson (1986) proposed a monophyletic Tri-
chiurinae (trichiurids of this study) based on nine
synapomorphies. The present study includes four of
these nine synapomorphies and increases the sup-
port for his hypothesis by adding 17 more (Fig. 4).

The sister group relationship of Diplospinus-
Paradiplospinus to the trichiurids (Fig. 4) agrees

66 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 28. A-D. Electron microscopy photographs and line drawing (scale bar = 2 mm) of the medial face of the left
sagittae of: (A) Diplospinus multistriatus; (B) Gempylus serpens ; (C) Nesiarchus nasutus (provided by Dr. D. Nolf); (D)
Paradiplospinus antarcticus.

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 67

Figure 28. E-G. Electron microscopy photographs and line drawing (scale bar = 2 mm) of the medial face of the left
sagittae of: (E) Aphanopus arigato; (F) Assurger anzac; (G) Benthodesmus pacificus.

\

68 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Figure 28. H-K. Electron microscopy photographs and line drawing (scale bar = 2 mm) of the medial face of the left
sagittae of: (H) Lepidopus caudatus ; (I) Lepidopus fitchi; (J) Lepturacanthus savala ; (K) Tricbiurus lepturus.

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 69

with the conclusion of Parin and Becker (1972),
who also considered these gempylid genera as the
closest relatives of the trichiurids. Previous to the
work of Parin and Becker (1972), most authors in-
cluded Diplospinus and Paradiplospinus within the
Trichiuridae. Recently, Carpenter et al. (1995) pre-
sented the results of several analyses of relation-
ships among the scombroids based on morpholog-
ical characters. They analyzed a revised version of
the data matrix of Johnson (1986), a combined
data set from Collette et al. (1984) and Johnson
(1986), and a data set in which the coding of one
of Johnson’s (1986) characters was changed. In all
of their analyses, the clade including Diplospinus
and Paradiplospinus appeared as the sister group
of the trichiurids.

Diplospinus and Paradiplospinus are so similar
in their osteology that they have been considered a
single genus (Russo, 1983). In addition, Fitch and
Gotshall (1972) questioned the validity of Paradip-
lospinus, indicating that the small differences in the
sagittae of these two genera seem to be specific
rather than generic in magnitude. However, in this
study, one feature defines the species P. antarcticus :
a posteriorly directed basibranchial attachment on
the urohyal. According to Russo’s (1983) hypoth-
esis of gempylid relationships, this basibranchial at-
tachment appears to have been lost once at the an-
cestral node separating Lepidocybium from re-
maining gempylids, to be regained independently in
Epinnula, Thyrsitoides, and Thyrsitops. In addi-
tion, the distinct condition of the pelvic girdle, in
which the external elements are absent and the bas-
ipterygium is well developed, also distinguishes this
genus from the closely related Diplospinus.

The loss of external elements of the pelvic girdle
has occurred twice independently within the tri-
chiurids: in Aphanopus and the clade Lepturacan-
thus-Trichiurus. This conclusion is supported by
the ontogenetic data of Gago (1997), which also
showed that larvae of Aphanopus bear a well-de-
veloped pelvic girdle with external elements. The
condition in Aphanopus is unique because the bas-
ipterygium is extremely reduced and consists of
only a small ossification under the pectoral girdle.
Lepturacanthus and Trichiurus lack all elements of
the pelvic girdle. Remaining trichiurids have re-
duced pelvic girdles, but they still retain the exter-
nal fin elements. Among the other gempylids, not
analyzed in this study, Promethichthys, Rexichthys
Parin and Astakhov 1987, and some species of
Rexea have also lost the external pelvic-girdle ele-
ments (Nakamura and Parin, 1993). Following the
hypothesis of gempylid relationships of Russo
(1983), this loss of external pelvic-girdle elements
appears to have occurred independently at least
three times. However, Russo (1983) did not de-
scribe the pelvic girdle of gempylids. Thus, I have
no evidence about the degree of reduction of the
basipterygium among these genera.

In this study, monophyly of Lepturacanthus and

Trichiurus is defined by six synapomorphies, in-
cluding two homoplasies (Fig. 4). Tucker (1956) in-
cluded these two genera within the subfamily Tri-
chiurinae Swainson 1839. In his tree, the Trichiu-
rinae appears as emerging at the same node with
his subfamily Lepidopodinae and a monophyletic
group including Aphanopus and Benthodesmus.
Furthermore, he questioned whether the subfamily
Trichiurinae appeared as a descendant of his Nes-
iarchus-Diplospinus bridge or if it had an earlier
divergence within the Gempylidae. Tucker (1956)
indicated that, if further examination of the pala-
tine of the gempylid Thyrsitoides marleyi Fowler
1929 indicated the presence of a villiform band of
teeth, it could be considered as evidence for an ear-
lier offshoot of the Trichiurinae. Matsubara and
Iwai (1958) indicated the absence of teeth in Mi-
masea ( = Thyrsitoides ), whereas Russo (1983) de-
scribed the presence of a single row of teeth on the
ventral margin of the palatine of this species. Mat-
subara and Iwai (1958: 33) concluded that, because
the presence of a number of small teeth on the pal-
atine “is rather variable by the species, it appears
likely that the palatine teeth are not so important
[a] characteristic in considering the gempylid-tri-
chiurid relationship,” as assumed by Tucker (1956).
Although Russo (1983) indicated that replacement
teeth were present dorsomedially to the ventral row
of teeth in his specimens of Thyrsitoides marleyi,
the condition is not comparable to that in Eupleu-
rogrammus, Lepturacanthus, and Trichiurus. The
specimens of these three genera analyzed in this
study are characterized by the presence of a villi-
form band of small teeth that covers most of the
length of the palatine.

Tucker (1956) also mentioned the presence of a
concave free margin of the subopercle as a char-
acter that is present not only in Lepturacanthus and
Trichiurus but also in some gempylids such as
Epinnula and Neoepinnula. Russo (1983) conclud-
ed that the apparent concave condition of the su-
bopercle in the gempylids Nealotus, Promethichth-
ys, Rexea, and Ruvettus is not comparable to that
of the trichiurids. Furthermore, he noted that Epin-
nula and Neoepinnula have a subopercle with a
convex posteroventral margin.

The presence of a postorbital ossification appears
as a homoplastic character (character 9, Appendix).
Although the condition of the postorbital ossifica-
tion in Lepturacanthus and Trichiurus (i.e., large,
thick, and triangular) is treated as being homolo-
gous to that of the other trichiurids possessing a
postorbital ossification (i.e., poorly ossified, thin,
and irregular in shape), it is possible that the two
are not homologous and that the ossification of
Lepturacanthus and Trichiurus represents a unique,
derived condition.

Complete loss of the pelvic girdle appears to have
occurred only once within the trichiurids. Although
absence of the basipterygium (character 39, Appen-

70 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

dix) is included in the analysis, five other synapo-
morphies support the single loss of all pelvic ele-
ments at the node uniting Lepturacanthus and Tri-
chiurus (Fig. 4).

Eupleurogrammus appears as the sister group to
Lepturacanthus and Trichiurus, and the monophyly
of these three genera is supported by nine synapo-
morphies including five homoplasies (Fig. 4). Gill
(1863) included Eupleurogrammus and Trichiurus
in the subfamily Lepturinae. His genus Trichiurus
included T. savala Cuvier 1829 ( = Lepturacanthus
savala). Goode and Bean (1895) had already noted
the similarity of Eupleurogrammus to Trichiurus by
calling it “a Chinese form” of the latter genus and
including both in their own family Trichiuridae.
Tucker (1956) excluded Eupleurogrammus from
the subfamily Lepturinae of Gill (1863). He consid-
ered that the presence of a median lateral line, eth-
mofrontal elevation, pelvic fins, a rounded opercle,
and a uniseriate row of palatine teeth represents
evidence of the close relationship of Eupleurogram-
mus to the lepidopodines and thus its exclusion
from the Trichiurinae. My observations differ from
those of Tucker (1956) because I found that Eu-
pleurogrammus shares with Lepturacanthus and
Trichiurus the presence of a villiform band of teeth
covering most of the length of the palatine. Fur-
thermore, the degree of elevation of the frontal
ridges on the ethmoidal region appears to be a con-
tinuous character among the trichiurid genera As-
surger, Evoxymetopon, Tentoriceps, and some spe-
cies of Lepidopus. Although Eupleurogrammus is
characterized by a frontal bone that is elevated in
the ethmoidal region, the condition is different
from that present in Assurger, Evoxymetopon, Ten-
toriceps, and some species of Lepidopus (character
23, Appendix). Thus, the original composition of
the subfamily Trichiurinae (= Lepturinae of Gill,
1863), including Eupleurogrammus, Lepturacan-
thus, and Trichiurus, is supported as a monophy-
letic group.

Loss of the caudal fin appears to have occurred
once within the trichiurids. Monophyly of the four
genera that lack a caudal fin ( Eupleurogrammus ,
Lepturacanthus, Tentoriceps, and Trichiurus ) is
supported by 13 synapomorphies (excluding loss of
the caudal fin elements and lack of flexion in the
ultimate centrum, characters 56 and 57, respective-
ly, Appendix), of which only two are homoplasies
(Fig. 4). However, some authors have questioned
the validity of using the presence and absence of a
caudal skeleton as a phylogenetically informative
character among the trichiurids. Starks (1911)
questioned the use of the presence of a caudal fin
as a character to define the family Lepidopidae of
Goode and Bean (1895), which included Aphano-
pus, Benthodesmus, Evoxymetopon, and Lepido-
pus. Tucker (1956) placed Tentoriceps and Eupleu-
rogrammus into two different groups separate from
Lepturacanthus and Trichiurus, which also lack a
caudal fin. Further support for the monophyly of

the four ecaudate genera comes from the apparent
gradual reduction of the caudal skeleton in Tento-
riceps, Eupleurogrammus, Lepturacanthus, and
Trichiurus, in that phyletic order. Tentoriceps bears
some reduced internal elements on the ventral and
dorsal margins of the unflexed ultimate centrum.
Eupleurogrammus bears a well-developed distally
expanded haemal spine on the ultimate centrum,
whereas Lepturacanthus and Trichiurus have a
thin, pointed haemal spine. Ontogenetic data
(Gago, 1997) show that larvae of Trichiurus and
Lepturacanthus bear one or two small cartilaginous
plates ventrally at the ultimate centrum. They also
have a few short ray-like elements that barely pen-
etrate the skin. This larval condition resembles the
reduced caudal skeleton of adult Tentoriceps.

Aphanopus appears as the sister group to the rest
of the trichiurids (Fig. 4). Parin and Becker (1972)
considered Aphanopus as the earliest diverging
branch within the trichiurids. Tucker (1956) includ-
ed Aphanopus, Benthodesmus, and Diplospinus
within the subfamily Aphanopodinae. Tucker
(1956; fig. 23) placed Aphanopus as an earlier off-
shoot of the branch leading to Benthodesmus, and
Diplospinus as a branch that diverges before an ap-
parent trichotomy at the base of his three trichiurid
subfamilies.

The subfamilies Aphanopodinae and Lepidopo-
dinae of Tucker (1956) appear as paraphyletic
groups in the most parsimonious trees obtained in
this study. Tucker’s (1956) subfamily Aphanopodi-
nae included Aphanopus and Benthodesmus, and
his subfamily Lepidopodinae included Assurger,
Eupleurogrammus, Evoxymetopon, Lepidopus,
and Tentoriceps. In the most parsimonious trees ob-
tained in this study, Aphanopus, Benthodesmus,
Lepidopus caudatus-L. fitchi, L. altifrons-Evoxy-
metopon, Assurger, and Tentoriceps appear sequen-
tially and in that phyletic order as sister groups to
the monophyletic Trichiurinae of Swainson (1839),
which includes Eupleurogrammus, Lepturacanthus,
and Trichiurus (Fig. 4).

No autapomorphies were found that define the
genus Lepidopus based on the three species avail-
able for this study. Some of the data indicate Lep-
idopus to be paraphyletic. All of the three most par-
simonious trees obtained separate L. caudatus and
L. fitchi from L. altifrons. In these three trees L.
altifrons appears forming a monophyletic group
with Evoxymetopon or a trichotomy with Evoxy-
metopon and a clade that includes Assurger, Eu-
pleurogrammus, Lepturacanthus, Tentoriceps, and
Trichiurus.

Three homoplastic characters support node V
and the separation of Lepidopus altifrons and
Evoxymetopon from L. caudatus and L. fitchi (Fig.
4). These three characters are all related to the el-
evation of the ethmofrontal region (characters 21,
23, and 26; Appendix). No evidence is presented
here for the independence of these characters, and
it is possible that these three characters are not in-

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 71

dependent of each other. Furthermore, as indicated
earlier, the degree of elevation of the ethmofrontal
region and the convexity of the interorbital space
appears to be gradual between Assurger, Eupleu-
rogrammus, Evoxymetopon, Tentoriceps, and Lep-
idopus.

Monophyly of the clade Evoxymetopon-Lepi-
dopus altifrons is supported in two of the three
most parsimonious trees (Fig. 5). However, support
for this clade is weak. Only a single character re-
versal (character 36) supports the monophyly of
these two genera. If character 36 is assumed to have
evolved independently within the L. caudatus-L.
fitchi clade (node IVa) and above node V (Fig. 5A),
then no support is offered for the monophyly of
Evoxymetopon and L. altifrons.

The first published description of a specimen of
Lepidopus altifrons is that of Tucker (1957) who
identified it as Evoxymetopon taeniatus. Parin and
Collette (1993; Collette, personal communication)
also found that most of the specimens in the USNM
and MCZ collections previously identified as E.
taeniatus were actually specimens of their new spe-
cies L. altifrons.

Parin and Collette (1993) noted that the place-
ment of Lepidopus altifrons within the genus Lep-
idopus is arbitrary because it does not fit the di-
agnostic characters presented by Tucker (1956).
Parin (personal communication) has indicated that
the species within the genus Lepidopus could pos-
sibly be separated into different genera. The simi-
larities between L. altifrons and Evoxymetopon
spp. might prove to be evidence in support of such
a hypothesis. However, in this study no attempt is
made at revising the classification of the genus Lep-
idopus. A revision of the genus Lepidopus should
await the availability of more specimens of all the
Lepidopus species for inclusion in an analysis of
relationships.

In two of the three most parsimonious trees ob-
tained in this study, Lepidopus caudatus and L. fit-
chi appear as a monophyletic group. Support for
the monophyly of L. caudatus-L. fitchi is based
only on the independent acquisition of the derived
condition of character 36 (Fig. 5A).

Rosenblatt and Wilson (1987) described the east-
ern Pacific material previously referred to as Lepi-
dopus xantusi as the new species L. fitchi. They
considered L. fitchi not to be conspecific with the
original holotype of L. xantusi from Cabo San Lu-
cas. They determined that L. xantusi represents a
synonym of the western Pacific-eastern Atlantic
species L. caudatus. Such a conclusion places the
holotype of L. xantusi as the only specimen of L.
caudatus from the eastern Pacific. Furthermore,
their meristic analyses of the number of vertebrae
and anal and dorsal-fin rays for the species of Lep-
idopus shows that L. fitchi and L. caudatus have
the lowest and highest ranges for these counts, re-
spectively. The ranges and 95% confidence limits
of the means of the meristic counts do not overlap.
Even though four other species of Lepidopus oc-

cupy an intermediate position between L. fitchi and
L. caudatus, there is no indication of a geographic
morphocline. The otoliths of these two species offer
more evidence of their morphological differentia-
tion. In addition to the presence of a postcaudal
trough, the sagittae of L. fitchi have longer, better-
defined rostra and antirostra and an overall shape
that is different from that of L. caudatus. However,
this potential character was not included in the
analysis because the definition of objective charac-
ters states with regard to the degree of extension of
the rostrum and antirostrum is difficult.

Parin and Collette (1993) noted that the presence
of a convex interorbital space and a sagittal crest
that extends onto the ethmoidal region are char-
acters that seem to change in a gradual manner in
the series Lepidopus manis and L. fitchi, L. cau-
datus and L. calcar, L. dubius, and L. altifrons. As-
surger, Evoxymetopon, and Tentoriceps could be
added to this series as the more derived conditions.
In this study, Tentoriceps is clearly separated from
Assurger, Evoxymetopon, and Lepidopus by 15
synapomorphies that include it in a clade with the
other three ecaudate trichiurid genera. Assurger is
also separated from Evoxymetopon and Lepidopus
and appears as the sister group to the ecaudate tri-
chiurids (Fig. 4, node VI). However, node VI is only
supported by a single synapomorphy: the anterior
neural spines are distally expanded and forked.

Ontogenetic data (Gago, 1997) have provided
evidence that supports several of the arguments of
character evolution proposed in this study (Fig. 29).
Monophyly of the trichiurids is strongly supported
by both the adult and larval data. Furthermore,
both data sets also support the monophyly and the
same phyletic sequence of the four ecaudate tri-
chiurid genera ( Tentoriceps , Eupleurogrammus,
Leptur acanthus, and Trichiurus). The adult data
place Aphanopus, Benthodesmus, the Lepidopus
fitchi-L. caudatus clade, the Lepidopus altifrons-
Evoxymetopon clade, and Assurger as the sister
groups, in that phyletic sequence, to the ecaudate
trichiurids. The larval data (Gago, 1997) also sup-
port the position of Aphanopus as the most basal
trichiurid, but the ontogenetic and adult characters
are incompatible with respect to the phyletic place-
ment of Assurger, Benthodesmus, and Lepidopus.

Johnson (1993) noted that specialized larval
characters and the patterns of chondrification and
ossification in ontogenetic series can be helpful in
phylogenetic studies by providing evidence for
monophyly and intrarelationships in studies at the
family level and below, and by testing hypotheses
of homology in adult characters.

In the larval data (Fig. 29), Assurger appears as
the sister group to the clade that includes Bentho-
desmus and Lepidopus. In contrast, in the adult
phytogeny (Fig. 4), Assurger appears as the sister
group of the ecaudate trichiurids, and Benthodes-
mus appears as the sister group to all the trichiur-
ids, except Aphanopus. Placement of Benthodes-

72 H Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

100

Trichiurus

Lepturacanthus

Eupleurogrammus

Tentoriceps

Benthodesmus

Lepidopus

Assurger

Aphanopus

Thyrsites

Thyrsitops

Gempylus

Nesiarchus

Promethichthys

Ruvettus

Diplospinus

Epinnula

Paradiplospinus

Neoepinnula

Rexea

Nealotus

Lepidocybium

Figure 29. Fifty percent majority rule consensus tree of the 87 equally most parsimonious hypotheses obtained by the
branch-and-bound analysis of the larval data of Gago (1997). Numbers indicate the percentage of resulting trees that
support those nodes.

mus as the sister group to all trichiurids, except
Aphanopus, is supported in the adult data matrix
by six synapomorphies. In contrast, the monophyly
of Benthodesmus and Lepidopus in the larval phy-
logeny is only supported by two synapomorphies
(i.e., the longitudinal ventral and anterior keels of
the pelvic and dorsal spines, respectively, are
smooth). Placement of Assurger as the sister group
to the ecaudate trichiurids in the adult phylogeny
(Fig. 4) is supported by one synapomorphy. The
sister group relationships of Assurger to the clade
of Benthodesmus-Lepidopus in the larval data is
also supported by a single synapomorphy (i.e.,
elongate first dorsal spine). However, the elonga-
tion of the first dorsal spine is a subjective character
since the degree of elongation among these genera
is variable. The larvae of Lepidopus have a first
dorsal-fin spine that is more than twice the length
of the other dorsal-fin spines. In contrast, the first
dorsal-fin spine of Assurger is only slightly longer

than the other dorsal-fin spines. Furthermore, the
condition in Benthodesmus has only been reported
for the larvae of B. pacificus (Ozawa, 1986). Vari-
ation in this character is difficult to evaluate be-
cause these elongate spines are usually broken dur-
ing collection.

In the hypothesis of Johnson (1986), the nine
genera of cutlassfishes are grouped in the subfamily
Trichiurinae within the family Gempylidae, a group
that includes the monophyletic Lepidocybiinae and
the apparently paraphyletic Gempylinae. Johnson
(1986) considered his Trichiurinae as a highly de-
rived branch of the gempylids. Although I also con-
sider the trichiurids as being closely related to the
gempylids, the development of a formal classifica-
tion for the cutlassfishes, at this point in time, is
unwarranted. I agree with Carpenter et al. (1995)
that a formal classification for the gempylids and
trichiurids (superfamily Trichiuroidea) must await
a combined analysis of both groups.

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 73

ACKNOWLEDGMENTS

This work benefited greatly from the collaboration and
support of many individuals and institutions. The follow-
ing individuals provided curatorial assistance and loans or
gifts of specimens: B.B. Collette, R.F. Feeney, S.L. Jewett,
G.D. Johnson, C. Klepadlo, R.J. Lavenberg, M. Mc-
Grouther, J. Nielsen, D. Noll, L. Palmer, N.V. Parin, J.R.
Paxton, W.J. Richards, R.H. Rosenblatt, J. Seigel, J. Tyler,
and H.J. Walker. Donald McNamee and Mark Herbert
provided extensive help during my library search. Special
thanks go to the past and present staff in Ichthyology at
the Natural History Museum of Los Angeles County for
their continuous help during the study: Dan Cohen, Ri-
chard Feeney, Guillermo Herrera, Robert Lavenberg, Ger-
ald McGowen, Margaret Neighbors, Debra Oda, Mason
Posner, Brenda and James Rounds, Helga Schwarz, Jeffrey
Seigel, Camm Swift, and Brian White. This work was
greatly improved by discussions with R. Bezy, S.L. Bower,
D.M. Cohen, B.B. Collette, R.F. Feeney, G.A. Herrera,
G.D. Johnson, R.J. Lavenberg, M. McFall-Ngai, G.E.
McGowen, B.G. Nafpaktitis, M. Posner, J. Seigel, J.D.
Stewart, B.N. White, and R.L. Zimmer. For comments on
portions of earlier drafts of this manuscript, I thank D.M.
Cohen, B.B. Collette, R.F. Feeney, G.A. Herrera, G.D.
Johnson, R.J. Lavenberg, M.A. Neighbors, and M. Posner.
Special thanks go to Bruce Collette, Robert Lavenberg,
and Basil Nafpaktitis for their continuous support and
guidance. Any errors of fact or interpretation are my own
and in no way reflect the views of those who generously
aided in this study. I thank the American Society of Ich-
thyologists and Herpetologists (Raney Award), the Uni-
versity of Southern California (Department of Biological
Sciences and The Trojan League Award), and the Natural
History Museum of Los Angeles County for their support
during this study.

LITERATURE CITED

Alexander, W.B. 1916. History of zoology in western Aus-
tralia. Part II. Journal of the Royal Society of West-
ern Australia 1:83-149.

Allis, E.P., Jr. 1903. The skull, and the cranial and first
spinal muscles and nerves in Scomber scomber. Jour-
nal of Morphology 1 8( 1— 2):45— 328.

Anderson, M.E., and G.M. Cailliet. 1975. Occurrence of
the rare north Pacific frostfish, Benthodesmus elon-
gatus pacificus Parin and Becker, 1970, in Monterey
Bay, California. California Fish and Game 61(3):
149-152.

Andriashev, A.P. 1960. Families of fishes new to the Ant-
arctic. 1. Paradiplospinus antarcticus gen. et sp. n.
(Pisces, Trichiuridae). Zoologicheskii Zhurnal 39(2):
244-249.

Barnard, K.H. 1948. Further notes on South African ma-
rine fishes. Annals of the South African Museum 36:
341-406.

Bleeker, P. 1860. Dertiende bijdrage tot de kennis der vis-
chrfauna van Borneo. Acta Societatis Scientiarum
Indo-Neerlandicae. 8(4):l-64.

Block, B.A. 1991. Evolutionary novelties: How fish have
built a heater out of muscle. American Zoologist 31:
726-742.

Block, B. A., J.R. Finnerty, A.F.R. Stewart, and J. Kidd.
1993. Evolution of endothermy in fish: Mapping
physiological traits on molecular phylogeny. Science
260(9 April) :2 10-2 14.

Boulenger, G.A. 1904. Teleostei. In Fishes, ed. S.F. Harmer

and A.E. Shipley, Cambridge Natural History 7:539-
727 . London: MacMillan and Co., Ltd.

Brauer, A. 1906. Die Tiefseefische. 1. Systematisther Teil.
Wissenschaftliche Ergebnisse der Deutschen Tiefsee-
Expedition auf dem Dampfer “ Valdavia ” 1898-
1899 15(1):1 — 432.

Capello, F. de B. 1868. Catalogo dos peixes de Portugal
que existen no Museu de Lisboa. Journal of Science,
Mathematics, Physics and Nature 1:233-264.

Carpenter, K.E., B.B. Collette, and J.L. Russo. 1995. Un-
stable and stable classifications of scombroid fishes.
Bulletin of Marine Science 56(2):379— 405.

Chaine, J., and J. Duvergier. 1934. Recherches sur les oto-
lithes des poissons etude descriptive et comparative
de la sagitta des teleostens. Actes de la Societe Lin-
neenne de Bordeaux 86:5-254.

Cocco, A. 1829. Su di alcuni nuovi pesci del mar di Mes-
sina. Giornale di scienze, lettere ed arti per la Sicilia
26(77):138— 147.

Collett, R. 1887 Aphanopus minor, en ny dybvandsfisk af
Trichiuridernes familie fra Gronland. Forhandlinger
i Videnskabs-selskabet i Christiana 19:1-7.

Collette, B.B., and L. Chao. 1975. Systematics and mor-
phology of the bonitos ( Sarda ) and their relatives
(Scombridae, Sardini). Fishery Bulletin, U.S. 73(3):
516-625.

Collette, B.B., and J.L. Russo. 1984. Morphology, system-
atics, and biology of the Spanish mackerels ( Scorn -
beromorus, Scombridae). Fishery Bulletin, U.S.
82(4):545-692.

Collette, B.B., and J.L. Russo. 1986. Systematic status of
the suborder Scombroidei. In Proceedings of the Sec-
ond International Conference on Indo-Pacific Fishes,
ed. T. Uyeno, R. Arai, T. Taniuchi, and K. Matsuura,
Abstract, p. 938. Tokyo: The Ichthyological Society
of Japan.

Collette, B.B., T. Potthoff, W.J. Richards, S. Ueyanagi, J.L.
Russo, and Y. Nishikawa. 1984. Scombroidei: De-
velopment and relationships. In Ontogeny and sys-
tematics of fishes, ed. H. G. Moser et al. American
Society of Ichthyologists and Herpetologists Special
Publication 1:591-620.

Coombs, S., J. Janssen, and J.F. Webb. 1987. Diversity of
lateral line systems: Evolutionary and functional
considerations. In Sensory biology of aquatic ani-
mals, ed. J. Atema, R.R. Fay, A.N. Popper, and W.N.
Tavolga, 553-593. New York: Springer- Verlag.

Cuvier, G. 1829. Ee regne animal distribue d’apres son
organisation, pour servir de base a d’histoire natu-
relle des animaux et P introduction a Panatomie com-
paree, vol. 2, 2nd edition. Paris: Deterville, 532 pp.

Cuvier, G., and A. Valenciennes. 1832. Histoire naturelle
des poissons, vol. VIII. Paris-Strasbourg: F.G. Lev-
rault, 509 pp.

Demestre, M., B. Moli, L. Recasens, and P. Sanchez. 1993.
Life history and fishery of Lepidopus caudatus (Pi-
sces: Trichiuridae) in the Catalan Sea (northwestern
Mediterranean). Marine Biology 115:23-32.

Dunn, J.R. 1983. The utility of developmental osteology
in taxonomic and systematic studies of teleost larvae:
A review. NO A A Technical Report NMFS Circular
450:1-19.

Euphrasen, B.A. 1788. Beskrifning pa trenne fiskar. Kon-
gliga Swenska Wetenskaps Academiens Handligar
Stockholm 9:51-55.

Fink, W.L. 1981. Ontogeny and phylogeny of tooth at-
tachment modes in actinopterygian fishes. Journal of
Morphology 167:167-184.

74 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Finnerty, J.R., and B.A. Block. 1995. Evolution of cyto-
chrome b in the Scombroidei (Teleostei): Molecular
insights into billfish (Istiophoridae and Xiphiidae) re-
lationships. Fishery Bulletin, U.S. 93:78-96.

Fitch, J.E., and D.W. Gotshall. 1972. First record of the
black scabbardfish, Aphanopus carbo, from the Pa-
cific Ocean with notes on other Californian trichiur-
id fishes. Bulletin of the Southern California Acade-
my of Sciences 71(1): 12-1 8.

Fowler, H.W. 1905. New, rare, or little known scom-
broids, I. Proceedings of the Academy of Natural
Sciences of Philadelphia (for 1904):757-771.

Fowler, H.W. 1929. New and little known fishes from the
Natal Coast. Annals of the Natal Museum 6(2):245-
264.

Frost, G.A. 1928. A comparative study of the otoliths of
the neopterygian fishes (continued). XVIII. Perco-
morphi (continued). Annals and Magazine of Natu-
ral History 10(2):328-331.

Fujita, K. 1990. The caudal skeleton of teleostean fishes.
Tokyo: Tokai University Press, 897 pp.

Gago, F.J. 1997. Character evolution and phylogeny of the
cutlassfishes: An ontogenetic perspective (Scombro-
idei: Trichiuridae). Bulletin of Marine Science 60(1):
161-191.

Garman, S. 1899. The fishes. In Reports on an exploration
off the west coast of Mexico, Central and South
America, and off the Galapagos Islands in charge of
Alexander Agassiz, by the U.S. Fish Commission
Steamer “ Albatross ” during 1891, Lieut. -Command-
er Z.L. Tanner, U.S. A. commanding. No. XXVI.
Memoirs of the Museum of Comparative Zoology of
Harvard 24:1-431.

Gilchrist, J.D.F. and C. von Bonde. 1924. Deep-sea fishes
procured by the S.S. “Pickle” (Part II). Fisheries and
Marine Biological Survey Report of South Africa
number 3 for the year 1922. 7(Special report):l-24.

Gill, T. 1862. On the limits and arrangement of the family
of scombroids. Proceedings of the Academy of Nat-
ural Sciences of Philadelphia 14(3— 4):124— 127.

Gill, T. 1863. Synopsis of the family of lepturoids, and
description of a remarkable new generic type. Pro-
ceedings of the Academy of Natural Sciences of Phil-
adelphia 1863:224-229.

Gill, T.N. 1893. A comparison of antipodal faunas. Mem-
oirs of the National Academy of Sciences 6:91-124.

Goode, G.B., and T.H. Bean. 1882. Benthodesmus a new
genus of deep-sea fishes allied to Lepidopus. Pro-
ceedings of the United States National Museum
4(241):379-383.

Goode, G.B., and T.H. Bean. 1895. Oceanic ichthyology,
a treatise on the deep-sea and pelagic fishes of the
world, based chiefly upon the collections made by
the steamers “Blake”, “Albatross” and “Fish Hawk”
in the northwestern Atlantic. Special Bulletin of the
United States National Museum 1895:1-553.

Goodrich, E.S. 1909. Vertebrata Craniata (First Fascicle:
Cyclostomes and Fishes). In A treatise of zoology,
vol. IX, ed. R. Lankester. London: Adam and
Charles Black, 518 pp.

Goiian, A. 1770. Historia piscium, sistens ipsorum ana-
tomen externam, internam, atque genera in classes
et ordines redacta. Strasbourg: A. Konig, 252 pp.,
pis. 1-4.

Gray, J.E. 1831. Description of twelve new genera offish,
discovered by Gen. Hardwicke, in India, the greater
part in the British Museum. The Zoological Miscel-
lany 1831:7-10.

Gregory, W.K. 1933. Fish skulls: a study of the evolution
of natural mechanisms. (Reissued) Laurel, Florida: E.
Lundberg (1959), 481 pp.

Gunther, A. 1860. Catalogue of the Acanthopterygian
fishes in the collection of the British Museum,
Squamipinnes, Cirrhitidae, Triglidae, Trachinidae,
Sciaenidae, Polynemidae, Sphyraenidae, Trichiuri-
dae, Scombridae, Carangidae, Xiphiidae. London:
Taylor and Francis, 548 pp.

Gunther, A. 1877. Preliminary notes on new fishes col-
lected in Japan during the expedition of H.M.S.
“Challenger”). Annals and Magazine of Natural His-
tory 4(20):433-446.

Gupta, M.V. 1966. Two new species of ribbon fishes of
the genus Trichiurus Linnaeus (Pisces: Trichiuridae)
from the Hooghly estuarine system. Proceedings of
the Zoological Society of Calcutta 19:169-171.

James, P.S.B.R. 1960. Instances of excessive thickening of
certain bones in the ribbon fish, Trichiurus lepturus
Linnaeus. Journal of the Marine Biological Associa-
tion of India 2 (2):253-258.

James, P.S.B.R. 1961. Comparative osteology of the rib-
bonfishes of the family Trichiuridae from Indian wa-
ters, with remarks on their phylogeny. Journal of the
Marine Biological Association of India 3:215-248.

Johnson, G.D. 1986. Scombroid phylogeny: an alternative
hypothesis. Bulletin of Marine Science 39( 1 ): 1 — 41 .

Johnson, G.D. 1993. Percomorph phylogeny: progress
and problems. Bulletin of Marine Science 52(1 ):3-
28.

Johnson, J.Y. 1862. Descriptions of some new genera and
species of fishes obtained at Madeira. Proceedings of
the Zoological Society of London 1862(2):167-180.

Johnson, J.Y. 1865. Descriptions of a new genus of tri-
chiuroid fishes obtained at Madeira ( Nealotus
tripes ), with remarks on the genus Dicrotus Gunther,
and on some allied genera of Trichiuridae. Proceed-
ings of the Zoological Society of London 1865(2):
434-437.

Jollie, M. 1986. A primer of bone names for the under-
standing of the actinopterygian head and pectoral
girdle skeletons. Canadian Journal of Zoology 64:
365-379.

Jordan, D.S. and C.H. Gilbert. 1882. Synopsis of the fishes
of North America. Bulletin of the United States Na-
tional Museum 16:1-1018.

Kamohara, T. 1936. Supplementary note on the fishes col-
lected in the vicinity of Koci-shi (x). Zoological Mag-
azine (Japan) 48(ll):929-935.

Kluge, A.G., and J.S. Farris. 1969. Quantitative phyletics
and the evolution of anurans. Systematic Zoology
18:1-32.

Klunzinger, C.B. 1884. Die Fische des Rothen Meeres.
Eine Kritische Revision mit Bestimmungstabellen.
Teil I . Acanthopteri veri Owen. Stuttgart: Schweiz-
erbart’sche Verlagshandlung, 133 pp.

Kusaka, T. 1974. The urohyal of fishes. Tokyo: University
of Tokyo Press, 320 pp.

Lacepede, B. 1798-1803. Histoire naturelle des poissons,
vols. 1-5. Paris: Plassan.

Leviton, A.E., R.H. Gibbs, Jr., E. Head, and C.E. Dawson.
1985. Standards in herpetology and ichthyology:
Part 1. Standard symbolic codes for institutional re-
source collections in herpetology and ichthyology.
Copeia 1985:802-832.

Linnaeus, C. 1758. Systema Naturae: Holmiae, vol. 1,
10th edition. 824 pp.

Lowe, R.T. 1838. Piscium Maderensium species quaedam

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 75

novae, vel minus rite cognitae breviter descriptae,
etc. Transactions of the Cambridge Philosophical So-
ciety 6:195-202.

Lowe, R.T. 1839. A supplement to a synopsis of the fishes
of Madeira. Proceedings of the Zoological Society of
London 7:76-92.

Maddison, W.P., and D.R. Maddison. 1992. MacClade.
Analysis of phytogeny and character evolution, ver.
3. Massachusetts: Sinauer Associates Inc.

Maddison, W.P., M.J. Donoghue, and D.R. Maddison.
1984. Outgroup analysis and parsimony. Systematic
Zoology 33(1 ):83— 103.

Matsubara, K., and T. Iwai. 1952. Studies on some Jap-
anese fishes of the family Gempylidae. Pacific Science
6(3): 1 93—212.

Matsubara, K., and T. Iwai. 1958. Anatomy and relation-
ships of the Japanese fishes of the family Gempyli-
dae. Memoirs of the College of Agriculture, Kyoto
University, Fisheries Series. Special Number June:23-
54.

Maul, G.E. 1948. Quatro peixes novos dos mares da Ma-
deira. Boletim do Museu Municipal do Funchal 3(6):
41-55.

Maul, G.E. 1953. Rediscovery of the trichiurid fish Ben-
thodesmus simonyi (Steindachner) off Madeira. Pro-
ceedings of the Zoological Society of London
123(1):167— 170.

McAllister, D.E. 1968. Evolution of branchiostegals and
classification of Teleostome fishes. National Museum
of Canada Bulletin 221:1-239.

Nakamura, I., and E. Fujii. 1983. A new genus and species
of Gempylidae (Pisces: Perciformes) from Tonga
Ridge. Seto Marine Biological Laboratory 27(4/6):
173-191.

Nakamura, I., and N.V. Parin. 1993. Snake mackerels and
cutlassfishes of the world (families Gempylidae and
Trichiuridae). An annotated and illustrated cata-
logue of the snake mackerels, snoeks, escolars, gem-
fishes, sackfishes, domine, oilfish, cutlassfishes, scab-
bardfishes, hairtails and frostfishes known to date.
In FAO Fisheries Synopsis, Techical editors: K.E.
Carpenter, A.L. Agnalt, and C. Sommer. FAO Spe-
cies Catalogue, vol. 15, no. 125, 1-136.

Nakamura, K., and H. Yamaguchi. 1991. Distribution of
scleral ossicles in teleost fishes. Memoirs of the Fac-
ulty of Fisheries, Kagoshima University 40:1-20.

Ozawa, T. 1986. The larvae of the family Trichiuridae in
the ocean of southern Japan. In Studies on the oce-
anic ichthyoplankton in the western north Pacific,
ed. T. Ozawa, 289-300. Fukuoka, Japan: Kyushu
University Press.

Parin, N.V. 1995. Three species and new records of cutlass
fishes of the genus Aphanopus (Trichiuridae). Jour-
nal of Ichthyology 35(2): 128—1 38.

Parin, N.V., and D.A. Astakhov. 1987. Rexichthys john-
paxtoni — A new fish of the family Gempylidae from
the Tasman Sea. Voprosy Ikhtiologii 27(1): 149-151.

Parin, N.V., and V.E. Becker. 1970. Materials for a revi-
sion of the trichiuroid fishes of the genus Bentho-
desmus, with the description of four new species and
one new subspecies. Proceedings of the Biological
Society of Washington 83 (33 ):351— 364.

Parin, N.V., and V.E. Becker. 1972. Materials on taxon-
omy and distribution of some trichiuroid fishes (Pi-
sces, Trichiuroidae: Scombrolabracidae, Gempyli-
dae, Trichiuridae). Transactions of the P.P. Shirshov
Institute of Oceanology 93:110-204.

Parin, N.V., and B.B. Collette. 1993. Results of the re-

search cruises of FRV “Walther Herwig” to South
America. LXIX. Lepidopus altifrons, a new species
of cutlassfish (Pisces, Scombroidei, Trichiuridae)
from the western Atlantic Ocean. Archiv fur Fis-
chereiwissenschaft 41 (3): 1 87— 1 95.

Parin, N.V., and S.V. Mikhailin. 1981. A new cutlassfish,
Lepidopus dubius Parin and Mikhailin (Trichiuri-
dae), from the eastern tropical Atlantic. Voprosy
Ikhtiologii 21(3):403-410.

Parin, N.V., and S.V. Mikhailin. 1982. Lepidopus calcar,
a new trichiurid fish from the Hawaiian underwater
ridge. Japanese Journal of Ichthyology 29( 1 ):27— 29.

Patterson, C., and G.D. Johnson. 1995. The intermuscular
bones and ligaments of teleostean fishes. Smithsoni-
an Contributions to Zoology 559:1-85.

Poey, F. 1854. Nuevo genero de peces escombrideos, Epin-
nula magistralis Poey. In Memorias sobre la historia
natural de la isla de Cuba, acompahadas de sumarios
latinos y extractos en frances Habana, Impresara de
Barcina, 281-463.

Potthoff, T. 1980. Development and structure of fins and
fin supports in dolphin fishes Coryphaena hippurus
and Coryphaena equiselis (Coryphaenidae). Fishery
Bulletin, U.S. 78(2):277-3 12.

Potthoff, T. 1984. Clearing and staining techniques. In
Ontogeny and systematics of fishes, special publica-
tion 1, ed. H.G. Moser et al., 35-37. American So-
ciety of Ichthyologists and Herpetologists. Lawrence.

Potthoff, T., S. Kelley, and J.C. Javech. 1986. Cartilage
and bone development in scombroid fishes. Fishery
Bulletin, U.S. 84(3):647-678.

Rafinesque, C.S. 1810. Caratteri di alcuni nuovi generi e
nuove specie di animali e piante della Sicilia, con
varie osservazioni sopra I medesimi. Palermo: San-
filippo, 105 pp.

Ramsay, E.P., and J.D. Ogilby. 1887. Notes on the genera
of Australian fishes. Proceedings of the Linnean So-
ciety of New South Wales 2(2): 1 8 1-1 84, 561-564.

Regan, C.T. 1909. On the anatomy and classification of
the scombroid fishes. Annals and Magazine of Nat-
ural History, ser. 8, 3:66-75.

Rose, A.F. 1793. Petri Artedi Angermannia-Sueci synon-
ymia nominum piscium fere omnium . . . Ichthyol-
ogiae, pars IV. Greifswald, 2nd edition. Grypeswal-
diae: Ant. Ferdin Rose, 140 pp.

Rosenblatt, R.H., and R.R. Wilson, Jr. 1987. Cutlassfishes
of the genus Lepidopus (Trichiuridae), with two new
eastern Pacific species. Japanese Journal of Ichthy-
ology 33(4):342-351.

Roule, L. 1921. Sur un nouveau poisson abyssal ( Scorn -
brolabrax heterolepis, nov. gen. nov. sp.) peche dans
les eaux de Pile Madere. Comptes Rendu Hebdo-
madaires des Seances de PAcademie des Sciences
172(24): 1534-1536.

Russo, J.L. 1983. Interrelationships of the gempylid fishes
(Teleostei, Scombroidei). Unpublished Ph.D. Disser-
tation, The George Washington University, Washing-
ton, D.C., 249 pp., 53 figs., 1 table.

Senta, T. 1975. Redescription of the trichiurid fish Ten-
toriceps cristatus and its occurrence in the South
China Sea and the Straits of Malacca. Japanese Jour-
nal of Ichthyology 21 (4):175— 1 82.

Smith, A. 1849. Illustrations of the zoology of South Af-
rica, consisting chiefly of figures and descriptions of
the objects of natural history collected during an ex-
pedition into the interior of South Africa in 1834-
1836, vol. 4. London: Smith, Elder and Co., 77 pp.

Smith-Vaniz, W.F., L.S. Kaufman, and J. Glowacki. 1995.

76 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Species-specific patterns of hyperostosis in marine
teleost fishes. Marine Biology 121:573-580.

Soot-Ryen, T. 1936. Aphanopus Schmidti Saemundsson. A
fish new to the Norwegian fauna. Nytt Magasin for
Naturvidenskapene 76:237-243.

South, J.F. 1845. Thunnus. In Encyclopedia Metropoli-
tana, vol. 25, ed. Smedley et al. 620-622. London:
B. Fellows.

Starks, E.C. 1911. The osteological characters of the
scombroid fishes of the families Gempylidae, Lepi-
dopidae and Trichiuridae. Leland Stanford Junior
University Publications, University Series 5:17-27.

Steindachner, F. 1891. Ichthyologische Beitrage. XV. Uber
einige seltene und neue Fischarten aus dem canar-
ischen Archipel. Sitzungsberichte der Kaiserlicben
Akademic der Wissenschaften in Wien 10(1):343-
374.

Stiassny, M.L., and J.A. Moore. 1992. A review of the
pelvic girdle of acanthomorph fishes, with comments
on hypotheses of acanthomorph intrarelationships.
Zoological Journal of the Linnean Society 104:209-
242.

Swainson, W. 1839. The natural history and classification
of fishes, amphibians and reptiles, or monocardian
animals, vol. 2. London: Longman, Orme, Brown,
Green and Longmans and John Taylor, 448 pp.

Swofford, D.L. 1993. PAUP: Phylogenetic Analysis Using
Parsimony, ver. 3.1.1. Champaign, Illinois: Comput-
er program distributed by the Illinois Natural His-
tory Survey.

Temminck, C.J., and H. Schlegel. 1850. Pisces. In Fauna
Japonica, sive descriptio animalium quae in itinere

per Japoniam suscepto annis 1823-30 collegit, notis
observationibus et adumbrationibus illustravit, ed.
P.F. de Siebold, 270-324. Lugduni Batavorum: A.
Arnz et Socios.

Tucker, D.W. 1953. The fishes of the genus Benthodesmus
(family Trichiuridae). Proceedings of the Zoological
Society of London 4(1):171— 197.

Tucker, D.W. 1956. Studies on the trichiuroid fishes — 3.
A preliminary revision of the family Trichiuridae.
Bulletin of the British Museum (Natural History)
Zoology 4(3):6-131.

Tucker, D.W. 1957. Studies on the trichiuroid fishes — IV.
A specimen of Evoxymetopon taeniatus (Poey) Gill,
from the Gulf of Mexico. Annals and Magazine of
Natural History, ser. 12, 10(114):425-428.

Waite, E.R. 1911. Additions of the fauna of New Zealand.
No. II. Proceedings of the New Zealand Institute
43(2):9— 5 1 .

Whitley, G.P. 1933. Studies in ichthyology, no. 7. Records
of the Australian Museum 19(1 ):60— 1 12.

Whitley, G.P. 1948. Studies in ichthyology, no. 13. Re-
cords of the Australian Museum 22(l):70-94.

Wiley, E.O., D. Siegel-Causey, D.R. Brooks, and V.A.
Funk. 1991. The compleat cladist. A primer of phy-
logenetic procedures. Lawrence, Kansas: University
of Kansas Museum of Natural History, Special Pub-
lication 19, 158 pp.

Ye, Y., and A. A. Rosenberg. 1991. A study of the dynam-
ics and management of the hairtail fishery, Trichiurus
haumela, in the East China Sea. Aquatic Living Re-
sources 4:65-75.

Received 16 June 1997; accepted 1 September 1998.

APPENDIX

List of Characters

Characters are grouped together according to structural
units or types (e.g., neurocranium). The numbers refer to
the character number as shown in Table 1. Character
states are within parentheses.

OPERCULAR SERIES

1. Posterior and ventral margins of opercle, subopercle,
and interopercle. Complete; only the dorsal flap of the
opercle and the posterior corner of the subopercle
may be slightly splintered (0). Strongly splintered or
fimbriate (1).

2. Lateral process on articular head of the opercle. Plate-
like and covering most of the articular fossa of the
bone (0). Elongate and flat in cross section (1). Elon-
gate and round in cross section (2).

3. Fimbriations on postero ventral corner of the subop-
ercle. Not much longer than the preceding ventral
ones, giving a convex appearance to the ventral mar-
gin of the bone (0). Much more elongate than the
preceding ventral ones, giving a slightly concave ap-
pearance to the ventral margin of the bone (1).

4. Articular process of subopercle. Extends dorsally at a
right angle, articulating mainly with the anteroventral
corner of the opercle (0). Extends anteriorly and ar-
ticulates mainly with the posterodorsal corner of the
interopercle (1).

5. Posterodorsal margin of preopercle. Convex (0). Con-
cave (1).

CIRCUMORBITAL SERIES

6. Circumorbital series. Complete (0). Incomplete with
a short gap between the first infraorbital elements and

the postorbitals (1). Extremely reduced; only the lach-
rymal and jugal are present (2).

7. Shape of ventral wing of lachrymal. Ovoid; fimbria-
tions in the anterior and posterior margins of the ven-
tral wing are not perpendicular to the dorsal, longi-
tudinal, lateral-line canal (0). Quadrilateral; fimbria-
tions in the anterior and posterior margins of the ven-
tral wing are nearly perpendicular to the dorsal,
longitudinal, lateral-line canal (1).

8. Posterodorsal plate of lachrymal. Short; ends before
or above the posterior pore of the longitudinal, dorsal
lateral-line canal (0). Elongate; extends past the pos-
terior pore of the longitudinal, dorsal lateral-line ca-
nal ending above the jugal (1).

9. Postorbital ossification. Absent (0). Reduced and thin
(1). Large and thick (2).

JAWS

10. Dentary and premaxillary fangs and serial teeth.
Smooth (0). Serrate (1).

SUSPENSORIUM

11. Posterodorsal process of quadrate. Long; extends well
past the ventral margin of the metapterygoid (0).
Short; does not extend well past the ventral margin
of the metapterygoid (1).

12. Posteromedial arm of ectopterygoid. Does not reach
the metapterygoid (0). Reaches the metapterygoid (1).

13. Well-developed condyle posterior to anteromedial

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 77

shelf of palatine. Absent (0). Present and visible in
lateral view (1).

14. Teeth on ventral margin of palatine. Large and few;
a single row covering only a very small portion of the
ventral margin of the bone (0). Small and numerous;
in several rows covering most of the ventral margin
of the bone (1).

HYOID COMPLEX

15. Anterodorsal corner of ceratohyal. Not pointed; does
not extend anteriorly (0). Pointed; extends anteriorly
abutting the dorsal margin of the dorsal hypohyal (1).

16. Shape of postero ventral processes of glossohyal in
ventral view. Neither triangular nor quadrilateral; do
not have their lateral corners pointing posteriorly (0).
Wing-like, with corners pointed posteriorly (1).
Quadrilateral (2). Triangular (3).

BRANCHIAL COMPLEX

17. Shape of articular head of first basibranchial. Tapers
to a point; not knob-like (0). Knob-like, but not bear-
ing dorsolateral processes (1). Knob-like and bearing
dorsolateral processes (2).

18. Laterally pointed processes on second basibranchial.
Absent (0). Present (1).

19. Shape of fourth ceratobranchial. Straight (0). Sigmoi-
dal (1).

NEUROCRANIUM
Ethmoidal Region

20. Anterior tip of nasal. Straight (0). Curved; giving the
bone a concave lateral margin in dorsal view (1).

21. Dorsal ridges on ethmoid. Absent or reduced; do not
extend well above the nasals in lateral view (0). Pre-
sent; extend well above the nasals in lateral view (1).

22. Connection of supraorbital and infraorbital lateral-
line canals. Supraorbital lateral-line canal connects
with the infraorbital canal through a dorsally or lat-
erally directed pore (0). Lateral bony tubular exten-
sion of supraorbital lateral-line canal to orbital rim
(1).

23. Elevation of frontal ridges on ethmoidal region. Not
elevated (0). Elevated; laterally convex (1). Extremely
elevated; appear as laterally flat sheets (2).

Orbital Region

24. Sclerotics. Present (0). Absent (1).

Otic Region

25. Supraoccipital crest. Reduced; runs close to and par-
allel to the epiotic ridges; dorsal profile of the neu-
rocranium is flat in lateral view (0). Dorsally expand-
ed; runs higher and not parallel to the epiotic ridges;
dorsal profile of the neurocranium is not flat in lateral
view (1).

26. Highest point of supraoccipital crest. Posterior to the
orbits (0). Above the orbits (1).

27. Pterotic. Ends before the posterior margin of the neu-
rocranium (0). Ends well past the posterior margin of
the neurocranium (1).

28. Intercalar. Visible dorsally (0). Not visible dorsally

(1).

29. Exoccipital ridge. Reaches the vagus foramen (0).
Does not reach the vagus foramen (1).

PECTORAL GIRDLE

30. Length of dorsal articular process of posttemporal.
Short; ends before or slightly past the anterior margin
of the supratemporal (0). More than twice the length
of the supratemporal (1).

31. Posteroventral process of posttemporal. Absent (0).
Reduced, flat in cross section; comes in contact or
close association with the articular head of the supra-
cleithrum (1). Long, round in cross section; does not
come in contact or close association with the articular
head of the supracleithrum (2).

32. Anteroventral process of posttemporal. Originates on
the posterior corner of the bone (0). Shares a common
origin with the posteroventral process; close to or on
the anterior half of the canal-bearing portion of the
bone (1). Separate from the posteroventral process;
originates close to or on the anterior half of the canal-
bearing portion of the bone (2).

33. Posterodorsal corner of posttemporal. Not expanded
(0). Expanded (1).

34. Articular head of the supracleithrum. Posteriorly ex-
panded; bears a completely enclosed canal that trans-
mits the lateral line to the posttemporal (0). Posteri-
orly expanded; bears an open canal or shelf connect-
ing the lateral line to the posttemporal (1). Not ex-
panded; lacks a canal (2).

35. Lateral process on articular head of supracleithrum.
Absent (0). Present (1).

36. Anteroventral process on articular head of supra-
cleithrum. Absent (0). Present (1).

37. Posteroventral plate on coracoid. Absent; ventral
margin of coracoid is flat; its posterior corner ending
before the fourth radial (0). Present; ventral margin is
round; its posterior corner ending past the fourth ra-
dial (1).

38. Length of pectoral-fin rays. Posterior fin rays are lon-
ger (0). Anterior fin rays are longer (1).

PELVIC GIRDLE

39. Basipterygium. Present (0). Absent (1).

40. External spinous elements of pelvic fin. Spine-like (0).
Scale-like (1).

41. Position of basipterygium. Thoracic; articular facet is
anterior to distal tip of ventral postcleithrum (0). Ab-
dominal; articular facet is posterior to distal tip of
ventral postcleithrum (1).

42. Central part of basipterygium. Dorsally inclined (0).
Nearly parallel to the ventral body wall (1).

43. Central part of basipterygium. Bilaterally divided (0).
Longitudinally fused (1).

AXIAL SKELETON
Vertebral Column

44. Total number of vertebrae. 30-55 (0). 57-67 (1). 84-
198 (2).

45. First neural spine. Not distally bifurcate (0). Distally
bifurcate (1).

46. Anterior neural spines. Not expanded or forked dis-
tally (0). Expanded and forked distally (1).

Intermusculars

47. Series of unattached epineurals and epicentrals ex-
tending into caudal region. Both present (0). Only the
epineural series present (1). Both absent (2).

78 ■ Contributions in Science, Number 476

Gago: Trichiurid Phylogeny

Dorsal and Anal Fins

48. Notch in the fin membrane separating the spinous and
soft portions of the dorsal fin. Present (0). Absent (1).

49. Spinous portion of the dorsal fin. XIX-XXXIX
spines; its base is longer than that of the soft portion
(0). XXXVIII-XLV spines; its length is only slightly
shorter than length of soft portion (1). XXXI-XLVI
spines; its base is less than half of the length of the
soft portion but not extremely short (2). Only III— X
spines; its base is extremely short compared to the
base of the soft portion (3).

50. Number of radials in soft dorsal-fin pterygiophores.
2; proximal-middle and distal (0). 3; proximal, mid-
dle, and distal (1).

51. Proximal-middle radial of first dorsal pterygiophore.
Does not extend above occiput (0). Extends above
occiput (1).

52. Foramen at antero ventral corner between the proxi-
mal and distal portions of the dorsal-fin proximal ra-
dials. Absent (0). Present (1).

53. Number of supernumerary elements on first anal-fin
pterygiophore. Two (0). One (1).

54. Anal-fin soft rays. Well-developed soft rays through-
out the whole length of the anal fin (0). Well-devel-
oped soft rays only in the posterior portion of the anal
fin (1). Modified as small spinule-like elements (2).
Reduced to small scale-like ossifications that do not
penetrate the epithelium (3). Modified into small
fused knobs that barely penetrate the epithelium (4).

55. Anal-fin pterygiophores. Two (proximal-middle and

distal) or three (proximal, middle, and distal) radials
present (0). The proximal, middle, and distal radials
are fused with each other and appear as a single unit
(1).

CAUDAL COMPLEX

56. Caudal complex. Well developed (0). Reduced to a
few small internal elements (1). Absent (2).

57. Ultimate centrum. Flexed and forming a urostyle (0).
Not flexed (1).

58. Number of epurals. Three (0). Two (1). One (2).

59. Haemal and neural spines of preural centrum 4. Long;
extends past or as far as the posterior margin of the
preural centrum 3 (0). Short; does not extend well
past the anterior margin of the preural centrum 3(1).

60. Hypural plates formula. I + II + III + IV + V (0).
I-II + III-IV + V (1). I-II + III-IV-V (2).

61. Hypural notch. Large (0). Small (1).

62. CMC cartilage. Absent (0). Present (1).

63. CPNPU3 cartilage. Present (0). Absent (1).

OTOLITHS

64. Excisura. Delimited by a well-developed rostrum and
antirostrum (0). Delimited by a reduced rostrum and
antirostrum (1).

65. Cristae superior and inferior. Reduced (0). Well de-
veloped and overhanging (1).

66. Longitudinal ridge on the ostium. Absent (0). Present

(1).

Contributions in Science, Number 476

Gago: Trichiurid Phylogeny ■ 79

Natural History Museum
of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

n

7 ' Number 477
(Vy\ 17 September 1999

Contributions
in Science

New Fanworm Species (Polychaeta: Sabellidae:
Fabriciinae) from Phuket, Thailand, with
Comments on Fabriciola flammula Rouse
and Fabriciola cri Rouse

Kirk Fitzhugh

Natural History Museum of Los Angeles County

Serial
Publications
of THE
Natural History
Museum of
Los Angeles
County

Scientific

Publications

Committee

John Heyning, Acting Deputy Director
for Research and Collections
John M. Harris, Committee Chairman
Brian V. Brown
Kenneth E. Campbell
Kirk Fitzhugh
Karen Wise

Robin A. Simpson and K. Victoria Brown,

Managing Editors

The scientific publications of the Natural History Museum
of Los Angeles County have been issued at irregular in-
tervals in three major series; the issues in each series are
numbered individually, and numbers run consecutively, re-
gardless of the subject matter.

# Contributions in Science, a miscellaneous series of tech-
nical papers describing original research in the life and
earth sciences.

# Science Bulletin, a miscellaneous series of monographs
describing original research in the life and earth sci-
ences. This series was discontinued in 1978 with the
issue of Numbers 29 and 30; monographs are now
published by the Museum in Contributions in Science.

# Science Series, long articles and collections of papers on
natural history topics.

Copies of the publications in these series are sold through
the Museum Book Shop. A catalog is available on request.

The Museum also publishes Technical Reports, a miscel-
laneous series containing information relative to scholarly
inquiry and collections but not reporting the results of
original research. Issue is authorized by the Museum’s Sci-
entific Publications Committee; however, manuscripts do
not receive anonymous peer review. Individual Technical
Reports may be obtained from the relevant Section of the
Museum.

Natural History Museum
of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

Printed at Allen Press, Inc., Lawrence, Kansas
ISSN 0459-8113

New Fanworm Species (Polychaeta: Sabellidae:
Fabriciinae) from Phuket, Thailand, with
Comments on Fabriciola flammula Rouse
and Fabriciola cri Rouse

Kirk Fitzhugh1

ABSTRACT. Two new fanworm species in the genera Fabriciola Friedrich and Fseudoaugeneriella Fitzhugh
are described from Panwa Bay, Phuket, Thailand. Fabriciola pbuketensis n. sp. is similar to Fabriciola sp.
cf. F. berkeleyi Banse, F. mediaseta Fitzhugh, F. rubra Fitzhugh, F. flammula Rouse, and F. cri Rouse in
having a low anterior peristomial ring collar, red peristomial and pygidial eyes, and abdominal neuropodial
pinhead setae. The new species is distinguished from F. mediaseta and F. rubra by the absence of thoracic
broadly hooded flagellate setae and differs from Fabriciola sp. cf. F. berkeleyi, F. flammula, and F. cri in
that it lacks crown and trunk pigmentation; it differs in collar construction as well. Fabriciola flammula
and F. cri were originally described as having broadly hooded flagellate setae. Examination of paratype
material reveals that all inferior thoracic notosetae are narrowly hooded; descriptions of the species are
appropriately emended. The total number of described Fabriciola species is increased to 16. Fseudoauge-
neriella brevirama n. sp. is the second species described in the genus and differs from F. unirama Fitzhugh
in having vascularized ventral filamentous appendages that are no more than half the length of the branchial
crown. Appendages in P. unirama are almost the same length as the crown. The monophyly of Pseudoau-
generiella is discussed, and cladistic relationships among Fabriciola species are presented.

INTRODUCTION

The present account describes a new species of Fa-
briciola Friedrich, 1939, and a new species of Fseu-
doaugeneriella Fitzhugh, 1998, from the intertidal
zone at Phuket Island, Thailand. Recognition of the
diversity of Fabriciola has especially increased dur-
ing the past several years. Including the species de-
scribed here, there have been seven Fabriciola spe-
cies described since 1990 (Fitzhugh, 1990a, 1992a,
1998; Rouse, 1993, 1996), giving a total of 16 spe-
cies. In the process of comparing the Fabriciola de-
scribed here to similar species, I found that the de-
scriptions of F. flammula Rouse, 1993, and F. cri
Rouse, 1996, must be emended with respect to the
type of thoracic notosetae in each.

Fseudoaugeneriella was originally described
from a single species, P. unirama Fitzhugh, 1998,
from Okinawa Island, Japan. Fitzhugh (1998) not-
ed that the genus resembles Augeneriella Banse,
1957, in that both have an anterior peristomial ring
collar only developed ventrally as a wide lobe, well-
developed triangular dorsal lips, and vascularized
ventral filamentous appendages. The difference be-
tween the two genera is that the filamentous ap-
pendages in Fseudoaugeneriella are unbranched,
whereas fully formed filaments in Augeneriella are

1. Invertebrate Zoology Section, Research and Collec-
tions Branch, Natural History Museum of Los Angeles
County, 900 Exposition Boulevard, Los Angeles, Califor-
nia 90007.

Contributions in Science, Number 477, pp. 1-17
Natural History Museum of Los Angeles County, 1999

branched to some extent. Fitzhugh (1998; see also
Fitzhugh, 1990b) pointed out, however, that fila-
ments in Augeneriella are initially unbranched,
with branching developing as animals grow. Based
on a cladistic analysis of Fabriciinae taxa, Fitzhugh
(1998) found the presence of unbranched vascular-
ized filaments to be plesiomorphic for the subfam-
ily, such that it is not possible to include P. unirama
in Augeneriella. Justification for the placement of a
second species in Fseudoaugeneriella as well as
monophyly of the genus are provided by an update
of the cladistic analysis of fabriciin genera and spe-
cies performed by Fitzhugh (1998).

SYSTEMATICS

Family Sabellidae Latrielle, 1825
Subfamily Fabriciinae, Rioja, 1923
Fabriciola Friedrich, 1939
Fabriciola pbuketensis n. sp.

Figures 1-2

MATERIAL EXAMINED. Indian Ocean, Andaman
Sea, Thailand, Phuket Island. Holotype: LACM-AHF
1896, Panwa Bay, just north (about 100 m) of Phuket
Marine Biological Center pier, scraping of low mat of de-
tritus and algae from shale rock, midintertidal zone, col-
lection made at low tide, depth 0 m, 18 August 1997, coll.
K. Fitzhugh. Paratypes: LACM-AHF 1897, 11 specimens
(10 complete, 1 missing crown), same locality as holotype.
Paratypes: LACM-AHF 1898, 11 specimens (2 complete,

9 lacking crown), same locality as holotype, scrapings off
dead, branching, Acropora- like coral, midintertidal zone,
collection made at low tide, depth 0 m, 18 August 1997,
coll. K. Fitzhugh.

ETYMOLOGY. The specific epithet refers to the
occurrence of the species at Phuket Island.

DESCRIPTION. Holotype a complete male
with 8 thoracic and 3 abdominal setigers; bran-
chial crown length, 0.45 mm; remainder of body,
1.70 mm long; maximum width, 0.12 mm (Fig.
1A). Body slender, nearly uniform in width except
for slight tapering posteriorly. Branchial crown
with 3 pairs of radioles, distal ends filamentous,
same width as pinnules. Branchial or “radiolar”
skeleton ( sensu Rouse 1993) not observed. Radi-
oles each with 3 pairs of pinnules, all terminating
at about same height as ends of radioles. Dorsal
lips erect, triangular, rounded distally, and distinct
from dorsal-most radioles; low, distally rounded,
ventral liplike process present at base of proximal-
most pinnule of ventral-most radioles (Fig. IB).
Nonvascularized ventral filamentous appendages
present, slightly shorter than total length of radi-
oles, surfaces smooth, about same width as pin-
nules (Figs. IB, 2B-C). Dorsal margins of bran-
chial lobes not fused to one another. Branchial
hearts present. Anterior margin of anterior peri-
stomial ring as low membranous collar (Fig. 2),
with distal margin smooth all around; middorsum
completely separated by narrow gap; collar height
uniform. Collar of even thickness throughout,
about same length as posterior peristomial ring.
Annulation between collar and posterior peristo-
mial ring only present ventrally. Anterior peristo-
mial ring, including collar, about same length as
posterior ring. Middorsal medial lobe just dorsal
to mouth almost same height as collar. Pair of
round, red eyes in posterior peristomial ring. Se-
tiger 1 about same length as posterior peristomial
ring, distinctly wider than long; setigers 2-6 each
successively longer, with setigers 4-8 longer than
they are wide. Setiger 9 about same length as 1,
with setigers 10-11 each slightly shorter than se-
tiger 9. Pygidium about same length as setiger 11,
posterior margin slightly tapered, rounded. Some
paratype specimens exhibit greater longitudinal
contraction, with most thoracic setigers about as
long as they are wide. Pair of round, bright red
pygidial eyes. Superior thoracic notosetae elon-
gate, narrowly hooded, 3-4 per fascicle. Inferior
thoracic notosetae in setigers 2-8 also narrowly
hooded but shorter, 1 per fascicle. Abdominal neu-
ropodia of setigers 9-11 with very elongate, nar-
rowly hooded setae, 1-2 per fascicle, and 2 pin-
head setae per fascicle. Thoracic uncini acicular,
main fang slender; teeth above main fang slender
and slightly decreased in size away from fang;
hood not observed; 4-5 uncini per fascicle in sin-
gle rows (6-7 in some paratypes). Abdominal un-
cini with 5-6 rows of teeth in profile, 3-5 teeth
per row; manubrium over 2 times longer than den-
tate region, slightly expanded proximally (Fig.

1C); uncini in setigers 9-11 number 13, 12, and
10, respectively. Anus midventral, along anterior
margin of pygidium. Males with spermiogenesis
occurring in setigers 4-8, oocytes in females in se-
tiger 4. Preserved specimens white, no pigmenta-
tion on crown or body wall. Tubes composed of
fine, flocculent detrital material, total tube width
about 2 times greater than body width. Brooding
of young not observed.

REMARKS. This species falls within the Fabri-
ciola species complex defined by the presence of
abdominal neuropodial pin-head setae ( sensu Ben-
Eliahu, 1975), which includes F. mediaseta Fitz-
hugh, 1990 (Aldabra Atoll, western Indian Ocean),
Fabriciola sp. cf. F. berkeleyi Banse, 1956 ( sensu
Fitzhugh, 1992a; California), F. flammula Rouse,
1993 (Belize), F. cri Rouse, 1996 (Papua New Guin-
ea), and F. rubra Fitzhugh, 1998 (Okinawa Island).
Fabriciola mediaseta and F. rubra have broadly
hooded, flagellate setae in thoracic notopodia,
whereas inferior notosetae in F. phuketensis, Fabri-
ciola sp. cf. F. berkeleyi, F. flammula, and F. cri have
elongate narrowly hooded setae. The body of F.
phuketensis lacks pigmentation, whereas Fabriciola
sp. cf. F. berkeleyi and F. cri have peristomial and
branchial crown pigmentation, respectively. The
posterior margin of the middorsal collar gap in F.
flammula (Rouse, 1993: figs. 1-2) and F. cri (Rouse,
1996: figs. 1-2) is noticeably expanded, whereas
the gap in F. phuketensis is uniformly narrow (Fig.
2A).

Rouse (1993, 1996) allied F. flammula and F. cri
with F. mediaseta on the basis of the fact that the
3 species have inferior, notopodial, broadly hood-
ed, flagellate setae {sensu Fitzhugh, 1990a: fig. 6C;
see also Fitzhugh, 1998: fig. IB) in some thoracic
setigers. Rouse (1993: fig. 3; 1996: fig. 3) described
these species as having these setae in setigers 2-8
and 2-6, respectively. After examining paratypes of
F. flammula and F. cri, I find that these species do
not have broadly hooded, flagellate setae. I com-
pletely concur with Rouse (1996) that in F. cri,
there are qualitative differences in hood shape in
some setigers, and, as much as possible, these dif-
ferences should be reported. Comparatively, how-
ever, I regard the setae in F. cri to come closest to
the condition of elongate narrowly hooded as op-
posed to flagellate. In the latter condition, there is
a marked disjunction between the hood and distal
continuation of the shaft. The setae in F. flammula,
F. cri, and F. phuketensis show a smooth transition
between hood and distal shaft. Emendations to F.
flammula and F. cri are provided in order to reflect
these changes.

Fabriciola flammula Rouse, 1993, Emended

Fabriciola flammula Rouse, 1993: 250-253, figs.

1-10, 48.

MATERIAL EXAMINED. Caribbean Sea, Belize, Car-
rie Bow Cay. Paratypes: LACM-AHF 1630, 3 specimens,

2 ■ Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms

Figure 1. Fabriciola phuketensis n. sp. A, entire animal, lateral view (holotype, LACM-AHF 1896); B, right half of
branchial crown, inner view (paratype, LACM-AHF 1897); C, abdominal uncinus from setiger 9 (paratype, LACM-AHF
1897). Abbreviations: bh, branchial heart; dl, dorsal lip; vfa, ventral filamentous appendage; vl, ventral liplike process.

Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms ■ 3

Figure 2. Fabriciola phuketensis n. sp. A-C, dorsal, lateral (right side), and ventral views of anterior end, respectively
(holotype, LACM-AHF 1896). Abbreviations: apr, anterior peristomial ring; bh, branchial heart; ppr, posterior peristomial
ring; vfa, ventral filamentous appendage.

south side of leeward jetty, algal turf on boulders and
Strombus shells, sta. F400, 22 May 1991, coll. G. Rouse.

REMARKS. Rouse (1993: 252) described inferi-
or thoracic notosetae in setigers 2-8 as broadly
hooded, flagellate. All inferior thoracic notosetae
are elongate, with narrow hoods, as seen in most
other Fabriciola species. It should be noted, how-
ever, that the hoods of inferior notosetae in setiger
1 of most Fabriciola are narrower than those seen
in subsequent setigers.

Rouse (1993: 251) stated that ventral lips are ab-
sent. At the base of each proximalmost pinnule of
the ventral radioles, there is a distinct, distally
rounded swelling, similar to that described here in
F. phuketensis (Fig. IB) and several other Fabriciola
(e.g., Fitzhugh, 1990a: fig. IE for F. baltica Fried-
rich, 1939; Fitzhugh, 1998: fig. 2B for F. rubra) as
a “ventral lip-like process.”

Fabriciola cri Rouse, 1996, Emended

Fabriciola cri Rouse, 1996: 1765-1768, figs. 1-12,

39, 42-43.

MATERIAL EXAMINED. Pacific Ocean, Papua New
Guinea, Madang Province. Paratypes: LACM-AHF 1798,
10 specimens, from log lying in mangroves opposite Riwo
village, sta. F645, 23 August 1993, coll. G. Rouse, K. Fau-
chald, L.A. Ward, and P. Scott.

REMARKS. Rouse (1996: 1767, fig. 3) described
inferior thoracic notosetae in setigers 2-6 as broad-
ly hooded, flagellate. I consider these setae to be
elongate, narrowly hooded, as seen in most other
Fabriciola species. As I noted in the remarks on F.
phuketensis, however, I agree with Rouse’s assess-
ment that the hoods of inferior setae in setigers 2-6
are somewhat broader than those seen in setigers 1
and 7-8.

4 ■ Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms

Rouse (1993: 251) stated that ventral lips are ab-
sent. The bases of the proximal-most pinnules of the
ventral radioles do show some slight swelling, but
this is not pronounced and does not impart the ap-
pearance of the “ventral lip-like process” described
above in F. phuketensis (Fig. IB) or several other
Fabriciola species (e.g., Fitzhugh, 1990a: fig. IE for
F. baltica ; Fitzhugh, 1998: fig. 2B for F. rubra).

REVISED KEY TO FABRICIOLA SPECIES

The following key is modified from Fitzhugh (1998). As
with that key, this one does not include F. spongicola
(Southern, 1921) or F. pacifica (Annenkova, 1934), as ma-
terial is unavailable and original descriptions are too in-
complete. Fabriciola toner ella Banse, 1959, is included
based on the original description.

la. Two abdominal setigers . . . F. minuta Rouse

b. Three abdominal setigers 2

2a. Pygidial eyes present 3

b. Pygidial eyes absent F. parvus Rouse

3a. Abdominal neuropodial pin-head setae pres-
ent, peristomial and pygidial eyes range from

red to brown 4

b. Pin-head setae absent, eyes black to light

brown 9

4a. Thoracic notopodia with inferior flagellate se-
tae in setigers 3-7 5

b. Flagellate setae absent 6

5a. Peristomial and pygidial eyes bright red . .

F. rubra Fitzhugh

b. Peristomial and pygidial eyes faint red

F. mediaseta Fitzhugh

6a. Peristomial and pygidial eyes bright red . . 7
b. Peristomial and pygidial eyes reddish-brown

Fabriciola sp. cf. F. berkeleyi Banse

7a. Body without pigmentation 8

b. Base of branchial crown with black pigment

F. cri Rouse

8a. Posterior margin of middorsal gap in peristo-
mial collar distinctly expanded

F. flammula Rouse

b. Middorsal collar gap narrow along entire

length F. phuketensis n. sp.

9a. Anterior peristomial ring collar relatively even

in height all around 10

b. Collar higher ventrally 11

10a. Middorsal gap in collar very wide

F. ghardaqa Banse

b. Collar gap narrow F. berkeleyi Banse

11a. Thoracic uncini few in number, 2-3 per fas-
cicle 12

b. Thoracic uncini more numerous, 5-8 per fas-
cicle 13

12a. Dorsal and lateral margins of anterior peristo-
mial ring collar relatively high

F. baltica Friedrich

b. Dorsal and lateral margins of collar very low

F. liguronis Rouse

13a. Branchial crown comprises 1/5 to 1/8 of total
body length ... F. brevibranchiata Fitzhugh
b. Branchial crown longer . . F. tonerella Banse

Pseudoaugeneriella Fitzhugh, 1998

Pseudoaugeneriella brevirama n. sp.

Figures 3-4

MATERIAL EXAMINED. Indian Ocean, Andaman
Sea, Thailand, Phuket Island. Holotype: LACM-AHF
1899, detrital scrapings off dead, branching, Acropora-
like coral, midhintertidal zone, collection made at low
tide, depth 0 m, 18 August 1997, coll. K. Fitzhugh. Par-
atypes: LACM-AHF 1900, 17 specimens (1 complete, 16
missing posterior end), same locality as holotype, scraping
of low mat of detritus and algae from shale rock, midin-
tertidal zone, collection made at low tide, depth 0 m, 18
August 1997, coll. K. Fitzhugh.

ETYMOLOGY. The specific epithet refers to the
short, vascularized, ventral filamentous appendages
of the crown.

DESCRIPTION. Holotype complete with 8 tho-
racic and 3 abdominal setigers; branchial crown
length, 0.4 mm; remainder of body, 1.3 mm long;
maximum width, 0.15 mm. Body slender, slightly ta-
pering anteriorly and posteriorly (Fig. 3A). Branchial
crown with 3 pairs of radioles, distal ends filamen-
tous, same width as pinnules. Radioles each with 4-
5 pairs of pinnules, terminating at or slightly below
distal ends of radioles. Dorsal lips erect but low,
broadly rounded distally, distinct from radioles; low,
distally rounded, ventral liplike processes present at
bases of proximalmost pinnules of ventral radioles
(Fig. 3B). Vascularized ventral filamentous append-
ages present, about one-half the total length of ra-
dioles, surfaces slightly to very wrinkled, about one-
third wider than pinnules; interior of each filament
occupied by large blood vessel (Figs. 3A-B, 4B-C).
Dorsal margins of branchial lobes not fused to one
another. Branchial hearts present. Anterior margin of
anterior peristomial ring is a low ridge dorsally and
laterally (Figs. 3 A, 4A-B). Collar developed ventrally
as low, triangular lobe; wide basally, tapering distally
to broadly rounded margin (Fig. 4C). Annulation be-
tween anterior and posterior peristomial rings visible
ventrally. Middorsal medial lobe just dorsal to
mouth low, triangular. Pair of reniform (dorsal view)
or rounded (lateral) black eyes in anterior half of
posterior peristomial ring. Posterior peristomial ring
is four to five times longer than anterior ring. Setiger
1 slightly shorter than posterior peristomial ring, dis-
tinctly wider than it is long; remaining setigers slight-
ly longer but all wider than they are long. Setiger 9
slightly shorter than 8; setigers 10-11 each about
one-half the length of 9. Pygidium about the same
length as setiger 11, posterior margin slightly ta-
pered, rounded. Pair of round, black pygidial eyes.
Superior thoracic notosetae elongate, narrowly
hooded, 3 per fascicle. Inferior thoracic notosetae in
setigers 2 and 7-8 also narrowly hooded but shorter,
1-2 per fascicle; setigers 3-6 each with 2 pseudos-
patulate setae (Fig. 3C). Abdominal neuropodia of
setigers 9-11 with very elongate, narrowly hooded
setae, 2-3 per fascicle. Thoracic uncini acicular, main
fang slender; single large tooth slightly offset over
main fang, followed by a series of smaller teeth;

Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms I 5

Figure 3. Pseudoaugeneriella brevirama n. sp. A, entire animal, lateral view (holotype, LACM-AHF 1899); B, right half
of branchial crown, inner view, entire crown length is 0.62 mm, ventral filamentous appendage length is 0.27 mm
(paratype, LACM-AHF 1900); C, inferior thoracic notopodial pseudospatulate seta from setiger 3 (paratype, LACM-
AHF 1900); D, thoracic uncinus from setiger 3 (paratype, LACM-AHF 1900). Abbreviations: dl, dorsal lip; vfa, ventral
filamentous appendage; vl, ventral liplike process.

hood present (Fig. 3D); 4-7 uncini per fascicle in
straight or irregular single rows. Abdominal uncini
with 7-8 rows of teeth in profile, 3-4 teeth per row;
manubrium about 1.5 times longer than dentate re-
gion, slightly expanded proximally; uncini in setig-
ers 9-11 number 17, 16, and 11, respectively. Anus
midventral, along anterior margin of pygidium. Oc-
currence of oocytes or sperm could not be deter-
mined. Branchial crown unpigmented in all speci-
mens. Dark to light brown pigment present in most
specimens (absent in holotype) in dorsum of pos-
terior peristomial ring and present dorsally, later-
ally, and ventrally in setigers 2 or 3; remainder of
body cream colored. Specimens in loose tubes com-

posed of detritus and mucus. No brooding of young
observed.

REMARKS. Pseudoaugeneriella brevirama is
nearly identical to P. unirama, known only from
Okinawa Island. Both species have inferior thoracic
pseudospatulate setae in setigers 3-6 and abdomi-
nal uncini with manubria that are about 1.5 times
longer than the dentate region. The only difference
between the two species lies in the length of the
ventral filamentous appendages. The appendages
extend to nearly the distal end of the crown in P.
unirama (Fitzhugh, 1998: fig. 10B), whereas in P.
brevirama, appendages are one-half the length of
the crown.

6 ■ Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms

0.1 mm

Figure 4. Pseudoaugeneriella brevirama n. sp. A-C, dorsal, lateral (left side), and ventral views of anterior end, respec-
tively (paratype, LACM-AHF 1900). Abbreviations: apr, anterior peristomial ring; bh, branchial heart; ppr, posterior
peristomial ring; vc, ventral lobe extension of anterior peristomial ring; vfa, ventral filamentous appendage.

The holotype of P. brevirama is one of the small-
er specimens in the type series. One of the largest
specimens, which is complete except for the abdo-
men, has a crown length of 0.53 mm and a thorax
length of 1.20 mm.

THE STATUS OF PSEUDOAUGENERIELLA

The description of the monotypic Pseudoauge-
neriella by Fitzhugh (1998) was basically developed
out of necessity as a means of accommodating a
species that could not be placed in Augeneriella.
With the discovery of a species nearly identical to
P. unirama, the question of the monophyly of Pseu-
doaugeneriella must be addressed. This is especially
critical since there are no features unique to Pseu-
doaugeneriella that are not also found in at least
some other Fabriciinae taxa. The most recent cla-
distic analysis of relationships among Fabriciinae
taxa was that of Fitzhugh (1998), which provided
the basis for recognizing Pseudoaugeneriella. The
data from that study were used to perform a cla-
distic analysis here, with the inclusion of P. brevir-
ama and Fabriciola pbuketensis.

A total of 21 characters were used (Appendix I),
comprising a total of 40 apomorphic states, these
being the same ones used by Fitzhugh (1998) in an
analysis of relationships among Fabriciinae genera
and species. In that analysis, Fitzhugh (1998) con-

ducted two separate analyses, taking into consider-
ation different outgroup conditions for the dentition
in thoracic uncini (character 11), and the following
states were treated as plesiomorphic in the respective
analyses: state 1, teeth gradually decrease in size
away from the main fang; and state 2, large tooth
slightly offset from midline, followed by a series of
smaller teeth. Separate analyses were also performed
in the present study. Taxa included 58 fabriciin spe-
cies among 13 genera. The recognition of Fabriciola
berkeleyi and Fabriciola sp. cf. berkeleyi as different
species and thus their inclusion as separate entities
in the analysis here was justified by Fitzhugh (1992a:
71). Character state assignments for species (Appen-
dix II) are the same as those used by Fitzhugh
(1998), except in the cases of P. flammula and P. cri,
which have been recoded with only narrowly hood-
ed inferior thoracic notosetae [state 13(0)] in accor-
dance with the emendations discussed above. Clad-
ograms were constructed using the program Hen-
nig86 (Farris, 1988), with the heuristic command
options “mhennig*” and “bb*.” Character state dis-
tributions among trees were examined using the Tree
Gardener program (Ramos, 1997).

Both analyses produced over 1,071 trees (maxi-
mum held in computer memory), each with a length
of 65 steps, a consistency index (ci) of 0.63, and a
retention index (ri) of 0.89. The consensus trees for
both analyses (Figs. 5-6) are similar to those found by

Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms ■ 7

Genus A

Manayunkia + Monroika
Pseudofabriciola
baltica
liguronis
minuta
parvus
berkeleyi
ghardaqa
brevibranchiata
tonerella

F. sp. cf. F. berkeleyi
flammula
cri

phuketensis, n. sp.

mediaseta
N rubra
\\\^ Pseudofabricia
Fabricia

\\ Parafabricia
^ Brifacia

Augeneriella

Fabhciola

umrama

brevirama , n. sp.

| Pseudoaugeneriella

Novafabricia

Fabricinuda

Figure 5. Strict consensus cladogram based on 1,071 cladograms from analysis with character 11 (thoracic uncini den-
tition) coded as state 1 (teeth gradually decrease in size away from main fang) for the outgroup. Monotypic genera are
indicated by dashed branches; species are shown for Fabriciola and Pseudoaugeneriella ; nonmonophyletic genera are
indicated by white bars.

8 ■ Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms

Genus A

Manayunkia + Monroika

Pseudofabriciola

baltica

liguronis

minuta

parvus

berkeleyi

ghardaqa

brevibranchiata

tonerella

F. sp. cf. F. berkeleyi

flammula

cri

phuketensis, n. sp.

mediaseta

rubra

Pseudofabhcia

Fabricia

Parafabricia

Brifacia

Augeneriella

Fabriciola

unirama

brevirama, n. sp.

| Pseudoaugeneriella

Novafabricia

Fabricinuda

Figure 6. Strict consensus cladogram based on 1,071 cladograms from analysis with character 11 (thoracic uncini den-
tition) coded as state 2 (large tooth above main fang followed by series of smaller teeth) for the outgroup. Monotypic
genera are indicated by dashed branches; species are shown for Fabriciola and Pseudoaugeneriella ; nonmonophyletic
genera are indicated by white bars.

Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms ■ 9

01 10 13

Monotypic genera are indicated by dashed branches; species-level relationships are shown for Novafabricia and Auge-
neriella. The distribution of states for characters 10 (thoracic pseudospatulate setae) and 13 (manubrium length in
abdominal uncini) is shown for terminal taxa. Note that the presence of pseudospatulate setae in setigers 3-6 [10(3)] is
a synapomorphy for Pseudoaugeneriella (cf. Fig. 8).

Fitzhugh (1998: figs. 18, 28), indicating the presence
of topologies in which Augeneriella and Novafabricia
Fitzhugh, 1990, are not monophyletic. As well, the
consensus tree for the analysis with state 11(2) pie-
siomorphic (Fig. 6) also allows for the possibility that
Fabriciola may not be monophyletic (see also Fitz-
hugh, 1998: figs. 28, 31). Relationships among Fa-
briciola species will be addressed in the next section.

In ail trees produced in both analyses, Fseudoau-
generiella is monophyletic, with P. unirama and P.
brevirama as sister taxa (e.g., Figs. 7-8). In all to-
pologies, Fseudoaugeneriella is sister group to a

clade that contains at least Augeneriella ; Novafa-
bricia; Farafabricia Fitzhugh, 1992; Brifacia Fitz-
hugh, 1998; Fabricia Blainville, 1828; and Fabri-
cinuda Fitzhugh, 1990. The lack of resolution in
the consensus tree is the result of movements of
Pseudofabricia Cantone, 1972, within and outside
this clade (see Fitzhugh, 1998).

The Fseudoaugeneriella clade is defined by either
state 10(3)-(distribution of inferior thoracic pseu-
dospatulate setae; Fig. 7) or 13(2)-(length of the
manubrium of abdominal uncini; Fig. 8), depend-
ing on the topology. The two species of Pseudoau-

10 ■ Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms

03 10 13

genera are indicated by dashed branches; paraphyletic Augeneriella is indicated by a white bar; species-level relationships
are shown for Novafabricia and Augeneriella. The distribution of states for characters 10 (thoracic pseudospatulate setae)
and 13 (manubrium length in abdominal uncini) is shown for terminal taxa. Note that the presence of a manubrium 1.5
times longer than the dentate region in abdominal uncini [13(2)] is a synapomorphy for the genus (cf. Fig. 7).

generiella have pseudospatulate setae in setigers 3-
6 [10(3)] and have a manubrium length 1.5 times
longer than the dentate region in abdominal uncini
[13(2)]. Other taxa with pseudospatulate setae lim-
ited to setigers 3-6 include two Augeneriella species
(A. hummelincki Banse, 1957; A. pectinata Fitz-
hugh, 1990) and two Novafabricia species [N. chi-
lensis (Hartmann-Schroder, 1962); N. labrus Fitz-
hugh, 1998]. Additional taxa with a manubrium
length like that of Pseudoaugeneriella include sev-
eral species in Pseudofabriciola Fitzhugh (P. Cali-
fornia Fitzhugh, 1991; P. peduncula Fitzhugh,

1996; P. sofla Fitzhugh, 1996) and Novafabricia
(N. tenuiseta Fitzhugh, 1990; N. triangularis Fitz-
hugh, 1990; N. exiguus Fitzhugh, 1998). The Pseu-
doaugeneriella clade can be defined by pseudospa-
tulates in setigers 3-6 only in those topologies in
which Augeneriella is not the most plesiomorphic
taxon in the clade that is most closely related to
Pseudoaugeneriella (Fig. 7). Manubrium length is a
synapomorphy for Pseudoaugeneriella in those to-
pologies in which Novafabricia is not the most ple-
siomorphic taxon in the clade that is most closely
related to Pseudoaugeneriella (Fig. 8).

Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms 111

Genus A

Pseudofabricia

baltica

liguronis

minuta

parvus

berkeleyi

ghardaqa

brevibrancbiata

tonerella

F. sp. cf. F. berkeleyi

flammula

cri

phuketensis , n. sp.

mediaseta

rubra

Manayunkia + Monroika

Pseudofabriciola

Pseudoaugeneriella

Fabriciola

Augeneriella

Fabricia

Parafabricia

Brifacia

Novafabricia

Fabricinuda

Figure 9. Single cladogram showing the placement of Pseudofabricia as sister group to Fabriciola, Manayunkia, Mon-
roika, and Pseudofabriciola. Note that this arrangement results in Fabriciola being paraphyletic, since nonvascularized
filamentous appendages [1(1)] are present both in Pseudofabricia and Fabriciola, and are thus plesiomorphic for the
clade. Monotypic genera are indicated by dashed branches; paraphyletic genera are indicated by white bars; species-level
relationships are only shown for Fabriciola and Augeneriella.

Fitzhugh (1998) noted that the increase in Fabri-
ciinae species and genera has not been followed by
a similar increase in character data, which has led
to greater topological instability in relationships as
well as ambiguity in the monophyly of some genera
(e.g., Augeneriella and Novafabricia). Unfortunate-
ly, while the definition of Pseudoaugeneriella has
not changed from that provided by Fitzhugh
(1998), the addition of a second species leaves the
issue of monophyly still unresolved.

CLADISTIC RELATIONSHIPS AMONG
FABRICIOLA SPECIES

The present cladistic analyses of fabriciin genera
and species show relationships among Fabriciola
similar to those described by Fitzhugh (1998: figs.
18, 28, 31). Fabriciola is monophyletic in all trees
in the analysis with state 11(1) as plesiomorphic
(Fig. 5), whereas some topologies in the analysis
with state 11(2) as plesiomorphic allow for Fabri-

12 ■ Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms

branchial ‘skeleton’
6(1)

baltica

liguronis

minuta

parvus

berkeleyi

brevibranchiata

ghardaqa

tonerella

F. sp. cf. F. berkeleyi
cri

flammula

phuketensis

mediaseta

rubra

baltica
liguronis
minuta
parvus
berkeleyi
brevibranchiata
ghardaqa
tonerella
F. sp. cf. F. berkeleyi
cri

flammula
phuketensis
mediaseta
rubra

baltica
liguronis
minuta
parvus
berkeleyi
brevibranchiata
ghardaqa
tonerella
F. sp. cf. F. berkeleyi
cri

flammula
phuketensis
mediaseta
rubra

parvus

baltica

liguronis

minuta

berkeleyi

brevibranchiata

ghardaqa

tonerella

F. sp. cf. F. berkeleyi
cri

flammula

phuketensis

mediaseta

rubra

Figure 10. A-C, relationships among Fabriciola species derived from data in Appendix IV. Character state changes are
shown in A and are the same in all trees. D, relationships among Fabriciola species derived from data in Appendix IV,
except that the absence of pygidial eyes [4(1)] is plesiomorphic. Relationships produced by this change are the same as
in A-C, except that F. parvus is plesiomorphic to the clade that includes F. baltica, F. liguronis, and F. minuta.

ciola to be paraphyletic relative to Manayunkia Lei-
dy, 1859; Monroika Hartman, 1951; and Pseudo-
fabriciola (Figs. 6, 9). These instances of paraphyly
in Fabriciola are due to the placement of Pseudo-
fabricia as sister group to Fabriciola, Manayunkia,
Monroika, and Pseudofabriciola, such that nonvas-
cularized ventral filamentous appendages [1(1)] are
plesiomorphic for this clade (Fig. 9). Given the ex-
tensive instability in the placement of Pseudofabri-
cia (see Fitzhugh, 1998), I consider the possibility
that Fabriciola may be paraphyletic to be an anom-
alous situation resulting from the lack of character
data sufficient to resolve the placement of Pseudo-
fabricia.

Current relationship patterns among Fabriciola
species were examined in a separate cladistic anal-

ysis using a subset of the characters from the above
analyses [i.e., six characters with a total of seven
apomorphic states (Appendix III), as well as the 14
species (Appendix IV)]. The outgroup condition for
pygidial eyes (character 4) was coded as either ab-
sent [4(2)] or black eyes present [4(0)], which re-
flects the possible plesiomorphic conditions for Fa-
briciola based on the earlier analyses. The exhaus-
tive search command “ie*” was used in Hennig86,
and this command produced three trees (Fig. 10),
regardless of the outgroup assignment for pygidial
eyes, each of which had a length of 7 steps and a
ci and ri of 1.00. The topologies produced are sim-
ilar to those reported by Fitzhugh (1998: fig. 34).
With black pygidial eyes coded as plesiomorphic
(Fig. 10A-C), there are two major clades — (i) F.

Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms I 13

baltica ; F. liguronis Rouse, 1993; F. minuta Rouse,
1996; and F. parvus Rouse, 1993; and (ii) F. ber-
keleyi ; F. brevibranchiata Fitzhugh, 1992; F. ghar-
daqa Banse, 1959; F. tonerella Banse, 1959; Fabri-
ciola sp. cf. F. berkeleyi ; F. cri ; F. flammula ; F. pbu-
ketensis ; F. mediaseta ; and F. rubra. The former
clade is defined by the presence of a branchial
“skeleton” [6(1)], the latter by the presence of red
peristomial [2(1)] and pygidial eyes [4(2)]. The
clade with red eyes also has an apomorphic clade
of species which have abdominal pinhead setae
[5(1)]. Differences between the trees is due to the
movement of F. tonerella (cf. Fig. 10A-C), since the
presence or absence of pinhead setae is unknown
for that species. Topologies are the same when py-
gidial eyes are coded as absent for the plesiom-
orphic condition (e.g., Fig. 10D), except within the
clade comprising F. baltica, F. liguronis, F. minuta,
and F. parvus. Since pygidial eyes are absent in F.
parvus, the presence of black eyes defines the more
inclusive clade comprising F. baltica, F. liguronis,
and F. minuta.

The only notable difference between the topolo-
gies obtained by Fitzhugh (1998: fig. 34) and those
reported here (Fig. 10) is the separation of F. cri
and F. flammula from the clade with F. mediaseta
and F. rubra. Fitzhugh’s (1998) analysis grouped
these species by the presence of inferior thoracic
flagellate setae, which were shown above to be ab-
sent in F. cri and F. flammula. It is possible that
greater resolution could be obtained for Fabriciola
relationships if peristomial collar construction were
taken into consideration. For instance, in several
cladistic analyses (Fitzhugh, 1991, 1992b, 1993) of
relationships among Fabriciinae genera and species
as well as among Fabriciola species (Fitzhugh,
1992a), collar height has been taken into consid-
eration. Within Fabriciola, most species have a low
membranous collar of even height all around. The
ventral collar margin is, however, distinctly higher
in F. baltica, F. liguronis, F. minuta, F. parvus, and
F. tonerella. I have opted to not take this feature
into consideration here, since there is complete am-
biguity as to the plesiomorphic collar state among
species of Fabriciola. Species of Manayunkia have
a membranous collar with a higher ventral margin,
but an exclusive sister group relationship with Fa-
briciola is not unambiguous (e.g., Figs. 5-6). Pseu-
dofabriciola, too, has a membranous collar, but it
is uniformly high all around. Remaining fabriciin
genera have a collar condition unlike that seen in
these three genera.

ACKNOWLEDGMENTS

I extend my thanks to the Phuket Marine Biological Cen-
ter for providing lab space. The species described here
were collected during the PMBC/DANIDA Polychaete
Workshop, organized by Danny Eibye-Jacobson, Zoolog-
ical Museum, Copenhagen. I thank Dr. Eibye-Jacobson for
his hospitality and material support. Reviews of the man-
uscript by Tom Perkins and Greg Rouse are greatly ap-
preciated.

LITERATURE CITED

Ben-Eliahu, N.M. 1975. Polychaete cryptofauna from
rims of similar intertidal vermetid reefs on the Med-
iterranean coast of Israel and in the Gulf of Elat:
Sabellidae (Polychaeta Sedentaria). Israel Journal of
Zoology 24:54-70.

Farris, J.S. 1988. Hennig86, version 1.5. [Computer soft-
ware and manual.] Available via http://www.vims.edu/
~mes/hennig/hennig.html.

Fitzhugh, K. 1990a. Revision of the Fabriciinae genus Fa-
briciola Friedrich, 1939 (Polychaeta: Sabellidae).
Zoologica Scripta 19:153-164.

— . 1990b. Revision of the Fabriciinae genus Auge-

neriella Banse, 1957 (Polychaeta: Sabellidae). Jour-
nal of Natural History 24:195-218.

. 1991. Further revisions of the Sabellidae subfam-
ilies and cladistic relationships among the Fabrici-
inae (Annelida: Polychaeta). Zoological Journal of
the Linnaean Society 102:305-332.

. 1992a. Species of Fabriciola Friedrich, 1939 (Po-
lychaeta: Sabellidae: Fabriciinae), from the Califor-
nia coast. Pacific Science 46:68-76.

. 1992b. On the systematic position of Monroika

africana (Monro) (Polychaeta: Sabellidae: Fabrici-
inae) and a description of a new fabriciin genus and
species from Australia. Proceedings of the Biological
Society of Washington 105:116-131.

— . 1993. Novafabricia brunnea (Hartman, 1969),

new combination, with an update on relationships
among Fabriciinae taxa (Polychaeta: Sabellidae).
Contributions in Science 438:1-12.

. 1998. New fan worm genera and species (Poly-
chaeta, Sabellidae, Fabriciinae) from the western Pa-
cific, and cladistic relationships among genera. Zool-
ogica Scripta 27:209-245.

Ramos, T.C. 1997. Tree Gardener, version 2.2. [Computer
software.] Available via http://www.vims.edu/~mes/
hennig/hennig.html.

Rouse, G.W. 1993. New Fabriciola species (Polychaeta,
Sabellidae, Fabriciinae) from the eastern Atlantic,
with a description of sperm and spermathecal ultra-
structure. Zoologica Scripta 22:249-261.

— . 1996. New Fabriciola and Manayunkia species (Fa-
briciinae: Sabellinae: Polychaeta) from Papua New
Guinea. Journal of Natural History 30:1761-1778.

14 ■ Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms

Appendix I. Characters and states used to determine cladistic relationships among Fabriciinae genera and species.

1. Ventral filamentous appendages: (0) absent; (1) nonvascularized, unbranched; (2) vascularized, unbranched; (3)
vascularized, branched.

2. Dorsal lips: (0) well-develped, triangular lobes, with dorsal margins well separated from proximalmost pinnules of
dorsal radioles; (1) dorsal margins fused with proximalmost radioles to some extent, forming low to moderately
narrow ridges; (2) absent.

3. Position of branchial crown: (0) extends over entire anterior end; (1) shifted dorsally to some extent.

4. Branchial lobe shape: (0) wide and short; (1) narrow and elongate, and/or with pedunclelike process.

5. Anterior peristomial ring collar: (0) low ridge dorsally and laterally, ventrally as narrow lobe; (1) membranous, of
varying height; (2) low ridge dorsally and laterally, ventrally as broad lobe; (3) low ridge all around.

6. Middorsal collar surface: (0) separate; (1) entire and distinctly grooved; (2) entire and smooth.

7. Middorsal collar margin: (0) separate; (1) entire; (2) notched or incised.

8. Anterior peristomal ring dimensions: (0) wider than long; (1) longer than wide.

9. Peristomal eyes: (0) red in live specimens, disappear in preservation; (1) red in live specimens, persist in preserva-
tion; (2) black, well developed, in live and preserved specimens; (3) black, poorly developed, in live and preserved
specimens.

10. Distribution of inferior thoracic pseudospatulate notosetae among setigers 2-8: (0) absent; (1) 2-5; (2) 3-5; (3) 3-
6; (4) 3-7; (5) 3-8; (6) broadly hooded, flagellate in 3-7 or 2-8.

11. Dentition above main fang of thoracic uncini: (0) series of uniformly small teeth; (1) teeth gradually decrease in
size away from main fang; (2) large tooth slightly offset from midline, followed by series of smaller teeth.

12. Dentition of abdominal uncini: (0) >1 row of teeth; (1) single row of teeth.

13. Manubrium of abdominal uncini: (0) Amphicorina- like; (1) about two times longer than dentate region; (2) about
1.5 times longer than dentate region; (3) same length as dentate region.

14. Pygidial eyes: (0) absent in live and preserved specimens; (1) black in live and preserved specimens; (2) red in live
specimens, persist in preservation; (3) red in live specimens, disappear in preservation.

15. Radioles: (0) 3 or more pairs; (1) 2 pairs.

16. Body-wall spicules: (0) absent; (1) present.

17. Branchial hearts: (0) absent; (1) present.

18. Displaced pinnules: (0) absent; (1) present.

19. Pinnule arrangement: (0) distinctly pectinate; (1) 2-4 pinnules at bases of branchial lobes.

20. Abdominal neuropodial pinhead setae: (0) absent; (1) present.

21. Branchial “skeleton”: (0) absent; (1) present.

Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms ■ 15

Appendix H. Character-state matrix for Fabriciinae genera and species based on character states presented in Appendix I.

1

1

1

1

1

1

1

1

1

1

2

2

1

2

3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

Outgroup

0

0

0

0

0

0

0

0

0

0

1/2

0

0

3

0

0

0

0

0

0

0

Manayunkia aestuarina

2

0

0

0

1

0

0

0

2

1

1

0

1

0

1

0

1

0

1

0

0

M. baicalensis

2

0

0

0

1

0

0

0

2

0

1

0

1

0

1

0

1

1

0

0

0

M. brasiliensis

2

0

0

0

1

0

0

0

2

1

1

0

1

0

1

0

1

1

1

0

0

M. polaris

2

0

0

0

1

0

0

0

2

3

1

0

1

0

1

0

1

0

1

0

0

M. speciosa

2

0

0

0

1

0

0

0

2

0

1

0

1

0

1

0

1

1

0

0

0

Monroika africana

2

0

0

0

1

0

0

0

?

2

2

0

3

?

1

0

1

1

0

0

0

Augenenella dubia

2

0

0

0

2

0

0

0

0

0

0

0

1

3

0

1

1

0

0

0

0

Genus A sp.

2

0

0

0

2

0

0

0

0

0

0

0

1

3

0

1

1

0

0

0

0

Fabriciola battica

1

0

0

0

1

0

0

0

2

0

1

0

1

1

0

0

1

0

0

0

1

F. berkeleyi

1

0

0

0

1

0

0

0

1

0

1

0

1

2

0

0

1

0

0

0

0

F. sp. cf. F. berkeleyi

1

0

0

0

1

0

0

0

1

0

1

0

1

2

0

0

1

0

0

1

0

F. brevibranchiata

1

0

0

0

1

0

0

0

1

0

1

0

1

2

0

0

1

0

0

0

0

F. cn

1

0

0

0

1

0

0

0

1

6

1

0

1

2

0

0

1

0

0

1

0

F. flammula

1

0

0

0

1

0

0

0

1

6

1

0

1

2

0

0

1

0

0

1

0

F. ghardaqa

1

0

0

0

1

0

0

0

1

0

1

0

1

2

0

0

1

0

0

0

0

F. liguronis

1

0

0

0

1

0

0

0

2

0

1

0

1

1

0

0

1

0

0

0

1

F. mediaseta

1

0

0

0

1

0

0

0

1

6

1

0

1

2

0

0

1

0

0

1

0

F. minuta

1

0

0

0

1

0

0

0

2

0

1

0

1

1

0

0

1

1

0

0

1

F. parvus

1

0

0

0

1

0

0

0

2

0

1

0

1

0

0

0

1

0

0

0

1

F, phuketensis n. sp.

1

0

0

0

1

0

0

0

1

0

1

0

1

2

0

0

1

0

0

1

0

F. rubra

1

0

0

0

1

0

0

0

1

6

1

0

1

2

0

0

1

0

0

1

?

F. tonerella

1

0

0

0

1

0

0

0

1

0

1

0

1

2

0

0

1

0

0

?

0

Pseudofabriciola analis

0

1

0

1

1

2

1

0

2

0

1

0

3

1

0

0

1

0

0

0

0

P. australiensis

2

0

0

1

1

2

2

0

2

0

1

0

3

1

0

0

1

0

0

0

0

P. califomica

0

0

0

1

1

1

1

0

2

0

2

0

2

1

0

0

1

0

0

0

0

P. capensis

0

0

0

1

1

1

1

0

2

0

2

0

3

1

0

0

1

0

0

0

0

P. filamentosa

0

7

0

1

1

2

1

c

2

0

1

0

1

1

0

0

1

0

0

0

0

P. incisura

2

0

0

1

1

2

2

0

2

0

1

0

3

1

0

0

1

0

0

0

0

P. tonga

0

1

0

1

1

2

1

0

2

0

1

0

3

1

0

0

1

0

0

0

0

P. longipyga

0

1

0

1

1

2

2

0

2

0

1

0

1

1

0

0

1

0

0

0

0

P. peduncula

0

1

0

1

1

2

1

0

2

0

1

0

2

1

0

0

1

0

0

0

0

P. quasiincisura

2

0

0

1

1

2

2

0

2

0

1

0

3

1

0

0

1

0

0

0

0

P. sofla

0

1

0

1

1

2

1

0

2

0

1

0

2

1

0

0

1

0

0

0

0

Fabncia stellans

0

0

0

0

2

0

0

0

2

4

2

0

1

1

0

0

1

0

0

0

0

Pseudofabricia aberrans

1

0

0

0

2

0

0

0

2

0

2

0

3

1

0

0

1

0

0

0

0

Augenenella basifurcata

3

0

0

0

2

0

0

0

3

4

2

0

3

1

0

0

1

0

0

0

0

A. hummelincki

3

0

0

0

2

0

0

0

2

3

2

0

1

1

0

0

1

0

0

0

0

A. lagunari

3

0

0

0

2

0

0

0

3

4

2

0

3

1

0

0

1

0

0

0

0

A. mossambica

3

0

0

0

2

0

0

0

2

4

2

0

3

1

0

0

1

0

0

0

0

A. pectinata

3

0

0

0

2

0

0

0

2

3

2

0

3

1

0

0

1

0

0

0

0

Novafabncia bilobata

0

1

0

0

2

0

0

0

2

4

2

0

3

1

0

0

1

0

0

0

0

N. brunnea

0

1

0

0

2

0

0

0

2

2

2

0

1

1

0

0

1

0

0

0

0

N. chilensis

0

1

0

0

2

0

0

0

2

3

2

1

3

1

0

0

1

0

0

0

0

N. exiguus

0

1

0

0

2

0

0

0

2

2

2

0

2

1

0

0

1

0

0

0

0

N. gerdi

0

1

0

0

2

0

0

0

2

2

2

1

3

1

0

0

1

0

0

0

0

N. infratorquata

0

1

0

0

2

0

0

0

2

2

2

0

1

1

0

0

1

0

0

0

0

N. labms

0

1

0

0

2

0

0

0

2

3

2

0

3

1

0

0

1

0

0

0

0

N. tenuiseta

0

1

0

0

2

0

0

0

2

0

2

0

2

1

0

0

1

0

0

0

0

N. triangularis

0

1

0

0

2

0

0

0

2

2

2

0

2

1

0

0

1

0

0

0

0

Parafabricia ventricingulata

0

0

0

0

2

0

0

0

2

4

2

0

3

1

0

0

1

0

0

0

0

Fabricinuda limnicola

0

2

0

0

3

0

0

1

2

5

2

0

3

1

0

0

1

0

0

0

0

F. bikinii

2

2

1

0

3

0

0

1

2

5

2

0

3

1

0

0

1

0

0

0

0

F. tnlobata

2

2

1

0

3

0

0

1

2

5

2

0

3

1

0

0

1

0

0

0

0

F. pseudocollaris

2

2

0

0

3

0

0

1

2

5

2

0

3

1

0

0

1

0

0

0

0

F. pseudopalpa

0

0

0

0

3

0

0

1

2

5

2

0

3

1

0

0

1

0

0

0

0

Pseudoaugeneriella unirama

2

0

0

0

2

0

0

0

2

3

2

0

2

1

0

0

1

0

0

0

0

P. brevlrama n. sp.

2

0

0

0

2

0

0

0

2

3

2

0

2

1

0

0

1

0

0

0

0

Bnfacia metastellaris

0

0

0

0

2

0

0

0

2

4

2

0

3

1

0

0

1

0

0

0

0

16 ■ Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms

Appendix III. Characters and states used to determine

cladistic relationships among Fabriciola species.

1. Ventral filamentous appendages: (0) vascularized, un-
branched; (1) nonvascularized, unbranched.

2. Peristomial eyes: (0) black, well developed; (1) red or
reddish brown.

3. Inferior thoracic notosetae: (0) elongate, narrowly
hooded in all setigers; (1) broadly hooded, flagellate
in some setigers.

4. Pygidial eyes: (0) black; (1) red or reddish brown; (2)
absent.

5. Abdominal pinhead setae: (0) absent; (1) present.

6. Branchial “skeleton”: (0) absent; (1) present.

Appendix IV. Character-state matrix for Fabriciola species
based on character states presented in Appendix III.

1

2

3

4

5

6

Outgroup

0

0

0

0

0

0

baltica

1

0

0

0

0

1

berkeleyi

1

1

0

1

0

0

F. sp. cf. F. berkeleyi

1

1

0

1

1

0

brevibranchiata

1

1

0

1

0

0

cri

1

1

0

1

1

0

flammula

1

1

0

1

1

0

ghardaqa

1

1

0

1

0

0

liguronis

1

0

0

0

0

1

mediaseta

1

1

1

1

1

0

minuta

1

0

0

0

0

1

parvus

1

0

0

2

0

1

phuketenisis n. sp.

1

1

0

1

1

0

rubra

1

1

1

1

1

p

tonerella

1

1

0

1

p

0

Contributions in Science, Number 477

Fitzhugh: Thailand Fanworms ■ 17

Natural History Museum
of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

H

A) it

Number 478
17 December 1999

Contributions
in Science

Larval Labrisomidae (Pisces: Blennioidei)
from the Galapagos Islands

Guillermo A. Herrera
Robert J. Lavenberg

Serial
Publications
of THE
Natural History
Museum of
Los Angeles
County

Scientific

Publications

Committee

John Heyning, Acting Deputy Director
for Research and Collections
John M. Harris, Committee Chairman
Brian V. Brown
Kenneth E. Campbell
Kirk Fitzhugh
Karen Wise

Robin A. Simpson and K. Victoria Brown,

Managing Editors

The scientific publications of the Natural History Museum
of Los Angeles County have been issued at irregular in-
tervals in three major series; the issues in each series are
numbered individually, and numbers run consecutively, re-
gardless of the subject matter.

# Contributions in Science, a miscellaneous series of tech-
nical papers describing original research in the life and
earth sciences.

# Science Bulletin, a miscellaneous series of monographs
describing original research in the life and earth sci-
ences. This series was discontinued in 1978 with the
issue of Numbers 29 and 30; monographs are now
published by the Museum in Contributions in Science.

# Science Series, long articles and collections of papers on
natural history topics.

Copies of the publications in these series are sold through
the Museum Book Shop. A catalog is available on request.

The Museum also publishes Technical Reports, a miscel-
laneous series containing information relative to scholarly
inquiry and collections but not reporting the results of
original research. Issue is authorized by the Museum’s Sci-
entific Publications Committee; however, manuscripts do
not receive anonymous peer review. Individual Technical
Reports may be obtained from the relevant Section of the
Museum.

Natural History Museum
of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

Printed at Allen Press, Inc., Lawrence, Kansas
ISSN 0459-8113

Larval Labrisomidae (Pisces: Blennioidei) from
the GalApagos Islands

Guillermo A. Herrera and Robert J.
Lavenberg

ABSTRACT. Larval Labrisomidae from the Galapagos Islands are described from field-collected speci-
mens. The species included are Starksia galapagensis, Dialommus fuscus, Labrisomus dendriticus, L. mul-
tiporosus, Malacoctenus tetranemus, and M. zonogaster. Identifications were based on meristic data and
on comparison of morphology between adults and larger larvae. Labrisomid larvae are elongate, slightly
compressed, with 30-47 myomeres, head small and generally rounded with a short snout, external mela-
nophores at anterior tip of the cleithral symphysis; at least one pair of melanophores on head; a ventral
row of melanophores on tail (between pterygiophores of anal fin); melanophores above gut (visible only
in early stages); and a large swimbladder that is lost during development. Characters that separate S.
galapagensis larvae from other labrisomids are a more advanced development at smaller size and a longer
preanal length (47-53% vs <45% of standard length). Larvae of Dialommus fuscus have a high number
of dorsal spines (24-27), a high number of anal rays (26-28), a high number of vertebrae (43), and a short
preanal distance (33% of standard length). Labrisomus and Malacoctenus larvae have no obvious syna-
pomorphies. Species are identified by particular combinations of characters: preopercular spines, and me-
lanophores on the upper jaw, dorsal margin of trunk, hypural border, and urostyle. In late larvae, the
relative size of spines and rays in the dorsal fin is useful in identification.

RESUMEN. Se describen las larvas de los labrisomidos de las Islas Galapagos, utilizando especimenes
recolectados en terreno. Las especies consideradas son: Starksia galapagensis, Dialommus fuscus, Labri-
somus dendriticus, L. multiporosus, Malacoctenus tetranemus, y M. zonogaster. La identificacion de las
larvas se establecio en base a informacion meristica y comparacion de morfologia de adultos con la de
larvas de mayor tamano. Las larvas de labrisomidos son elongadas, levemente comprimidas lateralmente,
con 30-47 miomeros, tienen una cabeza pequena y generalmente redondeada, con un hocico corto, con
un melanoforo externo en el extreme anterior de la sinfisis de los cleitros, con un par (al menos) de
melanoforos sobre la cabeza, con una hilera ventral de melanoforos en el margen ventral del cuerpo (entre
los pterigioforos de la aleta anal), con melanoforos sobre el intestine (visibles solo en etapas tempranas),
y una vejiga gaseosa grande que se pierde durante el desarrollo. Caracteres utiles en la separacion de las
larvas de S. galapagensis de las de otros labrisomidos son un desarrollo mas avanzado a una talla mas
pequena y una mayor longitud preanal (47-53% vs <45% de la longitud estandar). Las larvas de Di-
alommus fuscus tienen un alto numero de espinas en la aleta dorsal (24-27), un alto numero de radios
anales (26-28), un alto numero de vertebras (43), y una corta distancia preanal (33% de la longitud
estandar). Labrisomus and Malacoctenus no muestran obvias sinapomorfias larvales a nivel de genero. Las
especies se pueden identificar por combinaciones particulares de caracteres; espinas preoperculares, y me-
lanoforos en premaxilar, margen dorsal del tronco, borde de las placas hipurales, y urostilo. En larvas mas
grandes, la estructura de la aleta dorsal, i.e. el tamano relativo de espinas y radios es tambien util en la
identificacion.

INTRODUCTION

Labrisomid blennies are small (5-12 cm), demersal
reef and nearshore fishes, primarily of the New
World. The group is represented by 98 species (14
genera), with approximately half of the species in
the eastern Pacific (Nelson, 1984; Grove and Lav-
enberg, 1997).

Information about life history is available for
some members of the family. Matarese et al. (1984)
summarized briefly larval characters at a time when

Natural History Museum of Los Angeles County, Sec-
tion of Vertebrates (Ichthyology), 900 Exposition Boule-
vard, Los Angeles, California 90007-4000.

Contributions in Science, Number 478, pp. 1-14
Natural History Museum of Los Angeles County, 1999

not many larvae were described and the family was
not well defined. More recent and more significant
contributions include the works of Brogan (1992),
who studied the larvae of several species of labri-
somids and other blennies from the Gulf of Cali-
fornia, and Watson (1996), who described the lar-
vae of five species from the California Current Re-
gion.

We describe postflexion larvae of Starksia gala-
pagensis (Rosenblatt and Taylor, 1971), Dialom-
mus fuscus (Gilbert, 1891), Labrisomus multipo-
rosus (Hubbs, 1953), L. dendriticus (Reid, 1935),
Malacoctenus tetranemus (Cope, 1877), and M.
zonogaster (Heller and Snodgrass, 1903), based on
field-collected specimens from the Galapagos Is-

Table 1. Summary of meristic characters for the genera of eastern Pacific Labrisomidae.

Tribe/Genus

D

A

PI

P2

Vertebrae

Cryptotremini

Alloclinus

XXIV-XXVI, 9-11

II, 21-23

13-14

1,3

41-42

Auchenionchus

XXIV-XXVI, 11-12

II, 21-24

13-14

I, 3

42-43

Calliclinus

XXIV-XXV, 11-13

II, 20-22

15

1,4

41-43

Cryptotrema

XXVI-XXVIII, 11-12

II, 24-27

—

1,3

45-47

Labrisomini

Malacoctenus

XIX-XXII, 9-13

II, 17-23

13-16

I, 3

36-45

Labrisomus

XVII-XX, 10-13

II, 16-20

13-16

1,3

33-46

Mnierpini

Dialommus

XXIV-XVII, 12-14

I, 26-27

13

I, 3

43

Mnierpes

XXI-XXIII, 10-12

II, 22-23

12-13

1,3

39

Paraclinini

Exerpes

(III-IV)-(XXIII-XXVI), 1/2

II, 17-20

13-15

0-1, 2-3

35

Paraclinus

XXVI-XXXIII, 0-1

II, 16-21

11-15

0-1, 2-3

33-39

Starksiini

Starksia

XIX-XXII, 7-10

II, 16-21

12-15

1,3

30-35

Xenomedea

XX-XXIII, 8-11

II, 18-22

12-14

I, 3

34-37

lands. The larvae of another labrisomid species that
occurs in the Galapagos Islands, L. jenkinsi (Heller
and Sodgrass, 1903) were not found in the samples.
M. tetranemus and L. multiporosus range from the
Gulf of California to northern Chile, whereas the
other species, except L. dendriticus, are endemic to
the Galapagos Islands. L. dendriticus is an eastern
Pacific oceanic island endemic, known only from
Isla Malpelo and the Galapagos Islands (Grove and
Lavenberg, 1997).

MATERIALS AND METHODS

Plankton samples were collected by the R/V VELERO III
during the Allan Hancock expedition cruises to the eastern
tropical Pacific and Galapagos Islands. Most samples were
collected during 1933; however, a few specimens came
from the expeditions of 1932 and 1934. The collections
were made at night (at anchorage), using an electric light
and dipnets. Sampling sites were mainly at Espanola, San-
ta Maria, Santa Cruz, Isabela, Baltra, and Genovesa Is-
lands (Fraser, 1943).

The specimens (preserved in 70% ethanol) were sorted
from the larval fish collections at the Natural History Mu-
seum of Los Angeles County (LACM). Specimens were
identified by the size-series method (Powles and Markle,
1984), relying on the diagnostic meristics of each species.
Meristic data were obtained from Hubbs (1952, 1953),
Rosenblatt and Taylor (1971), Springer (1959), Stephens
and Springer (1973), Brogan (1992), and Hastings and
Springer (1994). Additional vertebral counts and fin mer-
istics were obtained from radiographs of juvenile and
adult specimens from the LACM holdings.

All measurements, standard length (SL) and preanal
length, were recorded to the nearest 0.1 mm using a Wild
M3 stereomicroscope. A representative size series of each
species, when available, was illustrated with the aid of a
camera lucida. Brief descriptions are given for species rep-
resented by a few larvae; they are usually about the same

size (e.g., Dialommus fuscus). Meristic data from radio-
graphs of specimens, combined with data from the liter-
ature (for labrisomid genera and species from the Gala-
pagos Islands), are summarized in Tables 1 and 2.

LABRISOMIDAE

The Labrisomidae are teleostean fishes of the sub-
order Blennioidei, which are considered to be par-
aphyletic on the basis of DNA sequences and al-
lozyme data (Stepien, Dixon, and Hillis, 1993).
Springer (1993) found no evidence to support the
monophyly of the Labrisomidae among the 34
characters he analyzed across all the Blennioidei.
Stepien et al. (1997) recently hypothesized that the
group composed of the Labrisomidae, Clinidae and
Chaenopsidae is monophyletic. The paraphyly in
the Labrisomidae is probably due to the inclusion
of some genera based on plesiomorphic characters
such as the presence of a scaled body and a lateral
line (Hastings and Springer, 1994).

Currently the labrisomids include 14 genera, al-
located into five tribes, and are diagnosed by a dor-
sal fin either entire or in a spinous and segmented
ray portion with more spines than segmented rays,
17-33 dorsal-fin spines, 0-13 simple segmented
dorsal-fin rays (never branched), two anal-fin spines
(except for Dialommus with one), 16-27 segment-
ed anal-fin rays, pectoral-fin rays unbranched,
scales cycloid when present with radii only in an-
terior field, free bony margins of opercular bones
not fimbriate, no projection of the pelvic girdle ex-
tending anteriorly in front of the juncture with the
cleithra, pelvic fins insert in advance of the pectoral
fins, dorsal- and anal-fin rays equal in number to
the vertebrae between them, and more than 30 ver-
tebrae (see Springer, 1993 for additional labrisomid

2 ■ Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae

_o

(2

-TO

3

3

rt

3

o

60

3

Q-

'rt

u

— 3

3

c e

« 3

^ +3

<u .a
§- &

6X3

*§ *

J’s

& ts
1 .2

S -a

5 c
ju *

3 - -

»-i m
-T3 N
C ON
rt
3

"•<-> C

•c &

3

6 a
to

©^
>• c

S *
E

|J3
<*> &

ur>
<S
o on

is

co *o vo io i— i

7 S T - T T T

^ O T-l oo

3" 3" 3" CO

I I

I I I

+ + I

+ +

Ci

I I +

I I

5 Q

+ i +

<N T-^

I + +

3

+ + +

3-| to VO LO

'"7 ' co ^ T T T ^

m ^ ^ -t co co ^

SI

SI

d ooS

ON _ O
_H O Ol
I ^ — ~
N I ON

» I h co

O o O

gSsiso

^SsxiR
m ® xix

xxx> J
xxxx£

x a

X X
K X

5 x|

X xl

Vo

'T1 CO NO

4- ^ CO

CO

VO NO OO
I CO CO CO
III

1 r+* L/n K.

c s

^ 32
§ S S
-St E

^ to CO

s

s

o

s s

3 S

Ss,

O K
.•CS. S

S 5

-a -o

3 3

►-J >-1

O o

.vo u
£ 3

*§ 3

*3 S

g 7^ O
O <U v+-t
■r; «

3 « B

g I u
■» a-§
^ c fc

cn ‘JJ *3
3 3 ^
3 -9 3

O g ~5
i; § “
c © §

3 (jT' — '
3 3 33
3 1_|
6 'P rt

t £ ©

co 0.3

g j7 £

B .3 m
tT1 6X3

-O <-^ 3

a ~ c

o s

(J " y

CO

3 >-l

_ O D

0 _c a

3 o ^
3 C co
.„ 3 3

C "5 "3

3 e <11

3 VlH co
CT o <U

aj w in

»h ^ Cm
o X
co VJ oj

CO

OJ V5 OJ
— ' 3 cj

O g

rt 3 2

co .5 -52

1 8 ”2

> co C

2 ^ 8

S -o' Oh

O Oh 3

X o 3

o aj

O ^ ^
^ + £
° s
^ 8 §
3 U

oj 3
& o

^ 1
3 ‘O
oj o

as

„ o

CO "TO

« 3

.s s

Dh OJ

CO >s J.

3 rj 3
3 3

M 3 Vh

3 c O ~V
3 3 3 „

3 © vh co
O' — ^ rt

3 Q „r
— — S

co >H O 3
co =3 B 3
3 3 3 B
— CO 3 ^

a s- a

33 J ^
O

3

CS

3 _. . .

-3 3 Jh — |

a 3 o g
o 3 -O o

3 u

3 _Q Q

Oh'

.. 73,

U, 3 >

03 0

3 3
3 —

C o

o 3 S

3

Iff I

<•5 &I

E X ^

features). These fishes are restricted to the tropical
and warm temperate waters of the New World and
the west coast of Africa (see Stepien 1992, figure 1;
we cannot verify the western Pacific record shown
in figure 1).

Labrisomids have two modes of spawning; ovo-
viviparity in the tribe Starksiini and oviparity in the
other four tribes. In the Starksiini the first anal-fin
spine is modified into an elongate genital papilla to
serve as an intromittent organ. Development pro-
ceeds in the follicles of the ovary, and the embryos
hatch at an advanced larval stage. To our knowl-
edge, all other labrisomids exhibit some degree of
courtship and spawn adhesive demersal eggs into a
nest site, which is guarded by the parents. At hatch-
ing, larvae have pigmented eyes and measure about
3-4 mm notochordal length (NL). Notochord flex-
ion usually occurs shortly thereafter.

Larvae are elongate, slender, and slightly laterally
compressed with at least 30-47 myomeres; gut ini-
tially straight, preanal length usually less than 50%
of standard length, up to 53% in Starksia\ a large
swimbladder (absent in adults); a small, rounded
head; a short snout; six branchiostegal rays; and
7+6 principal caudal rays. Pigmentation is weak or
light; melanophores occur primarily on dorsal cra-
nial surfaces; on the nape; anterior to the gut, be-
hind the cleithra; over the swimbladder; at the tip
of the cleithral symphysis; on the hind gut; and
along the ventral margin of the tail, especially be-
tween the pterygiophores of the anal-fin rays. La-
brisomid larvae show no pronounced morphologi-
cal specializations to pelagic life.

LARVAL DESCRIPTIONS

Starksia galapagensis (Rosenblatt
and Taylor, 1971)

Figure 1

GENERAL MORPHOLOGY. Based on 237
specimens, larvae and juveniles (11.6 to 14.2 mm).
Starksia galapagensis has single nasal, orbital, and
nuccal cirri and no preopercular spines (Table 2).
The smallest specimen (Fig. la., 11.6 mm) has com-
pleted notochord flexion. No significant changes in
body shape occur during subsequent development
(Fig. la-c). The largest larva was 14.2 mm (Fig.
lc), although metamorphosing specimens were
slightly smaller (Fig. ld-e). By 12.4 mm, all the
pores of the circumorbital series (10), and three in
the preoperculomandibular series (Fig. Id) are de-
veloped.

FIN DEVELOPMENT. All fin spines and rays are
present at 11.6 mm. The shape and relative pro-
portions of spines and rays are constant, except for
the posteriormost three to four anal-fin rays that
grow longer than the rest, forming a lobe nearly as
long as the caudal peduncle.

PIGMENTATION. Larval Starksia galapagensis
are unique in having a large ventral melanophore
on the surface of the basipterygium. A single small
melanophore may develop on the ventral margin of

Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae ■ 3

Figure 1. Field collected larvae and juvenile of Starksia galapagensis: (a) 11.6-mm larva (LACM 45644-12), (b) 13.0-
mm larva (LACM 45663-13), (c) 14.2-mm larva (LACM 45621-20), (d) 12.4-mm transforming specimen (LACM 45634-
15), (e) 12.6-mm juvenile (LACM 43688-1).

4 ■ Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae

the caudal peduncle. The 12.4-mm specimen (Fig.
Id) seems to be in the process of metamorphosis;
this specimen also has a small melanophore ventral
to the gut, at the pectoral-fin base level.

Part of the typical dark spots, composed of small
melanophores and observed in the head of the
adults, can be recognized in the two larger speci-
mens; these spots are in lips; in front of, below and
behind the eye; in the operculum; and in front of
the pectoral fins (Fig. ld-e).

At metamorphosis, small melanophores develop
along the anterior and posterior margins of all fin
spines and rays. Furthermore, there are blotches in
the dorsal fin membrane with higher concentration
of melanophores. In older specimens (Fig. le),
small punctate melanophores develop on the entire
body and fins; those dorsally on the head are a little
larger. The ventral melanophore on the basiptery-
gium is also lost.

REMARKS. Among the six Galapagos labrisom-
ids, only Starksia galapagensis has fully developed
larvae at sizes as small as 11.6 mm and metamor-
phic specimens smaller than 14 mm. Distinctive
characters are the elongate melanophore ventrally
on the basipterygium and a preanal length >47%
standard length (SL).

Dialommus fuscus (Gilbert, 1891)

Figure 2

GENERAL MORPHOLOGY. Description based
on 13 specimens; two larvae (8.4 and 10.2 mm)
described in detail. Compared with the other
known larval Galapagos labrisomids, the larvae of
this species are more slender, with a shorter preanal
length (Table 2). Neither specimen has developed
cirri. There are no preopercular spines, and appar-
ently they are absent in all larval Dialommus.

FIN DEVELOPMENT. Flexion and caudal-fin
ray development are complete at 8.4 mm, and the

dorsal-fin pterygiophores are developing posterior-
ly (Fig. 2a). At 10.2 mm, dorsal- and anal-fin spines
and rays are present, but most of the dorsal-fin
spines are poorly developed. Pectoral-fin rays are
not completely formed (Fig. 2b).

PIGMENTATION. Two well-separated melano-
phores develop on the principal caudal-fin rays near
the distal margins of the upper and lower hypural
plates but not in contact with them. A pair of small
melanophores develops in the interorbital region
anterior to the two large cephalic melanophores at
10.2 mm (Fig. 2b). Three small melanophores can
be found on the ventral margin of the caudal pe-
duncle.

REMARKS. Larvae of D. fuscus can be identified
by a long and slender body shape, a short preanal
distance, high myomere count (43), and distinctive
caudal-fin melanophores. In late larvae, the number
of spines and rays in the dorsal and anal fins is
diagnostic (Table 2).

Labrisomus dendriticus (Reid, 1935)
Figure 3

GENERAL MORPHOLOGY. Based on 176 lar-
vae (6.4-22.7 mm). Body is elongate with a shape
intermediate between the slender Dialommus fus-
cus and the more robust Starksia galapagensis.
Slope of the head is flat in early stages (Fig. 3a),
becoming more rounded (Fig. 3e). Preanal length
ranges from 40-45% SL. Cirri develop later than
in the other species of Labrisomus and Malacocten-
us. At 18.1 mm, nasal and orbital cirri are small
buds (Fig. 3d). By 22.7 mm, bifid nasal and orbital
cirri and three nuccal cirri are present (Fig. 3e); the
nuccal cirri are smaller in L. dendriticus than in L.
multiporosus and Malacoctenus.

FIN DEVELOPMENT. At 6.4 mm, a full com-
plement of anal-fin rays are developing, and the
dorsal-fin anlage is formed. Dorsal-fin spines begin

Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae ■ 5

6 ■ Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae

to develop posteriorly by 8.4 mm; all spines are
formed by 13.0 mm. By 18.1 mm, the first two
dorsal-fin spines are longer (Fig. 3d), as in adults.

SPINATION. Preopercular spines are developed
by 6.4 mm (Fig. 3a). They become embedded in
late-stage larvae and remain visible to at least 18.1
mm; they disappear by 22.7 mm (Fig. 3e).

PIGMENTATION. The most distinctive charac-
ter is the presence of a melanophore between the
hypural plates, close to the posterior border (Fig.
3a). A pair of melanophores develops on the upper
jaw at about 9.0 mm. A single pair of melano-
phores is present above the midbrain in larvae <7.0
mm, increasing to three pairs in larvae >16.0 mm
(Fig. 3a-d). At 6.4 mm, a ventral melanophore is
present on the gut (Fig. 3a), but it disappears before
8.4 mm.

At about 10 mm, two melanophores develop be-
hind the distal hypural margins. Two to four me-
lanophores are present on the ventral margin of the
caudal peduncle (Fig. 3a, 6.4 mm); one to two me-
lanophores develop on the dorsal margin after 10.0
mm (Fig. 3c-e). A continuous row of melanophores
develops along the pterygiophores of the dorsal fin,
spreading both caudad and cephalad from the an-
terior soft rays, beginning at 10.2 mm; in the larg-
est specimen, this row extends from the fourth or
fifth spine to the last dorsal-fin ray (Fig. 3e). At
22.7 mm, external melanophores form laterally on
the caudal peduncle (Fig. 3e), probably signaling
the beginning of metamorphosis.

REMARKS. Of the six species studied, Labriso-
mus dendriticus is the only one that has a mela-
nophore between the hypural plates and a contin-
uous dorsal row of melanophores on the trunk and
tail. Other distinguishing characters are small nuc-
cal cirri; presence of preopercular spines; a pair of
melanophores on the upper jaw; and a series of me-
lanophores along the posterior border of the hy-
pural plates. Also, L. dendriticus develops one or
two melanophores on the dorsal margin of the cau-
dal peduncle, a feature that is absent in larvae of
L. multiporosus.

The two species of Labrisomus have the same
configuration of spines and rays in the dorsal fin.
The terminal dorsal-fin spine is longer than the pre-
ceding spines and superficially appears to be part
of the segmented ray portion of the dorsal fin. Ma-
lacoctenus differs from Labrisomus in that the last
two dorsal-fin spines are longer than the preceding
ones and appear to belong to the segmented ray
portion of the fin.

Labrisomus multiporosus (Fiubbs, 1953)
Figure 4

GENERAL MORPHOLOGY. Based on 915 lar-
vae (5.4-18.4 mm). These larvae are morphologi-
cally similar to those of L. dendriticus. The largest
specimen, 18.4 mm, has bifid nasal cirri, a single
orbital cirri, and five nuccal cirri of moderate length
(Fig. 4f). A diagnostic feature of adults is a complex

cranial pore pattern, which begins to develop in
late-stage larvae. Pores form above the eye by 14
mm and proliferate as development proceeds, par-
ticularly around the orbit and laterally on the lower
jaw (Fig. 4e-f).

FIN DEVELOPMENT. The development of dor-
sal and anal fins is similar to that of L. dendriticus.
At 5.4 mm, only part of the dorsal-fin anlage is
present, and notochord flexion is complete. At 7.2
mm, dorsal and anal fins are partially formed. By
14.5 mm, the anteriormost four dorsal-fin spines
are of about uniform length (Fig. 4e— f), resulting in
the characteristic straight fin margin.

SPINATION. Three preopercular spines are pres-
ent in small larvae (5.4 mm), and the number in-
creases to four spines by 14.5 mm (Fig. 4e) and to
8 at 18.4 mm (Fig. 4f). The spines are not lost, as
occurs in other species; the adults of L. multipo-
rosus are characterized by the presence of preoper-
cular spines.

PIGMENTATION. Melanophores develop be-
tween the pterygiophores of the first three seg-
mented dorsal -fin rays at 11 mm (Fig. 4d). They
spread cephalad in a discontinuous row to the sev-
enth or eighth spine, and they spread caudad in a
continuous row to near the last ray.

Internal melanophores develop laterally on the
urostyle (Fig. 4e-f), and a series of external mela-
nophores develops along the posterior margin of
the hypurals (Fig. 4a) in larvae as small as 5.4 mm.
At 14.5 mm, a few small punctate melanophores
pepper the caudal fin membrane, mainly in the low-
er lobe (Fig. 4e). Two to five melanophores are
present on the ventral margin of the caudal pedun-
cle; these may coalesce, and by 18.4 mm, only two
large melanophores can be discerned (Fig. 4f).

Melanophores above the brain increase from one
pair at 5.4 mm to two pairs at 7.2 mm (Fig. 4a-b)
and to three pairs at 14.5 mm, with the appearance
of progressively more anterior pairs (Fig. 4e-f). At
5.4 mm, there is one ventral melanophore on the
gut and two below the anus; all of these disappear
before 7.2 mm (Fig. 4b). At 18.4 mm, three large
melanophores form ventrally on the gut and one
forms ventrolaterally below the pectoral base (Fig.

4f).

REMARKS. Among the six Galapagos labrisom-
id species, L. multiporosus is the only one that re-
tains preopercular spines as a juvenile. Further-
more, the number and size of spines increase in
larger larvae.

In contrast to L. dendriticus, L. multiporosus
lacks melanophores on the upper jaw and between
the hypural plates. Further, the development and
number of preopercular spines, the dorsal pigment
pattern, and the size of dorsal-fin spines and rays
serve to distinguish these two species. The two spe-
cies show differences in the time of development of
pigment along the hypural margin: in L. multipo-
rosus melanophores develop early (by 5.4 mm),
whereas in L. dendriticus they develop late (by 10
mm).

Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae ■ 7

Figure 4. Field collected larvae of Labrisomus multiporosus : (a) 5.4 mm (LACM 45623-5), (b)7.2 mm (LACM 45675-
4), (c) 8.3 mm (LACM 45675-4), (d) 11.8 mm (LACM 45621-6), (e) 14.5 mm (LACM 45623-5), (f) 18.4 mm (LACM
45623-5).

Malacoctenus tetranemus (Cope, 1877)

Figure 5

GENERAL MORPHOLOGY. Based on 228 lar-
vae (5.4-17.3 mm). The larval shape as in other
Labrisomus and Malacoctenus species. Preanal
length ranges from 41-45% SL. M. tetranemus de-
velops six long nuccal cirri, a bifid nasal cirrus, and
a bifid orbital cirrus. Two mandibular pores devel-
op laterally on the lower jaw in larger specimens
(Fig. 5d-e).

FIN DEVELOPMENT. Dorsal-fin spines begin to
develop at 9.0 mm. By about 16 mm, the spinous
dorsal fin is notched near its anterior and posterior
ends (Fig. 5d-e). In Malacoctenus, the first two el-
ements of the posteriorly lobed portion of the dor-

sal fin are spines. This feature is established by 10.2
mm in M. tetranemus, even before dorsal-fin spine
development is complete (Fig. 5 b-e).

SPINATION. At 5.4 mm, Malacoctenus tetra-
nemus has three preopercular spines (Fig. 5a); how-
ever, these disappear during development and are
no longer visible in larvae larger than 16 mm (Fig.
5d-e).

PIGMENTATION. A pair of melanophores de-
velops on the premaxillary in larvae as small as 5.4
mm (always present after 6.2 mm). Two pairs of
melanophores develop on the midbrain before 7.0
mm, and by 12.2 mm, two additional pairs have
appeared above the forebrain. These melanophores
increase in size with development (Fig. 5c-e).

A row of melanophores appears between the

8 ■ Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae

first pterygiophores of the segmented ray portion
of the dorsal fin by 10.2 mm (Fig. 5c) and subse-
quently spreads both cephalad and caudad (Fig.
5d-e). These melanophores are not arranged in
a continuous row, nor is their position on the
pterygiophores constant among individuals (Fig.
5d-e).

In larvae shorter than 7.0 mm, there is usually
no pigmentation on the ventral margin of the cau-
dal peduncle, but one small melanophore is present
in some specimens. Two melanophores are present
ventrally along the gut but disappear early by 10.2
mm (Fig. 5a). Melanophores typically are absent
along the hypural margin.

REMARKS. Among the six species treated here,
M. tetranemus has the largest and most numerous
(6) nuccal cirri. It is easily distinguished from other
species by the combination of preopercular spines,
the absence of pigmentation along the hypural mar-
gin, and the presence of melanophores on the upper
jaw. Further, M. tetranemus typically shows heavier
cranial pigmentation.

Malacoctenus zonogaster (Heller and
Snodgrass, 1903)

Figure 6

GENERAL MORPHOLOGY. Based on 414 lar-
vae (7.0-17.7 mm). Preopercular spines are absent.
Preanal length ranges from 39-41% SL. Malacoc-
tenus zonogaster develops a single orbital cirrus, a
bifid nasal cirrus, and three small nuccal cirri.

FIN DEVELOPMENT. Notochord flexion is

complete and segmented dorsal- and anal-fin rays
are forming by 7.0 mm (Fig. 6a). Dorsal-fin spines
begin to appear by 11.0 mm, and by 15.6 mm the
typical adult pattern of having the first two dorsal-
fin spines longer than the third and fourth spines is
present (Fig. 6d-e). The generic character of having
the last two dorsal-fin spines included in the seg-
mented ray portion of the dorsal fin is present at
12.2 mm (Fig. 6d).

PIGMENTATION. Cranial pigment consists of a
pair of melanophores on the frontals; another, larg-
er pair on the parietals; and an embedded mela-
nophore on the nape. A few smaller melanophores
sparsely scattered on the head increase in number
with growth (Fig. 6a-e).

A discontinuous row of melanophores develops
between the anterior pterygiophores of the seg-
mented ray portion of the dorsal fin at 12.2 mm
and spreads cephalad to the fifth dorsal-fin spine
and caudad (Figs. 6d-e), but not as far as the last
two segmented rays. Larvae larger than 15 mm
have a single, small, usually faint melanophore on
the upper jaw. In contrast, other labrisomids with
upper jaw pigment typically have a pair of mela-
nophores.

A series of two to four melanophores appears on
the ventral margin of the caudal peduncle after 7.0
mm (Fig. 6a), and this increases to four to six me-
lanophores after 9.0 mm (Fig. 6b-e). On the dorsal
peduncle margin, one or two melanophores devel-
op at about 10.0 mm and are retained in later lar-
val stages (Figs. 6b-e). Internal dorsal melano-

Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae ■ 9

Figure 5. Field collected larvae of Malacoctenus tetranemus : (a) 7.2 mm (LACM 45675-6), (b) 10.2 mm (LACM 45675-
6), (c) 12.2 mm (LACM 45623-7), (d) 16.1 mm (LACM 45643-5), (e) 17.3 mm (LACM 45625-14).

10 ■ Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae

Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae H 11

phores associated with the urostyle are present in
larvae larger than 9.0 mm (Fig. 6c-e). Hypural bor-
der pigmentation is present by 7.0 mm, initially as
one melanophore on the margin of each plate.
These increase in number and eventually overlap,
covering the entire hypural margin (Fig. 6d-e). At
18.4 mm, small melanophores develop on the lower
caudal rays (Fig. 6e).

REMARKS. Both species of Malacoctenus devel-
op pigmentation on the upper jaw, but M. zono-
gaster has a single medial melanophore that devel-
ops late (>15.0 mm), whereas M. tetranemus has
two that appear early (<6.0 mm SL). M. zonogas-
ter also differs from M. tetranemus in lacking pre-
opercular spines, in retaining the dorsal and ventral
pigmentation on the caudal peduncle throughout
larval life, and in having melanophores between
nearly all pterygiophores on the segmented-ray por-
tion of the dorsal fin, a pigmented hypural border,
smaller nuccal cirri, and a single orbital cirrus.

Those larvae of M. zonogaster that lack dorsal-
fin spines resemble the triplefin blenny, Lepidonec-
tes corallicola (Tripterygiidae). They can be distin-
guished from one another by the relative distance
between the snout and the origin of the soft portion
of the dorsal fin; this distance is 68-71% of SL in
M. zonogaster and 60-63% in L. corallicola.

DISCUSSION

Even though the coverage here is limited both in
number of species and stages of development ex-
amined, three larval types can be recognized ac-
cording to their shapes, which coincide with tradi-
tional labrisomid classification. For example, a typ-
ical feature is the preanal distance, which reaches
38-46% in species of Malacoctenus and Labriso-
mus (Labrisomini), 47-53% in Starksia (Starksiini),
and 33% in Dialommus (Mnierpini).

The larvae of Starksia galapagensis are morpho-
logically different from those of other labrisomids,
an observation that would be consistent with ex-
cluding the tribe Starksiini from the family Labri-
somidae. Based on molecular evidence, the tribe has
been considered to be more closely related to Clin-
idae (Stepien et al., 1993) or Chaenopsidae (Stepien
et al., 1997). The presence of a large and elongated
ventral melanophore at the basipterygium in S. gal-
apagensis, a striking feature of larval chaenopsids
(Brogan, 1992), suggests a close relationship with
the family Chaenopsidae.

The larvae of Dialommus fuscus differ from oth-
ers in this study in having a more elongate body
with a short preanal length (33%), a high number
of vertebrae (43), pigmented hypural plate margins,
a high number of spines in the dorsal fin (24-27),
and a high number of rays in the anal fin (26-28).
More comparisons are not possible because the
available larvae of this species were few and the size
range was narrow.

The species of Labrisomus and Malacoctenus
show specific arrays of preopercular spines and me-

lanophores on the upper jaw, dorsal margin of the
trunk, hypural plate border, and urostyle. In larger
larvae, the relative size of the first spines of the dor-
sal, and the number of spines included in the sec-
ond lobe of the dorsal fin are helpful in identifica-
tion. There are no characters that define Labriso-
mus and Malacoctenus as early larvae. However, in
larger specimens, an adult feature develops: the
posterior lobe of the dorsal fin includes one spine
in Labrisomus and two spines in Malacoctenus.

Watson (1996) described larvae of L. multipo-
rosus in the size range from 5.5-12.3 mm from the
California Current Region, which differ in some as-
pects with those from the Galapagos. These larvae
develop several relatively large melanophores
around the gut early in their development, whereas
larvae from the Galapagos do not have them in the
same size range. Some can be seen after 14.5 mm,
but this seems to be more a juvenile feature. Fur-
thermore, at the same stage of development, the
larvae of L. multiporosus from the Galapagos have
fewer preopercular spines than their counterparts
from the California Current Region. It remains to
be determined whether these differences are due to
variation within the same species, which has an ex-
tended geographic range in the Eastern Pacific, or
whether a separate species is present in the Gala-
pagos.

Brogan (1992) described larvae of two species of
Labrisomus (not identified) from the Gulf of Cali-
fornia, with one to three large ventral melano-
phores ventrally on the trunk and no preopercular
spines. Watson (1996) reports the same for L. xan-
ti. A generalized pattern for blennioids is the pres-
ence of a continuous series of melanophores asso-
ciated with the bases of the anal fin elements (Cav-
alluzzi, 1997), as it occurs in the two species from
the Galapagos illustrated in this study. Then, the
pigment pattern shared by the larvae of the three
species described by Brogan (1992) and Watson
(1996) would represent a derived condition sug-
gesting a closer relationship among them.

Brogan (1992) described the larvae of two spe-
cies of Malacoctenus and found that one of them,
M. hubbsi, never develops preopercular spines. Be-
cause this character is generalized in Malacoctenus
(and Labrisomus ), its absence suggests a closer re-
lationship between M. hubbsi and M. zonogaster.
The knowledge of the larvae of Malacoctenus and
Labrisomus is still limited. For each of these two
genera, there are about nine species in the eastern
Pacific and nine in the western Atlantic. With the
larvae of only five species of each genus known so
far, any generalization about relationships based on
ontogenetic characters is still speculative.

ACKNOWLEDGMENTS

We thank the staff of the Natural History Museum of Los
Angeles County (LACM) who have assisted us during our
studies of the Galapagos Islands fish larvae. We thank R.
Feeney and C. Thacker (LACM), M. Brogan (University
of Washington), and W. Watson (Southwest Fisheries Sci-

12 ■ Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae

ence Center) for critically reading the manuscript and of-
fering us numerous suggestions. We thank V. G. Springer
(National Museum of Natural History) for sharing his
knowledge of labrisomid fishes with us. We acknowledge
and thank the University of Southern California and the
Natural History Museum, for continued financial support.
Finally, we wish to acknowledge the National Science
Foundation for their support of larval fish curation at the
LACM, which made the collections available for study
(NSF DEB 8814791).

LITERATURE CITED

Brogan, M. W. 1992. Ecology of larval fishes around reefs
in the Gulf of California, Mexico. University of Ar-
izona, Tucson. Unpubl. Ph.D. Dissertation. 161 pp.
Cavalluzzi, M. 1997. Larvae of Gillelus jacksoni, G. ur-
anidea (Dactyloscopidae), Stahmonotus stabli tekla,
and S. hemphilli (Chaenopsidae), with comments on
the use of early life history characters for elucidating
relationships within the Blennioidei. Bull. Mar. Sci .,
60: 139-151.

Cope, E. D. 1877. Synopsis of the cold-blooded Vertebra-
ta secured by Prof. James Orton during his explo-
ration of Peru in 1876-77. Proc. Am. Phil. Soc. 17:
33-49.

Fraser, C. Me Lean. 1943. General account of the scien-
tific work of the Velero III in the eastern Pacific,
1931-41. Part III, a ten year list of the Velero III
collecting stations. Allan Hancock Pacific Expedi-
tions 1(3): 1-431.

Gilbert, C. H. 1891. A supplementary list of fishes col-
lected at the Galapagos Islands and Panama, with
description of one new genus and three new species.
Proc. U.S. Natl. Mus. 13(840): 449-445.

Grove, J. S., and R. J. Lavenberg. 1997. The Fishes of the
Galapagos Islands. Stanford, CA: Stanford Univer-
sity Press, 863 pp.

Hastings, P.A., and V.S. Springer 1994. Review of Stath-
monotus and phylogenetic analysis of the Chaen-
opsidae (Teleostei: Blennioidei). Smithsonian Contri-
butions to Zoology, 558: 1-48.

Heller, E., and R. E. Snodgrass. 1903. New fishes: Papers
of the Hopkins-Stanford Galapagos expedition
1898-1899. Proc. Wash. Acad. Sci. 5:189-229.
Hubbs, C. L. 1952. A contribution to the classification of
the blennioid fishes of the family Clinidae, with par-
tial revision of the Eastern Pacific forms. Stanford
Ichthyological Bulletin 4:41-165.

. 1953. Revision of the Eastern Pacific fishes of the

clinid genus Labrisomus. Zoologica 38:113-135.

Matarese, A. C., W. Watson, and E. G. Stevens. 1984.
Blennioidea: development and relationships. In On-
togeny and systematics of fishes, eds. H. G. Moser,
W. J. Richards, D. M. Cohen, M. P. Fahay, A. W.
Kendall, Jr., and S. L. Richardson, 565-573.
Lawrence, Kansas: American Society of Ichthyolo-
gists and Herpetologists.

Nelson, J. G. 1984. The Fishes of the World. New York:
Wiley. 523 pp.

Powles, H., and D. F. Markle. 1984. Identification of Lar-
vae. In Ontogeny and systematics of fishes, eds. H.
G. Moser, W. J. Richards, D. M. Cohen, M. P. Fahay,
A. W. Kendall, Jr., and S. L. Richardson, 31-33.
Lawrence, Kansas: American Society of Ichthyolo-
gists and Herpetologists.

Reid, E. D. 1935. Two new fishes of the families Dacty-
loscopidae and Clinidae from Ecuador and Galapa-
gos Islands. Copeia 1984(3):786-789.

Rosenblatt, R. H., and L. R. Taylor. 1971. The Pacific
species of the clinid fish tribe Starksiini. Pac. Sci. 25:
436-463.

Springer, V. G. 1959. A new species of Fabrisomus from
the Caribbean Sea, with notes on other fishes of the
subtribe Labrisomini. Copeia 1959:289-292.

. 1993. Definition of the suborder Blennioidei and

its included families (Pisces: Perciformes). Bulletin of
Marine Science 52(l):472-495.

Stephens, J. S., Jr., and V. G. Springer. 1973. Clinid fishes
of Chile and Peru, with description of a new species,
Myxodes ornatus, from Chile. Smithsonian Contri-
butions to Zoology 159:1-24.

Stepien, C. A. 1992. Evolution and biogeography of the
Clinidae (Teleostei, Blennioidei). Copeia 1992(2):
375-392.

Stepien, C. A., M. T. Dixon, and D. M. Hillis. 1993. Evo-
lutionary relationships of the blennioid fish families
Clinidae, Labrisomidae and Chaenopsidae: congru-
ence between DNA sequence and allozyme data.
Bulletin of Marine Science 52(1):496-515.

Stepien, C., A. Dillon, M. Brooks, K. Chase, and A. Hub-
ers. 1997. The evolution of Blennioid Fishes based
on an analysis of mitochondrial 12S rDNA. In Mo-
lecular Evolution of Fishes, eds. T. Kocher and C.
Stepien, 245-270.New York: Academic Press.

Watson, W. 1996. Blennioidei. In H.G. Moser (editor),
The Early Stages of Fishes in the California Current
Region, pp. 1148-199. CalCOFI Atlas No. 33.

Received 28 March 1994; accepted 16 July 1999.

Contributions in Science, Number 478

Herrera and Lavenberg: Larvae of the Galapagos Labrisomidae ■ 13

Natural History Museum

of Los Angeles County

900 Exposition Boulevard
Los Angeles, California 90007

A) Wr

Number 479
17 December 1999

Contributions
in Science

Upper Paleocene to Lower Eocene
(“Meganos Stage”) Marine Megafossils in
the Uppermost Santa Susana Formation,
Simi Valley, Southern California

Richard L. Squires

Natural History Museum

of Los Angeles County

Serial
Publications
ol THE
Natural History
Museum of
Los Angeles
County

Scientific

Publications

Committee

John Heyning, Acting Deputy Director
for Research and Collections
John M. Harris, Committee Chairman
Brian V. Brown
Kenneth E. Campbell
Kirk Fitzhugh
Karen Wise

Robin A. Simpson and K. Victoria Brown,

Managing Editors

The scientific publications of the Natural History Museum
of Los Angeles County have been issued at irregular in-
tervals in three major series; the issues in each series are
numbered individually, and numbers run consecutively, re-
gardless of the subject matter.

# Contributions in Science, a miscellaneous series of tech-
nical papers describing original research in the life and
earth sciences.

# Science Bulletin, a miscellaneous series of monographs
describing original research in the life and earth sci-
ences. This series was discontinued in 1978 with the
issue of Numbers 29 and 30; monographs are now
published by the Museum in Contributions in Science.

6 Science Series, long articles and collections of papers on
natural history topics.

Copies of the publications in these series are sold through
the Museum Book Shop. A catalog is available on request.

The Museum also publishes Technical Reports, a miscel-
laneous series containing information relative to scholarly
inquiry and collections but not reporting the results of
original research. Issue is authorized by the Museum’s Sci-
entific Publications Committee; however, manuscripts do
not receive anonymous peer review. Individual Technical
Reports may be obtained from the relevant Section of the
Museum.

Natural History Museum
of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

Printed at Allen Press, Inc., Lawrence, Kansas
ISSN 0459-8113

Upper Paleocene to Lower Eocene (“Meganos
Stage”) Marine Megafossils in the
Uppermost Santa Susana Formation,

Simi Valley, Southern California

Richard L. Squires1

ABSTRACT. Uppermost Paleocene to lowermost Eocene (“Meganos Stage”) marine rocks are rare on the
Pacific coast of North America, and the upper 100 m of the Santa Susana Formation in Simi Valley
represents the only known “Meganos Stage” rocks in southern California. This report concerns the first
detailed study of mega-invertebrate fossils in this part of the formation. The fossils are in thin lenses that
formed as storm-lag accumulations in a relatively shallow, offshore environment. Thirty localities yielded
a total of 38 megafossil taxa identifiable to species or subspecies. These include one isidid octocoral, one
solitary coral, two colonial corals, 17 gastropods, 14 bivalves, two crabs, and one spatangoid echinoid. A
new species of solitary coral, ? Antillopbyllia californica new species, is described and named. A possible
new species of the colonial coral Astrocoenia is described. Turritella andersoni susanae is locally abundant
and one of the most diagnostic species of the “Meganos Stage” in the Simi Valley area.

Most of the studied megafauna is known from elsewhere in Pacific coast Eocene strata, primarily in
California. The molluscan-stage ranges of approximately one third of the studied mega-invertebrates are
extended downward to the “Meganos Stage,” based on their presence in the upper 100 m of the Santa
Susana Formation. The molluscan-stage ranges of the gastropod Ringicula (R.) pinguis, the bivalve Saulella
undulifera, and the crab Cyclocorystes aldersoni are extended upward from the Paleocene “Martinez Stage”
to the “Meganos Stage.” In addition, a few of the mega-invertebrates have their geographic ranges ex-
tended. The isidid octocoral tMopsea sp., aff. M. costata is the first record of an isidid from Paleogene
rocks on the Pacific coast of North America. The studied megafauna is indicative of warm-water conditions,
and some of the taxa are conspecific with or closely allied to Old World Tethyan mollusks.

INTRODUCTION

Since the early 1900s, paleontologists have collect-
ed marine megafossils from the Santa Susana For-
mation in the Simi Valley area (Fig. 1), Ventura
County, southern California. These fossils, which
are predominantly mollusks, have been studied by
Waring (1917), Nelson (1925), Fantozzi (1955),
and Zinsmeister (1983a, 1983b), but their studies
dealt only with “Martinez Stage” rocks of late Pa-
leocene age in the lower part of the formation on
the south side of Simi Valley. Although various
workers have found a few mollusks in the upper
100 m of the Santa Susana Formation, to date there
has not been a comprehensive study of these fossils.
Poor exposures, absence of fossils at many locales,
and generally poor preservation have discouraged
such study, and prior to this study no one had ever
“walked out” the stratigraphic interval throughout
the Simi Valley area. In addition, outcrops of the
upper 100 m of the Santa Susana Formation have
the same color and general appearance of essen-

1. Department of Geological Sciences, California State
University, Northridge, California 91330-8266, and Re-
search Associate in Invertebrate Paleontology, Natural
History Museum of Los Angeles County, Los Angeles,
California 90007-4000.

Contributions in Science, Number 479, pp. 1-38
Natural History Museum of Los Angeles County, 1999

tially unfossiliferous outcrops stratigraphically low-
er in the Santa Susana Formation. Many previous
workers seemed to have assumed that the upper
100 m are not particularly fossiliferous. Locally,
however, megafossils are abundant there, and their
preservation can be good. The purpose of this pres-
ent study is to document the taxonomic composi-
tion of the megafossil content in these upper 100
m. This information will help greatly in refining the
molluscan stage ranges of these taxa because the
upper 100 m of this formation correlate to the
“Meganos Stage” of latest Paleocene to earliest Eo-
cene age. Rocks deposited on the Pacific coast of
North America during this time interval are rela-
tively rare, and those in the study area represent
the only known “Meganos Stage” rocks in southern
California.

The molluscan stage terminology used in this re-
port includes the upper Paleocene “Martinez
Stage,” the uppermost Paleocene to lowermost Eo-
cene “Meganos Stage,” the middle lower “Capay
Stage,” the upper lower to lower middle “Domen-
gine Stage,” the lower middle “Transition Stage,”
and the middle middle Eocene to upper Eocene “Te-
jon Stage.” The stage names are in quotes because
they are informal terms and in essence equivalent
to formation names. Clark and Vokes (1936) gave

a historical overview of these stage names. Givens
(1974) modified the use of the “Capay Stage,” and
it is in this modified sense that the “Capay Stage”
is used here. Saul (1983) and Squires (1984, 1987,
1988a) regarded the modified “Capay Stage” of
Givens (1974) as middle lower Eocene.

STRATIGRAPHY

The upper 100 m of the approximately 1000-m-
thick Santa Susana Formation in the Simi Valley
area consists mostly of gray, very fine-grained sand-
stone, with some gray muddy to sandy siltstone.
Locally, there are concentrations of fossil-shell
hash, and rocks surrounding these localized con-
centrations are barren of megafossils. The lower
part of the upper 100 m of the Santa Susana For-
mation has a gradational lithology from the under-
lying gray mudstone and siltstone. The Santa Su-

sana Formation is disconformably overlain by basal
conglomerate of the Llajas Formation (Fig. 2), and
the contact is an uneven erosion surface with as
much as 1 m of relief. Thin beds of laminated sand-
stone alternating with thin beds of bioturbated
sandstone containing vertical Ophiomorpha bur-
rows are present at the top of the Santa Susana
Formation at two places on the north side of the
valley. One location is at CSUN locality 968, and
the other location is just north of the Marr Ranch
(Fig. 3) in the northeast corner of section 31, T3N,
R17W, where the type section of the overlying Lla-
jas Formation is located (see Squires, 1981:fig. 3)
(Fig. 3).

Sedimentary rocks in the upper 100 m of the San-
ta Susana Formation are not resistant and are usu-
ally poorly exposed or covered. The best exposures
are on the north side of Simi Valley, where 19 fossil-

2 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

<L>

U)

03

+->

00

>>

03

Cl

03

u

Q)

O)

03

4->

00

1/3

o

c

03

CD

<D

<D

C

0)

o

o

_0>

03

CL

L_

<D

Q.

CL

ZD

(U

03

03

-t-J

00

N

CD

C

CL

u

CD

Q_

(~>

CD

CL

O

00

CL

CJ

00

CL

O

03

C

03

CO

=3

00

c

03

00

O)

c

o

O

co

ro

_Q

disconformity

►968

• LACMIP 26615

• 1349

• LACMIP 26611

• LACMIP 21550

• 1565

• 1347,1348

• 1344

• 958

• 959,960-967

• 1345,1346

• 967-973

• 1343

• 1342, LACMIP
71 24,UCMP
7009

20 m

base not measured

Figure 2. Stratigraphic column of the upper 100 m of the
Santa Susana Formation (and of the immediately under-
lying and overlying strata), showing subseries, provincial
molluscan stages, calcareous nannofossil biozones (CP
Zones), and stratigraphic position of each of the mega-
fossil-collecting localities. Age of the lower part of the Lla-
jas Formation from Squires (1984).

collecting localities have been found (Fig. 3). CSUN
localities 962 through 967 are approximately
aligned along the same strike and form a series of
localities that parallels the top of the Santa Susana
Formation. CSUN localities 969 through 973 are
similarly aligned. Four localities (CSUN 968, LAC-
MIP 21550, LACMIP 26611, and LACMIP 26615)
are near the top of the formation. At nearly all of
the localities on the north side of Simi Valley, fossils
are concentrated in lenses about 20 cm thick and
several meters in lateral extent. The tops and bot-
toms of the lenses are indistinct. The bivalves in the
lenses are almost always single valves and usually
broken, and many of the associated gastropods
consist of fragments. Although the fossils are most-
ly broken, none shows any obvious signs of abra-
sion. Specimens of Turritella andersoni susanae
Merriam, 1941 are plentiful and dominate the oth-
erwise usually meager megafauna. At CSUN local-

ity 959, there is a thin lens containing abundant
Turritella uvasana inf era Merriam, 1941 and the
lens is traceable laterally for about 6 m. This lo-
cality is now under home sites.

On the south side of Simi Valley, 11 fossil-col-
lecting localities have been found in the upper 100
m of the Santa Susana Formation (Figs. 1, 4). An
isolated locality (CSUN 1565) is in Bus Canyon,
and another isolated locality (UCMP loc. 7009) is
near Meier Canyon (Fig. 1); the other localities are
in two groups, with six localities found mostly near
the top of the Santa Susana Formation in the Run-
kle Canyon area and three localities just east of the
Runkle Canyon fault (Fig. 4). At most of these lo-
calities on the south side of Simi Valley, the fossils
are similar to those on the north side of the valley,
and there are concentrations of unabraded fossil-
shell hash in thin lenses whose tops and bottoms
are indistinct. Abundant weathered-out specimens
of the solitary coral ? Antillophyllia California new
species, abundant tips (apices) of turritellas, and
commonly occurring mostly right valves of the bi-
valve Pycnodonte ( Phygraea ) sp., aff. P. (P.) pacifica
Squires and Demetrion, 1990 were found in float
at CSUN locality 1343.

The relative stratigraphic position of each me-
gafossil locality in the upper 100 m of the Santa
Susana Formation, on both the north and south
side of Simi Valley, is show in Fig. 2.

DEPOSITIONAL ENVIRONMENT

Parker (1983) did the most current and detailed
study of the depositional environment of the Santa
Susana Formation and reported that the vertical se-
quence of the Simi Conglomerate, Las Virgenes
Sandstone, and Santa Susana Formation represents
a transition from nonmarine to deep-marine facies.
The distribution of these facies is, in part, defined
by the Runkle Canyon-Burro Flats fault zone in the
central part of the south side of Simi Valley. Parker
(1983) referred to strata west of this fault zone as
the “western facies,” and these include, from base
to top, braided river, meandering stream, near-
shore, transition zone, offshore to shelf, and slope
deposits. He referred to strata east of this fault zone
as the “eastern facies,” and these consist mostly of
slope and inner-fan deposits. The fault zone juxta-
poses coeval nonmarine and deep-marine rocks and
accounts for the fact that nonmarine to nearshore
Las Virgenes Sandstone, for example, is present
only west of the fault (Fig. 5).

Parker (1983) reported that although the Santa
Susana Formation was mostly deposited in deep
water, the sedimentology of the upper 100 m of the
formation throughout the Simi Valley area reflects
uplift to shelf depths prior to deposition of the non-
marine basal part of the Llajas Formation. New
sedimentologic information obtained during this
present study shows that the shallowing event Park-
er (1983) recognized produced, toward the top of
the Santa Susana Formation, sediments deposited

Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils ■ 3

Figure 3. Geologic map showing location of megafossil localities (CSUN localities unless otherwise noted) on the north
side of Simi Valley. Base map is U.S. Geological Survey, 7.5-minute, Santa Susana quadrangle, 1951 (photorevised 1969),
Ventura County, southern California. Geology from Squires (1983b).

in progressively more shallow water. Beds of alter-
nating laminated sandstone and bioturbated sand-
stone (e.g., at CSUN loc. 968) were found at the
top of the formation, and this alternation of lithol-
ogies, which is also present in the shallowest marine

part of the overlying Llajas Formation, is charac-
teristic of modern and ancient shoreface to upper
offshore environments (Squires, 1981).

Heitman (1983), on the basis of benthic forami-
niferal assemblages, also reported that the upper-

4 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

Figure 4. Geologic map showing location of megafossil localities (CSUN localities unless otherwise noted) on the south
side of Simi Valley. Base maps are U.S. Geological Survey, 7.5-minute Calabasas quadrangle, 1952 (photorevised 1967);
and Santa Susana quadrangle, 1951 (photorevised 1969), Ventura County, southern California. Geology from Squires
(1983b).

most part of the Santa Susana Formation represents
a shoaling event associated with basin filling that
deposited a silty sandstone just above the shelf-
slope break. Filewicz and Hill (1983), on the basis
of calcareous nannofossil datums and sediment-ac-
cumulation rates, reported that silty sandstones in
the upper Santa Susana Formation (20 m below the
base of the overlying Llajas Formation) were de-
posited just above the shelf-slope break.

Squires (1991a) and Saul and Squires (1997) in-
terpreted that megafossils in the upper 100 m of
the Santa Susana Formation on the north side of
Simi Valley are shallow-marine forms deposited as
storm-lag accumulations in a relatively shallow, off-
shore environment. This more detailed present
study further supports these earlier interpretations.

Squires (1990) suggested that the fossiliferous
lens at CSUN locality 1342, in the upper 100 m of
the formation on the south side of Simi Valley, rep-
resents a storm deposit in a middle to outer shelf
environment, and that the distance of postmortem

transport of the shallow-marine mollusks was not
great.

At all of the fossil localities mentioned in this
present report, fossils occur as concentrations in
small lenses of sandstone surrounded by siltstone
usually barren of megafossils. The tops and bot-
toms of these lenses are indistinct, and Squires
(1981) reported similar lenses in the overlying Lla-
jas Formation, where the shallow-marine facies
grades into outer shelf to slope facies. In this tran-
sitional paleoenvironment, the contacts of the fos-
siliferous lenses were rendered indistinct by the ac-
tivities of burrowing organisms shortly after depo-
sition of the sediments, undoubtedly just like those
in the upper 100 m of the Santa Susana Formation.

It is readily apparent that the fossils in the upper
100 m of the Santa Susana Formation have under-
gone some postmortem transport, based on their
broken condition and localized concentrations. The
distance of this transport was not great based on
the absence of any evidence of significant abrasion.

Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils H 5

Waring
1917 1

Clark

1918

Clark

1921,

Kew

1924

Nelson

1925

Clark

1926

Parker, 1983,
Saul, 1983, &
This Report

Mollus k
"Stages"

W E

W E

W E

Tejon

Martinez

Tejon

Meganos

Martinez

Meganos

Fm.

Domengine

Fm.

Domen-

gine

Fm.

Llajas

Fm.

"Domen-

gine"

"Capay"

Santa

Susana

Fm.

upper 1 00 m of
Santa Susana Fm.

"Meganos"

Martinez

Fm.

Martinez £
marine c
member c

r

C_

(

1

5

- Martinez
5 marine
i' member

j>

D

ȣ

5

£

Santa

Susana

Shales

Santa Susana
Fm.

3

re

"Martinez’

Las c

Virgenes

Sandstone

Las

Virgenes

Sandstone

er

a>

I^~

Simi Conglomerate

Simi Conglom.

Chico

Chico Fm

Chico

Chatsworth Fm.

Campanian

Figure 5. Comparative concepts of lower Paleogene stratigraphic units in the Simi Valley area. The letters “W” and “E”
refer to west and east of the Runkle Canyon fault (shown in Fig. 1).

In addition, the fossils are ecologically similar. Buri-
al was rapid because there is a scarcity of epibionts.
Most of the fossils represent storm-lag accumula-
tions in a middle-to-outer shelf environment, where
fine sands accumulated in close proximity to silt-
stone. Bioturbation must have obliterated any ini-
tially sharp contacts that would have been formed
along the bottoms of the fossiliferous lenses.

CSUN locality 1343, however, is unusual because
the abundant specimens of the solitary coral? An-
tillophyllia calif ornica new species found there con-
stitute a growth series. These corals must have lived
in close proximity to where they were buried and
experienced much less distance of postmortem
transport than the other associated fossils.

In summary, the overall fine-grained rock type,
the assemblages of benthic foraminifera, the local-
ized concentrations of unabraded shallow-marine
megafossils, and alternating rock types at the top
of the formation indicate that the bulk of the upper
100 m of the Santa Susana Formation in the Simi
Valley area was deposited in a relatively shallow,
offshore environment (middle-to-outer shelf
depths). Storm surges transported the megafossils,

but the distance of transport was not far. Beds at
the top of the formation were deposited in a shal-
lower environment (shoreface).

MEGAFAUNA

A total of 560 megafossil specimens were collected
from 30 localities. Preservation ranges from poor
to good, and many of the fossils are badly weath-
ered or in hard concretions. Thirty-eight megafossil
taxa were identified to genus and species/subspe-
cies. These include one octocoral, one solitary cor-
al, two colonial corals, 17 gastropods, 14 bivalves,
two crabs, and one spatangoid echinoid. Other
taxa too poorly preserved for generic identification
include one solitary coral, one scaphopod, three
gastropods (a naticid, a cymatiid, and a turrid), and
four bivalves (a pinnid, a lucinid, a pitarinid, and
a solenid). The poorly preserved solitary coral was
mentioned by Durham (1943:199, pi. 32, fig. 22),
who reported two specimens from UCMP locality
7000 (exact location not known). Only the exterior
morphology of this coral is known. It is alate and
somewhat flabelliform, but better preserved speci-

6 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

mens and studies of their interiors are needed to
determine the familial or generic assignments. No
new specimens were found during this present
study. Durham (1943) identified the coral as Fla-
bellum stantoni Durham, 1943, a name that he also
gave to specimens that Vaughan (1900:67-68, pi.
4, figs. 5, 6) had misidentified as Flabellum remon-
dianum Gabb (1864:207, pi. 26, fig. 199).
Vaughan’s specimens were probably collected from
“Martinez Stage” rocks near Benicia in northern
California. Durham (1943) was probably correct in
giving the name F. stantoni to Vaughan’s specimens
from Benicia, but whether or not this name is ap-
propriate for the two poorly known coral speci-
mens from the upper part of the Santa Susana For-
mation is an open question. These two coral spec-
imens are not the same as the new species of soli-
tary coral described here.

Although early workers (Clark, 1921:table 1;
1926:114-116; Kew, 1924:25; Nelson, 1925:pl. 61,
and foldout between pages 402 and 403) gave fau-
nal lists for so-called “Meganos” age-strata in the
Simi Valley area, there are serious problems in try-
ing to use these lists because (1) detailed locality
information is lacking, and (2) previously used for-
mation names (e.g., Meganos and Domengine) are
not exact equivalents of the upper 100 m of the
Santa Susana Formation (Fig. 5). In addition, be-
cause these species names have only been listed,
without illustration or reference to catalogued mu-
seum specimens, there is no way to verify the iden-
tifications.

Three previously named megafossil species have
their type localities in the upper 100 m of the Santa
Susana Formation in the Simi Valley area. They are
the gastropod Corsania (Januncia ) susana Saul and
Squires, 1997 and the bivalves Area (Area) filewiezi
Squires, 1991a and Netastoma squiresi Kennedy,
1993. In addition, the following two species could
probably be added to this list. Although the de-
scription of the type locality of the bivalve Veneri-
eardia ( Pacificor ) susanaensis Verastegui, 1953, is
somewhat vague, this locality is most likely in the
upper 100 m of the formation. As will be discussed
under “Age,” the type locality of the gastropod Tur-
ritella andersoni susanae was reported (Merriam,
1941) from the basal part of the Llajas Formation,
but the actual stratigraphic position of this locality
is probably in the upper 100 of the Santa Susana
Formation.

AGE

The upper 100 m of the Santa Susana Formation is
latest Paleocene through earliest Eocene (“Meganos
Stage”) in age, based on calcareous nannofossils,
mollusks, sporomorphs, and magnetostratigraphy.
Assignments to the calcareous nannofossil biozones
and provincial Pacific coast of North America low-
er Paleogene molluscan stages are shown in Fig-
ure 2.

The term “Meganos Stage” stems from Clark
Contributions in Science, Number 479

(1918), and the type section of this stage is in the
Deer Valley area north of Mount Diablo, Contra
Costa County, northern California, where Clark
(1918, 1921) subdivided the Meganos Formation
into five major lithologic members. Starting at the
base, he designated the members as Divisions A, B,
C, D, and E. The molluscan fauna from this series,
which was studied in detail by Clark and Woodford
(1927), comes from Division D. Almgren et al.
(1988: fig. 4) assigned Division D strata of the Me-
ganos Formation to the CP9 Zone (lowest Eocene)
of the standard calcareous nannofossil zonation.
Most of the “Meganos Stage” correlates with the
CP9 Zone, but the lower part of the stage correlates
to the CP8 Zone (uppermost Paleocene). The age
of the “Meganos Stage,” therefore, corresponds to
the latest Paleocene to earliest Eocene (Saul, 1983;
Squires, 1988a, 1997). Modern workers now refer
to the strata of Division D of the Meganos For-
mation as the Margaret Hamilton Sand (Edmond-
son, 1984).

Filewicz and Hill (1983:fig. 5) reported calcare-
ous nannofossils diagnostic of the lower Eocene
Diseoaster diastypus (CP9) Zone of the standard
calcareous nannofossil zonation from siltstones im-
mediately below the upper 100 m of the Santa Su-
sana Formation on the north side of Simi Valley.
The upper 100 m on the north side of Simi Valley
are barren of any calcareous nannofossils (Filewicz
and Hill, 1983), as well as any planktonic forami-
nifera (Heitman, 1983). Filewicz and Hill (1983)
did not study the upper 100 m of this formation on
the south side of Simi Valley. In the course of this
present study, I collected 12 microfossil samples
from the upper 100 m of the Santa Susana For-
mation on the south side of Simi Valley. The cal-
careous nannofossils are rare, poorly preserved,
and, as with those from the north side of Simi Val-
ley, only present near the bottom of the 100-m-
thick interval, where the siltstone content is higher.
Only two samples, those from CSUN localities
1342 and 1343, yielded any calcareous nannofossil
data pertinent to geologic age determination. Both
samples yielded a late Paleocene age, probably
equivalent to the Diseoaster multiradiatus (CP8)
Zone of the standard calcareous nannofossil zona-
tion (M. V. Filewicz, personal communication). No
planktonic foraminifera were recovered from any
of the 12 microfossil samples (H. L. Heitman, per-
sonal communication).

As mentioned earlier, the stratigraphic nomencla-
ture of the upper part of the Santa Susana Forma-
tion in the Simi Valley area has been confusingly
inconsistent. Similarly, the concept of the “Mega-
nos Stage” in this area has lacked biostratigraphic
precision. Saul’s (1983) study of the turritellas and
venericardias in the various Paleogene formations
in the Simi Valley area helped greatly in clarifying
the proper assignment of provincial molluscan stag-
es to these formations. She reported that Turritella
andersoni susanae, T. uvasana infera, T. megano-
sensis Merriam, 1941 and Venericardia (Paeifieor)

Squires: “Meganos Stage” Marine Megafossils ■ 7

hornii susanaensis are important megafaunal com-
ponents of the upper 100 m of the Santa Susana
Formation and enable correlation of these rocks to
the northern California-based concept of the “Me-
ganos Stage.” Turritella meganosensis, furthermore,
is also found at the type section (Margaret Hamil-
ton Sand) of this stage in northern California. Tur-
ritella andersoni susanae is locally plentiful in the
upper 100 m of the formation and is probably con-
fined to the “Meganos Stage.” A single specimen,
which is the holotype of this gastropod, however,
was reported from the basal part (“Capay Stage”)
of the overlying Llajas Formation on the north side
of Simi Valley. It seems highly probable that the
stratigraphic position of this locality is in error, and
that the holotype is actually from the upper part of
the Santa Susana Formation. Turritella uvasana in-
fera is found in both the “Meganos Stage” and in
the overlying “Capay Stage.”

The only well-described (Clark and Woodford,
1927) megafauna of “Meganos” age is that of the
Margaret Hamilton Sand at the type section of the
“Meganos Stage” in Deer Valley, Contra Costa
County, although strata (“Meganos Formation”) of
this age are also known from south of Round Val-
ley on the Middle Fork of Eel River in Mendocino
County, northern California. Other than Turritella
meganosensis, species found in both the Margaret
Hamilton Sand and the upper 100 m of the Santa
Susana Formation are Calyptraea diegoana (Con-
rad, 1855) [= Calyptraea ( Galerus ) calabasaensis
Nelson, 1925], Bracbyspbingus mammilatus Clark
and Woodford, 1927, Nuculana ( Saccella ) gabbii
(Gabb, 1869) [= Leda gabbii (Gabb, 1869)], and
Schizaster diabloensis Kew, 1920. Gemmula sp.,
aff. G. diabloensis Clark and Woodford, 1927 from
the upper 100 m of the Santa Susana Formation
has close affinity to G. diabloensis from the Mar-
garet Hamilton Sand.

Frederickson (1983) reported early Eocene spo-
romorphs from the upper 100 m of the Santa Su-
sana Formation on the north side of Simi Valley.
The presence of these early Eocene sporomorphs is
best documented near the top of the upper 100 m
of the formation.

Bottjer et al. (1991) reported that, in terms of
magnetostratigraphy, the lowermost part of the up-
per 100 m of the Santa Susana Formation on the
north side of Simi Valley is correlative to Chron
C24R, which is equivalent to an interval that en-
compasses the Paleocene-Eocene boundary (Berg-
gren et al., 1995).

Simi Valley is part of the Western Transverse
Ranges tectonic block, which has been rotated by
Neogene displacements. Before rotation, this block
lay adjacent to San Diego and Anaheim in southern
California. During the Miocene, the block was ro-
tated clockwise. Its northern end (which included
Simi Valley) acted as the pivotal area and remained
essentially in place (Atwater, 1998). The Eocene pa-
leolatitude of the Simi Valley area, therefore, was
probably not much different than it is today. De-

position of the upper 100 m of the Santa Susana
Formation was also coincident with an overall
global sea-level rise (supercycle TA2 of Haq et al.,
1987).

The latest Paleocene to earliest Eocene age (“Me-
ganos Stage”) of the upper 100 m of the Santa Su-
sana Formation corresponds to an absolute age of
53 to 55 Ma (Berggren et al., 1995). Based on their
presence in the upper 100 m of the Santa Susana
Formation, the molluscan-stage ranges of nine spe-
cies can be extended downward to the “Meganos
Stage.” These species are: V elates perversus (Gme-
lin, 1791), Pacbycrommium clarki (Stewart, 1927),
Arcbitectonica (A.) llajasensis Sutherland, 1966,
Cylicbnina tantilla (Anderson and Hanna, 1925),
Spondylus carlosensis Anderson, 1905, Macoma
rosa Hanna, 1927, Pitar uvasana coquillensis Turn-
er, 1938, Corbula ( Caryocorbula ) dickersoni Weav-
er and Palmer, 1922, and Corbula ( Caryocorbula )
parilis Gabb, 1864. Similarly, the presence of As-
trocoenia sp. and Pycnodonte { Phygraea ) sp., aff.
Pycnodonte (Pbygraea) pacifica Squires and De-
metrion, 1990, in the upper 100 m of the formation
represents the earliest occurrences of Astrocoenia
and Pbygraea in the Paleogene rock record of the
Pacific coast of North America.

The molluscan-stage ranges of three species in
the study area can be extended upward from the
Paleocene “Martinez Stage” to the “Meganos
Stage.” These species are: Ringicula ( R .) pinguis
(Gabb, 1864), Saulella undulifera (Gabb, 1869),
and Cyclocorystes alder soni Squires, 1980.

PALEOCLIMATE AND

PALEOBIOGEOGRAPHY

As reviewed by Squires (1987), the world climate
was relatively warm and equable during most of
the Paleocene through early middle Eocene time,
and a worldwide late Paleocene warming trend cul-
minated in a period of peak warming during the
early Eocene. Also reviewed by Squires (1998), dur-
ing late Paleocene to early middle Eocene time, hu-
mid tropical climatic conditions were prevalent in
coastal-lowland areas from Baja California, Mexi-
co, to southwestern Washington.

Megafossils in the upper 100 m of the Santa Su-
sana Formation strongly support the presence of
tropical waters. The hermatypic colonial coral ge-
nus Astrocoenia lives today only in the West Indies
and is a reef dweller in shallow, tropical seas (Dur-
ham, 1942). Although Paleocene and early Eocene
reef corals seldom formed true reefs, by middle
through late Eocene times they started to build
reefs, and Astrocoenia was one of these reef build-
ers in the Caribbean region (Budd et al., 1992). The
extinct gastropod genus Corsania is indicative of
warm climate (Saul and Squires, 1997), and mod-
ern Campanile is also indicative of warm waters
and very shallow depths (Squires, 1993). The bi-
valve Area s.s. most frequently inhabits tropical
waters today (Keen, 1971; Abbott and Dance,

8 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

1982), and the same is true of the bivalves Spon-
dylus (see Squires, 1984:table 5) and Fimbria (see
Nicol, 1950).

The late Paleocene and early Eocene were times
of major immigration of Old World Tethyan mega-
invertebrates into the Pacific coast region of North
America via a seaway, most likely through the Cen-
tral America seaway (Clark and Vokes, 1936; Giv-
ens, 1978, Zinsmeister, 1983a; Squires, 1984,
1987). One of the most important of these is the
gastropod Velates perversus, which dispersed west-
wardly from Pakistan into California (Squires,
1987; Squires and Demetrion, 1992). The new oc-
currence of this species in the “Meganos”-age upper
100 m of the Santa Susana Formation provides it
with a slightly earlier arrival date in California.

Additional mollusk taxa in the upper part of the
Santa Susana Formation that have been recognized
as Tethyan or of Tethyan affinity (Clark and Vokes,
1936; Squires, 1984; 1987, 1990, 1991a; Givens,
1989) and must have accompanied Velates perver-
sus into the waters of the Pacific coast of North
America are Pachycrommium, Area s.s., and Fim-
bria. Although Campanile is also indicative of Old
World Tethyan connections, this genus had already
arrived onto the Pacific coast region by the late Pa-
leocene (“Martinez Stage”) (Squires, 1993).

As will be indicated in the “Systematics” section,
some other megafossils found in the upper part of
the Santa Susana Formation are strongly suggestive
of Old World Tethyan connections. The octocoral
IMopsea sp., aff. M. costata Milne-Edwards and
Haime, 1850 is remarkably similar to M. costata
from the lower Eocene London Clay in southern
England. The crab Cyclocorystes aldersoni Squires,
1980 is most similar to C. pulchellus Bell, 1858,
also from the lower Eocene London Clay. Cyclo-
corystes is only known from these two species. The
gastropod ? Ancillarina sp., which might be the only
known record of this genus in the Western Hemi-
sphere, is most similar to A. canalifera (Lamarck,
1802) from the Paris Basin, France.

SYSTEMATIC MATERIALS AND METHODS

Systematic arrangement of the higher taxa follows that of
Bayer (1956) for the octocoral, Wells (1956) for the scler-
actinians, Vokes (1980) for the bivalves, and Glaessner
(1969) for the crabs. The higher classification of gastro-
pods is in a state of flux, and some of the categories used
here for suprafamilial names are referred to as superorders
and generally correspond to major clade names used by
Ponder and Lindberg (1996, 1997).

Synonymies (primarily including only figured speci-
mens), primary type material, molluscan stage range, and
geographic distribution data are given for the identifiable
species. Terms used to denote specimen abundance are de-
fined as follows (number of specimens in parentheses):
rare (1-4), uncommon (5-9), common ( 1 0—29), and abun-
dant (30 or more). Abbreviations for catalog and/or lo-
cality numbers are:

ANSP: Academy of Natural Sciences,

Philadelphia

ANSP CAS: California Academy of Sciences,

San Francisco

ANSP CSUN: California State University,

Northridge

ANSP LACMIP: Natural History Museum of Los
Angeles County, Invertebrate Paleontology
Section

ANSP UCMP: University of California, Museum
of Paleontology, Berkeley
ANSP UCR: University of California, Riverside
ANSP USNM: United States National Museum,
Washington, D.C.

The bulk of the collections used in this study are housed
at CSUN. New species primary type material and hypo-
types of the invertebrates used for illustrations in this re-
port are deposited at LACMIP.

SYSTEMATICS

Phylum Cnidaria Hatschek, 1888
Class Anthozoa Ehrenberg, 1834pqc
Subclass Octocorallia Haeckel, 1866
Order Gorgonacea Lamouroux, 1816
Family Isididae Lamouroux, 1812
Genus Mopsea Lamouroux, 1816

TYPE SPECIES. Isis dichotoma Linnaeus, 1758,
by subsequent designation Milne-Edwards and
Haime, 1850; Recent, Antarctica.

}Mopsea sp., aff. M. costata Milne-Edwards
and Haime, 1850
Figures 6, 7

LOCAL OCCURRENCE. CSUN loc. 1343.

REMARKS. Twenty-six fragments were collect-
ed, and they consist of straight-to-curved, cylindri-
cal calcareous internode axial stems up to 23 mm
long. The stems are longitudinally marked by close-
ly spaced ribs that bear small spines.

Few published reports exist describing fossil oc-
tocoral remains. The isidid octocorals are probably
the most common forms preserved because of ex-
tensive calcification of the axis (Kocurko, 1988).
Unfortunately, generic determinations of isidid oc-
tocorals cannot be reliably made using only calcar-
eous internode-stem material. Soft-part morpholo-
gy and microscopic spicular material are also need-
ed (Bayer, 1956, and personal communication).

The Santa Susana Formation specimens are re-
markably similar to the calcareous axial parts of
Mopsea costata Milne-Edwards and Haime, 1850,
from the lower Eocene London Clay in southern
England. The Santa Susana Formation specimens
differ from M. costata Milne-Edwards and Haime
(1850:42, pi. 7, figs. 3, 3a) only by having more
elongate spines. To a slightly lesser degree, the San-
ta Susana Formation specimens also resemble Mop-
sea encrinula (Lamarck, 1815) that lives today in
New Caledonian waters. The Santa Susana For-
mation specimens differ from M. encrinula, which

Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils ■ 9

10 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

has been illustrated by Bayer and Stefani (1987:65-
66, pi. 21, figs. 1-2), by having more closely spaced
ribs that bear blunter spines.

The Santa Susana Formation specimens of }Mop-
sea sp., aff. M. costata represent the first record of
an isidid octocoral from Paleogene rocks on the Pa-
cific coast of North America. The only other re-
ported octocoral from this region is the parisidid
Parisis batequensis Squires and Demetrion, 1992,
found in the lower Eocene part of the Bateque For-
mation of Baja California Sur, Mexico.

Subclass Zoantharia Blainville, 1830
Order Scleractinia Bourne, 1900
Family Astrocoeniidae Koby, 1890

Genus Astrocoenia Milne-Edwards and
Fiaime, 1848

TYPE SPECIES. Astrea numisma Defrance,
1826, by monotypy; upper Eocene (lower Bartoni-
an Stage), Gap, southeastern France (Maritime
Alps).

Astrocoenia sp.

Figure 8

LOCAL OCCURRENCE. CSUN loc. 1342.

REMARKS. Only a small fragment of a hemi-
spherical colony was found. The fragment, which
is weathered and possibly worn, is 40 mm long, 20
mm wide, and approximately 5 mm thick. The cor-
allites are mostly filled with hard matrix, but some
of them have been etched by weathering. The cor-
allites are polygonal in shape, and the inside di-
ameter of the calices are up to 1.75 mm. The thecal
walls are about 0.25 mm thick, and their upper
surfaces are irregular with short protuberances. The
calices have two cycles of septa, octamerally ar-
ranged in two subequal groups. The first cycle con-
sists of very thin septa that extend to the styliform
columella, which appears in the bottom of the cal-
ice as a tubercle. The upper margins of these septa
are beaded. The septa in the second cycle are very
short and rudimentary, consisting of trabecular
spines projecting inward from the thecal walls.

The presence of Astrocoenia sp. in the upper 100

m of the Santa Susana Formation represents the
earliest record of this genus on the Pacific coast of
North America and its first record in the “Meganos
Stage.” Only two other species of Astrocoenia have
been reported from this region. One is Astrocoenia
sp., aff. A. portoricoensis Vaughan, 1919, from
middle lower Eocene (“Capay Stage”) strata in the
Turritella uvasana infera fauna of the Juncal For-
mation in the Whitaker Peak area, Ventura County,
California. Squires (1987:19, fig. 6) described and
illustrated a specimen from the Juncal Formation.
Astrocoenia sp. from the Santa Susana Formation
differs from the Juncal Formation species by having
larger corallites, by the presence of secondary septa,
and by having eight primary septa. The Juncal For-
mation species usually has eight primary septa, but
rare specimens can have 10 primary septa.

The other species of Astrocoenia reported from
the Pacific coast of North America fossil record is
A. dilloni Durham (1942), from lower Eocene “Ca-
pay Stage” and possibly “Domengine Stage” strata
on the south side of the headwaters of Media Aqua
Creek, Kern County, California (Durham, 1942)
and from lower Eocene “Capay Stage” strata in the
Bateque Formation of Baja California Sur, Mexico
(Squires and Demetrion, 1992). Astrocoenia sp. dif-
fers considerably from A. dilloni Durham (1942:
505, pi. 44, fig. 3; Squires and Demetrion, 1992:
17, fig. 26) by having only eight rather than 10
septa and by having much weaker secondary septa.

Lower Tertiary astrocoenids are better represent-
ed in the Caribbean Sea region than on the Pacific
coast region of North America. Of the various fos-
sil astrocoenids reported by Duncan (1873),
Vaughan (1919), Wells (1934, 1945), Frost and
Langenheim (1974), and Budd et al. (1992) from
the Caribbean region, Astrocoenia sp. is most sim-
ilar to A. jukesbrownei Wells, 1945, and A. incrus-
tans (Duncan, 1873) [= A. guantanamensis
Vaughan, 1919]. These two species are discussed
below.

Astrocoenia jukesbrownei Wells (1945:3-4, pi. 1,
figs. 4, 5; Budd et al., 1992:575, fig. 2.9-2.10), is
known from middle Eocene strata in Barbados and
upper Eocene strata in Panama. Astrocoenia sp. dif-
fers from A. jukesbrownei by having thinner thecal

Figures 6-21. Octocoral, colonial corals, and solitary coral from the upper 100 m of the Santa Susana Formation, Simi
Valley. All specimens coated with ammonium chloride, unless otherwise noted. 6, 7. Octocoral.? Mopsea sp., aff. M.
costata Milne-Edwards and Haime, 1850, CSUN loc. 1343. 6. Side view, X3.9, LACMIP hypotype 12661. 7. Side view,
X3.3, LACMIP hypotype 12662. 8. Colonial coral. Astrocoenia sp., dorsal view of corallum, X4.3, LACMIP hypotype
12663, CSUN loc. 1342. 9-19. Solitary coral.? Antillophyllia californica new species, adult specimens unless otherwise
noted, CSUN loc. 1343. 9. Lateral view of juvenile showing basal attachment area, X2.1, LACMIP paratype 12665. 10.
Lateral view, X0.9, LACMIP paratype 12666. 11. Lateral view, Xl.l, LACMIP paratype 12667. 12. Lateral view, Xl.4,
LACMIP paratype 12668. 13. Lateral view, Xl.2, LACMIP paratype 12669. 14. Lateral view, Xl.5, LACMIP paratype
12670. 15, 16. LACMIP holotype 12664, Xl.3. 15. Dorsal view. 16. Lateral view. 17. Uncoated, polished section,
transverse view through lower fossa area, X3.9, LACMIP paratype 12671. 18. Uncoated, polished section, longitudinal
view through columella, X2, LACMIP paratype 12672. 19. Photomicrograph, longitudinal view through columella, X2,
LACMIP paratype 12673. 20, 21. Colonial coral. Archohelia clarki Vaughan, 1927, CSUN loc. 1348. 20. Lateral view,
X5.2, LACMIP hypotype 12674. 21. Dorsal view, X4, LACMIP hypotype 12675.

Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils 111

walls and shallower calices. Astrocoenia incrustans
(Duncan, 1873:554, pL 20, fig. 6; Budd et al.,
1992:575, fig. 2. 6-2. 8) is known with certainty
from middle Eocene strata in Cuba, St. Bartholo-
mew, and Chiapas, Mexico; upper Eocene strata in
Cuba and Antigua; and middle Miocene strata in
Panama (Budd et al., 1992). Astrocoenia sp. differs
from A. incrustans by having much weaker second-
ary septa and no secondary septa that extend to the
columella. The relation of Astrocoenia sp. to these
two species, as well as to other early Tertiary astro-
coenids of similar size, growth habit, septal num-
ber, and calicular structure, needs to be clarified.
More specimens and better preserved material of
Astrocoenia sp. are needed to fully describe the ex-
ternal and internal features. Only then will it be
possible to determine whether this astrocoenid rep-
resents a new species.

Family Faviidae Gregory, 1900
Genus Antillopbyllia Vaughan, 1932

TYPE SPECIES. Antiilia lonsdaleia Duncan,
1864, by original designation; Miocene, Dominican
Republic.

? Antillophyllia californica new species

Figures 9-19

DIAGNOSIS. Antillopbyllia-\ike with a shallow
fossa, a columella usually trabecular below and
sublamellar or nearly indiscernible above, a syn-
apticulate corallum wall, and a corallum with
prominent girdling bands.

COMPARISON. The new species is most similar
to ? Antillophyllia olssoni Clark and Durham
(1946) from upper Eocene strata in Colombia. The
similarity between the new species and ?A. olssoni
Clark and Durham (1946:80, pi. 25, figs. 8, 9) is
strong in terms of the large size of the corallum,
the prominent costae corresponding to septa, the
presence of a ring of synapticulae inside the wall,
and thickened inner ends of the septa near the col-
umella. The new species differs from ?A. olssoni by
having a slightly smaller size, a narrower base to
the corallum, a narrower columellar region, and the
presence of swollen, girdling bands on the coral-
lum.

The new species is most similar to Antillophyllia
sawkinsi (Vaughan in Vaughan and Hoffmeister,
1926) from the uppermost Oligocene and lower
Miocene La Quinta Formation in Chiapas, Mexico,
and lower Miocene strata (apparently the Brasso
Formation) in Trinidad (Frost and Langenheim,
1974). The swollen, girdling bands on the corallum
of the new species are similar to those reported by
Frost and Langenheim (1974) as epithecal bands on
specimens of A. sawkinsi. The new species differs
from A. sawkinsi (Vaughan in Vaughan and Hoff-
meister, 1926:118, pi. 2, figs. 6, 6a; Wells, 1934:pl.
28, figs. 6, 6a; Vaughan and Wells, 1943:fig. 305,
3a-3b; Frost and Langenheim, 1974:282, 285, pi.

106, figs. 3-8, pi. 108, figs. 1-8) by having a nar-
rower base, an elevated fossa (at least on some
specimens), and a columella that is trabecular be-
low and sublamellar above. Frost and Langenheim
(1974) reported that A. sawkinsi is closely related
to ? Antillophyllia olssoni Clark and Durham. Frost
and Langenheim (1974) did not questionably as-
sign Clark and Durham’s species to Antillophyllia,
although Clark and Durham (1946) originally did
question the generic assignment.

? Antillophyllia californica new species superfi-
cially resembles }Trochocyatbus striatus (Gabb,
1864) reported (Squires, 1984) from the middle Eo-
cene (“Domengine Stage”) part of the Llajas For-
mation on the north side of Simi Valley. The new
species differs from ?T. striatus (Gabb, 1864:207-
208, pi. 26, fig. 195; Squires, 1984:14, fig. 5c) by
having a larger size, a more compressed corallum,
much shallower fossa, many more septa, narrower
costae, a more horizontally elongate columella, and
girdling bands on the corallum. In addition, the
new species has no confirmed presence of pali,
whereas ?T. striatus has pali.

DESCRIPTION. Solitary, trochoid (basal angle
approximately 40 degrees) to subturbinate, up to
50 cm in height and 2.5 cm in diameter. A few
specimens moderately curved (ceratoid) and enlarg-
ing rapidly. Rare specimens bilobate. Corallum at-
tached in early juvenile stage, with small holdfast
conforming in shape to substrate; Pfree in mature
stage. Corallum external surface of most specimens
shows swollen and irregularly spaced, Pepithecal
girdling bands (1.5 to 3 mm wide). Corallum cos-
tate, septa lowly exsert. Costae tend to alternate in
size and correspond to all septal cycles. Calice el-
liptical to subcircular, usually elliptical. Fossa very
small to small, very slightly concave to shallowly
concave. Some specimens with an elevated fossa;
rare specimens with only a protruding central calice
area and no fossa. Columella very shallow to mod-
erately shallow; barely detectable in some speci-
mens (septa nearly fill the columellar area on these
specimens). Columella trabecular below and sub-
lamellar above; sublamellar columella usually most
obvious on juvenile specimens (less than about 18
mm in height). Wall septothecal? and parathecate
(endothecal dissepiments moderately common).
Septal margins moderately dentate; synapticulae in
corallum wall. Specimens about 22 mm high show
five cycles of septa: First cycle (six septa) and sec-
ond cycle (six septa) reach the columella; third cycle
(12 septa) reaches or nearly reaches the columella;
fourth cycle (24 septa) 75 to 80% as long as the
prosepta of the first cycle; and fifth cycle (48 septa)
short and only in the wall. Inner ends of the first
through third cycles of septa adjacent to the colu-
mella are swollen, and swellings (paliform lobes?)
much better developed on juvenile specimens (less
than about 18 mm in height).

HOLOTYPE DIMENSIONS. Height 24.3 mm
(incomplete), long diameter 23.2 mm, short diam-
eter 15.8 mm.

12 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

PRIMARY TYPE MATERIAL. LACMIP holo-
type 12664 (illustrated), LACMIP paratypes 12665
to 12673 (all illustrated); all types from CSUN loc.
1343.

TYPE LOCALITY. CSUN loc. 1343.

MOLLUSCAN STAGE RANGE. “Meganos.”

GEOGRAPHIC DISTRIBUTION. South side of
Simi Valley, California.

LOCAL OCCURRENCE. CSUN loc. 1343.

REMARKS. Specimens are abundant (155 col-
lected) and show a growth series that grades from
juvenile (8 mm in height) to very mature adult (5
cm in height). The base of nearly every specimen is
missing, but one juvenile (Fig. 9) shows a small
holdfast. Although weathering has affected all spec-
imens, external preservation is generally good. On
many specimens, however, the fossa area and the
corresponding septa have been crushed, infilled
with difficult to remove, well-cemented mudstone,
or both. In addition, no specimen is preserved well
enough in the fossa area to determine whether the
swollen inner ends of the major septa truly corre-
spond to paliform lobes. One of the main diagnos-
tic features of genus Antillophyllia is having pali-
form lobes, and until better preserved specimens
are found, the new species can only be questionably
assigned to this genus. Recrystallization and crush-
ing have obscured much of the internal features of
the specimens of the new species. There are “ghost
structures” that resemble septothecate structures,
and many of the dissepiments have been obliterat-
ed.

Antillophyllia was reported previously only from
upper Eocene strata in Colombia, South America,
Oligocene and Miocene strata in Chiapas, Mexico,
and lower Miocene strata in Trinidad and Florida
(Clark and Durham, 1946; Weisbord, 1971; Frost
and Langenheim, 1974). If the new species does be-
long to this genus, it would be the earliest and west-
ernmost occurrence of Antillophyllia.

ETYMOLOGY. The new species is named for
the state of California.

Family Oculinidae Gray, 1847a
Genus Archohelia Vaughan, 1919

TYPE SPECIES. Archohelia limonensis Vaughan,
1919, by original designation; Pliocene, Costa Rica.

Archohelia clarki Vaughan, 1927

Figures 20, 21

Archohelia clarki Vaughan, 1927:143, pi. 23, figs.

1-5.

PRIMARY TYPE MATERIAL. UCMP holotype
31414 and UCMP paratype 31415; exact location
unknown, “Meganos Formation,” north of Mt.
Diablo, Contra Costa County, California.

MOLLUSCAN STAGE RANGE. “Meganos.”

GEOGRAPHIC DISTRIBUTION. South side of
Simi Valley, California, and north of Mt. Diablo,
Contra Costa County, California.

LOCAL OCCURRENCE. CSUN loc. 1348.
REMARKS. Two specimens were found. After
removal from the enclosing rock, their state of pres-
ervation is somewhat poor. Nevertheless, parts of
the specimens show the branching, colonial form
and the septa. Archohelia clarki is the only known
species of this oculinid coral genus from the fossil
record of the North American Pacific coast.

Prior to this present study, Archohelia clarki was
only known from its type locality north of Mt. Dia-
blo in Contra Costa County, California.

Phylum Mollusca Linnaeus, 1758

Class Gastropoda Cuvier, 1797

Superorder Neritopsina Cox and Knight,
1960

Family Neritidae Rafinesque, 1815

Genus V elates Montfort, 1810

TYPE SPECIES. Nerita perversa Gmelin, 1791,
by original designation and monotypy; Eocene,
Paris Basin, France.

V elates perversus (Gmelin, 1791)

Figures 22, 23

Nerita perversa Gmelin, 1791:3686.

Velates perversus (Gmelin). Cox, 1931:36-37;
Vokes, 1935:382-383, pi. 25, figs. 1-5; pi. 26,
figs. 1-2; Clark and Vokes, 1936:875, pi. 1, figs.
7-8; Givens, 1974:61, pi. 5, figs. 5-6, 13;
Squires, 1984:16-17, figs. 6b-c; 1987:23-24,
figs. 15-19; 1991b:pl. 1, figs. 10, 11; Woods and
Saul, 1986:643-647, figs. 4.17, 5.20, 5.22-5.25,
6. 1-6.3, 6.8; Squires and Demetrion, 1992:26,
figs. 55, 56.

PRIMARY TYPE MATERIAL. Lamarck’s Cab-
inet in the Natural History Museum of Geneva,
Switzerland.

MOLLUSCAN STAGE RANGE. “Meganos”
and “Capay,” possibly “Domengine.”
GEOGRAPHIC DISTRIBUTION. Pakistan, In-
dia, Myanmar, Tibet, Middle East, northern Africa,
western Europe, Florida, PPanama, Baja California
Sur (Mexico), and southern California (possibly
south-central California).

LOCAL OCCURRENCE. CSUN loc. 1342,
LACMIP loc. 7124.

REMARKS. Two specimens were found. One,
from CSUN locality 1342 is a large internal mold,
but the diagnostic teeth on the inner lip are visible.
The second specimen, from LACMIP locality 7124,
is small, but it has the shell and shows all the di-
agnostic morphology.

See Squires (1987) for a more complete synony-
my of this cosmopolitan species. The synonymy
given here is primarily for the Pacific coast of North
America. This is the first report of Velates perversus
from the “Meganos Stage.” In the Simi Valley area,
it is also known from low in the Llajas Formation

Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils ■ 13

14 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

in “Capay Stage” rocks (Squires, 1984; Woods and
Saul, 1986).

The only other species of Velates in the Simi Val-
ley area is V. calif ornicus Vokes (1935), which has
been found at UCMP localities 7009 and 3792
(Vokes, 1935), as well as at LACMIP locality
23173 (Saul, 1983; Woods and Saul, 1986). The
stratigraphic position of UCMP locality 7009 is
known (Fig. 1), and is approximately on strike with
that of LACMIP locality 7124 (Fig. 4). The strati-
graphic positions of UCMP locality 3792 and LAC-
MIP 23173, however, are not known. These two
localities are probably in the upper part of the San-
ta Susana Formation on the south side of Simi Val-
ley, and the specimens are probably late Paleocene
in age (Woods and Saul, 1986).

Although Velates californicus and V. perversus
might have the same geologic range (late Paleocene
to early Eocene), they have never been collected at
the same locality. As Woods and Saul (1986) re-
ported, V. californicus always occurs stratigraphi-
cally below V. perversus, and the present study con-
firms this observation.

Genus Corsania Vidal, 1917

TYPE SPECIES. Corsania doubillei Vidal, 1917,
by original designation; late Early Cretaceous (Ap-
tian), Cors, Lerida, Spain.

Subgenus januncia Woods and Saul, 1986

TYPE SPECIES. Corsania { Januncia ) janus
Woods and Saul, 1986, by original designation; late
Paleocene?, Baja California Sur, Mexico.

Corsania ( Januncia ) susana Saul and Squires,
1997

Figures 24, 25

Corsania (Januncia ) susana Saul and Squires, 1997:

142-143, figs. 28-30.

PRIMARY TYPE MATERIAL. LACMIP holo-
type 7890, CSUN loc. 969; LACMIP paratype
7891, CSUN loc. 973; LACMIP paratype 6441,
CSUN loc. 966. All localities in the upper 100 m

of the Santa Susana Formation, north side of Simi
Valley, California.

MOLLUSCAN STAGE RANGE. “Meganos.”

GEOGRAPHIC DISTRIBUTION. North side of
Simi Valley, California.

LOCAL OCCURRENCE. CSUN Iocs. 966, 969,
973.

REMARKS. Three specimens were found. Only
the holotype shows the inner lip, which is promi-
nently set off from the deck area.

Superorder Caenogastropoda Cox, 1959
Family Turritellidae Woodward, 1851

TYPE SPECIES. Turbo terebra Linnaeus, 1758,
by monotypy; Recent, southwest Pacific.

Turritella meganosensis Clark and
Woodford, 1927
Figure 26

Turritella meganosensis Clark and Woodford,

1927:119-120, pi. 21, figs. 2-5; Merriam, 1941:

75, pi. 8, figs. 3, 4, 7, 9; Saul, 1983:pl. 1, fig. 13;

Clark, 1929:pl. 5, figs. 1, 11; Schenck and Keen,

1940:pl. 22, figs. 5-7.

PRIMARY TYPE MATERIAL. UCMP holotype
12445 and UCMP paratype 12441, UCMP loc.
3159, Margaret Hamilton Sand [= Clark and
Woodford’s (1927) Division D of the Meganos For-
mation], Deer Valley, Contra Costa County, Cali-
fornia); paratypes UMCP 31225-31226, UCMP
loc. 7000, upper Santa Susana Formation, north
side of Simi Valley, California.

MOLLUSCAN STAGE RANGE. “Meganos.”

GEOGRAPHIC DISTRIBUTION. North side of
Simi Valley, California; Deer Valley, Contra Costa
County, California; south of Covelo (Round Valley)
on Middle Fork of Eel River, Mendocino County,
California.

LOCAL OCCURRENCE. Upper 100 m of the
Santa Susana Formation (exact location not
known).

REMARKS. In the upper 100 m of the Santa Su-
sana Formation, this rare and large-sized turritellid

Figures 22-37. Gastropods from the upper 100 m of the Santa Susana Formation, Simi Valley. All specimens coated with
ammonium chloride. 22, 23. Velates perversus (Gmelin, 1791), Xl.8, LACMIP hypotype 12676, LACMIP loc. 7124.
22. Apertural view. 23. Abapertural view. 24, 25. Corsania Januncia) susana Saul and Squires, 1997, Xl.4, LACMIP
holotype 7890, CSUN loc. 969. 24. Apertural view. 25. Abapertural view. 26. Turritella meganosensis Clark and Wood-
ford, 1927, apertural view, X0.7, UCMP holotype 37430, UCMP loc. 7000. 27. Turritella andersoni susanae Merriam,
1941, apertural view, X3.2, LACMIP hypoptye 12677, CSUN loc. 969. 28, 29. Turritella buwaldana Dickerson, 1916.
28. Abapertural view, X3.7, LACMIP hypotype 10434 [= UCLA hypotype 59372], LACMIP loc. 26615. 29. Abapertural
view, X4.5, LACMIP hypotype 12678, CSUN loc. 1343. 30, 31. Turritella uvasana infera Merriam, 1941. 30. Abaper-
tural view, X3, LACMIP hypotype 12679, CSUN loc. 967. 31. Apertural view, X2, LACMIP hypotype 12680, CSUN
loc. 959. 32. Turritella susanensis Nelson, 1925 nomen dubium, abapertural view, X3.6, LACMIP hypotype 12681,
CSUN loc. 1343. 33. Campanile dilloni (Hanna and Hertlein, 1949), apertural view, X0.6, LACMIP hypotype 12335,
CSUN loc. 1565. 34. Calyptraea diegoana (Conrad, 1855), lateral view, X2.2, LACMIP hypotype 12682, CSUN loc.
958. 35, 36. Pachycrommium clarki (Stewart, 1927), X0.7, LACMIP hypotype 12683. 35. Apertural view. 36. Abaper-
tural view. 37.? Pbalium ( Semicassis ) tuberculiformis (Hanna, 1924), apertural view, Xl.7, LACMIP hypotype 12684,
CSUN loc. 967.

Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils ■ 15

(up to 90 mm in height) has been found only at
UCMP locality 7000, which is 7.5 m below the Lla-
jas Formation according to Saul (1983). The exact
location of this locality, however, is not known. No
additional specimens were found during this pres-
ent study. This species is the only Turritella in the
upper 100 m of the Santa Susana Formation that
is restricted to the “Meganos Stage.” Mid-adult
stage specimens have a pronounced posterior swell-
ing, but this feature becomes obsolete in late-adult
stage specimens (Clark and Woodford, 1927; Mer-
riam, 1941).

Turritella andersoni susanae Merriam, 1941

Figure 27

Turritella andersoni susanae Merriam, 1941:79, pi.

11, fig. 6; Saul, 1983:pl. 2, fig. 5.

Turritella andersoni n. subsp. Saul, 1983:pl. 1, figs.

15-18.

PRIMARY TYPE MATERIAL. UCMP holotype
15295, UCMP loc. A-993, basal Llajas Formation,
north side of Simi Valley, California.

MOLLUSCAN STAGE RANGE. “Meganos.”

GEOGRAPHIC DISTRIBUTION. Simi Valley,
California.

LOCAL OCCURRENCE. CSUN Iocs. 962, 963,
964, 965, 966, 969, 970, 971, 972, 973, 1347,
1349, LACMIP loc. 21551.

REMARKS. Specimens are abundant at most lo-
calities but generally have poor preservation be-
cause their weathered shells are chalky and disin-
tegrate when removed from the outcrop. The larg-
est specimens are up to 45 mm in height and were
found at locality 966.

This subspecies is characterized by adult whorls
with a concave whorl profile bearing two relatively
heavy spiral ribs anteriorly (the anteriormost is the
most prominent) and two primary sprial ribs pos-
teriorly. All of these ribs can be noded. The medi-
ally concave area between these two sets of primary
ribs is usually smooth, or, in some cases, with a
secondary and numerous tertiary spiral ribs. A su-
tural spiral rib is also present.

Squires (1984) regarded this subspecies as con-
specific with Turritella andersoni s.s. Dickerson,
1916. As stated by Merriam (1941), and later con-
firmed by Squires (1987), the two are separate taxa.
Turritella andersoni susanae differs from T. ander-
soni s.s. in the relatively stronger development of
the two posterior primary spiral ribs. In the syn-
onymy of T. andersoni s.s. given by Squires and
Goedert (1994), T. a. susanae was inadvertently in-
cluded as a synonym of T. andersoni.

Although Turritella andersoni susanae is locally
abundant and is one of the most diagnostic species
of the “Meganos Stage” in the Simi Valley area, the
holotype of this subspecies is reported to be from
the basal part of the overlying Llajas Formation, at
UCMP loc. A-993 on the north side of Simi Valley.
The holotype is the only specimen of T. a. susanae
ever reported from the Llajas Formation, and

Squires (1984), in his monographic work on the
Llajas Formation, never encountered any specimens
of this subspecies. The description of the type lo-
cality is imprecise and could apply equally to the
upper 100 m of the Santa Susana Formation or to
the lower part of the Llajas Formation. It is likely
that the type locality of T. a. susanae is in error and
should have been reported as from the upper 100
m of the Santa Susana Formation.

Turritella buwaldana Dickerson, 1916

Figures 28, 29

Turritella buwaldana Dickerson, 1916:500-501,
pi. 42, figs. 7a-b; Hanna, 1927:307:pl. 19, figs.
7-8, 12; Vokes, 1939:161; Kappeler et ah, 1984:
table 2; Merriam, 1941:86-87, pi. 21, figs. 3-9;
pi. 33, figs. 1-14; Stewart, 1946:pl. 11, fig. 24;
Givens, 1974:63, pi. 5, fig. 15; Saul, 1983:pl. 2,
figs. 13-15; Squires, 1983a:fig. 9f; 1984:18, fig.
6h; 1987:27, fig. 24; 1988b:10-ll, fig. 15;
1991b:pl. 1, fig. 14; Squires and Demetrion,
1992:27, fig. 61.

? Turritella uvasana Conrad. Dickerson, 1915:pl. 5,
figs, lc, 3, 4.

Turritella kewi Dickerson, 1916:501, pi. 42, fig. 8.
Turritella subuvasana Nelson, 1925:423, pi. 56,
figs. 5, 6, 7; Merriam, 1941:74, pi. 41, figs. 1-3.
Turritella buwaldana crooki Merriam and Turner,
1937:105, pi. 5, fig. 6; Merriam, 1941:87, pi. 21,
figs. 1, 2; Turner, 1938:85; Vokes, 1939:161-
162; Saul, 1983:pl. 2, figs. 2, 3.

Turritella ? buwaldana subuvasana Nelson. Saul,
1983:pl. 1, fig. 11.

PRIMARY TYPE MATERIAL. UCMP holotype
12130, UCMP loc. 672, Domengine Formation,
Fresno County, California.

MOLLUSCAN STAGE RANGE. “Meganos”
through “Tejon.”

GEOGRAPHIC DISTRIBUTION. San Ignacio
Lagoon area, Baja California Sur, Mexico, to Glide,
Oregon.

LOCAL OCCURRENCE. CSUN Iocs. 958, 959,
961, 962, 967, 970, 972, 1342, 1343, 1345, 1346,
LACMIP loc. 26615.

REMARKS. Specimens are rare to common and
are most abundant at locality 961. All specimens
are fragmentary and show moderately good pres-
ervation.

The range of variability of Turritella buwaldana
is “confusingly great,” as noted by Merriam (1941:
86), and he found variation among individuals
from the type locality, as well as variation within
and among other known regional assemblages.
Considerable variation of T. buwaldana is also
present in specimens found in the upper 100 m of
the Santa Susana Formation. Many of the speci-
mens fit the original description of T. buwaldana
and have three primary spiral ribs on the anterior
half of the mature whorls and two (in some cases
only one) secondary spiral ribs posterior to the pri-
mary ribs. Tertiary ribs, usually only one, are in the

16 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

interspaces. There are also rare specimens that have
three primary ribs and three secondary ribs. These
latter specimens are indistinguishable from T. bu-
waldana crooki Merriam and Turner. In the upper
100 m of the Santa Susana Formation, there are
also specimens of T. buwaldana whose posterior-
most primary rib is only as strong as a secondary
rib, thereby producing whorls with two primary
ribs and three secondaries.

Some specimens of Turritella buwaldana from
the upper Santa Susana Formation have minute
nodes on the primaries (usually only the posterior-
most one) and on the secondaries. Squires (1987,
1988b) and Squires and Demetrion (1992) also re-
ported the presence of minute nodes on specimens
of this species from elsewhere on the Pacific coast
of North America.

Nelson (1925: checklist opposite p. 402) report-
ed Turritella subuvasana Nelson, 1925, from
UCMP loc. 3791 [= CSUN loc. 1343] in the Santa
Susana Formation. Specimens of Nelson’s species
consist of only the tips (apices) of the shells, even
though Merriam (1941) reported that specimens
show adult- whorl sculpture. During the present
study, about 20 specimens were collected at CSUN
locality 1343, and they also consist of only the tips
of shells. The largest known specimens of T. su-
buvasana from this locality are only 12 mm in
height, and the sculpture on the largest whorls usu-
ally consists of five spiral ribs. The two posterior-
most ribs can be slightly weaker than the other
three, or all can be subequal. The sculptural pat-
terns of the tips of Paleogene turritellas from the
Pacific coast of North America are not species spe-
cific, but in the case of T. subuvasana, the tips are
sufficiently close to the variability of T. buwaldana
to allow identification as T. buwaldana. One of
these specimens is illustrated in Fig. 29.

Nelson (1925) and Merriam (1941) also reported
Turritella subuvasana from UCMP loc. 3796 in the
Santa Susana Formation. The exact stratigraphic
position of this locality, which is the type locality
of this gastropod, is not known because of struc-
tural complications caused by the nearness of the
Runkle Canyon fault zone.

Turritella uvasana inf era Merriam, 1941
Figures 30, 31

Turritella uvasana inf era Merriam, 1941:90, pi. 40,

figs. 2-4; Givens, 1974:65=66, pi. 6, figs. 5-7;

Saul, 1983:pl. 1, fig 19; pi. 2, fig. 4; Squires,

1984:19, fig. 6i; 1987:27-28, fig. 25.

PRIMARY TYPE MATERIAL. UCMP holotype
33993, UCMP loc. A-994, lower part of the Llajas
Formation, north side of Simi Valley, California.

MOLLUSCAN STAGE RANGE. “Meganos”
and “Capay.”

GEOGRAPHIC DISTRIBUTION. Simi Valley,
Whitaker Peak area, and Pine Mountain area, Ven-
tura County, California.

Contributions in Science, Number 479

LOCAL OCCURRENCE. CSUN Iocs. 959, 960,
961, 967, 1346, LACMIP loc. 21551.

REMARKS. Specimens are rare to abundant,
and, at all localities, they are moderately well pre-
served. They are most abundant at CSUN locality
959. This species is characterized by five equal and
strong primary spiral ribs on rounded whorls.

This species has been reported (Merriam, 1941;
Givens, 1974; Saul, 1983; Squires, 1984, 1987) as
present in the uppermost part of the Santa Susana
Formation in the Simi Valley area. Merriam (1941)
and Saul (1983) mentioned that these specimens
have a more rounded whorl profile and heavier rib-
bing than those from the type locality low in the
overlying Llajas Formation. Squires (1987), how-
ever, reported that the upper Santa Susana Forma-
tion specimens are indistinguishable from speci-
mens elsewhere.

Turritella susanaensis Nelson, 1925 nomen
dubium
Figure 32

Turritella susanaensis Nelson, 1925:423, pi. 56,
figs. 1, 2; Merriam, 1941:73-74, pi. 41, figs. 4,
10.

REMARKS. Nelson (1925) and Merriam (1941)
reported this taxon from the Santa Susana Forma-
tion at UCMP loc. 3791 [= CSUN loc. 1343] and
UCMP loc. 3796. Specimens are abundant and
moderately well preserved at locality 3791. As dis-
cussed under Turritella buwaldana, the exact strati-
graphic position of UCMP loc. 3796 within the
Santa Susana Formation is unclear. At both locali-
ties, only the broken-off tips of shells have been
found, even though Merriam (1941) reported that
specimens show adult- whorl sculpture. The largest
known specimens are only 13 mm in height.

So far, even after more than 60 years of collect-
ing, only tips of T. susanaensis are known. Their
sculptural pattern is not distinctive. The sculptural
pattern of the tips of Paleocene turritellas from the
Pacific coast of North America are, in fact, not spe-
cies specific. Nelson (1925) should not have based
his species on such material; therefore, T. susanaen-
sis becomes a nomen dubium.

Family Campanilidae Douville, 1904

Genus Campanile Fischer, 1884

TYPE SPECIES. Ceritbium giganteum Lamarck,
1804a, by subsequent designation (Sacco, 1895);
Eocene, Paris Basin, France.

Campanile dilloni (Hanna and Hertlein,
1949)

Figure 33

Campanilopa dilloni Hanna and Hertlein, 1949:
393, pi. 77, figs. 2, 4, text-fig. 1; Givens, 1974:
69, pi. 7, fig. 10; Squires and Advocate, 1986:
853, 855, fig. 2.1.

Squires: “Meganos Stage” Marine Megafossils ■ 17

Campanile dilloni Hanna and Hertlein. Squires,
1991b:pl. 1, fig. 18; 1993:327-329, figs. 6-11.

PRIMARY TYPE MATERIAL. CAS holotype
9425 and CAS paratypes 9428 and 9429, all from
CAS loc. 30667, Mabury Formation, Agua Media
Creek, Temblor Range, Kern County, California.

MOLLUSCAN STAGE RANGE. “Meganos”
and “Capay.”

GEOGRAPHIC DISTRIBUTION. Orocopia
Mountains, Riverside County, California, to Agua
Media Creek, Temblor Range, Kern County, Cali-
fornia.

LOCAL OCCURRENCE. CSUN loc. 1565.
REMARKS. A single specimen was found. It is
15.1 cm in height and well preserved.

Family Calyptraeidae Lamarck, 1809

Genus Calyptraea Lamarck, 1799

Type Species. Patella chinensis Linnaeus, 1758,
by monotypy; Recent, Europe.

Calyptraea diegoana (Conrad, 1855)
Figure 34

Trochita diegoana Conrad, 1855:7, 17; 1857:327,
pi. 5, fig. 42.

Galerus excentricus Gabb, 1864:136, pi. 20, fig.
95; pi. 29, fig. 232a; Arnold, 1907a:pl. 10, fig.
3a.

Calyptraea calabasaensis Nelson, 1925:419, pi. 54,
figs. 8a-b.

Calyptraea ( Galerus ) calabasaensis Nelson. Clark
and Woodford, 1927:120, pi. 21, figs. 10-13.
Calyptraea diegoana (Conrad). Stewart, 1927:340-
341, pi. 27, fig. 15; Turner, 1938:89-90, pi. 20,
figs. 1-2; Effinger, 1938:378; Weaver, 1943:351-
352, pi. 71, figs. 16, 20; pi. 103, fig. 3; 1953:29;
Stewart, 1946:pl. 11, fig. 5; Kleinpell and Weav-
er, 1963:186, pi. 24, fig. 7; Hickman, 1969:79,
82, pi. 11, figs. 7-8; 1980:33-34, pi. 2, figs. 18-
21; Demere et al., 1979:pl. 2, fig. 7; Squires,
1984:21, fig. 6q; 1987:32, fig. 4; 1988b:ll, fig.
19; 1 99 1 b:pl. 1, fig. 20; 1994:pl. 1, fig. 2; Squires
and Goedert, 1994:16, 18, fig. 36.

PRIMARY TYPE MATERIAL. USNM holotype
1856, Eocene strata (probably the Delmar Forma-
tion), San Diego, California.

MOLLUSCAN STAGE RANGE. “Martinez”
through lower Oligocene.

GEOGRAPHIC DISTRIBUTION. San Diego,
California, to Little River area, Grays Harbor
County, Washington.

LOCAL OCCURRENCE. CSUN Iocs. 958, 963,
1342, 1349.

REMARKS. Specimens are rare but moderately
well preserved.

Family Naticidae Forbes, 1838

Genus Pacbycrommium Woodring, 1928

Type Species. Amaura guppyi Gabb, 1873, by
original designation; Recent, Miocene, Dominican
Republic.

Pacbycrommium clarki (Stewart, 1927)
Figures 35, 36

Amauropsis alveata (Conrad). Arnold, 1910:pl. 4,
fig. 21; Dickerson, 1915:pl. 5, fig. 9; Waring,
1917:pl. 15, fig. 25 [Misidentifications]. Not
Amauropsis alveata (Conrad, 1855).

Amaurellina ( Euspirocrommium ) clarki Stewart,
1927:336-339, pi. 26, figs. 8, 9 [new name, in
part, for Amauropsis alveata (Conrad, 1855),
preoccupied and misidentified]; Clark, 1929:pl.

11, fig. 10; Turner, 1938:86, pi. 20, fig. 3; Weav-
er, 1943:345, pi. 70, figs. 10, 18; Kleinpell and
Weaver, 1963:188, pi. 27, fig. 15.

Amaurellina clarki Stewart. Gardner and Bowles,
1934:246, figs. 6, 8.

}Amaurellina multiangulata Vokes, 1939:174, pi.
22, figs. 2, 8, 13.

? Pacbycrommium clarki (Stewart). Vokes, 1939:
175, pi. 22, figs. 11, 30; Givens, 1974:73, pi. 8,
figs. 6, 10.

Amaurellina} ( Euspirocrommium ) clarki Stewart.

Stewart, 1946:pl. 11, fig. 3.

Pacbycrommium clarki (Stewart). Marincovich,
1977:238-241, pi. 20, figs. 4-10; Squires, 1983a:
fig. 9b; 1984:25, fig. 7e; 1987:36, fig. 44; 1988b:

12, fig. 25; 1991b:pl. 1, fig. 25; Squires et al.,
1992:pl. 1, fig. 18.

PRIMARY TYPE MATERIAL. UCMP holotype
31385 and UCMP paratype 31386 of Amaurellina
( Euspirocrommium ) clarki Stewart, both from
UCMP loc. 7004, Llajas Formation, north side of
Simi Valley, California.

MOLLUSCAN STAGE RANGE. “Meganos”
through “Tejon.”

GEOGRAPHIC DISTRIBUTION. Laguna San
Ignacio area, Baja California Sur, Mexico, to south-
western Washington.

LOCAL OCCURRENCE. CSUN Iocs. 958, 967,
1346, 1347.

REMARKS. Specimens are rare to uncommon
and most numerous at CSUN locality 967, where
they also show the best preservation. This is the
first report of this species from the “Meganos
Stage.”

Family Cassidae Swainson, 1832

Genus Pbalium Link, 1807

TYPE SPECIES. Buccinum glaucum Linnaeus,
1758, by subsequent designation (Dali, 1909); Re-
cent, Indo-Pacific.

18 1 Contributions in Science, Number 479

Squires: “Meganos Stage’’ Marine Megafossils

Subgenus Semicassis Morch, 1852

TYPE SPECIES. Cassis japonica Reeve, 1848, by
subsequent designation (Harris, 1897); Recent,
China and Japan.

}Phalium ( Semicassis ) tuberculiformis
(Hanna, 1924)

Figure 37

LOCAL OCCURRENCE. CSUN loc. 967.
REMARKS. A single specimen was found. It is
mostly an internal mold without the outer lip and
anterior end of the shell. Some shell is present on
the ventral surface, but it is weathered. The speci-
men has three carinae on the body whorl, and the
shell material shows fine spiral ribbing between the
carinae. Morphologically, the specimen resembles
comparably preserved specimens of Phalium (Sem-
icassis) tuberculiformis, from the “Domengine
Stage” “Stewart bed” in the Llajas Formation on
the north side of Simi Valley. This similarity is sig-
nificant enough to warrant tentative identification.

Order Neogastropoda Thiele, 1929

Family Buccinidae Rafinesque, 1815

Genus Brachysphingus Gabb, 1869

TYPE SPECIES. Brachysphingus sinuatus Gabb,
1869, by subsequent designation (Cossmann,
1901); Paleocene, California and Baja California,
Mexico.

Brachysphingus mammilatus Clark and
Woodford, 1927
Figure 38

Brachysphingus mammilatus Clark and Woodford,
1927:116-117, pi. 20, figs. 8-15; Clark, 1929:
13, pi. 4, figs. 3, 10; Schenck and Keen, 1940:
22, figs. 1, 2; Givens, 1974:84, pi. 10, fig. 3;
Squires, 1997:856, 858, figs. 5, 1-14.

Pseudoliva sp. Smith, 1975:pl. 1, figs. 14, 15.

PRIMARY TYPE MATERIAL. UCMP holotype

31234 from UCMP loc. 3157; UCMP paratype

31235 from UCMP loc. 3577; UCMP paratype

31236 from UCMP loc. 3159; UCMP paratype

31237 from UCMP loc. 3159, UCMP paratype

31238 from UCMP loc. 3577. All from Margaret
Hamilton Sand [= division D of Meganos Forma-
tion as used by Clark and Woodford (1927)], Deer
Valley, Contra Costa County, California.

MOLLUSCAN STAGE RANGE. “Meganos”
and “Capay.”

GEOGRAPHIC DISTRIBUTION. Simi Valley,
California, to south of Covelo (Round Valley) on
Middle Fork of Eel River, Mendocino County, Cal-
ifornia.

LOCAL OCCURRENCE. CSUN Iocs. 958,
1345.

REMARKS. Specimens are rare and well pre-
served. Squires (1997) did a detailed study of Bra-

Contributions in Science, Number 479

chysphingus, noting its occurrence in the upper 100
m of the Santa Susana Formation.

Brachysphingus mammilatus has been found as-
sociated with Turritella meganosensis in the area
south of Covelo (Round Valley) on the Middle Fork
of the Eel River, Mendocino County, California
(Merriam and Turner, 1937).

Family Olividae Latreille, 1825
Genus Ancillarina Bellardi, 1882

TYPE SPECIES. Ancilla canalifera Lamarck,
1802, by subsequent designation (Palmer, 1937);
Eocene, Paris Basin, France.

? Ancillarina sp.

Figures 39, 40

LOCAL OCCURRENCE. CSUN loc. 961.

REMARKS. Three specimens were found. The
best preserved of the three is crushed in the middle,
and most of the spire is represented as an internal
mold. The other two specimens are internal molds.
The best preserved specimen is similar in overall
shape and in the columella area to the Paris Basin,
France, Eocene Ancillarina canalifera (Lamarck,
1802). Ancillarina canalifera, which is the type spe-
cies of Ancillarina, has a total lack of callus on the
spires whorls and sutures, and, according to Kil-
burn (1981), this is a major diagnostic feature of
this genus. The specimen from CSUN locality 961
differs slightly from A. canalifera (Lamarck, 1802:
475, pi. 2, fig. 8; Cossmann and Pissarro, 1910-
1913:pl. 67, figs. 211-9, 211-9', 211-9"; Kilburn,
1981:figs. 24-27) by having a less distinct suture
between the penultimate whorl and body whorl.
The somewhat indistinct suture on the specimen
from locality 961 might be the result of poor pres-
ervation or of a slight amount of callus. Until better
preserved specimens are found, it is not possible to
assign this species with certainty to Ancillarina.

If the specimen from locality 961 does prove to
belong to Ancillarina, it would be the first record
of this genus in the Western Hemisphere. Wenz
(1943) reported the temporal range of this genus as
Eocene to Miocene and the geographic distribution
as confined to Europe.

The specimen of ? Ancillarina sp. superficially re-
sembles Ancilla burroensis Nelson (1925) from the
“Martinez marine member” (“Martinez Stage”)
part of the Santa Susana Formation on the south
side of Simi Valley. The specimen of ? Ancillarina
sp. differs from A. burroensis Nelson (1925:433, pi.
60, figs. 2, 3) by having columellar teeth that are
less prominent, longer, and more parallel to the
shell axis, as well as by having a more deeply
notched anterior sinus. In addition, if ? Ancillarina
sp. does have any callus on the spire, it is much
lighter than the heavily calloused spire found on A.
burroensis.

Squires: “Meganos Stage” Marine Megafossils ■ 19

20 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

Family Turridae Swainson, 1840
Genus Gemmula Weinkauff, 1875

Type Species. Gemmula hindsiana Berry, 1958
[= Pleurotoma gemmata Reeve, 1843], by subse-
quent designation (Cossmann, 1896); Recent,
southern Baja California, Mexico, to Colombia,
South America.

Gemmula sp., aff. G. diabloensis Clark and
Woodford, 1927
Figure 41

LOCAL OCCURRENCE. CSUN loc. 961.

REMARKS. A single specimen was found, and it
is missing the uppermost spire and tip of the ante-
rior canal. About half of the shell material is miss-
ing, and the half that present is weathered. The
specimen has close affinity to Gemmula diabloensis
Clark and Woodford (1927:107, pi. 19, figs. 3, 4)
from the “Meganos Stage” Margaret Hamilton
Sand [= Division D of the Meganos Formation as
used by Clark and Woodford, 1927] in Contra Cos-
ta County, California. The specimen from CSUN
locality 961 differs from G. diabloensis by having
12 rather than about 10 axial ribs on the penulti-
mate whorl, three rather than four spiral ribs be-
tween the shoulder angulation and the sutural col-
lar, and spiral ribs on the body whorl grading an-
teriorly from medium to fine rather than being dif-
ferentiated into pairs or sets of three.

Genus Turricula Schumacher, 1817

TYPE SPECIES. Turricula flammea Schumacher,
1817, by monotypy; Recent, Sri Lanka.

Turricula sp., aff. T. burroensis (Nelson,
1925)

Figure 42

LOCAL OCCURRENCE. CSUN loc. 969.

REMARKS. A single specimen was found, and it
is missing the upper part of the spire and about half

of the anterior canal. The specimen is also some-
what weathered, and the nodes on the spire are cor-
respondingly subdued. The specimen has affinity to
Turricula burroensis (Nelson, 1925) from the so-
called “Martinez Marine Member” (“Martinez
Stage”) part of the Santa Susana Formation on the
south side of Simi Valley. The specimen from CSUN
locality 969 differs from T. burroensis (Nelson,
1925:435, pi. 60, figs. 8, 9) by having more spiral
ribs (26 rather than 16) on the penultimate whorl
and more spiral ribs (about 60 rather than 45) on
the body whorl.

Nelson (1925) originally assigned his species to
Tunis Roding, 1798, but Zinsmeister (1983b) re-
assigned the species to Turricula based on the di-
agnostic presence of the anal sinus on the shoulder
slope.

Although Turricula calafia Nelson (1925:434, pi.
60, figs, la, lb), known only from the same locality
as T. burroensis, is similar to T. burroensis, the
specimen of Turricula from CSUN locality 969 is
more similar to T. burroensis in that it is slimmer.

Subclass Heterobranchia Gray, 1840

Order Heterostropha Fischer, 1885

Family Architectonicidae Gray, 1850

Genus Architectonica Roding, 1798

Type Species. Trochus perspectivus Linnaeus,
1758, by subsequent designation (Gray, 1847b);
Recent, Indo-Pacific.

Subgenus Architectonica s.s.
Architectonica ( Architectonica ) llajasensis
Sutherland, 1966
Figures 43, 44

Architectonica llajasensis Sutherland, 1966:1-4,
figs. 1, 2.

Architectonica ( Architectonica ) llajasensis Suther-
land. Squires, 1984:19, fig. 6k; Squires and De-
metrion, 1994:131-132, fig. 16.

Figures 38-56. Gastropods and bivalves from the upper 100 m of the Santa Susana Formation, Simi Valley. All specimens
coated with ammonium chloride. 38-48. Gastropods. 38. Brachysphingus mammilatus Clark and Woodford, 1927,
apertural view, Xl.4, LACMIP hypotype 12685, CSUN loc. 958. 39, 40.? Ancillarina sp., X2.5, LACMIP hypotype
12686, CSUN loc. 961. 39. Apertural view. 40. Abapertural view. 41. Gemmula sp., aff. G. diabloensis Clark and
Woodford, 1927, apertural view, X4.5, LACMIP hypotype 12687, CSUN loc. 969. 42. Turricula sp., aff. T. burroensis
(Nelson, 1925), abapertural view, X2, LACMIP hypotype 12688, CSUN loc. 961. 43, 44. Architectonica (A.) llajasensis
Sutherland, 1966, X2, LACMIP hypotype 12689, LACMIP loc. 26609. 43. Apical view. 44. Umbilical view. 45, 46.
Ringicula ( R .) pinguis (Gabb, 1864), X13.2, LACMIP hypotype 12690, LACMIP loc. 1344. 45. Apertural view. 46.
Right-lateral view. 47, 48. Cylichnina tantilla (Anderson and Lianna, 1925). 47. Apertural view, X5.4, LACMIP hypotype
12691, CSUN loc. 964. 48. Abapertural view, X4.8, LACMIP hypotype 12692, CSUN loc. 970. 49-54. Bivalves. 49.
Acila ( Truncacila ) decisa (Conrad, 1855), latex peel of external mold, right valve, X3.8, LACMIP hypotype 12693,
CSUN loc. 959. 50. Nuculana ( Saccella ) gabbii (Gabb, 1869), latex peel of external mold, left valve, X6, LACMIP
hypotype 12694, CSUN loc. 967. 51. Area (A.) filewiezi Squires, 1991a, left valve, Xl, LACMIP holotype 8365, CSUN
loc. 965. 52, 53. Spondylus carlosensis Anderson, 1905, X2.1, LACMIP hypotype 12695, CSUN loc. 1343. 52. Left
(free) valve. 53. Right (attached) valve. 54-56. Pycnodonte ( Phygraea ) sp., aff. Pycnodonte ( Phygraea ) pacifica Squires
and Demetrion, 1990, CSUN loc. 1343. 54. Left (lower) valve, Xl.4, LACMIP hypotype 12696. 55, 56. Right (upper)
valve, X2.5, LACMIP hypotype 12697. 55. Interior. 56. Exterior.

Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils ■ 21

PRIMARY TYPE MATERIAL. LACMIP holo-
type 1140, LACMIP loc. 461-B, Llajas Formation,
Simi Valley, California.

MOLLUSCAN STAGE RANGE. “Meganos”
and “Domengine” (no specimens known from “Ca-
pay Stage”).

GEOGRAPHIC DISTRIBUTION. Baja Califor-
nia Sur, Mexico, to Simi Valley, California.

LOCAL OCCURRENCE. CSUN Iocs. 966, 958.

REMARKS. A single specimen was found at
CSUN locality 966, and this is the first record of
Architectonica ( Arcbitectonica ) llajasensis from
“Meganos Stage” strata. An internal mold was
found at CSUN locality 958.

Family Ringiculidae Philippi, 1853
Genus Ringicula Deshayes, 1838
Subgenus Ringicula s.s.

Type Species. Auricula ringens Lamarck, 1804b,
by subsequent designation, Gray (1847b); Eocene,
Paris Basin, France.

Ringicula (Ringicula) pinguis (Gabb, 1864)

Figures 45, 46

Cinulia pinguis Gabb, 1864:112, pi. 29, figs. 221a,
221b.

Ringinella pinguis Gabb, 1869:175; Dickerson,

1914:17, figs. 4a, 4b.

Tornatellaea pinguis (Gabb). Nelson, 1925:436, pi.

60, figs. 5, 6; Stewart, 1927:433-434, pi. 25, fig.

10; Schenck and Keen, 1940:pl. 20, fig. 11.
Tornatella pinguis (Gabb). Zinsmeister, 1983a:pl.

4, fig. 31.

PRIMARY TYPE MATERIAL. ANSP lectotype
4265, Martinez Formation, “in the bluffs, a mile
west of Martinez” (Gabb, 1864), Contra Costa
County, California.

MOLLUSCAN STAGE RANGE. “Martinez” to
“Meganos.”

GEOGRAPHIC DISTRIBUTION. Southern
California to Martinez, California.

LOCAL OCCURRENCE. CSUN loc. 1344.

REMARKS. Eight specimens were found, all ear-
ly juveniles of minute size.

Ringicula s.s. is characterized by a small, low-
spire globose to subglobose shape with a sculpture
of spirally incised furrows, a greatly thickened out-
er lip, an internally dentate outer lip, a columella
with two strong folds, and an anterior notch to the
aperture (Sohl, 1964; Davies and Eames, 1971).
The Santa Susana Formation specimens have all of
these characteristics. Genus Tornatellaea Conrad,
1860, of family Acteonidae Orbigny, 1835, is sim-
ilar to Ringicula s.s., but Tornatellaea lacks a va-
rixlike outer lip and has no notch at the anterior
end of the aperture.

The presence of Ringicula (R.) pinguis at CSUN
loc. 1344 is the first record of this species in “Me-
ganos Stage” strata.

Order Opisthobranchia Milne-Edwards,
1848

Family Cylichnidae A. Adams, 1854

Genus Cylichnina Monterosato, 1884

Type Species. Bulla umbilicata Montagu, 1803,
by original designation; Recent, Norway.

Cylichnina tantilla (Anderson and Hanna,
1925)

Figures 47, 48

Cylichnella tantilla Anderson and Hanna, 1925:
140, pi. 7, figs. 4, 8, 9.

Cylichnina tantilla (Anderson and Hanna). Stewart,
1927:439-441, pi. 27, figs. 2-4; 1946:pl. 11, fig.
11; Turner, 1938:67-68, pi. 20, figs. 9, 10;
Vokes, 1939:110, pi. 16, figs. 28, 33, 39; Weaver,
1943:548-549, pi. 100, figs. 10-12, 14-15;
Squires, 1983a, fig. 9a; 1984:40, fig. 9p; 1988b:
17, fig. 43; 199 1 b:pl. 2, fig. 14; Squires et al.,
1992:pl. 1, fig. 24; Squires and Demetrion, 1992:
34, fig. 94.

PRIMARY TYPE MATERIAL. CAS holotype
958, CAS paratypes 959 and 960, all from CAS
locality 711, Tejon Formation, Grapevine Canyon,
southern end of San Joaquin Valley, California.

MOLLUSCAN STAGE RANGE. “Meganos”
through “Tejon.”

GEOGRAPHIC DISTRIBUTION. San Laguna
Ignacio area, Baja California Sur, Mexico, to south-
western Washington.

LOCAL OCCURRENCE. CSUN Iocs. 961?,
964, 967, 970, 1344, 1346.

REMARKS. Specimens are rare and generally
not well preserved. This is the first report of this
species from “Meganos Stage” strata.

Class Bivalvia Linnaeus, 1758

Order Nuculoida Dali, 1889

Family Nuculidae Gray, 1824

Genus Acila H. Adams and A. Adams, 1858

Type Species. Nucula divaricata Hinds, 1843, by
subsequent designation (Stoliczka, 1871); Miocene
to Pliocene, Japan; Recent, Japan, China, and Ko-
rea.

Subgenus Truncacila Grant and Gale, 1931

Type Species. Nucula castrensis Hinds, 1843, by
original designation; Pliocene to Pleistocene, Cali-
fornia; Recent, northeastern Pacific.

Acila (Truncacila) decisa (Conrad, 1855)

Figure 49

Nucula decisa Conrad, 1855:11-12; 1857:pl. 3, fig.
19.

Acila gabbiana Dickerson, 1916:481, pi. 36, fig. 1;
Anderson and Hanna, 1925:176, pi. 9, fig. 12.

22 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

Nucula ( Acila ) stillwaterensis Weaver and Palmer,
1922:6, pi. 8, fig. 8.

Acila lajollaensis Hanna, 1927:270, pi. 25, figs. 1,
3, 5, 7-8, 12, 15.

Acila ( Truncacila ) decisa (Conrad). Schenck, 1936:
53-56, pi. 3, figs. 1-9, 11-15; pi. 4, figs. 1-2;
text fig. 7 (22, 23, 25); Turner, 1938:41-42, pi.
5, figs. 2-3; Vokes, 1939:41, pi. 1, figs. 7-8;
Weaver, 1943:22-23, pi. 6, figs. 1, 4, 8; pi. 7,
figs. 8-9; Moore, 1968:30, pi. 13a; 1983:A10,
pi. 1, fig. 14; Givens, 1974:38, pi. 1, fig. 1;
Squires, 1984:41, fig. 10a; 1987:54, fig. 86;
1988b:17, fig. 44; 1991b:pl. 2, fig. 15; Throck-
morton, 1988:pl. 1, fig. 19; Squires and Goedert,
1997:fig. 2g.

PRIMARY TYPE MATERIAL. UCMP neotype
31132, designated by Schenck (1936), UCMP loc.
5062, Ardath Shale, San Diego County, California.

MOLLUSCAN STAGE RANGE. “Martinez”
through upper Eocene ( Turritella schencki dela-
guerrae Zone of Kleinpell and Weaver, 1963).

GEOGRAPHIC DISTRIBUTION. San Diego,
California, to Kamchatka, Russia.

LOCAL OCCURRENCE. CSUN Iocs. 958, 959,
961.

REMARKS. Specimens are scarce to uncommon
and are preserved as external molds. They are most
abundant at locality 961, where six specimens were
found. The Kamchatka occurrence of this species is
in lower Eocene strata along the shore of the Pen-
zhin Inlet (northern Sea of Okhotsk) and the nearby
Koryak Uplands to the east (Devyatilova and Vo-
lobueva, 1981).

Family Nuculanidae H. Adams and A.
Adams, 1858

Genus Nuculana Link, 1807

Type Species. Area rostrata Chemnitz, 1784, by
original designation; Recent, North Atlantic.

Subgenus Saccella Woodring, 1925

TYPE SPECIES. Area fragilis Chemnitz, 1784,
by original designation; Recent, Mediterranean Sea.

Nuculana ( Saccella ) gabbii (Gabb, 1869)
Figure 50

Leda} protexta Conrad. Gabb, 1864:199 (in part),
pi. 26, fig. 185.

Not Leda ? protexta Gabb, 1860:303, pi. 48, fig.
23.

Nuculana gabbii Conrad, 1866:3, nomen nudum.
Leda gabbii (Conrad). Gabb, 1869:197.

Leda gabbii (Conrad). Stanton, 1896:1041, pi. 64,
fig. 8; Arnold, 1907a:pl. 10, fig. 1; 1910:pl. 2,
fig. 8; Arnold and Anderson, 1910:pl. 24, fig. 8;
Waring, 1917:76, pi. 13, fig. 6; Dickerson, 1915:
pi. 1, fig. 1; 1916:pl. 36, fig. 3; Clark, 1929:pl.
3, fig. 12; Clark and Woodford, 1927:85-86, pi.
14, fig. 2.

Contributions in Science, Number 479

Leda vogdesi Anderson and Hanna, 1925:177-
179, pi. 2, figs. 8, 9.

Saccella gabbii (Gabb). Stewart, 1930:55-58, pi. 7,
fig. 3, pi. 10, fig. 4.

Nuculana ( Saccella ) gabbii (Gabb). Vokes, 1939:
41-42; Kleinpell and Weaver, 1963:195, pi. 28,
fig. 1; Givens, 1974:39, pi. 1, fig. 3; Moore,
1983:A16, pi. 2, figs. 7, 8; Squires, 1984:41, fig.
10b.

Nuculana gabbii (Gabb). Demere et al., 1979:pl. 1,
fig. 13.

? Nuculana ( Calorhadia ) gabbii (Gabb). Zinsmeis-
ter, 1983a:pl. 1, fig. 3.

PRIMARY TYPE MATERIAL. ANSP lectotype
ANSP 4476 (of Leda gabbii Gabb, 1869), desig-
nated by Stewart (1930), “Tejon Formation,” Mar-
tinez, California.

MOLLUSCAN STAGE RANGE. “Martinez”
through upper Eocene ( Turritella schencki dela-
guerrae Zone of Kleinpell and Weaver, 1963).

GEOGRAPHIC DISTRIBUTION. Simi Valley,
California, to Kamchatka, Russia.

LOCAL OCCURRENCE. CSUN Iocs. 958, 960,
961, 967, 1342?

REMARKS. Only a few poorly preserved inter-
nal molds were found, except at CSUN locality 967
where a well-preserved external mold was found.
The Kamchatka occurrence of this species is in low-
er Eocene strata along the shore of the northern Sea
of Okhotsk (Devyatilova and Volobueva, 1981).

Order Arcoida Stoliczka, 1871

Family Arcidae Lamarck, 1809

Genus Area Linnaeus, 1758

TYPE SPECIES. Area noae Linnaeus, 1758, by
subsequent designation (Schmidt, 1818); Recent,
Mediterranean Sea and northwest Africa.

Subgenus Area s.s.

Area (Area) filewiezi Squires, 1991a
Figure 51

Area (Area) filewiezi Squires, 1991a:68-69,
figs. 2-6.

PRIMARY TYPE MATERIAL. LACMIP holo-
type 8365 and LACMIP paratype 8366; both from
CSUN loc. 965, upper 100 m of the Santa Susana
Formation, north side of Simi Valley, California.
MOLLUSCAN STAGE RANGE. “Meganos.”
GEOGRAPHIC DISTRIBUTION. North side of
Simi Valley, California.

LOCAL OCCURRENCE. CSUN loc. 965
REMARKS. Two specimens were found; one is
closed valved.

Family Spondylidae Gray, 1826

Genus Spondylus Linnaeus, 1758

Type Species. Spondylus gaederopus Linnaeus,
1758, by subsequent designation (Schmidt, 1818);
Recent, Mediterranean Sea and northwest Africa.

Squires: “Meganos Stage” Marine Megafossils ■ 23

Spondylus carlosensis Anderson, 1905

Figures 52, 53

Spondylus carlosensis Anderson, 1905:194, pi. 13,
fig. 1; Arnold, 1910:pl. 2, figs. 6,7; Dickerson,
1915:pl. 1, fig. 7; Anderson and Hanna, 1925:
189-190, text fig. 10; Vokes, 1939:57, pi. 3, figs.
10, 13; Kleinpell and Weaver, 1963:199, pi. 31,
fig. 6; Squires, 1984:43, fig. 10); Moore, 1987:
C6-C7, pi. 1, fig. 5; Squires and Goedert, 1994:
23, fig. 55.

Spondylus cf. S. carlosensis Anderson. Squires,
1991b: pi. 2, fig. 17.

PRIMARY TYPE MATERIAL. CAS holotype
56, west and north of Coalinga, NW 1/4 of section
35, T 20 S, R 14 E, Domengine Formation, Fresno
County, California.

MOLLUSCAN STAGE RANGE. “Meganos”
through middle part of “Tejon.”

GEOGRAPHIC RANGE. Simi Valley, Califor-
nia, to southwestern Washington.

LOCAL OCCURRENCE. CSUN loc. 1343.
REMARKS. Two specimens were found. One is
small (height 20 mm), closed valved, and well-pre-
served overall. This specimen affords new morpho-
logic information because, unlike previously de-
scribed and illustrated specimens of this species, it
shows both the left (Fig. 52) and right (Fig. 53)
valves rather than only the left valve. The right (at-
tached) valve is less circular and much more convex
than the left. The radial ribs on the right valve are
less closely spaced, wider, less sharp sided, and
more spinose than those on the left. Usually, every
fourth or fifth radial rib on both valves (especially
on the right valve) is more prominent than the oth-
er ribs. Although the auricles are mostly missing on
the right valve, those on the left valve are intact (a
rare condition). The auricles on the left valve are
small, and the anterior one has coarse growth lines
and two strong and wide radial ribs, one of which
delineates the hinge-line. The posterior auricle on
the left valve is smooth, but it is set off from the
rest of the valve by a moderately strong radial rib.
On both valves, the beak is anterior of the valve
center.

The other specimen of Spondylus carlosensis
found at CSUN locality 1343 is a portion of a right
valve attached to a specimen of the solitary coral
? Antillopbyllia californica new species.

The presence of Spondylus carlosensis in the up-
per 100 m of the Santa Susana Formation is the
first record of this species from “Meganos Stage”
strata.

Order Ostreoida Ferussac, 1822
Family Gryphaeidae Vyalov, 1936

Genus Pycnodonte Fischer de Waldheim,
1835

TYPE SPECIES. Pyncodonte radiata Fischer de
Waldheim, 1835, by original designation; Late Cre-
taceous, Crimea.

Subgenus Phygraea Vyalov, 1936

TYPE SPECIES. Grypbaea ( Grypbaea ) sec. Phy-
graea frauscberi Vyalov, 1936, by original desig-
nation; late Paleocene, Austria.

Pycnodonte ( Pbygraea ) sp., aff. Pycnodonte

(Phygraea) pacifica Squires and Demetrion,
1990

Figures 54-56

LOCAL OCCURRENCE. CSUN loc. 1343.

REMARKS. A single left (lower valve) was
found, and 13 right (upper) valves were found. The
left valve is weathered, missing some shell material,
probably not complete, and infilled with very hard
matrix. It is smooth, very convex, and has a prom-
inent winglike extension separated from the main
part of the valve by a moderately shallow sulcus
that deepens ventrally. Only one of the right valves
is mostly complete. It is smooth, flattish, has ver-
micular anachomata, and has a prominent winglike
extension bearing a finely granular appearance be-
cause of vesicular shell structure. In addition, there
is a prominent ridge interiorly where the right valve
joins the left valve. The specimens have affinity
with Pcynodonte ( Pbygraea ) pacifica Squires and
Demetrion, 1990, from the “Capay Stage” through
the lower middle part of the “Tejon Stage” within
the Bateque Formation in Baja California Sur, Mex-
ico. The Santa Susana Formation specimens differ
from Pycnodonte ( Pbygraea ) pacifica Squires and
Demetrion (1990:386, fig. 3. 1-3.4) by having a
smaller size, more prominent anachomata, and, ap-
parently, a radial sulcus that originates farther from
the umbo. These differences, however, might be re-
lated to growth stage. It is possible that all the San-
ta Susana Formation specimens are juveniles. All
known specimens of Pycnodonte ( Phygraea ) paci-
fica are adults. Until more Santa Susana Formation
specimens (especially of the left valve) are found, it
cannot be positively determined if the already-col-
lected material represents a new species or whether
it represents only the juvenile stage of Pycnodonte
(Phygraea) pacifica.

The only other species of Pycnodonte (Phygraea)
known from the Paleogene rock record of the Pa-
cific coast of North America is P. (P.) cuarentaensis
Squires and Demetrion, 1994, from the “Capay
Stage” part of the Bateque Formation in Baja Cal-
ifornia Sur, Mexico. Pycnodonte (Pbygraea) cuar-
entaensis Squires and Demetrion (1994:132-133,
figs. 17-22) differs from both Pycnodonte (Phy-
graea) pacifica and the Santa Susana Formation
specimens by having fine radial ribbing on the left
valve.

The specimens of Pycnodonte (Phygraea) sp., aff.
Pycnodonte (Phygraea) pacifica in the upper 100 m
of the Santa Susana Formation in Simi Valley rep-
resent the earliest occurrence of Phygraea on the
Pacific coast of North America, and its first occur-

24 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

rence on the Pacific coast of North America outside
of Baja California Sur, Mexico.

Order Veneroida H. Adams and A. Adams,
1856

Family Fimbriidae Nicol, 1950

Genus Fimbria Megerle von Miihlfeld, 1811

TYPE SPECIES. Fimbria magna Megerle von
Miihlfeld, 1811 [= Venus fimbriata Linnaeus,
1758], by original designation; Recent, Indo-Pacif-
ic.

Fimbria susanensis Squires, 1990
Figure 57

Fimbria susanensis Squires, 1990:554, fig. 2. 1-2.3.

PRIMARY TYPE MATERIAL. UCMP holotype

38568 from UCMP loc. 7009; UCMP paratype

38569 from UCMP loc. 3792. Both specimens from
the upper 100 m of the Santa Susana Formation,
south side of Simi Valley, California.

MOLLUSCAN STAGE RANGE. “Meganos.”
GEOGRAPHIC RANGE. South side Simi Valley,
California.

LOCAL OCCURRENCE. CSUN loc. 1342 and
UCMP loc. 7009.

REMARKS. Seven specimens were found, and a
few show the hinge. Fimbria susanensis has been
reported from UCMP localities 3792 and 7009,
but, as mentioned under Velates perversus, the ex-
act location of UCMP locality 3792 is not known.

Squires (1990) incorrectly equated CSUN locality
1342 with UCMP loc. 3791. The latter locality,
which is in close proximity to CSUN locality 1342,
actually is the same as CSUN locality 1343.

Fimbria susanensis is the earliest occurrence of
Fimbria in North America (Squires, 1990).

Family Carditidae Fleming, 1828

Genus Venericardia Lamarck, 1801

TYPE SPECIES. Venericardia imbricata La-
marck, 1801 [= Venericardia imbricata Gmelin,
1791], by subsequent designation (Schmidt, 1818);
Eocene, Paris Basin France.

Subgenus Pacificor Verastegui, 1953

TYPE SPECIES. Venericardia mulleri Verastegui,
1953, by original designation; Paleocene, Califor-
nia.

Venericardia (Pacificor) calafia susanaensis
Verastegui, 1953
Figure 58

Venericardia ( Pacificor ) susanaensis Verastegui,
1953:22-23, pi. 5, figs. 1-4.

Venericardia ( Pacificor ) hornii susanaensis Veras-
tegui. Saul, 1983:pl. 1, fig. 14.

Venericardia ( Pacificor ) calafia susanaensis Veras-

Contributions in Science, Number 479

tegui. Moore, 1992:E19-E20, pi. 1, figs. 14, 16-
18.

PRIMARY TYPE MATERIAL. CAS holotype
8004, “Santa Susana shale,” McCray Oil Wells, Oil
Canyon, north side of Simi Valley (exact locality
unknown), California.

MOLLUSCAN STAGE RANGE. “Meganos.”
GEOGRAPHIC DISTRIBUTION. North side of
Simi Valley, California.

LOCAL OCCURRENCE. CSUN Iocs. 959, 967.
REMARKS. Only a few fragments were found at
localities 959 and 967. Elsewhere, in the northeast
corner of section 31, T 3 N, R17W, a few meters
below the base of the Llajas Formation, I found an
internal mold of a closed-valved specimen. No oth-
er megafossils were found with this internal mold.
In addition, new housing construction in the south-
west corner of section 32, T 3 N, R 17 W uncov-
ered a nearly complete, closed-valved specimen of
this bivalve (James Rohrer, Petras Company, per-
sonal communication), which I was able to inspect
and identify.

The holotype of Venericardia (P.) calafia susan-
aensis is incomplete. The exact location of its type
locality is not known, but it is most likely in the
upper 100 m of the Santa Susana Formation based
on the recent discovery of this subspecies in this
part of the formation (see above). The locality of
the specimen illustrated by Saul (1983:pl. 1, fig. 14)
is UCMP locality 7000, and the exact location of
this locality, which is 7.5 m below the Llajas For-
mation according to Saul (1983), is also not
known. Locality 7000 is also where the only spec-
imens of Turritella meganosensis known from the
upper Santa Susana Formation have been found.

Family Cardiidae Lamarck, 1809

Genus Nemocardium Meek, 1876

TYPE SPECIES. Cardium semiasperum Deshayes,
1858, by subsequent designation (Sacco, 1899); Eo-
cene, Paris Basin, France.

Nemocardium linteum (Conrad, 1855)
Figure 59

Cardium linteum Conrad, 1855:3, 9; 1857:pl. 2,
fig. 1; Anderson and Hanna, 1925:166-167, pi.
3, fig. 3.

Cardium cooperii Gabb, 1864:172, pi. 24, figs.
154-154a; Arnold, 1907b:pl. 38, figs. 2-2a;
Waring, 1917:pl. 13, fig. 3; Hanna, 1927:285, pi.
41, figs. 6, 7.

Cardium dalli Dickerson, 1913:289, pi. 14, fig.
4a-c.

Not Cardium dalli Heilprin, 1887:131, pi. 16a, fig.
70.

Cardium marysvillensis Dickerson, 1916:482 [new
name for Cardium dalli Dickerson, 1913, pre-
occupied].

Cardium ( Protocardium ) marysvillensis Dickerson.
Clark and Woodford, 1927:94, pi. 15, fig. 12.

Squires: “Meganos Stage” Marine Megafossils ■ 25

Figures 57-68. Bivalves, crabs, and spatangoid echinoid from the upper 100 m of the Santa Susana Formation, Simi
Valley. All specimens coated with ammonium chloride. 57-65. Bivalves. 57. Fimbria susanensis Squires, 1990, left valve,
XI, UCMP holotype 38568, UCMP loc. 7009. 58. Venericardia ( Pacificor ) calafia susanensis Verastegui, 1953, right
valve, X0.8, UCMP hypotype 37431, UCMP loc. 7009. 59. Nemocardium linteum (Conrad, 1855),? right valve, X2.4,
LACMIP hypotype 12698, LACMIP loc. 26611 [= CSUN loc. 965]. 60. Saulella undulifera (Gabb, 1869), latex peel of
external mold, right valve, X2.1, LACMIP hypotype 12699, LACMIP loc. 26610 [= CSUN loc. 967]. 61. Macoma rosa
Hanna, 1927, internal mold, left valve, X3.6, LACMIP hypotype 12700, CSUN loc. 1342. 62. Pitar uvasana coquillensis
Turner, 1938, left valve, X2.4, LACMIP hypotype 12701, CSUN loc. 1344. 63. Corbula ( Caryocorbula ) dickersoni

26 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

Nemocardium linteum (Conrad). Stewart, 1930:
275-277, pi. 8, fig. 6; Turner, 1938:52, pl.10, fig.
10; Vokes, 1939:76-77, pi. 11, figs. 6, 9; Weaver,
1943:159-160, pi. 38, fig. 3; 1953:28; Stewart,
1946:pl. 11, fig. 19; Moore, 1968:30, pi. 13, fig.
d; Zinsmeister, 1983a:pl. 2, fig. 7; Squires, 1984:
49-50, fig. 12c; 1987:65, 67, fig. 113; 1988b:19,
fig. 51; Squires et al., 1992:pl. 1, fig. 33; Squires
and Demetrion, 1992:42, fig. 121.

Cardium ( Nemocardium ) linteum Conrad. Klein-
pell and Weaver, 1963:202, pi. 34, fig. 4.

PRIMARY TYPE MATERIAL. USNM holotype
1834, Domengine Formation near Martinez, Cali-
fornia.

ILLUSTRATED SPECIMEN. LACMIP hypo-
type 12248.

MOLLUSCAN STAGE RANGE. “Martinez”
through “Tejon.”

GEOGRAPHIC RANGE. Eastern Laguna San
Ignacio area, Baja California Sur, Mexico, to Pulali
Point, Jefferson County, Washington.

LOCAL OCCURRENCE. CSUN Iocs. 965,
1553, 1553a, 1554, 1555.

REMARKS. Specimens are uncommon at CSUN
locality 1553 and rare at the other localities.

Family Tellinidae Blainville, 1814

Genus Saulella Zinsmeister, 1983b

TYPE SPECIES. Tellina undulifera Gabb, 1869,
by original designation; Paleocene, California.

Saulella undulifera (Gabb, 1869)

Figure 60

Tellina undulifera Gabb, 1869:183, pi. 3, fig. 74;

Dickerson, 1914:pl. 11, fig. 7a-7c.

“ Tellina ?” undulifera Gabb. Stewart, 1930:204-
205, pi. 7, fig. 8; Nelson, 1925:415, pi. 53, fig.
8a, 8b.

Saulella undulifera (Gabb). Zinsmeister, 1983b:
1288, fig. II, J.

PRIMARY TYPE MATERIAL. ANSP holotype
4551, from just “west of Martinez,” California.

MOLLUSCAN STAGE RANGE. “Martinez”
and “Meganos.”

GEOGRAPHIC DISTRIBUTION. Northern
Baja California, Mexico, to northern California.
LOCAL OCCURRENCE. CSUN loc. 967.
REMARKS. Two small specimens (up to 17 mm
in height) were found, and both are external molds.
They represent the first occurrence of this species

in “Meganos Stage” strata. Saulella undulifera has
long been used as a guide fossil of the Paleocene of
the Pacific coast of North America and is common
throughout the “Martinez Stage” in California
(Nelson, 1925).

Genus Macoma Leach, 1819

TYPE SPECIES. Macoma tenera Leach, 1819 [=
Tellina calcar ea Gmelin, 1791], by monotypy; Re-
cent, Arctic.

Macoma rosa Hanna, 1927
Figure 61

Macoma rosa Hanna, 1927:292, pi. 41, figs. 2-5,

8; Clark, 1929:pl 6, fig. 15; Squires, 1984:50,

fig. 12e.

PRIMARY TYPE MATERIAL. UCMP holotype
31094, Ardath Shale, UCMP loc. 3993; UCMP
paratype 31095, UCMP loc. 5089; UCMP para-
types 31096-31097, UCMP loc. 5085; all from the
Ardath Shale, San Diego County, California.

MOLLUSCAN STAGE RANGE. “Meganos”
and “Domengine” (no specimens known from “Ca-
pay Stage”).

GEOGRAPHIC DISTRIBUTION. San Diego to
Simi Valley, California.

LOCAL OCCURRENCE. CSUN loc. 1342.

REMARKS. A single internal mold was found,
and this specimen represents the first occurrence of
this species in “Meganos Stage” strata.

Family Veneridae Rafinesque, 1815
Genus Pitar Romer, 1857

TYPE SPECIES. Venus tumens Gmelin, 1791, by
monotypy; Recent, West Africa.

Pitar uvasana coquillensis Turner, 1938
Figure 62

Pitar uvasana coquillensis Turner, 1938:54, pi. 11,

figs. 14-17.

PRIMARY TYPE MATERIAL. UCMP holotype
33076, UCMP loc. A-836; UCMP paratypes
33077-33078, UCMP loc. A-838; all from Middle
Fork Coquille River, Coos County, Oregon.

MOLLUSCAN STAGE RANGE. “Meganos”
and “Capay.”

GEOGRAPHIC DISTRIBUTION. Simi Valley,
California, to Middle Fork Coquille River, Coos
County, Oregon.

Weaver and Palmer, 1922, left valve, X4.5, LACMIP hypotype 12702, CSUN loc. 1344. 64. Corbula { Caryocorbula )
parilis Gabb, 1864, left valve, X4.4, LACMIP hypotype 12703, CSUN loc. 958. 65. Netastoma squiresi Kennedy, 1993,
external (concave) mold of left valve, specimen lighted to give the effect of being a convex right valve, X3.8, LACMIP
holotype 8405, CSUN loc. 967. 66, 67. Crabs. 66. Cyclocorystes aldersoni Squires, 1980, dorsal view, X4, LACMIP
hypotype 12704, CSUN loc. 965. 67. Zantbopsis sp., aff. Z. hendersoni Rathbun, 1926, dorsal view, Xl.l, LACMIP
hypotype 12705, CSUN loc. 958. 68. Spatangoid echinoid. Scbizaster diabloensis Kew, 1920, crushed internal mold,
aboral view, Xl.9, LACMIP hypotype 12706, LACMIP loc. 26610 [= CSUN loc. 967].

Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils ■ 27

LOCAL OCCURRENCE. CSUN Iocs. 1344,
1346, 1348.

REMARKS. Specimens are rare to common and
most common at CSUN locality 1344. All speci-
mens are single valves and are mostly poorly pre-
served due to weathering. These specimens repre-
sent the first occurrence of this species in “Meganos
Stage” strata and its first occurrence outside of
southwestern Oregon.

Order Myoida Stoliczka, 1870

Family Corbulidae Lamarck, 1818

Genus Corbula Bruguiere, 1797

TYPE SPECIES. Corbula sulcata Lamarck, 1801,
by subsequent designation (Schmidt, 1818); Recent,
West Africa.

Subgenus Caryocorbula Gardner, 1926

TYPE SPECIES. Corbula alabamiensis Lea,
1833, by original designation; Eocene, Alabama.

Corbula ( Caryocorbula ) dicker soni Weaver
and Palmer, 1922
Figure 63

Corbula dicker soni Weaver and Palmer, 1922:24-
25, pi. 9, figs. 9-10; Clark, 1938:700, pi. 1, fig.
17; Weaver, 1943:257-258, pi. 61, figs. 13, 16-
17, 20; Demere et al., 1979:pl. 2, fig. 11.
Corbula ( Caryocorbula ) dicker soni Weaver and
Palmer. Vokes, 1939:98, pi. 16, figs. 1, 5, 9. Giv-
ens, 1974:57, pi. 4, fig. 7. Squires, 1984:53, fig.
12m; 1987:70-71, fig. 124.

PRIMARY TYPE MATERIAL. CAS holotype
7452, CAS paratypes 7452A-B, both from UW loc.
329, Cowlitz Formation, Lewis County, Washing-
ton.

ILLUSTRATED SPECIMEN. LACMIP hypo-
type 12249.

MOLLUSCAN STAGE RANGE. “Meganos”
through “Tejon.”

GEOGRAPHIC RANGE. San Diego, California,
to Little River area, Grays Harbor County, Wash-
ington.

LOCAL OCCURRENCE. CSUN loc. 1344.
REMARKS. A single specimen was found, and it
represents the first occurrence of this species in
“Meganos Stage” strata.

Corbula ( Caryocorbula ) parilis Gabb, 1864

Figure 64

Corbula parilis Gabb, 1864:150, pi. 29, figs. 239,
239a; Arnold, 1910:106, pi. 2, fig. 2; Dickerson,
1915:84, pi. 4, fig. 8; 1916:pl. 40, fig. 10; Han-
na, 1927:295, pi. 43, figs. 7-11, 13; Stewart,
1930:288-289, pi. 3, fig. 5; 1946:pl. 11, figs. 9,
10; Turner, 1938:65-66, pi. 8, figs. 11-14; Weav-
er, 1943:256, pi. 59, fig. 16.

Corbula ( Caryocorbula ) parilis Gabb. Vokes, 1939:

99, pi. 16, figs. 2-3, 6-7, 10; Givens, 1974:57,

pi. 4, fig. 9; Squires, 1987:71, fig. 125.

PRIMARY TYPE MATERIAL. UCMP holotype
33151, Eocene strata, Martinez, California.

MOLLUSCAN STAGE RANGE. “Meganos”
through “Transition.”

GEOGRAPHIC DISTRIBUTION. San Diego,
California, to southwestern Oregon.

LOCAL OCCURRENCE. CSUN Iocs. 958, 967.

REMARKS. Specimens are rare and are well-pre-
served single valves. These specimens represent the
first occurrence of this species in “Meganos Stage”
strata.

Family Pholadidae Lamarck, 1809
Genus Netastoma Carpenter, 1864

TYPE SPECIES. Pholas darwinii Sowerby, 1849,
by monotypy; Recent, southeastern Pacific.

Netastoma squiresi Kennedy, 1993

Figure 65

Netastoma squiresi Kennedy, 1993:400, 402, fig.

2.9, 2.10.

PRIMARY TYPE MATERIAL. LACMIP holo-
type 8405, CSUN loc. 967, upper 100 m of the
Santa Susana Formation, north side of Simi Valley,
California.

MOLLUSCAN STAGE RANGE. “Meganos.”

GEOGRAPHIC DISTRIBUTION. North side of
Simi Valley, California.

LOCAL OCCURRENCE. CSUN loc. 967.

REMARKS. Only the holotype is known for this
species. It is an external mold of a juvenile left
valve. Netastoma squiresi is the oldest known rep-
resentative of the genus, whose geologic range was
previously known as Pliocene to Recent (Kennedy,
1993).

Phylum Arthropoda Siebold and Stannius,
1848

Class Malacostraca Latreille, 1806
Order Decapoda Latreille, 1803

Family Xanthidae MacLeay, 1838
Genus Cyclocorystes Bell, 1858

TYPE SPECIES. Cyclocorystes pulchellus Bell,
1858, by original designation; early Eocene, Eng-
land.

Cyclocorystes aldersoni Squires, 1980

Figure 66

Cyclocorystes aldersoni Squires, 1980:474-475,

figs. 2, 3.

PRIMARY TYPE MATERIAL. LACMIP holo-
type 5893; LACMIP paratypes 5864-5866; all
from CSUN loc. 354, upper Santa Susana Forma-

28 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

tion, east-central Santa Monica Mountains, Cali-
fornia.

MOLLUSCAN STAGE RANGE. “Martinez”
and “Meganos.”

GEOGRAPHIC DISTRIBUTION. Garapito
Creek, east-central Santa Monica Mountains, Los
Angeles County, and south side of Simi Valley; both
in southern California.

LOCAL OCCURRENCE. CSUN Iocs. 965, 966.

REMARKS. Specimens are rare, moderately well
preserved, and missing their legs. These specimens
represent the first occurrence of this species in “Me-
ganos Stage” strata and its first occurrence outside
of the Santa Monica Mountains.

Genus Zanthopsis M’Coy, 1849

TYPE SPECIES. Cancer leachii Desmarest, 1822,
by original designation; early Eocene, England.

Zanthopsis sp., aff. Z. hendersoni Rathbun,
1926
Figure 67

LOCAL OCCURRENCE. CSUN loc. 958.

REMARKS. Two internal molds were found.
They show close affinity with Zanthopsis hender-
soni Rathbun, 1926, from Oligocene rocks in
Oregon and from the upper part of the Santa Su-
sana Formation on the north side of Simi Valley.
The exact location of the type locality of Z. hen-
dersoni is not known, but it is near Eugene in Lane
County, Oregon. The specimens from CSUN local-
ity 958 differ from Z. hendersoni Rathbun (1926:
53-54, pi. 10, figs. 5, 6) by having a frontal region
that is much less produced and not dentate and by
having a tubercle on each of the protogastric areas.
The frontal region on specimens from CSUN local-
ity 958 is essentially straight and without teeth. It
is possible that the material from the upper part of
the Santa Susana Formation represents a new spe-
cies. More specimens of Z. hendersoni are needed
to determine the full range of morphologic vari-
ability of this species. Only then will it be possible
to decide if the specimens of Z. sp., aff. Z. hender-
soni represent a new species.

The geologic range of genus Zanthopsis is Paleo-
cene to Oligocene, with distribution in Europe,
West Africa, West Indies, Panama, and North
America (Glaessner, 1969).

Phylum Echinodermata Klein, 1734
Class Echinoidea Leske, 1778
Order Spatangoida Claus, 1876
Family Schizasteridae Lambert, 1905
Genus Schizaster Agassiz, 1836

TYPE SPECIES. Schizaster studeri Agassiz, 1836,
by subsequent designation (ICZN, 1948:523-529,
opin. 209); late Eocene, Italy and southern France.

Schizaster diabloensis Kew, 1920
Figure 68

Schizaster diabloensis Kew, 1920:150-151, pi. 41,
fig. 5a-c; Clark and Woodford, 1927:123, pi. 22,
fig. 14; Clark, 1929:pl. 4, fig. 13; Squires, 1984:
56, fig. 13d; 1994:pl. 3, fig. 8.

PRIMARY TYPE MATERIAL. UCMP holotype
11387, UCMP loc. 1427, Eocene strata, south side
of Mount Diablo, California.

MOLLUSCAN STAGE RANGE. “Meganos”
through middle part of “Tejon.”

GEOGRAPHIC DISTRIBUTION. Simi Valley,
California, through Marysville Buttes, California.
LOCAL OCCURRENCE. LACMIP loc. 26610.
REMARKS. Three specimens were found. Two
are badly crushed internal molds. The other speci-
men (Fig. 68) is a partial external mold.

Clark and Woodford (1927) and Clark (1929)
reported Schizaster diabloensis as occurring in the
“Meganos horizon” at various places throughout
California, including Simi Valley, but they provided
few stratigraphic details. The specimens from LAC-
MIP locality 26610 confirm the presence of this
species in “Meganos Stage” strata. Squires (1994)
recently reported the youngest occurrence of S. dia-
bloensis to be in the Coldwater Sandstone (middle
part of the “Tejon Stage”) in upper Sespe Creek,
Ventura County, California.

LOCALITIES

All base maps are U.S. Geological Survey, 7.5-minute (un-
less otherwise stated), topographic quadrangles.

CAS LOCALITIES

711. “On the east side of Grapevine Canyon near the
point where the stream flows out upon the valley floor”
(Anderson and Hanna, 1925:39). Grapevine quadrangle,
Kern County, California.

30667. At elevation 800 m along crest of ridge on north
side of Media Agua Creek, 442 m north and 183 m east
of SW corner of section 27, T 28 S, R 19 E, La Yeguas
Ranch quadrangle, 1959, Kern County, south-central Cal-
ifornia.

CSUN LOCALITIES

NORTH SIDE OF SIMI VALLEY

All are in the upper 100 m of the Santa Susana For-
mation (“Meganos Stage”) and, unless otherwise stated,
in the Santa Susana quadrangle, 1951 (photorevised
1969), Ventura County, southern California. See Fig. 2 for
stratigraphic position relative to the base and top of the
upper 100 m of the Santa Susana Formation.

354. East bank of the south fork of Garapito Creek,
518m S20°E from the intersection of the San Bernardino
baseline and Los Angeles City boundary, Topanga quad-
rangle, 1952, Los Angeles County, southern California.
Collected by R. Squires, 1979.

958. Bulldozer-generated exposure now under houses,
at elevation of 346 m, just west of intersection of Chu-
mash Street and Indian Hills Drive, 381m south and 107

Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils ■ 29

m east of section 5, T 2 N, R 17 W. Collected by R.L.
Squires, Aug. 19, 1984.

959. At elevation of 347 m, 373 m east and 55 m south
of NW corner of section 5, T 2 N, R 17 W. Collected by
R.L. Squires, Aug. 24, 1984.

960. At elevation of 390 m, on south bank of Las Llajas
Canyon, 594 m east and 579 m north of SW corner of
section 32, T 3 N, R 17 W. Collected by R.L. Squires,
Aug. 24, 1984, and Feb. 28, 1986.

961. At elevation of 376 m, on north bank of Las Llajas
Canyon, 693 m east and 739 m north of SW corner of
section 32, T 3 N, R 17 W. Collected by R.L. Squires,
Aug. 24, 1984.

962. At elevation of 451 m, on east side of dirt road,
756 m east and 411 m south of NW corner of section 32,
T 3 N, R 17 W. Collected by R.L. Squires, February 28,
1986.

963. At elevation of 483 m, 735 m east and 305 m
south of NW corner of section 32, T 3 N, R 17 W. Col-
lected by R.L. Squires, February 28, 1986.

964. [= LACMIP 21551]. At elevation of 509 m, in
middle of dirt road, 792 m east and 171 m south of NW
corner of section 32, T 3 N, R 17 W. Collected by R.L.
Squires, February 28, 1986.

965. [= LACMIP 16111]. At elevation of 527 m, on
east side of dirt road, 792 m east and 94 m south of NW
corner of section 32, T 3 N, R 17 W. Collected by R.L.
Squires, Feb. 28, 1986.

966. [= LACMIP 16893 and LACMIP 26609]. At el-
evation of 529 m, on east side of dirt road, 792 m east
and 152 m south of NW corner of section 32, T 3 N, R
17 W. Collected by R.L. Squires, February 28, 1986.

967. [= LACMIP 12648 and LACMIP 26610]. At el-
evation of 533 m, just east of dirt road and on northeast
side of small hill, 762 m east and 183 m north of SW
corner of section 29, T 3 N, R 17 W. Collected by H.
Seiden, 1951, and R.L. Squires, Feb. 28, 1986.

968. At elevation of 427 m, 457 m east and 308 m
south of NW corner of section 32, T 3 N, R 17 W. This
localtiy is about 3 m stratigraphically below base of Llajas
Formation. Collected by R.L. Squires, Feb. 28, 1986.

969. [= LACMIP 16894]. At elevation of 381 m, on
east side of Chivo Canyon, 343 m west and 107 m north
of SE corner of section 30, T 3 N, R 17 W. Collected by
R.L. Squires, March 1, 1986.

970. At elevation of 415 m, on north side of small trib-
utary of east side of Chivo Canyon, 119 m west and 122
m north of SE corner of section 30, T 3 N, R 17 W.
Collected by R.L. Squires, March 1, 1986.

971. At elevation of 401 m, on north side of small trib-
utary of east side of Chivo Canyon, 21 m west and 107
m north of SE corner of section 30, T 3 N, R 17 W.
Collected by R.L. Squires, March 1, 1986.

972. At elevation of 407 m, on north side of small trib-
utary of east side of Chivo Canyon, 30 m west and 119
m north of SE corner of section 30, T 3 N, R 17 W.
Collected by R.L. Squires, March 1, 1986.

973. [= LACMIP 16895]. At elevation of 412 m, near
head of small tributary of east side of Chivo Canyon, 31
m west and 122 m north of SE corner of section 30, T 3
N, R 17 W.

SOUTH SIDE OF SIMI VALLEY

All are in the upper 100 m of the Santa Susana For-
mation (“Meganos Stage”) and, unless otherwise stated,
in the Calabasas quadrangle, 1952 (photorevised 1967),
Ventura County, southern California.

1342. Top of hill at elevation of 431 m, 38 m west and
762 m north of SE corner of section 22, T 2 N, R 18 W.
Collected by R.L. Squires, Aug. 14, 1989. This locality is
in close proximity to CSUN loc. 1343 and 3 m strati-
graphically below it.

1343. [= UCMP 3791]. On east side of hill at elevation
of 427 m, 23 m west and 785 m north of SE corner of
section 22, T 2 N, R 18 W. Collected by R.L. Squires,
Aug. 14, 1989. This locality is in close proximity to CSUN
loc. 1342 and 3 m stratigraphically above it.

1344. In saddle at elevation of 419 m, 183 m west and
823 m north of SE corner of section 22, T 2 N, R 18 W.
Collected by R.L. Squires, Aug. 14, 1989.

1345. On west side of Runkle Canyon, at elevation of
355 m, 664 m east and 168 m south of NW corner of
section 22, T 2 N, R 18 W. Collected by R.L. Squires,
Aug. 16, 1989.

1346. On west side of Runkle Canyon, at elevation of
335 m, 739 m east and 148 m south of NW corner of
section 22, T 2 N, R 18 W. Collected by R.L. Squires,
Aug. 16, 1989.

1347. On west side of Runkle Canyon, at elevation of
353 m, 655 m east and 125 m south of NW corner of
section 22, T 2 N, R 18 W. Collected by R.L. Squires,
Aug. 16, 1989.

1348. On west side of Runkle Canyon, at elevation of
325 m, 777 m east and 61 m south of NW corner of
section 22, T 2 N, R 18 W. Collected by R.L. Squires,
Aug. 16, 1989.

1349. On east side of Runkle Canyon, at elevation of
317 m, 579 m west and 1579 m south of NE corner of
section 14, T 2 N, R 18 W. Collected by R.L. Squires,
Aug. 16, 1989.

1565. At elevation of 340 m, along west side of Bus
Canyon, on north bank of an unnamed tributary that en-
ters Bus Canyon from the west, 274 m south and 503 m
west of NE corner of section 28, T 2 N, R 18 W, Thou-
sand Oaks quadrangle, 1950 (photorevised 1967), Ven-
tura County, southern California. Collected by A. I. Mar-
ro, 1985.

LACMIP LOCALITIES

461-B. On the northern slope of a small canyon inter-
secting Las Llajas Canyon from the east, Santa Susana
quadrangle, 1959 (photorevised 1961), Ventura County,
southern California. Collected by J.A. Sutherland, circa
early 1960s. This locality is 61 m (200 ft.) from the top
of the Llajas Formation.

7124. At elevation of 343 m, 610 m S45°W of 1480 ft.
hill, SW 1/4 of section 13, T 2 N, R 18 W, Calabasas
quadrangle, 1952 (photorevised 1967), Ventura County,
southern California. Collected by W.P. Popenoe and M.
Sperling, July 4, 1929.

12648. See CSUN 967.

21551. See CSUN 964.

23173. [Exact stratigraphic position not known.] About
15 m downslope from top of east-west trending ridge,
southwest of Runkle Canyon, approximately 884 m north
and 427 m west of SE corner of section 27, T 2 N, R 18
W, Calabasas quadrangle, 1952 (photorevised 1967), Ven-
tura County, southern California. Collected by J.H. Fan-
tozzi, Oct. 3, 1953.

26609. See CSUN 966.

26610. See CSUN 967.

26111. See CSUN 965.

26615. At elevation of 466 m, 518 m east of NW corner
of section 32, on section line between sections 29 and 32,

30 Q Contributions in Science, Number 479

Squires: “Meganos Stage’’ Marine Megafossils

T 3 N, R 17 W, Santa Susana quadrangle, 1959 (photo-
revised 1961), Ventura County, southern California. Col-
lected by H. Seiden, 1952.

UCMP LOCALITIES

672. South portion of crest of Parson’s Peak, SE 1/4 of
the NW 1/4 of section 24, T 18 S, R 14 E, Coalinga quad-
rangle, Fresno County, California.

1427. SW 1/4 of section 11, northeast of Wall Point,
south side of Mount Diablo, northern California.

3157. On ridge top on north side of Deer Valley, 792
m south and 411m west of NE corner of section 20, T 1
N, R 2 E, Antioch South quadrangle, 1980, Contra Costa
County, northern California. Margaret Hamilton Sand [ =
division D of Meganos Formation as used by Clark and
Woodford (1927)]. Collected by B.L. Clark and A.O.
Woodford, circa 1923.

3159. On same ridge top as UCMP loc. 3157, 1036 m
south and 46 m west of NE corner of section 20, TIN,
R 2 E, Antioch South quadrangle, 1980, Contra Costa
County, northern California. Margaret Hamilton Sand [ =
division D of Meganos Formation as used by Clark and
Woodford (1927)]. Collected by B.L. Clark and A.O.
Woodford, circa 1923.

3577. On ridge top 3399 m north and 610 m east of
SW corner of Brentwood quadrangle, 1978, Contra Costa
County, northern California. Margaret Hamilton Sand [ =
division D of Meganos Formation as used by Clark and
Woodford (1927)]. Collected by B.L. Clark and A.O.
Woodford, circa 1923.

3791. [see CSUN 1343]. Collected by R.N. Nelson, cir-
ca early 1920s.

3792. [Exact stratigraphic position not known]. West
of Runkle Canyon on same ridge as UCMP loc. 3791,
NE1/4 SE1/4 of section 22, T 2 N, R 18 W, Calabasas
quadrangle, 1952 (photorevised 1967), Ventura County,
southern California. Collected by R.N. Nelson, circa early
1920s.

3796. [Exact stratigraphic position not known]. On ridge
at elevation of 640 m, 1524 m N68°W of 2150 ft. hill,
south 1/2, NE 1/4 of section 27, T 2 N, R 18 W, Calabasas
quadrangle, 1952 (photorevised 1967), Ventura County,
southern California. Collected by R.N. Nelson, circa early
1920s.

3993. In bottom of Rose Creek where creek makes a
strong bend to west, .3 km (.2 mi.) south of Bench Mark
176, 3.2 km east of La Jolla, La Jolla quadrangle, San
Diego County, California.

5062. In sea cliff south of mouth of Soledad Valley, due
west of midpoint between “P” and “u” of “Pueblo,” La
Jolla quadrangle, San Diego County, California.

5085. 2.62 inches north of the top of the “S” of “So-
ledad Mountain,” on the north side of the creek, on a
small ridge formed by the creek and sea cliff, elevation 22
m, La Jolla quadrangle, San Diego, California.

5089. 91 m (300 ft.) north of the Scripps Institution
pier, in the conglomerate above the mudstone. In the sea
cliff, elevation 3 m, La Jolla quadrangle, San Diego Coun-
ty, California.

7000. [Exact location not known]. Las Llajas Canyon,
in first canyon on north side of road, .4 km east of point
where boundary line of Century Oil property crosses road,
Santa Susana quadrangle, Ventura County, California.

7004. About 91 m east of locality 7003 in next small
canyon that enters Llajas Canyon from the east just south
of the most northerly extension of the 717-m (1500-ft.)

Contributions in Science, Number 479

contour, Santa Susana quadrangle, Ventura County, Cal-
ifornia. Locality is equivalent to CSUN loc. 374.

7009. At elevation of 378 m in a small gulley, in sandy
shale about 100 m north of UCMP loc. 3759 [= 2134 m
south of BM 961 at Santa Susana well, flank of 717-m
(1500-ft.) hill] in sandy shale, 378 m elevation, Santa Su-
sana quadrangle, Ventura County, California. Collected
by R.B. Stewart.

A-836. About 3 km west of town of Remote, in bed of
Middle Fork of Coquille River opposite Roseburg-Coos
Bay Highway 42, White Rock quadrangle, Coos County,
southwestern Oregon.

A-838. West of town of Remote, in bed of Middle Fork
of Coquille River opposite Roseburg-Coos Bay Highway
42, White Rock quadrangle, Coos County, southwestern
Oregon.

A-993. Second gulley past Marrland Canyon (now
known as Las Llajas Canyon) at second small falls up
gulley approximately 183 m, Santa Susana quadrangle,
Ventura County, California.

A-994. About 69 to 91 m down the canyon from
UCMP loc. A-993 on west side of canyon about 18 m
from streambed, Santa Susana quadrangle, Ventura Coun-
ty, California.

UW LOCALITY

329. On north bank of the Cowlitz River at bend 1.5
to 2.5 km east of Vader, section 28, T 1 N, R 2 W, Castle
Rock quadrangle (15-minute), 1953, Lewis County,
Washington.

ACKNOWLEDGMENTS

Lindsey T. Groves (LACMIP) provided access to collec-
tions and loan of specimens, as well as obtaining some
difficult-to-find references. Frederick M. Bayer (Depart-
ment of Zoology, National Museum of Natural History,
Smithsonian Institution) shared his knowledge of isidid
octocorals and provided some key literature. George E.
Davis (Natural History Museum of Los Angeles County,
Crustacean Section) helped with the systematics of the
crabs. LouElla R. Saul (LACMIP) shared her knowledge
of early Tertiary mollusks and provided the photographic
negative of the Venericardia illustrated in this report. Don-
ald W. McNamee and Mark Herbert (Natural History
Museum of Los Angeles County) arranged for interlibrary
loans of critical references on corals. Mark V. Filewicz and
H.L. Heitman (Unocal Corporation, Houston, Texas) pro-
cessed rock samples and identified calcareous nannofossils
and foraminifera, respectively. Thomas A. Demere (San
Diego Natural History Museum) and LouElla Saul criti-
cally reviewed the manuscript.

LITERATURE CITED

Abbott, R.T., and S.P. Dance. 1982. Compendium of sea
shells. New York: E.P. Dutton, 411 pp.

Adams, A. 1854. Monograph of the family Bullidae. In
Thesaurus Conchyliorum, ed. G.B. Sowerby, 553-
608, pis. 119-125. Vol. 2. London.

Adams, H., and A. Adams. 1853-1858. The genera of
Recent Mollusca ; arranged according to their orga-
nization. 2 Vols. London: John van Voorst, 660 pp.
Agassiz, L.J.R. 1836. Prodrome d’une monographic des
radiaires ou echinodermes. Memoir es de la Societe
Science Naturelle Neuchatel 1:168-199.

Almgren, A.A., M.V. Filewicz, and H.L. Heitman. 1988.
Lower Tertiary foraminiferal and calcareous nan-

Squires: “Meganos Stage” Marine Megafossils ■ 31

nofossil zonation of California — an overview and
recommendation. In Paleogene stratigraphy, west
coast of North America, ed. M.V. Filewicz and R.L.
Squires, vol. 58, 83-106. Los Angeles: Pacific Sec-
tion, Society of Economic Paleontologists and Min-
eralogists.

Anderson, F.M. 1905. A stratigraphic study in the Mount
Diablo Range of California. Proceedings of the Cal-
ifornia Academy of Sciences, series 3, 2(2):155-248.

Anderson, F.M., and G.D. Hanna. 1925. Fauna and strati-
graphic relations of the Tejon Eocene at the type lo-
cality in Kern County, California. California Acad-
emy of Sciences, Occasional Papers 11:1-249.

Arnold, R. 1907a. Geology and oil resources of the Sum-
merland district, Santa Barbara County, California.
United States Geological Survey, Bulletin 321:1-93.

Arnold, R. 1907b. New and characteristic species of fossil
mollusks from the oil-bearing Tertiary formations of
southern California. Proceedings of the United
States National Museum 32(1545):525-546.

Arnold, R. 1910. Paleontology of the Coalinga district,
Fresno and Kings Counties, California. United States
Geological Survey, Bulletin 396:1-173.

Arnold, R., and R. Anderson. 1910. Geology and oil re-
sources of the Coalinga district, California. United
States Geological Survey, Bulletin 398:1-354.

Atwater, T.M. 1998. Plate tectonic history of southern
California with emphasis on the western Transverse
Ranges and northern Channel Islands. In Contribu-
tions to the geology of the northern Channel Islands,
southern California, ed. P.W. Weigand, Miscella-
neous Publication 45, 1-8. Los Angeles: Pacific Sec-
tion, American Association of Petroleum Geologists.

Bayer, F.M. 1956. Octocorallia. In Treatise on invertebrate
paleontology, part F, Coelenterata, ed. R. C. Moore,
F166-F231. Lawrence, Kansas: University of Kansas
Press.

Bayer, F.M., and J. Stefani. 1987. Isididae (Gorgonacea)
de Nouvelle-Caledonie nouvelle cle des genres de la
famille. Bulletin du Museum National d’Histoire
Naturelle, Paris, serie 4, section A, 1:47-106.

Bell, T. 1858. A monograph of the fossil malacostracous
Crustacea. Part 1, Crustacea of the London Clay.
London: Palaeontographical Society of London, 44

pp.

Bellardi, L. 1882. I molluschi dei terreni Terziarii del Pie-
monte e della Liguria. Bollettino dei Musei di Zool-
ogia ed Anatomia Comparata della R. Universitd di
Torino, Pt. 3 (Buccinidae, Cyclopsidae, Purpuridae,
Coralliophillidae, Olividae), 265 pp.

Berggren, W.A., D.V. Kent, C.C. Swisher, III, and M.-P.
Aubry, 1995. A revised Cenozoic geochronology and
chronostratigraphy. In Geochronology, time scales
and global stratigraphic correlation, ed. W.A. Berg-
gren, D.V. Kent, and J. Hardenbol, Special Publica-
tion 54, 129-212. Tulsa, Oklahoma: SEPM (Society
for Sedimentary Geology).

Berry, S.S. 1958. Notices of new eastern Pacific Mollusca.
II. Leaflets in Malacology l(15):83-90.

Blainville, H.M.D. 1814. Sur la classification methodique
des animaux mollusques, et etablissement d’une nou-
velle consideration pour y parvenir. Bulletin Sciences
Societe Philomathematiques, Paris, series 2, 4:174-
180.

Bottjer, D.J., S.P. Lund, J.E. Powers, M.C. Steele, and R.L.
Squires. 1991. Magnetostratigraphy of Paleogene
strata in San Diego and the Simi Valley, southern
California. In Eocene geologic history San Diego re-

gion, ed. P.L. Abbott and J. A. May, 115-124. Los
Angeles: Pacific Section, Society of Economic Pale-
ontologists and Mineralogists, Volume 68.

Bruguiere, J.G. 1797. Tableau encylclopedie et metho-
dique des trois regnes de la nature; vers testacees a
coquilles bivalves. Paris. Vol. 1, pp. 85-132.

Budd, A.F., T.A. Stemann, and R.H. Stewart. 1992. Eo-
cene Caribbean reef corals: A unique fauna from the
Gatuncillo Formation of Panama. Journal of Pale-
ontology 66( 4):570-594.

Carpenter, P.P. 1864. Supplementary report on the present
state of our knowledge with regard to the Mollusca
of the west coast of America. Report 33 of the Brit-
ish Association for the Advancement of Science for
1863, pp. 517-686. [Reprinted, 1872, Smithsonian
Miscellaneous Collections 10(252): 1-1 72.]

Chemnitz, J.H. 1784. Neues systematisches Conchylien-
Cabinet. Niirnberg: Bauer and Raspe, Vol. 7, 356

pp.

Clark, B.L. 1918. Meganos Group, a newly recognized
division in the Eocene of California. Bulletin of the
Geological Society of America 29:281-296.

Clark, B.L. 1921. The stratigraphic and faunal relations
of the Meganos Group, middle Eocene of California.
Journal of Geology 29:125-1 65.

Clark, B.L. 1926. The Domengine horizon, middle Eocene
of California. University of California Publications,
Bulletin of the Department of Geological Sciences
16(5):99— 1 18.

Clark, B.L. 1929. Stratigraphy and faunal horizons of the
Coast Range of California. Privately Published, 132

pp.

Clark, B.L. 1938. Fauna from the Markley Formation
(upper Eocene) on Pleasant Creek, California. Bul-
letin of the Geological Society of America 49(5):
683-730.

Clark, B.L., and J.W. Durham. 1946. Eocene faunas from
the Department of Bolivar, Colombia. The Geolog-
ical Society of America, Memoir 16:1-126.

Clark, B.L., and H.E. Vokes. 1936. Summary of marine
Eocene sequence of western North America. Geolog-
ical Society of America Bulletin 47(6):85 1—878.

Clark, B.L., and A.O. Woodford. 1927. The geology and
paleontology of the type section of the Meganos For-
mation (lower middle Eocene) of California. Univer-
sity of California, Publications, Bulletin of the De-
partment of Geological Sciences 17(2):63— 142.

Conrad, T.A. 1855. Report on the fossil shells collected in
California by W.P. Blake. In Preliminary geological
report of W. P. Blake. U.S. Pacific Railroad explo-
ration, United States 33rd Congress, First session,
House Executive Document 129, pp. 5-20. [Reprint-
ed in Dali, 1909, United States Geological Survey,
Professional Paper 59:163-171.]

Conrad, T.A. 1857. Descriptions of the fossil shells. In
Report of explorations in California for railroad
routes. U.S. Pacific Railroad exploration, United
States 33rd Congress, Second session, Senate Exec-
utive Document 78, House Executive Document 91,
5(2):317-329.

Conrad, T.A. 1860. Description of new species of Creta-
ceous and Eocene fossils of Mississippi and Ala-
bama. Journal of the Academy of Natural Sciences
of Philadelphia, series 2, 4:275-298.

Conrad, T.A. 1866. Check list of the invertebrate fossils
of North America. Eocene and Oligocene. Smithson-
ian Miscellaneous Collections 7(200): 1-41.

32 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

Cossmann, A.E.M. 1896. Essais de paleoconchologie
comparee, vol. 2. Paris: privately published, 178 pp.

Cossmann, A.E.M. 1901. Essais de paleoconchologie
comparee. vol. 4. Paris: privately published, 293 pp.

Cossmann, A.E.M., and G. Pissarro. 1910-1913. Icono-
graphie complete des coquilles fossiles de I’Eocene
des environs de Paris, vol. 2 (Gastropodes, etc.). Par-
is: Societe Geologique de France.

Cox, L.R. 1931. A contribution to the molluscan fauna
of the Laki and basal Khirthar Groups of the Indian
Eocene. Transactions of the Royal Society of Edin-
burgh 57(2):25-92.

Dali, W.H. 1909. Contributions to the Tertiary paleontol-
ogy of the Pacific Coast. I. The Miocene of Astoria
and Coos Bay, Oregon. United States Geological
Survey, Professional Paper 59:1-278.

Davies, A.M., and F.E. Eames. 1971. Tertiary faunas. Vol.
1. The composition of Tertiary faunas, 2nd ed. Lon-
don: George Allen and Unwin, 571 pp.

Defrance, M.J.L. 1826. Dictionnaire des sciences naturel-
les, vol. 39. Strasbourg: F.G. Levrault, 559 pp.

Demere, T.A., F.A. Sundberg, and F.R. Schram. 1979. Pa-
leoecology of a protected biotope from the Eocene
Mission Valley Formation, San Diego County, Cali-
fornia. In Eocene depositional systems, San Diego,
California, ed. P. L. Abbott, 97-102. Los Angeles:
Pacific Section, Society of Economic Paleontologists
and Mineralogists, Field Trip Guide. Geological So-
ciety of America Annual Meeting.

Desmarest, A.G. 1822. Les Crustaces proprement dits. In
Histoire naturelle des crustaces fossiles, 67-142. Par-
is.

Deshayes, G.P. 1838. In J.B.P. Lamarck’s Histoire des an-
imaux sans vertebres, vol. 8, 2nd ed, by G.P. De-
shayes and H. Milne-Edwards. Paris.

Deshayes, G.P. 1856-1866. Description des animaux sans
vertebres decouverts dans le bassin de Paris. 3 Vols.
Paris: J.-B. Bailliere et fils, 2536 pp., Atlas (2 Vols.).

Devyatilova, A.D., and V.I. Volobueva. 1981. Atlas of Pa-
leogene and Neogene fauna of the northeast USSR.
Moscow: Ministry of Geology and Northeast Indus-
trial Geological Society, 219 pp. [In Russian.]

Dickerson, R.E. 1913. Fauna of the Eocene at Marysville
Buttes, California. University of California Publica-
tions, Bulletin of the Department of Geology 7(12):
257-298.

Dickerson, R.E. 1914. Fauna of the Martinez Eocene of
California. University of California Publications,
Bulletin of Department of Geology 8(6):61— 1 80.

Dickerson, R.E. 1915. Fauna of the type Tejon: Its relation
to the Cowlitz phase of the Tejon Group of Wash-
ington. Proceedings of the California Academy of
Sciences, series 4, 5(3):33— 98.

Dickerson, R.E. 1916. Stratigraphy and fauna of the Tejon
Eocene of California. University of California Pub-
lications, Bulletin of the Department of Geological
Sciences 9(17):363-524.

Douville, H. 1904. Mollusques fossiles. In Mission Scien-
tifique in Perse, ed. J. de Morgan, 191-380, vol. 3,
(Etudes Geologique), pt. 4 (Paleontologie). Paris: E.
Leroux.

Duncan, P.M. 1864. On the fossil corals of the West In-
dian Islands. Quarterly Journal of the Geological So-
ciety of London 20:20-44.

Duncan, P.M. 1873. On the older Tertiary formations of
the West Indian islands. Quarterly Journal of the
Geological Society of London 29:548-565.

Durham, J.W. 1942. Reef corals from the California mid-

Contributions in Science, Number 479

die Eocene. Proceedings of the California Academy
of Sciences, series 4, 23(34):503-510.

Durham, J.W. 1943. Pacific coast Cretaceous and Tertiary
corals. Journal of Paleontology 17(2):196-202.

Edmondson, W.F. 1984. The Meganos gorge and the geo-
logic effects produced by compaction of the gorge
fill. In Paleogene submarine canyons of the Sacra-
mento Valley, California, eds. A. A. Almgren and P.D.
Hacker, vol. 1, 37-51. Los Angeles: Pacific Section,
Society of Economic Paleontologists and Mineralo-
gists, Symposium.

Effinger,W.L. 1938. The Gries Ranch fauna (Oligocene) of
western Washington. Journal of Paleontology 12(4):
355-390.

Fantozzi, J.H. 1955. The stratigraphy and biostratigraphy
of a portion of the Simi Hills of the south side of
Simi Valley, Ventura County, California. University
of California at Los Angeles, unpublished M.A. The-
sis, 68 pp.

Filewicz, M.V., and M.E. Hill, III. 1983. Calcarous nan-
nofossil biostratigraphy of the Santa Susana and Lla-
jas formations, north side of Simi Valley. In Cenozoic
geology of the Simi Valley area, southern California,
eds. R.L. Squires and M.V. Filewicz, 45-60. Los An-
geles: Pacific Section, Society of Economic Paleon-
tologists and Mineralogists, Fall Field Trip Volume
and Guidebook.

Fischer, P. 1880-1887. Manuel de conchyliologie et de pa-
leontologie conchyliologique ou histoire naturelle
des mollusques vivants et fossils. Paris: F. Savy, 1369

pp.

Fischer de Waldeheim, G. 1835. Lettre a M. le Baron de
Ferussac sur quelques genres de coquilles du Muse-
um Demidoff et en particulier sur quelques fossiles
de la Crimee. Societe Imperiale des Nautralistes
Moscow Bulletin 8:101-119.

Fleming, J. 1828. A history of British animals, etc. Edin-
burgh: Bell and Bradfute, 565 pp.

Forbes, E. 1838. Malacologia Monensis. A catalogue of
the Mollusca inhabiting the Isle of Man and the
neighboring sea. Edinburgh, 63 pp.

Frederickson, N.O. 1983. Late Paleocene and early Eo-
cene sporomorphs and thermal alteration of organic
matter in the Santa Susana Formation, southern Cal-
ifornia. In Cenozoic geology of the Simi Valley area,
southern California, eds. R.L. Squires and M.V. Fi-
lewicz, 23-31. Los Angeles: Pacific Section, Society
of Economic Paleontologists and Mineralogists, Fall
Field Trip Volume and Guidebook.

Frost, S.H., and R.L. Langenheim, Jr. 1974. Cenozoic reef
biofacies. Tertiary larger foraminifera and scleractin-
an corals from Chiapas, Mexico. De Kalb, Illinois:
Northern Illinois University Press, 388 pp.

Gabb, W.M. 1860. Description of new species of Ameri-
can Tertiary and Cretaceous fossils. Journal of the
Academy of Natural Sciences of Philadelphia, series
2, 4(4):375-406.

Gabb, W.M. 1864. Description of the Cretaceous fossils.
Geological Survey of California, Palaeontology 1:
55-243.

Gabb, W.M. 1869. Cretaceous and Tertiary fossils. Geo-
logical Survey of California, Palaeontology 2:1-299,
pis.

Gabb, W.M. 1873. On the topography and geology of San
Domingo. Transactions of the American Philosoph-
ical Society 15:49-259.

Gardner, J.A. 1926. The nomenclature of the superspecific
Squires: “Meganos Stage” Marine Megafossils ■ 33

groups of Corbula in the lower Miocene of Florida.
The Nautilus 4:41-47.

Gardner, J.A., and E. Bowles. 1934. Early Tertiary species
of gastropods from the Isthmus of Tehuantepec.
Journal of the Washington Academy of Sciences
24(6):241-248.

Givens, C.R. 1974. Eocene molluscan biostratigraphy of
the Pine Mountain area, Ventura County, California.
University of California, Publications in Geological
Sciences 109:1-107.

Givens, C.R. 1978. An occurrence of the Tethyan genus
Volutilithes (Gastropoda: Volutidae) in the Eocene of
California. Journal of Paleontology 52(1):104— 108.

Givens, C.R. 1989. First record of the Tethyan genus Vol-
utilithes (Gastropoda: Volutidae) in the Paleogene of
the Gulf Coastal Plain, with a discussion of Tethyan
molluscan assemblages in the Gulf Coastal Plain and
Florida. Journal of Paleontology 63(6)4852-856.

Glaessner, M.F. 1969. Decapoda. In Treatise on inverte-
brate paleontology, part R, Arthropoda 4, ed. R. C.
Moore, R400-R533. Lawrence, Kansas: University
of Kansas Press.

Gmelin, J.F. 1791. Systema naturae per regna tria naturae.
Editio decima tertia, aucta, reformata, vol. 1. Leip-
zig, pp. 3021-3910.

Grant, U.S., IV, and H.R. Gale. 1931. Catalogue of the
marine Pliocene and Pleistocene Mollusca of Cali-
fornia. San Diego Society of Natural History, Mem-
oirs 1:1-1036.

Gray, J.E. 1824. An account of the subjects of natural
history, shells. In W.E. Perry Supplement to the ap-
pendix of Captain Perry’s voyage for the discovery
of north-west passage in the years 1819-1820, pp.
181-310. London: John Murray.

Gray, J.E. 1826. On a recent species of the genus Hinnita
of Defrance and some observations on the shells of
the monomyaires of Lamarck. Annals of Philosophy,
new series, 12:103-106.

Gray, J.E. 1847a. An outline of an arrangement of stony
corals. Annals and Magazine of Natural History, se-
ries 1, 19:120-128.

Gray, J.E. 1847b. A list of the genera of Recent Mollusca,
their synonyma and types. Proceedings of the Zoo-
logical Record of London 15:129-219.

Gray, J.E. 1850. Explanations of plates and list of genera.
In Figures of molluscous animals, vol. 4. London: M.
E. Gray, 124 pp.

Gregory, J.W. 1900. The corals. Jurassic fauna of Cutch.
Paleontologia lndica (Calcutta), series 9, 2(pt. 2):1-
195.

Hanna, G.D. 1924. Rectifications of nomenclature. Pro-
ceedings of the California Academy of Sciences, se-
ries 4, 13(10):151-186.

Hanna, G.D., and L.G. Hertlein. 1949. Two new species
of gastropods from the middle Eocene of California.
Journal of Paleontology 23(4):392-394.

Hanna, M.A. 1927. An Eocene invertebrate fauna from
the La Jolla quadrangle, California. University of
California, Publications in Geological Sciences
16(8):247— 398.

Haq, B.U., J. Hardenbol, and P.R. Vail. 1987. Chronology
of fluctuating sea levels since the Triassic. Science
235:1156-1167.

Harris, G.F. 1897. Catalogue of Tertiary Mollusca in the
department of Geology, British Museum (Natural
History). Part 1. The Australasian Tertiary Mollusca.
London: British Museum (Natural History), 407 pp.

Heilprin, A. 1887. Explorations on the west coast of Flor-

ida, and in the Okeechobee wilderness, with special
reference to the geology and zoology of the Floridian
peninsula. Transactions of the Wagner Free Institute
of Science of Philadelphia 1:1-134.

Heitman, H.L. 1983. Paleoecological analysis and biostra-
tigraphy of the lower Paleogene Santa Susana For-
mation, northern Simi Valley, Ventura County, Cal-
ifornia. In Cenozoic geology of the Simi Valley area,
southern California, eds. R.L. Squires and M.V. Fi-
lewicz, 33-44. Los Angeles: Pacific Section, Society
of Economic Paleontologists and Mineralogists, Fall
Field Trip Volume and Guidebook.

Hickman, C. J. S. 1969. The Oligocene marine molluscan
fauna of the Eugene Formation in Oregon. Univer-
sity of Oregon Museum of Natural History, Bulletin
16:1-112, pis. 1-14.

Hickman, C.J.S. 1980. Paleogene marine gastropods of
the Keasey Formation in Oregon. Bulletins of Amer-
ican Paleontology 78:1-112.

Hinds, R.B. 1843. Descriptions of new shells from the col-
lection of Captain Sir Edward Belcher during the
years 1836-1842. Proceedings of the Zoological Re-
cord of London 11:36-49.

ICZN (International Commission of Zoological Nomen-
clature). 1948. Opinion 209. The Bulletin of Zoo-
logical Nomenclature 4(pts. 16-18):447-542.

Kappeler, K.A., R.L. Squires, and A.E. Fritsche. 1984.
Transgressive marginal-marine deposits of the Av-
enal Sandstone, Reef Ridge, central California. In
Kreyenhagen Formation and related rocks, ed. J.P.
Blueford, 9-27. Los Angeles: Pacific Section, Society
of Economic Paleontologists and Mineralogists.

Keen, A.M. 1971. Sea shells of tropical West America.
Second Edition. Stanford: Stanford University Press,
1064 pp.

Kennedy, G.L. 1993. New Cretaceous and Tertiary Pho-
ladidae (Mollusca: Bivalvia) from California. Journal
of Paleontology 67(3):397-404.

Kew, W.S.W. 1920. Cretaceous and Cenozoic Echinoidea
of the Pacific Coast region of North America. Uni-
versity of California Publications, Bulletin of the De-
partment of Geology 12:23-236.

Kew, W.S.W. 1924. Geology and oil resources of a part of
Los Angeles and Ventura Counties, California. Unit-
ed States Geological Survey, Bulletin 753:1-202.

Kilburn, R.N. 1981. Revision of the genus Ancilla La-
marck, 1799 (Mollusca: Olividae: Ancillinae). An-
nals of the Natal Museum 24(2):349-463.

Kleinpell, R.M., and D.W. Weaver. 1963. Oligocene bio-
stratigraphy of the Santa Barbara embayment, Cal-
ifornia. Part 2. Mollusca from the Turritella variata
zone. University of California, Publications in Geo-
logical Sciences 43:81-118.

Koby, F. 1890. Monographic des polypiers jurassiques de
la Suisse. Societe de la Paleontologie Suisse, Memo-
ires 16:457-582.

Kocurko, M.J. 1988. Notes on fossil octocorals and com-
parisons of some modern and ancient octocoral re-
mains. Tulane Studies in Geology and Paleontology
21(3):105-115.

Lamarck, J.B.P. 1799. Prodome d’une nouvelle classifica-
tion des coquilles. Memoires de la Societe d’Histoire
Naturelle de Paris 1:63-91.

Lamarck, J.B.P. 1801. Systeme des animaux san vertebres,
ou tableau general des classes, des ordres, et des
genres de ces animaux. Paris, 432 pp.

Lamarck, J.B.P. 1802. Suite de memoires sur les fossiles
des environs de Paris. Annales du Museum National

34 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

d’Histoire Naturelle, vol. 1, 474-475. (Reprinted
1978 by Paleontological Research Institution, Ithaca,
New York).

Lamarck, J.B.P. 1804a. Suite de memoires sur les fossiles
des environs de Paris. Annales du Museum National
d’Histoire Naturelle, vol. 3, 436-441. (Reprinted
1978 by Paleontological Research Institution, Ithaca,
New York).

Lamarck, J.B.P. 1804b. Suite de memoires sur les fossiles
des environs de Paris. Annales du Museum National
d’Histoire Naturelle, vol. 4, 429-436. (Reprinted
1978 by Paleontological Research Institution, Ithaca,
New York).

Lamarck, J.B.P. 1809. Philosophie zoologique, ou expo-
sition des considerations relatives a I’histoire natu-
relle des animaux. 2 Vols. Paris, 885 pp.

Lamarck, J.B.P. 1815. Sur les polypiers Corticiferes. Me-
moires du Museum National d’Histoire Naturelle,
Paris 1:401-416.

Lamarck, J.B.P. 1818. Histoire naturelle dex animaux sans
vertebres, vol. 5. Paris, 612 pp.

Lambert, J. 1905. Echinides du sud de la Tunisie environs
de Tatahouine. Bulletin de la Societe Geologique de
France, serie 4, 5:569-577.

Lamouroux, J.V.F. 1812. Extrait d’un memoire sur la clas-
sification des polypiers coralligenes non entierement
pierreux. Nouvelle Bulletin Sciences Societe Philom-
ath ematiques, Paris 3(63): 18 1-1 88.

Lamouroux, J.V.F. 1816. Histoire des polypiers corallige-
nes flexibles, vulgairement nommes zoophytes. F.
Caen Poisson, 560 pp.

Latreille, P.A. 1825. Families nautrelles du regne animal,
exposees succintement et dans un order anlytique,
avec Vindication de leurs genres, 2ed. Paris.

Lea, I. 1833. Contributions to geology. Philadelphia: Ca-
rey, Lea, and Blanchard, Til pp.

Leach, W.E. 1819. A list of invertebrate animals discov-
ered by H.M.S. ship Isabella, in a voyage to the arc-
tic regions. In Voyage of discovery. . .exploring Baf-
fin’s Bay, 252 pp. London.

Link, H.F. 1807. Beschreibung der Naturalien-Sammlung
der Universitdt zu Rostok, vol. 1. Variously paged.

Linnaeus, C. 1758. Sy sterna naturae per regna tria natu-
rae, secundum classes, ordines, genera, species, cum
characteribus, differentiis, synonymis, locis. Regnum
animale. Editio decima reformata, vol. 1. Laurentius
Salvius: Holmiae, 824 pp. [Facsimile, British Muse-
um (Natural History), 1956].

M’Coy, F. 1849. On the classification of some British fossil
Crustacea. Annals and Magazine of Natural History,
series 2, 4:161-179.

MacLeay, W.S. 1838. On the brachyurous decapod Crus-
tacea brought from the Cape by Dr. Smith, A. In
Illustrations of the Annulosa of South Africa, 53-72.
London.

Marincovich, L., Jr. 1977. Cenozoic Naticidae (Mollusca:
Gastropoda) of the northeastern Pacific. Bulletins of
American Paleontology 70(294): 1-494.

Meek, F.B. 1876. A report on the invertebrate Cretaceous
and Tertiary fossils of the upper Missouri country.
United States Geological Survey, Territory Annual
Report 9:1-629.

Megerle von Miihlfeld, J.K. 1811. Entwurf eines neuen
System’s der Schalthiergehause. Gesellschaft Natur-
forschender Freunde Magazin, (Berlin) 5:38-72.

Merriam, C.W. 1941. Fossil turritellas from the Pacific
coast region of North America. University of Cali-

Contributions in Science, Number 479

fornia Publications, Bulletin of the Department of
Geological Sciences 26( 1 ): 1— 2 14.

Merriam, C.W., and F.E. Turner. 1937. The Capay middle
Eocene of northern California. University of Cali-
fornia Publications, Bulletin of the Department of
Geological Sciences 24(6):91-114.

Milne-Edwards, H. and J. Haime. 1848. Recherches sur
les polypiers. Quatrieme memoire. Monographic des
Astreides. Annales des Sciences Naturelles, Paris, ser-
ie 3, 10:209-320.

Milne-Edwards, H., and J. Haime. 1850. A monograph
of the British fossil corals. Part 1 . Introduction; cor-
als from the Tertiary and Cretaceous formations.
London: Palaeontographical Society, 71 pp.

Montagu, G. 1803-1808. Testacea Britannica, or natural
history of British shells, marine, land and freshwater.
2 Vols. London, 789 pp.

Monterosato, T.A. 1884. Nomenclatura generica e speci-
fica di alcune conchiglie Mediterranee pel Marchese
di Monterosato. Palermo, 152 pp.

Montfort, P.D. 1810. Conchyliologie systematique et clas-
sification methodique des coquilles, vol. 2. Paris: F.
Schoell, 176 pp.

Moore, E.J. 1968. Fossil mollusks of San Diego County.
San Diego Society of Natural History, Occasional
Paper 15:1-76.

Moore, E.J. 1983. Tertiary marine pelecypods of Califor-
nia and Baja California: Nuculidae through Mallei-
dae. United States Geological Survey, Professional
Paper 1228-A:A1-A108.

Moore, E.J. 1987. Tertiary marine pelecypods of Califor-
nia and Baja California: Plicatulidae to Ostreidae.
United States Geological Survey, Professional Paper
1228-C:C1-C53.

Moore, E.J. 1992. Tertiary marine pelecypods of Califor-
nia and Baja California: Erycinidae through Cardi-
tidae. United States Geological Survey, Professional
Paper 1228-E:E1-E37.

Morch, O.A.L. 1852-1853. Catalogus conchyliorum quae
reliquit D. Alphonso d’ Aguirra et Gadea Comes de
Yoldi. 8 Vols. Hafniae.

Nelson, R.N. 1925. A contribution to the paleontology of
the Martinez Eocene of California. University of
California, Publications in Geological Sciences
15(11 ):397 — 466.

Nicol, D. 1950. Recent species of the lucinoid pelecypod
Fimbria. Journal of the Washington Academy of Sci-
ences 40(3):82— 87.

Orbigny, A.D. 1834-1847. Voyage dan I’Amerique Meri-
dional, Mollusques, vol. 5, pt. 3. Paris, 758 pp.

Palmer, K.V.W. 1937. The Claibornian Scaphopoda, Gas-
tropoda, and dibranchiate Cephalopoda of the
southeastern United States. Bulletins of American
Paleontology 7(32), pts. 1 and 2, 730 pp.

Parker, J.D. 1983. Lower Paleocene to lower Eocene, non-
marine to deep-marine strata of the Simi Hills, Ven-
tura County, California. In Cenozoic geology of the
Simi Valley area, southern California, eds. R.L.
Squires and M. V. Filewicz, 3-22. Los Angeles: Pa-
cific Section, Society of Economic Paleontologists
and Mineralogists, Fall Field Trip Volume and
Guidebook.

Philippi, R.A. 1853. Handbuch de Conchyliologie und
Malakozoologie. Halle, 547 pp.

Ponder, W.E, and D.R. Lindberg. 1996. Gastropod phy-
logeny — challenges for the 90s. In Origin and evo-
lutionary radiation of the Mollusca, ed. J. Taylor,
135-154. Oxford: Oxford University Press.

Squires: “Meganos Stage” Marine Megafossils ■ 35

Ponder, W.F., and D.R. Lindberg. 1997. Towards a phy-
logeny of gastropod molluscs: an analysis using mor-
phological characters. Geological Journal of the Lin-
nean Society 119:83-265.

Rafinesque, C.S. 1815. Analyse de la nature, ou tableau
de Vunivers et des corps organises. Palermo, 224 pp.

Rathbun, M.J. 1926. The fossil stalk-eyed Crustacea of
the Pacific slope of North America. United States
National Museum, Bulletin 138:1-155.

Reeve, L.A. 1843. Conchologica Iconica I, Pleurotoma,
vol. 1, 46 pp.

Reeve, L.A. 1848. Conchologica Iconica, vol. 5, 50 pp.

Roding, P.F. 1798. Museum Boltenianum sive catalogus
cimeliorum e tribus regnis naturae quae olim colle-
ger at. Hamburg: Johan Christi Trappii, 199 pp.

Romer, E. 1857. Kritische Untersuchung der Arten des
Molluskengeschechts Venus bei Linne und Gmelin
mit Beriicksichtigung der spater Beschriebenen Ar-
ten. Cassel: J.G. Luckhardt, 136 pp.

Sacco, F. 1895. I molluschi dei terreni Terziarii del Pie-
monte e della Liguria. Bollettino dei Musei di Zool-
ogia ed Anatomia Comparata della R. Universita di
Torino, Pt. 17 (Cerithiidae, Triforidae, Cerithiopsi-
dae e Diastomidae):86 pp.

Sacco, F. 1899. I molluschi dei terreni Terziarii del Pie-
monte e della Liguria. Bollettino dei Musei di Zool-
ogia ed Anatomia Comparata della R. Universita di
Torino, Pt. 27 (Unionidae through Chamidae):102

pp.

Saul, L.R. 1983. Notes on Paleogene turritellas, veneri-
cardias, and molluscan stages of the Simi Valley area,
California. In Cenozoic geology of the Simi Valley
area, southern California, eds. R.L. Squires and M.V.
Filewicz, 71-80. Los Angeles: Pacific Section, Society
of Economic Paleontologists and Mineralogists, Vol-
ume and Guidebook.

Saul, L.R., and R.L. Squires. 1997. New species of neritid
gastropods from Cretaceous and lower Cenozoic
strata of the Pacific slope of North America. The
Veliger 40(2):131-147.

Schenck, H.G. 1936. Nuculid bivalves of the genus Acila.
Geological Society of America, Special Papers 4:1-
149.

Schenck, H.G., and A.M. Keen. 1940. California fossils
for the field geologist. Stanford California: Stanford
University, 86 pp.

Schmidt, F.C. 1818. Versuch iiber die beste Einrichtung
zur Aufstellung, Behandlung und Aufbewahrung der
verschieden Naturkorper und Gegenstande der
Kunst. Gotha, 252 pp.

Schumacher, C.F. 1817. Essai d’un nouveau systeme des
habitations des ver testaces. Copenhagen, 287 pp.

Smith, J.T. 1975. Age, correlation, and possible Tethyan
affinities of mollusks from the Lodo Formation of
Fresno County, California. In Future energy horizons
of the Pacific Coast; Paleogene symposium and se-
lected technical papers, eds. D.W. Weaver, G.R. Hor-
naday, and A. Tipton, 464-483. Los Angeles: Pacific
Sections, American Association of Petroleum Geol-
ogists, Society of Economic Paleontologists and Min-
eralogists, and Society of Economic Geologists, An-
nual Meeting.

Sohl, N.F. 1964. Neogastropoda, Opisthobranchia and
Basommatophora from the Ripley, Owl Creek, and
Prairie Bluff Formations. Late Cretaceous gastro-
pods in Tennessee and Mississippi. U.S. Geological
Survey, Professional Paper 33 1-B: 1-344.

Sowerby, G.B., II. 1849. Monograph of the genus Pholas.

Monograph of the genus Triomphalia. Monograph
of the genus Xylophaga. Thesaurus Conchyliorum,
or figures and descriptions of Recent shells 2(10):
485-505.

Squires, R.L. 1980. A new species of brachyuran from the
Paleocene of California. Journal of Paleontology
54(2):472-476.

Squires, R.L. 1981. A transitional alluvial to marine se-
quence: the Eocene Llajas Formation, southern Cal-
ifornia. Journal of Sedimentary Petrology 51(3):
923-938.

Squires, R.L. 1983a. Eocene Llajas Formation, Simi Val-
ley, southern California. In Cenozoic geology of the
Simi Valley area, southern California, eds. R.L.
Squires and M.V. Filewicz, 81-96. Los Angeles: Pa-
cific Section, Society of Economic Paleontologists
and Mineralogists, Fall Field Trip Volume and
Guidebook.

Squires, R.L. 1983b. Geologic map of the Simi Valley
area, southern California. In Cenozoic geology of the
Simi Valley area, southern California, eds. R.L.
Squires and M.V. Filewicz, insert. Los Angeles: Pa-
cific Section, Society of Economic Paleontologists
and Mineralogists, Fall Field Trip Volume and
Guidebook.

Squires, R.L. 1984. Megapaleontology of the Eocene Lla-
jas Formation, Simi Valley, California. Natural His-
tory Museum of Eos Angeles County, Contributions
in Science 350:1-76.

Squires, R.L. 1987. Eocene molluscan paleontology of the
Whitaker Peak area, Los Angeles and Ventura Coun-
ties, California. Natural History Museum of Los An-
geles County, Contributions in Science 388:1-93.

Squires, R.L. 1988a. Geologic age refinements of west
coast Eocene marine mollusks. In Paleogene stratig-
raphy, west coast of North America, eds. M.V. Fi-
lewicz and R.L. Squires, vol. 58, 107-112, pis. 1-2.
Los Angeles: Pacific Section, Society of Economic Pa-
leontologists and Mineralogists, West Coast Paleo-
gene Symposium.

Squires, R.L. 1988b. Eocene macropaleontology of north-
ern Lockwood Valley, Ventura County, California.
Natural History Museum of Los Angeles County,
Contributions in Science 398:1-23.

Squires, R.L. 1990. New Paleogene Fimbria (Mollusca:
Bivalvia) from the Pacific coast of southwestern
North America. Journal of Paleontology 64(4):552-
556.

Squires, R.L. 1991a. New early Eocene species of Area
s.s. (Mollusca: Bivalvia) from southern California.
The Veliger 34(l):67-72.

Squires, R.L. 1991b. Molluscan paleontology of the lower
Eocene Maniobra Formation, Orocopia Mountains,
southern California. In Eocene geologic history San
Diego region, ed. P.L. Abbott, vol. 68, 217-226. Los
Angeles: Pacific Section, Society of Economic Pale-
ontologists and Mineralogist.

Squires, R.L. 1993. New reports of the large gastropod
Campanile from the Paleocene and Eocene of the
Pacific coast of North America. The Veliger 36(4):
323-331.

Squires, R.L. 1994. Macropaleontology of Eocene marine
rocks, upper Sespe Creek area, Ventura County,
southern California. In Sedimentology and paleon-
tology of Eocene rocks in the Sespe Creek area, Ven-
tura County, California, ed. A.E. Fritsche, Book 74,
39-56. Los Angeles: Pacific Section, SEPM (Society
for Sedimentary Geology).

36 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

Squires, R.L. 1997. Taxonomy and distribution of the
buccinid gastropod Bracbysphingus from uppermost
Cretaceous and lower Cenozoic marine strata of the
Pacific slope of North America. Journal of Paleon-
tology 71(5):847-861.

Squires, R.L. 1998. New information on morphology,
stratigraphy, and paleoclimatic implications of the
Eocene brackish-marine gastropod Loxotrema tur-
ritum Gabb, 1868, from the west coast of the United
States. The Veliger 41(4):297-313.

Squires, R.L., and D.A. Advocate. 1986. New early Eo-
cene mollusks from the Orocopia Mountains, south-
ern California. Journal of Paleontology 60(4):851-
864.

Squires, R.L., and R.A. Demetrion. 1990. New Eocene
marine bivalves from Baja California Sur, Mexico.
Journal of Paleontology 64(3):382-391.

Squires, R.L., and R.A. Demetrion. 1992. Paleontology of
the Eocene Bateque Formation, Baja California Sur,
Mexico. Natural History Museum of Los Angeles
County, Contributions in Science 434:1-55.

Squires, R.L., and R.A. Demetrion. 1994. New reports of
Eocene mollusks from the Bateque Formation, Baja
California Sur, Mexico. The Veliger 37(2): 125— 1 35.

Squires, R.L., and J.L. Goedert. 1994. Macropaleontology
of the Eocene Crescent Formation in the Little River
area, southern Olympic Peninsula, Washington. Nat-
ural History Museum of Los Angeles County, Con-
tributions in Science 444:1-32.

Squires, R.L., and J.L. Goedert. 1997. Eocene megafossils
from the Needles-Gray Wolf lithic assemblage of the
eastern “core rocks,” Olympic Peninsula, Washing-
ton. Washington Geology 25(4):25-29.

Squires, R.L., J.L. Goedert, and K.L. Kaler. 1992. Pale-
ontology and stratigraphy of Eocene rocks at Pulali
Point, Jefferson County, eastern Olympic Peninsula,
Washington. Washington Division of Geology and
Earth Resources, Report of Investigations 31, 27 pp.

Stanton, T.W. 1896. The faunal relations of the Eocene
and Upper Cretaceous on the Pacific Coast. United
States Geological Survey, Annual Report 17:1005-
1060.

Stewart, R.B. 1927. Gabb’s California fossil type gastro-
pods. Proceedings of the Academy of Natural Sci-
ences of Philadelphia 78:287-447.

Stewart, R.B. 1930. Gabb’s California Cretaceous and
Tertiary type lamellibranchs. Academy of Natural
Sciences of Philadelphia, Special Publication 3:1-
314.

Stewart, R.B. 1946. Geology of Reef Ridge, Coalinga dis-
trict, California. United States Geological Survey,
Professional Paper 205-C:81-115.

Stoliczka, F. 1870-1871. Cretaceous fauna of southern In-
dia. The Pelecypoda, with a review of all known gen-
era of this class, fossil and Recent. Geological Survey
of India, Palaeontologia Indica, series 6, 3:1-535.

Sutherland, J.A. 1966. A new species of Architectonica
from the Santa Susana Mountains, Ventura County,
California. Natural History Museum of Los Angeles
County Contributions in Science 117:1-4.

Swainson, W. 1820-1833. Zoological illustrations, or
original figures and descriptions of new, rare, or in-
teresting animals. 2 Series, 3 Vols. London, 136 pp.

Swainson, W. 1840. A treatise on malacology; or the nat-
ural classification of shells and shell-fish. London:
Longman et al., 419 pp.

Throckmorton, C.K. 1988. Depositional environments
and molluscan biostratigraphy of the Tesla Forma-

Contributions in Science, Number 479

tion, central California. In Paleogene stratigraphy,
west coast of North America, ed. M.V. Filewicz and

R. L. Squires, vol. 58, 209-223. Los Angeles: Pacific
Section, Society of Economic Paleontologists and
Mineralogists, West Coast Paleogene Symposium.

Turner, F.E. 1938. Stratigraphy and Mollusca of the Eo-
cene of western Oregon. Geological Society of Amer-
ica Special Papers 10:1-130.

Vaughan, T.W. 1900. Eocene and lower Oligocene coral
faunas of the United States with a few doubtfully
Cretaceous species. United States Geological Survey,
Monograph 39:1-263.

Vaughan, T.W. 1919. Fossil corals from Central America,
Cuba, and Porto Rico, with an account of the Amer-
ican Tertiary, Pleistocene, and Recent coral reefs. U.

S. National Museum, Bulletin 103:189-524.

Vaughan, T.W. 1927. A new species of Eocene coral, Ar-

chohelia clarki, from California. University of Cali-
fornia Publications, Bulletin of the Department of
Geological Sciences 17:143-144.

Vaughan, T.W. 1932. Antillophyllia, a new coral generic
name. Journal of the Washington Academy of Sci-
ences 22:506-510.

Vaughan, T.W., and J.E. Hoffmeister. 1926. Miocene cor-
als from Trinidad. Carnegie Institution of Washing-
ton, Publication 344:105-134.

Vaughan, T.W., and J.W. Wells. 1943. Revision of the sub-
orders, families, and genera of the Scleractinia. Geo-
logical Society of America, Special Papers 44:1-363.

Verastegui, P. 1953. The pelecypod genus Venericardia in
the Paleocene and Eocene of western North America.
Palaeontographica Americana 3 (25): 1-1 13.

Vidal, L.M. 1917. Nota paleontologica sobre el Cretaceo
de Cataluna. Asociacion Espahola progreso de las
ciencias, Congreso de Sevilla. Barcelona. Tomo V
(con varias lamlinas), 19 pp.

Vokes, H.E. 1935. The genus Velates in the Eocene of
California. University of California, Publications,
Bulletin of the Department of Geological Sciences
23( 12):38 1—390.

Vokes, H.E. 1939. Molluscan faunas of the Domengine
and Arroyo Hondo formations of the California Eo-
cene. Annals of the New York Academy of Sciences
38:1-246.

Vokes, H.E. 1980. Genera of the Bivalvia: a systematic
and bibliographic catalogue (revised and updated).
Ithaca: Paleontological Research Institution, 307 pp.

Vyalov, O.S. 1936. Sur la classification des huitres. Aca-
demie des Sciences URSS, Comptes-Rendus (Dokla-
dy), new series, 4:17-20.

Waring, C.A. 1917. Stratigraphic and faunal relations of
the Martinez to the Chico and Tejon of southern
California. Proceedings of the California Academy
of Sciences, series 4, 7(4):41— 125.

Weaver, C.E. 1943. Paleontology of the marine Tertiary
formations of Oregon and Washington. University of
Washington, Publications in Geology 5(1— 3):1— 789.

Weaver, C.E. 1953. Eocene and Paleocene deposits at
Martinez, California. University of Washington Pub-
lications in Geology 7:1-102.

Weaver, C. E., and K. V. W. Palmer. 1922. Fauna from the
Eocene of Washington. University of Washington,
Publications in Geology 1 (3): 1—56.

Weinkauff, H.C. 1875. Uber eine kritische Gruppe des ge-
nus Pleurotoma Lam. sensu stricto. Jahrbuch der
Deutschen Malakozoologie Geschichte 2:285-292.

Weisbord, N.E. 1971. Corals from the Chipola and Jack-
son Bluff formations of Florida. State of Florida, De-

Squires: “Meganos Stage” Marine Megafossils ■ 37

partment of Natural Resources, Geological Bulletin
53:1-100.

Wells, J.W. 1934. Eocene corals from Cuba. Bulletins of
American Paleontology 20(70B, pt. 1 ): 147— 1 58.

Wells, J.W. 1945. Part 2- West Indian Eocene and Miocene
corals. Geological Society of America, Memoir 9:1-
25.

Wells, J.W. 1956. Scleractinia. In Treatise on invertebrate
paleontology, part F, Coelenterata, ed. R.C. Moore,
F328-F444. Lawrence, Kansas: University of Kansas
Press.

Wenz, W. 1943. Subfamilia Olivinae. In Handbuch de Pa-
laozoologie, ed. O.H. Schindewolf, 1272-. Berlin:
Gebriider Borntraeger.

Woodring, W.P. 1925. Miocene Mollusca from Bowden,
Jamaica, pelecypods and scaphopods. Carnegie In-
stitution of Washington, Publication 366:1-564.

Woodring, W.P. 1928. Miocene mollusks from Bowden,
Jamaica. Part 2, Gastropods and discussion of re-
sults. Carnegie Institution of Washington, Publica-
tion 385:1-564.

Woods, A.J.C., and L.R. Saul. 1986. New Neritidae from
southwestern North America. Journal of Paleontol-
ogy 60(3):636-655.

Woodward, S.P. 1851. A manual of the Mollusca: Or ru-
dimentary treatise of Recent and fossil shells. Lon-
don: John Weale, 330 pp.

Zinsmeister, W.J. 1983a. Late Paleocene (“Martinez pro-
vincial stage”) molluscan fauna from the Simi Hills,
Ventura County, California. In Cenozoic geology of
the Simi Valley area, southern California, eds. R.L.
Squires and M.V. Filewicz, 61-70. Los Angeles: Pa-
cific Section, Society of Economic Paleontologists
and Mineralogists, Fall Field Trip Volume and
Guidebook.

Zinsmeister, W.J. 1983b. New late Paleocene molluscs
from the Simi Hills, Ventura County, California.
Journal of Paleontology 57(6):1282-1303.

Received 23 April 1998; accepted 11 August 1999.

Natural History Museum

of Los Angeles County

900 Exposition Boulevard
Los Angeles, California 90007

38 ■ Contributions in Science, Number 479

Squires: “Meganos Stage” Marine Megafossils

Number 480
17 December 1999

Contributions
in Science

Middle Eocene Brackish-Marine
Mollusks from the Matilija Sandstone
at Matilija Hot Springs, Ventura
County, Southern California

Richard L. Squires

of Los Angeles County

Natural History Museum

Serial
Publications
of THE
Natural History
Museum of
Los Angeles
County

Scientific

Publications

Committee

John Heyning, Acting Deputy Director
for Research and Collections
John M. Harris, Committee Chairman
Brian V. Brown
Kenneth E. Campbell
Kirk Fitzhugh
Karen Wise

Robin A. Simpson and K. Victoria Brown,

Managing Editors

The scientific publications of the Natural History Museum
of Los Angeles County have been issued at irregular in-
tervals in three major series; the issues in each series are
numbered individually, and numbers run consecutively, re-
gardless of the subject matter.

# Contributions in Science, a miscellaneous series of tech-
nical papers describing original research in the life and
earth sciences.

# Science Bulletin, a miscellaneous series of monographs
describing original research in the life and earth sci-
ences. This series was discontinued in 1978 with the
issue of Numbers 29 and 30; monographs are now
published by the Museum in Contributions in Science.

# Science Series, long articles and collections of papers on
natural history topics.

Copies of the publications in these series are sold through
the Museum Book Shop. A catalog is available on request.

The Museum also publishes Technical Reports, a miscel-
laneous series containing information relative to scholarly
inquiry and collections but not reporting the results of
original research. Issue is authorized by the Museum’s Sci-
entific Publications Committee; however, manuscripts do
not receive anonymous peer review. Individual Technical
Reports may be obtained from the relevant Section of the
Museum.

Natural History Museum
of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

Printed at Allen Press, Inc., Lawrence, Kansas
ISSN 0459-8113

Middle Eocene Brackish-Marine Mollusks

FROM THE MATILIJA SANDSTONE AT MATILIJA

Hot Springs, Ventura County,
Southern California

Richard L. Squires1

ABSTRACT. This study is the first detailed account of within-habitat, brackish-marine Eocene mollusks
in the Transverse Ranges of southern California. The fossils are from the lower middle Eocene (“Transition
Stage”) upper part of the Matilija Sandstone at Matilija Hot Springs, near Ojai, Ventura County, southern
California. Sixteen species (eight gastropods and eight bivalves) were found, and three of these are new: a
gastropod Tympanotonos (T.) californicus new species, and two bivalves Neotrapezium californicum new
species and Corbicula jestesi new species. This is the first confirmed record of Tympanotonos in North
America and the first record of Neotrapezium in North America. The megafauna contains the earliest
known record of the gastropod “ Melanatria ” markleyensis and the latest known records of the bivalves
Barbatia (B.) morsel, uTeilina" joaquinensis, and “ Tellina ” domenginensis.

The local brackish-marine section, which is approximately 55 m thick, was deposited on the upper part
of a deltaic complex and consists of lagoonal mudstones and siltstones alternating with beach or barrier-
bar sandstones. Within the lagoonal rocks are interbeds of coastal-sabkha limestone and gypsum, as well
as subaerial? redbeds. Mollusks are abundant within the lagoonal rocks and represent parautochthonous
assemblages that have undergone varying amounts of postmortem transport but were not moved out of
their original lagoonal habitat of mud and silt. Other megafossils are rare. Some of the molluscan assem-
blages consist of up to 13 species of mollusks. All of the shells are unabraded, and many are complete.
Other assemblages consist entirely of concentrations of either the bivalve Pelecyora aequilateralis or the
bivalve Guneocorhula torreyensis. Both types of concentrations consist of tightly packed, unabraded single
valves. Within some of the beach and barrier-bar sandstones are fragments of the oyster Acutostrea id-
riaensis idriaensis, which were transported out of their muddy lagoonal habitat.

INTRODUCTION

Brackish-marine rocks are uncommon in the rock
record because they are highly susceptible to ero-
sion. A local, 55-ni' thick section of middle Eocene
(“Transition Stage”) brackish-marine rocks in the
upper part of the Matilija Sandstone at Matilija
Hot Springs, located 6.5 km northwest of the city
of Ojai in Ventura County, southern California
(Fig. 1A), was preserved because it underwent sub-
sidence and was overlain by a protective cover of
deeper marine deposits. Kerr and Schenck (1928)
were the first to recognize the brackish-marine as-
pect of these rocks. They reported a “lignitic facies”
with abundant specimens of mollusks near Matilija
Hot Springs, but they believed that the facies was
confined to a single bed. Jestes (1963) was the first
to more fully recognize the extent of this brackish-
marine paleoenvironment, which he documented
by means of a preliminary study of the fossil mol-
lusks. Link (1975) and Link and Welton (1982) did

1. Department of Geological Sciences, California State
University, North ridge, California 91330-8266, and Re-
search Associate in Invertebrate Paleontology, Natural
History Museum of Los Angeles County, Los Angeles,
California 90007-4000.

Contributions in Science, Number 480, pp. 1-30
Natural History Museum of Los Angeles County, 1999

sedimentological studies that confirmed Jestes’ in-
terpretation. They utilized Jestes’ preliminary mol-
luscan studies but did not elaborate on them. Mol-
luscan fossils are abundant and, in many cases, well
preserved in these rocks, but until this present
study, they were not analyzed in detail. Squires
(1991a, 1998) worked on two of the gastropod spe-
cies from these rocks, namely Potamides (Potami-
dopsis) calif ornica Squires, 1991a, and Loxotrema
turritum Gabb, 1868. In recent years, I have be-
come increasingly interested in brackish-marine Eo-
cene rocks, and my students and I have returned
on many occasions to the Matilija Hot Springs sec-
tion to undertake more detailed studies. The goal
of this present article is to fully document, for the
first time, ail the mollusks and to give the details of
their stratigraphic distribution in the study area.

The molluscan stage terminology used here stems
from Clark and Vokes (1936), who proposed five
moll usk- based provincial Eocene stages: “Mega-
nos” (uppermost Paleocene to lowermost Eocene),
“Capay” (middle lower Eocene), “Domengine” (up-
per lower to lower middle Eocene), “Transition”
(lower middle Eocene), and “Tejon” (middle middle
Eocene to upper Eocene). Givens (1974) modified
the use of the “Capay Stage.” The stage names are

in quotes because they are informal terms. Squires
(in press) correlated all the stages, except the upper
part of the “Tejon Stage,” to the standard calcare-
ous nannofossil zonation.

In this present article, the term “brackish marine”
refers to restricted waters with salinities lower than
those of normal ocean waters. Furthermore, the
term “brackish marine” refers to waters landward
of beaches or barrier bars but with some connec-
tion to the shallow-marine environment. The term
“shallow marine” refers to unrestricted, nearshore
waters of normal ocean salinity seaward of beaches
or barrrier bars.

The following institutional acronyms are used:

ANSP Academy of Natural Sciences,

Philadelphia

CSUN California State University,

Department of Geological Sci-
ences, North ridge

LACM Natural History Museum of

Los Angeles County, Section
of Malacology

LACMIP Natural History Museum of

Los Angeles County, Section
of Invertebrate Paleontology
MCZ Museum of Comparative Zo-

ology, Harvard University
UCMP University of California Mu-

seum of Paleontology, Berke-
ley

UCR University of California, Riv-

erside

USGS United States Geological Sur-

vey, Reston, Virginia

STRATIGRAPHY

The study area rocks are in the upper part of the
Matilija Sandstone and crop out in a small area
bounded on the south by a side road leading to the
Matilija Hot Springs and on the north by the river
bed of the North Fork Matilija Creek, a distance
of 270 m (Fig. IB). This general area is also the
type section of the Matilija Sandstone, named by
Kerr and Schenck (1928). The study area rocks
consist of resistant beds of sandstone, 1 to 5 m
thick, alternating with nonresistant, finer grained
intervals, about 2 to 7 m thick. These alternating
rock types will be referred to as the “restricted-
coastal facies.” The finer grained intervals consist
of complexly interbedded mudstone, fossiliferous
mudstone, siltstone, claystone, limestone, and gyp-
sum, as well as stringers of sandstone. Although
Link and Welton (1982: fig. 3, section 5) provided
a columnar section of the restricted-coastal facies,
they did not indicate which beds contain fossils.
This present study revealed, for the first time, the
stratigraphic distribution of these fossils (Figs. 2, 3).

The most accessible and best exposed section of
the restricted-coastal facies is in a roadcut along the
north side of the side road leading to Matilija Hot

Springs (Fig. IB). The roadcut exposes a continu-
ous section of nearly vertical beds, and this is where
Link and Welton (1982) and Jestes (1963) focused
their studies. This is also where I measured my
main section (Fig. 2). I did a microstratigraphic
study and recorded every change in lithology and
every place where fossils were found. The side of
the roadcut is steep, and access to some of the beds
is extremely limited.

An auxiliary section was measured along the
south side of the riverbed of the North Fork of Ma-
tilija Creek (Figs. IB, 3). The lower part of the sec-
tion is accessible, but the middle part is along the
cut-bank side of the creek. Access is difficult when
there is considerable water flow in the creek. The
upper part of the section is covered. Very steep
slopes and dense brush prevent “walking out” of
individual beds between the main and the auxiliary
measured sections. Although individual beds could
not be correlated between the two sections, six
stratigraphic units are recognizable at both sec-
tions. These units, which are denoted on Figures 2
and 3 as units 1 through 6, consist of alternating
sandstone and finer grained units that are similar
in terms of lithology and overall fossil content. The
finer grained units thin toward the auxiliary sec-
tion. Additional complex stratigraphic units (7
through 10) are present at the main section, but at
the auxiliary section only a single, very thick sand-
stone and an overlying thick covered interval cor-
respond to units 7 through 10. Unit 7 at the main
section is an interval of “red-beds” consisting of un-
fossiliferous sandstone, siltstone, and claystone
beds. In addition to having some red color, rocks
in this interval show much variation in color, with
gray, maroon, bluish gray, and greenish gray also
present.

A few meters of fossiliferous mudstone and in-
terbedded sandstone were temporarily uncovered
by recent bulldozing activity in an active rock quar-
ry on the north side of the North Fork of Matilija
Creek (Fig. IB), but private property and safety re-
strictions make this area inaccessible. Extensive
slope wash and dense brush prevent the detection
of any more exposures of the study area rocks im-
mediately south of the roadcut along the road to
Matilija Hot Springs and immediately north of the
rock quarry.

The sandstones in the restricted-coastal facies
usually are fine grained, tabular units that are near-
ly structureless. Some of them are horizontally lam-
inated, and a few have horizontal burrows near
their lower contact. Most of the sandstones are not
fossiliferous (Figs. 2, 3). The only fossils found in
them are scarce fragments of the oyster Acutostrea
idriaensis idriaensis (Gabb, 1869). The most abun-
dant oyster fragments are in a 10-cm-thick limy
sandstone interbed near the base of a 4-m-thick
sandstone (unit 3) in the main measured section. In
this interbed, the oysters form a coquina consisting
of fragments of single valves, which are thick
shelled and concave down.

Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks ■ 3

UNIT

t additional Matilija Sandstone
(shallow marine)

v VVVVV".

1446

=Trrp

• 1445

xC

• 1444

g xP

g

WA

xP

• 1453
xC
XP

y.-

5m

— 0

additional Matilija Sandstone
(shallow marine)

limestone (black aphanitic)

claystone

siltstone

silty mudstone to muddy siltstone

sandstone (very fine to fine grained)

gypsum

mudcracks

horizontal burrows

horizontal laminae

Acutostrea idriaensiss.s. coquina

CSUN fossil locality

Cuneocorbula torreyensis coquina

Pelecyora aequilateralis coquina

4 ■ Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

covered

additional Matiiija Sandstone
(shallow marine)

5m

0

Pt) limestone (black aphanitic)

silty mudstone to muddy siltstone

• 1452

xC

• 1451

• 1450

xP

sandstone alternating with thin beds
of siltstone

sandstone (very fine to fine grained)
g gypsum

Acutostrea idriaensis s.s fragments
• 1 450 CS U N fossi I local ity

x C Cuneocorbula torreyensis coquina

x P Peiecyora aequilateralis coquina

1

additional Matiiija Sandstone
(shallow marine)

Figure 3. Stratigraphy and CSUN megafossil localities of the restricted-coastal facies in the upper part of the Matiiija
Sandstone along the auxilary measured section in the Matiiija Hot Springs area. See Figure 1 for location of this measured
section.

The finer grained intervals in the restricted-coast-
al facies show much complexity and variation in
the vertical sequence of rock types. It is not unusual
to have 10 to 12 changes in lithology in just one
meter of vertical section, especially where limestone

and gypsum are present. The best examples are in
units 2 (upper half), 4, 6 (lower half), 8, and 10. In
these units, aphanitic limestone and gypsum are
complexly interbedded with mudstone, gypsiferous
mudstone, limy sandstone, thin sandstone, and, in

Figure 2. Stratigraphy and CSUN megafossil localities of the restricted-coastal facies in the upper part of the Matiiija
Sandstone along the main measured section in the Matiiija Hot Springs area. See Figure 1 for location of this measured
section.

Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks ■ 5

Table 1. Checklist and abundance of brackish-marine mollusks from CSUN localities in the restricted-coastal facies,
upper part of the Matilija Sandstone at Matilija Hot Springs. Localities listed from left to right in ascending stratigraphic
order. Localities 1450 and 1451 are 270 m (885 ft.) to the north of the other localities. A >30 specimens, C = 10-29
specimens, UC = 5-9 specimens, R = 1-4 specimens, — = not found.

Localities and abundance

Taxa

1444

1450

1451

1452

1453

1445

1446

Gastropoda

Crepidula inornata

R

C

—

—

—

C

R

Crommium sp. cf. C. andersoni

UC

R

—

—

—

—

—

Loxotrema turritum

C

A

K

R

—

C

R

“ Melanatria ” markleyensis

—

—

—

—

—

A

UC

Neverita (Neverita) globosa

R

R

R

—

—

—

—

Potamides ( Potamidopsis ) californica

A

A

C

—

C

UC

C

Pyrgulfera (P.) lajollaensis

—

UC

—

R

R

UC

R

Typanotonos (T.) calif ornicus new species

C

A

C

—

A

—

—

Bivalvia

Acutostrea idriaensis idriaensis

C

C

c

R

—

c

UC

Barbatia ( Barbatia ) morsei

R

C

R

—

—

— ■

—

Corbicula jestesi new species

—

—

—

R

—

A

-

Cuneocorbula torreyensis

—

—

—

A

—

A

—

Neotrapezium californicum new species

UC

A

R

—

UC

UC

R

Pelecyora aequilateralis

A

A

R

—

UC

C

A

“Tellina” domenginesis

R

C

—

—

R

R

“ Tellina ” joaquinensis

UC

A

UC

UC

c

A

UC

some cases, fossiliferous mudstone. Fossils in the
finer grained intervals are confined to mudstone
and siltstone (Figs. 2, 3), but not every mudstone
or siltstone contains fossils. Fossils in the finer
grained intervals are found in either muddy, thin
coquina beds or in less densely packed, muddy fos-
siliferous beds that immediately overlie most of the
coquina beds. The latter contain huge numbers of
bivalve specimens of either Pelecyora aequilateralis
(Gabb, 1869) or Cuneocorbula torreyensis (Flanna,
1927), and the specimens are always stacked one
upon the other and tightly packed. Although many
of these specimens have been crushed by compac-
tion, they are otherwise complete and unworn sin-
gle valves, and, in these respects, differ from usual
coquinas found in the rock record. In the less
densely packed fossiliferous beds associated with
nearly all of these coquina beds, fossils are abun-
dant but more widely spaced. There is much more
diversity, with up to 13 species present, and artic-
ulated bivalves are common, as well as nearly com-
plete growth series of unworn bivalves and gastro-
pods. Seven of these fossiliferous beds were found
in the restricted-coastal facies and represent the lo-
calities shown on Figures IB, 2, and 3 as CSUN
localities 1444, 1445, 1446, 1450, 1451, 1452, and
1453. Each locality overlies a coquina bed, except
locality 1451, which overlies a limestone bed.
There is similarity in the taxonomic composition of
the fossils at these localities, but the abundance of
the species varies greatly from locality to locality
(Table 1).

At CSUN locality 1453, the relationship between

the underlying coquina bed and the overlying fossil
bed is especially clear-cut. The coquina bed consists
of tightly packed and crushed specimens of Pele-
cyora aequilateralis in a sandy mudstone, and the
coquina bed has a sharp contact with an overlying
10-cm- thick black mudstone containing abundant
specimens of the gastropod Tympanotonos (T.) cal-
ifornicus new species These gastropods, which pos-
sess sharp spines, are complete, and specimens
range from just a few millimeters to 23 mm in
length. There are small patches consisting of dense
concentrations of only juveniles, some of which
show a slight preferred orientation.

A coquina consisting entirely of unworn and un-
broken single valves of Cuneocorbula torreyensis
directly underlies CSUN locality 1445. Some spec-
imens are concave up, some are concave down, and
a few are vertical. This coquina bed is distinctive
because the specimens of the bivalves form a “shell
pavement” along the bedding planes. In the imme-
diately overlying fossiliferous silty mudstone at
CSUN locality 1445, there are other mollusks in
addition to abundant specimens of C. torreyensis.
Most notable is the bivalve Corbicula jestesi new
species Some shells of this species are articulated,
and others are “butterflied,” (i.e., the opposing
valves are open and lying adjacent to each other on
the bedding plane). Many of the specimens of Pe-
lecyora aequilateralis, “ Tellina ” joaquinensis Ar-
nold, 1909, and Neotrapezium californicum new
species in the fossiliferous mudstone are articulated
individuals. There are also growth series of P. ae-
quilateralis and “T.” joaquinensis.

6 ■ Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

At CSUN locality 1450, a Pelecyora aequilater-
alis coquina directly underlies a fossiliferous mud-
stone in which many specimens of “ Tellina ” joa-
quinensis, Neotrapezium californicum, as well as P.
aequilateralis, were found articulated. Growth se-
ries of these three bivalves and of the gastropods
Crepidula inornata Dickerson, 1916, and Loxotre-
ma turritum were also found in this fossiliferous
mudstone.

Link and Welton (1982:fig. 3, section 5) reported
lignite in the restricted-coastal facies and graphi-
cally depicted several lignite beds scattered
throughout their measured section, which is located
along the same traverse as the main measured sec-
tion of this present study. I was able to find car-
bonaceous material (rare, very small pieces of car-
bonized wood), although only at CSUN locality
1445 and near the top of unit 3 in the auxiliary
section. Blackish, lignitic-looking stains are associ-
ated with some of the mudstones near CSUN lo-
cality 1446, but these stains are related to postde-
positional processes. The mudstone at CSUN lo-
cality 1453 is black and superficially resembles
coal.

FAUNA

A total of about 1270 molluscan specimens iden-
tifiable to species were collected. Preservation of the
fossils ranges from poor to moderately good, but
many are in poor condition because of crushing,
weathering, being coated with tightly adhering mud
matrix, or a combination of all three factors. The
bivalve specimens generally do not lend themselves
for cleaning of the hinge, and, therefore, identifi-
cation of some of the bivalves was particularly dif-
ficult. Many of the bivalves, especially “ Tellina ”
joaquinensis and Neotrapezium californicum, are
very fragile because the shells are particularly thin.
The hinges are easily destroyed when attempts are
made to remove tightly adhering mudstone.

Although I collected most of the studied speci-
mens, I also used Jestes’ (1963) collection, which is
now stored at LACMIP. Diversity of the entire
megafauna is low, with only 16 identifiable species,
but some of the species are represented by extreme-
ly abundant specimens. Each locality has one to
several dominant species, and these vary from lo-
cality to locality (Table 1).

Although Link and Welton (1982) reported tur-
ritellas and the freshwater bivalve Unio ? from the
study area, these taxa are not present. The so-called
“turritellas” are the potamidid Potamides ( Potam -
idopsis) calif ornica, and the Unio ? is Neotrapezium
californicum.

The only other megafossils found in the restrict-
ed-coastal facies were a few, minute-sized fish scales
(in mudstone at CSUN locality 1450) and rare spec-
imens of encrusting bryozoans on oyster shells (at
CSUN locality 1444).

TAPHONOMY

Both Link (1975) and Link and Welton (1982) uti-
lized the unpublished work of Jestes (1963) in con-
cluding that the coquinas are made up of brackish-
water species that are essentially in place. Squires
(1991a, 1998) studied two of the gastropod species
from these coquinas and also reported them to have
been brackish-water dwellers that have not under-
gone any significant postmortem transport.

Using the taphonomic terminology of Kidwell et
al. (1986), the megafossil assemblages in the finer
grained intervals represent parautochthonous ass-
semblages. These are ones that underwent some
postmortem transport but were deposited within
their original habitat. The amount of postmortem
transport is not the same, however, for all the mol-
lusks in the finer grained intervals. Those at each
of the seven collecting localities (Figs. 1-3) show
little or no obvious evidence of any postmortem
transport. They have unworn delicate sculpture,
nearly complete growth series of the more abun-
dant species, and are unbroken. There is also a high
percentage of articulated and/or “butterflied” spec-
imens, especially of Pelecyora aequilateralis, “ Tel-
lina” joaquinensis, and Neotrapezium californicum.
These assemblages are closely analogous to paleo-
communities, but they cannot be referred to as such
because no specimens were found in life position.
According to Kidwell et al. (1986), paleocommun-
ities (autochthonous assemblages) are composed of
specimens preserved in life position. Evidently, the
megafossils found at each of the seven collecting
localities experienced postmortem transport of a
very short distance. At CSUN locality 1453, there
is, in fact, some evidence of postmortem transport
where localized concentrations of only juvenile
Tympanotonos (T.) californicus are found. They
show a low degree of preferred orientation. It is not
uncommon to find adult shells lying next to each
other, with their apices pointing in opposite direc-
tions. This is commonly associated with wave sort-
ing of shells, but the amount of postmortem trans-
port must have been slight because the very delicate
apical tips are present on many specimens, and
sharp projecting nodes are present on nearly every
specimen. These localized concentrations grade into
mudstone, containing other mollusks that show no
obvious signs of postmortem transport.

Burial of all the mollusks must have been rapid
because the shells show no clionid sponge bore-
holes, algal boreholes, or corrosion and only rare
cases of epifaunal incrustations by bryozoans. The
only boreholes found were on rare specimens of
Tympanotonos (T.) californicus, Pelecyora aequi-
lateralis, and “ Tellina ” joaquinensis. These bore-
holes most likely were made by the carnivorous
gastropods Crommium sp. cf. C. andersoni (Dick-
erson, 1914) and Neverita (N.) globosa Gabb,
1869.

The fossils in the muddy coquinalike beds, which
usually directly underlie the seven main collecting

Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks ■ 7

localities, were more affected by postmortem trans-
port, as evidenced by their concentration in large
numbers of essentially one species (e.g., Pelecyora
aequilateralis or Cuneocorbula torreyensis). Nev-
ertheless, postmortem transport was within the
original environment, based on the unworn and un-
broken condition of the specimens and the presence
of these same species in the immediately overlying
assemblages that show little or no obvious signs of
postmortem transport.

Using the taphonomic terminology of Kidwell et
al. (1986), the fossils in the sandstones represent
allochthonous assemblages transported out of their
original habitat. These fossils consist only of scat-
tered fragments of the oyster Acutostrea idriaensis
idriaensis that were transported out of their nearby
original muddy environment (see discussion below)
and deposited in sand, which represented a foreign
substratum. Their transported condition is based
on their fragmented condition (mostly small-sized
fragments). Only in the lower part of unit 3 in the
main measured section are the fragments abundant
enough to form an oyster hash. The source of the
oysters was local, based on the presence of unworn,
large single valves (up to 80 mm long) and a few
articulated specimens of the oysters in the subjacent
finer grained intervals. When the oysters were
transported, they survived the high-energy process-
es to some degree because of their stout shells.

DEPOSITIONAL ENVIRONMENT

In the region surrounding the study area, the Ma-
tilija Sandstone consists of deep-marine to shallow-
marine deltaic facies associated with a major re-
gressive event. In the lower part of the formation,
sand-rich proximal turbidite deposition took place
in outer neritic, bathyal depths, or both on the
flanks of a prograding delta. As the delta prograded
and filled the basin, the turbidites were covered by
shallow-marine shelf deposits that grade upward
into the restricted-coastal deposits, which formed
at the delta top. These latter deposits are overlain,
in turn, by shallow-marine shelf deposits that make
up the uppermost part of the Matilija Sandstone.
These uppermost beds are transitional with the
deep-water Cozy Dell Shale and record a rapid ba-
sin deepening (Link, 1975; Link and Welton,
1982).

Link (1975) reported that the finer grained inter-
vals in the restricted-coastal facies represent a low-
energy lagoonal, bay, or estuarine environment. He
also reported that the interbedded limestone and
gypsum, in association with mudcracks and red
beds suggest very shallow, quiet-water deposition
in a high-evaporation environment. Furthermore,
he reported that the sandstones formed in much
higher energy conditions associated with narrow
beaches or tidal channels. Link and Welton (1982)
and Squires (in press) refined these interpretations
and reported that the mudstone and coquina in the
finer grained intervals represent a low-energy, la-

goonal environment where brackish-marine mol-
lusks lived, and that the limestone, evaporites, and
“red beds” formed in a sabkha environment along
the margin of the lagoon. The “red beds” probably
indicate subaerial exposure, and furthermore the
well-sorted, horizonally laminated, very fine to fine-
grained sandstones represent beach and barrier-bar
washover deposits brought by storms into the la-
goon. The limestone to gypsum evaporite sequences
represent increasingly hypersaline conditions along
the shoreline of the lagoon, and the complex alter-
ations of these lithologies indicate rapidly fluctu-
ating conditions. Squires (1998) reported that the
formation of these evaporites coincided with a re-
gional change of climate from humid subtropical or
tropical to seasonal semiarid conditions.

The megafauna in the restricted-coastal rocks at
Matilija Hot Springs is strongly indicative of brack-
ish-water conditions. Some of the species are
known elsewhere only from brackish-marine rocks
and include: Potamides ( Potamidopsis ) californica,
Loxotrema turritum, Pyrgulifera (P.) lajollaensis
(Hanna, 1927), Pelecyora aequilateralis, Cuneocor-
bula torreyensis, and “Te//m<z” joaquinensis. Each
of these species is discussed below.

Potamides ( Potamidopsis ) californica has been
found elsewhere only in brackish-marine rocks in
the Matilija Sandstone at Beartrap Creek of the
Pine Mountain area, Ventura County, southern
California (Squires, 1991a). The paleoenvironment
of this species is consistent with modern analogues.
Today, potamidids feed on surface detritus that ac-
cumulates on mud surfaces of enclosed intertidal
flats in warm waters (Morton and Morton, 1983).

Loxotrema turritum is widespread and ranges
from southern California to northwestern Kam-
chatka, Russia. Nearly all of the northeastern Pa-
cific early Eocene specimens of this species under-
went downslope postmortem transport, most likely
from deltaic areas, into deeper waters and became
mixed with shallow-marine mollusks. The north-
eastern Pacific middle Eocene specimens of this spe-
cies lived in brackish-marine lagoons or bays within
deltaic complexes (Squires, 1998). The northwest-
ern Kamchatka middle Eocene specimens are found
in coastal-marine rocks (Devyatilova and Volob-
ueva, 1981).

Pyrgulifera (P.) lajollaensis is present in brackish-
marine rocks in San Diego County, southern Cali-
fornia (Hanna, 1927; Givens and Kennedy, 1979).
In modern usage, these rocks are referred to as the
Delmar Formation. This species is also known from
one locality (UCR loc. 4747) in the Matilija Sand-
stone at Beartrap Creek (Givens, 1974). At this lo-
cality, the species is associated with Loxotrema tur-
ritum, Potamides (P?) carbonicola Cooper, 1894,
and the gastropod Nerita ( Theliostyla ) triangulata
Gabb, 1869, as well as with the bivalve Acutostrea
idriaensis idriaensis. These latter two species are
common constituents of northeastern Pacific brack-
ish-marine and very shallow-marine Eocene mol-
luscan faunas (e.g., Vokes, 1939; Squires, 1984;

8 ■ Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

Squires, 1992). Pyrgulifera, long believed to be a
freshwater genus, was reported by Bandel and Rie-
del (1994) to be a brackish-water genus that could
tolerate freshwater inflow.

Pelecyora aequilateralis and Cuneocorbula tor-
reyensis have been found elsewhere in three for-
mations. One is the brackish-marine Delmar For-
mation near San Diego (Hanna, 1927; Givens and
Kennedy, 1979). The second is a brackish-marine
part of the Domengine Sandstone in the Vallecitos
syncline near New Idria in central California, and
some of these rocks contain Loxotrema turritum
and Potamides (P.?) carbonicola, as well as some
coal beds (Vokes, 1939). The third is a section of
rocks referred to by modern workers as the White-
tail Ridge Formation near Glide in southwestern
Oregon (Turner, 1938; Niem et al., 1992). Utilizing
the work of Niem et al. (1992), Squires (1998) as-
signed these latter rocks to a deltaic (mixed fluvial
and shallow marine) origin. Cuneocorbula torrey-
ensis is also recorded from “Transition” age brack-
ish-marine strata in the upper Juncal Formation in
the Pine Mountain area in southern California
(Givens, 1974).

“ Tellina ” joaquinensis has been found elsewhere
with certainty only in localized brackish-marine
rocks in Coalmine Canyon near Coalinga, central
California (Arnold, 1909; Arnold and Anderson,
1910; Vokes, 1939). In modern usage, these rocks
are referred to as the Domengine Formation. Two
of the associated species in the Coalmine Canyon
rocks are Loxotrema turritum and Potamides (P.?)
carbonicola.

Crommium andersoni, which is most likely pre-
sent at Matilija Hot Springs, as well as Neverita
(N.) globosa, Acutostrea idriaensis idriaensis, and
Barbatia ( B .) morsei Gabb, 1864, which are all pre-
sent at Matilija Hot Springs, are similar in that they
all have been found in brackish-marine strata, as
well as in shallow-marine environments (Arnold,
1909; Vokes, 1939; Givens, 1974; Givens and Ken-
nedy, 1976, 1979; Squires, 1987; Nesbitt, 1995).
These species, which appear to have been euryha-
line, indicate that the muddy and silty environs in
the restricted-coastal facies probably had some con-
nection to the open ocean. Nesbitt (1995) reported
an Acutostrea idriaensis idriaensis paleocommunity
from the upper middle Eocene Cowlitz Formation
of southwestern Washington, and she inferred that
this oyster “inhabited a shallow-water, soft-bottom
embayment of a delta shore in which the water
temperatures and salinites were seasonally very var-
iable.”

The gastropod Crepidula inornata and the bi-
valve domenginensis Vokes, 1939, are the

only mollusks in the restricted-coastal facies at Ma-
tilija Hot Springs that have not been previously
found in brackish-marine deposits. Possibly dis-
placed (brackish-marine?) specimens of C. inorna-
ta, however, have been found in rocky shoreline de-
posits in the basal part of the Tejon Formation at
the Edmonston Pumping Plant, Tehachapi Moun-

Contributions in Science, Number 480

tains, south-central California (Lindberg and
Squires, 1990). The numerous and well-preserved
juvenile through adult, growth-stage specimens of
this gastropod at Matilija Hot Springs strongly in-
dicate that this species inhabited the brackish-ma-
rine environment. Like modern analogues, C. in-
ornata would have been a hard-substrate-dwelling
gastropod with a sedentary, epifaunal suspension
(filter) feeding mode of life. Most Crepidula spp.
are generalists with respect to temperature and sa-
linity, and this has allowed them to be stable species
in unstable environments (Hoagland, 1977), which
would be the norm for brackish-marine conditions.

“ Melanatria ” markleyensis (Clark, 1938) was
known previously only from the northeastern Pa-
cific region at a single locality in the Markley For-
mation in northern California. This locality con-
tains mostly shallow-marine mollusks in coarse-
grained sandstone with lenses of conglomerate, but
Clark (1938) reported that the megafauna has a
brackish-water element, as shown by “ Melanatria ”
markleyensis and species belonging to Corbicula
and the gastropod Elimia. In the northwestern
Kamchatka area, “M.” markleyensis has been
found in middle Eocene coastal-marine rocks (Dev-
yatilova and Volobueva, 1981). Today, Melanatria
is found in rivers and streams in Madagascar (Star-
miihlner, 1969; Brown, 1980). The presence of
“M.” markleyensis at the Matilija Hot Springs sec-
tion could be explained in two ways: During the
Eocene, either this species was a brackish-water
dweller, or the specimens in the section were trans-
ported there from nearby freshwater sources. The
former seems more likely because the specimens are
unabraded.

The three new species found in the restricted-
coastal facies at Matilija Hot Springs belong to gen-
era that can be present in brackish-marine condi-
tions. Tympanotonos is moderately common in Eo-
cene and lower Oligocene strata of France, and Git-
ton et al. (1986) reported the genus from lower
Oligocene lagoonal-marine strata in the Paris basin.
The genus is found today in coastal-marine man-
grove swamps (Bouchet, 1977; Plaziat, 1977).
Modern species of Neotrapezium are commonly
found attached by their byssus to hard substrates,
and at least a few species (such as Neotrapezium
liratum; Reeve, 1843) live in warm waters of low
salinity (Kira, 1965; Morton, 1979; Morton and
Morton, 1983). Neotrapezium californicum might
have lived attached to the oyster Acutostrea id-
riaensis idriaensis. The byssate epifaunal bivalve
Barbatia {B.) morsei and hard-substrate-dwelling
gastropod Crepidula inornata could have done like-
wise.

Although fossil forms of Corbicula are found in
brackish-marine, freshwater, and shallow-marine
strata, modern forms are found only in brackish-
marine and freshwater environments (Keen and
Casey, 1969).

The depositional scenario of the restricted-coast-
al facies of the Matilija Sandstone at Matilija Hot

Squires: Brackish-Marine Eocene Mollusks ■ 9

Springs agrees closely with what has been observed
in modern lagoons that form behind barriers. As
summarized by Boggs (1987), modern lagoons are
typically low-energy environments, although tidal
currents move into the lagoons through inlets be-
tween barriers, winds create some wind action
along shorelines, and storms provide occasional pe-
riods of high-energy waves that can bring in wash-
over deposits from the barrier beach. Interbedded
sands are generally horizontally laminated. Faunas
are highly variable and generally characterized by
low diversity. The salinity conditions largely dictate
the taxonomic composition of the faunas. Lagoons
with normal salinity show faunas similar to those
of the open ocean, whereas brackish-marine faunas
dominate more restricted lagoons (Boggs, 1987).
Fluge numbers of specimens are commonly associ-
ated with these brackish-marine conditions (Bandel
and Riedel, 1994). Carbonate deposition can pre-
vail if somewhat hypersaline conditions are present,
and if these conditions become very arid, then
evaporites (mainly gypsum) form. Very hypersaline
lagoons contain few organisms (Boggs, 1987).

AGE

Link and Welton (1982) reported a middle Eocene
(Pll and P12 Zones of the standard planktonic zo-
nation) age for the overlying Cozy Dell Formation
in the Matilija Hot Springs area. The Pll Zone is
equivalent to the CPI 3b and CPI 3c Zones of the
standard calcareous nannoplankton zonation
(Berggren et al., 1995). The restricted-coastal fa-
cies, therefore, are no younger than the CPI 3b or
CPI 3c Zones. Squires (in press) assigned the re-
stricted-coastal facies to the lower middle Eocene
“Transition Stage,” which is equivalent to the
CP13a Zone.

SYSTEMATIC MATERIALS AND METHODS

Systematic arrangement of higher taxa of the gastropods
generally follows Ponder and Waren (1988), and that of
the bivalves follows Vokes (1980) and Coan and Scott
(1997). The synonymies are selective. Usually, only works
that include illustrations of the species are listed. In a few
cases, however, works that only have faunal lists are in-
cluded if they add significant geographic information
about a species. The figured specimens used in this report,
as well as the material collected by Jestes (1963), are on
deposit in the Natural History Museum of Los Angeles
County, Invertebrate Paleontology Section. Additional un-

figured specimens are on deposit in the Department of
Geological Sciences Paleontology collection, California
State University, Northridge.

SYSTEMATICS

Class Gastropoda Cuvier, 1797

Superorder Caenogastropoda Cox, 1959

Order Neotaenioglossa Haller, 1882

Superfamily Cerithioidea Ferussac, 1819

Family Potamididae Adams and Adams,
1854

Genus Potamides Brongniart, 1810

TYPE SPECIES. Potamides lamarcki Brongniart,
1810, by monotypy; Oligocene, St. Michiel, France.

Subgenus Potamidopsis Munier-Chalmas,
1900

TYPE SPECIES. Cerithium tricarinatus Lamarck,
1804, by original designation; middle Eocene, Paris
Basin, France.

Potamides (Potamidopsis) californica
Squires, 1991a
Figures 4, 5

Potamides sp. Jestes, 1963:223.

Potamides aff. P. tricarinata (Lamarck). Jestes,
1963:225.

Potamides ( Potamidopsis ) californica Squires,
1991a:356-358, figs. 2-5.

PRIMARY TYPE MATERIAL. LACMIP holo-
type 11300, LACMIP paratypes 11301, 11302; all
from the Matilija Sandstone, Beartrap Creek, Pine
Mountain area, Ventura County, southern Califor-
nia, LACMIP loc. 7226.

ILLUSTRATED SPECIMENS. LACMIP hypo-
types 12440-12441.

MOLLUSCAN STAGE RANGE. “Transition” to
lower part of “Tejon.”

GEOGRAPHIC DISTRIBUTION. Matilija Hot
Springs and Beartrap Creek, Ventura County,
southern California.

LOCAL OCCURRENCE. CSUN Iocs. 1444,
1445, 1446, 1450, 1451, 1453.

REMARKS. This is the most ubiquitous gastro-
pod in the restricted-coastal facies and is very abun-

Figures 4-15. Gastropods from Matilija Hot Springs area upper part of the Matilija Sandstone. CSUN loc. 1453 unless
otherwise indicated. All specimens coated with ammonium chloride. 4, 5. Potamides ( Potamidopsis ) californica Squires,
1991. 4. Abapertural view, X2.5, LACMIP hypotype 12440. 5. Apertural view, X2.6, LACMIP hypotype 12441. 6-13.
Tympanotonos (T.) californicus new species. 6. Apertural view, X3.9, LACMIP paratype 12443. 7. Apertural view, X3.5,
LACMIP paratype 12444. 8. Abapertural view, X3.8, LACMIP paratype 12445. 9. Apertural view, X3.7, LACMIP
holotype 12442. 10. Abapertural view, x 3, LACMIP paratype 12446. 11. Apertural view, X3.6, LACMIP paratype
12447. 12, 13. LACMIP paratype 12448. 12. Abapertural view, x 3. 13. Oblique view of partial left side and anterior
end, X3. 14. Pyrgulifera (P.) lajollaensis (Hanna, 1927), abapertural view, X2.4, LACMIP hypotype 12449, CSUN loc.
1445. 15. uMelanatria'” markleyensis (Clark, 1938), abapertural view, X3.9, LACMIP hypotype 12450, CSUN loc. 1445.

10 ■ Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks ■ 1 1

dant at localities 1444 and 1450. At the other lo-
calities, between six and 12 specimens were found.
Specimens are usually encased in brittle mudstone
or silty mudstone, and extraction from the outcrop
almost always results in loss of the uppermost spire
and the aperture. Although no complete specimens
were found, a few nearly complete specimens (up
to 32 mm high) were recovered. This species is
characterized by a turritelliform shape, concave
whorls with reticulate sculpture, and a noded Ca-
rina near the anterior suture. A sutural spiral rib
immediately posterior to the suture can be strong
on some specimens, such as the exceptionally well-
preserved specimen illustrated in Figure 5. This
same specimen also shows that the anterior half of
the body whorl has three spiral ribs, all of which
are noded. None of the specimens at any of the
localities shows evidence of postmortem transport.
The nodes on the carina are always sharp and un-
worn.

Potamidopsis is known only from upper Paleo-
cene and middle Eocene brackish-marine strata in
France and lower middle Eocene brackish-marine
strata in southern California (Squires, 1991a). Saul
and Squires (1998) reported a possible Early Cre-
taceous (Hauterivian Stage) species of Potamidop-
sis from the Ogo Member of the Budden Canyon
Formation along the North Fork Cottonwood
Creek, Shasta County, northern California. To date,
Potamides ( Potamidopsis ) californica is known
only from the Matilija Sandstone in Ventura Coun-
ty, southern California.

Genus Tympanotonos Schumacher, 1817

TYPE SPECIES. Tympanotonos fluviatilis Schu-
macher, 1817 [ = Mur ex fuscatus Linnaeus, 1758],
by monotypy; Recent, West Africa.

Subgenus Tympanotonos s.s.
Tympanotonos ( Tympanotonos ) calif ornicus new
species
Figures 6-13

DIAGNOSIS. A Tympanotonos s.s. in which the
upper spire has rounded whorls with three equal
spiral ribs showing small nodes and the lower spire
has angulate whorls with cancellate sculpture con-
sisting of two to three spiral ribs and numerous
axial ribs.

COMPARISON. Tympanotonos (T.) calif ornicus
new species is most similar to T. (T.) fuscatus rad-
ula, which lives today in mangroves along the coast
of West Africa. Some earlier workers, such as Thie-
le (1929-1935) considered T. (T.) fuscatus radula
to be a distinct species, but modern workers such
as Plaziat (1977) consider it a variety of T. (T.) fus-
catus. Plaziat (1977) found both in the same gen-
eral brackish-water habitat but reported that radula
lives higher than fuscatus in the estuarine and del-
taic mangrove swamps near Douala in Cameroon,
Africa. A specimen of T. (T.) radula illustrated by
Thiele (1929:fig. 202) and specimens of T. (T.) fus-

catus radula illustrated by Plaziat (1977:figs. 6a-
6b, 7a-7c) are useful for comparative purposes
with the new species, which differs by having
slightly more rounded whorls on the lower spire.
The new species is similar in the upper spire sculp-
ture to T. (T.) fuscatus fuscatus from Liberia, Africa
[LACM lot 51-4], but the new species differs by
having much more subdued sculpture on the rest of
the shell rather than having the large and projecting
spines that characterize the carinate shoulder of the
lower spire and body whorl of T. (T.) fuscatus fus-
catus.

DESCRIPTION. Medium in size, up to 23 mm
high (estimated), turreted-conical, approximately
15 whorls; high-spired with spire about two-thirds
of shell height. Suture impressed, immediately an-
terior to a spiral riblet. Protoconch missing, apical
area rapidly tapering (acicular), pleural angle ap-
proximately 20°. Whorl convexity changes with
growth; earlier spire whorls rounded and grada-
tional into angulate whorls on more mature spire
and body whorl. Teleoconch sculpture consisting of
spiral ribs crossed by numerous axial ribs. Sculp-
ture changes with growth. Upper spire with three
equal spiral ribs; beaded to noded where crossed by
axial ribs. On middle and lower spires, posterior-
most spiral rib weakens, with the other two spiral
ribs becoming carinate, more strongly noded, and
more prominent with growth. Sculpture cancellate
with nodes projected somewhat; interareas between
the two rows of nodes rather deep. On rare speci-
mens, weak posterior-most spiral rib obsolete or
nearly so. Mature whorls of very rare specimens
with a fourth spiral rib, moderately prominent and
noded, near the anterior suture. Posterior half of
body whorl with three spiral ribs bearing strong
nodes; posterior-most spiral rib weakest. Anterior
half of body whorl flattish, with approximately six
unnoded spiral ribs, strength decreasing anteriorly.
Aperture somewhat roundish; inner lip smooth and
twisted. Anterior end of aperture with a narrow but
distinct notch. Outer lip not seen.

HOLOTYPE DIMENSIONS. 18.9 mm high, 7.5
mm wide.

PRIMARY TYPE MATERIAL. LACMIP holo-
type 12442 (illustrated), LACMIP paratypes 12443
to 12448 (all illustrated); all from CSUN loc. 1453.
TYPE LOCALITY. CSUN loc. 1453.
MOLLUSCAN STAGE RANGE. “Transition.”
GEOGRAPHIC DISTRIBUTION. Matilija Hot
Springs, southern California.

LOCAL OCCURRENCE. CSUN Iocs. 1444,
1450, 1451, 1453.

REMARKS. The new species is most abundant
at CSUN locality 1453, where it forms coquinas. It
is common at the other localities. Specimens range
from 5 to 25 mm in height. Whether or not the
largest specimens found represent fully mature in-
dividuals cannot be resolved, but it seems unlikely.
On the bedding planes of some hand-specimens of
rock, there are densely packed patches of mostly
same-sized specimens less than 10 mm high. These

12 ■ Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

patches represent concentrations of juvenile speci-
mens. These small specimens have rounded whorls
with beaded sculpture on three equal spiral ribs and
could easily be mistakenly identified as a separate
species if more mature specimens (e.g., Figs. 8, 9,
12) showing a transition from rounded whorls with
beaded sculpture to angulate whorls with cancellate
sculpture were not present. It is uncommon to find
specimens that show this transition in sculpture be-
cause nearly all the more mature specimens are
missing their tips. This might be the result of break-
age that occurred either during postmortem trans-
port or during removal of the larger specimens
from the rock. The mudstone containing the spec-
imens of the new species is brittle and highly frac-
tured. It easily falls apart, and removal of the larger
specimens is extremely difficult. In addition, no
specimens of the new species were found that show
the outer lip, and only rare specimens show the out-
line of the aperture. Most likely these features were
either crushed by postburial compaction or broken
off when the rock was split to initially reveal the
specimens. It is also possible that, at least in some
cases, the more delicate parts of the shells were bro-
ken off prior to burial.

Wenz (1939) reported the geologic range of Tym-
panotonos to be Late Cretaceous (Turonian) to Re-
cent. During the Eocene, the genus was relatively
diverse in France, especially in the Paris Basin
(Cossmann and Pissarro, 1910-1913; Le Renard
and Pacaud, 1995). The French species belong to
either subgenus Eotympanotonus Chavan, 1952, or
to subgenus Diptychocbilus Cossmann in Don-
cieux, 1908. The former is characterized by whorls
whose posterior-most spiral rib develops spines,
whereas the latter has a smooth but tabulate carina
on the shoulder of the whorls. Neither subgenus is
characterized by cancellate sculpture such as that
seen on the new species.

The new species, which belongs to Tympanoto-
nos s.s., is the only confirmed record of Tympano-
tonos in North America. Flynn et al. (1989:fig. 3
[la-lbj) reported Tympanotonos sp. aff. T. papalis
from lower Eocene (“Capay Stage”) strata of the
Bateque? Formation near the village of “El Rosa-
rio” [i.e., Rosarito] in Baja California Sur, Mexico.
Squires and Demetrion (1992), however, reported
that the northernmost exposures of the Bateque
Formation are far south of the area studied by
Flynn et al. (1989). The specimens of the so-called
Tympanotonos sp. aff. T. papalis from the Rosarito
area lack apertures and upper spires. These speci-
mens are quite unlike those of T. ( Eotympanoton-
us) papalis (Deshayes, 1833; Cossmann and Pissar-
ro, 1910-19 13:pl. 29, fig. 151bis-5), known from
lower Eocene (Cuisian Stage) strata of the Paris Ba-
sin, France. The Mexican specimens have a lowly
noded carina on the shoulder of the whorls, and
the remaining parts of the whorls are smooth. The
Mexican specimens are somewhat similar to Paris
Basin species of Tympanotonos ( Eotympanotonus ),
as well to certain species of the cerithiid genus Ser-

Contributions in Science, Number 480

ratoceritbium Vignal, 1897, but positive identifi-
cation of the Mexican specimens awaits better pre-
served material.

ETYMOLOGY. The species is named for the
state of California.

Genus Pyrgulifera Meek, 1877

TYPE SPECIES. Pyrgulifera humerosa Meek,
1877, by monotypy; Upper Cretaceous (Cenoman-
ian), Bear River Formation, near Bear River, south-
western Wyoming.

Subgenus Pyrgulifera s.s.

Pyrgulifera ( Pyrgulifera ) lajollaensis
(Hanna, 1927)

Figure 14

Tricbotropis (?) lajollaensis Hanna, 1927:311-312,

pi. 48, figs. 4-6, 9, 11.

Tricbotropsis ? lajollaensis Hanna. Clark, 1929:pl.

10, fig. 12.

Tricbotropis{}) sp. Jestes, 1963:226.

“ Tricbotropis ” lajollaensis Hanna. Givens, 1974:

70, pi. 6, fig. 18.

Gyrineum (?) sp. Jestes, 1963:226.

Pyrgulifera lajollaensis (Hanna). Givens and Ken-
nedy, 1979:95, table 2.

PRIMARY TYPE MATERIAL. UCMP holotype
30906, UCMP loc. 3992; UCMP paratype 30907,
UCMP loc. 5084; UCMP paratype 30908, UCMP
loc. 3992; all from the Delmar Formation, south of
Del Mar, San Diego County, southern California.

ILLUSTRATED SPECIMENS. LACMIP hypo-
type 12449.

MOLLUSCAN STAGE RANGE. “Domengine”
to lower part of “Tejon.”

GEOGRAPHIC DISTRIBUTION. San Diego
and Matilija Hot Springs, southern California.

LOCAL OCCURRENCE. CSUN Iocs. 1445,
1446, 1450, 1452, 1453.

REMARKS. A total of 21 specimens were found.
Eight were found at locality 1450, and they range
from 13 to 28 mm in height. Seven specimens were
found at locality 1445, and they range from 18 to
25 mm in height. At the other localities, only two
or three specimens were found. Nearly all the col-
lected specimens are poorly preserved. Some are
only internal molds.

This species is characterized by a short, stout
shell with angulate whorls, a wide ramp, numerous
pointed axial nodes, and strong spiral ribs covering
the entire shell. The axial nodes on the Matilija Hot
Springs specimens are sharp and unworn, thereby
indicating no signs of postmortem transport.

The familial assignment of Pyrgulifera is tenu-
ous. Traditionally, it has been assigned to family
Thiaridae, but Bandel and Riedel (1994) recently
placed it in family Potamididae. Wenz (1939) re-
ported that the geologic range of Pyrgulifera is Late
Cretaceous (Cenomanian) to Eocene. Two species
are known from estuarine rocks in the Upper Cre-

Squires: Brackish-Marine Eocene Mollusks ■ 13

taceous (Cenomanian) Bear River Formation in
southwestern Wyoming (White, 1895; Yen, 1954,
1958). Stephenson (1952) described two species of
Pyrgulifera from the Cenomanian Woodbine For-
mation of Texas, but Bandel and Riedel (1994) con-
sidered the generic identification as doubtful. One
of these Woodbine Formation species, Pyrgulifera
ornata Stephenson (1952:157-158, pi. 37, figs. 9-
13), however, does look much like a Pyrgulifera.

Givens (1974:70) reported Pyrgulifera lajollaen-
sis from the Ectinocbilus canalifer megafaunal bio-
zone of the Matilija Sandstone at Beartrap Creek
of the Pine Mountain area, Ventura County, south-
ern California. Squires (in press) reported that the
Matilija Sandstone in the Beartrap Creek area is
early middle Eocene in age and equivalent to the
lower part of the “Tejon Stage.”

Family Thiaridae Troschel, 1857
Genus Melanatria Bowdich, 1822

TYPE SPECIES. Buccinum flumineum Gmelin in
Linnaeus, 1767, by original designation?; Recent,
Madagascar, rivers and streams.

“ Melanatria ” markleyensis (Clark, 1938)

Figures 15-17

Thiara ( Melanoides ) markleyensis Clark, 1938:

706, pi. 3, figs. 24, 30; table 1.

Cerithium sp. Jestes, 1963:222.

Bittium (?) dumblei (Dickerson). Jestes, 1963:226.
Melania markleyensis (Clark). Devyatilova and Vo-

lobueva, 1981:114, pi. 9, figs. 16-18.

PRIMARY TYPE MATERIAL. UCMP holotype
30891; UCMP paratype 30892; both from the
Markley Formation near Vacaville, Solano County,
northern California, UCMP loc. A1297.

ILLUSTRATED SPECIMENS. LACMIP hypo-
types 12450 and 12451.

MOLLUSCAN STAGE RANGE. “Transition” to
“Tejon.”

GEOGRAPHIC DISTRIBUTION. Matilija Hot
Springs, southern California; Pleasant Creek near
Vacaville, northern California; and northwestern
Kamchatka, Russia.

LOCAL OCCURRENCE. CSUN Iocs. 1445,
1446.

REMARKS. A total of 45 specimens were found,

and 30 of these are from locality 1445. Although
most specimens from locality 1445 are fragments,
usually consisting of just the apices, a few represent
nearly complete specimens that range from 11 to
26.5 mm in height. Jestes, who must have picked
up weathered-out material, collected these frag-
mentary specimens. I collected specimens from this
same locality by removing rock from the outcrop,
and these specimens are mostly complete, although
the apertures are always poorly preserved. Only
five specimens of this species were found at CSUN
locality 1446. Two consist of moderately large frag-
ments, and the other three are weathered tips.

The specimen that best shows the aperture is il-
lustrated in Figure 16, but the outer lip and anterior
end of the aperture are missing. This species is char-
acterized by tabulate whorls on most of the spire
and body whorl, by prominent axial nodes a short
distance below the suture, and by beaded spiral
sculpture. The apertural features are poorly known
for this species, but the anterior end appears to be
unnotched. Strength of the axial nodes is variable,
as is their extent along the sides of the whorls.

I use quotation marks for the generic assignment
of Clark’s (1938) species, which cannot be assigned
with certainty to any genus because apertural de-
tails are not fully known. I tentatively assign
Clark’s species to Melanatria rather than to Thiara
Bolton in Roding, 1798 [= Melania Lamarck, 1799
fide Wenz (1939)], or Melanoides Olivier, 1804, be-
cause the species has more morphologic similarity
to the type species of Melanatria than to the type
species of the other two genera. Similar to Melan-
atria fluminea, which is the type species of Melan-
atria, Clark’s species has a turreted-conical shape,
whorls bearing prominent spiral ornament and ax-
ial ribs, and an apparently unnotched aperture. Al-
though the type species of Thiara, which is Thiara
(T.) amarula (Linnaeus, 1758), is somewhat similar
to Clark’s species, T. (T.) amarula has an oval-tur-
reted shape, a row of backward-directed spines on
the body whorl shoulder, smooth or fine spiral or-
nament, and a feebly notched anterior end of the
aperture (Davies and Eames, 1971). The type spe-
cies of Melanoides, which is M. (M.) tuberculata
(Muller, 1774), differs significantly from Clark’s
species by having a fusiform shape, rounded
whorls, and cancellate sculpture (especially on the
spire).

Figures 16-28. Gastropods and bivalves from Matilija Hot Springs area upper part of the Matilija Sandstone. CSUN
loc. 1450 unless otherwise indicated. All specimens coated with ammonium chloride. 16-26. Gastropods. 16, 17. “Me-
lanatria'’'’ markleyensis (Clark, 1938), X5.2, LACMIP hypotype 12451, CSUN loc. 1445. 16. Apertural view. 17. Aba-
pertural view. 18, 19. Loxotrema turritum Gabb, 1868. 18. Apertural view, Xl.8, LACMIP hypotype 12452. 19. Aba-
pertural view, X2.6, LACMIP hypotype 6451, CSUN loc. 1446. 20, 21. Crepidula inornata Dickerson, 1916, X2.3,
LACMIP hypotype 12453. 20. Dorsal view. 21. Right-side view. 22, 23. Crommium sp. cf. C. andersoni (Dickerson,
1914), X2.6, LACMIP hypotype 12454. 22. Apertural view. 23. Abapertural view. 24, 25. Neverita ( Neverita ) globosa
Gabb, 1869, X2.8, LACMIP hypotype 12455. 24. Apertural view. 25. Abapertural view. 26-28. Bivalves. 26. Barbatia
{Barbatia) morsei Gabb, 1864, left valve, X4.8, LACMIP hypotype 12456. 27, 28. Acutostrea idriaensis idriaensis (Gabb,
1869). 27. Left valve, Xl.7, LACMIP hypotype 12457. 28. Left valve, Xl.4, LACMIP hypotype 12458.

14 ■ Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks ■ 15

“ Melanatria ” markleyensis resembles Thiara
{ Melanoides ) calafi Clark (1938:706, pi. 3, figs. 24,
30). Both are from the same bed in the upper Eo-
cene Markley Formation in northern California.
“ Melanatria ” markleyensis differs from T. (M.) cal-
afi by having whorls with a much stronger tabulate
shoulder, coarser and fewer nodes on the shoulder,
much closer proximity between the shoulder and
the suture, and better development of beaded sculp-
ture on the spiral ribs on the whorls. Clark (1938)
reported that both of his species are closely related
to Potamides fettkei Weaver (1912:36, pi. 2, figs.
23, 24; 1942 [1943]:379-380, pi. 75, figs. 18, 21,
22, 26) from the upper middle Eocene Cowlitz For-
mation in southwestern Washington. Givens and
Kennedy (1976:964) reassigned “ Potamides ” fett-
kei to genus Melanoides. Although “ Melanatria ”
markleyensis resembles Melanoides fettkei, espe-
cially in the prominent nodes on the shoulder, T.
(M. ) calafi is more like Melanoides fettkei. “ Melan-
atria” markleyensis differs from Melanoides fettkei
by having a much closer proximity of the tabulate
shoulder and the suture, more closely spaced nodes
on the shoulder, coarser beaded sculpture on the
spiral ribs on the whorls, and no secondary spiral
ribs.

Houbrick (1991) assigned Melanatria to the fam-
ily Thiaridae. On the basis of a shared unique and
unusual foliate gastric structure, he implied that the
phylogenetic placement of Melanatria is close to the
genera Faunus Montfort, 1810, and Melanopsis
Ferussac, 1807. He also reported that more work
is needed to clarify their phylogenetic relationships.

Devyatilova and Volobueva (1981) reported this
species from coastal-marine rocks of the middle Eo-
cene Kamchik Formation along the shores of Pen-
zhin Inlet at the head of the Sea of Okhotsk, north-
western Kamchatka. Their illustrated specimens are
identical to those from the type locality in northern
California. To date, there are no precise correlation
data on how the Kamchatka Eocene rocks compare
biostratigraphically with northeastern Pacific Eo-
cene rocks. The presence of “M.” markleyensis in
the Matilija Sandstone at Matilija Hot Springs,
however, lowers the northeastern Pacific temporal
range of this species into the “Transition Stage.”
The species previously was known only from the
“Tejon Stage” in northern California.

Wenz (1939) reported the geologic range of Me-
lanatria to be Paleocene?, Eocene to Recent, with
all the confirmed fossil occurrences in Europe. If
“ Melanatria ” markleyensis is proven eventually to
belong to genus Melanatria, this species would ex-
tend the geographic range of this genus to the New
World.

Family Melanopsidae Adams and Adams,
1854

Genus Loxotrema Gabb, 1868

TYPE SPECIES. Loxotrema turritum Gabb,
1868, by original designation; early to middle Eo-

cene, California, Oregon, Washington, and Kam-
chatka.

Loxotrema turritum Gabb, 1868

Figures 18, 19

Loxotrema turrita Gabb, 1868:147, pi. 14, fig. 21;
1869:168, pi. 28, fig. 49; Arnold, 1909:14, pi. 4,
fig. 17; Arnold and Anderson, 1910:71, pi. 26,
fig. 17; Clark, 1929:pl. 10, fig. 3. Hanna, 1927:
312, pi. 50, figs. 5-8; Vokes, 1939:159, pi. 20,
figs. 15-19; Schenck and Keen, 1940:pl. 24, figs.
10-13; Weaver 1942 [1943]:374, pi. 75, figs. 1-
3; pi. 103, fig. 18; Devyatilova and Volobueva,
1981:114-115, pi. 9, figs. 19-23.

Strutkiolaria ( Loxotrema ) turrita (Gabb). Tryon,
1883:196, pi. 60, fig. 95.

Loxotrema turritum Gabb. Stewart, 1927:347-
348, pi. 26, figs. 3, 4; Turner, 1938:tables 2, 4,
8, p. 81, pi. 17, figs. 12, 13; Moore, 1968:26,
pi. 11, fig. d; Givens, 1974:70, pi. 6, fig. 17. Giv-
ens and Kennedy, 1976:963, pi. 1, figs. 5-8;
Squires, 1998.

Pachychilus ( Loxotrema ) turritum (Gabb). Wenz,
1939:686, fig. 1968.

Loxotrema cf. L. turritum Gabb. Jestes, 1963:225.
? Loxotrema turritum Gabb. Squires, 1991b:table
1, pi. 1, fig. 16.

PRIMARY TYPE MATERIAL. Lectotype ANSP
4228, designated by Stewart (1927); “Tejon Group,
10 miles west of Griswold’s, between San Juan and
New Idria” (Gabb, 1869:168); Domengine For-
mation, Vallecitos syncline area, San Benito Coun-
ty, central California.

ILLUSTRATED SPECIMENS. LACMIP hypo-
types 6451 and 12452.

MOLLUSCAN STAGE RANGE. “Meganos” to
lower part of “Tejon.”

GEOGRAPHIC DISTRIBUTION. Southern to
northern California; southwestern Oregon; Cres-
cent Bay, Olympic Peninsula, southwestern Wash-
ington; and northwestern Kamchatka.

LOCAL OCCURRENCE. CSUN Iocs. 1444,
1445, 1446, 1450, 1451, 1452.

REMARKS. This species is abundant only at lo-
cality 1450, where specimens range from 18 to 30
mm in height. At the other localities, between 4 to
10 specimens were found. Specimens are usually
encased in brittle mudstone, and extraction from
the bedrock almost always results in loss of the up-
permost spire and anterior end of the aperture.
Many of the specimens have been crushed. This
turreted species is characterized by tabulate whorls,
a large cylindrical body whorl with strong spiral
ribs, and a notched aperture. The small nodes on
the tabulate body whorl are distinct on the Matilija
Hot Springs specimens and show no signs of wear
resulting from postmortem transport. Squires
(1998) reported also that the specimens of Loxo-
trema turritum from Matilija Hot Springs show no
evidence of postmortem transport. See Squires
(1998) for a review of this species.

16 ■ Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

Devyatilova and Volobueva (1981) reported this
species from middle Eocene coastal-marine rocks
along the shores of Penzhin Inlet at the head of the
Sea of Okhotsk, northwestern Kamchatka. These
rocks are within the Kamchik Formation and the
lower part of the Tkapravayam subformation. The
specimens they illustrated are identical to those
from the type locality in central California.

Superfamily Calyptraeoidea Lamarck, 1809
Family Calyptraeidae Lamarck, 1809
Genus Crepidula Lamarck, 1799

TYPE SPECIES. Patella fornicata Linnaeus,
1758, by monotypy; Recent, eastern United States
and northwestern Europe.

Crepidula inornata Dickerson, 1916
Figures 20, 21

Crepidula inornata Dickerson, 1916:489, pi. 38,

figs. 5a, 5b; Lindberg and Squires, 1990:579.
Crepidula ( Spirocrypta ) inornata Dickerson. Vokes,

1939:165-166, pi. 21, figs. 10, 11.

Crepidula sp. Jestes, 1963:223.

PRIMARY TYPE MATERIAL. UCMP holotype
11804, Domengine Formation near Coalinga, Fres-
no County, central California, UCMP loc. 672.

ILLUSTRATED SPECIMEN. LACMIP hypoty-
pe 12453.

MOLLUSCAN STAGE RANGE. “Domengine”
to “Transition.”

GEOGRAPHIC DISTRIBUTION. Matilija Hot
Springs, Ventura County, southern California; Ed-
monston Pumping Plant, near Grapevine, Kern
County; and Coalinga area, Fresno County, central
California.

LOCAL OCCURRENCE. LACMIP Iocs. 1444,
1445, 1446, 1450.

REMARKS. Specimens of this species are abun-
dant at localities 1445 and 1450, where specimens
range from 11 to 17 mm in length. Most are well
preserved, but a few are crushed. Nearly all show
delicate growth lines. At all the other localities,
only single specimens were found.

Although some workers (e.g., Stewart, 1927;
Turner, 1938; Hoagland, 1977) considered Crepi-
dula inornata as conspecific with C. pileum (Gabb,
1864:137, pi. 29, figs. 233, 233a, 233b) from upper
Eocene to middle Oligocene strata in California
through Washington (Squires, 1987), other workers
(e.g., Clark, 1938; Vokes, 1939; Kleinpell and
Weaver, 1963) considered them to be separate spe-
cies. Crepidula inornata is distinguished from C.
pileum by having a smaller size, an elevated rather
than a submarginal spire, and a bulbous penulti-
mate whorl. The Matilija Hot Springs specimens
have all the diagnostic characters of C. inornata.
Lindberg and Squires (1990) reported this species
from “Transition Stage” strata at the Edmonston
Pumping Plant at the south end of the San Joaquin

Contributions in Science, Number 480

Valley, central California. The presence of C. inor-
nata at Matilija Hot Springs is the first record of
this species from brackish-marine strata.

Superfamily Naticoidea Forbes, 1838

Family Naticidae Forbes, 1838

Genus Crommium Cossmann, 1888

Type Species. Ampullaria willemetii Deshayes,
1825, by original designation; Eocene, France.

Crommium sp. cf. C. andersoni
(Dickerson, 1914)

Figures 22, 23

ILLUSTRATED SPECIMEN. LACMIP hypo-
type 12454.

LOCAL OCCURRENCE. CSUN Iocs. 1444,
1450.

REMARKS. Specimens are rare. Four specimens
were found at locality 1444 and two at locality
1450. They range from 10 to 20 mm in height, and
all are crushed internal molds. They resemble
Crommium andersoni (Dickerson, 1914:120, pi.
12, figs. 2a, 2b) from lower to middle Eocene strata
of California and southwestern Oregon. Marincov-
ich (1977:225-227, pi. 18, figs. 3-7) also illustrated
this species. The Matilija Hot Springs specimens,
such as C. andersoni, have tabulated whorls, a
smooth globose body whorl, and a lowly elevated
spire. Whether or not the Matilija Hot Springs
specimens have a narrow umbilical slit bounded by
a narrow cordlike angulation, such as that found
on C. andersoni, cannot be determined because the
anterior end of the Matilija Hot Springs specimens
are either incomplete or poorly preserved. If the
Matilija Hot Springs specimens are C. andersoni,
they would be the youngest record of this species.

Genus Neverita Risso, 1826

TYPE SPECIES. Neverita josephina Risso, 1826,
by monotypy; Eocene to Recent, Europe.

Neverita ( Neverita ) globosa Gabb, 1869
Figures 24, 25

Neverita globosa Gabb, 1869:161, pi. 27, fig. 39;
Dickerson, 1916:pl. 39, figs. 5a-b; Stewart,
1927:326-327, pi. 28, fig. 6; Clark and Wood-
ford, 1927:121-122, pi. 22, figs. 5-10; Turner,
1938:89, pi. 19, figs. 6-7, 13-15; Vokes, 1939:
169, pi. 21, figs. 9, 15-19; Schenck and Keen,
1940:pl. 24, figs. 2-5; Givens and Kennedy,
1979:tables 1-3.

Neverita weaveri Dickerson, 1915:57, pi. 4, figs.
lOa-b.

Neverita nomlandi Dickerson, 1917:173-174, pi.
30, figs. 2a-b.

Polinices weaveri (Dickerson). Turner, 1938:86, pi.
20, figs. 14, 16.

Neverita globosa reefensis Vokes, 1939:169, pi. 21,
figs. 24-25.

Squires: Brackish-Marine Eocene Mollusks ■ 17

Polinices ( Neverita ) globosa (Gabb). Weaver, 1942
[1943]:339, pi. 68, figs. 21, 24; pi. 69, figs. 5-6;
pi. 100, fig. 29.

Polinices ( Neverita ) nomlandi (Dickerson). Weaver,
1942 [1943]:340, pi. 69, figs. 8-9, 12.

Neverita ( Neverita ) globosa Gabb. Givens, 1974:
76; Marincovich, 1977:312-31 6, pi. 28, figs. 10-
15; pi. 29, figs. 1-3; Squires, 1984:25, fig. 7g;
1987:37-38, fig. 47.

Neverita ( Glossaulax ?) globosa Gabb. Givens and
Kennedy, 197 6:965-966, pi. 2, figs. 5-14, 16,
18-19.

PRIMARY TYPE MATERIAL. MCZ holotype
27859, Domengine? Formation, 16 km west of
Griswold’s, on the road from San Juan to New Id-
ria, and southeast of “Sheep Well,” T 15 S, R 9 E,
Priest Valley quadrangle, San Benito County, cen-
tral California.

ILLUSTRATED SPECIMEN. LACMIP hypo-
type 12455.

MOLLUSCAN STAGE RANGE. “Meganos”
through “Tejon.”

GEOGRAPHIC DISTRIBUTION. San Diego,
California, through southwestern Washington.

LOCAL OCCURRENCE. CSUN Iocs. 1444,
1450, and 1451.

REMARKS. A specimen was found at each lo-
cality. Collectively, they range from 11 to 17 mm
in height. They are poorly preserved and are miss-
ing the outer lip.

Class Bivalvia Linnaeus, 1758

Subclass Pteriomorphia Beurlen, 1944

Order Arcoida Stoliczka, 1871

Superfamily Arcoidea Lamarck, 1809

Family Arcidae Lamarck, 1809

Genus Barbatia Gray, 1847

TYPE SPECIES. Area barbata Linnaeus, 1758,
by original designation; Recent, Mediterranean to
northwestern Africa.

Subgenus Barbatia s.s.

Barbatia ( Barbatia ) morsei Gabb, 1864
Figure 26

Barbatia morsei Gabb, 1864:216, pi. 32, fig. 286;
Arnold, 1909:13, 16, pi. 3, fig. 8; Arnold and
Anderson, 1910:70, 73, pi. 25, fig. 8; Hanna,
1927:272, pi. 25, figs. 2, 10, 11, 13, 14. Clark,
1929:pl. 6, fig. 3; Stewart, 1930:87, pi. 8, fig. 7.
Barbatia ( Obliquarca ) morsei Gabb. Vokes, 1939:
49-50, pi. 1, figs. 25, 26, 28, 29; Reinhart, 1943:
30-32, pi. 1, fig. 4.

Barbatia ( Barbatia ) morsei Gabb. Moore, 1983:34,
pi. 5, fig. 7.

PRIMARY TYPE MATERIAL. UCMP lectotype
11984, designated by Reinhart (1943); Eocene of

the San Diego region (exact locality unknown), San
Diego County, southern California.

ILLUSTRATED SPECIMEN. LACMIP hypoty-
pe 12456.

MOLLUSCAN STAGE RANGE. “Domengine”
to “Transition.”

GEOGRAPHIC DISTRIBUTION. San Diego,
Matilija Hot Springs, and Pine Mountain area,
southern California; Coalmine Canyon near Coal-
inga, central California.

LOCAL OCCURRENCE. CSUN Iocs. 1444,
1450.

REMARKS. Eleven specimens were found at lo-
cality 1450, and they range from 3 to 10.3 mm in
height. Three specimens were found at locality
1444. Preservation at both localities is excellent to
good, although most specimens have been slightly
crushed. Both right and left valves are present, but
no articulated specimens were found. The thinness
of the valves makes them extremely fragile, and
specimens usually break apart during extraction
from the rock at the outcrop.

This species is characterized by the nearness of
the beak to the anterior margin. The illustrated
specimen (Fig. 26) shows the exterior sculpture bet-
ter than any other known specimen of this species.
It shows that there can be radial ribs on the pos-
terodorsal area and that these ribs are the widest
spaced of any on the shell. This specimen, which
has a sharp angle where the posterior margin meets
the hinge, shows no sign of any abrasion due to
postmortem transport.

The Matilija Hot Springs specimens of this spe-
cies are the geologically youngest known for this
species.

Order Ostreoida Ferussac, 1822

Superfamily Ostreoida Rafinesque, 1815

Family Ostreidae Rafinesque, 1815

Genus Acutostrea Vialov, 1936

TYPE SPECIES. Ostrea acutirostris Nilsson,
1827, by original designation; Upper Cretaceous,
Europe and North America.

Acutostrea idriaensis idriaensis (Gabb, 1869)
Figures 27, 28

Ostrea idriaensis Gabb, 1869:203, pi. 33, figs.
103b-d; pi. 34, figs. 103, 103a; Hanna, 1927:
276, pi. 30, figs. 1-2; pi. 31, figs. 3-4; Stewart,
1930:126-127, pi. 8, fig. 3; pi. 17, fig. 1; Vokes,
1935:291-304, pi. 22-24; Turner, 1938:46, pi. 6,
fig. 9; Weaver, 1942 [1943]:78-79, pi. 15, fig. 5;
Schenck and Keen, 1940:pl. 23, figs. 3, 4; Givens,
1974:44; Givens and Kennedy, 1979:tables 2, 4;
Squires, 1984:45, fig. 101.

Ostrea Columbiana Weaver and Palmer, 1922:13-
14, pi. 8, figs. 15, 16.

Ostrea oregonensis Packard, 1923:4-6, pis. 1-4.
Ostrea sp. Jestes, 1963:223, 225.

18 H Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

Acutostrea idriaensis idriaensis (Gabb). Moore,

1987:31-32, figs. 2, 3; pi. 14, fig. 6; pi. 16, fig.

3; pi. 29, figs. 3-5; Lindberg and Squires, 1990:

579.

Not Ostrea idriaensis Gabb. Devyatiliova and Vo-

lobueva, 1981:57, pi. 5, fig. 3.

PRIMARY TYPE MATERIAL. MCZ lectotype
15048, Domengine Formation, about 3 km east of
New Idria, N 1/2 of section 15, T 17 S, R 12 E,
Priest Valley quadrangle, San Benito County, cen-
tral California.

ILLUSTRATED SPECIMENS. LACMIP hypo-
types 12457 and 12458.

MOLLUSCAN STAGE RANGE. “Capay” to
“Tejon.”

GEOGRAPHIC DISTRIBUTION. San Diego,
California, through southwestern Washington.

LOCAL OCCURRENCE. CSUN Iocs. 1444,
1445, 1446, 1450, 1451, 1453.

REMARKS. This species is one of the most ubiq-
uitous bivalves in the restricted-coastal facies, and
it is abundant in muddy rocks at all the localities,
except at locality 1452 where it is rare. It is also
present as a few scattered fragments in the beach
and barrier-bar sandstones. At localities where it is
abundant, specimens range from 25 to 75 mm in
height. Preservation at all localities ranges from
good to poor, and nearly all the specimens are dis-
articulated. They consist of only the left (lower)
valve and are somewhat elongate, lowly arched,
and usually have low radial ribs crossed by growth
rugae. A few specimens are roundish and arched.
Articulated specimens are rare.

Subclass Heterodonta Neumayr, 1884
Order Veneroida Adams and Adams, 1856

Superfamily Arcticoidea Newton, 1891
Family Trapezidae Lamy, 1920
Genus Neotrapezium Habe, 1951

TYPE SPECIES. Cardita sublaevigata Lamarck,
1819, by original designation; Recent, Indo-Pacific.

Neotrapezium californicum new species
Figures 29-33

Unio{ ?) torreyensis Hanna. Jestes, 1963:224.
Lithophaga (?) sp. Jestes, 1963:226.

DIAGNOSIS. A small-sized Neotrapezium with
a moderately produced anterior end, usually
straight posterior hinge-line area, and prominent
growth lines.

COMPARISON. The new species is similar to
the living Neotrapezium liratum (Reeve, 1843), an
Indo-Pacific species introduced to various locations
in the northeastern Pacific region (Bonnot, 1935 [as
Cypricardia lyrata ]; Solem, 1954; Hanna, 1966).
Kira (1965:148, pi. 53, fig. 29) illustrated this spe-
cies. The new species differs by having a slightly
more produced anterior end.

Contributions in Science, Number 480

The new species is similar to Neotrapezium grig-
nonensis (Deshayes, 1824:64-65, pi. 9 figs. 18-19)
from middle and upper Eocene (Lutetian and Bar-
tonian stages) strata of the Paris Basin, France.
Cossmann and Pissarro (1904-1906:pl. 15, figs.
63-1 through 63-10) figured this Paris Basin spe-
cies, as well as seven other closely related “species”
from the same region. Le Renard and Pacaud
(1995) regarded N. grignonensis as conspecific with
all of these seven other “species.” The new species
is similar to N. grignonensis in that there is a range
in morphology from elongate-ovate to subquadra-
te, but the new species differs by having an anterior
end that is more produced, a posterior end that is
never as produced, growth lines that are usually
more accentuated, and a shell that never is as nar-
row or smooth as in some specimens of N. grig-
nonensis.

DESCRIPTION. Small-sized (up to 13.2 mm
high), thin-shelled, moderately elongate-ovate to
subquadrate, equivalved, and inequilateral. Liga-
ment external, opisthodetic. Beaks near anterior
end, prosogyrous, and moderately elevated. Ante-
rior margin slightly pointed; posterior hinge-line
straight to slightly arched. Posterior margin steeply
sloping. Very faint radial lirae rare on posterior
area. Right valve with two cardinal teeth and long
posterior lateral. Left valve with long posterior
socket (cardinals unseen). Growth lines prominent,
commonly clustered into bands; on a few specimens
growth-line bands merge into very fine concentric
ribs, strongest on posterior area and/or medial area.
Growth rugae near ventral margin on some speci-
mens. Irregular to distorted growth on some spec-
imens.

HOLOTYPE DIMENSIONS. Height 7.7 mm,
length 13.9 mm.

PRIMARY TYPE MATERIAL. LACMIP holo-
type 12459 (illustrated), LACMIP paratypes 12460
to 12463 (all illustrated); all from CSUN loc. 1450,
except paratype 12462 (from CSUN loc. 1445).

TYPE LOCALITY. LACMIP loc. 1450.

MOLLUSCAN STAGE RANGE. “Transition.”

GEOGRAPHIC DISTRIBUTION. Upper part of
Matilija Sandstone at Matilija Hot Springs area.

LOCAL OCCURRENCE. CSUN Iocs. 1444,
1445, 1446, 1450, 1451, 1453.

REMARKS. The new species is abundant at lo-
cality 1450, where specimens range from 1.3 to
13.2 mm in height. Specimens are uncommon to
rare at the other localities. Preservation is generally
good, and most are single valves. At locality 1450,
however, about 13% of the specimens are articu-
lated. Nearly all specimens are extensively crushed.
The new species shows a considerable degree of
variation in shape, based on 56 specimens pre-
served well enough to determine valve shape. Of
these, 52 (93%) are elongate-ovate, and the other
4 (7%) are subquadrate. A few of the latter even
show distorted (irregular or disjunct) growth,
where the growth lines delineate a progression of
different shapes on a valve. Consequently, the valve

Squires: Brackish-Marine Eocene Mollusks >19

Squires: Brackish-Marine Eocene Mollusks

20 ■ Contributions in Science, Number 480

appears to be a composite of several individual
valves, one on top of the other. An example is il-
lustrated in Figure 31. Modern species of Neotra-
pezium are known to be nestlers or shallow infau-
na, and the nestling mode of life commonly causes
distortion during growth (Kira, 1965; Morton,
1979). Of the 52 elongate-ovate specimens men-
tioned above, 25 have pronounced growth lines, 20
are somewhat smooth, and 7 have growth lines that
are very accentuated posteriorly. Only the holotype
of the new species shows any radial sculpture, and
it is very faint. The subquadrate-shaped valves of
the new species all have pronounced growth lines.
The new species is the first record of Neotrapezium
in the fossil record of North America. Although
Trapezium ( Schedotrapezium ) carinatum Gabb
(1864:170, pi. 23, fig. 150; Stewart, 1930:174-175,
pi. 5, fig. 5; pi. 17, fig. 4) was reported from Upper
Cretaceous (Campanian Stage fide LouElla Saul,
personal communication, 1997) rocks in Placer
County, northern California, this species is no lon-
ger considered to belong to the family Trapezidae
and is now placed (Keen, 1969a) in Schedotrape-
zium Stewart, 1930, of the family Arcticidae New-
ton, 1891.

A bivalve referred to as Trapezium claibornense
Dali (1900a:1498, pi. 43, figs. 9, 10; Harris, 1919:
154, pi. 48, figs. 3, 4) was reported as a very rare
species from middle Eocene (Claiborne Stage) strata
of Alabama. Maestrati and Lozouet (1995), how-
ever, regarded Dali’s species as probably being an
astartid. The new species differs from Dali’s species
by having a lower height relative to the length of
the valves, more variable concentric sculpture, a
more rounded posterior end, and a posterior mar-
gin that is not obliquely angular.

The new species is also the first record of sub-
genus Neotrapezium in the fossil record of North
America. This article follows Maestrati and Lo-
zouet (1995) and Coan and Scott (1997) in regard-
ing Neotrapezium as a distinct genus within the
family Trapezidae.

ETYMOLOGY. The new species is named for
the state of California.

Superfamily Corbiculoidea Gray, 1847
Family Corbiculidae Gray, 1847

Genus Corbicula Mergele von Miihlfeld,
1811

TYPE SPECIES. Tellina fluminalis Muller, 1774;
Recent, Africa, Asia, Australia, and introduced into
the United States of America (Brandt, 1974).

Corbicula jestesi new species

Figures 34-39

Corbicula williamsoni Anderson and Hanna, 1925.

Jestes, 1963:223.

Pitar sp. Jestes, 1963:222.

Macrocallista sp. Jestes, 1963:223.

DIAGNOSIS. A Corbicula with circular-rounded
shape, moderately inflated beaks, moderately
strong concentric ribbing, and a very slight umbo-
nal ridge.

COMPARISON. The new species was compared
to all other fossil species of Corbicula known from
the northern Pacific region (including Kamchatka).
The new species is most similar to C. williamsoni
Anderson and Hanna (1925:164-165, pi. 1, fig. 4;
pi. 3, fig. 2) from the “Tejon Group” at Grapevine
Canyon. Only a single specimen, a left valve, is
known of this species. The new species differs from
C. williamsoni by having concentric ribbing over
the entire left valve (not just the anterior half of the
valve), more closely spaced and weaker concentric
ribbing, a much weaker umbonal ridge, and a less
sloping anterior margin. The other northern Pacific
Eocene Corbicula species differ from the new spe-
cies by having much more produced beaks, usually
more central beaks, and nearly obsolete sculpture.

DESCRIPTION. Medium sized (up to 27 mm
high), equivalved, slightly inequilateral; circular-
rounded shape, longer than high. Thick shelled.
Beaks prosogyrous, slightly anterior of center, mod-
erately inflated, and elevated above hinge line. Very
slight posterior umbonal ridge. Anterior dorsal
margin gently sloped and concave; posterodorsal
margin straight. Anterior, ventral, and posterior
margins broadly and regularly rounded. Posterior
very slightly produced on some specimens. Three
divergent cardinal teeth in each valve, middle tooth
bifid in right valve. Left valve with a long anterior
and a long posterior lateral tooth, the latter mi-
nutely serrated. Right valve with long anterior lat-
eral socket; posterior part of hinge area unknown.
Entire external surface of both valves with moder-
ately strong, closely spaced concentric ribbing.

HOLOTYPE DIMENSIONS. Height 18.5 mm,
length 22.6 mm (a “butterflied” specimen consist-
ing of both valves).

PRIMARY TYPE MATERIAL. LACMIP holo-
type 12464 (illustrated), LACMIP paratypes 12465
to 12468 (all illustrated); all from CSUN loc. 1445.

TYPE LOCALITY. CSUN loc. 1445.

Figures 29-38. Bivalves from Matilija Hot Springs area upper part of Matilija Sandstone. All specimens coated with
ammonium chloride. 29-33. Neotrapezium californicum new species, CSUN loc. 1450, unless otherwise indicated. 29.
Left valve, X4.3, LACMIP holotype 12459. 30. Left valve, X3.6, LACMIP paratype 12460. 31. Right valve, X2.9,
LACMIP hypotype 12461. 32. Dorsal view showing ligamental area, LACMIP paratype 12462, X2.4, CSUN loc. 1445.
33. Right-valve hinge, X6.3, LACMIP paratype 12463. 34-39. Corbicula jestesi new species, CSUN loc. 1445. 34. Left
valve, Xl.8, LACMIP paratype 12465. 35. Right valve, X2, LACMIP holotype 12464. 36. Left-valve hinge, X2.7,
LACMIP paratype 12466. 37. Right-valve hinge, X3.2, LACMIP paratype 12467. 38, 39. Left-valve hinge, LACMIP
paratype 12468. 38. X3. 39. Enlargement (of a portion of Figure 38) showing serrations, X10.

Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks ■ 2 1

MOLLUSCAN STAGE RANGE. “Transition.”
GEOGRAPHIC DISTRIBUTION: Upper part of
Matilija Sandstone at Matilija Hot Springs.

LOCAL OCCURRENCE. CSUN Iocs. 1445,
1452.

REMARKS. Specimens are abundant at CSUN
locality 1445, where they range from 12.5 to 27
mm in height. Most are single valves, but they are
unbroken and unworn. Specimens are rare at
CSUN loc. 1452.

Worldwide, the temporal range of Corbicula is
Early Cretaceous to Recent (Keen and Casey,
1969). On the Pacific coast of North America, the
only Cretaceous Corbicula are two Late Cretaceous
species. One is Turonian in age and the other is
Maastrichtian in age (Dailey and Popenoe, 1966).
No Paleocene Corbicula species are known from
this area. Only three species of northern Pacific Eo-
cene Corbicula are known to be older than the new
species. They are C. oregonensis Turner, 1938, from
lower Eocene (“Capay Stage”) rocks now assigned
to the Whitetail Ridge Formation near Glide in
southwestern Oregon; C. triangula Volobueva in
Devyatilova and Volobueva, 1981, from lower Eo-
cene rocks of the Central Amaam subformation in
the Koryak Uplands north of Kamchatka; and C.
carlosensis Vokes, 1939, from middle Eocene (“Do-
mengine Stage”) rocks within the Domengine For-
mation in central California. The highest number
of northern Pacific Eocene Corbicula species are
found in middle to upper Eocene rocks in Wash-
ington (Weaver, 1942 [1943]) and in middle Eocene
rocks of northwestern Kamchakta (Devyatilova
and Volobueva, 1981).

ETYMOLOGY. The species is named for Ed-
ward C. Jestes, who did the initial paleontological
investigations of the restricted-coastal facies at Ma-
tilija Hot Springs and who found many of the spec-
imens of this new species.

Superfamily Veneroidea Rafinesque, 1815

Family Veneridae Rafinesque, 1815

Genus Pelecyora Dali, 1902

TYPE SPECIES. Cytherea hatch etigbeensis Al-
drich, 1886; Eocene, Wilcox, Alabama.

Pelecyora aequilateralis (Gabb, 1869)

Figures 40-41

Venus aequilateralis Gabb, 1869:184, pi. 30, fig.

76; Dickerson, 1916:pl. 37, figs. 2a, 2b.

Pitaria aequilateralis (Gabb). Hanna, 1927:288, pi.
39, figs. 1-5, 9, 12.

Pelecyora aequilateralis (Gabb). Stewart, 1930:
237-238, pi. 8, fig. 13; Vokes, 1939:87, pi. 14,
figs. 4, 6, 7, 8, 11; Weaver, 1942 [1943]:194, pi.
45, fig. 9; pi. 46, figs. 3, 6; pi. 104, fig. 6; Givens
and Kennedy, 1979:table 2.

Pelecyora aequilateralis (Gabb) var. Turner, 1938:
57, pi. 10, figs. 1-4.

Spisula (?) sp. Jestes, 1963:224, 226.

Pitar sp. Jestes, 1963:225.

Thyasira (?) sp. Jestes, 1963:226.

(l)Venus { Antigona ) sp. Jestes, 1963:227.

PRIMARY TYPE MATERIAL. MCZ lectotype
15039, designated by Stewart (1930); Delmar? For-
mation, San Diego area (exact location unknown),
San Diego County, southern California.

ILLUSTRATED SPECIMENS. LACMIP hypo-
types 12469 and 12470.

MOLLUSCAN STAGE RANGE. “Domengine”
to “Transition.”

GEOGRAPHIC DISTRIBUTION. San Diego,
southern California, to southwestern Oregon.

LOCAL OCCURRENCE. CSUN Iocs. 1444,
1445, 1446, 1450, 1451, 1453, and scattered co-
quinas.

REMARKS. Specimens are extremely abundant
at localities 1444, 1446, and 1450. At the first two
localities, specimens range from 6 to 24 mm in
height. At locality 1450, they range from 1.5 to 20
mm in height. Preservation at these three localities
is excellent to good, although there are some inter-
nal molds. Many specimens are single valves, but
articulated specimens are common. At the other
main localities, Pelecyora aequilateralis is common
to rare. Coquinas consisting almost entirely of this
species are, however, scattered throughout the sec-
tion. One of the best examples is the coquina bed
immediately below locality 1453.

The specimens of Pelecyora aequilateralis in the
study area show variation in overall shape of the
valves, as well as in the strength of the concentric
ribbing, but this species is characterized by such
variation (Hanna, 1927; Turner, 1938; Vokes,
1939).

Superfamily Tellinoidea Blainville, 1814
Family Tellinidae Blainville, 1814
Genus Tellina Linnaeus, 1758

TYPE SPECIES. Tellina radiata Linnaeus, 1758,
by subsequent designation, Children, 1823; Recent,
Caribbean.

“ Tellina ” joaquinensis Arnold, 1909

Figures 42-43

Tellina joaquinensis Arnold, 1909:49, pi. 2, fig. 11;

Arnold and Anderson, 1910:70, pi. 24, fig. 11.

Vokes, 1939:90, pi. 14, figs. 15, 19, 20.

Tellina sp. Jestes, 1963:222 (in part).

Gari sp. Jestes, 1963:226.

Tellina (?) sp. Jestes, 1963:225.

PRIMARY TYPE MATERIAL. USNM holotype
165619, Domengine Formation, Fresno County,
central California, USGS loc. 4801.

ILLUSTRATED SPECIMENS. LACMIP hypo-
types 12471 and 12472.

MOLLUSCAN STAGE RANGE. “Domengine”
through “Transition.”

GEOGRAPHIC DISTRIBUTION. Matilija Hot

22 ■ Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

Figures 40-49. Bivalves from Matilija Hot Springs area upper part of Matilija Sandstone. All specimens coated with
ammonium chloride. 40-41. Pelecyora aequilateralis (Gabb, 1869), CSUN 1450. 40. Left valve, X2.7, LACMIP hypotype
12469. 41. Right valve, X2.8, LACMIP hypotype 12470. 42-43. “Tellina” joaquinensis Arnold, 1909. 42. Left valve,
X3, LACMIP hypotype 12471, CSUN loc.1445. 43. Right valve, X2, LACMIP hypotype 12472, CSUN loc. 1445. 44-48.
“Tb//i«tf” domenginensis Vokes, 1939. 44. Left valve, Xl.5, LACMIP hypotype 12473, CSUN loc. 1445. 45. Left valve
showing concentric ribs near umbo, posterior margin incomplete, X2.5, LACMIP hypotype 12474, CSUN loc. 1450. 46.
Right valve, internal mold showing pallial sinus, X2.2, LACMIP hypotype 12475. 47. Latex peel of internal mold of
left-valve hinge (incomplete), X3.6, LACMIP hypotype 12476, CSUN loc. 1450. 48. Right-valve hinge (incomplete), X6,
LACMIP hypotype 12477, CUN loc. 1446. 49. Cuneocorbula torreyensis (Hanna, 1927), left valve, X5.1, LACMIP
hypotype 12478, CSUN loc. 1445.

Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks ■ 23

Springs, southern California; Coalmine Canyon,
central California; and possibly Middle Fork of the
Coquille River, southwestern Oregon.

LOCAL OCCURRENCE. CSUN Iocs. 1444,
1445, 1446, 1450, 1451, 1452, 1453.

REMARKS. This species is the only mollusk
found at all of the six main localities in the restrict-
ed-coastal facies at Matilija Hot Springs. Specimens
are abundant at localities 1445 and 1450 and com-
mon to uncommon at the other localities. At local-
ity 1445, specimens range from 10 to 20 mm in
height, and at locality 1450 they range from 13 to
22 mm in height. Preservation is usually poor.
Nearly every specimen has been crushed. Articulat-
ed specimens are plentiful, especially at locality
1445. A specimen at this latter locality has a naticid
borehole.

Arnold (1909) was somewhat contradictory in
his description of this species. He stated that it is
inequilateral, but in an accompanying paragraph he
reported that it has approximate bilateral symme-
try. The Matilija Hot Springs specimens confirm the
latter. In fact, on some specimens, this bilateral
symmetry makes it difficult to ascertain which is the
left valve and which is the right valve.

I use quotation marks for the generic assignment
of this tellinid. The species cannot be assigned with
certainty to Tellina because the hinge is not known.

“ Tellina ” joaquinensis differs from “T.” domen-
ginensis by having a much less elongate shape, less
inflated and thinner valves, an absence of concen-
tric ribs near the umbones, and a rounded rather
than a pointed posterior.

Turner (1938:61, pi. 7, fig. 9) tentatively reported
(as Tellina cf. joaquinensis) a single, poorly pre-
served specimen of this species in Eocene rocks
along the Middle Fork Coquille River, Coos Coun-
ty, southwestern Oregon.

The presence of “ Tellina ” joaquinensis at Matil-
ija Hot Springs is the youngest and southernmost
record of this species.

“ Tellina ” domenginensis Vokes, 1939

Figures 44-48

Tellina domenginensis Vokes, 1939:91, pi. 14, figs.

12, 14, 16, 18.

Tellina sp. Jestes, 1963:222 (in part).

Gari sp. Jestes, 1963:223.

PRIMARY TYPE MATERIAL. UCMP holotype
15694, UCMP loc. 3315; UCMP paratypes 15695,
15696, both from UCMP loc. A-975; UCMP par-
atype 15697, UCMP loc. A-1220; all from the Do-
mengine Formation, Fresno County, central Cali-
fornia.

ILLUSTRATED SPECIMENS. LACMIP hypo-
types 12473 to 12477.

MOLLUSCAN STAGE RANGE. “Domengine”
through “Transition.”

GEOGRAPHIC DISTRIBUTION. Matilija Hot
Springs, southern California, and between Oil City

and Domengine Creek on west side of San Joaquin
Valley, north of Coalinga, central California.

LOCAL OCCURRENCE. CSUN Iocs. 1444,
1445, 1450.

REMARKS. Specimens are common only at lo-
cality 1450, where they range from 12 to 20 mm
high. A few specimens are articulated. Specimens
are rare at the other two localities. Study area spec-
imens reveal new information about this species’
morphology, both exterior and interior. Fine con-
centric ribs are present near the umbones, although
Vokes (1939) reported that the surface ornamen-
tation consists only of coarse concentric growth
lines. The pallial sinus of the right valve is revealed
for the first time. It is broadly rounded, the dorsal
margin ascends posteriorly, and the anterior margin
is not close to the anterior adductor scar (Fig. 46).
The relationship between the ventral margin of the
right-valve pallial sinus and the pallial line is not
clear, but they seem to be coalescent. The shape of
the left-valve pallial sinus remains unknown. The
cardinal teeth of both valves are also revealed for
the first time. The left-valve cardinal teeth are di-
vergent, with the anterior tooth strong and the pos-
terior tooth very thin and lamellar (Fig. 47). The
cardinal teeth on the right valve are both strong
(Fig. 48), but whether or not these teeth are bifid
cannot be determined. Unfortunately, poor preser-
vation prevents study of the rest of the hingeline,
where lateral teeth might be present. Until the pres-
ence of lateral teeth is confirmed, the species cannot
be assigned with certainty to Tellina.

Among the tellinids, “Tellina” domenginensis has
some important similarities to the extant Peronidia
albicans (Gmelin, 1791; Afshar, 1969:84, pi. 35,
figs. 1-5), which is the type species of Peronidia
Dali, 1900b. Both have the following features: an
ovate-trigonal shape with nearly central umbones;
the posterior cardinal of the left valve is thin, la-
mellar, and mostly fused with the nymph; the pos-
terior cardinal of the right valve is larger than the
anterior one; and the pallial sinus is large and does
not touch the anterior adductor scar.

There is also a close resemblance between “ Tel-
lina” domenginensis and the flat, almost equilateral
Peronidia nysti (Deshayes, 1860; Baldi, 1973:225-
226, pi. 21, figs. 1-2, 4) from upper Oligocene
rocks in Hungary. Only the external features are
known for P. nysti, and “T.” domenginensis differs
from it by having concentric ribs near the umbones
and a posterior margin that is rounded rather than
slightly angular.

Better specimens of “Tellina” domenginensis are
needed to determine whether or not this species is
related to Peronidia. Additionally, there has been
little agreement on the taxonomic position of Per-
onidia. It has been regarded as a subgenus of Tel-
lina by some workers (Keen, 1969b; Coan, 1971;
Coan and Scott, 1997); as a subgenus of Macoma
by Afshar (1969); and as a subgenus of Angulus by
Baldi (1973).

The presence of “ Tellina ” domenginensis at Ma-

24 ■ Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

tilija Hot Springs is the youngest and the southern-
most record of this species.

Subclass Asthenodonta Dali, 1895

Order Myoida Goldfuss, 1820

Suborder Myina Goldfuss, 1820

Superfamily Myoidea Lamarck, 1809

Family Corbulidae Lamarck, 1818

Genus Cuneocorbula Cossmann, 1886

TYPE SPECIES. Corbula pelseneeri Glibert and
van de Poel, 1966 [— Corbula biangulata Deshay-
es, 1861]; upper Paleocene (Thanetian Stage), Paris
Basin, France.

Cuneocorbula torreyensis (Hanna, 1927)
Figure 49

Corbula torreyensis Hanna, 1927:296-297, pi. 44,
figs. 6-10, 15-16; Clark and Yokes, 1936:875,
figs. 9, 11; Turner, 1938:66, pi. 8, figs. 6, 7;
Weaver 1942 [1943]:259-260, pi. 61, fig. 12.
Cuneocorbula torreyensis (Hanna). Vokes, 1939:
101-102, ph 16, figs. 16, 20, 21; Jestes, 1963:
222, 225; Givens and Kennedy, 1979:table 2.
Corbula { Cuneocorbula ) torreyensis Hanna. Giv-
ens, 1974:58.

PRIMARY TYPE MATERIAL. UCMP holotype
31115; UCMP paratypes 31116-31119; all from
Delmar Formation, San Diego area, San Diego
County, southern California, UCMP loc. 3981.

ILLUSTRATED SPECIMEN. LACMIP hypo-
type 12478.

MOLLUSCAN STAGE RANGE. “Domengine”
to “Transition.”

GEOGRAPHIC RANGE. San Diego and Matil-
ija Hot Springs, southern California; Vallecitos Syn-
cline, central California; and Glide, southwestern
Oregon.

LOCAL OCCURRENCE. CSUN Iocs. 1445,
1452, and scattered coquinas.

REMARKS. Specimens are abundant at localities
1445 and 1452. At both localities, specimens range
from 5 to 6 mm in height, and preservation is good.
Specimens are unworn, unbroken, and mostly sin-
gle valves. Articulated specimens are very rare.
Throughout the section, specimens of Cuneocor-
bula torreyensis form coquinas consisting almost
entirely of compacted specimens of this species.

LOCALITIES

CSUN LOCALITIES

All are in the upper part of the Matilija Sandstone (lower
middle Eocene “Transition Stage”) in the NE 1/4 of the
SE 1/4 of section 29, T 5 N, R 23 W, USGS topographic
quadrangle, Matilija, California, 7.5 minute, 1952 (pho-
torevised 1967), 1:24,000.

1444. [= LACMIP 16961]. Roadcut on north side of a
short, paved road that leads from Highway 33 to Matilija

Contributions In Science, Number 480

Hot Springs, about 48 m (157 ft.) west of sharp bend in
this road, near bottom of restricted-coastal facies.

1445. [= LACMIP 24259]. Roadcut on north side of a
short, paved road that leads from Highway 33 to Matilija
Hot Springs, about 26 m (85 ft.) west of sharp bend in
this road.

1446. [= LACMIP 24258]. Roadcut on north side of a
short, paved road that leads from Highway 33 to Matilija
Hot Springs, about 12 m (39 ft.) west of sharp bend in
this road.

1450. [= LACMIP 16963]. On south bank of North
Fork of Matilija Creek, near bottom of restricted-coastal
rocks, about 260 m (852 ft.) west of junction of Highway
33 and a short, paved road that leads to Matilija Hot
Springs.

1451. [= LACMIP 16964]. On south bank of North
Fork of Matilija Creek, about 250 m (820 ft.) west of
junction of Highway 33 and a short, paved road that leads
to Matilija Hot Springs.

1452. [= LACMIP 16965]. On south bank of North
Fork of Matilija Creek, about 230 m (754 ft.) west of
junction of Highway 33 and a short, paved road that leads
to Matilija Hot Springs.

1453. [= LACMIP 16962]. Roadcut on north side of a
short, paved road that leads from Highway 33 to Matilija
Hot Springs, about 43 m (141 ft.) west of sharp bend in
this road.

LACMIP LOCALITY7

7226. In the vicinity of Beartrap Creek, just east of hill
4560 along an unmaintained trail and downslope for
about 15 m (49ft) from trail, at section line between sec-
tions 24 and 25, T 7 N, R 23 W, USGS topographic quad-
rangle, Reyes Peak, California, 7.5 minute, 1943, 1:
24,000.

UCMP LOCALITIES

672. SE 1/4 of the NW 1/4 of section 24, T 18 S, R 14
E, USGS topographic quadrangle, Joaquin Rocks, Cali-
fornia, 7.5 minute, 1969, 1:24,000.

3315. Base of Domengine Formation, immediately
south of Domengine Creek, USGS topographic quadran-
gle, Domengine Ranch, California, 7.5 minute, 1956
(photorevised 1979), 1:24,000.

3981. At 15 m (50 ft.) above high-tide level in a small
gulley .4 km (.24 mi.) south of mouth of Soledad Valley,
USGS topographic quadrangle, Del Mar, California, 7.5
minute, 1967, 1:24,000.

3992. In sea cliff about .8 km (.5 mi.) south of the
mouth of Soledad Valley at high-tide level, USGS topo-
graphic quadrangle, Del Mar, California, 7.5 minute,
1967, 1:24,000.

5084. At 2.55 inches due north of top of the “S” of
Soledad Mountain, in sea cliff, elevation 3 m (10 ft.), fos-
sils in the conglomerate above the mudstone, USGS to-
pographic quadrangle, La Jolla, California, 7.5 minute,
1967, 1:24,000.

A-975. Second “reef” above base of Domengine For-
mation in draw across ridge to south of Big Tar Canyon,
USGS topographic quadrangle, Garza Peak, California,
7.5 minute, 1953, 1:24,000.

A- 1220. At base of Domengine Formation in small
draw cutting long ridge, 53 m (175 ft.) north of line be-
tween sections 9 and 16, T 19 S, R 15 E, USGS topo-
graphic quadrangle, Domengine Ranch, California, 7.5
minute, 1956 (photorevised 1979), 1:24,000.

A-1297. From sandstone cliff on northeast bank of

Squires: Brackish-Marine Eocene Mollusks ■ 25

Pleasants Creek opposite Brink ranch house about 1.2 km
(.75 mi.) east of bench mark 258, and 3.2 km (2 mi.)
south of Putah Creek, USGS topographic quadrangle, Mt.
Vaca, California, 7.5 minute, 1951 (photorevised 1968),
1:24,000.

UCR LOCALITY

4747. Just east of elevation 4072 on ridge south of
mouth of Beartrap Creek, 594 m (1948 ft.) north and 457
m (1498 ft.) east of southwest corner of section 23, T 7
N, R 23 W, USGS topographic quadrangle, Reyes Peak,
California, 7.5 minute, 1943, 1:24,000.

USGS LOCALITY

4801. About 4.8 km (3 mi.) northwest of Coalinga, at
Coalmine Canyon in NW 1/4 of section 26, T 20 S, R 14
E, USGS topographic quadrangle, Alcalde Hills, Califor-
nia, 7.5 minute, 1969, 1:24,000.

ACKNOWLEDGMENTS

Lindsey T. Groves (LACM and LACMIP) provided access
to the collections and obtained some literature. James H.
McLean (LACM) shared his knowledge of cerithioid gas-
tropods. LouElla R. Saul (LACMIP) shared her knowledge
of bivalves. Eugene V. Coan (Department of Invertebrate
Zoology, California Academy of Sciences, San Francisco)
shared his knowledge of literature on trapeziid bivalves.
Louie Marincovich (Department of Geology, California
Academy of Sciences, San Francisco) provided an English
translation of Devyatilova and Volobueva (1981). Gian
Carlo Shammas (CSUN) provided important details about
the measured sections. Daniel Geiger (University of South-
ern California) assisted in translating German. Jean
DeMouthe (California Academy of Sciences, San Francis-
co) and Karen Grycewicz (UCMP) loaned type specimens.
The manuscript benefited from reviews by LouElla R. Saul
and Charles R. Givens (Department of Physical Sciences,
Nicholls State University, Thibodaux, Louisiana.

LITERATURE CITED

Adams, H., and A. Adams. 1853-1858. The genera of
Recent Mollusca ; arranged according to their orga-
nization, 2 vols. London: John van Vorst, 660 pp.
Afshar, F. 1969. Taxonomic revision of the superspecific
groups of the Cretaceous and Cenozoic Tellinidae.
The Geological Society of America Memoir 1 19:1—
215.

Aldrich, T.H. 1886. Preliminary report on the Tertiary fos-
sils of Alabama and Mississippi, Part 1. Geological
Survey of Alabama Bulletin 1:1-85.

Anderson, F.M., and G.D. Hanna. 1925. Fauna and strati-
graphic relations of the Tejon Eocene at the type lo-
cality in Kern County, California. California Acad-
emy of Sciences, Occasional Papers 11:1-249.
Arnold, R. 1909. Paleontology of the Coalinga district. U.

S. Geological Survey Bulletin 396:1-173.

Arnold, R., and R. Anderson. 1910. Geology and oil re-
sources of the Coalinga district, California. U. S.
Geological Survey Bulletin 398:1-354.

Baldi, T. 1973. Mollusc fauna of the Hungarian upper
Oligocene (Egerian), studies in stratigraphy, palaeoe-
cology, palaeogeography and systematics. Budapest:
Akademiai Kiado, 511 pp.

Bandel, K., and F. Riedel. 1994. The Late Cretaceous gas-
tropod fauna from Ajka (Bakony Mountains, Hun-

gary): A revision. Annalen des Naturhistorischen
Museums in Wien 96A:l-65.

Berggren, W.A., D.V. Kent, C.C. Swisher, III, and M.-P.
Aubry. 1995. A revised Cenozoic geochronology and
chronostratigraphy. In Geochronology, time scales,
and global stratigrapic correlation, eds. W.A. Berg-
gren, D.V. Kent, M.-P. Aubry, and J. Hardenbol. Tul-
sa, Oklahoma: SEPM (Society for Sedimentary Ge-
ology) Special Publication 54.

Blainville, H.M.D. 1814. Memoire sur la classification
methodique des animaux mollusques, et etablisse-
ment d’une nouvelle consideration pour y parvenir.
Societe Philomathique de Paris, Bulletin, 2nd serie,
4(2):175— 1 80.

Boggs, S., Jr. 1987. Principles of sedimentology and stra-
tigraphy. Columbus, Ohio: Merrill Publishing Com-
pany, 784 pp.

Bonnot, P. 1935. A recent introduction of exotic species
of mollusks into California waters from Japan. Nau-
tilus 49(1): 1-2.

Bouchet, P. 1977. Distribution des mollusques dan les
mangroves du Senegal. Malacologia 16(l):67-74.

Bowdich, T.E. 1822. Elements of conchology, including
the fossil genera and animals, 2 vols. Paris-London,
119 pp.

Brandt, R.A.M. 1974. The non-marine aquatic Mollusca
of Thailand. Archiv fiir Molluskende 105(1— 4):1—
423, pis. 1-29.

Brongniart, A. 1810. Sur les terrains qui paroissent avoir
formes sous l’eau douce. Annales du Museum Na-
tional d’Histoire Naturelle de Paris 15:357-405.

Brown, D.S. 1980. Freshwater snails of Africa and their
medical importance. London: Taylor and Francis,
487 pp.

Chavan, A. 1952. Quelques interessants types de cerithes.
Cahiers Geologiques de Thoiry 15:103-104, 113,
114, 128.

Children, J.G. 1822-1823. Lamarck’s genera of shells,
translated from the French, with plates from original
drawings by Miss Anna Children. Quarterly Journal
of Science 14:64-86, 298-322 [1822]; 15:23-52,
216-258 [1823]; 16:49-79, 241-264 [1823].

Clark, B.L. 1929. Stratigraphy and faunal horizons of the
Coast Ranges of California, with illustrations of in-
dex fossils of Tertiary horizons. Privately Published,
50 pp.

Clark, B.L. 1938. Fauna from the Markley Formation
(upper Eocene) on Pleasant Creek, California. Geo-
logical Society of America Bulletin 49:683-730.

Clark, B.L., and H.E. Vokes. 1936. Summary of the ma-
rine Eocene sequence of western North America.
Geological Society of America Bulletin 47(6):851-
878.

Clark, B.L., and A.O. Woodford. 1927. The geology and
paleontology of the type section of the Meganos for-
mation (lower middle Eocene) of California. Univer-
sity of California Publications, Bulletin of the De-
partment of Geological Sciences 17(2):63-142.

Coan, E.V. 1971. The northwest American Tellinidae. The
Veliger 14(suppl):l-63.

Coan, E.V., and P.H. Scott. 1997. Checklist of the marine
bivalves of the northeastern Pacific Ocean. Santa
Barbara Museum of Natural History, Contributions
in Science 1:1-128.

Cooper, J.G. 1894. Catalogue of Californian fossils, parts
2-5. California State Mining Bureau, Bulletin 4:5-
65.

Cossmann, A.E.M. 1886. Catalogue illustre des coquilles

26 ■ Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

fossiles de 1; Eocene des environs de Paris. Premier
fascicule. Annales de la Societe Roy ale Malacolo-
gique de Belgique 21:17-186.

Cossmann, A.E.M. 1888. Catalogue illustre des coquilles
fossiles de P Eocene des environs de Paris. Troisieme
fascicule. Annales de la Societe Royale Malacolo-
gique de Belgique 23:3-324.

Cossmann, A.E.M. , and G. Pissarro. 1904-1913. Icono-
graphie complete des coquilles fossiles de (’Eocene
des environs de Paris. Paris: H. Bouillant, [Vol. 1
(1904-1906), Pelecypodes, 45 pis.; Vol. 2 (191 0—
1913), Gastropodes, etc.].

Dailey, D.H., and W.P. Popenoe. 1966. Mollusca from the
Upper Cretaceous Jalama Formation, Santa Barbara
County, California. University of California Publi-
cations in Geological Sciences 65:1-27.

Dali, W.H. 1900a. Contributions to the Tertiary fauna of
Florida with especial reference to the silex beds of
Tampa and the Pliocene beds of the Caioosahatchie
River, including in many cases a complete revision
of the generic groups treated of and their American
Tertiary species. Part 5. Teleodesmacea: Solen to Di-
plodonta. Wagner Free Institute of Science of Phil-
adelphia, Transactions 3(pt. 5):949— 121 8.

Dali, W.H. 1900b. Synopsis of the family Tellinidae and
of the North American species. Proceedings of the
U. S. National Museum 23(1210):285-326.

Dali, W.H. 1902. Synopsis of the family Veneridae and of
the North American Recent species. Proceedings of
the United States National Museum 26(131 3 ):335-
412.

Davies, A.M. and F.E. Eames. 1971. Tertiary faunas. A
text-book for oilfield paleontologists and students of
geology. Vol. 1. The composition of Tertiary faunas.
London: George Allen and Unwin, 571 pp.

Deshayes, G.-P. 1824-1837. Description des coquille fos-
siles des environs de Paris, 2 vols. (text). Paris: EG.
Levrault, 1172 pp.

Deshayes, G.-P. 1856-1866. Description des animaux san
vertebres decouveris dans bassin de Paris, 3 vols.
(text), 2 vols. (atlas). Paris: J.-B. Bailliere et fils, 2536

pp.

DeVyatilova, A.D., and V.I. Volobueva. 1981. Atlas of Pa-
leogene and Neogene fauna of the northeast USSR.
Central Combined Thematic Expedition of the
Northeast Industrial Geologicial Society, 219 pp. [In
Russian].

Dickerson, R.E. 1914. The fauna of the Siphonalia sutter-
ensis Zone in the Roseburg quadrangle, Oregon.
Proceedings of the California Academy of Sciences,
series 4, 4:113-128.

Dickerson, R.E. 1915. Fauna of the type Tejon: Its relation
to the Cowlitz phase of the Tejon Group of Wash-
ington. Proceedings of the California Academy of
Sciences, series 4, 5(3):33-98.

Dickerson, R.E. 1916. Stratigraphy and fauna of the Tejon
Eocene of California. University of California Pub-
lications, Bulletin of the Department of Geological
Sciences 9(17):363-524.

Dickerson, R.E. 1917. Climate and its influence upon the
Oligocene faunas of the Pacific coast, with descrip-
tions of some new species from the Molopophorus
lincolnensis zone. Proceedings of the California
Academy of Sciences, series 4, 7(6):157— 192.

Doncieux, L. 1908. Catalogue descriptif des fossiles num-
mulitiques de PAude et de PHerault. Deuxieme par-
tie (fascicule 1). Corbieres septentrionales. Annales

Contributions in Science, Number 480

de I’Universite de Lyon, nouvelle serie, I. Sciences,
Medecine, fascicule 22:1-288.

Ferussac, J.B.L. 1807. Essai d’une methode conchyliolo-
gique appliquee aux mollusques fluviatiles et terres-
tres. Paris.

Flynn, J.J., R.M. Cipolletti, and M.J. Novacek. 1989.
Chronology of early Eocene marine and terrestrial
strata, Baja California, Mexico. Geological Society
of America Bulletin 101:1182-1196.

Forbes, E. 1838. Malacologia Monensis. A catalogue of
the Mollusca inhabiting the Isle of Man and the
neighboring sea. Edinburgh, 63 pp.

Gabb, W.M. 1864. Description of the Cretaceous fossils.
California Geological Survey, Palaeontology 1:57-
243.

Gabb, W.M. 1868. An attempt at a revision of the two
families Strombidae and Aporrhaidae. American
Journal of Conchology 4:137-149.

Gabb, W.M. 1869. Cretaceous and Tertiary fossils. Cali-
fornia Geological Survey, Palaeontology 2:1-299.

Gitton, j.L., P. Lozouet, and P. Maestrati. 1986. Biostra-
tigraphie et paleoecologie des gisements types du
Stampien de la region d’Etampes (Essone). Bureau
de recherches Geologiques et Minieres 1:1-101.

Givens, C.R. 1974. Eocene molluscan biostratigraphy of
the Pine Mountain area, Ventura County, California.
University of California Publications in Geological
Sciences 109:1-107.

Givens, C.R., and M.P. Kennedy. 1976. Middle Eocene
mollusks from northern San Diego County, Califor-
nia. Journal of Paleontology 50(5):954-975.

Givens, C.R., and M.P. Kennedy. 1979. Eocene molluscan
stages and their correlation, San Diego area, Cali-
fornia. In Eocene depositional systems, San Diego
ed. P. L. Abbott, 81-95. Los Angeles: Pacific Section,
Society of Economic Paleontologists and Mineralo-
gists, Field Trip Guide.

Glibert, M., and L. van de Poel. 1966. Les Bivalvia fossiles
du Cenozoi'que etranger des collections de Plnstitut
Royal des Sciences Naturelles de Belgique. III. Het-
eroconchia. Pt. 1: Laternulidae a Chamidae. Institut
Royal des Sciences Naturelles de Belgique, Memo-
ires, Deuxieme serie, fascicule 81:1-82.

Gmelin, J.F. 1791. Caroli a Linnaeus Sy sterna naturae per
regna tria naturae. Editio decima tertia. Vol. 1, pt.
6. Lipsiae, 888 pp.

Gray, J.E. 1 847. A list of the genera of Recent Mollusca,
their synonyms and types. Proceedings of the Zoo-
logical Society of London 15:129-219.

Habe, T. 1951. Genera of Japanese shells. Pelecypoda (2).
Kyoto, 89 pp.

Hanna, G.D. 1966. Introduced mollusks of western North
America. California Academy of Sciences, Occasion-
al Papers 48:1-108.

Hanna, M.A. 1927. An Eocene invertebrate fauna from
the La Jolla quadrangle, California. University of
California, Publications in Geological Sciences
16(8):247— 398.

Harris, G.D. 1919. Pelecypoda of the St. Maurice and
Claiborne stages. Bulletins of American Paleontolo-
gy 6(31):1— 268.

Hoagland, K.E. 1977. Systematic review of fossil and re-
cent Crepidula and discussion of evolution of the
Calyptraeidae. Malacologia 16(2):353-420.

Houbrick, R.S. 1991. Anatomy and systematic placement
of Faunus Montfort, 1810 (Prosobranchia: Melan-
opsinae). Malacological Review 24:35-54.

Jestes, E.C. 1963. A stratigraphic study of some Eocene

Squires: Brackish-Marine Eocene Mollusks I 27

sandstones, northeastern Ventura basin, California.
University of California, Los Angeles, unpublished
Ph.D. Dissertation, 253 pp.

Keen, A.M. 1969a. Superfamily Arcticacea Newton,
1891. In Treatise on invertebrate paleontology, part
N, Bivalvia 2 (of 3), ed. R. C. Moore, 644-657.
Lawrence: University of Kansas Press.

Keen, A.M. 1969b. Superfamily Tellinacea de Blainville,
1814. In Treatise on invertebrate paleontology, part
N, Bivalvia 2 (of 3), ed. R. C. Moore, 613-643.
Lawrence: University of Kansas Press.

Keen, A.M., and R. Casey. 1969. Family Corbiculidae. In
Treatise on invertebrate paleontology, part N, Biv-
alvia 2 (of 3), ed. R. C. Moore, 665-669. Lawrence:
University of Kansas Press.

Kerr, P.F., and H.G. Schenck. 1928. Significance of the
Matilija overturn. Geological Society of America
Bulletin 39:1087-1102.

Kidwell, S.M., F.T. Fiirisch, and T. Aigner. 1986. Concep-
tual framework for the analysis and classification of
fossil concentrations. Palaios 1:228-238.

Kira, T. 1965. Shells of the western Pacific in color. Vol.
1. Osaka, Japan: Hoikusha Publishing Co., 224 pp.

Kleinpell, R.M., and D.W. Weaver. 1963. Oligocene bio-
stratigraphy of the Santa Barbara embayment, Cal-
ifornia. Part 2. Mollusca from the Turritella variata
zone. University of California, Publications in Geo-
logical Sciences 43:81-118.

Lamarck, J.B. 1799. D’une nouvelle classification des co-
quilles. Memoir es de la Societe d’Histoire Naturelle
de Paris 1(1):63-91.

Lamarck, J.B. 1804. Memoire sur les fossiles des environs
de Paris. Annales du Museum National d’Histoire
Naturelle 3:266-274.

Lamarck, J.B. 1809. Philosophie zoologique, ou exposi-
tion des considerations relatives a I’histoire naturelle
des animaux, 2 vols. Paris, 895 pp.

Lamarck, J.B. 1818-1819. Histoire naturelle des animaux
sans vertebres. Paris: Verdiere, Deterville, &c Chez
l’auteur, vol. 5, 612 pp.; vol. 6, no.l, 343 pp.

Lamy, E. 1920. Revision des Cypricardiacea et des Iso-
cardiacea vivants du Museum d’Histoire Naturelle
de Paris. Journal de Conchyliologie 64(4):259-307.

Le Renard, J. and J.-M. Pacaud. 1995. Revision des moll-
usques Paleogenes du bassin de Paris. II — Liste des
references primaires des especes. Cossmanniana 3(3):
65-132.

Lindberg, D.R., and R.L. Squires. 1990. Patellogastropods
(Mollusca) from the Eocene Tejon Formation of
southern California. Journal of Paleontology 64(4):
578-587, figs. 1-9.

Link, M.H. 1975. Matilija Sandstone: A transition from
deep-water turbidite to shallow-marine deposition in
the Eocene of California. Journal of Sedimentary Pe-
trology 45(1) :63-7 8.

Link, M.H., and J.E. Welton. 1982. Sedimentology and
reservoir potential of Matilija Sandstone: An Eocene
sand-rich deep-sea fan and shallow-marine complex,
California. American Association of Petroleum Ge-
ologists Bulletin 66( 10): 1514— 1534.

Linnaeus, C. 1758. Sy sterna naturae per regna tria natu-
rae. Editio decima, reformata. Vol. 1, Regnum ani-
male. Stockholm: Laurentii Salvii, 824 pp.

Linnaeus, C. 1767. Sy sterna naturae per regna tria natu-
rae. Editio duodecima, reformata. Vol. 1, pt. 2, “Ver-
mes Testacea.” Stockholm: Laurentii Salvii, 794 pp.

Maestrati, P., and P. Lozouet. 1995. Revision des genres

de la famille Trapeziidae (Mollusca, Bivalvia) du
Cenozoi'que. Geobios 28(2):185-197.

Marincovich, L., Jr. 1977. Cenozoic Naticidae (Mollusca:
Gastropoda) of the northeastern Pacific. Bulletins of
American Paleontology 70(294): 169-494.

Meek, F.B. 1877. Paleontology. U. S. Geological Explo-
ration 40th Parallel Report 4(pt. 1): 1-197, pis. 1-
17.

Mergele von Miihlfeld, J.K. 1811. Entwurf eines neuen
System’s der Schalthiergehause. Gesellschaft Natur-
forschender Freunde Magazin, (Berlin) 5:38-72.

Montfort, P.D. 1810. Conchyliologie systematique et clas-
sification methodique de coquilles. Coquilles uni-
valves, vol. 2. Paris: F. Schoell, 676 pp.

Moore, E.J. 1968. Fossil mollusks of San Diego County.
San Diego Society of Natural History Occasional
Paper 15:1-76.

Moore, E.J. 1983. Tertiary marine pelecypods of Califor-
nia and Baja California: Nuculidae through Mallei-
dae. U. S. Geological Survey Professional Paper
1228-A:1-108.

Moore, E.J. 1987. Tertiary marine pelecypods of Califor-
nia and Baja California: Plicatulidae to Ostreidae. U.
S. Geological Survey Professional Paper 1228-G1-
53.

Morton, B. 1979. Some aspects of the biology and func-
tional morphology of Trapezium ( Neotrapezium )
sublaevigatum (Lamarck) (Bivalvia: Arcticacea). Pa-
cific Science 33(2):177-194.

Morton, B., and J. Morton. 1983. The sea shore ecology
of Hong Kong. Hong Kong: Hong Kong University
Press, 350 pp.

Muller, O.F. 1774. Vermium terrestrium et fluviatilium,
seu animalium infusoriorum, helminth icorum et tes-
tascerorum, non marinorum, succinta historia. 2
Vols. Havniae: Heineck and Faber, 421 pp.

Munier-Chalmas, E. 1900. In P. J. Chedeville, Liste gener-
ale, et synonymique des fossiles Tertiaires du Bassin
de Paris. Bulletin de la Societe d’ Etude des Sciences
Naturelles d’Elbeuf 18-19:1-226.

Nesbitt, E.A. 1995. Paleoecological analysis of molluscan
assemblages from the middle Eocene Cowlitz For-
mation, southwestern Washington. Journal of Pale-
ontology 69(6):1060— 1073.

Newton, R.B. 1891. Systematic list of the F. E. Edwards
collection of British Oligocene and Eocene Mollusca
in the British Museum (Natural History), with ref-
erences to the type-specimens from similar horizons
contained in other collections belonging to the Geo-
logical Department of the Museum. London: British
Museum (Natural History), 365 pp.

Niem, A.R., I.-C. Ryu, and W.A. Niem. 1992. Geologic
interpretation of the schematic fence diagram of the
southern Tyee basin, Oregon Coast Range. State of
Oregon Department of Geology and Mineral Indus-
tries, Oil and Gas Investigation 18:1-40.

Nilsson, S. 1827. Petrificata suecana formationis Creta-
ceae, descripta et iconibus illustrata, vol. 1. London:
Gothorum, 39 pp.

Olivier, G.A. 1804. Voyage dans V empire Ottoman,
I’Egypte et la Perse, vol. 2. Paris: H. Agasse.

Packard, E.L. 1923. An aberrant oyster from the Oregon
Eocene: University of Oregon Publication 2(4):3— 14.

Plaziat, J.-C. 1977. Les cerithides tropicaux et leur poly-
morphisme lie a l’ecologie littorale des mangroves.
Malacologia 16( 1 ):35— 44.

Ponder, W.F., and A. Waren. 1988. Appendix. Classifica-
tion of the Caenogastropoda and Heterostropha — a

28 ■ Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks

list of the family-group names and higher taxa. Mal-
acoiogical Review (supplement 4):288-326.

Rafinesque, C.S. 1815. Analyse de la nature, ou tableau
de V waivers et des corps organises. Palermo, 224 pp.

Reinhart, P.W. 1943. Mesozoic and Cenozoic Arcidae
from the Pacific slope of North America. Geological
Society of America Special Papers 47:1-117.

Reeve, L.A. 1843. Monograph of the genus Cypricardia.
Conchologia Iconia; or, illustrations of the shells of
molluscous animals, vol. 1. London: L. Reeve and
Company 122 pis.

Risso, A. 1826. Histoire naturelle des principals produc-
tions de l’ Europe meridionale et particular ement de
cedes des environs de Nice et des Alpes maritimes,
vol. 4. Paris: R G. Levrault, 439 pp.

Roding, P.R 1798. Museum Boltenianum sive catalogus
cimeliorum e tribus regnis nature quae olim code-
gerat Joa. Fried Bolten. Pars Secunda. Hamburg: Jo-
han. Christi Trappii., 199 pp. [Reprinted 1906 by C.
D. Sherborn and E. R. Sykes, 1986 by American
Malacologies! Union].

Saul, L.R., and R.L. Squires. 1998. New Cretaceous Gas-
tropoda from California. Palaeontology 41 (3 1:46 1 -
488.

Schenck, H.G., and A.M. Keen. 1940. California fossils
for the field geologist. Stanford, California: Stanford
University Press, 86 pp.

Schumacher, C.F. 1817. Essai d’un nouveau systeme des
habitations des vers testaces. Copenhague.

Solem, A. 1954. Living species of the pelecypod family
Trapeziidae. Proceedings of the Malacological Soci-
ety of London 31 (2): 64-82.

Squires, R.L. 1984. Megapaleontology of the Eocene I. la -
fas Formation, Simi Valley, California. Natural His-
tory Museum of Los Angeles County, Contributions
in Science 350:1-76.

Squires, R.L. 1987. Eocene molluscan paleontology of the
Whitaker Peak area, Los Angeles and Ventura Coun-
ties, California. Natural History Museum of Los An-
geles County, Contributions in Science 388:1-93.

Squires, R.L. 1991a. A new middle Eocene potamidid gas-
tropod from brackish-marine depostis, southern Cal-
ifornia. The Veliger 34(4):354-359.

Squires, R.L. 1991b. Molluscan paleontology of the lower
Eocene Maniobra Formation, Orocopia Mountains,
southern California. In Eocene geologic history San
Diego region, ed. P. L. Abbott, vol. 68, 217-226. Los
Angeles: Pacific Section, Society of Economic Pale-
ontologists and Mineralogists.

Squires, R.L. 1992. New morphologic and geographic data
on the neritid gastropod Nerita ( Theliostyla ) triangu-
lata Gabb, 1869, from the Eocene of the Pacific coast
of North America. The Veliger 35(4):323~329.

Squires, R.L. 1998. New information on morphology,
stratigraphy, and paleoclimate implications of the
Eocene brackish-marine gastropod Loxotrema tur-
ritum Gabb, 1868, from the west coast of the United
States. The Veliger 41(4):297-313.

Squires, R.L., and R.A. Demetrion. 1992. Paleontology of
the Eocene Bateque Formation, Baja California Sur,
Mexico. Natural History Museum of Los Angeles
County, Contributions in Science 434:1-55.

Starmiihlner, F. 1969. Ergebnissse der osterreichischen Ma-
dajaskar Expedition 1958. Die Gastropoden der Mad-
agasischen Binnengewasser. Schale, Makro- und Mik-

roanatomie des Weichkorpers Lebensraum und geo-
graphische Verbeitung. Malacologia 8 (1-2): 1-434.

Stephenson, L.W. 1952. Larger invertebrate fossils of the
Woodbine Formation (Cenomanian) of Texas. U. S.
Geological Survey Professional Paper 242:1-22 6.

Stewart, R.B. 1927. Gabb’s California fossil type gastro-
pods. Proceedings of the Academy of Natural Sci-
ences of Philadelphia 78:287-447.

Stewart, R.B. 1930. Gabb’s California Cretaceous and Ter-
tiary type lamellibranchs. Academy of Natural Scienc-
es of Philadelphia, Special Publication 3:1-314.

Thiele, J. 1929-1935. Handbuch der Systematischen
Weichtierkunde. Teil 1-4. Jena: Gustave Fischer,
1154 pp. [In 1992, translated into English by Smith-
sonian Institution Libraries and the National Science
Foundation, Washington, D. C.].

Troschel, F.H. 1856-1863. Das Gebiss der Schnecken zur
Begrundung einer natiirlichen Classification, vol. 1.
Berlin, 661 pp.

Tryon, G.W., Jr. 1883. Structural and systematic conchol-
ogy: an introduction to the study of the Mollusca,
vol. 2. Philadelphia: Privately published, 430 pp.

Turner, F.E. 1938. Stratigraphic and Mollusca of the Eo-
cene of western Oregon. Geological Society of Amer-
ica Special Papers 10:1-130.

Vialov, O.S. 1936. Sur la classification des huitres. Aca-
demic des Sciences de I’URSS, Comptes Rendus
[Doklady], new series, 4 [13] (1[105]):17— 20.

Vignal, L. 1897. Note sur quelques coquilles de Cerithidae
de 1’ Eocene parisien. La Feuille des Jeunes Natural-
istes, serie 3, 27:183-184, 195-197.

Vokes, H.E. 1935. Notes on the variation and synonymy
of Ostrea idriaensis Gabb. University of California
Publications Bulletin of the Department of Geolog-
ical Sciences 23(9):291-304.

Vokes, H.E. 1939. Molluscan faunas of the Domengine
and Arroyo Hondo formations of the California Eo-
cene. Annals of the New York Academy of Sciences
38:1-246, pis. 1-22.

Vokes, H.E. 1980. Genera of the Bivalvia: A systematic
and bibliographic catalogue (revised and updated).
Ithaca, New York: Paleontological Research Insti-
tution, 307 pp.

Weaver, C.E. 1912. A preliminary report on the Tertiary
paleontology of western Washington. Washington
Geological Survey Bulletin 15:1-80.

Weaver, C.E. 1942 [1943]. Paleontology of the marine
Tertiary formations of Oregon and Washington.
University of Washington, Publications in Geology
5(1— 3):1— 789.

Weaver, C.E., and K.V.W. Palmer. 1922. Fauna from the
Eocene of Washington. University of Washington,
Publications in Geology 1 (3):1— 56.

Wenz, W. 1938-1944. Gastropoda. In Handbuch de pa-
laozoologie, vol. 6, in 7 parts, ed. O. H. Schinde-
wolf. Berlin: Gebriider Borntraeger, 1639 pp.

White, C.A. 1895. The Bear River formation and its char-
acteristic fauna. U. S. Geological Survey Bulletin
128:1-108.

Yen, T.-C. 1954. Nonmarine mollusks of Late Cretaceous
age from Wyoming, Utah and Colorado. U. S. Geo-
logical Survey Professional Paper 254-B:45-64.

Yen, T.-C. 1958. Systematics and distributions of Pyrgu-
lifera Meek. Annalen des Naturhistorischen Muse-
ums Wien 62:193-209, pis.

Submitted 22 July 1997; accepted 11 August 1999.

Contributions in Science, Number 480

Squires: Brackish-Marine Eocene Mollusks ■ 29

Natural History Museum

of Los Angeles County

900 Exposition Boulevard
Los Angeles, California 90007

/_£T?X

A) H

Number 481
20 April 2000

Contributions
in Science

A New Hemphillian (Late Miocene)
Mammalian Fauna from Ho ye Canyon,
West Central Nevada

Thomas S. Kelly

Serial
Publications
of THE
Natural History
Museum of
Los Angeles
County

Scientific

Publications

Committee

John Heyning, Deputy Director
for Research and Collections
John M. Harris, Committee Chairman
Brian V. Brown
Kenneth E. Campbell
Kirk Fitzhugh
Karen Wise

Robin A. Simpson and K. Victoria Brown,

Managing Editors

The scientific publications of the Natural History Museum
of Los Angeles County have been issued at irregular in-
tervals in three major series; the issues in each series are
numbered individually, and numbers run consecutively, re-
gardless of the subject matter.

# Contributions in Science, a miscellaneous series of tech-
nical papers describing original research in the life and
earth sciences.

# Science Bulletin, a miscellaneous series of monographs
describing original research in the life and earth sci-
ences. This series was discontinued in 1978 with the
issue of Numbers 29 and 30; monographs are now
published by the Museum in Contributions in Science.

6 Science Series, long articles and collections of papers on
natural history topics.

Copies of the publications in these series are sold through
the Museum Book Shop. A catalog is available on request.

The Museum also publishes Technical Reports, a miscel-
laneous series containing information relative to scholarly
inquiry and collections but not reporting the results of
original research. Issue is authorized by the Museum’s Sci-
entific Publications Committee; however, manuscripts do
not receive anonymous peer review. Individual Technical
Reports may be obtained from the relevant Section of the
Museum.

Natural History Museum
of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

Printed at Allen Press, Inc., Lawrence, Kansas
ISSN 0459-8113

A New Hemphillian (Late Miocene) Mammalian
Fauna from Hoye Canyon, West Central
Nevada

Thomas S. Kelly1

ABSTRACT. A new late Hemphillian (late Miocene) fossil mammalian assemblage, the Hoye Canyon
Local Fauna, is now recognized from an unnamed formation exposed along the western flanks of the
Wellington Hills, Douglas County, Nevada. The fauna was recovered from the lower part of the unnamed
formation and consists of the following taxa: Leporidae, Pronotolagus nevadensis n. sp.; Sciuridae, Sper-
mophilus wellingtonensis; Sciuridae, Marmota korthi n. sp.; Geomyidae, Parapliosaccomys oregonensis;
Camelidae, Hemiauchenia vera ; PAntilocapridae, gen. and sp. indet.; Rhinocerotidae, gen. and sp. indet.;
and Equidae, Dinohippus sp. indet. The middle and upper portions of the unnamed formation previously
yielded late Blancan (late Pliocene) and early Irvingtonian (early Pleistocene) faunas. Discovery of the Hoye
Canyon Local Fauna now indicates that the formation spans the late Hemphillian to the early Irvingtonian
or about 7 to 1.8 million years before present.

INTRODUCTION

Kelly (1997) documented the first records of fossil
vertebrates from the Wellington Hills-Antelope
Valley area of Douglas County, Nevada. The fossils
were recovered from an unnamed formation ex-
posed along the western flanks of the Wellington
Hills from the vicinity of Hoye Canyon in the north
to Risue Canyon in the south (Fig. 1). Kelly (1997)
provisionally recognized two late Cenozoic mam-
malian faunas from the unnamed formation: the
late Blancan Wellington Hills Local Fauna and the
early Irvingtonian Topaz Lake Local Fauna. Kelly
(1997) reported only one locality in the Hoye Can-
yon area, UCMP V-95013 (= LACM 6993), that
did not yield any age-diagnostic fossils at the time.
Because the northern outcrops of the unnamed for-
mation in the Hoye Canyon area are isolated from
the southern outcrops by foothills covered with
Quaternary alluvium, Kelly (1997, fig. 3) only
questionably correlated UCMP V-95013 with the
late Blancan localities to the south.

Quarrying at LACM 6993 and the discovery of
an additional locality, LACM 6994, that occurs
slightly higher in the section above LACM 6993,
has resulted in the recovery of age-diagnostic fos-
sils. These new fossils indicate that the fauna from
the Hoye Canyon localities is late Hemphillian (late
Miocene) in age and, thus, considerably older than
the fauna from the late Blancan (late Pliocene) lo-
calities of the southern outcrops. The assemblages
from LACM 6993 and LACM 6994 are herein
named the Hoye Canyon Local Fauna.

1. Museum Associate, Research and Collections
Branch, Vertebrate Paleontology Section, Natural History
Museum of Los Angeles County, 900 Exposition Blvd.,
Los Angeles, California 90007.

The purpose of this report is to document the
new records of Hemphillian mammalian fossils
from the Hoye Canyon area and reevaluate the bio-
stratigraphy of the unnamed formation exposed
along the western flanks of the Wellington Hills.

METHODS

Larger mammal teeth and appendicular elements were
measured with a vernier caliper to the nearest 0.1 mm and
those of smaller mammals were measured with an optical
micrometer disc to the nearest 0.01 mm. All teeth were
measured along their greatest anteroposterior and trans-
verse enamel dimensions. Metric abbreviations, dental ter-
minology, and dental formulas follow standard usage.
Measurements and calculations of the degree of deflection
of the posterior external reentrants in the rabbit lower
premolars follows those of White (1987, 1991). Speci-
mens previously collected from the unnamed formation of
the Wellington Hills-Antelope Valley area reported on by
Kelly (1997) are housed in the University of California,
Berkeley, Museum of Paleontology, whereas all new spec-
imens recovered during this study have been deposited in
the Vertebrate Paleontology Section of the Natural His-
tory Museum of Los Angeles County.

Abbreviations are as follows: AER, anterior external re-
entrant; AIR, anterior internal reentrant; ANT, anterior;
A-P, greatest anteroposterior dimension; AR, anterior re-
entrant; d, deciduous; D-V, dorsoventral; L, left; Ma, mil-
lion years before present; PER, posterior external reen-
trant; PIR, posterior internal reentrant; POST, posterior;
R, right; s. s., sensu stricto; TR, greatest transverse di-
mension. Institutional acronyms are as follows: LACM,
Natural History Museum of Los Angeles County; UCMP,
University of California, Berkeley, Museum of Paleontol-
ogy; V-, UCMP vertebrate fossil locality.

GEOLOGY AND FOSSIL OCCURRENCES

Halsey (1953) first reported the occurrence of a
thick deposit of Tertiary sediments along the west-

Contributions in Science, Number 481, pp. 1-21
Natural History Museum of Los Angeles County, 2000

119° 25'

Figure 1. Map of Antelope Valley- Wellington Hills area showing geographic extent of unnamed formation (stippled
areas) and locations of Hoye Canyon fossil localities (1 = LACM 6994, 2 = LACM 6993). Base map: U.S.G.S. 15
minute series, Desert Creek Peak, Nevada-California Quadrangle (scale = 1:62,000, contour interval = 400 ft).

ern flanks of the Wellington Hills from Risue Can-
yon northward to Hoye Canyon. Moore (1969) re-
garded these sediments as probably middle or early
late Miocene (Barstovian or Clarendonian) in age.
However, Kelly (1997) provided biostratigraphic
evidence that these sediments are, in part, late Pli-
ocene to early Pleistocene (late Blancan to early Ir-

vingtonian) in age. Kelly (1997) regarded these sed-
iments as an unnamed formation because they are
a single continuously deposited rock unit comprised
of stream and lake deposits of relatively homoge-
neous lithologies.

The unnamed formation of the Wellington Hills
area is comprised of alternating sequences of lacus-

2 ■ Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada

trine, braided fluvial, and overbank sediments that
are composed of tuffaceous mudstone, diatoma-
ceous shale, siltstone, sandstone, and conglomerate
(Kelly, 1997). In Hoye Canyon, the unnamed for-
mation unconformably overlies Miocene andesite
interbedded with minor sedimentary deposits,
probably a correlative of the Kate Peak Formation
of Gianella (1936), and is unconformably overlain
by Quaternary alluvium.

The mammalian fossils were recovered from la-
custrine deposits exposed on the south side of Hoye
Canyon (detailed locality data on file at the
LACM). Locality LACM 6993 occurs about 170 m
above the contact with the Miocene andesite in a
1.6-m-thick sandstone bed that contains small
mudstone clasts and thin lenses of conglomerate.
The precise stratigraphic position of LACM 6993
relative to the base of the formation is difficult to
determine because a small alluvium-filled valley
separates the section that contains LACM 6993
from the lowermost portion of the formation. Lo-
cality LACM 6994 occurs near the top of a 3.9-m-
thick bentonitic clay and mudstone bed, about 30
m stratigraphically higher in the section than
LACM 6993.

SYSTEMATIC PALEONTOLOGY
Order Lagomorpha Brandt, 1855

Family Leporidae Fischer de Waldheim,
1817

Genus Pronotolagus White, 1991

Pronotolagus nevadensis, new species

Figure 2, Table 1

HOLOTYPE. Partial dentary with L P3-M3,
LACM 145952.

TYPE LOCALITY. LACM 6993.

HYPODIGM. From LACM 6993: partial den-
tary with L dP3_4, Mj.3, LACM 145953; partial
dentary with L P4-M3, LACM 145954; partial den-
tary with R P4-M2, LACM 145955; R P3, LACM
145956.

DISTRIBUTION AND AGE. Known only from
the type locality, late Hemphillian.

ETYMOLOGY. Named for its occurrence in Ne-
vada.

DIAGNOSIS. Pronotolagus nevadensis differs
from Pronotolagus albus Voorhies and Timperley,
1997, by the following characteristics: (1) much
smaller size (mean P3 A-P 62% smaller than that
of P. albus ); (2) better developed and deeper P3 AIR
(depth averages 22.2% of TR occlusal dimension);
(3) a distinct P3 PIR present and cement filled; (4)
deeper P3 AER (depth averages 23% of TR occlusal
dimension); and (5) shallower P3 PER (depth av-
erages 44.2% of TR occlusal dimension). It differs
from Pronotolagus apachensis (Gazin, 1930) by the
following characteristics: (1) much smaller size
(mean P3 A-P 50% smaller than that of P. apacb-
ensis ); (2) deeper P3 AIR; (3) more distinct and

deeper P3 PIR (depth averages 8.5% of TR occlusal
dimension). It differs from Pronotolagus whitei
Korth, 1998, by the following characteristics: (1)
much smaller size (mean P3 A-P 53% smaller than
that of P. whitei); (2) deeper P3 AIR; (3) better de-
veloped P3 PIR; (4) deeper P3 AER; and (5) shal-
lower P3 PER that is inclined posteriorly.

DESCRIPTION. Of the five specimens of Pron-
otolagus nevadensis, LACM 145953 retains dP3_4,
indicating that this specimen represents an imma-
ture individual, whereas P3s of the holotype and the
referred specimen (LACM 14 5956) are in early
wear indicating that these specimens represent
young adults. Although P3s of LACM 145954 and
LACM 145955 are missing, the remaining cheek
teeth are in early wear and early moderate wear,
respectively, indicating that they also represent
young adults. The dentary is of typical leporid
structure. The anterior margin of the masseteric
fossa extends to a point below the middle of Mx.
Numerous small foramina are present on the lateral
side of the horizontal ramus below P3.

DP3 is molariform and rooted, with the trigonid
and talonid being transversely expanded. An oval-
shaped anterior conid is present that is connected
to the middle of the anterior face of the trigonid by
an isthmus. Likewise, the trigonid is connected to
the middle of the anterior face of the talonid by an
isthmus. DP4 also is molariform and rooted but dif-
fers from dP3 by having an anterior marginal crest
along the anterior border of the trigonid instead of
a distinct conid. Thin enamel bands are present
along the posterolabial margins of the dP3_4 trigo-
nids and talonids. Hypoconulids are lacking on
both deciduous premolars.

P3 of the holotype (LACM 145952, Fig. 2B) ex-
hibits the following characteristics: (1) size small,
as compared with other species of Pronotolagus;
(2) an AIR and PIR are present with the AIR shal-
lower than the PIR; (3) the AIR depth is 20.8% of
the TR occlusal dimension; (4) the PIR depth is
8.3% of the TR occlusal dimension; (5) the AER
depth is 25% of the TR occlusal dimension; (6) the
PER is 45.8% of the TR occlusal dimension; (7) an
AR is lacking; (8) the thick enamel of the PER is
relatively straight with a slight posterior deflection
(9.2 degrees); (9) the thin enamel of the PER is
smooth, lacking crenulations; and (10) cement is
present in the AIR, PIR, AER, and PER. P3 of the
holotype was broken off 2.8 mm from the crown
at the alveolar border during preparation that al-
lowed examination of the enamel pattern farther
down the crown. The enamel pattern (Fig. 2C) at
this point is very similar to the occlusal enamel pat-
tern, primarily differing by a slight increase in the
depth of the AIR and PIR (21.2% and 10.2% of
the TR occlusal dimension, respectively) and a
slightly more posteriorly deflected PER (9.3 de-
grees). The P3 occlusal enamel pattern of LACM
145956 (Fig. 2D) is very similar to the holotype,
primarily differing by having a slightly shallower

Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada ■ 3

F

Figure 2. Pronotolagus nevadensis new species. A, partial dentary with L P3-M3, holotype, LACM 145952, lateral view.
B, L P3, holotype, LACM 145952, occlusal view, anterior left. C, L P3, holotype, LACM 145952, cross-section 2.8 mm
below crown, anterior left. D, R P3, LACM 145956, occlusal view, anterior right. E, R P3, LACM 145956, enamel pattern
at base of tooth, anterior left. F, R dP3, LACM 145953, occlusal view, anterior right. Upper scale for A = 1 mm, lower
scale for B-F = 1 mm.

PIR, AER, and PER. The reentrants at the occlusal
surface of LACM 145956 exhibit the following
characteristics: (1) the AIR is 20.9% of the TR oc-
clusal dimension; (2) the PIR is 4.3% of the TR
occlusal dimension; (3) the AER is 20.9% of the
TR occlusal dimension; (4) the PER is 43% of the
TR occlusal dimension; (5) the thick enamel of the

PER is almost straight with a slight posterior de-
flection (10 degrees); and (6) cement is present in
the AIR, PIR, AER, and PER. The enamel pattern
at the base of LACM 145956 (Fig. 2E) is similar
to the occlusal enamel pattern, primarily differing
by having a deeper AIR, PIR, AER, and PER
(26.1%, 11.3%, 21.7%, and 43.5% of the TR di-

4 ■ Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada

Table 1. Measurements (in mm)

of specimens of Pronotolagus nevadensis new

species from Hoye Canyon.

Tooth position/
dimension

LACM

145953

Holotype

LACM

145952

LACM

145954

LACM LACM

145955 145956

dP3 A-P

1.06

ANT-TR

0.70

POST-TR

1.01

dP4 A-P

1.00

ANT-TR

1.05

POST-TR

1.06

P3 A-P

1.20

1.19

TR

1.10

1.04

P4 A-P

1.28

1.34

1.33

ANT-TR

1.33

—

1.34

POST-TR

1.21

1.55

1.31

Mj A-P

1.02

1.45

1.32

1.33

ANT-TR

1.09

1.35

1.50

1.45

POST-TR

1.10

1.30

1.45

1.44

M2 A-P

1.05

1.36

1.33

1.39

ANT-TR

1.06

1.34

1.36

1.35

POST-TR

1.04

1.23

1.33

1.34

M3 A-P

0.46

0.68

0.70

TR

0.50

1.00

0.99

dP3-M3 alveolar A-P

5.25

P3-M3 alveolar A-P

6.31

6.61

Depth of dentary

below P4

3.98

5.05

4.91

5.09

mension, respectively) and a less posteriorly deflect-
ed PER (9.7 degrees).

The lower molars are of typical leporid structure
with oval-shaped trigonids and talonids that are
connected by an isthmus, prominent anterior mar-
ginal crests along the anterior borders of the tri-
gonids, and enamel along the posterolabial borders
of the trigonids and talonids. In all of the partial
dentaries, the molar crowns are well above their
alveolar borders.

DISCUSSION. Species of rabbits are primarily dif-
ferentiated by the morphologies of P2 and P3 (White,
1987, 1991). Recently, Voorhies and Timperley
(1997) emended the diagnosis of Pronotolagus,
wherein they listed the following diagnostic charac-
teristics: (1) small- to medium-sized leporines; (2) an
AR is lacking on P3; (3) the P3 AIR is more deeply
incised than the PIR when the latter is present; (4)
the P3 AER is shallow and wide; and (5) the P3 PER
ranges from 40 to 58% of the transverse occlusal
surface. The Hoye Canyon rabbit specimens exhibit
all of these diagnostic characters and can be assigned
confidently to Pronotolagus.

Although only two P3s are known for Pronoto-
lagus nevadensis, they both exhibit very similar oc-
clusal enamel patterns. These enamel patterns also
remain rather consistent down the crown as indi-
cated by the cross-sectional pattern of the holotype
2.8 mm below the occlusal surface and by the
enamel pattern at the base of LACM 145956 4.5

mm below the occlusal surface (Fig. 2B-E). Also,
the A-P and TR dimensions vary little down the
crowns. In the holotype, the P3 A-P and TR dimen-
sions vary from 1.20 mm and 1.10 mm at the oc-
clusal surface to 1.23 mm and 1.12 mm at the al-
veolar border, respectively. In LACM 145956, the
A-P and TR dimensions vary from 1.19 mm and
1.04 mm at the occlusal surface to 1.21 mm and
1.06 mm at the base of the crown, respectively.
Thus, even though P3s are in early wear, their oc-
clusal dimensions and enamel patterns would
change little with additional wear. Dalquest (1979)
noted that P3 is diagnostic in almost all instances,
even in immature rabbits where the occlusal surface
is unworn by using the enamel pattern at the base
of the tooth or cross-sectioned patterns.

The dimensions of rabbit cheek teeth vary with
age, wherein P3s of immature individuals usually
exhibit a marked increase in size from the occlusal
surface to the base of the tooth (Dice and Dice,
1935, 1941; Wood, 1940; White, 1991). This is
true especially for immature P3s that are unworn or
just beginning to wear. For example, in an imma-
ture P3 of Nekrolagus progressus (Hibbard, 1939),
the A-P dimension at the base of the tooth is 38%
larger than the occlusal dimension (Hibbard, 1963,
fig. lb-d). Similarly, in an immature P3 of Pratile-
pus kansasensis Hibbard, 1939, the A-P dimension
at the base of the tooth is 38% greater than the
crown dimension (Hibbard, 1963, fig. 2a-a'). In an

Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada I 5

unworn P3 of Pronotolagus whitei, the A-P crown
dimension is about 22% smaller than the mean A-
P dimension of adult P3s (Korth, 1998, fig. 14D-G,
table 14). In Palaeolagus Leidy, 1856, and Lepus
Linnaeus, 1758, the P3 occlusal dimensions in early
wear average about 18% and 10% smaller, respec-
tively, than those of well-worn teeth (Dice and
Dice, 1935; Wood, 1940; Hibbard, 1963). As noted
above, P3s of Pronotolagus nevadensis do not ex-
hibit any significant differences in the A-P and TR
dimensions from the occlusal surface to the base of
the crowns. However, because of the small sample
size, the possibility that older individuals could
have larger P3s cannot be ruled out. Even if the P3
A-P dimensions of more mature individuals of P.
nevadensis were discovered to be 30% larger than
those of the holotype and referred specimen, they
would still be 45%, 40%, and 30% smaller than
the means of those of Pronotolagus albus, Prono-
tolagus whitei, and Pronotolagus apachensis, re-
spectively. Moreover, the fact that the A-P dimen-
sion of dP4 of P. nevadensis is 48% smaller than
the mean dP4 A-P of P. whitei (Korth, 1998, table
14) further supports the conspicuous size difference
between P. nevadensis and the other species of
Pronotolagus.

Voorhies and Timperley (1997) noted the follow-
ing evolutionary trends in Pronotolagus with de-
creasing geologic age: (1) a decrease in body size;
(2) a deepening of the P3 AIR; (3) an increase in the
amount of cement in the P3 AIR; and (4) a more
distinct P3 PIR. At the time, only two species of
Pronotolagus were known, Pronotolagus albus
from the late Barstovian of Nebraska and Prono-
tolagus apachensis from the Clarendonian of Cali-
fornia and the early Hemphillian of Nebraska. Sub-
sequently, Korth (1998) described a third species,
Pronotolagus whitei from late Clarendonian Pratt
Quarry of the Merritt Dam Member of the Ash
Hollow Formation, Nebraska. Although P. whitei
is similar in size to P. apachensis, it differs from it
by having a deeper P3 PER that is inclined anteri-
orly instead of posteriorly (Korth, 1998). Korth
(1998) noted that the early Hemphillian specimens
from the LeMoyne Quarry of Nebraska that White
(1991) referred to P. apachensis differ from the to-
potypic Clarendonian sample of P. apachensis from
California by having anteriorly inclined P3 PERs,
like those of P. whitei. The P3 AIRs of the LeMoyne
Quarry specimens are also deeper and more persis-
tent than the topotypic samples of P. apachensis
and P. whitei (White, 1991; Korth, 1998). Korth
(1998) suggested that the LeMoyne Quarry sample
might represent a distinct species more closely re-
lated to P. whitei than P. apachensis. P3s of P. nev-
adensis differ from those of the early Hemphillian
LeMoyne Quarry sample by having the following:
(1) smaller size; (2) more persistent PIRs; and (3)
shallower PERs that are inclined posteriorly rather
than anteriorly.

Pronotolagus nevadensis is the smallest species of
the genus and its P3 AIR and PIR are particularly
well developed, as compared with those of Prono-

tolagus albus, Pronotolagus whitei, and Pronoto-
lagus apachensis (White, 1991; Voorhies and Tim-
perley, 1997; Korth, 1998). If the evolutionary
trends noted by Voorhies and Timperley (1997) ac-
tually represent derived character transformations,
then P. nevadensis is the most derived species of
Pronotolagus. As noted above, the age of the Hoye
Canyon Local Fauna is late Hemphillian and, there-
fore, P. nevadensis is also the youngest known spe-
cies of Pronotolagus.

Order Rodentia Bowdich, 1821
Family Sciuridae Fischer de Waldheim, 1817
Genus Spermophilus Cuvier, 1825

Spermophilus wellingtonensis Kelly, 1997
Figure 3, Table 2

SPECIMENS. From LACM 6993: partial skull
with L and R T-M3 and associated appendicular
elements, LACM 145957; partial dentary with R
I„ LACM 145958.

DESCRIPTION. The cranial morphology and
upper dentition of Spermophilus wellingtonensis
were previously unknown. The posterior portion of
the partial skull is missing, being broken off dor-
sally across the parietals and ventrally across the
palatines (Fig. 3 A, B). The anterior tips of the na-
sals and premaxillaries and the zygomatic arches
are also missing. Many small fractures are present,
making it difficult to identify sutures and individual
bones. The nasals taper posteriorly and the naso-
frontal sutures form an obtuse angle with the apex
pointing posteriorly. The partial parietals are slight-
ly depressed, but the skull roof appears to have
been relatively flat. The rostrum is elongate, taper-
ing slightly anteriorly. The infraorbital foramen is
an oval slit (2.3 mm D-V, 1.3 mm A-P) on the max-
illa and positioned 1.4 mm anteriorly from the an-
terior margin of the P3 alveolus. The maxillary root

Table 2. Measurements (in mm) of upper dentition of
Spermophilus wellingtonensis from Hoye Canyon.

Tooth position/
dimension

LACM 145957

Right

Left

P3

A-P

1.28

1.29

TR

1.25

1.27

P4

A-P

2.36

2.28

TR

2.74

2.79

M1

A-P

2.59

2.71

TR

3.17

3.20

M2

A-P

2.64

2.63

TR

3.34

3.42

M3

A-P

3.11

3.13

TR

3.17

3.16

p3— 4

A-P

3.23

3.32

A-P

8.09

8.12

P4-M3

A-P

10.17

10.20

P3-M3

A-P

10.77

11.35

6 ■ Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada

c —

Figure 3. Spermopbilus wellingtonensis. A-C, partial skull, LACM 145957. A, right lateral view. B, palatal view. C, L
P3-M3, occlusal view, anterior left. Upper scale for A, B = 10 mm; lower scale for C = 1 mm.

of the zygomatic arch extends from a point above
the anterior portion of P4 to one above the middle
of M1. The palate is broad, and the tooth rows are
nearly parallel. The anterior margin of the pala-
tine-maxillary suture extends to a point below the
middle of M1. The posterior palatine foramen is
positioned below the middle of M2. The matrix
within the orbits was not removed to allow the
morphology of the foramina to be observed be-
cause of the fragile condition of the skull.

The teeth of LACM 145957 are well preserved
and only moderately worn (Fig. 3C). P3 is single-
rooted and has an oval occlusal outline. A single
prominent cusp is positioned anteriorly, and a well-
developed posterior cingulum is present along the
posterolingual border of the tooth.

P4 is trapezoidal in occlusal outline and moder-
ately reduced in size relative to M‘~2. The P4 ante-
rior cingulum extends anterolabially from the an-
terior base of the protocone to the anterolabial cor-
ner of the tooth where a distinct, well-developed
parastyle is present. The protoloph is complete,
connecting the protocone and paracone. A small,

but distinct, mesostyle is present. The metaloph is
almost complete, separated from the protocone by
a very shallow notch that would disappear with
slightly more wear. The protoloph and metaloph
are separated by a deep valley and are distinctly
taller than the anterior and posterior cingulae. The
metaconule is well developed. The posterior cin-
gulum extends from the posterolabial base of the
protocone to the metacone with a narrow valley
separating it from the metaloph.

M1 has a trapezoidal occlusal outline. The ante-
rior cingulum extends from the anterolabial base of
the protocone to the anterolabial corner of the
tooth, where a moderately well-developed parastyle
is present. The protoloph and metaloph are com-
plete, connecting the protocone to the paracone
and the protocone to the metacone, respectively.
The loph connecting the protocone to the metaco-
nule is narrower than the loph that connects the
protocone to the protoloph. A small, but distinct,
mesostyle is present. A metaconule is developed as
a distinct swelling on the metaloph. The metalophs
and protolophs are considerably taller than the an-

Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada ■ 7

terior and posterior cingulae. The posterior cingu-
lum is similar in morphology to that of P4 with a
narrow valley between it and the metaloph. M2 is
very similar in structure to M1 but differs in having
an anterior cingulum that is not as anteriorly ex-
panded at its anterolabial margin and a narrower
valley between the protoloph and metaloph.

M3 is expanded anteroposteriorly, almost as long
as it is wide. The anterior cingulum is very similar
in structure to that of M2, but the parastyle is not
as prominent. The protoloph is complete. A small
protoconule is developed as a slight swelling on the
protoloph. A short metaloph is present and extends
posterolabially from the protocone to a little more
than a third of the way across the occlusal surface.
The posterior cingulum extends posterolabially
from the middle of the posterior base of the pro-
tocone to form a posterolabially expanded shelf
and then curves anteriorly where it continues to the
base of the paracone as a low loph. A metacone is
lacking.

The partial dentary (LACM 145958) is damaged,
with the posterior portion broken off at about the
level of the M2 alveolus. The diastema between the
lower incisor and P4 is 7.9 mm. Kelly (1997) al-
ready has described the morphology of the lower
incisor of Spermophilus wellingtonensis.

DISCUSSION. Kelly (1997) described Sper-
mophilus wellingtonensis from LACM 6993 ( =
UCMP V-95013) based on the holotype (a partial
dentary with P4-M3, UCMP 141314) and an ad-
ditional partial dentary (UCMP 141341). The par-
tial skull (LACM 145957) was recovered from the
quarry at LACM 6993 within 0.5 m of where the
holotype of S. wellingtonensis was discovered. Be-
cause both the holotype dentary and partial skull
were recovered from the same locality and the size
of their dentitions is compatible, the partial skull is
referred to S. wellingtonensis.

Based on dental morphology, Kelly (1997) re-
garded Spermophilus wellingtonensis as most close-
ly related to the late Hemphillian Spermophilus
shotwelli (Black, 1963) of the McKay Reservoir Lo-
cal Fauna of Oregon. Although similar in dental
morphology, Kelly (1997) noted that 5. wellingto-
nensis can be easily distinguished from 5. shotwelli
by the following characteristics: (1) the P4 antero-
conid is well developed; (2) the P4 protoconid and
metaconid are separated by a relatively deep notch
or groove; (3) the M1-3 metalophids are less com-
plete, with the trigonids open to the talonids at an
earlier wear stage; and (4) M3 is larger relative to
Mj_2. With the discovery of the upper dentition of
5. wellingtonensis, the following additional char-
acteristics can now be used to distinguish 5. wel-
lingtonensis from 5. shotwelli: (1) much greater an-
terior expansion of the P4 anterior cingulum; (2) the
presence of a well-developed P4 parastyle; (3) more
complete M1-2 metalophs, attaching to the proto-
cones at an earlier wear stage; and (4) slightly more
prominent M1-2 mesostyles.

Genus Marmota Blumenbach, 1779

Marmota korthi new species

Figures 4, 5, Table 3

Marmota or Cynomys, sp. indet.: Kelly, 1997:15.

HOLOTYPE. Associated partial maxilla with
partial L P4-M3, complete L M2 3, and partial den-
tary with partial L I1? complete L P4-M2, and par-
tial L M3, LACM 145959.

TYPE LOCALITY. LACM 6993.

HYPODIGM. From LACM 6993: partial skull
with L M13, isolated R M2, and associated partial
skeleton, LACM 145961; partial maxilla with L
P4-M3, LACM 145960; partial dentary with bro-
ken R P4-M2, UCMP 141313.

DISTRIBUTION AND AGE. Known only from
the type locality, late Hemphillian.

ETYMOLOGY. Named in honor of William W.
Korth of the Rochester Institute of Vertebrate Pa-
leontology in recognition of his many contributions
to our understanding of rodent phylogeny.

DIAGNOSIS. Marmota korthi differs from Mar-
mota vetus (Marsh, 1871) by the following char-
acteristics: (1) size larger (mean P4 A-P 38% larger
than that of M. vetus ); (2) I-P4 diastema relatively
longer; (3) lower incisor lacking median groove; (4)
P4 with incipient mesoconid present and more elon-
gated anteroposteriorly relative to Mx_2 anteropos-
terior lengths and wider transversely relative to
Mj_2 transverse widths; and (5) M1-2 metalophids
more complete with better developed and deeper
trigonid valleys that are completely closed off from
talonids. It differs from Marmota minor (Kellogg,
1910) by the following characteristics: (1) size larg-
er (mean P4 A-P 12% larger than that of M. minor );

(2) P4 anterior cingulum less expanded anteriorly;

(3) M1-3 metaconules better separated from proto-
cones; (4) M3 metaloph better developed; (5) P4 tri-
gonid open anteriorly and relatively narrower
transversely; (6) P4 less elongated anteroposteriorly
relative to Mx_2 anteroposterior lengths and trans-
verse width narrower relative to M1-2 transverse
widths; (7) M1-2 metalophids more complete with
trigonid valleys completely closed off from talonids;
(8) M3_2 relatively less anteroposteriorly com-
pressed; and (9) M2 anteroposterior length more
elongated relative to Mt anteroposterior length.
Marmota korthi can be easily distinguished from
all other late Blancan to Recent species of Marmota
by the following characteristics: (1) cheek teeth
lower crowned; (2) M1-2 metalophs less complete
and distinctly separated from protocone until late
wear; (3) P4 less molariform, trigonid less trans-
versely expanded, and less elongated anteroposte-
riorly relative to Mt_2 anteroposterior lengths; and

(4) M1-2 relatively less anteroposteriorly com-
pressed.

DESCRIPTION. The partial skull (LACM
145961) of Marmota korthi is badly damaged (Fig.
4). L M1-3 of the skull are well worn. An isolated
R M2 was found within the small block of matrix

8 ■ Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada

c

Figure 4. Marmota korthi new species. A-C, partial skull, LACM 145961. A, dorsal view. B, right lateral view. C, ventral
view. Scale =10 mm and all have anterior right.

Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada ■ 9

Figure 5. Marmota korthi new species. A, L P4-M*, LACM 145960, occlusal view. B, L P4, partial M1, M2-3, holotype,
LACM 145959, occlusal view. C, L P3-M2, partial M3, holotype, LACM 145959, occlusal view. D, partial dentary,
holotype, LACM 145959, lateral view. E, partial dentary, holotype, LACM 145959, occlusal view. Upper scale for A-C
= 1 mm; lower scale for D, E = 10 mm, and all have anterior left.

10 ■ Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada

Table 3. Measurements (in mm) of dentition oiMarmota
korthi new species from Hoye Canyon (e = estimated).

Tooth

position/

dimension

Holotype

LACM

LACM

LACM

145959

145960

145961

Right

Left

P4

A-P

4.30e

4.27

TR

4.68

4.64

M1

A-P

4.55e

4.62

4.50

TR

—

5.14

5.33

M2

A-P

4.99

4.87

4.86

TR

5.34

5.48

5.49

M3

A-P

5.48

5.63

TR

5.25

5.55

M1

-3

14.64e

14.65

P4-

M3

18. 75e

P4

A-P

4.87

ANT-TR

3.67

POST-TR

4.42

Mj

A-P

4.42

ANT-TR

4.81

POST-TR

4.94

m2

A-P

4.83

ANT-TR

5.54

POST-TR

5.56

m3

A-P

—

ANT-TR

—

POST-TR

-----

p4-

M3 A-P

18.69e

that yielded the skull. It is indistinguishable in wear
and morphology from M2 attached to the skull and
is assumed to have broken off from the skull prior
to or during burial. Because of the poor condition
of the skull, it provides only limited information on
the cranial morphology of M. korthi. The overall
morphology of the partial skull appears to be very
similar to those of Recent species of Marmota. The
total A-P length of the specimen from the occipital
condyles to the anterior edge of the broken nasals
is 72.4 mm. The TR widths across the auditory bul-
lae and the postorbital constriction of the frontals
are 44.8 mm and 19.5 mm, respectively. Although
broken off at the tips, the supraorbital processes of
the frontals are well developed. Weakly developed
frontal crests extend posteriorly from the supraor-
bital processes to converge with the sagittal crest.
These crests appear to be slightly less developed
than those of late Blancan to Recent species of Mar-
mota. The auditory bullae are moderately inflated.
The pterygoid processes are well developed.

P4 of the holotype (Fig. 5B) is damaged with part
of the anterior cingulum and paracone missing,
whereas, in the referred specimen, P4 is missing
only a portion of the enamel along the lingual bor-
der. P4 is molariform with a trapezoidal-shaped oc-
clusal outline. The anterior cingulum extends an-
terolabially from the protocone to a distinct para-

style, which is separated from the paracone by a
shallow notch. A deep valley is present between the
protoloph and anterior cingulum. The well-devel-
oped protoloph extends labially from the large pro-
tocone to connect with the paracone. There is no
indication of a protoconule on the protoloph. The
protoloph is slightly lower in height than the me-
taloph. The metaloph is short and thick, connecting
the well-developed metaconule to the metacone.
The metaconule is well separated from the proto-
cone by a distinct valley between the trigon and
talon. A well-developed mesostyle is present be-
tween the paracone and metacone. The posterior
cingulum is well developed and extends lingually
from the protocone to connect with the posterior
base of the metacone. The posterior cingulum is
lower in height than the protoloph and metaloph.

In the holotype, M2-3 are complete, but M1 is
damaged, with part of the paracone, metacone, and
anterior cingulum broken off (Fig. 5B). In the re-
ferred specimens, M's are complete (Figs. 4C, 5A).
M1 and M2 are almost identical in structure, where-
as M3 differs from M1 2 primarily by having an en-
larged and posteriorly expanded posterior cingu-
lum. All the upper molars exhibit the following
characteristics: (1) the anterior cingulum is a low,
complete crest that extends labially from the pro-
tocone to the paracone, forming a distinct valley
between the anterior cingulum and the protoloph;
(2) a parastyle is present as a small cuspule on the
anterior cingulum; (3) the well-developed proto-
loph extends from the large protocone to the para-
cone and is the tallest loph on the occlusal surface;
(4) a protoconule is lacking; and (5) the well-de-
veloped metaconule is connected to the paracone
by a thick metaloph but is separated from the pro-
tocone by a distinct valley until late wear. A small
distinct mesostyle is present on M1-2, whereas, on
M3, only a very small mesostyle is present. The
M1-2 posterior cingulae are low, complete crests
that connect the protocones to the metacones. On
M3, the posterior cingulum extends posterolabially
from the protocone to about the middle of the
tooth, where a slight notch occurs and after which
the cingulum thickens and increases in height along
its posterolabial border. The enlarged M3 posterior
cingulum is separated from the metaloph by a deep
valley.

The dorsal surface of the dentary drops steeply
anterior to P4, forming a sharply curved diastema
(Fig. 5D). The anterior margin of the masseteric
fossa is somewhat bulbous and ends anteriorly be-
low the anterior margin of M,. The enamel band
on the lower incisor extends laterally from the an-
teromedial edge to the dorsal lateral border, and its
anterior surface is smooth, with no indication of a
median groove.

P4 is moderately enlarged relative to M3_2 (Fig.
5C). A small, low anterior cingulid extends lin-
gually from the base of the protoconid to the base
of the metaconid, resulting in an anteriorly open
trigonid. The protoconid and metaconid are the

Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada ■ 1 1

tallest cusps and positioned relatively close to each
other, resulting in a small trigonid. The metalophid
is slightly damaged but appears to have been com-
plete with a slight notch near its attachment to the
metaconid. A small metastylid is present along the
lingual border near the metaconid. The talonid is
open lingually but closed off labially by a complete
ectolophid between the protoconid and hypoconid.
A shallow talonid trench is present just lingual to
the ectolophid. A small, but distinct, mesoconid is
present on the ectolophid between the protoconid
and hypoconid. The hypoconid is well developed,
whereas the entoconid is represented only by a very
small cuspulid on the posterior cingulid. The pos-
terior cingulid is a continuous, low lophid that ex-
tends from the hypoconid to the entoconid. A very
low lophid extends anteriorly from the entoconid
toward the metastylid but is separated from the me-
tastylid by a small notch.

Mj_2 have parallelogram-shaped occlusal outlines
and are essentially identical in structure, except that
M2 is larger than (Fig. 5C). The metaconid is
the tallest primary cusp, followed by, in decreasing
height, the protoconid, hypoconid, and entoconid.
The anterior cingulid is lower than the metalophid
but is complete, connecting the protoconid to the
metaconid and closing off the trigonid anteriorly.
The metalophid is a high lophid connecting the
metaconid to the protoconid, resulting in a deep
and well-developed trigonid basin that is complete-
ly closed off from the talonid. A complete, well-
developed ectolophid is present and positioned
deep in from the labial border of the tooth. A me-
soconid is lacking on the ectolophid. A shallow tal-
onid trench is present along the ectolophid and me-
talophid margins. The posterior cingulid extends
lingually as a low lophid from the well-developed
hypoconid to the posterolingual corner of the tooth
wherein it turns anteriorly and extends as a low
lophid to a very small metastylid. Because of the
low lophid along the lingual border, the talonid is
open lingually. The entoconid is a relatively indis-
tinct cuspulid on the posterior cingulid near the
posterolingual corner of the tooth.

M3 is badly damaged, with only the anterior cin-
gulid and part of the trigonid present. The trigonid
appears to have been similar to those of M3_2 with
a complete anterior cingulid and metalophid, re-
sulting in a deep, completely enclosed trigonid val-
ley.

DISCUSSION. Kelly (1997) referred a partial
right dentary with badly damaged teeth (UCMP
141313) from LACM 6993 (= UCMP V-95013) to
an indeterminate species of Marmota Blumenbach,
1779, or Cynomys Rafinesque, 1817. He could not
make a definitive generic diagnosis because of the
poor condition of the teeth. With the new marmot
specimens reported here, UCMP 141313 can now
be referred confidently to Marmota korthi.

Marmota korthi exhibits certain similarities to
species of Paenemarmota Hibbard and Schultz
(1948). Hibbard and Schultz (1948) described the

type species Paenemarmota barbouri from the
Blancan Rexroad Formation of Kansas. In a de-
tailed review of P. barbouri, Repenning (1962)
synonymized Marmota mexicanus Wilson, 1949,
with P. barbouri. Voorhies (1988) referred Mar-
mota sawrockensis Hibbard, 1964, to Paenemar-
mota and provided a revised diagnosis of the ge-
nus. The following characteristics have been used
previously to distinguish Paenemarmota from
Marmota (Hibbard and Schultz, 1948; Repenning,
1962; Voorhies, 1988): (1) the cheek teeth are
much larger; (2) the cheek teeth are relatively
more hypsodont and have more inflated cusps; (3)
the P4-M2 posterior cingulae are high crests that
are separated from the metalophs by deep valleys;
(4) P4 is larger than M1; (5) the P4 metaconule is
slightly more developed; (6) the M3 metaloph is a
distinct, high loph that is separated from the pro-
tocone until late wear; (7) the lower incisors have
distinct longitudinal striations and their bases ex-
tend well behind M3; (8) the P4-M3 protoconids
are large, equaling or exceeding the metaconids in
height and basal area; (9) P4-M3 have deep talonid
trenches present along the metalophid and ecto-
lophid margins; and (10) the P4-M3 talonid basins
have well-developed accessory ridges and cuspules
resulting in strongly rugose surfaces. Marmota
korthi is similar to P. sawrockensis and P. barbou-
ri by having a well-developed M3 metaloph that is
separated from the protocone until late wear and
lower cheek teeth with talonid trenches present.
However, the talonid trenches of M. korthi are not
as well developed as those of P. sawrockensis and
P. barbouri. Talonid trenches are also present in
species of Cynomys and Spermophilus Cuvier,
1825. Marmota korthi differs from P. sawrock-
ensis and P. barbouri by the following character-
istics: (1) the cheek teeth are much smaller; (2) P4
is smaller than M1; (3) the P4-M2 posterior cin-
gulae are not as well developed and lack deep val-
leys between the cingulae and metalophs; (4) the
lower incisors lack prominent longitudinal
grooves; (5) the P4 protoconid and metaconid are
less well separated; 6) the M1-3 metalophids are
more complete; (7) the P4-M3 ectolophids are rel-
atively deeper; and (8) the P4-M3 talonid basins
lack heavy rugosity. It further differs from P. bar-
bouri by having a much less molariform P4. Except
for the more complete Mt_3 metalophids, all of the
characters that distinguish M. korthi from P. saw-
rockensis and P. barbouri are also present in other
species of Marmota and support its referral to the
genus. Thus, well-developed M3 metalophs and
P4-M3 talonid trenches no longer appear to rep-
resent valid diagnostic characters to differentiate
Paenemarmota from Marmota.

Kellogg (1910) described Marmota nevadensis
based on a partial dentary with P4-Mi from the
middle Hemphillian Thousand Creek Formation of
Nevada. Subsequent investigators (Hibbard and
Schultz, 1948; Repenning, 1962; Black, 1963;
Voorhies, 1988) have noted the following similari-

12 ■ Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada

ties in the lower cheek teeth of M. nevadensis and
species of Paenemarmota: (1) large size; (2) prom-
inent longitudinal grooves on the lower incisors; (3)
well-developed talonid trenches on P4-Mx; and (4)
heavy rugosity in the P4-M1 talonid basins. Voor-
hies (1988, p. 171) stated that, when better known,
M. nevadensis “may eventually prove to represent
Paenemarmota rather than Marmota."" Marmota
nevadensis differs from and appears less derived
than Paenemarmota sawrockensis and Paenemar-
mota barbouri by its slightly smaller size and by
having a less molariform P4 that is smaller relative
to Mj. Until better known and following Korth
(1994), it appears best to refer M. nevadensis to
? Paenemarmota. Marmota korthi differs from ?P.
nevadensis by the following characteristics: (1)
much smaller cheek teeth; (2) lack of prominent
longitudinal grooves on the lower incisors; (3) lack
of heavy rugosity in the P4-M, talonid basins (M2_3
unknown for ?P. nevadensis ); (4) a more complete
Mi metalophid; (5) a relatively deeper Mj ectolo-
phid; and (6) a less developed M, posterior cingu-
lid.

Pre-Pleistocene marmot fossils are rare (Black,
1963). Marmota first appears in the Clarendonian,
where it is represented by a single species, Marmota
vetus (Marsh, 1871; Black, 1963; Korth, 1994).
Prior to this study, the only described species of
Marmota from the Hemphillian was Marmota mi-
nor from the middle Hemphillian Thousand Creek
Local Fauna of Nevada (Black, 1963). Hay (1921)
described a third pre-Pleistocene species, Marmota
arizonae, based on a partial skull from the late
Blancan Anita Fauna of Arizona (Kurten and An-
derson, 1980). Morphologically, M. arizonae is
very similar to the late Rancholabrean to Recent
Marmota flaviventris (Audubon and Bachman,
1841), differing only in the morphology of the
snout (Kurten and Anderson, 1980). The only other
North American species of Marmota recorded from
the Pleistocene (late Irvingtonian to late Rancho-
labrean) is extant Marmota monax Linnaeus, 1758.

Black (1963) regarded the following dental trans-
formations as derived for Marmota based on the
Pleistocene to Recent species: (1) an increase in size;

(2) an enlargement of P4 relative to the lower mo-
lars; (3) a reduction of the posterior portion of M3;
and (4) further anteroposterior compression of
M3_2. Additionally, the Pleistocene to Recent species
exhibit a moderate increase in crown height of the
cheek teeth relative to the Clarendonian and Hem-
phillian species. Black (1963) regarded Marmota
minor as having the following synapomorphies
with Recent species of Marmota ; (1) P4 is longer
than M1-2; (2) the metalophids of M,_3 are reduced;

(3) the posterior portion of M3 is reduced; and (4)
the diastemal depression is deep anterior to P4.
Marmota korthi is derived relative to Marmota ve-
tus by having the following characteristics: (1) in-
creased size; (2) higher crowned cheek teeth; (3)
greater enlargement of P4 relative to M,_2; and (4)

more complete M3_3 metalophids, with the trigo-
nids completely separated from the talonids. Mar-
mota korthi and M. vetus exhibit about the same
degree of anteroposterior compression of Mt_2.
Marmota korthi is derived relative to M. minor by
the following characteristics: (1) increased size; and
(2) more complete M1-2 metalophids with the tri-
gonids completely separated from the talonids, re-
sulting in deep, enclosed trigonid valleys. Marmota
korthi further differs from M. minor by the follow-
ing characteristics: (1) the P4 anterior cingulum is
less expanded anteriorly; (2) the M1-3 metaconules
are better separated from the protocones; (3) the
M3 metaloph is better developed, forming a contin-
uous crest from the metaconule to the metacone;

(4) the ratio of the P4 anteroposterior length to the
M]_2 anteroposterior lengths is slightly less (aver-
aging 1.06 for M. korthi versus 1.14 for M. minor);

(5) the ratio of the P4 transverse width relative to
the M1-2 greatest transverse widths is slightly less
(averaging 0.84 for M. korthi versus 0.93 for M.
minor); and (6) M1-2 are slightly less anteroposte-
riorly compressed (the ratio of the M1-2 greatest
transverse widths to the M,_2 anteroposterior
lengths averaging 1.14 for M. korthi versus 1.26
for M. minor). Marmota korthi can be easily dis-
tinguished from all late Blancan to Recent species
of Marmota by the following characteristics: (1)
cheek teeth and body size slightly smaller; (2) cheek
teeth lower crowned; (3) the M1-3 metalophs are
well separated from the protocones by distinct val-
leys until late wear; (4) P4 is less enlarged relative
to M1-2; and (5) M1-2 are less anteroposteriorly
compressed.

Family Geomyidae Bonaparte, 1845
Genus Parapliosaccomys Shotwell, 1967

Parapliosaccomys oregonensis
Shotwell, 1967
Figure 6, Table 4

SPECIMENS. From LACM 6993: R P4, LACM
145962; L M1 or2, LACM 145963; partial dentary
with L I„ dP4-M3, LACM 145964; R dP4, LACM

145965; partial dentary with R I3— P4, LACM

145966; R P4, LACM 145967; R P4, LACM

145968; R P4, LACM 145969; L P4, LACM

145970; L P4, LACM 145971; R Mj or 2, LACM
145972; R Mlor2, LACM 145973.

DISCUSSION. The dental sample of Paraplio-
saccomys from Hoye Canyon (Fig. 6) is indistin-
guishable from the topotypic sample of Paraplio-
saccomys oregonensis from McKay Reservoir,
Oregon, and can be referred confidently to this spe-
cies. Because Shotwell (1967) has already provided
detailed descriptions of the dental morphology of
P. oregonensis, a morphological description of the
dental sample from Hoye Canyon is not included
here. It should be noted that the occlusal dimen-
sions of the cheek teeth of P. oregonensis vary with

Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada ■ 13

cfooocQoO

BCE G

D F H

Figure 6. Parapliosaccomys oregonensis. A, R dP4, LACM 145964, occlusal view, reversed, B, R P4, LACM 145962, C,
D, L P4, LACM 145970. E, F, R P4, LACM 145968. G, H, R P4, LACM 145969. B, C, E, and G, occlusal views; D,
labial view; F and H, lingual views, and all have anterior left. Scale = 1 mm.

wear, so that teeth in early wear have significantly
smaller occlusal dimensions than those in late wear
(Shotwell, 1967; Kelly and Lugaski, 1999).

Kelly (1997) questionably referred UCMP
141344 (L P4, well worn) and UCMP 141343 (L
P4) from LACM 6993 (= UCMP V-95013) to }Ner-
terogeomys sp. indet. and ?Tbomomys sp. indet.,
respectively. With the discovery of the additional
geomyid material from LACM 6993, these speci-
mens can now be assigned confidently to Paraplio-
saccomys oregonensis.

Three species have been assigned previously to
Parapliosaccomys: the type species, Parapliosacco-

mys oregonensis, from the late Hemphillian McKay
Reservoir Fauna of Oregon and the late Hemphil-
lian Churchill Butte Local Fauna of Nevada; Par-
apliosaccomys bibbardi (Storer, 1973) from the
Clarendonian WaKeeney Local Fauna of Kansas;
and Parapliosaccomys annae Korth, 1987, from the
Barstovian Crookston Bridge Member of the Val-
entine Formation of Nebraska. However, Korth
and Reynolds (1994) recently described the genus
Phelosaccomys and referred P. bibbardi and P. an-
nae to their new genus. Thus, Parapliosaccomys is
a monotypic genus restricted to the late Hemphil-
lian. The presence of P. oregonensis in the Hoye

Kelly: Hemphillian Fauna from Nevada

14 ■ Contributions in Science, Number 481

Table 4. Measurements (in mm) of teeth of selected specimens of Parapliosaccomys oregonensis from Hoye Canyon;
all measurements taken at the occlusal surface (a = approximate, r = tooth just erupting).

Tooth position/ LACM LACM LACM LACM LACM LACM LACM LACM LACM LACM

dimension 145962 145963 145964 145965 145968 145969 145970 145966 145967 145971

P4

A-P

1.44

ANT-TR

1.22

POST-TR

1.60

Mlor2

A-P

TR

dP4

A-P

ANT-TR

POST-TR

P4

A-P

ANT-TR

POST-TR

Mj

A-P

TR

m2

A-P

TR

m3

A-P

TR

0.87

1.66

2.20 2.24

1.10 1.07

1.26 1.35

1.09
1.63a

1.10
1.67
0.99r
1.55r

1.93

2.47

1.71

1.23

1.15

1.25

1.66

1.60

1.36

1.39

1.73

1.50

1.30

1.46

1.26

1.61

1.80

1.49

Canyon Local Fauna indicates the fauna is late
Hemphillian in age.

Order Artiodactyla Owen, 1848
Family Camelidae Gray, 1821

Genus Hemiauchenia Gervais and
Ameghino, 1880

Hemiauchenia vera (Matthew, 1909)
Figure 7

SPECIMENS. From LACM 6993: partial den-
tary with partial L P3^, complete Ml5 LACM
146521. From LACM 6994: partial first phalanx,
LACM 145975; partial first phalanx, LACM
145976.

DISCUSSION. The partial dentary (LACM
146521) has the P3 crown broken off at the alve-
olus, the anterior portion of the P4 crown missing,
and Mj complete (Fig. 7). The partial dentary can
be confidently referred to Hemiauchenia vera (Mat-
thew, 1909, in Matthew and Osborn, 1909) be-
cause the teeth exhibit the following diagnostic
characters (Webb, 1974; Kelly, 1998b): (1) relative-
ly low-crowned and small in size, as compared with
all other species of Hemiauchenia ; (2) P3 is two-
rooted; (3) P4 has a single posterior fossettid and
appears to have had a simple triangular occlusal
outline; (4) P4 is anteroposteriorly compressed and
indented into the anterior occlusal surface of Mt;
and (5) Mj has very weak internal stylids and a
small protostylid present. Measurements of the
cheek teeth of LACM 146521 are as follows: P3
alveolar A-P = 7.2 mm; P, A-P = 11.7 mm (esti-
mated), TR = 6.6 mm; M, A-P = 20.3 mm, TR =
12.9 mm.

The partial first phalanges are characterized by
their small size, as indicated by the following mea-
surements: LACM 145975, distal condylar A-P =
10.5 mm, distal condylar TR = 12.6 mm; LACM
145976, distal condylar A-P = 11.5 mm, distal
condylar TR = 11.8 mm, and midshaft A-P = 13.4
mm. The camel phalanges from Hoye Canyon are
indistinguishable from those of Hemiauchenia vera
(Kelly, 1998b). The presence of a dental specimen
of H. vera in the Hoye Canyon Local Fauna strong-
ly suggests that the phalanges also represent this
species. For these reasons, the Hoye Canyon pha-
langes are provisionally referred to H. vera.

Hemiauchenia vera has been recorded previously
from the type locality in Hemphillian deposits of
the Ogallala Group of Long Island, Kansas, the late
Hemphillian Yerington and Silver Springs local fau-
nas of Nevada, the latest Hemphillian Buis Ranch
Local Fauna of Oklahoma, and the late Hemphil-
lian Upper Bone Valley Fauna of Florida (Webb,
1974; Tedford et al., 1987; Kelly, 1998b). The pres-
ence of H. vera in the Hoye Canyon Local Fauna
indicates the fauna is Hemphillian in age, probably
late Hemphillian.

Family Antilocapridae Gray, 1866
? Antilocapridae, gen. and sp. indet.

SPECIMEN. From LACM 6994: partial lower L
I2or3, LACM 145977.

DISCUSSION. The partial lower incisor is com-
plete, except that the tip of the root has been bro-
ken off. The root is robust relative to the crown,
indicating that the tooth is not a deciduous incisor.
The small, spatulate incisor is most similar to those
of the Antilocapridae. The tooth differs from those

Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada ■ 15

A

B

Figure 7. Hemiauchenia vera. A, B, partial dentary with partial L P3_4, M1? LACM 146521. A, occlusal view, anterior
left. B, labial view, anterior left. Scale =10 mm.

of the smallest Hemphillian camel, Hemiauchenia
vera, by its smaller size and differs from those of
the Tayassuidae by its smaller size and more spat-
ulate shape. The lower incisor probably represents
a member of the Antilocapridae, to which it is as-
signed very questionably.

Order Perissodactyla Owen, 1848

Family Rhinocerotidae Owen, 1845

Rhinocerotidae, gen. and sp. indet.

Figure 8

SPECIMEN. From LACM 6994: associated par-
tial L lower premolar, partial L Mj_3, LACM
145978.

DISCUSSION. The teeth are badly damaged
(Fig. 8). Mj and M3 are fairly complete, but the
partial lower premolar only consists of a partial ec-
tolophid and M2 by only the posterior enamel wall.
Mj and M3 exhibit the following characteristics: (1)

size small, as compared with other early to middle
Hemphillian rhinos; (2) moderately hypsodont,
considering their degree of wear; and (3) lingual
cingulids lacking. Measurements of the lower mo-
lars are as follows: M, A-P = 50.2 mm, TR — 33.3
mm; M3 A-P = 52.9 mm (broken), TR = 36.2 (bro-
ken).

The Hoye Canyon rhino teeth are most similar
in size and morphology to those from the late Hem-
phillian Washoe Local Fauna of Nevada that were
referred to Teleoceras sp. indet. by Kelly (1997). A
small species of Teleoceras Hatcher, 1894, also oc-
curs in the late Hemphillian Silver Springs Local
Fauna of Nevada (Kelly, 1998b). It appears that a
dwarf species of Teleoceras occurred during the late
Hemphillian in Nevada (Kelly, 1997, 1998b). Pre-
molar and molar lingual cingulids are generally
lacking in Teleoceras but commonly present in
Aphelops Cope, 1873 (Osborn, 1904; Douglas,
1908; Matthew, 1932; Tanner, 1967, 1975). The
Hoye Canyon rhino teeth lack lingual cingulids,

16 ■ Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada

A

B

Figure 8. Rhinocerotidae, gen. indet. A, B, L. Mj and M3, LACM 145978. A, occlusal views, anterior left. B, labial

views, anterior left. Scale = 1 0 mm.

suggesting that they might represent Teleoceras. It
is possible that the Hoye Canyon rhino is conspe-
cific with the small, late Hemphillian species of Te-
leoceras. However, a generic assignment cannot be
made without determining if the Hoye Canyon rhi-
no possessed upper incisors. As such, the Hoye
Canyon rhino is referred to Rhinocerotidae, gen.
and sp. indet.

Family Equidae Gray, 1821
Genus Dinohippus Quinn, 1955
Dinobippus sp. indet.

Figure 9, Table 5

SPECIMEN. From LACM 6993: associated par-
tial L E_2, P2-M3, LACM 145974.

DISCUSSION. The teeth are somewhat dam-
aged, with most having some part of the base miss-
ing (Fig. 9). The crowns are complete in all respects
but the following: (1) I2 is missing a small portion
of the enamel at the lateral edge of the tooth; (2)
P3 is missing a small portion of the anterior hypo-
conid; and (3) Mj is missing part of the posterior
half of the tooth, broken off 22.5 mm below the
occlusal surface. P2_4 and M3 are unworn, and M1-2
are in early wear, which, in the extant domestic
horse, would indicate an age of about 2 years old.

The lower incisors are of typical equid structure,
with well-defined central cusps, convex anterior
enamel borders, and notable lateral tapering of the
occlusal outlines. When placed together, the inci-

Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada ■ 17

B

Figure 9. Dinohippus sp. indet. A, B, partial P2-M3, LACM 145974. A, occlusal view, anterior left. B, labial view,
anterior left. Scale = 10 mm.

sors appear to have formed a rounded dental ar-
cade.

The cheek teeth of the Hoye Canyon horse are
characterized by having the following: (1) size
large, estimated P2-M3 A-P about 170 mm; (2) hyp-
sodont (mesostylar crown height of P4 = 64.6 mm,
estimated crown height of about 70 mm); (3)
the cement layer is thick; (4) the P2 4 ectolophids
do not penetrate the isthmuses between metaconids
and metastylids; (5) the M1-2 occlusal enamel pat-
terns are simple (the other teeth are unworn, but
appear also to have simple enamel patterns, based
on the cross-sectional patterns at the broken bases);
(6) the M|_3 metaconids are notably smaller than

Table 5. Measurements (in mm) of lower teeth of Di-
nohippus sp. indet. (LACM 145974) from Hoye Canyon
(a = approximate).

Tooth

position

A-P

TR

P2

31.8

17.2

P3

29.2

15.1

P4

30.7

15.9

Ma

27.6a

14.5

m2

31.2

12.1

m3

24.0

10.0

metastylids (especially evident about half way
down the crowns from the occlusal surfaces and at
the bases of the teeth); and (7) the M1-3 ectoflexids
are deep, completely penetrating the isthmuses be-
tween the metaconids and metastylids.

The cheek teeth of the Hoye Canyon horse are
indistinguishable in size and morphology from
those of the late Hemphillian “ Dinohippus ” inter-
polatus and Dinohippus leidyanus. These two spe-
cies have very similar lower cheek teeth and are
differentiated from each other primarily by the
morphology of the facial fossae (Kelly, 1998a),
which is unknown for the Hoye Canyon horse. Pre-
vious investigators have suggested that these two
species are conspecific, but their taxonomic status
must await a complete revision of Dinohippus
(Hulbert, 1993; Kelly, 1998a). The lower cheek
teeth of the Hoye Canyon horse differ from those
of Equus Linnaeus, 1758, by having the M1-3 me-
tastylids notably smaller than the metaconids. They
differ from those of the Hipparionini Quinn, 1955,
by having the following characteristics: (1) larger
size; (2) lacking protostylids; (3) relatively smaller
and less separated P2-M3 metaconids and meta-
stylids with the M3_3 metastylids notably smaller
than the metaconids; and (4) deeper Mx_3 ectoflex-
ids, completely penetrating the isthmuses between
the metaconids and metastylids. They differ from
those of the Protohippini Quinn, 1955, by having

Kelly: Hemphillian Fauna from Nevada

18 ■ Contributions in Science, Number 481

the following characteristics: (1) larger size; (2)
more hypsodont; (3) protostylids lacking; and (4)
the P3 4 metaconids and metastylids are about equal
in size and position. The Hoye Canyon horse ap-
pears to represent either “D.” interpolatus or D.
leidyanus. However, a definitive specific assignment
must await the discovery of more complete mate-
rial.

AGE OF FAUNA

The Hoye Canyon Local Fauna consists of the fol-
lowing taxa: the rabbit Pronotolagus nevadensis n.
sp.; the ground squirrel Spermophilus wellingto-
nensis Kelly, 1997; the marmot Marmota korthi n.
sp.; the gopher Parapliosaccomys oregonensis Shot-
well, 1967; the camel Hemiaucheria vera\ PAntilo-
capridae, gen. and sp. indet.; Rhinocerotidae, gen.
and sp. indet.; and the horse Dinohippus sp. indet.

The age of the Hoye Canyon Local Fauna can be
determined by the shared occurrences of certain
taxa within the fauna. The geomyid, Parapliosac-
comys oregonensis, previously was known only
from the late Hemphillian McKay Reservoir Fauna
of Oregon and the late Hemphillian Churchill Butte
Local Fauna from the Desert Mountains of west
central Nevada (Shotwell, 1967; Kelly and Lugaski,
1999). Marmota korthi n. sp. is more derived than
the only known Clarendonian species, Marmota ve-
tus (Kellogg, 1910), but less derived than the Blan-
can to Recent marmot species. Similarly, Pronoto-
lagus nevadensis n. sp. appears to be more derived
than the Clarendonian to early Hemphillian Pron-
otolagus apachensis (Gazin, 1930). The camel,
Hemiaucheria vera, is restricted to the Hemphillian
(Webb, 1974). Most investigators generally regard
the extinction of the Rhinocerotidae in North
America as one of the events to mark the end of
the Hemphillian (Tedford et al., 1987). However,
Madden and Dalquest (1990) reported finding a
single fragment of a rhinoceros tooth as a result of
screen washing matrix from the Blancan Yellow
Quarry, Scurry County, Texas (Beck Ranch Local
Fauna). With the exception of this record, no other
Blancan occurrence of the Rhinocerotidae is known
(Prothero, 1998). Dinohippus sp. indet. from Hoye
Canyon appears to represent either “ Dinohippus ”
interpolatus (Cope, 1893) or Dinohippus leidyanus
(Osborn, 1918), both of which are known only
from the late Hemphillian (Azzaroli, 1988; Hul-
bert, 1993; Kelly, 1998a). The combined presence
of the Rhinocerotidae, Pronotolagus nevadensis n.
sp., Marmota korthi n. sp., Parapliosaccomys ore-
gonensis, Hemiaucheria vera, and Dinohippus sp.
(either “D.” interpolatus or D. leidyanus) indicates
that the fauna is late Hemphillian (late Miocene) in
age.

CONCLUSIONS

An unnamed formation exposed along the western
flanks of the Wellington Hills, Douglas County, Ne-
vada, has previously yielded two mammalian fau-

nas: the late Blancan Wellington Hills Local Fauna
and the early Irvingtonian Topaz Lake Local Fauna
(Kelly, 1997). The northern outcrops of this un-
named unit occur in the Hoye Canyon area, and
the southern outcrops occur from Risue Canyon to
about 3 km north of Long Dry Canyon (Kelly,
1997). Foothills covered with Quaternary alluvium
separate the northern outcrops from the southern
outcrops. Previously, only a single locality (LACM
6993 = UCMP V-95013) was known from the
Hoye Canyon area, and this locality did not pro-
duce any age-diagnostic fossils. Because the section
containing LACM 6993 is isolated and cannot be
traced laterally, Kelly (1997) only questionably cor-
related LACM 6993 with the late Blancan localities
in the southern exposures. Quarrying at LACM
6993 and the discovery of an additional locality in
the Hoye Canyon area has now produced new
specimens that indicate the assemblage from Hoye
Canyon is late Hemphillian and not Blancan in age.
This assemblage is named the Hoye Canyon Local
Fauna.

The recognition of a late Hemphillian fauna from
the Hoye Canyon section allows a reevaluation of
the geologic history and biostratigraphy of the un-
named formation of the Wellington Hills. Based on
the faunas, the unnamed formation appears to have
been deposited from at least the late Hemphillian
(late Miocene) to the early Irvingtonian (early Pleis-
tocene) or about 7 to 1.8 Ma. The Hoye Canyon
localities of the northern outcrops must occur strat-
igraphically below the late Blancan localities of the
southern outcrops. This fact indicates that Kelly
(1997) underestimated the total thickness of the un-
named formation by at least several hundred me-
ters.

The unnamed formation of the Wellington Hills
dips westward and, in the northern exposures, un-
conformably overlies uplifted Miocene andesitic
rocks, while, in the southern exposures, it uncon-
formably overlies uplifted Mesozoic granitic and
meta volcanic basement rocks (Kelly, 1997). The
steep escarpment along the eastern border of the
Wellington Hills demarcates the position of a large
northerly-trending normal fault zone that extends
from the Stillwater Range in the south to the Vir-
ginia Range in the north (Gilbert and Reynolds,
1973). Gilbert and Reynolds (1973) proposed that
Quaternary activity along the frontal fault zone re-
sulted in the uplifting of the Wellington Hills. As
the Wellington Hills were uplifted, so was the un-
named formation, as indicated by its westward tilt-
ing. Based on biostratigraphic and lithologic evi-
dence, Kelly (1997) proposed that the major uplift-
ing of the Wellington Hills could have occurred
from sometime prior to the late Blancan, the age of
the oldest known fauna at the time, to as late as or
later than the early Irvingtonian, the age of the
youngest fauna. With the recognition of a late
Hemphillian fauna from the lower part of the un-
named formation of the Wellington Hills, it now
appears that the major uplifting could have oc-

Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada ■ 19

curred from the late Hemphillian to as late as or
later than the early Irvingtonian.

ACKNOWLEDGMENTS

I am indebted to John M. Harris of the George C. Page
Museum, William W. Korth of the Rochester Institute of
Paleontology, E. Bruce Lander of Paleo Environmental As-
sociates, Inc. and a Research Associate at the Natural His-
tory Museum of Los Angeles County (LACM), and Don-
ald R. Prothero of Occidental College for their construc-
tive comments and advice on the original draft of this
report. Special thanks are given to Samuel A. McLeod and
David P. Whistler of the LACM for their considerate help
in curating the fossil specimens. Collections at the LACM
and University of California, Museum of Paleontology,
were made available by David P. Whistler and Patricia
Holroyd, respectively.

LITERATURE CITED

Audubon, J.J., and J. Bachman. 1841. Descriptions of new
species of quadrupeds inhabiting North America.
Proceedings of the Academy of Natural Sciences,
Philadelphia 1:92-104.

Azzaroli, A. 1988. On the equid genera Dinohippus
Quinn 1955 and Pliohippus Marsh 1874. Bollettino
della Societa Palaeontologica Italiana 27:61-72.
Black, C.C. 1963. A review of the North American Ter-
tiary Sciuridae. Bulletin of the Museum of Compar-
ative Zoology 130:109-248.

Blumenbach, J.F. 1779. Handbuch der Naturgeschichte.

Gottigen: Johann Christian Dieterich, 559 pp.

Cope, E.D. 1873. On some new fossil Ungulata. Proceed-
ings of the Academy of Natural Sciences, Philadel-
phia 19:258-2 63.

— . 1893. A preliminary report on the vertebrate pa-
leontology of the Llano Estacado. Geological Survey
of Texas Annual Report 4:1-136.

Cuvier, F. 1825. Des dents des mammiferes considerees
comme caracteres zoologiques. Paris: Deterville,
255 pp.

Dalquest, W.W. 1979. Identification of genera of Ameri-
can rabbits of Blancan age. Southwestern Naturalist
24:275-278.

Dice, L.R., and D.S. Dice. 1935. The lower cheek teeth of
the fossil hare Palaeolagus haydeni. Papers of the
Michigan Academy of Science, Arts, and Letters 20:
455-463.

-. 1941. Age changes in the teeth of the cottontail

rabbit. Papers of the Michigan Academy of Science,
Arts, and Letters 26:219-228.

Douglas, E. 1908. Rhinoceroses from the Oligocene and
Miocene deposits of North Dakota and Montana.
Annals of the Carnegie Museum 4:256-266.

Gazin, C.L. 1930. A Tertiary vertebrate fauna from the
upper Cuyama drainage basin, California. Carnegie
Institution of Washington, Publication 404:55-76.
Gervais, H., and F. Ameghino. 1880. Les mammiferes fos-
siles de L’Amerique du Sud. Paris: Sary, 225 pp.
Gianella, V.P. 1936. Geology of the Silver City district and
the southern portion of the Comstock Lode, Nevada.
University of Nevada, Bulletin 30:1-108.

Gilbert, C.M., and M.W. Reynolds. 1973. Character and
chronology of basin development, western margin of
the Basin and Range Province. Geological Society of
America, Bulletin 84:2489-2510.

Halsey, J.H. 1953. Geology of parts of the Bridgeport,
California, and Wellington, Nevada, Quadrangles.

Ph.D. dissertation, University of California, Berkeley,
498 pp.

Hatcher, J.B. 1894. A median-horned rhinoceros from the
Loup Fork beds of Nebraska. American Geologist
13:149-150.

Hay, O.P. 1921. Descriptions of Pleistocene Vertebrata,
types or specimens of which are preserved in the
United States National Museum. Proceedings of the
United States National Museum 59:617-638.

Hibbard, C.W. 1939. Four new rabbits from the upper
Pliocene of Kansas. American Midland Naturalist
21:506-413.

— . 1963. The origin of the P3 pattern of Sylvilagus,

Caprolagus, Oryctolagus, and Lepus. Journal of
Mammalogy 44:1-15.

Hibbard, C.W., and C.B. Schultz. 1948. A new sciurid of
Blancan age from Kansas and Nebraska. Bulletin of
the University of Nebraska State Museum 3:19-29.

Hulbert, R.C. 1993. Taxonomic evolution in North Amer-
ican Neogene horses (subfamily Equinae): The rise
and fall of an adaptive radiation. Paleobiology 19:
216-234.

Kellogg, L. 1910. Rodent fauna of the late Tertiary beds
of Virgin Valley and Thousand Creek, Nevada. Uni-
versity of California, Bulletin of the Department of
Geological Sciences 5:411-437.

Kelly, T.S. 1997. Additional late Cenozoic (latest Hem-
phillian to earliest Irvingtonian) mammals from
Douglas County, Nevada. Paleobios 18:1-31

— - — . 1998a. New Miocene equid crania from Califor-
nia and their implications for the phylogeny of the
Equini. Natural History Museum of Los Angeles
County, Contributions in Science 473:1-44.

. 1998b. New Miocene mammalian faunas from

west central Nevada. Journal of Paleontology 72:
137-149.

Kelly, T.S., and T. Lugaski. 1999. A Hemphillian (late
Miocene) fauna from the Desert Mountains, west
central Nevada. Bulletin of the Southern California
Academy of Science 98:1-14.

Korth, W.W. 1987. New rodents (Mammalia) from the
late Barstovian (Miocene) Valentine Formation, Ne-
braska. Journal of Paleontology 61:1058-1064.

— . 1994. The Tertiary record of rodents in North

America, topics in geobiology 12. New York: Ple-
num Press, xi + 319 pp.

— . 1998. Rodents and lagomorphs (Mammalia) from

the late Clarendonian (Miocene) Ash Hollow For-
mation, Brown County, Nebraska. Annals of the
Carnegie Museum 67:299-348.

Korth, W.W., and R. Reynolds. 1994. A hypsodont gopher
(Rodentia, Geomyidae) from the Clarendonian
(Miocene) of California. San Bernardino County
Museum Association Special Publication 94-1:91-
95.

Kurten, B., and E. Anderson. 1980. Pleistocene mammals
of North America. New York: Columbia University
Press, 442 pp.

Leidy, J. 1856. Notices of remains of extinct Mammalia,
discovered by Dr. F. V. Hayden in Nebraska Terri-
tory. Proceedings of the Academy of Natural Scienc-
es, Philadelphia 8:88-90.

Linnaeus, C. 1758. Systema naturae per regna tria natu-
rae, secundum clases, ordines, genera, species, cum
characteribus, differentiis, synonymis, locis. Tomus
I. Editio decima, reformata. Stockholm: Laurentii
Salvii, 824 pp.

Madden, C.T., and W.W. Dalquest. 1990. The last rhi-

20 ■ Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada

noceros in North America. Journal of Vertebrate Pa-
leontology 10:266-267.

Matthew, W.D. 1932. A review of the rhinoceroses with
a description of Aphelops material from the Pliocene
of Texas. University of California, Bulletin of the
Department of Geological Sciences 20:411-444.

Matthew, W.D., and H.F. Osborn. 1909. Faunal lists of
the Tertiary Mammalia of the West. United States
Geological Survey, Bulletin 361:91-138.

Marsh, O.C. 1871. Notice of some new fossil mammals
and birds from the Tertiary formation of the West.
American Journal of Science 2:120-127.

Moore, J.G. 1969. Geology and mineral deposits of Lyon,
Douglas, and Ormsby Counties, Nevada. Nevada
Bureau of Mines and Geology, Bulletin 75:1-45.

Osborn, H.F. 1904. New Miocene rhinoceroses with re-
vision of known species. Bulletin of the American
Museum of Natural History 13:307-326.

. 1918. Equidae of the Oligocene, Miocene, and

Pliocene of North America, iconographic type revi-
sion. Memoirs of the American Museum of Natural
History 2:1-326.

Prothero, D.R. 1998. Rhinocerotidae. In Evolution of Ter-
tiary mammals of North America, ed. C.M. Janis,
K.M. Scott, and L.L. Jacobs, 595-605. Cambridge,
New York: Cambridge University Press, x + 691 pp.

Quinn, J.H. 1955. Miocene Equidae of the Texas Gulf
Coastal Plain. University of Texas, Bureau of Eco-
nomic Geology Publication 5516:1-102.

Rafinesque, C.S. 1817. Extracts from the journal of Mr.
Charles Le Roge. American Monthly Magazine 2:45.

Repenning, C.A. 1962. The giant ground squirrel Paene-
marmota. Journal of Paleontology 36:540-556.

Shotwell, J.A. 1967. Late Tertiary geomyoid rodents of
Oregon. University of Oregon, Bulletin of the Mu-
seum of Natural History 9:1-51.

Storer, J.E. 1973. The entoptychne geomyid Lignimus
(Mammalia: Rodentia) from Kansas and Nebraska.
Canadian Journal of Earth Sciences 10:72-83.

Tanner, L.G. 1967. A new species of rhinoceros, Aphelops
kimballensis, from the latest Pliocene of Nebraska.
Bulletin of the University of Nebraska State Museum
6:1-16.

— . 1975. Cenozoic mammals from the central Great

Plains, part 2. Stratigraphic occurrences of Teleocer-
as, with a new Kimballian species from Nebraska.
Bulletin of the University of Nebraska State Museum
10:23-33.

Tedford, R.H., T. Galusha, M.F. Skinner, B.E. Taylor, R.W.
Fields, J.R. Macdonald, J.M. Rensberger, S.D. Webb,
and D.P. Whistler. 1987. Faunal succession and
biochronology of the Arikareean through Hemphil-
lian interval (late Oligocene through earliest Plio-
cene), North America. In Cenozoic mammals of
North America, ed. M.O. Woodburne, 153-210.
Berkeley: University of California Press, 336 pp.

Voorhies, M.R. 1988. The giant marmot Paenemarmota
sawrockensis (new combination) in Hemphillian de-
posits of northeastern Nebraska. Transactions of the
Nebraska Academy of Sciences 16:165-172.

Voorhies, M.R., and C.L. Timperley. 1997. A new Pron-
otolagus (Lagomorpha: Leporidae) and other lepor-
ids from the Valentine Railway Quarries (Barstovian,
Nebraska), and the archaeolagine-leporine transi-
tion. Journal of Vertebrate Paleontology 17:725-
737.

Webb, S.D. 1974. Pleistocene llamas of Florida, with a
brief review of the Lamini. In Pleistocene mammals
of Florida, ed. S.D. Webb, 170-213. Gainesville:
University Presses of Florida, 270 pp.

White, J.A. 1987. The Archaeolaginae (Mammalia, La-
gomorpha) of North America, excluding Archaeo-
lagus and Panolax. Journal of Vertebrate Paleontol-
ogy 7:425-450.

— . 1991. North American Leporinae (Mammalia,

Lagomorpha) from the late Miocene (Clarendonian)
to latest Pliocene (Blancan). Journal of Vertebrate
Paleontology 11:67-89.

Wilson, R.W. 1949. Rodents of the Rincon Fauna, west-
ern Chihuahua, Mexico. Carnegie Institution of
Washington, Publication 584:165-176.

Wood, A.E. 1940. The mammalian fauna of the White
River Oligocene, part III. Lagomorpha. Transactions
of the American Philosophical Society, new series 28:
271-362.

Received 29 June 1999; accepted 13 October 1999.

Contributions in Science, Number 481

Kelly: Hemphillian Fauna from Nevada 121

Natural History Museum

of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

3X
A) H

Number 482
20 April 2000

Contributions
in Science

Revision of the “Apocephalus miricauda-
Group” of Ant-Parasitizing Flies
(Diptera: Phoridae)

Brian V. Brown

-sWtf SON/%

W 10

Natural History Museum

of Los Angeles County

Serial
Publications
ol THE
Natural History
Museum ol
Los Angeles
County

Scientific

Publications

Committee

John Heyning, Deputy Director
for Research and Collections
John M. Harris, Committee Chairman
Brian V. Brown
Kenneth E. Campbell
Kirk Fitzhugh
Karen Wise

Robin A. Simpson and K. Victoria Brown,

Managing Editors

The scientific publications of the Natural History Museum
of Los Angeles County have been issued at irregular in-
tervals in three major series; the issues in each series are
numbered individually, and numbers run consecutively, re-
gardless of the subject matter.

# Contributions in Science, a miscellaneous series of tech-
nical papers describing original research in the life and
earth sciences.

# Science Bulletin, a miscellaneous series of monographs
describing original research in the life and earth sci-
ences. This series was discontinued in 1978 with the
issue of Numbers 29 and 30; monographs are now
published by the Museum in Contributions in Science.

# Science Series, long articles and collections of papers on
natural history topics.

Copies of the publications in these series are sold through
the Museum Book Shop. A catalog is available on request.

The Museum also publishes Technical Reports, a miscel-
laneous series containing information relative to scholarly
inquiry and collections but not reporting the results of
original research. Issue is authorized by the Museum’s Sci-
entific Publications Committee; however, manuscripts do
not receive anonymous peer review. Individual Technical
Reports may be obtained from the relevant Section of the
Museum.

Natural History Museum
of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

Printed at Allen Press, Inc., Lawrence, Kansas
ISSN 0459-8113

Revision of the “Apocephalus miricauda-G roup
of Ant-Parasitizing Flies
(Diptera: Phoridae)

Brian V. Brown1

CONTENTS

ABSTRACT

INTRODUCTION

METHODS AND MATERIALS

SYSTEMATICS

Apocephalus Coquillett

Apocephalus miricauda-groxxp

A. paraponerae- subgroup

A. paraponerae- series

A. paraponerae Borgmeier

A. deceptus new species

A. strongylus new species

A. melinus new species

A. roeschardae new species

A. conecitonis- series

A. conecitonis new species

A. constr ictus new species

A. crassilatus new species

A. dracodermus new species

A. indeptus new species

A. inpalpahilis new species

A. reticulatus new species

Other A. paraponerae- subgroup Species

A. persecutor Borgmeier

A. curtinotus new species

A. secus new species

A. spiculus new species

A. torulus new species

A. succineus new species

A. spatulatus- subgroup

A. incomptus new species

A. striativentris new species

A. hrochus new species

A. fuscipalpis Borgmeier

A. digitalis Borgmeier

A. denotatus new species

A. pachycondylae new species

A. atrimarginatus new species

A. batillus new species

A. emargilatus new species

A. magnicauda new species

A. triangularis new species

1. Natural History Museum of Los Angeles County,
900 Exposition Boulevard, Los Angeles, California 90007.
E-mail: bbrown@nhm.org.

55

2

3

3

4

5

5

5

6

6

8

9

9

10

10

10

12

12

13

13

13

14

14

14

15

16

16

17

17

18

19

20

20

21

21

22

22

23

23

24

25

25

Contributions in Science, Number 482, pp. 1-62
Natural History Museum of Los Angeles County, 2000

A. quadratus Brown 26

A. acanthus new species 26

A. spatulatus Borgmeier 26

A. miricauda-subgroup 27

A. miricauda Borgmeier 27

A. kungae new species 27

A. flexus new species 29

A. orbiculus new species 29

A. meniscus- subgroup 29

A. meniscus new species 29

A. harbiventris new species 31

A. amplidiscus new species 31

A. paldiae new species 32

A. cyclodiscus new species 32

A. lopesi (Borgmeier) 33

A. funditus- subgroup 33

A. funditus new species 33

A. intonsus new species 34

A. spatulicauda- subgroup 35

A. spatulicauda Borgmeier 35

A. inimicus Borgmeier 35

A. brevifrons new species 36

Other Taxa (subgroup unknown) 36

A. densepilosus Borgmeier 36

A. comosus new species 37

A. lobicauda new species 38

A. globosus new species 39

A. maculosus new species 40

A. glabriventris new species 40

A. minutus Borgmeier 40

A. cardiacus new species 41

A. petiolus new species 42

A. gigantivorus new species 42

A. piliventris Borgmeier 42

A. annulatus new species 44

A. contortiventris new species 44

A. eurydomus new species 45

A. conformalis new species 45

A. fenestratus new species 45

A. asyndetus new species 47

A. catholicus new species 48

A. lyratus Borgmeier 49

A. trifidus new species 49

A. tanyurus new species 50

A. contracticauda new species 50

A. indistinctus new species 51

A. dinoponerae new species 51

A. latinsulosus new species 52

Key to females 54

BEHAVIORAL ASPECTS 60

ACKNOWLEDGMENTS 61

LITERATURE CITED 61

ABSTRACT. The “ Apocephalus miricauda- group”, a paraphyletic assemblage, is revised and diagnosed.
Seventy-three species are recognized, including the following 58 new to science: A. acanthus, amplidiscus,
annulatus, asyndetus, atrimarginatus, harbiventris, batillus, brevifrons, brochus, cardiacus, catholicus, com-
osus, conecitonis, conformalis, constrictus, contortiventris, contracticauda, crassilatus, curtinotus, cyclodis-
cus, deceptus, denotatus, dinoponerae, dracodermus, emargilatus, eurydomus, fenestratus, flexus, funditus,
gigantivorus, glabriventris, globosus, incomptus, indeptus, indistinctus, inpalpabilis, intonsus, kungae, la-
tinsulosus, lobicauda, maculosus, magnicauda, melinus, meniscus, orbiculus, pachycondylae, paldiae, pe-
tiolus, reticulatus, roeschardae, secus, spiculus, striativentris, strongylus, tanyurus, torulus, triangularis, and

2 ■ Contributions in Science, Number 482

Brown: Revision of Apocephalus miricauda- group

trifidus. The previously described species, all named by Borgmeier, are A. densepilosus, digitalis, fuscipalpis,
inimicus, lopesi, lyratus, minutus, miricauda, paraponerae, persecutor, piliventris, quadratus, spatulatus,
and spatulicauda. Additionally, a fossil species, A. succineus new species, is described from Dominican
amber. The genus Anaclinusa is considered a synonym of Apocephalus, and the one species, A. lopesi, is
transferred to Apocephalus (new combination). The species A. angularis Borgmeier is considered a synonym
of A. minutus (new synonym). Monophyletic groups within the A. miricauda-gronp include the A. para-
ponerae- subgroup, the A. spatulatus- subgroup, the A. meniscus- subgroup, the A. miricauda-suhgroup, the
A. funditus-subgroup, and the A. spatulicauda- subgroup. The “A. miricauda-growp'’’ is paraphyletic with
respect to the A. attophilus- group. Most species are parasitoids of injured ponerine ants.

INTRODUCTION

The genus Apocephalus Coquillett is a large group
of parasitoid phorid flies, including the original
“ant-decapitating flies” of the New World. Tradi-
tionally, the group has been organized into two
subgenera: Apocephalus s. s. and Mesophora. Sub-
genus Mesophora was treated previously (Brown,
1993, 1994, 1996, 1997a), but only one subsection
of subgenus Apocephalus has been revised (Brown,
1997b).

In this paper I treat the “A. miricauda- group” of
species, a paraphyletic group I proposed previously
(Brown, 1997b; hereafter the quotation marks
around the name of this group, which indicate non-
monophyly, will be omitted). These flies are para-
sitoids, mostly of ponerine ants (Hymenoptera:
Formicidae: Ponerinae), and mostly of injured or
distressed individual workers.

METHODS AND MATERIALS
METHODS

Methods are the same as my previous works on Apo-
cephalus, but a comment on species concepts is necessary.
Some of the proposed new species in this revision are ex-
tremely similar to each other; for instance A. flexus new
species and A. orbiculus new species differ mostly by the
shape of a small internal sclerite (Figs. 51 and 53). Al-
though these differences seem marginal and possibly in-
significant, they are based on characters that are not
known to differ significantly within species. Additionally,
recent work in phorid taxonomy has used extremely nar-
row species definitions (e.g., Disney, 1989), and some ev-
idence exists for cryptic species within otherwise well-de-
fined taxa. In summary, the new species descriptions con-
tained herein are proposals that can be reevaluated when
additional specimens have been collected.

The female ovipositor is described in a similar manner
as that of A. attop hilus-gr oup species (Brown, 1997b), but
one additional structure is present in many A. miricauda-
group species, namely the internal, sclerotized loop de-
rived from sternite 9. In many species it is a round struc-
ture with a broad, moderately sclerotized process (Fig.
106), but in some others it is darkly sclerotized and of
different structure (Figs. 107-109).

PHYLOGENETIC ANALYSIS. The character states of
the A. spatulatus-svbgroxip (Table 1) were analyzed using
the computer program HENNIG-86 (Farris, 1989). Other
cladograms were constructed by hand.

TERMS AND NAMES. The nomenclature of ant spe-
cies was checked against Bolton (1995). One notable
change from normal use is that the more commonly used
Eciton “ burchelli ” (Westwood) has been changed to the
correct E. burchellii, following Bolton.

Geographical coordinates are quoted as decimal de-

Contributions in Science, Number 482

grees, rather than degrees, minutes and seconds (e.g.,
90.5°W, rather than 90°30'W; Crawford, 1983).

BARCODES. In addition to the usual insect labels re-
cording locality information, specimens were labeled with
barcoded insect labels (Thompson, 1994) and data were
recorded in a database. All barcoded labels that begin
with the abbreviation “LACM ENT,” indicate that the
Natural History Museum of Los Angeles County (LACM)
is the institution where the data are stored. Specimens
with barcoded labels beginning “INBIO” have their data
stored at LACM and the Instituto Nacional de Biodiver-
sidad in Costa Rica. To make later recognition of holo-
types easier, I list their individual barcode number in
square brackets.

MATERIALS

This revision is based on adult female specimens. A few
species are also known from males and immatures, but
because of the highly incomplete record of these forms,
they are not treated at this time.

Specimens belong to the following institutions (codens
from Arnett et ah, 1993; curator or collection manager
names in parentheses):

AMNH Department of Entomology, American Muse-
um of Natural History, Central Park West at
79th Street, New York, NY 10024-5192,
U.S.A. (D. Grimaldi)

BHMH Universidade Federal de Minas Gerais, Caixa
Postal 486, 30.161-970, Belo Horizonte, MG,
Brazil (R. Parentoni)

DEBU Department of Environmental Biology, Uni-
versity of Guelph, Guelph, ON, Canada NIG
2W1 (S. Marshall)

EMUS Department of Biology, Utah State University,
Logan, Utah 84322-5305, U.S.A. (W.J. Han-
son)

INBC Instituto Nacional de Biodiversidad, A.P. 22-
3100, Santo Domingo, Heredia, Costa Rica
(M. Zumbado)

INPA Instituto Nacional de Pesquisas da Amazonia,
Estrada do Aleixo, 1756, C.P. 478, 69.011
Manaus, Brazil (J. Rafael)

LACM Entomology Section, Natural History Muse-
um of Los Angeles County, 900 Exposition
Boulevard, Los Angeles, CA 90007, U.S.A. (B.
Brown)

LACM-IP Invertebrate Paleontology Section, Natural
History Museum of Los Angeles County, 900
Exposition Boulevard, Los Angeles, CA
90007, U.S.A. (L. Groves)

MCZC Museum of Comparative Zoology, Harvard
University, Cambridge, MA 02138, U.S.A. (on
indefinite loan to B. Brown)

MIUP Museo de Invertebrados Graham B. Fairchild,
Universidad de Panama, Estafeta Universitar-
ia, Panama (D. Quintero)

Brown: Revision of Apocephalus miricauda- group ■ 3

MUCR Museo de Insectos, Universidad de Costa
Rica, San Pedro, San Jose, Costa Rica (P.S.
Hanson)

MUSM Museo de Historia Natural, Universidad Na-
cional Mayor de San Marcos, Av. Arenales
1267, Apartado 14-0434, Lima-14, Peru (G.
Lamas)

MZSP Museu de Zoologia, Universidade de Sao Pau-
lo, Av. Nazare 481, CP 7172, 01051 Sao Pau-
lo, Brazil (F.C. do Val)

QCAZ Quito Catholic Zoology Museum, Departa-
mento de Biologia, Pontificia Universidad Ca-
tolica del Ecuador, 12 de Octubre y Carrion,
Apto. 2184, Quito, Ecuador (G. Onore)
ROME Department of Entomology, Royal Ontario
Museum, 100 Queen’s Park, Toronto, ON,
Canada M5S 2C6 (D.C. Darling)

TAMU Department of Entomology, Texas A&M Uni-
versity, College Station, TX 77843, U.S.A. (R.
Wharton)

UCMS Department of Ecology and Evolutionary Bi-
ology, Box U-43, University of Connecticut,
Storrs, CT 06269-3043 (J. O’Donnell)

UGGG University of Guyana, Georgetown, Guyana
(M. Tamessar)

UNCB Museo de Historia Natural, Instituto de Cien-
cias Naturales, Universidad Nacional de Co-
lombia, Apto. 7495, Santa Fe de Bogota, Co-
lombia (E. Flores)

USNM United States National Museum, Smithsonian
Institution, Washington, DC 20560, U.S.A.
(on indefinite loan to B. Brown)

USNM-IP Department of Paleobiology, United States Na-
tional Museum, Smithsonian Institution, Wash-
ington, DC 20560, U.S.A. (C. Labandeira)

Additionally, the abbreviation ALAS refers to the Ar-
thropod Survey of La Selva Biological Station, Costa Rica.

SYSTEMATICS

The monophyly of the many species groups of Apo-
cephalus is still insufficiently established. In my pre-
vious work, I proposed that the A. attophilus- group
was characterized by a separated apical sclerite
(Brown, 1997b). Based on a more thorough knowl-
edge of the A. miricauda-group , this character
needs some refinement to allow it to diagnose the
A. attop hilus-gr oup. Furthermore, as I previously
suggested, there is no indication that the A. miri-
cauda-group is monophyletic with respect to the A.
attopbilus-group. Instead, it appears to be a grade
group relative to the A. attophilus- group.

There is some evidence that the A. attopbilus-
group and the A. miricauda-group together form a
monophylum. Both have a distinctive ovipositor
structure, in which most of the ovipositor is evenly,
often lightly sclerotized, but apically there are dark-
ened areas, particularly laterally. This condition is
taken to represent the major synapomorphy of the
species belonging to these two groups.

Another character state shared by the species of
the two groups is that the cercus is straight in
males, in contrast to the curved cercus found in
many other species of the genus. The outgroup con-
dition for this character is a straight cercus; there-

4 ■ Contributions in Science, Number 482

fore this is probably a plesiomorphic character
state.

PHYLOGENETIC HYPOTHESIS

The following character states were analyzed to
provide an explanation of the relationships within
the A. miricauda-group and the A. attophilus- group
(Fig. 119):

1. Ovipositor with apical, lateral darkening (prim-
itive state: without darkenings or darkening not
lateral).

2. Ovipositor elongate (primitive state: ovipositor
relatively short).

3. Dufour’s mechanism elongate, with relatively
large cells (primitive state: short, rounded Du-
four’s mechanism; cells small in center).

This character state is found in some members
of the A. miricauda-group and all the A. atto-
philus-group species I have examined, such as
A. pseudocercus Brown and A. dicbromatus
Brown (Figs. 114-115). Within the A. miricau-
da-group, the primitive state is found in the A.
paraponerae (Fig. 117), A. spatulatus, and A.
miricauda- subgroups. The outgroup condition is
found in other Apocepbalus, such as species of
the A. grandipalpis- group (Fig. 116) and sub-
genus Mesophora.

4. Venter of ovipositor with separate apical sclerite
(primitive state: venter not separated apically).

Some A. miricauda- group species look super-
ficially like they should be placed in the A. at-
topbilus-group, based on the structure of the
dorsum of the ovipositor. Ventrally, however,
these species have sternite 7 completely contig-
uous and unaffected by the separation of the
apical sclerite from the ovipositor.

5. Apical sclerite with sclerotized connection to
ovipositor restricted to thin, medial strip (prim-
itive state: connection more extensive).

6. Anterior section of v-shaped darkening fused
into a single process (primitive state: anterior
section parallel, joining only at apex).

HOST-PARASIT OID RELATIONSHIPS

Most species of the A. miricauda- group are par-
asitoids of injured ants of the subfamily Ponerinae.
In contrast, species of the A. attopbilus-group at-
tack healthy (noninjured) ants of the tribe Attini,
subfamily Myrmecinae. Therefore, within the A. at-
topbilus-group (including the A. miricauda- group),
parasitism of ponerines is a plesiomorphic charac-
ter state.

The hypothesized phylogenetic relationships
among the phorid flies considered herein reflect
similar ideas about ant phylogeny. Ponerines and
myrmecines have been considered closely related in
the past, but more recent reviews have discounted
this relationship (e.g., Baroni Urbani et al., 1992).
Ward (1994), however, has cast doubt upon confi-
dence in the monophyly of Ponerinae and thus the

Brown: Revision of Apocepbalus miricauda-group

rejection of a sister-group relationship between po-
nerines and myrmecines.

If the Ponerinae is indeed paraphyletic relative to
the Myrmecinae, it would lend support to the idea
that host ants and parasitoid flies have coevolved,
at least on a broad scale. This scenario would
match the phylogeny in Fig. 119, in which parasites
of the ponerines are paraphyletic with respect to
parasitoids of the myrmecines. Many details need
to be worked out, however, before this correlation
between host and parasite phylogenies can be ac-
cepted as significant. For instance, although the A.
attophilus- group species attack myrmecine ants,
they attack only one small group, species of the
fungus-growing ant genera Acromyrmex and Atta,
although I have suggested that at least one species
of the closely related Trachymyrmex probably is
also attacked (Brown, 1997b). If coevolution of the
groups was prevalent, then one would expect fur-
ther myrmecines to be hosts. Because most species
of the A. attophilus- group have unknown hosts,
such a scenario of wider host range is still possible.

KEY TO GROUPS OF SUBGENUS
APOCEPHALUS

1 Tergite 6 enlarged, wider than tergite 5, extended

laterally on segment 6, often with greatly en-
larged lateral setae A. pergandei-gmwp

- Tergite 6 of normal size, usually smaller and nar-

rower than tergite 5, usually without greatly en-
larged lateral setae 2

2 Ovipositor dorsally concave, spatulate, with up-
turned apices (Figs. 37, 45)

.............. A. miricauda-gmup (in part)

- Ovipositor dorsally flat or convex ........ 3

3 Ovipositor with apical section of ovipositor well

differentiated from proximal section with clearly
demarked narrowing at junction between the
two, both dorsally and ventrally; sections artic-
ulating by at most a thin, median, sclerotized
strip; v-shaped darkening usually anteriorly
fused to form a long, single process; parasitoids
of attine ants A. attophilus-group

- Ovipositor with apical section not concurrently

differentiated ventrally and dorsally; usually not
differentiated into proximal and apical sections
ventrally; anterior apex of v-shaped darkening
not fused to form long, forward-directed, single
process 4

4 Ovipositor usually blunt-ending dorsally, often
subparallel throughout length, with separated or
differentiated proximal and apical sections dor-
sally; apical section often with lateral darken-
ings; some species with pointed, elongate ventral
postapical sclerite on ovipositor that contrasts
strongly with shape of dorsal apex; mostly par-
asitoids of injured ponerine ants ............

.............. A. miricauda- group (in part)

- Ovipositor usually pointed dorsoapically, usually
tapered posteriorly, not differentiated into sepa-
rate anterior and posterior sections; if not point-
ed apically, then without lateral darkenings; ovi-

Contributions in Science, Number 482

positor ventrally without pointed, postapical
sclerite that contrasts strongly with shape of dor-
sal apex other Apocephalus

Apocepbalus Coquillett

Apocephalus Coquillett, 1901:501, fig. 1. Type spe-
cies: A. pergandei Coquillett, by original desig-
nation. Gender masculine (Ride et ah, 1985: Ar-
ticle 30a iii).

Pseudoplastophora Schmitz, 1915:327, figs. 6, 7.
Type species: P. caudataria Schmitz, by monoty-
py. Synonymized by Borgmeier, 1968.
Pleurophorina Borgmeier, 1969:66, figs. 40-42.
Type species: F. turgida Borgmeier, by original
designation. Synonymized by Brown, 1997b.
Zyziphora Peterson and Robinson, 1976:119, figs.
1-5. Type species: Z. hirtifrons Peterson and
Robinson, by original designation. Synonymized
by Brown, 1992.

Anaclinusa Borgmeier, 1969:63-64, figs. 35-37.
Type species: Anaclinusa lopesi Borgmeier, by
original designation. New synonymy.

The genus Anaclinusa is here considered a synonym
of Apocephalus , based on the single, unusual spe-
cies A. lopesi (below).

Apocephalus miricauda-group

This is not a monophyletic group. It is here recog-
nized in an informal sense to allow discussion of a
group of species of a similar evolutionary grade.
Almost all are parasitoids of injured ponerine ants,
in contrast to their hypothesized relatives, the A.
attophilus- group, which are parasitoids of healthy,
uninjured attine ants.

Apocephalus paraponerae- subgroup

DIAGNOSIS. Ovipositor with ventral postapical
sclerite that is usually apically pointed and trian-
gular in shape (two species have apex truncate).
Lateral, medially directed bars of sclerotization pre-
sent at apex ventrally.

FOSSIL RECORD. There are seven pieces of Do-
minican amber (Oiigocene Miocene) that contain
fossil phorids closely resembling A. paraponerae
Borgmeier or A. deceptus new species. One of these
pieces contains four specimens; another contains
two. These are the oldest known fossils of this ge-
nus (Brown, 1999) and are treated in detail below.

PHYLOGENETIC RELATIONSHIPS. Species of
this subgroup are apparently the most basal taxa in
the A. attophilus- group.

Within the A. paraponerae- subgroup, three fur-
ther divisions are apparent, based on the following
characters (see also Fig. 120):

1. Dorsal, paired, preapical sclerites present on
ovipositor (primitive state: no separate sclerites).
This character state defines a group, herein
called the A. paraponerae- series, containing A.
paraponerae, A. deceptus, A. strongylus new
species, A. melinus new species, and possibly A.

Brown: Revision of Apocephalus miricauda- group ■ 5

roeschardae new species (see discussion in spe-
cies treatment of A. roeschardae). The highly ab-
errant A. tanyurus new species (see below)
might also belong here.

2. Apex of ventral postapical sclerite truncate
(primitive state: apex pointed).

This character state is shared by A. melinus and
A. roeschardae.

3. Sternite 7 with anterior process (primitive state:
sternite 7 broad).

This character state is shared by a number of
species, herein called the A. conecitonis- series:
A. conecitonis, A. constr ictus, A. crassilatus, A.
dracodermus, A. indeptus, A. inpalpabilis, and
A. reticulatus (all new species).

Apocephalus paraponerae- series

Apocephalus paraponerae Borgmeier

(Figs. 1-2, 117)

Apocephalus paraponerae Borgmeier, 1958:324,

figs. 8, 15.

HOLOTYPE. 9, PANAMA: Barro Colorado Is-
land, 19.vi.l 956, C.W. & M.E. Rettenmeyer, Par-
aponera clavata nest entrance (University of Kan-
sas; not examined).

SPECIES RECOGNITION. The ovipositor of
this species is distinctive, with its long, thin, parallel
lateral darkenings. The similar A. deceptus, below,
has thicker lateral darkenings, smaller, thinner
preapical sclerites, and usually more ventral setae
on segment 6.

DESCRIPTION. Body length 1. 4-2.1 mm. Frons
yellow, anterior margin relatively straight. Two
pairs of supra-antennal setae present; lower pair
markedly smaller than upper pair to subequal in
size to upper pair. Flagellomere 1 yellow, round.
Proboscis normal, small. Palpus brown. Dorsum of
thorax light brown; pleuron yellow. Anterior scu-
tellar seta small, fine, subequal to posterior setulae
of scutum. Legs yellowish-brown; apex of hind fe-
mur slightly darker on anterior face. Mean costal
length 0.57 wing length; range 0.53-0.59. Wing
vein R2+3 present. Flalter brown. Abdominal ter-
gites dark brown, except tergite 6, which is partly
to completely yellowish. Venter of abdomen yellow
to gray. Abdominal tergites of normal form. Tergite
3 evenly colored. Tergite 6 completely divided, with
long seta at posterolateral corner. Venter of seg-
ments 3-5 bare. Abdomen without dense lateral se-
tae. Ventral setae of segment 6 long, consisting usu-
ally of single median pair but sometimes up to four
setae present. Ovipositor (Figs. 1-2) straight in lat-
eral view, slightly sclerotized, but with small pair
of darker preapical sclerites. Lateral darkening
thin, subparallel, margin dark, complete. Dorsoap-
ical sclerite of ovipositor not differentiated. Api-
codorsal margin of ovipositor straight. Apicoven-
tral margin of ovipositor straight. Ovipositor with
triangular ventral postapical sclerite. Sternite 7
broad, lightly sclerotized. Dufour’s mechanism

6 ■ Contributions in Science, Number 482

round (Fig. 117). Abdominal glands in segment 5
white, inconspicuous in cleared specimens. Internal
sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Wide-
spread in lowland rain forest throughout the Neo-
tropical Region.

WAY OF LIFE. The life history of this species
was studied by Brown and Feener (1991a, 1991b)
and Feener et al. (1996). It is a parasitoid of injured
and dying workers of Paraponera clavata but also
attacks other ponerine ants such as Ectatomma
tuberculatum (Olivier) and Pachycondyla spp. At
least on the Osa Peninsula in Costa Rica, A. para-
ponerae must subsist entirely on E. tuberculatum,
because P. clavata does not occur there (D. Feener,
personal communication). No flies have been
reared to adulthood from any host except P. cla-
vata, but I have reared larvae from Pachycondyla
villosa (Fabricius) and E. tuberculatum.

Females of A. paraponerae seem to adjust their
clutch size to the size of the available host. When
ovipositing in the much larger workers of P. cla-
vata, females laid an average of 2.09 eggs/host (n
= 11, sd = 1.14), whereas they laid an average of
only 1.13 eggs on the smaller E. tuberculatum (n
= 5), P. apicalis (n = 1; 1 egg), and P. villosa (n =
2); the difference between the number of eggs laid
on P. clavata versus the pooled mean for the smaller
ant species was statistically significant (95%; T :‘=
2.65, p = 0.021, df = 12). As with other species
in the A. miricauda- group, individual flies may be
egg-layers or feeders (see Behavioral Aspects, be-
low).

Based on behavioral characteristics and body
size, Morehead and Feener (1997) have proposed
that what is currently recognized as A. paraponerae
may be a group of at least two races that are un-
dergoing sympatric speciation. Specifically, they
have proposed that individuals attacking P. clavata
and E. tuberculatum are of different races. If this
is true, then differences in the average clutch size
of ovipositing females might be a characteristic of
each race, not of individual choice by females. I
have been unable to find consistent structural char-
acters that separate females attacking P. clavata
from those attacking other hosts.

MATERIAL EXAMINED. BRAZIL: Amazonas:
Manaus, Reserva Ducke, 3.13°S, 60.02°W, 29,8-
15.iv.1992, J. Vidal, Arm. Cola. 1-B-lm (INPA);
Rio de Janeiro, near Desengano State Park, site #2,
21.92°S, 41.80°W, 119, lO.v.1999, B. Brown, in-
jured Pachycondyla apicalis, 200 m (LACM,
MZSP), near Desengano State Park, site #3,
21.96°S, 41.81°W, 1 9, 7.V.1999, B. Brown, injured
Ectatomma tuberculatum, 200 m (LACM), 29,
7.V.1999, B. Brown, injured Ectatomma lugens,
200 m (LACM), 6 9, 7.v. 1999, B. Brown, injured
Pachycondyla villosa, 200 m (LACM); Rondonia:
62 km SE Ariquemes, Fazenda Rancho Grande,
10.35°S, 62.80°W, 5 9, 14-25.ix.1993, B. Harris,
165 m, injured Paraponera clavata (LACM). CO-
LOMBIA: Amazonas: Amacayacu National Park,

Brown: Revision of Apocephalus miricauda-group

Figures 1-9. Ovipositors. Figures 1-2. Apocepbalus paraponerae Borgmeier. 1. Dorsal. 2. Ventral. Figures 3-4. Apoce-
pbalus deceptus new species. 3. Dorsal. 4. Ventral. 5. Apocepbalus strongylus new species, dorsal. Figures 6-7. Apoce-
pbalus melinus new species. 6. Dorsal. 7. Ventral. Figures 8-9. Apocepbalus roeschardae new species. 8. Dorsal. 9. Ventral.

3.82°S, 70.26°W, 17?, 30.viii.1997, B. Brown, G.
Kung, injured Paraponera clavata (LACM), 2?,
2.ix.l997, B. Brown, G. Kung, injured Pachycon-
dyla apicalis (LACM), 8 ? , 3.ix.l997, B. Brown, G.
Kung, injured Paraponera clavata (LACM, UNCB),
1?, 3.ix.l997, B. Brown, G. Kung, injured Pachy-
condyla villosa (LACM), 19, 5.ix.l997, B. Brown,
G. Kung, injured Dolichoderus attelaboides
(LACM), 7 km W Leticia, 4.13°S, 69.9° W, 19,
26.viii.1997, B. Brown, G. Kung, injured Pachy-
condyla crassinoda (LACM), 49, 26.viii.1997, B.
Brown, G. Kung, injured Pachycondyla apicalis
(LACM), 22 km NW Leticia, 4.04°S, 69.99° W, 19,
6.ix.l997, 59, 7.ix.l997, B. Brown, G. Kung, in-
jured Pachycondyla apicalis (LACM); Valle: Rio
Raposo, 3.67°N, 77.08°W, 29, x.1964, 19,
vi.1965, V.H. Lee, light trap (LACM). COSTA
RICA: Guanacaste: Finca Posmompa, near Pitilla,
11.05°N, 85.43°W, 8 9, 5-6.vii.1997, B. Brown, in-

Contributions in Science, Number 482

jured Pachycondyla villosa (LACM), 26 9, 5-
6.vii.l997, B. Brown, J. Paldi, injured Paraponera
clavata (LACM); Heredia: La Selva Biological Sta-
tion, 10.43°N, 84.02°W, 23d, 279, 19-26.V.1988,
B. Brown, injured Paraponera clavata (LACM,
MCZC), 29, 23-26. v.1988, B. Brown, Malaise
trap SSO 50 (LACM) 3d, 3 9, 22.iv.1989, B.
Brown, D. Feener, on Paraponera clavata (LACM),
19, 8-15.V.1989, B. Brown, D. Feener, Malaise
trap SOR@SHO (LACM), 209, 24.vi.1993, at-
tracted to injured Ectatomma tuberculatum
(LACM), 19, 24.vi.1993, B. Brown, injured Pa-
chycondyla apicalis (LACM), 1 9, 26.vi-l.vii. 1993,
B. Brown and D. Feener, Malaise trap #3 (LACM),
19, 27.vi.1993, B. Brown, injured Pachycondyla
villosa (LACM), 519, 27.vi.1993, D. Feener, Par-
aponera experiment vouchers (LACM), 49,
28. vi. 1993, B. Brown, ovipositing on Ectatomma
tuberculatum (LACM), 3 9, 2.vii.l993, B. Brown,

Brown: Revision of Apocepbalus miricauda-group H 7

injured Pachycondyla impressa, 18?, 2.vii.l993,
D.H. Feener, attracted to injured Ectatomma tub-
erculatum (LACM), 1?, 1-15. iii. 1993, Malaise
trap M/10/41, 1 9 , 15.iv-l.v.l993, Malaise trap M/
10/89, 29, 1-15.V.1993, Malaise trap M/10/104,
29, 18.V.1993, ALAS, Malaise trap, M/01/96, M/
02/97 (INBC), 19, 15.v-l.vi.1993, Malaise trap
M/08/114, 19, 15.v-l.vi. 1993, Malaise trap M/10 /
116, 29, l-15.vi.1993, Malaise trap M/10/132,
19, 15.vi-l.vii.1993, Malaise trap M/08/142, 19,

15. vi-l.vii.1993, ALAS, Malaise trap M/10/144
(INBC), 99, 15.vi-l.vii.1993, ALAS, Malaise trap
M/12/146 (INBC), 19, 15.vi-l.vii.1993, ALAS,
Malaise trap M/06/151 (INBC), 19, 15.vii.1993,
ALAS, Malaise trap M/12/162 (INBC), 49,
3-viii.l993, ALAS, Malaise trap M/01/164, M/12/
174 (INBC), 19, 16.ix.1993, ALAS, Malaise trap
M/08/214 (INBC), 19, 15.X.1993, ALAS, Malaise
trap M/12/246 (INBC), 19, l.xii.1993, ALAS,
Malaise trap M/01/276 (INBC), 19, 17.vii.1995,
ALAS, Malaise trap M/10/407 (INBC), 19,

16. X.1995, ALAS, Malaise trap M/07/476 (INBC),

69, 25.vi.1997, B. Brown, J. Paldi, injured Ecta-
tomma tuberculatum (LACM), 29, 7.vii.l998,
ALAS, light L/l 8/415 (INBC), 19, 15.L1998,
ALAS, light L/PP/274 (INBC), 19, 9.vii.l998,
ALAS, light L/09/418 (INBC), 29, 22.vii.1998,
ALAS, light L/l 7/426 (INBC), 19, 15.X.1998,
ALAS, light L/08/489, Rara Avis, 10.28°N,
84.04°W, 19, 18-235.1989, D.A. Grimaldi
(AMNH), 29, 10.vii.1993, B. Brown, attracted to
injured Paraponera clavata (LACM), Estacion
Magsasay, Parque Nacional Braulio Carrillo, 1 9 ,
iii. 1991, A. Fernandez (INBC); Puntarenas: Coo-
pemarti, 8 km S puente de Rio Rincon, 8.63°N,
83.47°W, 19, ii.1991, P. Hanson, Malaise trap, 30
m, primary rainforest (LACM), 3 km SW Rincon,
8.68°N, 83.48°W, 19, xii.1991, P. Hanson, Mal-
aise trap (LACM), 5 km SW Rincon, 8.7°N,
83.51°W, 109, 4.vi.l998, B. Brown, injured Pa-
chycondyla apicalis (LACM), Sirena, 8.48°N,
83.60°W, 169, 9.vii.l993, D. Feener, injured Ec-
tatomma tuberculatum (LACM). ECUADOR:
Napo: Yasuni Biological Research Station, 0.67°S,
76.39°W, 1(3, 3 9, 23.V.1996, B. Brown, injured
Paraponera clavata (LACM), 19, 26.V.1996, J.
Roschard, injured Pachycondyla apicalis (LACM);
Pichincha: Rio Palenque Science Center, 0.60°S,
79.35°W, 19, 25.iv-6.vi.1996, P. Hibbs, Malaise
trap, 200 m (LACM); Sucumbios: Anaga, 0.48°S,
76.38°W, 169, 10.ix.1997, P. DeVries, injured Par-
aponera clavata (LACM), Limoncocha, 0.40°S,
76.58°W, 3c3, 109, 27.vii.1970, C. Rettenmeyer,
Paraponera clavata, #4606 (LACM), Sacha Lodge,
0.50°S, 76.50°W, 19, 22.ii-4.iii.1994, 3 9, 13-
23.vi.1994, 19, 13-25.vii.1994, 19, 10-

21. xi. 1994, P. Hibbs, Malaise trap (LACM,
QCAZ). GUYANA: Berbice: Dubulay Ranch,
5.68°N, 57.86°W, 8 9, 185.1999, B. Brown, injured
Paraponera clavata (LACM, UGGG), Warniabo
Creek, Dubulay Ranch, 5.66°N, 57.88°W, 16-
205.1999, B. Brown, M. Sharkey, Malaise trap #9

8 ■ Contributions in Science, Number 482

(LACM). PANAMA: Canal Zone: Barro Colorado
Island, 9.17°N, 79.83°, 19 [no collector or date],
associated with Ectatomma tuberculatum (LACM),
3 9, 27.vi.1956, C. and M. Rettenmeyer, Parapo-
nera clavata (LACM), 19, vii.1967, W.W. Wirth,
light trap (USNM), 1(3, 59, 105.1985, D.H. Fee-
ner, on Paraponera, #0873 (LACM), 19, 10-
17.iii.1993, J. Pickering, Malaise trap #957
(LACM), 6(3, 69, 23.vi.1996, S. Morehead, in-
jured Paraponera clavata (LACM); Darien: Cruce
de Mono, 7.92°N, 77.62°W, 19, 65i-4.iii.1993, R.
Cambra, J. Coronado, Malaise trap (MIUP); San
Bias: Nusagandi Reserve, 9.33°N, 79°W, 19, 16-
23.iv.1994, J. Pickering, Malaise trap #2862
(LACM). PERU: Madre de Dios: Pakitza, 11.94°S,
71.28°W, 99, 13. ii. 1992, 69, 14.ii.1992, B.
Brown, D. Feener, injured Paraponera clavata
(LACM, MUSM, USNM), 49, 27.ii.1992, B.
Brown, D. Feener, injured Pachycondyla crassinoda
(LACM), 2 9 , 28.ii-4.iii. 1992, B. Brown, D. Feener,
Malaise trap #1 (LACM). VENEZUELA: Amazon-
as: Rio Mavaca Camp, 2.03°N, 65. 10°W, 19, 16-
27.iii.1989, D.A. Grimaldi (AMNH).

Apocephalus deceptus new species

(Figs. 3-4)

SPECIES RECOGNITION. This species most
closely resembles A. paraponerae but has thicker
lateral darkenings and more ventral setae on seg-
ment 6.

DESCRIPTION. Body length 1.5-1. 8 mm. Frons
yellow, anterior margin relatively straight. Two
pairs of supra-antennal setae present; lower pair
markedly smaller than upper pair. Flagellomere 1
yellow, round. Proboscis normal, small. Palpus
brown. Dorsum of thorax yellow; pleuron white.
Anterior scutellar seta small, fine, subequal to pos-
terior setulae of scutum. Legs yellowish-brown;
apex of hind femur with abrupt darkening on an-
terior face. Mean costal length 0.53 wing length;
range 0.53-0.54. Wing vein R2+3 present. Halter
brown. Abdominal tergites dark brown, except ter-
gite 6, which is partly to completely yellowish. Ven-
ter of abdomen gray. Abdominal tergites of normal
form. Tergite 3 evenly colored. Tergite 6 anteriorly
and posteriorly emarginate, with long seta at pos-
terolateral corner. Venter of segments 3-5 bare. Ab-
domen without dense lateral setae. Ventral setae of
segment 6 long, consisting of several setae in a
straight row. Ovipositor (Figs. 3-4) straight in lat-
eral view, slightly sclerotized, but with small pair
of darker preapical sclerites. Lateral darkening
thin, subparallel (but slightly broader than those of
A. paraponerae ), margin dark, complete. Dorsoap-
ical sclerite of ovipositor not differentiated. Api-
codorsal margin of ovipositor straight. Apicoven-
tral margin of ovipositor straight. Ovipositor with
triangular ventral postapical sclerite. Sternite 7
broad, lightly sclerotized. Abdominal glands in seg-
ment 5 white, inconspicuous in cleared specimens.
Internal sclerite not seen.

Brown: Revision of Apocephalus miricauda-group

GEOGRAPHICAL DISTRIBUTION. This spe-
cies is known only from Amazonian Ecuador.

WAY OF LIFE. Females of this species are at-
tracted to injured workers of Pachycondyla com -
mutata (Roger), the presumed host.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for deceive, referring to the
deceptive similarity of this species to A. paraponer-
ae.

HOLOTYPE. 9, ECUADOR: Sucumbios: Ana-
gu, 0.48°S, 76.38°W, ll.ix.1997, P. DeVries, in-
jured Pachycondyla commutata [LACM ENT
024208] (LACM).

PARATYPES. ECUADOR: Napo: Cuyabeno,
0.2°S, 76.3°W, 19, 14.vi.1996, J. Roschard, injured
Pachycondyla commutata (LACM); 1 9 , same data
as holotype (LACM).

Apocephalus strongylus new species

(Fig. 5)

SPECIES RECOGNITION. The combination of
broad lateral darkenings and the pair of preapical,
dorsal sclerites of the ovipositor serve to distinguish
this species from its closest relatives.

DESCRIPTION. Body length 1.4 mm. Frons yel-
low, anterior margin relatively straight. Two pairs
of supra-antennal setae present; lower pair sub-
equal in size to upper pair. Flagellomere 1 light
brown, round. Proboscis normal, small. Palpus
brown. Dorsum of thorax light brown; pleuron
white. Anterior scutellar seta small, fine, subequal
to posterior setulae of scutum. Legs yellowish-
brown; apex of hind femur with abrupt darkening
on anterior face. Mean costal length 0.5 wing
length. Wing vein R2+3 present. Halter brown. Ab-
dominal tergites dark brown, except tergite 6,
which is partly to completely yellowish. Venter of
abdomen yellow. Abdominal tergites of normal
form. Tergite 3 evenly colored. Tergite 6 anteriorly
emarginate, with long seta at posterolateral corner.
Venter of segments 3-5 bare. Abdomen without
dense lateral setae. Ventral setae of segment 6 long,
consisting of a complete ventral and lateral row.
Ovipositor (Fig. 5) straight in lateral view, slightly
sclerotized, but with small pair of darker preapical
sclerites. Lateral darkening broadened, rounded lat-
erally, margin dark, complete. Dorsoapical sclerite
of ovipositor not differentiated. Apicodorsal mar-
gin of ovipositor straight. Apicoventral margin of
ovipositor drawn out into pointed process. Ovipos-
itor with triangular ventral postapical sclerite. Ster-
nite 7 broad, lightly sclerotized. Abdominal glands
in segment 5 white, inconspicuous in cleared spec-
imens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in Brazil.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Greek strongylos for round,
referring to the rounded apical portion of the ovi-
positor.

Contributions in Science, Number 482

HOLOTYPE. 9, BRAZIL: Para: Tucurui,
3.83°S, 49.67°W, 20.vh-8.viii.1982, J. Vidal, CDC
trap [LACM ENT 038362] (INPA).

Apocephalus melinus new species

(Figs. 6-7)

SPECIES RECOGNITION. This species can be
recognized by the brown-colored lateral darkenings
and the large, black, round preapical sclerites of the
ovipositor.

DESCRIPTION. Body length 1.0-1. 5 mm. Frons
yellow, anterior margin relatively straight. Two
pairs of supra-antennal setae present; lower pair
markedly smaller than upper pair. Flagellomere 1
yellow, round. Proboscis normal, small. Palpus
brown. Dorsum of thorax light brown; pleuron yel-
low. Anterior scutellar seta small, fine, subequal to
posterior setulae of scutum. Legs yellowish-brown;
apex of hind femur with abrupt darkening on an-
terior face. Mean costal length 0.54 wing length;
range 0.51-0.6. Wing vein R2+3 present. Halter
brown. Abdominal tergites dark brown, except ter-
gite 6, which is partly to completely yellowish. Ven-
ter of abdomen yellow to gray. Abdominal tergites
of normal form. Tergite 3 evenly colored. Tergite 6
completely divided, with short setae at posterolat-
eral corner. Venter of segments 3-5 bare. Abdomen
without dense lateral setae. Ventral setae of seg-
ment 6 long, consisting of four long setae emanat-
ing from a small sternite. Ovipositor (Figs. 6-7)
straight in lateral view, slightly sclerotized, but with
small pair of darker preapical sclerites. Lateral
darkening thin, subparallel, margin yellowish
brown, complete. Dorsoapical sclerite of ovipositor
not differentiated. Apicodorsal margin of oviposi-
tor straight. Apicoventral margin of ovipositor
straight. Ovipositor ventrally with apically truncate
postapical sclerite. Sternite 7 broad, lightly sclero-
tized. Dufour’s mechanism round. Abdominal
glands in segment 5 white, inconspicuous in cleared
specimens.

GEOGRAPHICAL DISTRIBUTION. Known
from Amazonian Ecuador and Colombia.

WAY OF LIFE. Males and females were attracted
to injured workers of Dolichoderus attelahoides
(Fabricius), and in one instance, D. decollatus
Smith. Larvae were reared from D. attelahoides.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for honey-colored, referring
to the yellowish-brown color of the ovipositor’s lat-
eral darkenings.

HOLOTYPE. 9, ECUADOR: Napo: Yasuni Bi-
ological Research Station, 20.V.1 996, B. Brown, in-
jured Dolichoderus attelahoides [LACM ENT
053807] (LACM).

PARATYPES. COLOMBIA: Amazonas: Ama-
cayacu National Park, 3.82°S, 70.26°W, 19,
3 1 .viii. 1997, B. Brown, G. Kung, injured Dolicho-
derus decollatus (LACM), Id, 99, 5.ix.l997, B.
Brown, G. Kung, injured D. attelahoides (LACM,
UNCB), 22 km NW Leticia, 4.04°S, 69.99° W, 19,

Brown: Revision of Apocephalus miricauda-gvoxxp ■ 9

28.viii.1997, B. Brown, G. Kung, injured Pachy-
condyla commutata (LACM). ECUADOR: 58,
12$, same data as holotype (LACM, MCZC,
QCAZ, USNM).

Apocepbalus roeschardae new species

(Figs. 8-9)

SPECIES RECOGNITION. This species can be
recognized by the truncate apex of the ventral post-
apical sclerite, its dark overall color, and the large,
medial sclerotization instead of paired sclerites dor-
sally.

DESCRIPTION. Body length 1. 4-2.0 mm. Frons
yellow, anterior margin relatively straight. One pair
of supra-antennal setae present. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus brown.
Dorsum of thorax yellow; pleuron light brown. An-
terior scutellar seta small, fine, subequal to poste-
rior setulae of scutum. Legs yellowish-brown; apex
of hind femur of even color anteriorly. Mean costal
length 0.52 wing length; range 0.49-0.55. Wing
vein R2+3 present. Halter mostly yellow but with
dark brown spot on knob. Abdominal tergites dark
brown. Venter of abdomen dark gray. Abdominal
tergites of normal form. Tergite 3 evenly colored.
Tergite 6 anteriorly emarginate, with long seta at
posterolateral corner. Venter of segments 3-5 bare.
Abdomen without dense lateral setae. Ventral setae
of segment 6 long, consisting of several setae em-
anating from a small sternite. Ovipositor (Figs. 8-
9) straight in lateral view, with broad median scler-
ite. Lateral darkening thin, subparallel, margin
dark, complete. Dorsoapical sclerite of ovipositor
not differentiated. Apicodorsal margin of oviposi-
tor straight. Apicoventral margin of ovipositor
straight. Ovipositor ventrally with apically truncate
postapical sclerite. Sternite 7 round. Abdominal
glands in segment 5 white, inconspicuous in cleared
specimens. Internal sclerite a simple, round loop.

GEOGRAPHICAL DISTRIBUTION. Known
from Atlantic coastal Brazil, Ecuador, and Colom-
bia.

WAY OF LIFE. Adult females are attracted to
injured workers of the myrmicine ant, Cephalotes
atratus (Linnaeus). I observed one female ovipos-
iting in the abdomen of the host.

PHYLOGENETIC RELATIONSHIPS. It is pos-
sible that A. roeschardae is the sister-species of A.
melinus, although this is only one of the most par-
simonious conclusions from the data presented
above. The relationship is supported by the pecu-
liarly truncate ventral postapical sclerite in these
two species. It is contradicted by the lack of paired
sclerites in A. roeschardae. The abdomen of A.
roeschardae is unusually colored, being completely
dark brown, in contrast to that in other A. miri-
cauda-group species, which have at least the ventral
membrane yellow. Also, the sclerotized portions of
the ovipositor are unusually robust. Such changes
might have modified the “missing” character states

10 ■ Contributions in Science, Number 482

to states unrecognizable as homologous with those
found in A. melinus.

DERIVATION OF SPECIFIC EPITHET. This
species is dedicated to Ms. Jacqueline Roschard,
who helped with field work in Ecuador and inde-
pendently collected many new parasitic phorid flies.

HOLOTYPE. $, COLOMBIA: Amazonas:
Amacayacu National Park, 3.82°S, 70.26°W,
31.viii.1997, B. Brown, G. Kung, injured Cephalo-
tes atratus [LACM ENT 093613] (UNCB).

PARATYPES. BRAZIL: Rio de Janeiro: near De-
sengano State Park, site #1, 21.87°S, 41.80°W, 2$,
5.V.1999, 6$, 9.V.1999, B. Brown, injured Cepha-
lotes atratus, 200 m (LACM, MZSP). COLOM-
BIA: Amazonas: Leticia, 4.19°S, 69.93°W, 1$,
25.viii.1997, 1$, 8.ix.l997, B. Brown, G. Kung,
injured C. atratus (LACM), 22 km NW Leticia,
4.04°S, 69.99°W, 3$, 28.viii.1997, injured C. atra-
tus (LACM, UNCB). ECUADOR: Napo: Jatun Sa-
cha, 1.07°S, 77.6°W, 1$, 16.ix.1996, J. Roschard,
injured C. atratus (LACM), Yasuni Biological Re-
search Station, 0.67°S, 76.39°W, 1$, 21.V.1996,
1$, 22.V.1996, B. Brown, injured C. atratus, 220
m (LACM, QCAZ). PERU: Madre de Dios: Pakit-
za, 11.94°S, 71.28°W, 1 $ , 4.ix.l991, T. Erwin, M.
Pogue, fog tree #184 (USNM).

Apocephalus conecitonis- series

Apocepbalus conecitonis new species

(Ftg. 10)

SPECIES RECOGNITION. Although similar to
A. secus, this species differs by the shape of the
ovipositor and setation of the venter, as outlined in
the key. The apex of the ovipositor also is markedly
downturned in this species.

DESCRIPTION. Body length 1.0-1. 5 mm. Frons
dark brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 brown, round. Proboscis normal, small.
Palpus yellow. Dorsum of thorax light brown; pleu-
ron yellow. Anterior scutellar seta small, fine, sub-
equal to posterior setulae of scutum. Legs yellow-
ish-brown; apex of hind femur of even color ante-
riorly. Mean costal length 0.46 wing length; range
0.46-0.48. Wing vein R2+3 present. Halter brown.
Abdominal tergites dark brown. Venter of abdo-
men yellow. Abdominal tergites of normal form.
Tergite 3 evenly colored. Tergite 6 anteriorly emar-
ginate, with long seta at posterolateral corner. Ven-
ter of segments 4-5 with row of setae on posterior
margin. Abdomen without dense lateral setae. Ven-
tral setae of segment 6 long, consisting of several
setae in a straight row. Ovipositor (Fig. 10) down-
turned apically, lightly but evenly sclerotized dor-
sally. Lateral darkening short, margin dark, com-
plete. Dorsoapical sclerite of ovipositor not differ-
entiated. Apicodorsal margin of ovipositor pointed.
Apicoventral margin of ovipositor straight. Ovi-
positor with triangular ventral postapical sclerite.
Sternite 7 short, narrow. Dufour’s mechanism not

Brown: Revision of Apocephalus miricauda- group

10,

11.

12,

13.

Figures 10-18. Ovipositors. 10. Apocephalus conecitonis new species, dorsal. Figures 11-12. Apocephalus constrictus
new species. 11. Dorsal. 12. Ventral. 13. Apocephalus crassilatus new species, dorsal, 14. Apocephalus dracodermus new
species, dorsal. Figures 15-16. Apocephalus indeptus new species. 15. Dorsal. 16. Ventral. Figures 17-18. Apocephalus
inpalpabilis new species. 17. Dorsal. 18. Ventral.

seen. Abdominal glands in segment 5 white, incon-
spicuous in cleared specimens. Internal sclerite not
seen.

GEOGRAPHICAL DISTRIBUTION. Known
only from the area of La Selva, Costa Rica.

WAY OF LIFE. Females of this species are asso-
ciated with army ant raids, but it is not clear which
ants they parasitize. It is possible that they are as-
sociated with army ants as a means to procure their
actual hosts (Brown and Feener, 1998). On one oc-
casion I collected these flies in association with a
raid of Eciton lucanoides Emery on Pachycondyla
obscuricornis (Emery), a ponerine ant and possible
host; however, the army ants also were attacking
Paratrechina longicornis (Latreille) and Aphaeno-
gaster araneoides Emery.

DERIVATION OF SPECIFIC EPITHET. The
name is from the Latin word con, for “with,” and

Contributions in Science, Number 482

eciton, the name of army ants, referring to the as-
sociation of these flies with ants of the genus Eciton
Latreille.

HOLOTYPE. 9, COSTA RICA: Heredia: La
Selva Biological Station, 10.43°N, 84.02°W,
ll.v.1989, B. Brown, Eciton burchellii raid [LACM
ENT 012251] (LACM).

PARATYPES. COSTA RICA: Heredia: Chila-
mate, 10.45°N, 84.08°W, 29, v.1989, 19, iv-
vi.1990, P. Hanson, Malaise trap (LACM), La Selva
Biological Station, 10.43°N, 84.02°W, 29,
25.iv.1989, 449, ll.v.1989, 269, 13.V.1989, 19,
14.V.1989, 249, 16.V.1989, 159, 20.V.1989, B.
Brown, D. Feener, Eciton burchellii raid (LACM,
MCZC, MUCR, MZSP, USNM), 19, 2.m.l993,
ALAS, M/05/20 (INBC), 19, l-15.iii.1993, ALAS,
Malaise trap M/07/38, 29, l-15.iv.1993, ALAS,
Malaise trap M/05/68, M/ll/74 (INBC), 19,

Brown: Revision of Apocephalus miricauda-group 111

l.vi.1993, ALAS, Malaise trap M/12/118 (INBC),
49, 15.vi-l.vii.1993, ALAS, Malaise trap M/10/
144 (INBC), 2 9, 3.vii.l993, B. Brown, E. lucano-
ides raid (LACM), 19, 4.iv.l994, ALAS, Malaise
trap M/ll/389 (INBC).

Apocephalus constrictus new species

(Figs. 11-12)

SPECIES RECOGNITION. The unusually
shaped ovipositor, with its extreme narrowing at
midlength, is distinctive for this species (Figs. 11-
12).

DESCRIPTION. Note that this description is
somewhat fragmentary because it is based on a sin-
gle, air-dried specimen. I hope that additional spec-
imens will be found so that the missing character
states can be included.

Body length 1.3 mm. Frons dark brown, anterior
margin relatively straight. Two pairs of supra-an-
tennal setae present; lower pair subequal in size to
upper pair. Flagellomere 1 brown, oval. Proboscis
normal, small. Palpus yellow. Dorsum of thorax
light brown; pleuron light brown. Anterior scutellar
seta small, fine, subequal to posterior setulae of scu-
tum. Legs yellowish-brown; apex of hind femur of
even color anteriorly. Mean costal length 0.54 wing
length. Wing vein R2+3 present. Halter brown. Ab-
dominal tergites dark brown. Venter of abdomen
gray. Abdominal tergites of normal form. Tergite 3
evenly colored. Ovipositor (Figs. 11-12) constrict-
ed at midlength, slightly downturned apically, light-
ly but evenly sclerotized dorsally. Lateral darkening
broadened, rounded laterally, margin dark, com-
plete. Dorsoapical sclerite of ovipositor not differ-
entiated. Apicodorsal margin of ovipositor straight.
Apicoventral margin of ovipositor straight. Ovi-
positor with triangular ventral postapical sclerite.
Sternite 7 triangular, with narrow anterior projec-
tion. Internal sclerite rounded, with a short process.

GEOGRAPHICAL DISTRIBUTION. Known
from a single, mid-elevation site in Costa Rica.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is from a Latin word for constricted, referring
to the marked narrowing of the ovipositor at mid-
length.

HOLOTYPE. 9, COSTA RICA: Puntarenas:
Monteverde, 20-24.vi.1986, W. Hanson, G. Bohart
[LACM ENT 012778] (EMUS).

Apocephalus crassilatus new species

(Fig. 13)

SPECIES RECOGNITION. This species can be
recognized easily by the broad, curved lateral mar-
gins of ovipositor and dense ventral setae of the
abdomen.

DESCRIPTION. Body length 1.6-2. 3 mm. Frons
yellow, anterior margin relatively straight. Two
pairs of supra-antennal setae present; lower pair
subequal in size to upper pair. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus brown.

12 ■ Contributions in Science, Number 482

Dorsum of thorax light brown; pleuron white. An-
terior scutellar seta small, fine, subequal to poste-
rior setulae of scutum. Legs yellowish-brown; apex
of hind femur with abrupt darkening on anterior
face. Mean costal length 0.55 wing length; range

O. 52-0.58. Wing vein R2+3 present. Halter brown.
Abdominal tergites yellow, posteriorly dark brown;
tergite 6 completely yellow. Venter of abdomen yel-
low. Abdominal tergites of normal form. Tergite 3
evenly colored. Tergite 6 completely divided, with
long seta at posterolateral corner. Venter of seg-
ments 3-5 with long, dense setae concentrated me-
dially. Abdomen without dense lateral setae. Ven-
tral setae of segment 6 long, consisting of a com-
plete ventral and lateral row. Ovipositor (Fig. 13)
straight in lateral view, lightly but evenly sclerotized
dorsally. Lateral darkening broadened, rounded lat-
erally, margin dark, complete. Dorsoapical sclerite
of ovipositor not differentiated. Apicodorsal mar-
gin of ovipositor straight. Apicoventral margin of
ovipositor straight. Ovipositor with triangular ven-
tral postapical sclerite. Sternite 7 broad, lightly
sclerotized. Dufour’s mechanism round. Abdominal
glands in segment 5 white, inconspicuous in cleared
specimens.

GEOGRAPHICAL DISTRIBUTION. There are
records of this species from Costa Rica, Panama,
Colombia, and Brazil.

WAY OF LIFE. Females were attracted to injured
workers of various species of Pacbycondyla, includ-
ing P. apicalis (Latreille), P. impressa (Roger), P.
unidentata Mayr, and P. villosa.

DERIVATION OF SPECIFIC EPITHET. The
name is a combination of the Latin words crassus
and latus, referring to the wide lateral darkenings.

HOLOTYPE. 9, COSTA RICA: Heredia: La
Selva Biological Station, 10.43°N, 84.02°W,
22.vi.1993, B. Brown, injured Pacbycondyla villosa
[LACM ENT 001551] (LACM).

PARATYPES. BRAZIL: Minas Gerais: Belo Ho-
rizonte, Estacao Ecologica, UFMG campus, 19,2-
5.vii.l993, S.D. Gaimari, Malaise trap (BHMH).
COLOMBIA: Amazonas: Amacayacu National
Park, 3.82°S, 70.26°W, 29, 2.ix.l997, B. Brown,
G. Kung, injured Pacbycondyla apicalis (LACM,
UNBC). COSTA RICA: Alajuela: 20 km S Upala,
19, 11-21. vi.1991, F.D. Parker [Malaise trap]
(EMUS); Guanacaste: Finca Posmompa, near Pitil-
la, 11.05°N, 85.43°W, 29, 6.vii.l997, B. Brown, J.
Paldi, injured P. villosa (LACM); Heredia: La Selva
Biological Station, 10.43°N, 84.02°W, 19,
24.vi.1993, 29, 6.vii.l993, B. Brown, injured P.
apicalis (LACM) 19, l.xi.1993, ALAS, Malaise
trap M/02/249 (INBC), 29, 21.iii.1995, B. Brown,
injured P. villosa (LACM), Plastico, 10.28°N,
84.02°W, Id, 3 9, 1 1 .vii.1993, B. Brown, injured

P. unidentata (LACM), Rara Avis, 12 km SW Hor-
quetas, 19, 18-235.1989, D.A. Grimaldi, 550 m
(AMNH). PANAMA: Canal Zone: Barro Colorado
Island, 9.17°N, 79.83°W, 19, 25.viii-l.ix.1993, J.
Pickering, Malaise trap #1676 (MIUP), 19,

Brown: Revision of Apocephalus miricauda-gr oxxp

19.viii.1996, S. Morehead, injured P. impressa
(LACM).

Apocephalus dracodermus new species

(Fig. 14)

SPECIES RECOGNITION. This species can be
diagnosed by the reticulate surface and the slight
narrowing at the midpoint of the ovipositor.

DESCRIPTION. Body length 1.1 mm. Frons
dark brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 brown, round. Proboscis normal, small.
Palpus yellow. Dorsum of thorax light brown; pleu-
ron light brown. Anterior scutellar seta small, fine,
subequal to posterior setulae of scutum. Legs yel-
lowish-brown; apex of hind femur slightly darker
on anterior face. Mean costal length 0.44 wing
length. Wing vein R2+3 present. Halter brown. Ab-
dominal tergites dark brown. Venter of abdomen
gray. Abdominal tergites of normal form. Tergite 3
evenly colored. Tergite 6 completely divided, with
long seta at posterolateral corner. Venter of seg-
ments 3-5 with row of setae on posterior margin.
Abdomen without dense lateral setae. Ventral setae
of segment 6 long, consisting of a complete ventral
and lateral row. Ovipositor (Fig. 14) straight in lat-
eral view, lightly but evenly sclerotized dorsally.
Lateral darkening broadened, rounded laterally,
margin dark, complete. Dorsoapical sclerite of ovi-
positor not differentiated. Apicodorsal margin of
ovipositor emarginate on either side, with long me-
dial process. Apicoventral margin of ovipositor
drawn out into pointed process. Ovipositor without
ventral postapical sclerite. Sternite 7 broad, lightly
sclerotized. Abdominal glands in segment 5 white,
inconspicuous in cleared specimens.

GEOGRAPHICAL DISTRIBUTION. Known
only from a single site in Amazonian Peru.

WAY OF LIFE. The single specimen was collect-
ed over a raid of the army ant Labidus spininodis
(Emery). Whether it was attracted to the army ants
or to some victim of their raid is unknown (see
Brown and Feener, 1998).

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin word draco for lizard
and the Greek derma for skin, referring to the re-
ticulate pattern on the ovipositor.

HOLOTYPE. 9, PERU: Madre de Dios: Zona
Reserva Manu, Pakitza, 17.ii.1992, B. Brown, D.
Feener, raid Labidus spininodis, 360 m [LACM
ENT 011897] (MUSM).

Apocephalus indeptus new species

(Figs. 15-16)

SPECIES RECOGNITION. This species has a
slightly rounded look to the apex of the ovipositor.
It is separated from the similar A. inpalpabilis by
having two pairs of supra-antennal setae and by the
lateral bars of the venter of the ovipositor, which
meet medially (Fig. 16).

DESCRIPTION. Body length 1. 0-1.1 mm. Frons
Contributions in Science, Number 482

dark brown, anterior margin relatively straight.
Two pairs of supra-antennal setae present; lower
pair markedly smaller than upper pair. Flagel-
lomere 1 brown, round. Proboscis normal, small.
Palpus yellow. Dorsum of thorax light brown; pleu-
ron white. Anterior scutellar seta small, fine, sube-
qual to posterior setulae of scutum. Legs yellowish-
brown; apex of hind femur of even color anteriorly.
Mean costal length 0.45 wing length; range 0.45-

0. 46. Wing vein R2+3 present. Halter brown. Ab-
dominal tergites dark brown. Venter of abdomen
gray. Abdominal tergites of normal form. Tergite 3
evenly colored. Tergite 6 anteriorly emarginate,
with long seta at posterolateral corner. Venter of
segments 3-5 with a few, scattered setae. Abdomen
without dense lateral setae. Ventral setae of seg-
ment 6 long, consisting of a complete ventral and
lateral row. Ovipositor (Figs. 15-16) slightly down-
turned apically, lightly but evenly sclerotized dor-
sally. Lateral darkening broadened, rounded later-
ally, extended anteriorly on ovipositor; margin
dark, complete. Dorsoapical sclerite of ovipositor
not differentiated. Apicodorsal margin of oviposi-
tor straight. Apicoventral margin of ovipositor
drawn out into pointed process. Ovipositor with
triangular ventral postapical sclerite. Sternite 7
round. Abdominal glands in segment 5 white, in-
conspicuous in cleared specimens. Internal sclerite
not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from single sites in Costa Rica and Ecuador.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for reached or attained, re-
ferring to the lateral darkenings that extend farther
anteriorly onto the ovipositor than those of some
related species.

HOLOTYPE. 9, COSTA RICA: Heredia: La
Selva Biological Station, 10.43°N, 84.02°W, 26.vi-

1. vii.1993, B. Brown, D. Feener, Malaise trap #3
[LACM ENT 013220] (LACM).

PARATYPES. COSTA RICA: Heredia: La Selva
Biological Station, 10.43°N, 84.02°W, 3 9, 16. ii-
2.iii.l993, ALAS, Malaise trap M/07/22, M/05/20,
19, 15.iii.1993, ALAS, Malaise trap M/07/38
(INBC). ECUADOR: Pichincha: 17 km E Santo
Domingo, Tinalandia, 29, 6-13.V.1987, B. Brown,
710 m, windows (LACM).

Apocephalus inpalpabilis new species

(Figs. 17-18)

SPECIES RECOGNITION. This species differs
from the similar A. indeptus by having a single pair
of supra-antennal setae and the lateral bars not
reaching the center of the ovipositor (Fig. 18).

DESCRIPTION. Body length 1.3 mm. Frons
dark brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 brown, oval. Proboscis normal, small. Pal-
pus yellow. Dorsum of thorax light brown; pleuron
yellow. Anterior scutellar seta small, fine, subequal

Brown: Revision of Apocephalus miricauda-groxxp ■ 13

to posterior setulae of scutum. Legs yellow; apex
of hind femur slightly darker on anterior face.
Mean costal length 0.47 wing length. Wing vein
R2+3 present. Halter brown. Abdominal tergites
dark brown. Venter of abdomen yellow. Abdominal
tergites of normal form. Tergite 3 evenly colored.
Tergite 6 anteriorly emarginate, with long seta at
posterolateral corner. Venter of segments 3-5 with
a few, scattered setae. Abdomen without dense lat-
eral setae. Ventral setae of segment 6 long, consist-
ing of a complete ventral and lateral row. Ovipos-
itor (Figs. 17-18) slightly downturned apically,
lightly but evenly sclerotized dorsally. Lateral dark-
ening broadened, rounded laterally, margin dark,
complete. Dorsoapical sclerite of ovipositor not dif-
ferentiated. Apicodorsal margin of ovipositor
straight. Apicoventral margin of ovipositor
straight. Ovipositor with triangular ventral postap-
ical sclerite. Sternite 7 broad, lightly sclerotized.
Abdominal glands in segment 5 white, inconspicu-
ous in cleared specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
only from La Selva, Costa Rica.

WAY OF LIFE. This fly was collected over a
swarm raid of the army ant Eciton burchellii. Its
host is unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin prefix in, meaning not,
combined with palpabilis, meaning touchable, re-
ferring to the ventral, transverse sclerites of the ovi-
positor, which do not touch.

HOLOTYPE. 9, COSTA RICA: Heredia: La
Selva Biological Station, 10.43°N, 84.02°W,
13.V.1989, B. Brown, Eciton burchellii swarm raid
[LACM ENT 011326] (LACM).

Apocephalus reticulatus new species

(Figs. 19-20)

SPECIES RECOGNITION. The namesake retic-
ulations of the ovipositor are visible with light mi-
croscopy; otherwise, the forked ventral process
(Fig. 20) is distinctive.

DESCRIPTION. Body length 1.1-1 .4 mm. Frons
dark brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 brown, round. Proboscis normal, small.
Palpus yellow. Dorsum of thorax light brown; pleu-
ron light brown to white. Anterior scutellar seta
small, fine, subequal to posterior setulae of scutum.
Legs yellowish-brown; apex of hind femur with
abrupt darkening on anterior face (restricted to api-
cal margin). Mean costal length 0.47 wing length;
range 0.44-0.5. Wing vein R2+3 present. Halter
brown. Abdominal tergites dark brown. Venter of
abdomen gray. Abdominal tergites of normal form.
Tergite 3 evenly colored. Tergite 6 anteriorly emar-
ginate, with long seta at posterolateral corner. Ven-
ter of segments 3-4 bare, segment 5 with a single
row of setae on posterior margin. Abdomen with-
out dense lateral setae. Ventral setae of segment 6
long, consisting of several setae in a straight row.

14 ■ Contributions in Science, Number 482

Ovipositor (Figs. 19-20) slightly downturned api-
cally, lightly but evenly sclerotized dorsally and
with reticulate sculpture. Lateral darkening short,
margin dark, complete. Dorsoapical sclerite of ovi-
positor not differentiated. Apicodorsal margin of
ovipositor emarginate on either side, with long me-
dial process. Apicoventral margin of ovipositor
straight. Ovipositor with triangular ventral postap-
ical sclerite. Sternite 7 broad, with forked anterior
process. Dufour’s mechanism not seen. Abdominal
glands in segment 5 white, inconspicuous in cleared
specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
only from a mid-elevation site in Costa Rica.

WAY OF LIFE. Unknown. One specimen was
collected over a raid of the army ant Labidus prae-
dator (F. Smith).

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for net-like, referring to the
pattern on the dorsal surface of the ovipositor.

HOLOTYPE. 9, COSTA RICA: San Jose: Zur-
qui de Moravia, 10.05°N, 84.02°W, l-15.vi.1993,
P. Hanson, Malaise trap, 1600 m [LACM ENT
053562] (LACM).

PARATYPES. COSTA RICA: San Jose: Zurqui
de Moravia, 10.05°N, 84.02°W, 19, ix-x.1990,
19, h.1991, 19, xii.1991-ii.1992, 69, v.1992,
5 9 , vii.1992, 1 9 , iv-v.1993, 5 9 , vi.1993, 3 9 , ix-
x.1993, 19, i.l 996, P. Hanson, Malaise trap, 1600
m (INBC, LACM, MCZC, MUCR, USNM), 19,
8.iii.l995, B. Brown, J. Cantley, over raid of La-
bidus praedator (LACM).

Other A. paraponerae- subgroup Species

Apocephalus persecutor Borgmeier

(Fig. 20)

Apocephalus persecutor Borgmeier, 1961:44, figs.

54, 80.

HOLOTYPE. 9, BRAZIL: Goias: Campinas,
26.V.1933, J.S. Schwarzmaier, with Nomomyrmex
esenbecki (MZSP; examined).

SPECIES RECOGNITION. This species is easily
recognized by the triangular ovipositor. It is similar
to A. dichocercus Borgmeier (1958), a species I did
not examine. Borgmeier (1971) separated these two
species on the basis of the shorter costa in A. per-
secutor (0.38, versus 0.57 for A. dichocercus ). Fur-
ther study of A. dichocercus is needed.

DESCRIPTION. Body length 1. 0-1.1 mm. Frons
dark brown, anterior margin relatively straight.
Two pairs of supra-antennal setae present; lower
pair markedly smaller than upper pair. Flagello-
mere 1 light brown, round. Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron light brown. Anterior scutellar seta
small, fine, subequal to posterior setulae of scutum.
Legs yellowish-brown; apex of hind femur slightly
darker on anterior face. Mean costal length 0.39
wing length; range 0.38-0.40. Wing vein R2+3 pre-
sent. Halter brown. Abdominal tergites dark
brown. Venter of abdomen yellow to gray. Abdom-
inal tergites of normal form. Tergite 3 evenly col-
ored. Tergite 6 completely divided, with setae of

Brown: Revision of Apocephalus miricauda-group

Figures 19-27. Figures 19-20. Apocephalus reticulatus new species. 19. Dorsal. 20. Ventral. 21. Apocephalus persecutor
Borgmeier, dorsal. 22. Apocephalus curtinotus new species, dorsal. Figures 23-24. Apocephalus secus new species. 23.
Dorsal. 24. Ventral. Figures 25-26. Apocephalus spi cuius new species. 25. Dorsal. 26. Ventral. 27. Apocephalus torulus,
dorsal.

medium length along posterior margin. Venter of
segments 3-5 with row of setae on posterior mar-
gin. Abdomen without dense lateral setae. Ventral
setae of segment 6 long, consisting of several setae
in a straight row. Ovipositor (Fig. 20) straight in
lateral view, lightly sclerotized anteriorly; posteri-
orly with darkly sclerotized triangle; expanded.
Lateral darkening not differentiated. Dorsoapical
sclerite of ovipositor not differentiated. Apicodor-
sal margin of ovipositor straight. Apicoventral mar-
gin of ovipositor straight. Ovipositor with trian-
gular ventral postapical sclerite. Sternite 7 not dif-
ferentiated. Dufour’s mechanism not seen. Abdom-
inal glands in segment 5 white, inconspicuous in
cleared specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from Ecuador and Costa Rica.

WAY OF LIFE. The host of this species is un-
known, but it has been collected with various spe-
cies of army ants.

Contributions in Science, Number 482

MATERIAL EXAMINED. COSTA RICA: He-
redia: La Selva Biological Station, 10.43°N,
84.02°W, 1$, ix.1992, P. Hanson, Malaise trap
(LACM), 2$, l-15.iv.1993, ALAS, Malaise trap
M/04/67, M/05/68, (INBC), 1?, 15.v-l.vi.1993,
ALAS, Malaise trap M/8/114 (INBC). ECUADOR:
Napo: Limoncocha, 0.40°S, 76.58°W, 1$,

9.xi.l967, C. and M. Rettenmeyer, with Eciton
mexicanum, E-528, #3673 (UCMS), 1$,

13.xii.1967, with E. lucanoides, E-618, #4108
(UCMS); Pichincha: Rio Palenque Science Center,
3$, 1-3.V.1987, E. hurchellii bivouac, B. Brown
(LACM).

Apocephalus curtinotus new species

(Fig. 22)

SPECIES RECOGNITION. This species has a
peculiar, medial darkening of the dorsum, as well
as a rounded apical region of the ovipositor.

DESCRIPTION. Body length 1.3 mm. Frons

Brown: Revision of Apocephalus miricauda-group M 15

light brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 brown, round. Proboscis normal, small.
Palpus yellow. Dorsum of thorax light brown; pleu-
ron white. Anterior scutellar seta small, fine, sub-
equal to posterior setulae of scutum. Legs yellow-
ish-brown; apex of hind femur with abrupt dark-
ening on anterior face. Mean costal length 0.47
wing length. Wing vein R2+3 present. Halter brown.
Abdominal tergites dark brown, except tergite 6,
which is partly to completely yellowish. Venter of
abdomen yellow. Abdominal tergites of normal
form. Tergite 3 evenly colored. Tergite 6 complete,
with long seta at posterolateral corner. Venter of
segments 3-5 bare. Abdomen without dense lateral
setae. Ventral setae of segment 6 long, consisting of
several setae in a straight row. Ovipositor (Fig. 22)
slightly downturned apically, with broad median
sclerite. Lateral darkening broadened, rounded lat-
erally, margin dark, complete. Dorsoapical sclerite
of ovipositor not differentiated. Apicodorsal mar-
gin of ovipositor straight. Apicoventral margin of
ovipositor straight. Ovipositor with triangular ven-
tral postapical sclerite. Sternite 7 broad, lightly
sclerotized. Abdominal glands in segment 5 dark,
enlarged, elongate, clearly visible in cleared speci-
mens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in Amazonian Brazil.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin words curtus for short
and nota for mark, referring to the short lateral
darkenings.

HOLOTYPE. $, BRAZIL: Amazonas: Manaus,
Reserva Ducke, 3.13°S, 60.02°W, 6-17.vii.1992, J.
Vidal, Arm. Cola, l-B-20 m [LACM ENT 008385]
(INPA).

Apocephalus secus new species

(Figs. 23-24)

SPECIES RECOGNITION. This species is most
similar to A. conecitonis, from which it differs by
the shape of the ovipositor and the ventral setation,
as outlined in the key. It also resembles A. spiculus
new species, which has a much shorter ventral post-
apical sclerite.

DESCRIPTION. Body length 1.0 mm (both spec-
imens). Frons light brown, anterior margin rela-
tively straight. One pair of supra-antennal setae
present. Flagellomere 1 light brown, round. Pro-
boscis normal, small. Palpus light brown. Dorsum
of thorax light brown; pleuron white to brown. An-
terior scutellar seta small, fine, subequal to poste-
rior setulae of scutum. Legs yellowish-brown; apex
of hind femur slightly darker on anterior face.
Mean costal length 0.47 wing length; range 0.47-
0.48. Wing vein R2+3 present. Halter brown. Ab-
dominal tergites dark brown. Venter of abdomen
yellow to gray. Abdominal tergites of normal form.
Tergite 3 evenly colored. Tergite 6 anteriorly emar-

16 ■ Contributions in Science, Number 482

ginate, with long seta at posterolateral corner. Ven-
ter of segments 3-5 with a few, scattered setae. Ab-
domen without dense lateral setae. Ventral setae of
segment 6 long, consisting of a complete ventral
and lateral row; lateral setae shorter. Ovipositor
(Figs. 23-24) straight in lateral view, lightly but
evenly sclerotized dorsally. Lateral darkening short,
margin dark, complete. Dorsoapical sclerite of ovi-
positor not differentiated. Apicodorsal margin of
ovipositor emarginate on either side, with long me-
dial process. Apicoventral margin of ovipositor
straight. Ovipositor with triangular ventral postap-
ical sclerite. Sternite 7 not differentiated. Dufour’s
mechanism not seen. Abdominal glands in segment
5 white, invisible in cleared specimens. Internal
sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
only from a single, mid-elevation site in Costa Rica.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for different, referring to the
fact that although this species is closely similar to
A. conecitonis, A. reticulatus, and A. torulus new
species, it differs in a few key characters.

HOLOTYPE. $ , COSTA RICA: Puntarenas: Las
Alturas, 8.95°N, 82.38°W, i.1992, P. Hanson, Mal-
aise trap [LACM ENT 016169] (LACM).

PARATYPE. 1 9 , same data as holotype, except
v.1992 (LACM).

Apocephalus spiculus new species

(Figs. 25-26)

SPECIES RECOGNITION. This species is rec-
ognized by the distinctive short ventral postapical
sclerite of the ovipositor (Fig. 26).

DESCRIPTION. Body length 1.0 mm. Frons
dark brown, anterior margin relatively straight.
Two pairs of supra-antennal setae present; lower
pair markedly smaller than upper pair. Flagello-
mere 1 light brown, round. Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron white. Anterior scutellar seta small,
fine, subequal to posterior setulae of scutum. Legs
yellowish-brown; apex of hind femur with abrupt
darkening on anterior face. Mean costal length
0.47 wing length. Wing vein R2+3 present. Halter
light brown. Abdominal tergites dark brown. Ven-
ter of abdomen white. Abdominal tergites of nor-
mal form. Tergite 3 evenly colored. Tergite 6 an-
teriorly emarginate, with long seta at posterolateral
corner. Venter of segments 3-5 with row of setae
on posterior margin, but rows somewhat irregular.
Abdomen without dense lateral setae. Ventral setae
of segment 6 long, consisting of several setae in a
straight row. Ovipositor (Figs. 25-26) straight in
lateral view, lightly but evenly sclerotized dorsally.
Lateral darkening short, apically rounded, diver-
gent, margin dark, complete. Dorsoapical sclerite
of ovipositor not differentiated. Apicodorsal mar-
gin of ovipositor emarginate on either side, with
long medial process. Apicoventral margin of ovi-

Brown: Revision of Apocephalus miricauda-group

positor straight. Ovipositor with small, triangular
ventral postapical sclerite. Sternite 7 broad, without
anterior process. Dufour’s mechanism round. Ab-
dominal glands in segment 5 white, inconspicuous
in cleared specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in Panama.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a diminutive of the Latin word spica, for
spike, referring to the small apical triangle of the
ovipositor.

HOLOTYPE. $, PANAMA: San Bias: Nusagan-
di Reserve, 9.33°N, 79.0°W, 5-12.ii.1994, J. Pick-
ering, Malaise trap #2042 [LACM ENT 101299]
(LACM).

Apocephalus torulus new species

(Fig. 27)

SPECIES RECOGNITION. This species can be
separated from other small species similar to A. co-
necitonis by the rounded, lobe-shaped apices of the
lateral darkenings.

DESCRIPTION. Body length 0. 9-1.0 mm. Frons
dark brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 light brown, round. Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron yellow. Anterior scutellar seta
small, fine, subequal to posterior setulae of scutum.
Legs yellowish-brown; apex of hind femur with
abrupt darkening on anterior face. Mean costal
length 0.47 wing length; range 0.44-0.49. Wing
vein R2+3 present. Halter brown. Abdominal ter-
gites dark brown. Venter of abdomen yellow to
gray. Abdominal tergites of normal form. Tergite 3
evenly colored. Tergite 6 anteriorly emarginate,
with long seta at posterolateral corner. Venter of
segments 3-5 with row of setae on posterior mar-
gin. Abdomen without dense lateral setae. Ventral
setae of segment 6 long, consisting of a complete
ventral and lateral row; lateral setae shorter. Ovi-
positor (Fig. 27) straight in lateral view, lightly but
evenly sclerotized dorsally. Lateral darkening short,
apically rounded, divergent, margin dark, com-
plete. Dorsoapical sclerite of ovipositor not differ-
entiated. Apicodorsal margin of ovipositor emar-
ginate on either side, with long medial process. Ap-
icoventral margin of ovipositor straight. Ovipositor
with triangular ventral postapical sclerite. Sternite
7 broad, without anterior process. Dufour’s mech-
anism not seen. Abdominal glands in segment 5
white, inconspicuous in cleared specimens. Internal
sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from two sites in western Colombia and Ecuador.

WAY OF LIFE. Unknown. Most specimens were
collected with a raid of Labidus praedator, but we
did not observe any oviposition attempts directed
at the ants.

DERIVATION OF SPECIFIC EPITHET. The
Contributions in Science, Number 482

name is a Latin word for a small, rounded projec-
tion, referring to the lateral darkenings of the ovi-
positor.

HOLOTYPE. $ , ECUADOR: Esmeraldas: Bilsa
Biological Station, 0.34°N, 79.71°W, 8.V.1996, B.
Brown, P. Hibbs, J. Cantley, raid Labidus praedator
[LACM ENT 025275] (LACM).

PARATYPES. COLOMBIA: Valle: Rio Raposo,
1$, viii.1965, V. Lee, light trap (USNM). ECUA-
DOR: 6 $, same data as holotype (LACM, QCAZ).

Apocephalus succineus new species

SPECIES RECOGNITION. This is the only spe-
cies of Apocephalus in Dominican Republic amber
that has a broad ovipositor, similar in appearance
to that of A. paraponerae. There are other, unde-
scribed species of Apocephalus in this amber, but
all have much narrower ovipositors.

Unlike the extant species that it resembles, A.
succineus apparently lacks dorsal, paired, preapical
sclerites. It has a row of long setae on the posterior
margins of the venter of abdominal segments 5 and
6 as well as scattered, slightly smaller setae.

DESCRIPTION. All specimens of this species are
preserved in Dominican Republic amber. All show
some evidence of decay and shrivelling, probably
because they decomposed to some extent after be-
ing caught on the surface of the amber but before
being covered with a final coating of resin. Because
of this imperfect preservation, a number of char-
acter states cannot be seen clearly. When a partic-
ularly important character state can be seen best in
a single specimen other than the holotype, it is not-
ed in the following description.

Two pairs of supra-antennal setae present. Fla-
gellomere 1 appearing elongate, narrow (possibly
an artifact of shrivelling). Palpus normal, small.
Anterior scutellar seta enlarged, longer than one-
half posterior seta (easily seen in specimen DR-14-
105). Mean costal length 0.44 wing length; range
0.44-0.45. Wing vein R2+3 present. Abdominal ter-
gites of normal form. Tergite 6 undivided, tapered
posteriorly; without large posterior setae (best seen
in JW, March 1996-10). Venter of abdominal seg-
ments 5 and 6 with scattered, long setae; longest
setae in posterior row of each segment (seen in DR-
14-21 1). Abdomen without dense lateral setae.
Ovipositor broad, straight in lateral view, lightly
sclerotized, apparently without preapical sclerites
(similar to Fig. 1, but lacking paired, preapical
sclerites). Lateral darkening thin, subparallel, mar-
gin dark, complete. Apicodorsal margin of ovipos-
itor straight. Apicoventral margin of ovipositor not
seen but apparently with triangular, postapical
sclerite. The actual separation of the ovipositor and
the triangular postapical sclerite cannot be seen, so
it cannot be ruled out that the postapical sclerite is
actually fused to the ovipositor, which would be a
relatively primitive character state. Sternite 7 not
differentiated. Dufour’s mechanism, abdominal
glands, and internal sclerite not seen.

Brown: Revision of Apocephalus miricauda- group ■ 17

AGE OF FOSSILS. Dominican amber is assumed
to have been deposited in the early to middle Mio-
cene, 15-20 million years ago (Iturralde-Vinent and
MacPhee, 1996).

WAY OF LIFE. Unknown. Given that this species
is similar to A. paraponerae, a possible host would
be the Dominican amber fossil ant Paraponera die-
teri Baroni Urbani (1994). The genus Paraponera,
with its single extant species P. clavata (Fabricius),
no longer occurs on Hispaniola. Presumably, nei-
ther do species of the A. paraponerae- series, al-
though collecting has been so limited in this region
that such an assumption could not be made with
absolute certainty.

Phorids that parasitize injured hosts are expected
to be relatively common in amber (Brown, 1997c).
Hosts that were caught in amber probably emitted
alarm pheromones that attracted the parasitoids,
which in turn became trapped.

DERIVATION OF SPECIFIC EPITHET. The
name is from a Latin word, succinum, for amber.

HOLOTYPE. 9, DOMINICAN REPUBLIC
AMBER. Santiago/Puerto Plata area (LACM-IP).

PARATYPES. DOMINICAN REPUBLIC AM-
BER. 1?, north mines, DR-14-211, 4$, north
mines, DR-14-105 (AMNH). 2$, #3804 (USNM-
IP). 1$, DM, March 1993-5, 19, JW, March
1996-10 (Private collection of Mr. Pat Craig).

A. spatulatus- subgroup

DIAGNOSIS. This group can be diagnosed by
the first two synapomorphies listed below (under
Phylogenetic relationships). The ovipositors of
many of the included species do not resemble other
A. miricauda-group species, but some still retain
the distinctive lateral darkenings (e.g., Figs. 32, 33,
43, 44, 46).

PHYLOGENETIC RELATIONSHIPS. A hy-
pothesis of the relationships of some of the species
within this group is possible (Table 1, Fig. 121).
The following are proposed synapomorphic char-
acter states:

1. Ovipositor spatulate, dorsally concave (primi-
tive state: ovipositor flat or convex).

2. Ovipositor with sclerotization of dorsal, apical
region extending ventrally, completely encir-
cling the ovipositor (primitive state: dorsal and
ventral sclerotization not so confluent).

3. Ovipositor with medial, narrow, elongate,
bluntly ending, dorsoapical process (primitive
state: dorsal apex of ovipositor without pro-
cess).

4. Ventral apex of ovipositor drawn out into long,
extremely thin, filament-like process; substates:
(1) long, (2) short (primitive state: apex ter-
minating much more bluntly).

According to the cladogram, the short pro-
cess in A. acanthus new species and A. spatu-
latus Borgmeier is the result of secondary re-
duction. In A. striativentris new species and A.

18 ■ Contributions in Science, Number 482

Table 1. Character state matrix for A. spatulatus- group.

Outgroup

00000000000

incomptus

11000000000

striativentris

11000000000

brochus

11000000001

fuscipalpis

11000000001

digitalis

11100000000

denotatus

11100000000

pachycondylae

11010000000

atrimarginatus

11010000000

batillus

11010000000

emargilatus

11011000000

magnicauda

11011000000

triangularis

11010110000

quadratus

11010111000

acanthus

11021112110

spatulatus

11020112110

hrochus new species, there is a pointed ventral
apex but no elongate process.

5. Apicolateral region of v-shaped darkening
thickened (primitive state: apicolateral region
thin).

This state appears in A. emargilatus new
species, A. magnicauda new species and, ap-
parently convergently, in A. acanthus new spe-
cies.

6. Median carina present on dorsum of ovipositor
(primitive state: carina absent).

7. Venter of abdomen densely setose (primitive
state: abdominal setation sparser or lacking).

8. Median carina with (1) bluntly raised area, (2)
sharp spine (primitive state: median carina
without differentiated raised area).

The polarization of this character state — and
thus the sister group relationship between A.
quadratus Brown and A. acanthus + A. spa-
tulatus— is speculative. I assume that the high,
sharply pointed spine in A. acanthus and A.
spatulatus was preceded by a lower spine, such
as that found in A. quadratus.

9. Anterior portion of ovipositor darkly sclero-
tized (anterior portion of ovipositor lightly
sclerotized, except for lateral, v-shaped dark-
ening).

10. Ovipositor greatly expanded at midlength
(Figs. 44-46) (primitive state: ovipositor at
most slightly expanded at midlength).

11. Anterolateral corner of apical region of ovi-
positor with dorsal, tooth-like swelling (prim-
itive state: without tooth-like swelling).

The outgroup for this analysis was the A. para-
p oner ae-subgroup , plus other Apocephalus outside
of the A. attophilus- and A. miricauda- groups.
Analysis of these character states with HENNIG-
86 resulted in a single tree of length 14, consistency
index 92, retention index 95 (Fig. 121).

Brown: Revision of Apocephalus miricauda-group

Figures 28-37. Ovipositors. 28. Apocephalus incomptus new species, dorsal. 29. Apocephalus striativentris new species,
dorsal. 30. Apocephalus brochus new species, dorsal. 31. Apocephalus digitalis Borgmeier, left lateral. 32. Apocephalus
denotatus new species, dorsal. Figures 33-34. Apocephalus pachycondylae new species. 33. Dorsal. 34. Ventral. 35.
Apocephalus atrimarginatus new species, dorsal. Figures 36-37. Apocephalus batillus new species. 36. Dorsal. 37. Left
lateral.

Apocephalus incomptus new species

(Fig. 28)

SPECIES RECOGNITION. This relatively plain
species can be recognized by the lack of ventral se-
tae and the unmodified ovipositor.

DESCRIPTION. Body length 1.5 mm. Frons
dark brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 light brown, round. Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron yellow to white. Anterior scutellar
seta small, fine, subequal to posterior setulae of scu-
tum. Legs yellowish-brown; apex of hind femur
with abrupt darkening on anterior face. Mean cos-

Contributions in Science, Number 482

tal length 0.5 wing length; range 0.49-0.50. Wing
vein R2+3 present. Flatter brown. Abdominal ter-
gites dark brown, yellow medially and anteriorly,
tergite 6 yellow. Venter of abdomen yellow. Ab-
dominal tergites of normal form. Tergite 3 evenly
colored. Tergite 6 anteriorly emarginate, with seta
of medium length at posterolateral corner. Venter
of segments 3-5 bare. Abdomen without dense lat-
eral setae. Ventral setae of segment 6 long, consist-
ing of small, ventrolateral group and smaller lateral
setae. Ovipositor (Fig. 28) straight in lateral view,
lightly sclerotized, but with darker apical area. Lat-
eral darkening not differentiated; apical region
dark, rectangular. Apicodorsal margin of ovipositor
straight. Apicoventral margin of ovipositor drawn

Brown: Revision of Apocephalus miricauda- group ■ 19

out into pointed process. Ovipositor without ven-
tral postapical sclerite. Sternite 7 not differentiated
but anteriorly encircling segment. Dufour’s mech-
anism not seen. Abdominal glands in segment 5
white, inconspicuous in cleared specimens. Internal
sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from Amazonian Brazil and Ecuador.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for unadorned, referring to
the relatively simple ovipositor.

HOLOTYPE. 9, ECUADOR: Sucumbios: Sacha
Lodge, 0.5°N, 76.5°W, 25.vii-3.viii. 1994, P. Hibbs,
Malaise trap, 270 m [LACM ENT 040968]
(LACM).

PARATYPES. BRAZIL: Amazonas: Manaus, Re-
serva Ducke, 3.13°S, 60.02°W, 19, 6-17.vii.1992,
J. Vidal, Arm. Cola, 1-B 1 m (LACM). ECUADOR:
Sucumbios: Sacha Lodge, 0.5°N, 76.5° W, 19, 4-
14.V.1994, P. Hibbs, Malaise trap, 270 m (LACM).

Apocephalus striativentris new species

(Fig. 29)

SPECIES RECOGNITION. This species can be
immediately recognized by the darkly sclerotized
striations of intersegment 6-7. Other species have
these striations (for instance, see Brown, 1992, fig.
35B-D), but they are not black with sclerotization.
The dorsum of the ovipositor of this species also
has a small, apical ridge (Fig. 29).

DESCRIPTION. Body length 1.5-1. 8 mm. Frons
dark brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 light brown, round. Proboscis normal,
small. Palpus light brown. Dorsum of thorax light
brown; pleuron yellow to brown. Anterior scutellar
seta small, fine, subequal to posterior setulae of scu-
tum. Legs yellowish-brown; apex of hind femur of
even color anteriorly. Mean costal length 0.51 wing
length; range 0.49-0.52. Wing vein R2+3 present.
Halter light brown. Abdominal tergites dark
brown, except tergite 6, which is partly to com-
pletely yellowish. Venter of abdomen yellow. Ab-
dominal tergites of normal form. Tergite 3 evenly
colored. Tergite 6 complete, with short setae at pos-
terolateral corner. Venter of segments 3-5 bare. Ab-
domen without dense lateral setae. Ventral setae of
segment 6 long, consisting of several setae in a
straight row. Intersegment 6-7 ventrally with dark-
ly sclerotized, longitudinal striations. Ovipositor
(Fig. 29) slightly upturned apically, dorsally con-
cave, with a median ridge posteriorly; lightly but
evenly sclerotized dorsally. Lateral darkening short,
margin dark, complete. Dorsoapical sclerite of ovi-
positor not differentiated. Apicodorsal margin of
ovipositor a rounded point. Apicoventral margin of
ovipositor drawn out into pointed process. Ovipos-
itor without ventral postapical sclerite. Sternite 7
broad, without anterior process. Dufour’s mecha-
nism not seen. Abdominal glands in segment 5

20 ■ Contributions in Science, Number 482

white, inconspicuous in cleared specimens. Internal
sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from Brazil, Costa Rica, Ecuador, and Peru.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name refers to the darkened, ventral striations of
abdominal intersegment 6-7.

HOLOTYPE. 9 , ECUADOR: Sucumbios: Sacha
Lodge, 0.5°S, 76.5°W, 24.v-3.vi.1994, P. Hibbs,
Malaise trap, 270 m [LACM ENT 038105]
(LACM).

PARATYPES. BRAZIL: Amazonas: Reserve
Ducke, 3.13°S, 60.02°W, 49, 8-15.iv.1992, J. Vi-
dal, Arm-Cola 1-B-l m, 1 9 Arm. Oleo 1-A-l m
(INPA, LACM). COSTA RICA: Heredia, La Selva
Biological Station, 10.43°N, 84.02°W, 19,
2.iii.l993, ALAS, Malaise trap, M/05/20 (INBC);
Puntarenas: San Vito, Las Cruces, 8.78°N, 83.0°W,
19, iii.1988, P. Hanson, Malaise trap, 1300 m
(LACM). ECUADOR: Sucumbios: Sacha Lodge,
0.5°S, 76.5°W, 19, 23.iv-3.v.l994, 19, 4-
14.V.1994, 19, 14-24.V.1994, 49, 24.v-3.vi.1994,
1 9, 3-16.viii.1994, P. Hibbs, Malaise trap, 270 m
(LACM, QCAZ). PERU: Madre de Dios: Zona Re-
serva Manu, Pakitza, 11.95°S, 71.28°W, 29, 7-
9.iii.l992, R. Cambra, Malaise trap (LACM,
MUSM).

Apocephalus brochus new species

(Fig. 30)

SPECIES RECOGNITION. This species is dis-
tinctive in the broad, dark, pointed, medial process
of the dorsum of the ovipositor (Fig. 30). It is most
closely related to A. fuscipalpis Borgmeier, based on
the shared presence of anterior, tooth-like processes
on the ovipositor.

DESCRIPTION. Body length 1.4-1. 6 mm. Frons
yellow to light brown, anterior margin relatively
straight. Two pairs of supra-antennal setae present;
lower pair markedly smaller than upper pair. Fla-
gellomere 1 yellow, round. Proboscis normal, small.
Palpus brown. Dorsum of thorax yellow; pleuron
yellow. Anterior scutellar seta small, fine, subequal
to posterior setulae of scutum. Legs yellowish-
brown; apex of hind femur with abrupt darkening
on anterior face. Mean costal length 0.5 wing
length; range 0.47-0.51. Wing vein R2+3 present.
Halter brown. Abdominal tergites dark brown, ex-
cept tergite 6, which is partly to completely yellow-
ish. Venter of abdomen yellow. Abdominal tergites
of normal form. Tergite 3 evenly colored. Tergite 6
completely divided, with short setae at posterolat-
eral corner. Venter of segments 3-5 with a few, scat-
tered setae. Abdomen without dense lateral setae.
Ventral setae of segment 6 short, consisting of a
complete ventral and lateral row. Ovipositor (Fig.
30) slightly upturned apically, lightly but evenly
sclerotized dorsally, with anterolateral tooth-like
swelling. Lateral darkening convergent to apical
point, margin dark, complete. Dorsoapical sclerite

Brown: Revision of Apocephalus miricauda-gxoup

of ovipositor not differentiated. Apicodorsal mar-
gin of ovipositor pointed, broad, heavily sclero-
tized. Apicoventral margin of ovipositor drawn out
into point, but without long, filament-like process.
Ovipositor without ventral postapical sclerite. Ster-
nite 7 not differentiated but anteriorly encircling
segment. Dufour’s mechanism round. Abdominal
glands in segment 5 white, inconspicuous in cleared
specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from Brazil and Colombia.

WAY OF LIFE. Females of this species attacked
healthy Camponotus banghaasi Emery and C. ra-
pax (Fabricius) that were attracted to tuna fish
baits. Flies hovered over the ants and darted at the
posterior part of the heads of their hosts.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for projecting, referring to
the extended dorsal apex of the ovipositor.

HOLOTYPE. 9, COLOMBIA: Amazonas: 22
km NW Leticia, 4.04°S, 69.99° W, 6.ix.l997, B.
Brown, G. Kung, over baited Camponotus ban-
ghaasi (UNCB) [LACM ENT 102257].

PARATYPES. BRAZIL: Para: Oriximina, 1.8°S,
53.83°W, 19, 13.X.1992, J. Rafael, Malaise trap
(INPA). COLOMBIA: 19, same locality and col-
lectors as holotype, 28.viii.1997, over baited Cam-
ponotus rapax (LACM), 219, same data as holo-
type (LACM, MCZC, MZSP, UNCB, USNM).

Apocephalus fuscipalpis Borgmeier
(Fig. 118)

Apocephalus fuscipalpis Borgmeier, 1958:325, figs.

9-10, 13.

HOLOTYPE. 9, BRAZIL: Santa Catarina:
Nova Teutonia, iv.1950, F. Plaumann [LACM ENT
033880] (MZSP; examined).

SPECIES RECOGNITION. This species differs
from its closest relative, A. brochus, by the less
sclerotized and produced dorsal apex of the ovi-
positor. Female specimens of A. fuscipalpis are also
significantly larger in size than those of A. brochus.

DESCRIPTION. Body length 1.8-1. 9 mm. Frons
light brown, anterior margin relatively straight.
Two pairs of supra-antennal setae present; lower
pair slightly smaller than upper pair. Flagellomere
1 light brown, round. Proboscis normal, small. Pal-
pus brown. Dorsum of thorax light brown. Pleuron
light brown. Anterior scutellar seta small, fine, sub-
equal to posterior setulae of scutum. Legs yellow-
ish-brown. Apex of hind femur with abrupt dark-
ening on anterior face. Mean costal length 0.49
wing length (no variation). Wing vein R2+3 present.
Halter brown. Abdominal tergites dark brown, ex-
cept tergite 6, which is partly to completely yellow-
ish. Venter of abdomen dark gray. Abdominal ter-
gites of normal form. Tergite 3 evenly colored. Ter-
gite 6 completely divided, with long seta at pos-
terolateral corner. Venter of segments 3-5 with few,
medial setae. Abdomen without dense lateral setae.
Ventral setae of segment 6 long, consisting of a

Contributions in Science, Number 482

complete ventral and lateral row. Ovipositor
straight in lateral view, lightly but evenly sclerotized
dorsally, with anterolateral tooth-like swelling (Fig.
118). Lateral darkening not differentiated. Dor-
soapical sclerite of ovipositor not differentiated.
Apicodorsal margin of ovipositor pointed. Apico-
ventral margin of ovipositor drawn out into point-
ed process. Ovipositor without ventral postapical
sclerite. Ovipositor without separate sclerites lat-
erally. Sternite 7 not differentiated but anteriorly
encircling segment. Dufour’s mechanism not seen.
Abdominal glands in segment 5 white, inconspicu-
ous in cleared specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in southeastern Brazil.

WAY OF LIFE. Unknown.

OTHER MATERIAL EXAMINED. 19 para-
type, BRAZIL: Santa Catarina: Nova Teutonia,
ix.1958, F. Plaumann (MZSP).

Apocephalus digitalis Borgmeier

(Fig. 31)

Apocephalus digitalis Borgmeier, 1971:101, figs.
138-139.

HOLOTYPE. 9, BRAZIL: Santa Catarina:
Nova Teutonia, F. Plaumann [LACM ENT 093873]
(MZSP; examined).

SPECIES RECOGNITION. This species is ex-
tremely close to A. denotatus new species, but the
holotype of A. digitalis has a distinctive dorsal pro-
cess on the ovipositor that is lacking in A. deno-
tatus. Both have extremely lightly sclerotized ovi-
positors, except for the heavily sclerotized, promi-
nent, dorsal process.

DESCRIPTION. Body length 1.5 mm. Frons yel-
low, anterior margin relatively straight. Two pairs
of supra-antennal setae present; lower pair mark-
edly smaller than upper pair. Flagellomere 1 yellow,
round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax light brown; pleuron yellow. An-
terior scutellar seta small, fine, subequal to poste-
rior setulae of scutum. Legs yellowish-brown; apex
of hind femur with abrupt darkening on anterior
face. Mean costal length 0.54 wing length. Wing
vein R2+3 present. Halter brown. Abdominal ter-
gites yellow, posteriorly dark brown; tergite 6 com-
pletely yellow. Venter of abdomen yellow. Abdom-
inal tergites of normal form. Tergite 3 evenly col-
ored. Tergite 6 completely divided, with short setae
at posterolateral corner. Venter of segments 3-5
with a few, scattered setae. Abdomen without dense
lateral setae. Ventral setae of segment 6 long, con-
sisting of a complete ventral and lateral row. Ovi-
positor (Fig. 31) upturned apically, lightly but even-
ly sclerotized dorsally; with medial, setose process.
Lateral darkening short, margin dark, complete.
Dorsoapical sclerite of ovipositor not differentiat-
ed. Apicodorsal margin of ovipositor pointed,
greatly elongate. Apicoventral margin of ovipositor
drawn out into pointed process. Ovipositor without
ventral postapical sclerite. Sternite 7 not differen-

Brown: Revision of Apocephalus miricauda-group ■ 21

tiated. Dufour’s mechanism not seen. Abdominal
glands in segment 5 white, inconspicuous in cleared
specimens. Internal sclerite small, round.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in southeastern Brazil.

WAY OF LIFE. Unknown.

Apocephalus denotatus new species

(Fig. 32)

SPECIES RECOGNITION. This species differs
from the extremely similar A. digitalis by the lack
of a dorsomedial process on the ovipositor.

DESCRIPTION. Body length 1.3-1. 4 mm. Frons
light brown, anterior margin relatively straight.
Two pairs of supra-antennal setae present; lower
pair markedly smaller than upper pair. Flagello-
mere 1 light brown, round. Proboscis normal,
small. Palpus brown. Dorsum of thorax light
brown; pleuron white. Anterior scutellar seta small,
fine, subequal to posterior setulae of scutum. Legs
yellowish-brown; apex of hind femur with abrupt
darkening on anterior face. Mean costal length
0.59 wing length; range 0.59-0.60. Wing vein R2+3
present. Halter brown. Abdominal tergites dark
brown, yellowish anteriorly, tergite 6 yellow. Venter
of abdomen yellow. Abdominal tergites of normal
form. Tergite 3 evenly colored. Tergite 6 completely
divided, with a pair of large, posterior setae. Venter
of segments 3-5 bare. Abdomen without dense lat-
eral setae. Ventral setae of segment 6 long, consist-
ing of a complete ventral and lateral row. Ovipos-
itor (Fig. 32) slightly upturned apically, lightly but
evenly sclerotized dorsally. Lateral darkening not
differentiated. Apicodorsal margin of ovipositor
pointed, greatly elongate. Apicoventral margin of
ovipositor drawn out into pointed process; this pro-
cess thicker and more rounded than in related spe-
cies. Ovipositor without ventral postapical sclerite.
Sternite 7 not differentiated. Dufour’s mechanism
round. Abdominal glands in segment 5 white, in-
conspicuous in cleared specimens. Internal sclerite
rounded, with a short process.

GEOGRAPHICAL DISTRIBUTION. Known
from lowland Costa Rica.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for conspicuous, referring to
the large ventromedial projection of the ovipositor.

HOLOTYPE. 9, COSTA RICA: Puntarenas: 3
km SW Rincon, 8.68°N, 83.48°W, iii-v.1989, P.
Hanson, Malaise trap, 10 m [LACM ENT 050925]
(LACM).

PARATYPES. COSTA RICA: Heredia: La Selva
Biological Station, 10.43°N, 84.02°W, 2$, 1-
15.iv.1993, ALAS, Malaise trap M/04/67, M/15/78
(INBC); Puntarenas: 23 km NW Puerto Jimenez,
8.67°N, 83.45°W, 19, i-iv.1991, P. Hanson, Mal-
aise trap, 10 m (LACM), 3 km SW Rincon, 8.68°N,
83.48°W, 49, iii.1989, P. Hanson, Malaise trap, 10
m (LACM, MUCR).

22 ■ Contributions in Science, Number 482

Apocephalus pachycondylae new species

(Figs. 33-34)

SPECIES RECOGNITION. This species is some-
what nondescript. It differs from other species with
a long, filamentlike, ventral process by the lack of
ornamentation of the ovipositor and the distinct
lateral darkenings.

DESCRIPTION. Body length 1.1-1. 5 mm. Frons
light brown, anterior margin relatively straight.
Two pairs of supra-antennal setae present; lower
pair markedly smaller than upper pair. Flagello-
mere 1 yellow, round. Proboscis normal, small. Pal-
pus brown. Dorsum of thorax light brown; pleuron
light brown to white. Anterior scutellar seta small,
fine, subequal to posterior setulae of scutum. Legs
yellowish-brown; apex of hind femur with abrupt
darkening on anterior face. Mean costal length
0.55 wing length; range 0.54-0.56. Wing vein R2+3
present. Halter brown. Abdominal tergites dark
brown, except tergite 6, which is partly to com-
pletely yellowish. Venter of abdomen yellow. Ab-
dominal tergites of normal form. Tergite 3 evenly
colored. Tergite 6 completely divided, with a pair
of large, posterior setae. Venter of segments 3-5
with few, medial setae. Abdomen without dense lat-
eral setae. Ventral setae of segment 6 long, consist-
ing of a complete ventral and lateral row. Ovipos-
itor (Figs. 33-34) slightly downturned apically,
lightly but evenly sclerotized dorsally. Lateral dark-
ening short, ill-defined, margin dark, complete.
Dorsoapical sclerite of ovipositor not differentiat-
ed. Apicodorsal margin of ovipositor straight. Ap-
icoventral margin of ovipositor drawn out into
pointed process. Ovipositor without ventral post-
apical sclerite. Sternite 7 broad, lightly sclerotized.
Dufour’s mechanism round. Abdominal glands in
segment 5 white, inconspicuous in cleared speci-
mens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from one site each in Panama and Costa Rica.

WAY OF LIFE. Most specimens were collected
over a raid of army ants, Eciton lucanoides. The
army ants were raiding Pachycondyla obscuricornis
(Emery), among others; at one point I observed the
flies circling around army ants that were trying to
pry open a hole on a tree trunk. One specimen was
attracted to an injured worker of P. apicalis.

DERIVATION OF SPECIFIC EPITHET. The
name is based on that of the presumed hosts, spe-
cies of the ant genus Pachycondyla (F. Smith).

HOLOTYPE. 9, COSTA RICA: Heredia: La
Selva Biological Station, 10.43°N, 84.02°W,
3.vii.l993, B. Brown, over raid Eciton lucanoides
[LACM ENT 012797] (LACM).

PARATYPES. COSTA RICA: Heredia: La Selva
Biological Station, 10.43°N, 84.02°W, 19,
24.vi.1993, B. Brown, injured Pachycondyla api-
calis (LACM), Id, 79, same data as holotype
(male collected in copula with one of the females)
(INBC, LACM, MUCR). PANAMA: San Bias: Nu-
sagandi Reserve, 9.33°N, 79°W, 19, 16-

Brown: Revision of Apocephalus miricauda- group

23.iv.1994, J. Pickering, Malaise trap #2862
(LACM).

Apocepbalus atrimarginatus new species

(Fig. 35)

SPECIES RECOGNITION. This species is simi-
lar to A. batillus new species but has an ovipositor
with a much darker posterior margin.

DESCRIPTION. Body length 1.4-1. 8 mm. Frons
light brown, anterior margin relatively straight.
Two pairs of supra-antennal setae present; lower
pair markedly smaller than upper pair. Flagello-
mere 1 yellow, round. Proboscis normal, small. Pal-
pus brown. Dorsum of thorax light brown; pleuron
white. Anterior scutellar seta small, fine, subequal
to posterior setulae of scutum. Legs yellowish-
brown; apex of hind femur with abrupt darkening
on anterior face. Mean costal length 0.52 wing
length; range 0.50-0.55. Wing vein R2+3 present.
Halter brown. Abdominal tergites dark brown, ex-
cept tergite 6, which is partly to completely yellow-
ish. Venter of abdomen yellow. Abdominal tergites
of normal form. Tergite 3 evenly colored. Tergite 6
completely divided, with long seta at posterolateral
corner. Venter of segments 3-5 with few, medial
setae. Abdomen without dense lateral setae. Ventral
setae of segment 6 long, consisting of a complete
ventral and lateral row. Ovipositor (Fig. 35) slightly
upturned apically, lightly sclerotized, but with
darker apical area. Lateral darkening not differen-
tiated; apical region dark, especially posterior mar-
gin. Apicodorsal margin of ovipositor sinuous. Ap-
icoventral margin of ovipositor drawn out into ex-
tremely long, pointed process. Ovipositor without
ventral postapical sclerite. Sternite 7 not differen-
tiated but anteriorly encircling segment. Dufour’s
mechanism round. Abdominal glands in segment 5
white, inconspicuous in cleared specimens. Internal
sclerite small, oval.

GEOGRAPHICAL DISTRIBUTION. Known
from a few sites on the Osa Peninsula in Costa
Rica.

WAY OF LIFE. Females were attracted to injured
workers of Pachycondyla unidentata. One was at-
tracted to an injured Odontomachus laticeps Roger,
but it was collected at the same time as another
anomalous species for this host, A. comosus new
species. It is possible that there was some contam-
ination from fluids of other ants that I was crushing
on this day. One was also attracted to an injured
Odontomachus bauri Emery.

DERIVATION OF SPECIFIC EPITHET. The
name is based on Latin words for black and mar-
gin, referring to the darkened posterodorsal margin
of the ovipositor.

HOLOTYPE. 9, COSTA RICA: Puntarenas: 5
km SW Rincon, 8.7°N, 83.51°W, 6.vi.l998, B.
Brown, injured Pachycondyla unidentata [LACM
ENT 116694] (LACM).

PARATYPES. COSTA RICA: Puntarenas: Coo-
pemarti, 8.63°N, 83.47°W, 1 9, ii.1991, P. Hanson,

Contributions in Science, Number 482

Malaise trap (LACM), 3 km SW Rincon, 8.68°N,
83.48°W, 1$, vi— viii.1989, 19, x-xu.1990, 2 9,
x.1991, P. Hanson, Malaise trap (LACM, MUCR),
5 km SW Rincon, 8.7°N, 83.51°W, 19, 3.vi.l998,
B. Brown, injured Odontomachus laticeps
(LACM), 19, 4.vi.l998, O. bauri (LACM), 3 9,
5.vi.l998, 29, 6.vi.l998, B. Brown, injured Pachy-
condyla unidentata (INBC, LACM).

Apocepbalus batillus new species

(Figs. 36-37)

SPECIES RECOGNITION. This species most
closely resembles A. atrimarginatus but has a less
darkened posterior margin of the ovipositor. Fe-
males of A. incomptus (Fig. 28) are somewhat sim-
ilar but lack the long, filament-like process on the
ventral apex of the ovipositor.

DESCRIPTION. Body length 1.0-1. 8 mm. Frons
light brown, anterior margin relatively straight.
Two pairs of supra-antennal setae present; lower
pair markedly smaller than upper pair. Flagello-
mere 1 yellow, round. Proboscis normal, small. Pal-
pus brown. Dorsum of thorax light brown; pleuron
white. Anterior scutellar seta small, fine, subequal
to posterior setulae of scutum. Legs yellowish-
brown; apex of hind femur with abrupt darkening
on anterior face. Mean costal length 0.51 wing
length; range 0.49-0.53. Wing vein R2+3 present.
Halter brown. Abdominal tergites dark brown, yel-
low medially and anteriorly, tergite 6 yellow. Venter
of abdomen yellow. Abdominal tergites of normal
form. Tergite 3 evenly colored. Tergite 6 completely
divided, with long seta at posterolateral corner.
Venter of segments 3-5 with few, medial setae. Ab-
domen without dense lateral setae. Ventral setae of
segment 6 long, consisting of a complete ventral
and lateral row. Ovipositor (Figs. 36-37) slightly
upturned apically, lightly sclerotized, but with
darker apical area. Lateral darkening not differen-
tiated; apical region dark, with small, lateral inci-
sion. Apicodorsal margin of ovipositor pointed.
Apicoventral margin of ovipositor drawn out into
extremely long, thin, pointed process. Ovipositor
without ventral postapical sclerite. Sternite 7 not
differentiated but anteriorly encircling segment.
Dufour’s mechanism round. Abdominal glands in
segment 5 white, inconspicuous in cleared speci-
mens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
only from Barro Colorado Island, Panama.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for shovel, referring to the
shape of the ovipositor.

HOLOTYPE. 9, PANAMA: Canal Zone: Barro
Colorado Island, 9.15°N, 79.85°W, 10-17.iii.1993,
J. Pickering, Malaise trap #957 [LACM ENT
094017] (LACM).

PARATYPES. PANAMA: Canal Zone: Barro
Colorado Island, 9.15°N, 79.85°W, 19, 13-
205.1993, Malaise trap, 3 9, 20-275.1993, Mai-

Brown: Revision of Apocepbalus miricauda-gxoup □ 23

41.

42.

Figures 38-46. Ovipositors. Figures 38-40. Apocephalus emargilatus new species. 38. Dorsal. 39. Ventral. 40. Tip of
ovipositor, ventral. 41. Apocephalus magnicauda new species, dorsal. 42. Apocephalus triangularis new species, dorsal.
43. Apocephalus quadratus Brown, dorsal. Figures 44-45. Apocephalus acanthus new species. 44. Dorsal. 45. Left lateral.
46. Apocephalus spatulatus Borgmeier, dorsal.

aise trap #736, 29, 24-3 l.iii. 1993, Malaise trap,
#959, 1 9, 4-ll.viii.1993, Malaise trap #1670, 19,
6-13.iv.1994, Malaise trap #2419, 4 9, 20-
27.iv.1994, Malaise trap #2421, J. Pickering
(LACM, MCZC, MIUP, MZSP, USNM).

Apocephalus emargilatus new species

(Figs. 38-40)

SPECIES RECOGNITION. This species can be
recognized by the expanded proximal portion of
the ovipositor and lateral emargination of the api-
cal portion of the ovipositor.

DESCRIPTION. Body length 1.3-1. 8 mm. Frons

24 ■ Contributions in Science, Number 482

yellow, anterior margin relatively straight. Two
pairs of supra-antennal setae present; lower pair
slightly smaller than upper pair. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus brown.
Dorsum of thorax light brown; pleuron white to
light brown. Anterior scutellar seta small, fine, sub-
equal to posterior setulae of scutum. Legs yellow-
ish-brown; apex of hind femur of even color ante-
riorly. Mean costal length 0.5 wing length; range
0.49-0.51. Wing vein R2+3 present. Halter brown.
Abdominal tergites dark brown, yellow medially
and anteriorly, tergite 6 yellow. Venter of abdomen
yellow. Abdominal tergites of normal form. Tergite

Brown: Revision of Apocephalus miricauda-group

3 evenly colored. Tergite 6 anteriorly and posteri-
orly emarginate, with long seta at posterolateral
corner. Venter of segments 3-5 with few, medial
setae. Abdomen without dense lateral setae. Ventral
setae of segment 6 long, consisting of a complete
ventral and lateral row; lateral setae much shorter.
Ovipositor (Figs. 38-39) slightly upturned apically,
lightly sclerotized, but with darker apical area. Lat-
eral darkening not differentiated; apical region
dark, with transverse ridges and laterally emargin-
ate. Apicodorsal margin of ovipositor straight. Ap-
icoventral margin of ovipositor drawn out into
pointed process (Fig. 40). Ovipositor without ven-
tral postapical sclerite. Sternite 7 not differentiated
but anteriorly encircling segment. Dufour’s mech-
anism round. Abdominal glands in segment 5
white, inconspicuous in cleared specimens. Internal
sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from two sites on the Osa Peninsula in Costa Rica.

WAY OF LIFE. One specimen was attracted to
an injured worker of Camponontus sericeiventris
( Guerin-Mene ville ) .

DERIVATION OF SPECIFIC EPITHET. The

name is based on the lateral emargination of ovi-
positor.

HOLOTYPE. $ , COSTA RICA: Puntarenas: Sir-
ena, 8.48°N, 83.6°W, 1$, ll.vii.1993, D. Feener,
injured Camponotus sericeiventris [LACM ENT
024158] (LACM).

PARATYPES. COSTA RICA: Puntarenas, 3$,
Coopemarti, 8.63°N, 83.47°W, ii. 1991, P. Hanson,
Malaise trap (LACM).

Apocephalus magnicauda new species

(Fig. 41)

SPECIES RECOGNITION. The ovipositor of
this species is apically expanded, without lateral
emarginations; otherwise it is similar to A. emar-
gilatus.

DESCRIPTION. Body length 1. 7-2.1 mm. Frons
yellow, anterior margin relatively straight. Two
pairs of supra-antennal setae present; lower pair
markedly smaller than upper pair. Flagellomere 1
yellow, round. Proboscis normal, small. Palpus
brown. Dorsum of thorax light brown; pleuron yel-
low to white. Anterior scutellar seta small, fine,
subequal to posterior setulae of scutum. Legs yel-
lowish-brown; apex of hind femur with abrupt
darkening on anterior face. Mean costal length
0.52 wing length; range 0.51-0.53. Wing vein R2+3
present. Halter brown. Abdominal tergites dark
brown, yellow medially and anteriorly, tergite 6
yellow. Venter of abdomen yellow. Abdominal ter-
gites of normal form. Tergite 3 evenly colored. Ter-
gite 6 anteriorly and posteriorly emarginate, with
setae of medium length along posterior margin.
Venter of segments 3-5 with few, medial setae. Ab-
domen without dense lateral setae. Ventral setae of
segment 6 long, consisting of a complete ventral
and lateral row; lateral setae slender. Ovipositor

Contributions in Science, Number 482

(Fig. 41) slightly upturned apically, lightly sclero-
tized, but with darker apical area. Lateral darken-
ing not differentiated; apical region dark, posteri-
orly expanded. Apicodorsal margin of ovipositor
pointed. Apicoventral margin of ovipositor drawn
out into pointed process. Ovipositor without ven-
tral postapical sclerite. Sternite 7 not differentiated
but anteriorly encircling segment. Dufour’s mech-
anism round. Abdominal glands in segment 5
white, inconspicuous in cleared specimens. Internal
sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from Amazonian Ecuador and Peru.

WAY OF LIFE. I collected two females that were
flying above Camponotus sericeiventris attracted to
a tuna fish bait. The ants were agitated, running
around in a seeming panic response.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word magnus, for large, referring
to the apically enlarged ovipositor.

HOLOTYPE. 9 , ECUADOR: Sucumbios: Sacha
Lodge, 0.5°N, 76.5°W, 3-13.vi.1994, P. Hibbs,
Malaise trap, 270 m [LACM ENT 027779]
(LACM).

PARATYPES. ECUADOR: Sucumbios: Sacha
Lodge, 0.5°N, 76.5°W, 29, 22.ii^l.iii.l994, 19,
13-23.iv.1994, 19, 1 3-25. vii. 1994, P. Hibbs, Mal-
aise trap, 270 m (LACM, QCAZ). PERU: Madre
de Dios: Zona Reserva Manu, Pakitza, 11.95°S,
71.28°W, 2 9 , 25. ii. 1992, B. Brown, D. Feener, over
baited Camponotus sericeiventris (LACM,
MUSM).

Apocephalus triangularis new species

(Fig. 42)

Apocephalus fuscipalpis Borgmeier, 1958:325, figs.

9-10, 13 (in part).

SPECIES RECOGNITION. This species is easily
recognized by the triangular apex of the ovipositor
and its distinctive dark markings and by the prom-
inent, medial ridge.

Specimens of this species were among Borg-
meier’s paratypes of A. fuscipalpis.

DESCRIPTION. Body length 1.8 mm. Frons
light brown, anterior margin relatively straight.
Two pairs of supra-antennal setae present; lower
pair markedly smaller than upper pair. Flagello-
mere 1 yellow, round. Proboscis normal, small. Pal-
pus brown. Dorsum of thorax light brown; pleuron
light brown. Anterior scutellar seta small, fine, sub-
equal to posterior setulae of scutum. Legs yellow-
ish-brown; apex of hind femur with abrupt dark-
ening on anterior face. Mean costal length 0.46
wing length. Wing vein R2+3 present. Halter brown.
Abdominal tergites dark brown, yellow medially
and anteriorly, tergite 6 yellow. Venter of abdomen
yellow. Abdominal tergites of normal form. Tergite
3 evenly colored. Tergite 6 completely divided, with
three long setae on posterior margin of each scler-
ite. Venter of segments 3-5 with small, dense setae
medially. Abdomen without dense lateral setae.

Brown: Revision of Apocephalus miricauda-group ■ 25

Ventral setae of segment 6 long, consisting of a
complete ventral and lateral row. Ovipositor (Fig.
42) slightly upturned apically, lightly sclerotized,
but with darker apical area and prominent, medial
ridge. Lateral darkening not differentiated; much of
apical region dark, triangular with lateral, darker
patches. Apicodorsal margin of ovipositor pointed.
Apicoventral margin of ovipositor drawn out into
pointed process; process extremely long and heavily
sclerotized. Ovipositor without ventral postapical
sclerite. Sternite 7 not differentiated but anteriorly
encircling segment. Dufour’s mechanism not seen.
Abdominal glands in segment 5 white, inconspicu-
ous in cleared specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
only from Brazil and Mexico.

WAY OF LIFE. Unknown.

HOLOTYPE. 9, MEXICO: Chiapas: Montebel-
lo, vii.1969, W. Mason, Malaise trap (LACM).

PARATYPE. BRAZIL: Nova Teutonia, 1$,
xi.1958, F. Plaumann (MCZC).

Apocepbalus quadratus Brown

(Fig. 43)

Apocepbalus quadratus Brown, 1997b:47-48, fig.

67.

HOLOTYPE. 9 , ECUADOR: Sucumbios: Sacha
Lodge, 0.5°S, 76.5°W, 12-22.ii.1994, P. Hibbs,
Malaise trap, 270 m [LACM ENT 006850]
(LACM; examined).

SPECIES RECOGNITION. This species has an
extremely distinctive ovipositor, with a broad api-
cal region and with large, dark, heavily sculpted
lateral darkenings. There is a medial ridge with a
low prominence or spine at midlength.

I considered this a species of the A. attophilus-
group (Brown, 1997b) before my more precise def-
inition of the A. miricauda-group was formulated.

DESCRIPTION. See Brown, 1997b.

GEOGRAPHICAL DISTRIBUTION. Ecuador,
Brazil.

WAY OF LIFE. Unknown.

NEW MATERIAL EXAMINED. BRAZIL: Mi-
nas Gerais: Belo Horizonte, Estacao Ecologica,
UFMG campus, 19, 12-15.vii.1993, S. Gaimari,
Malaise trap, 800 m (BHMH).

Apocepbalus acanthus new species

(Figs. 44-45)

SPECIES RECOGNITION. This species, like A.
spatulatus, has an ovipositor with a dorsal spine
and distinctive lateral darkenings. The ovipositor of
A. acanthus differs in many small ways, however,
the most easily qualified being in the apex of the
ovipositor, which is obliquely truncate.

DESCRIPTION. Body length 1 .4 mm. Frons yel-
low, anterior margin relatively straight. Two pairs
of supra-antennal setae present; lower pair mark-
edly smaller than upper pair. Flagellomere 1 yellow,
round. Proboscis normal, small. Palpus yellow to

light brown. Dorsum of thorax light brown; pleu-
ron yellow. Anterior scutellar seta small, fine, sub-
equal to posterior setulae of scutum. Legs yellow-
ish-brown; apex of hind femur with abrupt dark-
ening on anterior face. Mean costal length 0.5 wing
length; range 0.49-0.50. Wing vein R2+3 present.
Halter mostly yellow, but with dark brown spot on
knob. Abdominal tergites dark brown, yellow me-
dially and anteriorly, tergite 6 yellow. Venter of ab-
domen yellow. Abdominal tergites of normal form.
Tergite 3 evenly colored. Tergite 6 completely di-
vided, with a pair of large, posterior setae. Venter
of segments 3-5 with small, dense setae. Abdomen
without dense lateral setae. Ventral setae of seg-
ment 6 long, consisting of a complete ventral and
lateral row, but lateral setae shorter. Ovipositor
(Figs. 44-45) slightly upturned apically, lightly
sclerotized, but with darker apical area; dorsally
with medial spine. Lateral darkening thin, subpar-
allel, margin dark, complete. Dorsoapical sclerite of
ovipositor not differentiated. Apicodorsal margin
of ovipositor straight, laterally with oblique trun-
cation. Apicoventral margin of ovipositor straight.
Ovipositor without ventral postapical sclerite. Ster-
nite 7 not differentiated but anteriorly encircling
segment. Dufour’s mechanism not seen. Abdominal
glands in segment 5 white, inconspicuous in cleared
specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from two lowland sites in Costa Rica.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is from a Greek word for spine, referring to
the mid-dorsal spine of the ovipositor.

HOLOTYPE. COSTA RICA: Puntarenas: 3 km
SW Rincon, 8.68°N, 83.48°W, viii.1991, P. Han-
son, Malaise trap, 10 m [LACM ENT 048759]
(LACM).

PARATYPE. COSTA RICA: Guanacaste: A1 Ta-
boga Forest Reserve, 9 km SW Canas, 19, 17-
27.ii.1987, W.L. Rubink, Malaise trap (EMUS).

Apocepbalus spatulatus Borgmeier
(Fig. 46)

Apocepbalus spatulatus Borgmeier, 1958:322, figs.

7, 12.

HOLOTYPE. 9, BRAZIL: Santa Catarina:
Nova Teutonia, F. Plaumann [LACM ENT 061158]
(MZSP; examined).

SPECIES RECOGNITION. This species, like the
preceding one, has a distinctive dorsal spine on the
ovipositor. Unlike A. acanthus, however, A. spatu-
latus has the posterior corners of the apex of the
ovipositor squared, not obliquely truncate (Fig. 46).

DESCRIPTION. The single specimen of this spe-
cies apparently was air-dried and has not retained
its natural color. Some references to color were
therefore made by referring to the original descrip-
tion.

Body length 1.8 mm. Frons yellow, anterior mar-
gin relatively straight. Two pairs of supra-antennal

26 ■ Contributions in Science, Number 482

Brown: Revision of Apocepbalus miricauda- group

setae present; lower pair markedly smaller than up-
per pair. Flagellomere 1 yellow, round. Proboscis
normal, small. Palpus yellow. Dorsum of thorax
yellow; pleuron yellow. Anterior scutellar seta
small, fine, slightly larger than posterior setulae of
scutum. Legs yellowish-brown; apex of hind femur
with abrupt darkening on anterior face. Mean cos-
tal length 0.45 wing length. Wing vein R2+3 present.
Halter brown. Abdominal tergites dark brown, yel-
low medially and anteriorly, tergite 6 yellow. Venter
of abdomen yellow. Abdominal tergites of normal
form. Tergite 3 evenly colored. Tergite 6 completely
divided, with several long setae on posterior mar-
gin. Venter of segments 3-5 with small, dense setae.
Abdomen without dense lateral setae. Ventral setae
of segment 6 long, consisting of a complete ventral
and lateral row. Ovipositor (Fig. 46) slightly up-
turned apically, with dark anterior and lighter pos-
terior sclerotization; dorsally with medial spine.
Lateral darkening thin, subparallel, margin dark,
complete. Dorsoapical sclerite of ovipositor not dif-
ferentiated. Apicodorsal margin of ovipositor
straight. Apicoventral margin of ovipositor drawn
out into pointed process. Ovipositor without ven-
tral postapical sclerite. Sternite 7 not differentiated
but anteriorly encircling segment. Dufour’s mech-
anism not seen. Abdominal glands in segment 5
white, inconspicuous in cleared specimens. Internal
sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in Brazil.

WAY OF LIFE. Unknown.

Apocephalus miricauda-subgroup

DIAGNOSIS. Ovipositor strongly bent dorsally
at midlength.

PHYLOGENETIC RELATIONSHIPS. Un-
known.

WAY OF LIFE. Two species are known to attack
injured ants of the genus Dinoponera Roger. The
other two species, A. flexus and A. orbiculus, occur
outside the published distribution of Dinoponera
species (Kempf, 1971) and must use different hosts.

Apocephalus miricauda Borgmeier

(Figs. 47-49)

Apocephalus miricauda Borgmeier 1971:111; figs.
150-152.

HOLOTYPE. 9, BRAZIL: Para: Utinga,
xii.1966, S.J. de Oliveira (MZSP; examined).

SPECIES RECOGNITION. This species can be
recognized by the curved ovipositor in lateral view
(Fig. 49) and the extremely expanded sternite 7,
which is visible projecting laterally from above (Fig.
47).

DESCRIPTION. Body length 2.4-2. 9 mm. Frons
yellow, anterior margin relatively straight. Two
pairs of supra-antennal setae present; lower pair
markedly smaller than upper pair. Flagellomere 1

yellow, round. Proboscis normal, small. Palpus yei-

Contributions in Science, Number 482

low. Dorsum of thorax yellow; pleuron yellow. An-
terior scutellar seta small, fine, subequal to poste-
rior setulae of scutum. Legs yellow; apex of hind
femur of even color anteriorly. Mean costal length
0.54 wing length; range 0.53-0.55. Wing vein R2+3
present. Halter brown. Abdominal tergites dark
brown, lighter medially. Venter of abdomen yellow.
Abdominal tergites of normal form. Tergite 3 even-
ly colored. Tergite 6 anteriorly and posteriorly
emarginate, with short setae at posterolateral cor-
ner. Venter of segments 3-5 with a few, scattered
setae. Abdomen without dense lateral setae. Ventral
setae of segment 6 short, consisting of single me-
dian pair. Ovipositor (Figs. 47-49) deflected dor-
sally posterior to midpoint, with triangular median
sclerite. Lateral darkening posteriorly enlarged,
truncate, margin dark, complete. Dorsoapical scler-
ite of ovipositor triangular. Apicodorsal margin of
ovipositor straight. Apicoventral margin of ovipos-
itor straight. Ovipositor without ventral postapical
sclerite. Sternite 7 triangular, apically flared with
small lateral projections. Dufour’s mechanism
round. Abdominal glands in segment 5 white, in-
conspicuous in cleared specimens. Internal sclerite
a simple, round loop.

GEOGRAPHICAL DISTRIBUTION. Known
only from Brazil. The host ant is known from Brazil
and Peru (Kempf, 1971).

WAY OF LIFE. The life history of this species
was described by Silveira-Costa and Moutinho
(1996), who found that it was attracted to injured
workers of Dinoponera gigantea (Perty).

PHYLOGENETIC RELATIONSHIPS. This spe-
cies and A. kungae new species are hypothesized to
be sister-species, based on the enlarged, broadened
sternite 7 with particularly distinct fringes of lateral
processes.

MATERIAL EXAMINED. BRAZIL: Para: Par-
agominas, 2.92°S, 47.58°W, 1$, v.1994, 3 9,
lO.v.1995, A. Silveira-Costa, injured Dinoponera
gigantea (LACM).

Apocephalus kungae new species

(Figs. 50, 108)

SPECIES RECOGNITION. This species differs
from the others in this group by the extremely
broad area of medium sclerotization between the
lateral darkenings of the ovipositor (Fig. 50). It is
similar in many respects to A. dinoponerae new
species, with which it was collected, but A. dino-
ponerae does not have a dorsally flexed ovipositor
and its sternite 7 is of a different form.

DESCRIPTION. Body length 1.9 mm. Frons yel-
low, anterior margin relatively straight. One pair of
supra-antennal setae present. Flagellomere 1 light
brown, round. Proboscis normal, small. Palpus yel-
low. Dorsum of thorax light brown; pleuron white
to light brown. Anterior scutellar seta large, bristle-
like. Legs yellowish-brown; apex of hind femur
with abrupt, but small, darkening on anterior face.
Mean costal length 0.56 wing length. Wing vein

Brown: Revision of Apocephalus miricauda-group M 27

50.

Figures 47-53. Ovipositors. Figures 47-49. Apocepbalus miricauda Borgmeier. 47. Dorsal. 48. Ventral. 49. Left lateral.
50. Apocepbalus kungae new species, dorsal. Figures 51-52. Apocepbalus flexus new species. 51. Dorsal (showing inter-
nal, sclerotized loop). 52. Ventral. 53. Apocepbalus orbiculus new species, dorsal (showing internal, sclerotized loop).

R2+3 present. Flalter brown. Abdominal tergites yel-
lowish-brown. Venter of abdomen yellow. Abdom-
inal tergites of normal form. Tergite 6 anteriorly
emarginate, with short setae at posterolateral cor-
ner. Venter of segments 3-5 with a few, scattered
setae. Abdomen without dense lateral setae. Ventral
setae of segment 6 long, consisting of a complete
ventral and lateral row. Ovipositor (Fig. 50) deflect-
ed dorsally at midpoint, lightly but evenly sclero-
tized dorsally. Lateral darkening apically conver-
gent and lighter in color, margin dark, complete.
Dorsoapical sclerite of ovipositor not differentiat-
ed. Apicodorsal margin of ovipositor straight. Ap-
icoventral margin of ovipositor rounded. Oviposi-
tor without ventral postapical sclerite. Sternite 7
triangular, apically flared with small lateral projec-

28 ■ Contributions in Science, Number 482

tions. Abdominal glands in segment 5 white, incon-
spicuous in cleared specimens. Internal sclerite a
simple, round loop (Fig. 108).

GEOGRAPHICAL DISTRIBUTION. Known
only from southeastern Colombia. The host ant is
known from Brazil and Peru (Kempf, 1971).

WAY OF LIFE. An injured worker of Dinopo-
nera longipes Emery attracted this fly.

PHYLOGENETIC RELATIONSHIPS. See A.
miricauda.

DERIVATION OF SPECIFIC EPITHET. This
species is named after Ms. Giar-Ann Kung, who
helped me study its way of life and collect the ho-
lotype.

HOLOTYPE. 9, COLOMBIA: Amazonas:
Amacayacu National Park, 3.82°S, 70.26°W,

Brown: Revision of Apocepbalus miricauda- group

5.ix.l997, B. Brown, G. Kung, injured Dinoponera
longipes [LACM ENT 102143] (UNCB).

Apocephalus flexus new species

(Figs. 51-52, 107)

SPECIES RECOGNITION. This species is ex-
tremely similar to A. orbiculus but differs in having
a much larger sclerotized loop (compare Figs. 51
and 53).

DESCRIPTION. Body length 1.5 mm. Frons yel-
low, anterior margin relatively straight. One pair of
supra-antennal setae present. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax light brown; pleuron white. An-
terior scutellar seta small, fine, subequal to poste-
rior setulae of scutum. Legs yellowish-brown; apex
of hind femur with abrupt darkening on anterior
face. Mean costal length 0.47 wing length. Wing
vein R2+3 present. Halter brown. Abdominal ter-
gites dark brown. Venter of abdomen yellow. Ab-
dominal tergites of normal form. Tergite 3 evenly
colored. Tergite 6 anteriorly emarginate, with a
pair of large, posterior setae. Venter of segments 3-
5 with few, medial setae. Abdomen without dense
lateral setae. Ventral setae of segment 6 long, con-
sisting of several setae in a straight row. Ovipositor
(Figs. 51-52) deflected dorsally posterior to mid-
point, lightly but evenly sclerotized dorsally. Lateral
darkening posteriorly enlarged, truncate, margin
dark, complete. Dorsoapical sclerite of ovipositor
rounded anteriorly. Apicodorsal margin of ovipos-
itor straight. Apicoventral margin of ovipositor
straight. Ovipositor without ventral postapical
sclerite. Sternite 7 narrow but apically expanded.
Abdominal glands in segment 5 white, inconspicu-
ous in cleared specimens. Internal sclerite rounded,
with long process (Fig. 107).

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in western Ecuador.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for bend, referring to the ovi-
positor, which is curved dorsally in lateral view.

HOLOTYPE. 9, ECUADOR: Pichincha: 17 km
E Santo Domingo, Tinalandia, 6-13.V.1987, B.
Brown, windows, 710 m [LACM ENT 012773]
(LACM).

Apocephalus orbiculus new species

(Fig. 53)

SPECIES RECOGNITION. This species is ex-
tremely similar to A. flexus but has a slightly dif-
ferent preapical sclerite and a much smaller scler-
otized loop (compare Figs. 51 and 53).

DESCRIPTION. Body length 1.3 mm. Frons
light brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 light brown, oval. Proboscis normal,
small. Palpus light brown. Dorsum of thorax light
brown; pleuron yellowish brown. Anterior scutellar
seta large, bristle-like. Legs yellowish-brown; apex

Contributions in Science, Number 482

of hind femur with abrupt darkening on anterior
face. Mean costal length 0.44 wing length. Wing
vein R2+3 present. Halter brown. Abdominal ter-
gites dark brown. Venter of abdomen gray. Abdom-
inal tergites of normal form. Tergite 3 evenly col-
ored. Tergite 6 anteriorly emarginate, divided, with
short setae at posterolateral corner. Venter of seg-
ments 3-5 with a few, scattered setae. Abdomen
without dense lateral setae. Ventral setae of seg-
ment 6 long, consisting of a complete ventral and
lateral row. Ovipositor (Fig. 53) deflected dorsally
at midpoint, lightly but evenly sclerotized dorsally.
Lateral darkening posteriorly enlarged, truncate,
margin dark, complete. Dorsoapical sclerite of ovi-
positor rectangular. Apicodorsal margin of ovipos-
itor straight. Apicoventral margin of ovipositor
straight. Ovipositor without ventral postapical
sclerite. Sternite 7 narrow but apically expanded.
Dufour’s mechanism not seen. Abdominal glands in
segment 5 white, inconspicuous in cleared speci-
mens. Internal sclerite small, round.

GEOGRAPHICAL DISTRIBUTION. Known
from a single lowland site in Costa Rica.

WAY OF LIFE. Unknown. Almost all of the Mal-
aise trap captures were from a single site (trap 12).

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for small circle, referring to
the sclerotized loop, which is much smaller than
that of the similar A. flexus.

HOLOTYPE. 9, COSTA RICA: Heredia: La
Selva Biological Station, 10.43°N, 84.02°W,
3.V.1993, ALAS, Malaise trap M/12/91 [INBI-
OCRI002273758] (INBC).

PARATYPES. COSTA RICA: Heredia: La Selva
Biological Station, 10.43°N, 84.02°W, 19,
16.iii.1993, ALAS, Malaise trap M/12/43, 19,
2.iv.l993, ALAS, Malaise trap M/12/59, 19,
3.V.1993, ALAS, Malaise trap M/12/91, 3 9,
19.V.1993, ALAS, Malaise trap M/12/00 6, 19,
4.iv.l994, ALAS, Malaise trap M/09/387, 19,
17.xi.1998, ALAS, light L/08/513 (INBC, LACM).

A. meniscus- subgroup

DIAGNOSIS. Dorsomedial preapical sclerite of
ovipositor attached to lateral darkenings by dis-
tinctive bar of sclerotization; in most species, lateral
postapical sclerites present.

PHYLOGENETIC RELATIONSHIPS. The rela-
tionships within this group are not resolved.

Although it is divergent in structure from all oth-
er A. miricauda- group species, A. lopesi (Borg-
meier) clearly belongs in this group. It has the post-
apical sclerites present but has apparently lost the
dorsal preapical sclerite.

WAY OF LIFE. The two species with known
hosts are parasitoids of injured ants of the genus
Odontomachus Latreille.

Apocephalus meniscus new species

(Fig. 54)

SPECIES RECOGNITION. This species can be
recognized by the long, dense ventral setae and the

Brown: Revision of Apocephalus tniricauda-group ■ 29

Figures 54-63. Ovipositors. 54. Apocephalus meniscus new species, dorsal. Figures 55-56. Apocephalus barbiventris
new species. 55. Dorsal. 56. Ventral. 57. Apocephalus amplidiscus new species, dorsal. Figures 58-59. Apocephalus
paldiae new species. 58. Dorsal. 59. Ventral. Figures 60-61. Apocephalus cyclodiscus new species. 60. Dorsal. 61. Ventral.
Figures 62-63. Apocephalus lopesi (Borgmeier). 62. Dorsal. 63. Ventral.

preapical sclerite that does not markedly extend an-
terior to the lateral bars of sclerotization.

DESCRIPTION. Body length 1.1-1. 3 mm. Frons
yellow to light brown, anterior margin relatively
straight. One pair of supra-antennal setae present.
Flagellomere 1 light brown, round. Proboscis nor-
mal, small. Palpus yellow. Dorsum of thorax light
brown; pleuron white. Anterior scutellar seta small,
fine, subequal to posterior setulae of scutum. Legs
yellowish-brown; apex of hind femur with abrupt
darkening on anterior face. Mean costal length
0.49 wing length; range no range. Wing vein R2+3
present. Halter brown. Abdominal tergites dark
brown. Venter of abdomen yellow. Abdominal ter-
gites of normal form. Tergite 3 evenly colored. Ter-
gite 6 anteriorly emarginate, with short setae at

30 ■ Contributions in Science, Number 482

posterolateral corner. Venter of segments 3-5 with
long, dense setae concentrated medially. Abdomen
without dense lateral setae. Ventral setae of seg-
ment 6 long, consisting of a complete ventral and
lateral row. Ovipositor (Fig. 54) straight in lateral
view, lightly but evenly sclerotized dorsally. Lateral
darkening thin, subparallel, margin dark, complete.
Dorsoapical sclerite of ovipositor trapezoidal. Ap-
icodorsal margin of ovipositor straight. Apicoven-
tral margin of ovipositor drawn out into pointed
process. Ovipositor without ventral postapical
sclerite. Sternite 7 narrow but apically expanded.
Abdominal glands in segment 5 white, inconspicu-
ous in cleared specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
only from Pakitza, Peru.

Brown: Revision of Apocephalus miricauda- group

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The

name is based on the Greek word meniskos for
crescent, referring to the shape formed by the
preapical sclerite and the lateral bars of the ovi-
positor.

HOLOTYPE. 9, PERU: Madre de Dios, Zona
Reserva Manu, Pakitza, 11.95°S, 71.28°W, 13-
18. ii.1992, B. Brown, D. Feener, Malaise trap #1
[LACM ENT 012211] (MUSM).

PARATYPE. 1 9 , same data as holotype
(LACM).

Apocephalus barbiventris new species

(Figs. 55-56)

SPECIES RECOGNITION. This species is also
densely setose ventrally, similar to A. meniscus , but
has a preapical sclerite that is shaped differently,
projecting farther anteriorly on the ovipositor. It
also has lateral postapical sclerites, which are lack-
ing in A. meniscus.

DESCRIPTION. Body length 1.2-1. 4 mm. Frons
yellow, anterior margin relatively straight. One pair
of supra-antennal setae present. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax yellow; pleuron white. Anterior
scutellar seta small, fine, subequal to posterior se-
tulae of scutum (slightly larger). Legs yellowish-
brown; apex of hind femur slightly darker on an-
terior face. Mean costal length 0.48 wing length;
range 0.47-0.5. Wing vein R2+3 present. Halter
brown. Abdominal tergites dark brown. Venter of
abdomen yellow. Abdominal tergites of normal
form. Tergite 3 evenly colored. Tergite 6 anteriorly
emarginate, with short setae at posterolateral cor-
ner. Venter of segments 3-5 with long, dense setae.
Abdomen without dense lateral setae. Ventral setae
of segment 6 long, consisting of a complete ventral
and lateral row. Ovipositor (Figs. 55-56) straight
in lateral view, lightly but evenly sclerotized dor-
sally. Lateral darkening posteriorly enlarged, trun-
cate, margin dark, complete. Dorsoapical sclerite of
ovipositor rectangular, anteriorly emarginate; with
lateral bars extended to anterior apex of lateral
darkenings. Apicodorsal margin of ovipositor
straight. Apicoventral margin of ovipositor drawn
out into pointed process. Ovipositor without ven-
tral postapical sclerite. Sternite 7 thin, with lighter,
expanded area at midlength. Dufour’s mechanism
elongate. Abdominal glands in segment 5 white, in-
conspicuous in cleared specimens. Internal sclerite
with small opening and broad, moderately sclero-
tized process.

GEOGRAPHICAL DISTRIBUTION. Known
from a few sites on the Osa Peninsula in Costa
Rica.

WAY OF LIFE. One female was attracted to in-
jured Odontomacbus bauri.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin words barba and ven-

Contributions in Science, Number 482

tris, for beard and belly, referring to the setose ven-
ter of the abdomen.

HOLOTYPE. 9, COSTA RICA: Puntarenas: 24
km W Piedras Blancas, 8.77°N, 83.4°W, ii.1992, P.
Hanson, Malaise trap, 200 m [LACM ENT
012698] (LACM).

PARATYPES. COSTA RICA: Puntarenas: 10 km
W Piedras Blancas, 8.75°N, 83.3°W, 29, iii-
v.1989, P. Hanson, Malaise trap (INBC, LACM),
24 km W Piedras Blancas, 8.77°N, 83.4°W, 19,

ii. 1992, P. Hanson, Malaise trap, 200 m (LACM),
3 km SW Rincon, 8.68°N, 83.48°W, 19, ii-

iii. 1989, 19, iii.1989, 19, iii— iv. 1991, P. Hanson,
Malaise trap, 10 m (LACM, MUCR), 5 km SW
Rincon, 8.7°N, 83.51°W, 1 9, 4.vi.l998, B. Brown,
injured Odontomacbus bauri (LACM).

Apocephalus amplidiscus new species

(Fig. 57)

SPECIES RECOGNITION. This species has a
distinctive large preapical sclerite that fills most of
the space between the lateral darkenings.

DESCRIPTION. Body length 1.4 mm. Frons
light brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 yellow, round. Proboscis normal, small.
Palpus yellow. Dorsum of thorax light brown; pleu-
ron white. Anterior scutellar seta small, fine, sub-
equal to posterior setulae of scutum. Legs yellow-
ish-brown; apex of hind femur with abrupt dark-
ening on anterior face. Mean costal length 0.48
wing length. Wing vein R2+3 present. Halter brown.
Abdominal tergites dark brown. Venter of abdo-
men yellow. Abdominal tergites of normal form.
Tergite 3 evenly colored. Tergite 6 completely di-
vided, with short setae at posterolateral corner.
Venter of segments 3-5 with long, dense setae con-
centrated medially. Abdomen without dense lateral
setae. Ventral setae of segment 6 long, consisting of
a complete ventral and lateral row. Ovipositor (Fig.
57) straight in lateral view, lightly but evenly scler-
otized dorsally. Lateral darkening thin, subparallel,
margin dark, with separate apical sclerite. Dor-
soapical sclerite of ovipositor large, round, dark,
shiny. Apicodorsal margin of ovipositor straight.
Apicoventral margin of ovipositor straight. Ovi-
positor without ventral postapical sclerite. Sternite
7 narrow but apically expanded. Dufour’s mecha-
nism elongate. Abdominal glands in segment 5
white, inconspicuous in cleared specimens. Internal
sclerite rounded, small, with a short process.

GEOGRAPHICAL DISTRIBUTION. Known
from a single lowland site in Costa Rica.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin words amplus and dis-
cus, for large and plate, referring to the large preap-
ical sclerite of this species.

HOLOTYPE. 9, COSTA RICA: Limon: 7 km
SW Bribri, 9.58°N, 82.88°W, xii.1989-ii.1990, P.

Brown: Revision of Apocephalus miricauda- group ■ 3 1

Hanson, Malaise trap, 50 m [LACM ENT 047431]
(LACM).

Apocephalus paldiae new species

(Figs. 58-59)

SPECIES RECOGNITION. This species is simi-
lar to A. barbiventris, but the ventral setae of the
abdomen are much shorter and fewer.

DESCRIPTION. Body length 1.3-1. 9 mm. Frons
yellow to light brown, anterior margin relatively
straight. One pair of supra-antennal setae present.
Flagellomere 1 basally yellow, apically brown,
round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax yellow; pleuron white. Anterior
scutellar seta small, fine, subequal to posterior se-
tulae of scutum. Legs yellowish-brown; apex of
hind femur with abrupt darkening on anterior face.
Mean costal length 0.47 wing length; range 0.45-
0.51. Wing vein R2+3 present. Halter brown. Ab-
dominal tergites dark brown. Venter of abdomen
yellow. Abdominal tergites of normal form. Tergite
3 evenly colored. Tergite 6 anteriorly emarginate,
with short setae at posterolateral corner. Venter of
segments 3-5 with short setae, concentrated medi-
ally. Abdomen without dense lateral setae. Ventral
setae of segment 6 long, consisting of a complete
ventral and lateral row. Ovipositor (Figs. 58-59)
straight in lateral view, lightly but evenly sclerotized
dorsally. Lateral darkening posteriorly enlarged,
truncate, margin dark, with separate apical sclerite.
Dorsoapical sclerite of ovipositor heart-shaped;
with lateral bars extended to anterior apex of lat-
eral darkenings. Apicodorsal margin of ovipositor
straight. Apicoventral margin of ovipositor
straight. Ovipositor without ventral postapical
sclerite. Sternite 7 triangular. Dufour’s mechanism
elongate. Abdominal glands in segment 5 white, in-
conspicuous in cleared specimens. Internal sclerite
not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in northern Costa Rica.

WAY OF LIFE. These flies are attracted to, and
oviposit in, injured workers of Odontomachus
chelifer (Latreille). Pairs of flies arrive in copula,
but the male immediately departs when they land
near an injured ant. Both egg-layers and feeders
were observed in this species. A single egg (n = 6),
or in one instance two eggs, are laid in the abdo-
men, and newly hatched larvae apparently migrate
internally through the ant’s body to the head, where
feeding takes place. Larval feeding ended after x =
4.75 days (n = 4: 4, 3, 6, 6 days).

DERIVATION OF SPECIFIC EPITHET. This
species is named after Ms. Jill Paldi, who helped
me study its way of life and collect the type series.

HOLOTYPE. 9 , COSTA RICA: Guanacaste: Es-
tacion Pitilla, 11.0°N, 85.43°W, 5.vii.l997, J. Paldi,
injured Odontomachus chelifer [LACM ENT
101744] (LACM).

PARATYPES. COSTA RICA: Guanacaste: Esta-
cion Biologia Pitilla, 11.0°N, 85.43°W, 29,

3.vii.l997, 1 <5 , 13 9 , 4.vii.l997, 9 $ , 5.vii.l997, B.
Brown, J. Paldi, injured Odontomachus chelifer
(INBC, LACM, MCZC, MUCR, USNM).

Apocephalus cyclodiscus new species

(Figs. 60-61)

SPECIES RECOGNITION. This species has a
distinctive ventral sclerite on segment 7, being ex-
panded and rounded anteriorly, then narrowed pos-
teriorly (Fig. 61).

DESCRIPTION. Body length 1.2-1 .4 mm. Frons
light brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 brown, round. Proboscis normal, small.
Palpus yellow. Dorsum of thorax light brown; pleu-
ron yellow to white. Anterior scutellar seta small,
fine, subequal to posterior setulae of scutum. Legs
yellowish-brown; apex of hind femur with abrupt
darkening on anterior face. Mean costal length
0.46 wing length; range 0.45-0.47. Wing vein R2+3
present. Halter brown. Abdominal tergites dark
brown. Venter of abdomen yellow. Abdominal ter-
gites of normal form. Tergite 3 evenly colored. Ter-
gite 6 anteriorly emarginate, with long seta at pos-
terolateral corner. Venter of segments 3-5 with a
few, scattered setae. Abdomen without dense lateral
setae. Ventral setae of segment 6 long, consisting of
several setae in a straight row. Ovipositor (Figs. 60-
61) straight in lateral view, lightly but evenly scler-
otized dorsally. Lateral darkening thin, subparallel,
apically divergent, margin dark, with separate api-
cal sclerite. Dorsoapical sclerite of ovipositor rect-
angular, anteriorly emarginate; with lateral bars ex-
tended to anterior apex of lateral darkenings. Api-
codorsal margin of ovipositor straight. Apicoven-
tral margin of ovipositor straight. Ovipositor
without ventral postapical sclerite. Sternite 7 ante-
riorly rounded, dark; posteriorly narrow. Dufour’s
mechanism elongate. Abdominal glands in segment
5 white, inconspicuous in cleared specimens. Inter-
nal sclerite with small opening and broad, moder-
ately sclerotized process.

GEOGRAPHICAL DISTRIBUTION. Known
from two sites in Panama.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is based on two Latin words, cyclos for round
and discus for plate, referring to the shape of ster-
nite 7.

HOLOTYPE. $, PANAMA: San Bias: Nusagan-
di Reserve, 9.33°N, 79.0°W, 16-23.iv.1994, J. Pick-
ering, Malaise trap #2862 [LACM ENT 101530]
(LACM).

PARATYPES. PANAMA: Canal Zone: Barro
Colorado Island, 9.17°N, 79.83°W, 1$, 17-
24.ii.1993, 1$, 24-31.iii.1993, J. Pickering, Mal-
aise trap #740, #959 (LACM); San Bias: Nusagandi
Reserve, 9.33°N, 79.0°W, 19, 5-12.ii.1994, J.
Pickering, Malaise trap #2042 (MIUP).

32 ■ Contributions in Science, Number 482

Brown: Revision of Apocephalus miricauda-group

Apocephalus lopesi (Borgmeier)

(Figs. 62-63, 104)

Anaclinusa lopesi Borgmeier, 1969:64-65, figs. 35-

37; 1971:5-6, fig. 6, new combination.

HOLOTYPE. 9, BRAZIL: Para: Belem, vii.1965,
H.S. Lopes (MZSP; examined).

SPECIES RECOGNITION. This species is ex-
tremely distinctive, with the reclinate supra-anten-
nal setae on the unusually modified frons (in both
sexes) and the characteristic broad ovipositor. It
was previously considered to belong in a separate
genus.

DESCRIPTION. Body length 1.5 mm. Frons yel-
low, anterior margin produced between antennae
(Fig. 104). One pair of divergent, reclinate supra-
antennal setae present. Flagellomere 1 yellow,
round. Proboscis normal, small. Palpus brown.
Dorsum of thorax light brown; pleuron yellow. An-
terior scutellar seta small, fine, subequal to poste-
rior setulae of scutum. Legs yellowish-brown; apex
of hind femur with abrupt darkening on anterior
face. Mean costal length 0.47 wing length. Wing
vein R2+3 present. Halter brown. Abdominal ter-
gites dark brown, except tergite 6, which is partly
to completely yellowish. Venter of abdomen yellow.
Abdominal tergites of normal form. Tergite 3 even-
ly colored. Tergite 6 completely divided, with long
seta at posterolateral corner. Venter of segments 3-
5 with long, dense setae concentrated medially. Ab-
domen without dense lateral setae. Ventral setae of
segment 6 long, consisting of a complete ventral
and lateral row. Ovipositor (Figs. 62-63) straight
in lateral view, lightly but evenly sclerotized dor-
sally. Lateral darkening posteriorly narrowed, mar-
gin dark, complete. Dorsoapical sclerite of ovipos-
itor not differentiated. Apicodorsal margin of ovi-
positor wedge-shaped, posteriorly widened. Api-
coventral margin of ovipositor straight. Ovipositor
without ventral postapical sclerite. Sternite 7
round. Dufour’s mechanism round. Abdominal
glands in segment 5 white, inconspicuous in cleared
specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from lowland sites in Brazil, Colombia, Costa Rica,
and Guyana.

WAY OF LIFE. Females were attracted to
crushed workers of Odontomachus haematodus
(Linnaeus), the presumed host, in Colombia and
Guyana. This ant, whose taxonomic history is par-
ticularly convoluted (Brown, 1976), apparently oc-
curs only in Amazonian South America. Therefore,
A. lopesi must attack another host in Costa Rica.

MATERIAL EXAMINED. BRAZIL: Para: Utin-
ga, near Belem, Id, 19, xii.1966, Malaise trap,
S.J. de Oliveira (MZSP). COLOMBIA: Amazonas:
Amacayacu National Park, 3.82°S, 70.26°W, 19,
4.ix.l997, B. Brown, G. Kung, injured Odonto-
machus haematodus (LACM). COSTA RICA: Li-
mon: 16 km W Guapiles, 10.15°N, 83.92°W, 19,
ii. 1989, P. Hanson, Malaise trap, 400 m (LACM);
Puntarenas: Cerro Rincon, 8.52°N, 83.47°W, 19,

Contributions in Science, Number 482

iii.1991, P. Hanson, Malaise trap, 745 m [LACM
ENT 017042] (LACM). GUYANA: Berbice: Du-
bulay Ranch, 5.68°N, 57.86°W, 3d, 17-225.1999,
B. Brown, M. Sharkey, Malaise trap #10, #12, #14
(LACM), 49, 185.1999, B. Brown, injured O. hae-
matodus (LACM, UGGG).

Apocephalus funditus- subgroup

DIAGNOSIS. Setation of dorsum of abdomen
short, bristle-like. Ovipositor unusually darkly
sclerotized throughout; strongly downturned at
midlength.

PHYLOGENETIC RELATIONSHIPS. All three
of the diagnostic characters are potential synapo-
morphies of a clade including these two species.

Apocephalus funditus new species
(Fig. 64-65)

SPECIES RECOGNITION. This species has ex-
tensive sclerotization, with all sclerotized portions
of the ovipositor appearing shiny, dark brown. Un-
like A. intonsus new species, it lacks a separate
preapical sclerite.

DESCRIPTION. Body length 1.4-1. 6 mm. Frons
yellow to light brown, anterior margin relatively
straight. One pair of supra-antennal setae present.
Flagellomere 1 yellow, round. Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron yellow to light brown. Anterior
scutellar seta small, fine, subequal to posterior se-
tulae of scutum. Legs yellowish-brown; apex of
hind femur slightly darker on anterior face. Mean
costal length 0.44 wing length; range 0.4-0.46.
Wing vein R2+3 present. Halter light brown to
brown. Abdominal tergites dark brown (yellowish
brown in one Costa Rican specimen). Venter of ab-
domen yellow. Abdominal tergites of normal form.
Tergite 3 evenly colored. Tergite 6 completely di-
vided, with short setae at posterolateral corner.
Venter of segments 3-5 bare. Abdomen without
dense lateral setae. Ventral setae of segment 6 long,
consisting of a complete ventral and lateral row.
Ovipositor (Fig. 65) deflected ventrally at midpoint,
darkly sclerotized, with sclerotization extended
completely to posterior apex. Lateral darkening
posteriorly enlarged, inner margin extended medi-
ally, margin dark, complete. Dorsoapical sclerite of
ovipositor not differentiated. Apicodorsal margin
of ovipositor rounded. Apicoventral margin of ovi-
positor straight. Ovipositor without ventral postap-
ical sclerite. Sternite 7 broad, narrowed apically but
widened abruptly at apex. Abdominal glands in
segment 5 white, inconspicuous in cleared speci-
mens. Internal sclerite with small opening and
broad, moderately sclerotized process.

GEOGRAPHICAL DISTRIBUTION. Known
from Brazil, Costa Rica, Ecuador, and Peru.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for complete or wholly, re-

Brown: Revision of Apocephalus miricauda-group ■ 33

Figures 64-71. Ovipositors. Figures 64-65. Apocephalus funditus new species. 64. Dorsal. 65. Left lateral. 66. Apoce-
phalus intonsus new species, dorsal. 67. Apocephalus spatulicauda Borgmeier, dorsal. 68. Apocephalus inimicus Borg-
meier, dorsal. 69. Apocephalus brevifrons new species, dorsal. 70. Apocephalus densepilosus Borgmeier, dorsal. 71. Apo-
cephalus comosus new species, dorsal.

ferring to the extensive, dark sclerotization of the
ovipositor.

HOLOTYPE. $, COSTA RICA: Puntarenas: 24
km W Piedras Blancas, 8.77°N, 83.4°W, ii.1992, P.
Hanson, Malaise trap, 200 m [LACM ENT
012697] (LACM).

PARATYPES. BRAZIL: Roraima: Ilha de Mara-
ca, 3.37°N, 61.43°W, 5$, 2-13.V.1987, J. Rafael,
Malaise trap (INPA, LACM). COSTA RICA: 1$,
same data as holotype. ECUADOR: Sucumbios: Sa-
cha Lodge, 0.5°S, 76.5°W, 1$, 12-22.ii.1994, 1 9,
24.hi-3.iv.1994, 2$, 21-31.X.1994, P. Hibbs, Mal-
aise trap, 270 m (LACM, QCAZ). PERU: Madre
de Dios, Zona Reserva Manu, Pakitza, 11.95°S,
71.28°W, 1?, 7-9.iii.1992, R. Cambra, Malaise
trap (LACM).

34 ■ Contributions in Science, Number 482

Apocephalus intonsus new species

(Figs. 66)

SPECIES RECOGNITION. The similarites of
this species to A. funditus are discussed above. It is
also similar to A. indistinctus new species, except
that the ovipositor is downturned sharply at mid-
length.

DESCRIPTION. Body length 1.1-1. 3 mm. Frons
light brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 light brown, round. Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron yellow. Anterior scutellar seta
small, fine, subequal to posterior setulae of scutum.
Legs yellowish-brown; apex of hind femur with

Brown: Revision of Apocephalus miricauda- group

abrupt darkening on anterior face. Mean costal
length 0.44 wing length; range 0.42-0.45. Wing
vein R2+3 present. Halter brown. Abdominal ter-
gites dark brown, with short, bristly setae. Venter
of abdomen yellow. Abdominal tergites of normal
form. Tergite 3 evenly colored. Tergite 6 anteriorly
emarginate, with short setae at posterolateral cor-
ner. Venter of segments 3-5 with few, medial setae.
Abdomen without dense lateral setae. Ventral setae
of segment 6 long, consisting of a complete ventral
and lateral row. Ovipositor (Fig. 66) deflected ven-
trally at midpoint, lightly but evenly sclerotized
dorsally. Lateral darkening posteriorly enlarged, in-
ner margin extended medially (slightly), margin
dark, complete. Dorsoapical sclerite of ovipositor
rectangular, anteriorly emarginate. Apicodorsal
margin of ovipositor straight. Apicoventral margin
of ovipositor drawn out into pointed process. Ovi-
positor without ventral postapical sclerite. Sternite
7 broadly, evenly developed across entire venter of
ovipositor. Dufour’s mechanism not seen. Abdom-
inal glands in segment 5 white, inconspicuous in
cleared specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in Panama.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for unshaven referring to the
short, bristly setae of the tergites.

HOLOTYPE. 9, PANAMA: Canal Zone: Barro
Colorado Island, 9.17°N, 79.83°W, 11-18.V.1994,
J. Pickering, Malaise trap #2424 [LACM ENT
093710] (LACM).

PARATYPES. PANAMA: Canal Zone: Barro
Colorado Island, 9.17°N, 79.83°W, 19, 10-
17.iii.1993, 19, 6-13.iv.1994, 1 9, 15-22.vi.1994,
J. Pickering, Malaise trap #957, #2419, #2384
(LACM, MIUP).

A. spatulicauda-subgrowp

DIAGNOSIS. Frons short. Ovipositor with dis-
tinctive line across lateral darkenings, indicating
abrupt downturn of direction. Lateral darkenings
broad. Ventral apex of ovipositor with dark, heavi-
ly sclerotized, transverse sclerite.

PHYLOGENETIC RELATIONSHIPS. The char-
acters in the diagnosis are all potential synapomor-
phies of the group. The relationships among the
three species are not resolved.

WAY OF LIFE. Unknown.

Apocepbalus spatulicauda Borgmeier

(Fig. 67)

Apocepbalus spatulicauda Borgmeier, 1961:45,

figs. 64, 73, 85.

HOLOTYPE. 9, BRAZIL: Rio de Janeiro: Pe-
tropolis, 15.vi.1923, T. Borgmeier [LACM ENT
122183] (MZSP; examined).

SPECIES RECOGNITION. This species has a
distinct ovipositor, with broad lateral darkenings
and a narrow preapical sclerite.

Contributions in Science, Number 482

DESCRIPTION. Unfortunately, the unique fe-
male holotype was heavily damaged in the mail.
Little remains except the abdomen, so the following
description is fragmentary and based in part on the
original description of the species.

Body length 1.5 mm. One pair of supra-antennal
setae present. Flagellomere 1 yellow, round. Pro-
boscis normal, small. Palpus yellow. Dorsum of
thorax light brown. Anterior scutellar seta small,
fine, subequal to posterior setulae of scutum. Mean
costal length 0.44 wing length. Halter yellow. Ven-
ter of segments 3-5 with a few, scattered setae. Ab-
domen without dense lateral setae. Ovipositor (Fig.

67) slightly downturned apically, lightly but evenly
sclerotized dorsally. Lateral darkening broadened,
with apical concavity, margin dark, complete. Dor-
soapical sclerite of ovipositor elongate, anteriorly
emarginate. Apicodorsal margin of ovipositor
straight. Apicoventral margin of ovipositor drawn
out into pointed process. Ovipositor without ven-
tral postapical sclerite. Sternite 7 narrow but api-
cally expanded and with short, broad, black, apical
darkening. Dufour’s mechanism not seen. Abdom-
inal glands in segment 5 white, inconspicuous in
cleared specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in southeastern Brazil.

WAY OF LIFE. Unknown.

Apocepbalus inimicus Borgmeier

(Fig. 68)

Apocepbalus inimicus Borgmeier, 1961:42, fig. 63.

HOLOTYPE. 9, BRAZIL: Santa Catarina:
Nova Teutonia, vii.1952, F. Plaumann, with Labi-
dus coecus [LACM ENT 122468] (MZSP; exam-
ined).

SPECIES RECOGNITION. This species has
broad lateral darkenings and an extremely broad,
heavily sclerotized preapical sclerite.

DESCRIPTION. Body length 1.3-1. 4 mm. Frons
light brown (bleached from alcohol), anterior mar-
gin relatively straight, frons short. Two pairs of su-
pra-antennal setae present; lower pair markedly
smaller than upper pair, greatly reduced. Flagello-
mere 1 brown, oval. Proboscis normal, small. Pal-
pus yellow. Dorsum of thorax light brown; pleuron
light brown. Anterior scutellar seta small, fine, sub-
equal to posterior setulae of scutum. Legs yellow-
ish-brown; apex of hind femur with abrupt dark-
ening on anterior face. Mean costal length 0.42
wing length; range 0.42-0.43. Wing vein R2+3 pre-
sent. Halter light brown. Abdominal tergites dark
brown. Venter of abdomen gray. Abdominal ter-
gites of normal form. Tergite 3 evenly colored. Ter-
gite 6 completely divided, with short setae at pos-
terolateral corner. Venter of segments 3-5 with a
few, scattered setae. Abdomen without dense lateral
setae. Ventral setae of segment 6 long, consisting of
a complete ventral and lateral row. Ovipositor (Fig.

68) slightly downturned apically, lightly but evenly
sclerotized dorsally. Lateral darkening broadened,

Brown: Revision of Apocepbalus miricauda- group ■ 35

with apical concavity, margin dark, complete. Dor-
soapical sclerite of ovipositor rectangular, anteri-
orly emarginate. Apicodorsal margin of ovipositor
straight. Apicoventral margin of ovipositor drawn
out into pointed process. Ovipositor without ven-
tral postapical sclerite. Sternite 7 narrow but api-
cally expanded and with short, broad, black apical
darkening. Dufour’s mechanism not seen. Abdom-
inal glands in segment 5 white, inconspicuous in
cleared specimens. Internal sclerite a simple, round
loop.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in southeastern Brazil.

WAY OF LIFE. Although the type specimens
were collected with army ants, the host of this spe-
cies is unknown.

OTHER MATERIAL EXAMINED. Paratype <3,
9, same data as holotype (MZSP).

Apocepbalus brevifrotts new species

(Figs. 69, 105)

SPECIES RECOGNITION. This species is easily
recognized by the extremely enlarged lateral dark-
enings of the ovipositor, the lack of wing vein R2+3
and the lack of lower fronto-orbital setae.

DESCRIPTION. Body length 1.3-1. 4 mm. Frons
(Fig. 105) light brown, anterior margin relatively
straight. One pair of supra-antennal setae present.
Flagellomere 1 light brown, oval (slightly pointed
in the male, shown in Fig. 105). Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron light brown. Anterior scutellar seta
small, fine, subequal to posterior setulae of scutum.
Legs yellowish-brown; apex of hind femur slightly
darker on anterior face. Mean costal length 0.46
wing length; range 0.43-0.49, slightly thickened.
Wing vein R2+3 absent. Halter brown. Abdominal
tergites dark brown. Venter of abdomen yellow to
gray. Abdominal tergites of normal form. Tergite 3
evenly colored. Tergite 6 completely divided, with
setae of medium length along posterior margin.
Venter of segments 3-5 with short setae, concen-
trated medially. Abdomen without dense lateral se-
tae. Ventral setae of segment 6 long, consisting of
a complete ventral and lateral row. Ovipositor (Fig.
69) slightly downturned apically, lightly but evenly
sclerotized dorsally. Lateral darkening broadened,
with apical concavity, margin dark, complete. Dor-
soapical sclerite of ovipositor not differentiated.
Apicodorsal margin of ovipositor pointed. Apico-
ventral margin of ovipositor straight. Ovipositor
without ventral postapical sclerite. Sternite 7 nar-
row but apically expanded and with short, broad,
black apical darkening. Dufour’s mechanism not
seen. Abdominal glands in segment 5 white, incon-
spicuous in cleared specimens. Internal sclerite a
simple, round loop.

GEOGRAPHICAL DISTRIBUTION. Known
from mid-elevation sites in Mexico and Costa Rica.

WAY OF LIFE. Unknown.

36 ■ Contributions in Science, Number 482

DERIVATION OF SPECIFIC EPITHET. This
species is named for the short frons.

HOLOTYPE. 9, COSTA RICA: San Jose: Zur-
qui de Moravia, 10.05°N, 84.02°W, ix-x.1993, P.
Hanson, Malaise trap, 1600 m [LACM ENT
014242] (LACM).

PARATYPES. MEXICO: Chiapas: Montebello,
19, 16.12°N, 91.67°W, 19.vi.1969 (LACM). COS-
TA RICA: Puntarenas: Monteverde Biological Sta-
tion, 10.32°N, 84.8°W, Id, 25-28.V.1998, B.
Brown, V. Berezovskiy, Malaise trap #4, 1500 m
(LACM); San Jose: Zurqui de Moravia, 10.05°N,
84.02°W, Id, ii.1989, 19, iii.1989. Id, 19, x-
xii.1990, Id, 19, i.1991, 2d, iii.1992, 19,
v.1992, 5d, vi.1992, 10d, 39, vii.1992, 3d, iv-
v.1993, 12d, 19, l-15.vi.1993, 9d, 59, ix-
x.1993, 5d, 19, ii. 1 994, 6d, 39, iii.1994, 3d,
19, iv.1994, 4d, v.1994, 3d, i.1996, P. Hanson,
Malaise trap (INBC, LACM, MCZC, MUCR,
MZSP, USNM).

Other Taxa (subgroup unknown)

Apocepbalus densepilosus Borgmeier

(Figs. 70, 112)

Apocepbalus densepilosus Borgmeier, 1971:110,

figs. 147-148.

HOLOTYPE. 9, BRAZIL: Para: Utinga,
xii.1966, S.J. de Oliveira [LACM ENT 049341]
(MZSP; examined).

SPECIES RECOGNITION. This species has dis-
tinctive extremely dense setation laterally on ab-
dominal segments 4 and 5, unlike other species that
have none at all on segment 4 and usually segment 5.

DESCRIPTION. Body length 1. 5-2.1 mm. Frons
yellow, anterior margin relatively straight. One or
two pairs of supra-antennal setae present; lower
pair smaller than upper pair. Flagellomere 1 yellow,
round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax light brown; pleuron yellow. An-
terior scutellar seta small, fine, subequal to poste-
rior setulae of scutum. Legs yellowish-brown; apex
of hind femur with abrupt darkening on anterior
face. Mean costal length 0.55 wing length; range
0.53-0.56. Wing vein R2+3 present. Halter brown.
Abdominal tergites dark brown. Venter of abdo-
men yellow. Abdominal tergites of normal form.
Tergite 3 evenly colored. Tergite 6 anteriorly and
posteriorly emarginate, with short setae at postero-
lateral corner. Venter of segments 3-5 with few, me-
dial setae. Abdomen with dense lateral setae on seg-
ments 4 and 5 (Fig. 112). Ventral setae of segment
6 long, consisting of a complete ventral and lateral
row. Ovipositor (Fig. 70) slightly downturned api-
cally, with broad median sclerite. Lateral darkening
thin, subparallel, margin dark, complete. Dorsoap-
ical sclerite of ovipositor heart-shaped. Apicodorsal
margin of ovipositor straight. Apicoventral margin
of ovipositor straight. Ovipositor without ventral
postapical sclerite. Sternite 7 narrow but apically
expanded. Abdominal glands in segment 5 white,

Brown: Revision of Apocepbalus miricauda-group

inconspicuous in cleared specimens. Internal scler-
ite with small opening and broad, moderately scler-
otized process.

GEOGRAPHICAL DISTRIBUTION. Known
from Brazil, Colombia, Ecuador, and Peru. The
host ant has also been recorded from French Gui-
ana, Trinidad, and Venezuela.

WAY OF LIFE. Females are attracted to injured
workers of the presumed host species, Pachycon-
dyla crassinoda (Latreille).

MATERIAL EXAMINED. BRAZIL: Roraima:
Ilha de Maraca, 3.37°N, 61.43°W, 29, 2-
13.V.1987, J. Rafael, Malaise trap (INPA, LACM).
COLOMBIA: Amazonas: 7 km W Leticia, 4.13°S,
69.9° W, 19, 26.viii.1997, B. Brown, G. Kung, in-
jured Pachycondyla crassinoda (LACM). ECUA-
DOR: Napo: Yasuni Biological Research Station,

O. 67°S, 76.36°W, 1 9, 24.V.1996, B. Brown, injured

P. crassinoda (LACM); Sucumbios: Anagu, 0.48°S,
76.38°W, 1 9 , 9.ix.l997, P. DeVries, injured P. cras-
sinoda (LACM), Sacha Lodge, 0.5°S, 76.5°W, 19,
14-24.iii.1994, 19, 14-24.V.1994, P. Hibbs, Mal-
aise trap, 270 m (LACM, QCAZ). PERU: Madre
de Dios, Zona Reserva Pakitza, 11.94°S, 71.28°W,
19, 14.ii.1992, 19, 27.ii.1992, B. Brown, D. Fee-
ner, injured P. crassinoda (MUSM, USNM).

Apocephalus comosus new species

(Figs. 71, 113)

SPECIES RECOGNITION. This species is easily
recognized by the lateral group of setae on abdom-
inal segment 5 and the distinctive shape of the
preapical sclerite.

DESCRIPTION. Body length 1.8-2. 5 mm. Frons
yellow to dark brown, anterior margin produced
between antennae. One pair of supra-antennal se-
tae present. Flagellomere 1 yellow to light brown,
round. Proboscis elongate; anterior margin of frons
produced anteriorly. Palpus brown. Dorsum of tho-
rax light brown; pleuron white, strongly and sharp-
ly contrasting with scutum. Anterior scutellar seta
slightly enlarged, bristle-like. Legs yellowish-
brown; apex of hind femur with abrupt darkening
on anterior face. Mean costal length 0.58 wing
length; range 0.55-0.61. Wing vein R2+3 present.
Halter brown. Abdominal tergites dark brown.
Venter of abdomen yellow. Abdominal tergites 2
extremely short; 3-5 elongate in compensation.
Tergite 3 dark, with round, yellow spot laterally.
Tergite 6 anteriorly emarginate, with short setae at
posterolateral corner. Venter of segments 3-5 bare.
Abdomen with patch of long, dense lateral setae on
a small evagination on segment 6 (Fig. 113). Ven-
tral setae of segment 6 long, consisting of single
median pair. Ovipositor (Fig. 71) slightly upturned
apically, evenly sclerotized, with long, thin, pos-
teromedial process dorsally. Lateral darkening thin,
subparallel, margin dark, complete. Dorsoapical
sclerite of ovipositor spade-shaped. Apicodorsal
margin of ovipositor straight. Apicoventral margin
of ovipositor rounded. Ovipositor without ventral

Contributions in Science, Number 482

postapical sclerite. Sternite 7 broad, with pointed,
mediolateral projection. Dufour’s mechanism
round. Abdominal glands in segment 5 dark, en-
larged, elongate, clearly visible in cleared speci-
mens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. This spe-
cies is widely distributed in the lowlands through-
out the Neotropical Region, from southern Mexico
to Ecuador.

WAY OF LIFE. One female of this species was
attracted to injured Ectatomma tuberculatum
workers; one was attracted to an injured Odonto-
machus laticeps. The actual host ant of this rela-
tively commonly collected species is still unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for hairy, referring to the
highly setose abdomen.

HOLOTYPE. 9, COSTA RICA: Puntarenas: 3
km SW Rincon, 8.68°N, 83.48°W, iii-v.1991, P.
Hanson, Malaise trap, 10 m [LACM ENT 012879]
(LACM).

PARATYPES. COSTA RICA: Alajuela: San Pe-
dro de la Tigra, 10.37°N, 84.57°W, 1 9 , u.1990, P.
Hanson, Malaise trap, 200 m (LACM); Guana-
caste: Volcan Cacao, Cerro Pedregal, 10.93°N,
85.48°W, 19, ii— iv. 1989, 1. Gauld, D. Janzen, Mal-
aise trap, 1000 m (LACM); Heredia: La Selva Bi-
ological Station, 10.43°N, 84.02°W, 29, 1-
8.V.1989, 19, 8-15.V.1989, 19, 15-21.V.1989, B.
Brown, D. Feener, Malaise trap, SHO@SOR
(LACM), 29, iii.1991, P. Hanson, Malaise trap
(LACM, MUCR), 19, 15.ii-l.iii.1993, ALAS,
Malaise trap M/10/25 (INBC), 19, l-15.iii.1993,
ALAS, Malaise trap M/10/41 (INBC), 3 9, 1-
15.iv.1993, ALAS, Malaise trap M/08/71, M/12/
75, M/15/78 (INBC), 29, 1-15.V.1993, ALAS,
Malaise trap M/10/104 (INBC), 19, 15.vi-
l.vii.1993, ALAS, Malaise trap M/06/151 (INBC),
19, 15.vi-l.vii.1993, ALAS, Malaise trap M/12/
118 (INBC), 1 9, 22-24.vi.1993, B. Brown, D. Fee-
ner, Malaise trap #1 (LACM), 19, 24.vi.1993, B.
Brown, injured Ectatomma tuberculatum (LACM),
29, l.vii.1993, ALAS, Malaise trap M/12/146
(INBC), 29, 15.vii.1993, ALAS, Malaise trap MJ
12/162 (INBC), 29, 3.viii.l993, ALAS, Malaise
trap, M/12/174 (INBC), 19, 15.xi.1993, ALAS,
Malaise trap M/04/267 (INBC), 49, 15.xi.1993,
ALAS, Malaise trap M/12/274 (INBC), 29,
3.i.l994, ALAS, Malaise trap, M/01/304, M/09/
311 (INBC), 19, 15.i. 1994, ALAS, Malaise trap
M/08/32 6 (INBC), 19, 15.ii.1994, ALAS, Malaise
trap, M/05/352 (INBC), 19, l.iii.1994, ALAS,
Malaise trap, M/04/363 (INBC), 19, 30.vi.1995,
ALAS, Malaise trap M/01/388 (INBC), 19,
l.xii.1995, ALAS, Malaise trap M/01/507 (INBC),
19, 15.X.1997, ALAS, Malaise trap M/18/686
(INBC); Limon: 4 km NE Bribri, 9.63°N, 82.82°W,
29, xii.1989-iii.1990, P. Hanson, Malaise trap, 50
m (LACM), 7 km SW Bribri, 9.58°N, 82.88°W, 2 9 ,
ix-x.1989, P. Hanson, Malaise trap (LACM), 16
km W Guapiles, 10.15°N, 83.92°W, 1 9, i-iv.1991,
P. Hanson, Malaise trap, 400 m (LACM); Puntar-

Brown: Revision of Apocephalus miricauda- group I 37

enas: Coopemarti, 8.63°N, 83.47°W, 6 9, ii.1991,
P. Hanson, Malaise trap, 30 m (LACM), 5 km W
Piedras Blancas, 8.77°N, 83.28°W, 1$, vi-
viii.1989, 2 9 , vi.1991, 1 9 , vii.1991, 1 9 , viii.1991,
P. Hanson, Malaise trap, 100 m (LACM), 10 km
W Piedras Blancas, 8.75°N, 83.3°W, 2$, iii-
v.1989, P. Hanson, Malaise trap, 100 m (LACM),
24 km W Piedras Blancas, 8.77°N, 83.4°W, 1 9 , iii-
v.1989, 19, xii.1991, P. Hanson, Malaise trap
(LACM), 5 km NW Puerto Jimenez, 8.55°N,
83.35°W, 1 9, v.1991, P. Hanson, Malaise trap, 10
m (LACM), 23 km N Puerto Jimenez, 8.67°N,
83.45°W, 19, i— iv. 1991, 29, viii.1991, P. Hanson,
Malaise trap, 10 m (LACM), 3 km SW Rincon,
8.68°N, 83.48°W, 19, iii.1989, 19, xii.1989, 19,
m-vi.l990, 19, vii-ix.1990, 69, iii-v.1991, 29,
viii.1991, 19, x.1991, 19, ii.1992, P. Hanson,
Malaise trap, 10 m (LACM, MUCR), 5 km SW
Rincon, 8.7°N, 83.51°W, 49, 31.v-7.vi.1998, B.
Brown, V. Berezovskiy, Malaise trap #3, Malaise
trap #5 (LACM), 1 9, 3.vi.l998, B. Brown, injured
Odontomacbus laticeps (LACM), Rio Piro, 8.28°N,
83.32°W, 1 9, ii.1991, P. Hanson, Malaise trap, 75
m (LACM); San Jose: Zurqui de Moravia, 10.05°N,
84.02°W, 19, vii.1990, P. Hanson, Malaise trap,
1600 m (LACM). ECUADOR: Sucumbios: Sacha
Lodge, 0.5°S, 76.5°W, 19, 3-13.vi.1994, P. Hibbs,
Malaise trap, 270 m (QCAZ), Tiputini Station,
0.67°S, 76.25°W, 19, 28-30.vi.1998, E. Holscher,
C. Carter, Malaise trap (LACM). MEXICO: Vera-
cruz: 33 km NE Catemaco, Los Tuxtlas Biological
Station, 19, l.vii— 1 .viii.1983, S. and J. Peck, FIT,
rain forest, 160 m (LACM). PANAMA: Canal
Zone: Barro Colorado Island, 9.15°N, 79.85°W,
1 9, 20-275.1993, J. Pickering, Malaise trap #736
(LACM), 19, 10-17.iii.1993, J. Pickering, Malaise
trap #957, 99, 24-31.iii.1993, J. Pickering, Mal-
aise trap #935, #959 (LACM, MIUP), 19, 285v-
5.V.1993, J. Pickering, Malaise trap #940 (LACM),
3 9, 5-12.V.1993, J. Pickering, Malaise trap #941
(LACM), 19, 9-23.vi.1993, J. Pickering, Malaise
trap #946 (LACM), 19, 25.viii-l.ix.1993, J. Pick-
ering, Malaise trap #1676 (LACM), 19, 8-
15. ix. 1993, J. Pickering, Malaise trap #1672
(LACM), 23 9, 6-13.X.1993, J. Pickering, Malaise
trap #1693 (LACM, MCZC, MIUP, MUSP,
USNM), 1 9 , 2-9.ii.1994, J. Pickering, Malaise trap
#2366 (LACM), 19, 11-18.V.1994, J. Pickering,
Malaise trap #2424 (LACM), 19, 22-29.vi.1994,
J. Pickering, Malaise trap #2385 (LACM), 29, 24-
31.viii.1994, J. Pickering, Malaise trap #2394
(LACM), 49, 6-13.xi.1996, J. Pickering, Malaise
trap #7046 (LACM); San Bias: Nusagandi Reserve,
9.33°N, 79°W, 19, 14-21.V.1994, J. Pickering,
Malaise trap #2871 (LACM).

Apocephalus lobicauda new species

(Fig. 72)

SPECIES RECOGNITION. This species is rec-
ognized by its short, broad, rounded lateral dark-
enings.

38 ■ Contributions in Science, Number 482

DESCRIPTION. Body length 1.1-1. 4 mm. Frons
yellow to light brown, anterior margin relatively
straight. One pair of supra-antennal setae present.
Flagellomere 1 yellow, round. Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron yellow. Anterior scutellar seta
small, fine, subequal to posterior setulae of scutum.
Legs yellowish-brown; apex of hind femur with
abrupt darkening on anterior face. Mean costal
length 0.43 wing length; range 0.40-0.45. Wing
vein R2+3 present. Halter light brown. Abdominal
tergites dark brown. Venter of abdomen yellow to
gray. Abdominal tergites of normal form. Tergite 3
evenly colored. Tergite 6 anteriorly emarginate,
with short setae at posterolateral corner. Venter of
segments 3-5 bare. Abdomen without dense lateral
setae. Ventral setae of segment 6 long, consisting of
a complete ventral and lateral row. Ovipositor (Fig.
72) slightly downturned apically, lightly but evenly
sclerotized dorsally. Lateral darkening broadened,
enlarging posteriorly, apically rounded, margin
dark, complete. Dorsoapical sclerite of ovipositor
trapezoidal, but anteriorly emarginate. Apicodorsal
margin of ovipositor straight. Apicoventral margin
of ovipositor straight. Ovipositor without ventral
postapical sclerite. Sternite 7 narrow but apically
expanded. Abdominal glands in segment 5 white,
inconspicuous in cleared specimens. Internal scler-
ite with small opening and broad, moderately scler-
otized process.

GEOGRAPHICAL DISTRIBUTION. Known
from a number of lowland sites in Costa Rica.

WAY OF LIFE. Females are attracted to injured
Ectatomma tuberculatum workers.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin words lobus and cauda,
for lobe and tail, referring to the shape of the lateral
darkenings.

HOLOTYPE. 9, COSTA RICA: Limon: 7 km
SW Bribri, 9.58°N, 82.88°W, ix-xi.1989, P. Han-
son, Malaise trap [LACM ENT 005279] (LACM).

PARATYPES. COSTA RICA: Heredia: La Selva
Biological Station, 10.43°N, 84.02°W, 2 9, 1-
15.iv.1993, ALAS, Malaise trap M/15/78, M/01/64
(INBC), 19, l.vi.1993, ALAS, Malaise trap M/7/
113 (INBC), 19, l.vii.1993, ALAS, Malaise trap
M/06/151 (INBC), 19, l-15.vii.1993, ALAS, Mal-
aise trap M/10/160 (INBC), 19, l.xi.1993, ALAS,
Malaise trap M/01/248 (INBC), 109, l.xii.1993,
ALAS, Malaise trap M/01/276 (INBC, LACM),
8 9, 35.1994, ALAS, Malaise trap M/01/304
(INBC, MCZ, USNM) 1 9 , l.iii.1994, ALAS, Mal-
aise trap M/01/360 (INBC), 29, 45v.l994, ALAS,
Malaise trap M/01/384 (INBC), 29, 15.vii.1995,
29, 19.vii.1995, D. Feener, injured Ectatomma
tuberculatum (LACM); Limon: 4 km NE Bribri,
9.63°N, 82.82°W, 19, xii.1989-iii.1990, P. Han-
son, Malaise trap, 50 m (LACM), 7 km SW Bribri,
9.58°N, 82.88°W, 49, ix-xi.1989, P. Hanson, Mal-
aise trap (LACM, MUCR); Puntarenas: 24 km W
Piedras Blancas, 8.77°N, 83.4°W, 1 9, iv-v.1991, P.
Hanson, Malaise trap (LACM), 3 km SW Rincon,

Brown: Revision of Apocephalus miricauda- group

76. 77.

78. 79.

Figures 72-79. Ovipositors. 72. Apocepbalus lobicauda new species, dorsal. 73. Apocephalus globosus new species,
dorsal. 74. Apocephalus maculosus new species, dorsal. 75. Apocephalus glabriventris new species, dorsal. 76. Apoce-
phalus minutus Borgmeier, dorsal. 77. Apocephalus cardiacus new species, dorsal. Figures 78-79. Apocephalus petiolus
new species. 78. Dorsal. 79. Ventral.

8.68°N, 83,48°W, 2 ? , x-xii.1990, P. Hanson, Mal-
aise trap (LACM), Sirena, 8.48°N, 83.6°W, 1$,
9.vii.l993, D. Feener, injured E. tuberculatum
(LACM).

Apocepbalus globosus new species

(Fig. 73)

SPECIES RECOGNITION. This species is rec-
ognized by the distinctive large rounded preapical
sclerite.

DESCRIPTION. Body length 1.4 mm. Frons
light brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel
lomere 1 light brown, round. Proboscis normal,
small. Palpus yellow. Dorsum of thorax light

Contributions in Science, Number 482

brown; pleuron white. Anterior scutellar seta small,
fine, subequal to posterior setulae of scutum. Legs
yellowish-brown; apex of hind femur with abrupt
darkening on anterior face. Mean costal length
0.47 wing length. Wing vein R2+3 present. Halter
brown. Abdominal tergites dark brown. Venter of
abdomen yellow. Abdominal tergites of normal
form. Tergite 3 evenly colored. Tergite 6 anteriorly
emarginate, with short setae at posterolateral cor-
ner. Venter of segments 3-5 with short setae, con-
centrated medially. Abdomen without dense lateral
setae. Ventral setae of segment 6 long, consisting of
a complete ventral and lateral row. Ovipositor (Fig.
73) straight in lateral view, lightly but evenly scler-
otized dorsally. Lateral darkening broadened, en-

Brown: Revision of Apocephalus miricauda- group ■ 39

larging posteriorly, apically rounded, margin dark,
complete. Dorsoapical sclerite of ovipositor large,
round, dark, shiny. Apicodorsal margin of ovipos-
itor straight. Apicoventral margin of ovipositor
straight. Ovipositor without ventral postapical
sclerite. Sternite 7 triangular, apically tridentate and
with apical, triangular sclerite. Abdominal glands
in segment 5 white, inconspicuous in cleared spec-
imens. Internal sclerite a simple, round loop.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in northern Costa Rica.

WAY OF LIFE. The single specimen was attract-
ed to injured Pachycondyla villosa workers.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for round, referring to the
shape of the preapical sclerite.

HOLOTYPE. 9, COSTA RICA: Guanacaste: Es-
tacion Biologia Pitilla, 11.0°N, 85.43°W, 5.vii.l997,
B. Brown, injured Pachycondyla villosa [LACM
ENT 099871] (LACM).

Apocephalus maculosus new species

(Figs. 74, 110)

SPECIES RECOGNITION. This species is distin-
guished by light-colored spots on tergite 3 and the
unusual structure of the ovipositor.

DESCRIPTION. Body length 1.5 mm. Frons yel-
low, anterior margin relatively straight. One pair of
supra-antennal setae present. Flagellomere 1
brown, round. Proboscis normal, small. Palpus yel-
low. Dorsum of thorax light brown; pleuron yel-
low. Anterior scutellar seta small, fine, subequal to
posterior setulae of scutum. Legs yellow; apex of
hind femur with abrupt darkening on anterior face.
Mean costal length 0.55 wing length. Wing vein
R2+3 present. Halter brown. Abdominal tergites
dark brown. Venter of abdomen yellow. Abdominal
tergites of normal form. Tergite 3 dark, with round,
yellow spot laterally (Fig. 110). Tergite 6 anteriorly
emarginate, with short setae at posterolateral cor-
ner. Venter of segments 3-5 bare, except for small
patch of setae mediolaterally on segment 5. Abdo-
men without dense lateral setae. Ventral setae of
segment 6 long, consisting of a complete ventral
and lateral row. Ovipositor (Fig. 74) straight in lat-
eral view, lightly but evenly sclerotized dorsally.
Lateral darkening broad, margin dark, complete.
Dorsoapical sclerite of ovipositor trapezoidal, but
anteriorly emarginate. Apicodorsal margin of ovi-
positor straight. Apicoventral margin of ovipositor
straight. Ovipositor without ventral postapical
sclerite. Sternite 7 triangular. Dufour’s mechanism
elongate. Abdominal glands in segment 5 white, in-
conspicuous in cleared specimens. Internal sclerite
rounded, with a short process.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in Amazonian Ecuador.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin word maculosus for

40 ■ Contributions in Science, Number 482

spotted, referring to the light-colored, round spots
on tergite 3.

HOLOTYPE. 9 , ECUADOR: Sucumbios: Sacha
Lodge, 0.5°S, 76.5°W, l-31.xii.1994, P. Hibbs,
Malaise trap, 270 m [LACM ENT 050741]
(LACM).

Apocephalus glabriventris new species

(Fig. 75)

SPECIES RECOGNITION. This species belongs
to a group of similar species with small preapical
sclerites on the ovipositor. It is recognized by the
large notch in the anterior margin of the sclerite, as
well as the bare venter of the abdomen.

DESCRIPTION. Body length 1.0 mm. Frons yel-
low, anterior margin relatively straight. One pair of
supra-antennal setae present. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax light brown; pleuron light
brown. Anterior scutellar seta small, fine, subequal
to posterior setulae of scutum. Legs yellowish-
brown; apex of hind femur with abrupt darkening
on anterior face. Mean costal length 0.47 wing
length. Wing vein R2+3 present. Halter brown. Ab-
dominal tergites dark brown. Venter of abdomen
gray. Abdominal tergites of normal form. Tergite 3
evenly colored. Tergite 6 anteriorly emarginate,
with short setae at posterolateral corner. Venter of
segments 3-5 bare. Abdomen without dense lateral
setae. Ventral setae of segment 6 long, consisting of
a complete ventral and lateral row. Ovipositor (Fig.
75) straight in lateral view, lightly but evenly scler-
otized dorsally, darker than in similar species. Lat-
eral darkening posteriorly enlarged, truncate, mar-
gin dark, complete. Dorsoapical sclerite of ovipos-
itor heart-shaped. Apicodorsal margin of ovipositor
straight. Apicoventral margin of ovipositor
straight. Ovipositor without ventral postapical
sclerite. Sternite 7 narrow but apically expanded.
Abdominal glands in segment 5 white, inconspicu-
ous in cleared specimens. Internal sclerite with
small opening and broad, moderately sclerotized
process.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in western Mexico.

WAY OF LIFE. The single known specimen of
this species was attracted to a pair of fighting Ec-
tatomma ruidum (Roger)

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin words glaber and ven-
tris, for bare and belly, referring to the bare venter
of the abdomen.

HOLOTYPE. 9, MEXICO: Jalisco: Chamela,
19.52°N, 105.08°W, l.x.1990, W. Eberhard, over
fighting Ectatomma ruidum [LACM ENT 012756]
(LACM).

Apocephalus minutus Borgmeier
(Fig. 76)

Apocephalus minutus Borgmeier, 1958:329, figs.

20, 20a, 39.

Brown: Revision of Apocephalus miricauda-group

Apocephalus angular is Borgmeier, 1971:106-107,

fig. 149, new synonymy.

HOLOTYPE. ?, BRAZIL: Rio de Janeiro: Ja-
carepagua, 1957, T. Borgmeier (MZSP; not exam-
ined).

SPECIES RECOGNITION. This species is distin-
guished by the small heart-shaped preapical sclerite
and the downturned apex of the ovipositor.

In his last key to Apocephalus species, Borgmeier
(1971) stated that A. angularis Borgmeier had an
apically pointed ovipositor, unlike A. minutus,
which was stated to have a small, triangular exci-
sion. This apparent excision, however, was an ar-
tifact of the drying of the intersegmental membrane
posterior to the ovipositor. Examination of speci-
mens of A. angularis showed that they are identical
to A. minutus.

DESCRIPTION. Body length 1.0 mm. Frons yel-
low, anterior margin relatively straight. One pair of
supra-antennal setae present. Flagellomere 1 light
brown, round. Proboscis normal, small. Palpus yel-
low. Dorsum of thorax light brown; pleuron yel-
low. Anterior scutellar seta small, fine, subequal to
posterior setulae of scutum. Legs yellowish-brown;
apex of hind femur slightly darker on anterior face.
Mean costal length 0.46 wing length. Wing vein
R>+3 present. Halter brown. Abdominal tergites
dark brown. Venter of abdomen yellow. Abdominal
tergites of normal form. Tergite 3 evenly colored.
Tergite 6 completely divided, with long seta at pos-
terolateral comer. Venter of segments 3-5 bare. Ab-
domen without dense lateral setae. Ventral setae of
segment 6 long, consisting of a complete ventral
and lateral row. Ovipositor (Fig. 76) slightly down-
turned apically, lightly but evenly sclerotized dor-
sally. Lateral darkening posteriorly enlarged, trun-
cate, margin dark, complete. Dorsoapical sclerite of
ovipositor elongate, anteriorly emarginate. Apico-
dorsal margin of ovipositor straight. Apicoventral
margin of ovipositor straight. Ovipositor without
ventral postapical sclerite. Sternite 7 thin, with ligh-
ter, expanded area at midlength. Du four’s mecha-
nism not seen. Abdominal glands in segment 5
white, inconspicuous in cleared specimens. Internal
sclerite with small opening and broad, moderately
sclerotized process.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in southeastern Brazil.

WAY OF LIFE. Unknown.

MATERIAL EXAMINED. BRAZIL: Rio de Ja-
neiro: Jacarepagua, 1$ [paratype], 18.xi.1957, T.
Borgmeier (MCZC); Sao Paulo: Nova Teutonia,
2? [holotype and paratype of A. angularis ], F.
Plaumann (MZSP).

Apocephalus cardiacus new species

(Fig. 77)

SPECIES RECOGNITION. This species can be
recognized by the heart-shaped preapical sclerite
and the small, internally directed arms of the apex
of the lateral darkenings. It differs from the most

Contributions in Science, Number 482

similar species, A. petiolus new species, by the rel-
atively short, sparse setae on the venter of the ab-
domen.

DESCRIPTION. Body length 1.4-1. 8 mm. Frons
yellow, anterior margin relatively straight. One pair
of supra-antennal setae present. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax light brown; pleuron white. An-
terior scutellar seta large, bristle-like. Legs yellow-
ish-brown; apex of hind femur with abrupt dark-
ening on anterior face. Mean costal length 0.52
wing length; range 0.50-0.56. Wing vein R2+3 pre-
sent. Halter brown. Abdominal tergites dark brown
(some specimens with tergite 3 light brown). Venter
of abdomen yellow to gray. Abdominal tergites of
normal form. Tergite 3 evenly colored. Tergite 6
completely divided, with short setae at posterolat-
eral corner. Venter of segments 3-5 with short se-
tae, concentrated medially. Abdomen without dense
lateral setae. Ventral setae of segment 6 relatively
short, consisting of a complete ventral and lateral
row. Ovipositor (Fig. 77) slightly downturned api-
cally, lightly but evenly sclerotized dorsally. Lateral
darkening posteriorly enlarged, truncate, margin
dark, complete. Dorsoapical sclerite of ovipositor
heart-shaped. Apicodorsal margin of ovipositor
straight. Apicoventral margin of ovipositor
straight. Ovipositor without ventral postapical
sclerite. Sternite 7 broad, narrowed apically, but
widened abruptly at apex. Dufour’s mechanism
elongate. Abdominal glands in segment 5 white, in-
conspicuous in cleared specimens. Internal sclerite
rounded, with long process.

GEOGRAPHICAL DISTRIBUTION. Known
from several lowland sites in Costa Rica.

WAY OF LIFE. Host unknown. This species was
collected almost exclusively by Malaise trap #10 in
the 16-trap ALAS survey. In Brown and Feener
(1995, as “species 143”), the increase in abundance
of this species in trap #10 was hypothesized to be
correlated with the onset of the rainy season.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Greek word kardia for heart,
referring to the shape of the preapical sclerite.

HOLOTYPE. ?, COSTA RICA: Heredia: La
Selva Biological Station, 10.43°N, 84.02°W, 1-
15.V.1993, ALAS, Malaise trap M/10/104 [INBI-
OCRI001264537] (INBC).

PARATYPES. COSTA RICA: Heredia: La Selva
Biological Station, 10.43°N, 84.02°W, 3$, ix.1992,
P. Hanson, Malaise trap (LACM, MUCR), 2$,
15.ii-l.iii.1993, ALAS, Malaise trap M/10/25
(INBC), 2$, 15.iv-l.v.l993, ALAS, Malaise trap
M/10/89 (INBC), 12?, 1-15.V.1993, ALAS, Mal-
aise trap M/10/104 (INBC), 17?, 15.v-l.vi.1993,
ALAS, Malaise trap M/10/116 (INBC, LACM),
11?, l-15.vl.1993, ALAS, Malaise trap M/10/132
(INBC, LACM), 1 ?, 15.vi-l.vii.1993, ALAS, Mal-
aise trap M/04/139 (INBC), 25?, 15.vi-l.vii.1993,
ALAS, Malaise trap M/10/144 (INBC, LACM),
12?, l-15.vii.1993, ALAS, Malaise trap M/10/160
(INBC, LACM), 17?, 15.vii-3.viii.1993, ALAS,

Brown: Revision of Apocephalus miricauda-gxonp ■ 41

Malaise trap M/10/172 (INBC, MCZC, USNM),
3$, 3.viii.l993, ALAS, Malaise trap M/12/174
(INBC), 14$, 3-14.viii.1993, ALAS, Malaise trap
M/10/188 (INBC), 11$, 15.viii-l.ix.1993, ALAS,
Malaise trap M/10/200 (INBC), 1$, l.xi.1993,
ALAS, Malaise trap M/02/249 (INBC), 1$,
2. i. 1996, ALAS, Malaise trap M/01/531 (INBC),
Rara Avis, 12 km SW Horquetas, 1$, 18-
234.1989, D.A. Grimaldi, 550 m (AMNH); Pun-
tarenas: 24 km W Piedras Blancas, 8.77°N, 83.4°W,
1$, i. 1992, P. Hanson, Malaise trap, 200 m
(LACM), 23 km N Puerto Jimenez, 8.67°N,
83.45°W, 1$, viii.1991, P. Hanson, Malaise trap
(LACM).

Apocephalus petiolus new species

(Fig. 78-79, 109)

SPECIES RECOGNITION. This species is rec-
ognized by the shape of the preapical sclerite and
by the densely setose, “hairy” appearance of the
venter of the abdomen.

DESCRIPTION. Body length 1.4 mm. Frons yel-
low, anterior margin relatively straight. One pair of
supra-antennal setae present. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax light brown; pleuron yellow. An-
terior scutellar seta small, fine, subequal to poste-
rior setulae of scutum. Legs yellow; apex of hind
femur with abrupt darkening on anterior face.
Mean costal length 0.55 wing length. Wing vein
R2+3 present. Halter brown. Abdominal tergites
dark brown. Venter of abdomen yellow. Abdominal
tergites of normal form. Tergite 3 evenly colored.
Tergite 6 complete, with short setae at posterolat-
eral corner. Venter of segments 3-5 with long,
dense setae. Abdomen without dense lateral setae.
Ventral setae of segment 6 long, consisting of a
complete ventral and lateral row. Ovipositor (Figs.
78-79) straight in lateral view, lightly but evenly
sclerotized dorsally. Lateral darkening posteriorly
enlarged, inner margin extended medially, margin
dark, complete. Dorsoapical sclerite of ovipositor
small, triangular. Apicodorsal margin of ovipositor
straight. Apicoventral margin of ovipositor
straight. Ovipositor without ventral postapical
sclerite. Sternite 7 broad, narrowed apically. Du-
four’s mechanism elongate. Abdominal glands in
segment 5 white, inconspicuous in cleared speci-
mens. Internal sclerite rounded, with long process
(Fig. 109).

GEOGRAPHICAL DISTRIBUTION. Known
from a few lowland sites in Costa Rica.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for small stem, referring to
the sclerotized loop.

HOLOTYPE. $ , COSTA RICA: Puntarenas: Rio
Piro, 8.28°N, 83.32°W, ii.1991, P. Hanson, Malaise
trap, 75 m [LACM ENT 004854] (LACM).

PARATYPES. COSTA RICA: Heredia: La Selva
Biological Station, 10.43°N, 84.02°W, 1$,

42 ■ Contributions in Science, Number 482

2.iii. 1993, ALAS, Malaise trap, M/05/20, 1$,
4.iv.l994, ALAS, Malaise trap, M/ll/389 (INBC);
Puntarenas: 24 km W Piedras Blancas, 8.77°N,
83.4°W, 1 $, xi.1990, P. Hanson, Malaise trap, 200
m (LACM).

Apocephalus gigantivorus new species

(Figs. 80-81)

SPECIES RECOGNITION. This species has a
distinctive preapical sclerite that is much longer
than broad.

DESCRIPTION. Body length 1.9-2. 3 mm. Frons
yellow, anterior margin relatively straight. One pair
of supra-antennal setae present. Flagellomere 1 ba-
sally yellow, apically brown, round. Proboscis nor-
mal, small. Palpus yellow. Dorsum of thorax yel-
low; pleuron white. Anterior scutellar seta large,
bristle-like. Legs yellowish-brown; apex of hind fe-
mur with abrupt darkening on anterior face. Mean
costal length 0.52 wing length; range 0.51-0.54.
Wing vein R2+3 present. Halter brown. Abdominal
tergites dark brown, lighter medially. Venter of ab-
domen yellow. Abdominal tergites of normal form.
Tergite 3 evenly colored. Tergite 6 completely di-
vided, with short setae at posterolateral corner.
Venter of segments 3-5 bare. Abdomen without
dense lateral setae. Ventral setae of segment 6 long,
consisting of a complete ventral and lateral row.
Ovipositor (Figs. 80-81) slightly downturned api-
cally, lightly but evenly sclerotized dorsally. Lateral
darkening posteriorly enlarged, inner margin ex-
tended posteriorly, margin dark, complete. Dor-
soapical sclerite of ovipositor elongate, rounded an-
teriorly. Apicodorsal margin of ovipositor straight.
Apicoventral margin of ovipositor straight. Ovi-
positor without ventral postapical sclerite. Sternite
7 triangular. Dufour’s mechanism elongate. Ab-
dominal glands in segment 5 white, inconspicuous
in cleared specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
only from Brazil, but the host ant is also known
from Peru (Kempf, 1971).

WAY OF LIFE. Females were attracted to injured
Dinoponera gigantea. They were mentioned by Sil-
veira-Costa and Moutinho (1996, p. 94) in their
paper about phorid parasitoids of Dinoponera.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the species name of the host and
the Latin word voro, for eat.

HOLOTYPE. $, BRAZIL: Paragominas: 3°S,
47.5°W, lO.v.1995, A. Silveira-Costa, over Dino-
ponera gigantea [LACM ENT 006557] (MZSP).

PARATYPES. BRAZIL: Paragominas: 3°S,
47.5°W, 1$, v.1994, 5$, lO.v.l 995, A. Silveira-
Costa, over Dinoponera gigantea (LACM, MZSP).

Apocephalus piliventris Borgmeier

(Fig. 82)

Apocephalus piliventris Borgmeier, 1925:186, figs.

18, 19, Plate VIII, fig. 39.

Lectotype (here designated). $, BRAZIL: Rio de
Brown: Revision of Apocephalus miricauda-gxoxxp

80. 81.

82. 83.

84.

Figures 80-90. Ovipositors. Figures 80-81. Apocephalus gigantivorus new species. 80. Dorsal. 81. Ventral. 82. Apoce-
pbalus piliventris Borgmeier, dorsal. Figures 83-84. Apocephalus annulatus new species. 83. Dorsal. 84. Ventral. Figures
85-86. Apocephalus contortiven.tr is new species. 85. Dorsal. 86. Ventral. Figures 87-88. Apocephalus eurydomus new
species. 87. Dorsal. 88. Ventral Figures 89-90. Apocephalus conformalis new species. 89. Dorsal. 90. Ventral.

Janeiro, Petropolis, 22.52°S, 43.17°W [LACM ENT
121126] (MZSP).

SPECIES RECOGNITION. This species is most
easily recognized by the thin, divergent lateral hark-
enings of the ovipositor.

DESCRIPTION. Body length 1.4-1. 6 mm. Frons
yellow, anterior margin relatively straight. One to
two pairs of supra-antennal setae present; lower
pair markedly smaller than upper pair. Fiageilo-
mere 1 brown, round. Proboscis normal, small. Pal-
pus yellow. Dorsum of thorax light brown; pleuron
yellow. Anterior scutellar seta small, fine, subequal
to posterior setulae of scutum. Legs yellow; apex
of hind femur with abrupt darkening on anterior
face. Mean costal length 0.51 wing length; range
0.50-0.52. Wing vein R2+3 present. Halter brown.

Contributions in Science, Number 482

Abdominal tergites dark brown. Venter of abdo-
men yellow. Abdominal tergites of normal form.
Tergite 3 evenly colored. Tergite 6 completely di-
vided, with short setae at posterolateral comer.
Venter of segments 3-5 with long, dense setae con-
centrated medially. Abdomen without dense lateral
setae. Ventral setae of segment 6 long, consisting of
a complete ventral and lateral row. Ovipositor (Fig.
82) straight in lateral view, lightly and evenly scler-
otized dorsally with thin, dark, median line. Lateral
darkening thin, apically divergent, margin dark,
complete. Dorsoapical sclerite of ovipositor heart-
shaped. Apicodorsal margin of ovipositor straight.
Apicoventral margin of ovipositor straight. Ovi-
positor without ventral postapical sclerite. Sternite
7 triangular. Abdominal glands in segment 5 white,

Brown: Revision of Apocephalus miricauda-gtowp ■ 43

inconspicuous in cleared specimens. Internal scler-
ite with small opening and broad, moderately scler-
otized process.

GEOGRAPHICAL DISTRIBUTION. Known
from a few sites in southeastern Brazil.

WAY OF LIFE. Females were attracted to injured
workers of Pachycondyla striata Smith in the urban
park next to the Museu de Zoologia, Sao Paulo,
Brazil

MATERIAL EXAMINED. BRAZIL: Minas Ger-
ais: Congonhas, 19, 22-24.ii.1990, S.A. Marshall,
pan trap (DEBU); Rio de Janeiro: Petropolis, 19,
2.iii.l923, 19, 7.iv.l923, Id, 24.iv.1923, 19,
2.V.1923, Ronchi (MCZC, USNM), Id, 3 9, no
other data (LACM, MZSP); Santa Catarina: Nova
Teutonia, 1 9 [no date], F. Plaumann (MZSP); Sao
Paulo: Parque do Ipiranga, 23.59°S, 46.61°W, 29,
29.iv.1999, 69, l.v.1999, B. Brown, injured Pa-
chycondyla striata (LACM, MZSP).

Apocephalus annulatus new species

(Figs. 83-84)

SPECIES RECOGNITION. This species is rec-
ognized by the posterior cleft in the dorsal preapical
sclerite and by the rounded lateral darkenings.

DESCRIPTION. Body length 1.5-1. 8 mm. Frons
yellow, anterior margin relatively straight. One pair
of supra-antennal setae present. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax light brown; pleuron white. An-
terior scutellar seta small, fine, subequal to poste-
rior setulae of scutum. Legs yellowish-brown; apex
of hind femur with abrupt darkening on anterior
face. Mean costal length 0.47 wing length; range 0.
Wing vein R2+3 present. Halter brown. Abdominal
tergites yellow, posteriorly dark brown; tergite 6
completely yellow. Venter of abdomen yellow. Ab-
dominal tergites of normal form. Tergite 3 evenly
colored. Tergite 6 completely divided, with short
setae at posterolateral corner. Venter of segments
3-5 with short setae, concentrated medially. Ab-
domen without dense lateral setae. Ventral setae of
segment 6 moderately long, consisting of a com-
plete ventral and lateral row. Ovipositor (Figs. 83-
84) straight in lateral view, lightly but evenly scler-
otized dorsally. Lateral darkening posteriorly en-
larged, inner margin extended medially, margin
dark, complete. Dorsoapical sclerite of ovipositor
rectangular, posteriorly cleft. Apicodorsal margin
of ovipositor straight. Apicoventral margin of ovi-
positor straight. Ovipositor without ventral postap-
ical sclerite. Sternite 7 triangular, apically tridentate
and with apical, triangular sclerite. Abdominal
glands in segment 5 white, inconspicuous in cleared
specimens. Internal sclerite rounded, with a short
process.

GEOGRAPHICAL DISTRIBUTION. Known
from two sites in eastern Costa Rica.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The

44 ■ Contributions in Science, Number 482

name is a Latin word for circular, referring to the
rounded lateral darkenings of this species.

HOLOTYPE. 9, COSTA RICA: Limon: 16 km
W Guapiles, 10.15°N, 83.92°W, i-iv.1991, P. Han-
son, Malaise trap, 400 m [LACM ENT 013196]
(LACM).

PARATYPES. COSTA RICA: Limon: 4 km NE
Bribri, 9.63°N, 82.82°W, 19, vii-ix.1990, P. Han-
son, Malaise trap, 50 m (LACM).

Apocephalus contortiventris new species

(Figs. 85-86)

SPECIES RECOGNITION. This species can be
recognized by the extremely complex shape of ster-
nite 7.

DESCRIPTION. Body length 1.5 mm. Frons
light brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 light brown, round. Proboscis normal,
small. Palpus brown. Dorsum of thorax light
brown; pleuron light brown. Anterior scutellar seta
small, fine, subequal to posterior setulae of scutum.
Legs yellowish-brown; apex of hind femur with
abrupt darkening on anterior face. Mean costal
length 0.5 wing length. Wing vein R2+3 present.
Halter brown. Abdominal tergites dark brown.
Venter of abdomen gray. Abdominal tergites of nor-
mal form. Tergite 3 evenly colored. Tergite 6 com-
pletely divided, with several long setae on posterior
margin. Venter of segments 3-5 with long, dense
setae concentrated medially. Abdomen without
dense lateral setae. Ventral setae of segment 6 long,
consisting of a complete ventral and lateral row.
Ovipositor (Figs. 85-86) straight in lateral view,
lightly but evenly sclerotized dorsally. Lateral dark-
ening posteriorly enlarged, inner margin extended
medially, margin dark, complete. Dorsoapical scler-
ite of ovipositor rectangular. Apicodorsal margin of
ovipositor straight. Apicoventral margin of ovipos-
itor straight. Ovipositor without ventral postapical
sclerite. Sternite 7 anteriorly triangular, posteriorly
narrowed, then broadly expanded to an oval with
numerous lateral setae, then narrowed and expand-
ed to broad posterior apex. Abdominal glands in
segment 5 white, inconspicuous in cleared speci-
mens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in western Ecuador.

WAY OF LIFE. The single specimen was attract-
ed to an injured worker of Pachycondyla impressa.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin words contortus and
venter, for complex and belly, referring to the com-
plicated structure of sternite 7.

HOLOTYPE. 9, ECUADOR: Esmeraldas: Bilsa
Biological Station, 0.34°N, 79.71°W, lO.v.l 996, B.
Brown, injured Pachycondyla impressa, 500 m
[LACM ENT 053897] (LACM).

Brown: Revision of Apocephalus miricauda-group

Apocepbalus eurydomus new species

(Figs. 87-88, 106, 111)

Apocephalus sp., Brown 1992, fig. 35B-D.

SPECIES RECOGNITION. This species is best
recognized by the medial ventral setae of the ab-
domen, the apically expanded sternite 7 and the
small preapical sclerite. The most similar species is
A. conformalis new species, which differs in the
form of sternite 7.

DESCRIPTION. Body length 1.3-1. 6 mm. Frons
yellow to light brown, anterior margin relatively
straight. One pair of supra-antennal setae present.
Flagellomere 1 yellow, round. Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron yellow to light brown. Anterior
scutellar seta slightly enlarged, bristle-like. Legs yel-
lowish-brown; apex of hind femur with abrupt
darkening on anterior face. Mean costal length
0.49 wing length; range 0.47-0.5. Wing vein R2+3
present. Flalter brown. Abdominal tergites dark
brown. Venter of abdomen yellow. Abdominal ter-
gites of normal form. Tergite 3 evenly colored. Ter-
gite 6 completely divided, with short setae at pos-
terolateral corner. Venter of segments 3-5 with
short setae, concentrated medially (Fig. 111). Ab-
domen without dense lateral setae. Ventral setae of
segment 6 long, consisting of a complete ventral
and lateral row. Ovipositor (Figs. 87-88) slightly
downturned apically, lightly but evenly sclerotized
dorsally. Lateral darkening posteriorly enlarged,
truncate, margin dark, complete. Dorsoapical scler-
ite of ovipositor trapezoidal. Apicodorsal margin of
ovipositor straight. Apicoventral margin of ovipos-
itor straight. Ovipositor without ventral postapical
sclerite. Sternite 7 narrow but apically expanded.
Abdominal glands in segment 5 white, inconspicu-
ous in cleared specimens. Internal sclerite with
small opening and broad, moderately sclerotized
process (Fig. 106).

GEOGRAPHICAL DISTRIBUTION. Known
from the southern USA to Panama.

WAY OF LIFE. Presumably, this species is a par-
asitoid of Pachycondyla harpax (Fabricius). Fe-
males were attracted to a chemical extract of work-
ers of this ant in Texas.

DERIVATION OF SPECIFIC EPITHET. The

name is based on the Greek words eurys, for wide-
spread, and doma for home. It refers to the large
geographic range of this species, the only member
of the A. miricauda- group to occur in the Nearctic
Region.

HOLOTYPE. 9, USA: Texas: Travis County,
Austin 30.3°N, 97.78°W, 9.vii.l994, D. Feener, at-
tracted to Pachycondyla harpax extract [LACM
ENT 031026] (LACM).

PARATYPES. COSTA RICA: Cartago: Duke
Nombre, 9.83°N, 83.92°W, 2$, vi-viii.1993, P.
Hanson, Malaise trap, 1400 m (LACM); San Jose:
Ciudad Colon, 9.92°N, 84.25°W, 3 9, h.1990, 7$,
iii— iv. 1990, 29, iv-v.1990, P. Hanson, Malaise
trap, 800 m (INBC, LACM, MUCR). PANAMA:

Contributions in Science, Number 482

Canal Zone: Balboa, Ancon Hill, 29, 26-
29. iv. 1983, G. Otis, pan trap (LACM). USA: Texas:
Bastrop County, Bastrop State Park, 30.12°N,
97.35°W, 19, l-12.iii.1991, R. Wharton, Malaise
trap (TAMU), Travis County, Austin, 30.3°N,
97.78°W, 19, 9.vii.l994, D. Feener, attracted to
Pachycondyla harpax extract (LACM).

Apocephalus conformalis new species

(Figs. 89-90)

SPECIES RECOGNITION. This species is simi-
lar to A. eurydomus but has a different form of
sternite 7.

DESCRIPTION. Body length 1.4-1. 6 mm. Frons
light brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 light brown, round. Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron white to light brown. Anterior scu-
tellar seta small, fine, subequal to posterior setulae
of scutum. Legs yellowish-brown; apex of hind fe-
mur with abrupt darkening on anterior face. Mean
costal length 0.47 wing length; range 0.45-0.48.
Wing vein R2+3 present. Halter light brown. Ab-
dominal tergites dark brown. Venter of abdomen
yellow. Abdominal tergites of normal form. Tergite

6 completely divided, with short setae at postero-
lateral corner. Venter of segments 3-5 with short
setae, concentrated medially. Abdomen without
dense lateral setae. Ventral setae of segment 6 long,
consisting of a complete ventral and lateral row.
Ovipositor (Figs. 89-90) straight in lateral view,
lightly but evenly sclerotized dorsally. Lateral dark-
ening posteriorly enlarged, truncate, margin dark,
complete. Dorsoapical sclerite of ovipositor trape-
zoidal. Apicodorsal margin of ovipositor straight.
Apicoventral margin of ovipositor straight. Ovi-
positor without ventral postapical sclerite. Sternite

7 broad, narrowed apically but widened abruptly
at apex. Dufour’s mechanism elongate. Abdominal
glands in segment 5 white, inconspicuous in cleared
specimens. Internal sclerite with small opening and
broad, moderately sclerotized process.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in Amazonian Brazil.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for similar, referring to the
close similarity of this species to A. eurydomus.

HOLOTYPE. 9, BRAZIL: Roraima: Ilha de
Maraca, 3.37°N, 61.43°W, 2-13.V.1987, J. Rafael,
Malaise trap [LACM ENT 012754] (INPA).

PARATYPES. 30 9, same data as holotype
(INPA, LACM, MCZC, MZSP, USNM).

Apocephalus fenestratus new species

(Fig. 91)

SPECIES RECOGNITION. This species is in-
stantly recognizable by the peculiar clear sections
in the ovipositor.

DESCRIPTION. Body length 1.3 mm. Frons
Brown: Revision of Apocephalus miricauda-group M 45

95.

Figures 91-101. Ovipositors. 91. Apocephalus fenestratus new species, dorsal. 92. Apocephalus asyndetus new species,
dorsal. 93. Apocephalus catholicus new species, dorsal. 94. Apocephalus lyratus Borgmeier, dorsal. 95. Apocephalus
trifidus new species, dorsal. 96. Apocephalus tanyurus new species, dorsal. Figures 97-98. Apocephalus contracticauda
new species. 97. Dorsal. 98. Ventral. 99. Apocephalus indistinctus new species, dorsal. Figures 100-101. Apocephalus
dinoponerae new species. 100. Dorsal. 101. Ventral.

light brown, anterior margin relatively straight.
One pair of supra-antennal setae present. Flagel-
lomere 1 light brown, oval. Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron light brown. Anterior scutellar seta
small, fine, subequal to posterior setulae of scutum.
Legs yellowish-brown; apex of hind femur with
abrupt darkening on anterior face. Mean costal
length 0.48 wing length. Wing vein R2+3 present.
Halter brown. Abdominal tergites dark brown.
Venter of abdomen white. Abdominal tergites of
normal form. Tergite 3 evenly colored. Tergite 6
completely divided, with several long setae on pos-
terior margin. Venter of segments 3-5 with a few,

46 ■ Contributions in Science, Number 482

scattered setae. Abdomen without dense lateral se-
tae. Ventral setae of segment 6 long, consisting of
a complete ventral and lateral row. Ovipositor (Fig.
91) slightly upturned apically, evenly sclerotized,
but with a pair of clear areas. Lateral darkening
not differentiated. Dorsoapical sclerite of oviposi-
tor not differentiated. Apicodorsal margin of ovi-
positor straight. Apicoventral margin of ovipositor
straight. Ovipositor without ventral postapical
sclerite. Sternite 7 narrow but apically expanded.
Abdominal glands in segment 5 white, inconspicu-
ous in cleared specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from a single mid-elevation site in Costa Rica.

Brown: Revision of Apocephalus miricauda-group

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The

name is from a Latin word, fenestra, for window.

HOLOTYPE. 9, COSTA RICA: Cartago: La
Cangreja, 9.8°N, 83.97°W, iii— v. 1992, P. Hanson,
Malaise trap, 1950 m [LACM ENT 013053]
(LACM).

PARATYPE. COSTA RICA: Cartago: La Can-
greja, 9.8°N, 83.97°W, 1 9, vi-vii.1992, P. Hanson,
Malaise trap, 1950 m (LACM).

Apocephalus asyndetus new species

(Fig. 92)

SPECIES RECOGNITION. This widespread spe-
cies is easily recognized by the distinctive oviposi-
tor. Specimens from South America are consistently
lighter in color, especially in flagellomere 1, but do
not differ from Central American specimens in any
other substantial way.

DESCRIPTION. Body length 1.4-1 .9 mm. Frons
yellow to dark brown, anterior margin relatively
straight. Two pairs of supra-antennal setae present;
lower pair extremely small. Flagellomere 1 light
brown to brown, oval, enlarged. Proboscis normal,
small. Palpus brown. Dorsum of thorax yellow to
light brown; pleuron yellow to white. Anterior scu-
tellar seta small, fine, subequal to posterior setulae
of scutum. Legs yellowish-brown; apex of hind fe-
mur with abrupt darkening on anterior face. Mean
costal length 0.49 wing length; range 0.45-0.52.
Wing vein R2+3 present. Halter brown. Abdominal
tergites dark brown to yellow, posteriorly dark
brown; tergite 6 completely yellow. Venter of ab-
domen yellow. Abdominal tergites of normal form.
Tergite 3 evenly colored. Tergite 6 completely di-
vided, with long seta at posterolateral corner. Ven-
ter of segments 3-5 with short setae, concentrated
medially. Abdomen without dense lateral setae.
Ventral setae of segment 6 long, consisting of a
complete lateral row and a medial group centered
on a small sclerite. Ovipositor (Fig. 92) straight in
lateral view, with broad median sclerite. Lateral
darkening thin, extended posteriorly from apex of
ovipositor, margin dark, complete. Dorsoapical
sclerite of ovipositor not differentiated. Apicodor-
sal margin of ovipositor straight. Apicoventral mar-
gin of ovipositor straight. Ovipositor without ven-
tral postapical sclerite. Sternite 7 broad, lightly
sclerotized. Abdominal glands in segment 5 white,
inconspicuous in cleared specimens. Internal scler-
ite not seen.

GEOGRAPHICAL DISTRIBUTION. Lowland
Brazil, Costa Rica, Ecuador, Panama, and Peru.
The host ant, Gnamptogenys bispinosa (Emery), is
known from Costa Rica and Colombia (Lattke,
1995).

WAY OF LIFE. One specimen was collected at-
tacking workers of Gnamptogenys bispinosa as
they were being raided by the army ant Eciton va-
gans (Olivier) (Brown and Feener, 1998).

DERIVATION OF SPECIFIC EPITHET. The

Contributions in Science, Number 482

name is derived from the Greek asyndetos for un-
connected, referring to the freely ending lateral dar-
kenings.

HOLOTYPE. 9, COSTA RICA: Heredia: La
Selva Biological Station, 10.43°N, 84.02°W, 6-
ll.vii.1993, B. Brown, D. Feener, Malaise trap #1
[LACM ENT 003497] (LACM).

PARATYPES. BRAZIL: Para: Tucurui, 3.83°S,
49.67°W, 19, 20.vii-8.viii.1982, J. Vidal, CDC
trap (INPA); Roraima: Ilha de Maraca, 3.37°N,
61.43°W, 3 9, 2-13.V.1987, J. Rafael, Malaise trap
(INPA, LACM). COSTA RICA: Guanacaste: 14 km
S Canas, 19, 24-31.V.1990, F.D. Parker (EMUS),
Estacion Pitilla, 11.0°N, 85.43°W, 19, iv.1989, P.
Hanson, Malaise trap, 200 m (LACM), Santa Rosa
National Park, 10.95°N, 85.62°W, 19, 21. ii-
14.iii.1987, 3 9, 14.iii-4.iv.1987, I. Gauld and D.
Janzen, Malaise trap, 300 m (LACM); Heredia: La
Selva Biological Station, 10.43°N, 84.02°W, 19,
17.V.1988, B. Brown, over Gnamptogenys bispi-
nosa raided by Eciton vagans (LACM' 29, 21. i-
3.ii.l991, J. Noyes, Malaise trap (LACM), 3 9,
15.iv.1993, ALAS, Malaise trap, M/01/64 (INBC),
19, l-15.iv.1993, ALAS, Malaise trap, M/ll/74
(INBC), 29, 15.iv-l.v.l993, ALAS, Malaise trap,
M/8/87 (INBC), 169, 1-15.V.1993, ALAS, Malaise
trap, M/04/83, M/8/102 (INBC), 29, 1-
15.vi.1993, ALAS, Malaise trap, M/8/130 (INBC),
19, 6-ll.vii.1993, B. Brown, D. Feener, Malaise
trap #1 (LACM), 8 9, 15.X.1993, ALAS, Malaise
trap M/09/243 (INBC), 19, 3.L1994, ALAS, Mal-
aise trap M/03/306 (INBC), 29, 4.iv.l994, ALAS,
Malaise trap M/01/384, M/09/387 (INBC), 19,
2. i. 1996, ALAS, Malaise trap M/01/531 (INBC),
19, 4.ix.l997, ALAS, Malaise trap M/17/679
(INBC), 19, 30.X.1997, ALAS, M/17/698 (INBC);
Limon: 4 km NE Bribn, 9.63°N, 82.82°W, 1 9 , vii-
ix.1990, P. Hanson, Malaise trap, 50 m (LACM),
Pandora, Estrella Valley, 19, 28. iii. 1984, G.V.
Manley, Malaise trap (LACM); Puntarenas: Cerro
Rincon, 8.52°N, 83.47°W, 3 9, i.1991, 1 9, ii.1991,
19, iii.1991, P. Hanson, Malaise trap, 745 m
(LACM), 24 km W Piedras Blancas, 8.77°N,
83.4°W, 19, x.1990, P. Hanson, Malaise trap, 200
m (LACM), 3 km SW Rincon, 8.68°N, 83.48°W,
2 9 , ix-xi.1989, 1 9 , x.1991, 8 9 , xi. 1991, P. Han-
son, Malaise trap, 10 m (LACM, MUCR). ECUA-
DOR: Pichincha, E. Santo Domingo [Tinalandia],
19, 8-14.V.1988, Bohart and Hanson (EMUS);
Sucumbios: Sacha Lodge, 0.5°S, 76.5°W, 19,4-
14. iii. 1994, 29, 24.v-3.vi.1994, 19, 27.viii-
10. ix. 1994, 29, 1 0-21. x. 1994, 19, 21.xi-

I. xii.1994, P. Hibbs, Malaise trap, 270 m (LACM,
QCAZ). PANAMA: Darien: Cruce de Mono,
7.92°N, 77.62°W, 19, 6.ii-4.iii.l993, R. Cambra,
J.Coronado, Malaise trap (LACM). PERU: Madre
de Dios: Manu National Park, Cocha Cashu Sta-
tion, 19, 23-30.viii.1986, D.C. Darling, Malaise
trap, 380 m (ROME), Zona Reserva Pakitza,

II. 94°S, 71.28°W, 19, 18.ii.1992, B. Brown, D.
Feener, Malaise trap #4 (LACM) 49, 13-
18.ii.1992, D. Quintero, Malaise trap (MIUP,

Brown: Revision of Apocephalus miricauda-group ■ 47

USNM), 19, 855.1992, B. Brown, D. Feener,
blacklight trap (LACM). TRINIDAD: Asa Wright
Nature Center, 1$, 15. i. 1981, G.E. Bohart
(EMUS).

Apocephalus catbolicus new species

(Fig. 93)

SPECIES RECOGNITION. This species is easily
recognized by the apicodorsal region of the ovipos-
itor, which has relatively broad lateral darkenings
and medial sclerotization that expands posteriorly.

DESCRIPTION. Body length 1.1-1. 6 mm. Frons
yellow to dark brown, anterior margin relatively
straight. One to two pairs of supra-antennal setae
present; lower pair markedly smaller than upper
pair. Flagellomere 1 light brown to brown, round.
Proboscis normal, small. Palpus yellow to brown.
Dorsum of thorax light brown; pleuron white. An-
terior scutellar seta slightly enlarged, bristle-like.
Legs yellowish-brown; apex of hind femur with
abrupt darkening on anterior face. Mean costal
length 0.57 wing length; range 0.55-0.6. Wing vein
R2+3 present. Halter brown. Abdominal tergites
dark brown. Venter of abdomen yellow to gray.
Abdominal tergites of normal form. Tergite 3 even-
ly colored. Tergite 6 completely divided, with short
setae at posterolateral corner. Venter of segments
3-5 with a few scattered setae. Abdomen without
dense lateral setae. Ventral setae of segment 6 long,
consisting of a complete ventral and lateral row.
Ovipositor (Fig. 93) straight in lateral view, lightly
but evenly sclerotized dorsally. Lateral darkening
broad, margin dark, complete. Dorsoapical sclerite
of ovipositor not differentiated. Apicodorsal mar-
gin of ovipositor wedge-shaped, posteriorly wid-
ened. Apicoventral margin of ovipositor straight.
Ovipositor without ventral postapical sclerite. Ster-
nite 7 wedge-shaped, narrowed toward apex but
abruptly widened at apex. Dufour’s mechanism
round. Abdominal glands in segment 5 white, in-
conspicuous in cleared specimens. Internal sclerite
a simple, round loop.

GEOGRAPHICAL DISTRIBUTION. This spe-
cies is known from lowland Brazil, Colombia, Cos-
ta Rica, Ecuador, and Panama.

WAY OF LIFE. Females of this species have been
attracted to injured ants of a number of species,
including Ectatomma goninion Kugler and Brown,
Odontomachus bauri, O. chelifer (Latreille), O.
hastatus (Fabricius), O. laticeps, Pachycondyla
crassinoda, P. harpax, and P. impressa. All of the
specimens collected at Bilsa Biological Station in
Ecuador were feeders (see Behavioral Aspects).

At Bilsa, we conducted preference trials using Pa-
chycondyla impressa and Odontomachus bauri.
One bait with three injured P. impressa and one
with three injured O. bauri was offered simulta-
neously. The number of flies landing on each bait
was recorded for twelve 15-minute periods. Ac-
cording to these trials, the flies were much more
attracted to injured P. impressa (x = 2.25 flies/trial)

48 ■ Contributions in Science, Number 482

than to injured O. bauri (x = 0.33 flies/trial); the
difference was highly significant (99%; T = —4.60,
p = 0.0003, df = 16).

DERIVATION OF SPECIFIC EPITHET. The

name is a Latin word for universal or general, re-
ferring to the wide variety of ants that attracted
females of this species.

HOLOTYPE. 9, COSTA RICA: Limon: 7 km
SW Bribri, 9.58°N, 82.88°W, ix-xi.1989, P. Han-
son, Malaise trap [LACM ENT 005282] (LACM).

PARATYPES. BRAZIL: Amazonas: Manaus, Re-
serva Ducke, 3.13°S, 60.02°W, 19, 8-15.iv.1992,
Arm-Cola, 1-B-l m, J. Vidal (INPA). COLOMBIA:
Amazonas: 22 km NW Leticia, 0.4°S, 69. 99°W,
19, 28.viii.1997, B. Brown, G. Kung, injured Pa-
chycondyla crassinoda (LACM); Valle: Rio Rapo-
so, 1 9, v.1965, V.H. Lee, light trap (USNM). COS-
TA RICA: Alajuela: La Virgen del Socorro, near
Carriblanca, 19, 16. ii. 1989, D. Grimaldi, P. De-
Vries, 700 m (AMNH); Guanacaste: Estacion Pitil-
la, 1TN, 85.43°W, 29, 4.vii.l997, B. Brown, J.
Paldi, injured Odontomachus chelifer (LACM);
Heredia: La Selva Biological Station, 10.43°N,
84.02°W, 1 9, 15.iii-l.iv.1993, ALAS, Malaise trap
M/10/57 (INBC), 1 9 , 4.vii.l993, B. Brown, injured
Pachycondyla harpax (LACM), 19, 15. i. 1994,
ALAS, Malaise trap M/08/326 (INBC), 19,
25.1996, ALAS, Malaise trap M/01/531 (INBC),
Rara Avis, 12 km SW Horquetas, 19, 18-

235.1989, D.A. Grimaldi, 550 m (AMNH); Limon:
4 km NE Bribri, 9.63°N, 82.82°W, 79, xii.1989-

111.1990, 1 9, vii-ix.1990, P. Hanson, Malaise trap,
50 m (LACM), 7 km SW Bribri, 9.58°N, 82.88°W,
29, ix-xi.1989, P. Hanson, Malaise trap (LACM),
16 km W Guapiles, 10.15°N, 83.92°W, 19, viii-
ix.1989, 39, iii— v. 1990, 29, i-iv.1991, P. Hanson,
Malaise trap, 400 m (LACM); Puntarenas: Cerro
Rincon, 8.52°N, 83.47°W, 19, ii.1991, P. Hanson,
Malaise trap, 745 m (LACM), 3 km SW Rincon,
8.68°N, 83.48°W, 29, iii.1989, 19, ix-xi.1989,
19, xii.1991, P. Hanson, Malaise trap, 10 m
(LACM, MUCR), 5 km SW Rincon, 8.7°N,
83.5TW, 89, 3.vi.l998, 169, 6.vi.l998, B.
Brown, injured Pachycondyla impressa (LACM),
49, 3.vi.l998, B. Brown, injured Odontomachus
laticeps (LACM), 2 9, 4.vi.l998, 29, 5.v i.1998,
49, 6.vi.l998, B. Brown, injured Odontomachus
bauri (LACM). ECUADOR: Esmeraldas: Bilsa Bi-
ological Station, 0.34°N, 79.71°W, 19, 8.V.1996,
B. Brown, injured Ectatomma goninion (LACM),
3c3, 209, 8. v. 1996, B. Brown, injured Pachycon-
dyla impressa (LACM, QCAZ, UNCB), 20 9,
8.v.l 996, B. Brown, injured Odontomachus bauri
(LACM, QCAZ), 19, 8.V.1996, B. Brown, injured
Odontomachus hastatus (LACM), 1 9 , 9.V.1996, B.
Brown, injured Odontomachus bauri (LACM), Id,
199, lO.v.1996, B. Brown, injured Pachycondyla
impressa , 500 m (LACM, MCZC, QCAZ), 29,
lO.v.1996, B. Brown, injured Odontomachus bauri
(LACM); Napo: Yasuni Biological Research Sta-
tion, 0.67°S, 76.36°W, 19, 24.V.1996, B. Brown,
injured Pachycondyla crassinoda (LACM); Pichin-

Brown: Revision of Apocephalus miricauda-gxoxxp

cha: Maquipucuna Biological Reserve, 0.12°N,
78.63°W, 1$, 3.V.1996, B. Brown, injured Pachy-
condyla impressa (LACM), 47 km S Santo Domin-
go, Rio Palenque Science Center, 1?, 2--4.V.1987,
B. Brown, L.Coote, FIT, 180 m, primary rain forest
(LACM), 17 km E Santo Domingo, Tinalandia,
3 9, ll.v.1987, B. Brown, injured Pachycondyla
impressa (LACM). PANAMA: Darien: Cruce de
Mono, 7.92°N, 77.62°W, 19, 6.ii-4.iii.l993, R.
Cambra, J. Coronado, Malaise trap (MIUP).

Apocephalus lyratus Borgmeier
(Fig. 94)

Apocephalus lyratus Borgmeier, 1971:100; fig. 137.

HOLOTYPE. 9, BRAZIL: Santa Catarina:
Nova Teutonia, F. Plaumann [LACM ENT 122350]
(MZSP; examined).

SPECIES RECOGNITION. This species is rec-
ognized by the broad, flat ovipositor with thin,
light-colored lateral darkenings and a narrow scler-
ite between them.

DESCRIPTION. Borgmeier (1971) lists four
male paratypes, but the manner in which they were
associated with the females is unknown. Therefore,
I am skeptical that they belong in the same species.

Body length 1.2-1. 4 mm. Frons yellow, anterior
margin relatively straight. Two pairs of supra-an-
tennal setae present; lower pair markedly smaller
than upper pair. Flagellomere 1 yellow, round. Pro-
boscis normal, small. Palpus yellow. Dorsum of
thorax yellow; pleuron yellow. Anterior scutellar
seta small, fine, subequal to posterior setulae of scu-
tum. Legs yellowish-brown; apex of hind femur
with abrupt darkening on anterior face. Mean cos-
tal length 0.47 wing length; range 0.46-0.5. Wing
vein R2+3 present. Halter brown. Abdominal ter-
gites yellow, posteriorly dark brown; tergite 6 com-
pletely yellow. Venter of abdomen yellow. Abdom-
inal tergites of normal form. Tergite 3 evenly col-
ored. Tergite 6 completely divided, with several
long setae on posterior margin. Venter of segments
3-5 with long, dense setae concentrated medially.
Abdomen without dense lateral setae. Ventral setae
of segment 6 long, consisting of several setae in a
straight row. Ovipositor (Fig. 94) straight in lateral
view, evenly sclerotized, with long, thin, postero-
medial process dorsally. Lateral darkening thin,
subparallel, apically divergent, margin dark, com-
plete. Dorsoapical sclerite of ovipositor not differ-
entiated. Apicodorsal margin of ovipositor con-
cave. Apicoventral margin of ovipositor concave.
Ovipositor without ventral postapical sclerite. Ster-
nite 7 anteriorly rounded, dark, posteriorly narrow.
Abdominal glands in segment 5 white, inconspicu-
ous in cleared specimens. Internal sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from Brazil and Ecuador.

WAY OF LIFE. Unknown.

MATERIAL EXAMINED. BRAZIL: Amazonas:
Manaus, Reserva Ducke, 3.13°S, 60.02°W, 3 9,8-
15.iv.1992, Arm-Cola, 1-B-l m, 29, 6-17.vii.1992,

Contributions in Science, Number 482

10 m, J. Vidal (INPA, LACM); Santa Catarina:
Nova Teutonia, 19, F. Plaumann (MZSP). ECUA-
DOR: Sucumbios: Sacha Lodge, 0.5°S, 76.5° W, 19,
12-22. ii. 1994, 19, 4-14.iii.1994, 19, 23.vi-
3.vii.l994, 29, 16-27.viii.1994, 19, 27.viii-
10.ix.1994, 29, 31.x-10.xi.1994, P. Hibbs, Mal-
aise trap, 270 m (LACM, QCAZ).

Apocephalus trifidus new species

(Fig. 95)

SPECIES RECOGNITION. The distinctive ovi-
positor makes this species instantly recognizable:
the median projection of the ovipositor is subequal
in length and breadth to the lateral, sclerotized
margins, making the apex of the ovipositor appear
three-pronged.

DESCRIPTION. Body length 1.3-2. 5 mm. Frons
yellow, anterior margin relatively straight. One pair
of supra-antennal setae present. Flagellomere 1 yel-
low, slightly pyriform (pointed). Proboscis normal,
small. Palpus yellow. Dorsum of thorax light
brown; pleuron white. Anterior scutellar seta small,
fine, subequal to posterior setulae of scutum. Legs
yellowish-brown; apex of hind femur with abrupt
darkening on anterior face. Mean costal length
0.56 wing length; range 0.53-0.59. Wing vein R2+3
present. Halter brown. Abdominal tergites dark
brown. Venter of abdomen gray to white. Abdom-
inal tergites of normal form. Tergite 3 evenly col-
ored. Tergite 6 completely divided, with short setae
at posterolateral corner. Venter of segments 3-5
bare. Abdomen without dense lateral setae. Ventral
setae of segment 6 long, consisting of a complete
ventral and lateral row. Ovipositor (Fig. 95)
straight in lateral view, evenly sclerotized, with
long, thin posteromedial process dorsally. Lateral
darkening thin, subparallel, margin dark, complete.
Dorsoapical sclerite of ovipositor not differentiat-
ed. Apicodorsal margin of ovipositor straight. Ap-
icoventral margin of ovipositor straight. Ovipositor
without ventral postapical sclerite. Sternite 7 thin.
Dufour’s mechanism round. Abdominal glands in
segment 5 white, inconspicuous in cleared speci-
mens. Internal sclerite with small opening and
broad, moderately sclerotized process.

GEOGRAPHICAL DISTRIBUTION. This spe-
cies is known from Brazil, Colombia, Ecuador,
Guyana, and Peru. The host ant also has been re-
corded from French Guiana, Trinidad, and Vene-
zuela.

WAY OF LIFE. Females of this species are at-
tracted to, and oviposit in, injured Pachycondyla
crassinoda workers. An average of 3.67 eggs were
laid per host (n = 3 oviposition events, sd = 1.16
eggs), and larval feeding averaged 4.21 days (n =
14 larvae, sd = 1.05 days).

DERIVATION OF SPECIFIC EPITHET. The
name is a Latin word for trifurcated, referring to
the structure of the ovipositor.

HOLOTYPE. 9, PERU: Madre de Dios: Zona
Reserva Manu, Pakitza, 27.ii.1992, B. Brown, D.

Brown: Revision of Apocephalus miricauda- group ■ 49

Feener, injured Pachycondyla crassinoda [LACM
ENT 012199] (MUSM).

PARATYPES. BRAZIL: Para: Belem, 1$,
ix.1970, T.H.G. Aitken, sticky trap (USNM); Ro-
raima: Ilha de Maraca, 3.37°N, 61.43°W, 29,2-
13.V.1987, J. Rafael, Malaise trap (INPA, LACM).
COLOMBIA: Amazonas: 22 km NW Leticia,
4.04°S, 69.99°W, 1 9 , 26.viii.1997, 11 9, 27.viii.1997,
279, 28.viii.1997, 19, 6.ix.l997, B. Brown, G.
Kung, injured Pachycondyla crassinoda (LACM,
MCZC, MZSP, UNCB, USNM). ECUADOR:
Napo: Yasuni Biological Research Station, 0.67°S,
76.36°W, 3 9, 21.V.1996, B. Brown, injured Pachy-
condyla crassinoda (LACM, QCAZ); Sucumbios:
Anagu, 0.48°S, 76.38°W, 3 9, 9.ix.l997, P.J. De-
Vries, injured Pachycondyla crassinoda (LACM,
QCAZ). GUYANA: Berbice: Dubulay Ranch,
5.68°N, 57.86°W, 3 9 , 23.L1999, B. Brown, injured
Pachycondyla crassinoda (LACM), Warniabo
Creek, Dubulay Ranch, 5.66°N, 57.88°W, 109,
16.i.l999, B. Brown, injured Pachycondyla crassi-
noda (LACM, UGGG). PERU: Madre de Dios,
Zona Reserva Manu, Pakitza, 11.95°S, 71.28°W,
19, 13-18. ii. 1992, D. Quintero, Malaise trap
(MIUP), 29, 14. ii. 1992, 49, 17.ii.1992, 59,
27.ii.1992, B. Brown, D. Feener, injured Pachycon-
dyla crassinoda (LACM, MUSM), 19, 7.iii.l992,
B. Brown, D. Feener, blacklight trap (LACM).

Apocephalus tanyurus new species

(Flg. 96)

SPECIES RECOGNITION. This species has an
extremely aberrant ovipositor. It can be recognized
by the pair of large preapical sclerites and the un-
usual elongate shape of the other sclerotized por-
tions of the ovipositor.

DESCRIPTION. Body length 1.6-1. 8 mm. Frons
yellow, anterior margin relatively straight. One pair
of supra-antennal setae present. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax light brown; pleuron white. An-
terior scutellar seta small, fine, subequal to poste-
rior setulae of scutum. Legs yellowish-brown; apex
of hind femur with abrupt darkening on anterior
face. Mean costal length 0.49 wing length; range
0.47-0.5. Wing vein R2+3 present. Halter brown.
Abdominal tergites dark brown, except tergite 6,
which is partly to completely yellowish. Venter of
abdomen yellow. Abdominal tergites of normal
form. Tergite 3 evenly colored. Tergite 6 completely
divided, with long seta at posterolateral corner.
Venter of segments 3-5 bare. Abdomen without
dense lateral setae. Ventral setae of segment 6 short,
consisting of a patch of setae, progressively increas-
ing in size posteriorly. Ovipositor (Fig. 96) straight
in lateral view, slightly sclerotized, but with small
pair of darker preapical sclerites. Lateral darkening
greatly elongate, posteriorly diverging, margin
dark, complete. Dorsoapical sclerite of ovipositor
not differentiated. Apicodorsal margin of oviposi-
tor straight. Apicoventral margin of ovipositor

50 ■ Contributions in Science, Number 482

drawn out into pointed process. Ovipositor without
ventral postapical sclerite. Sternite 7 broad, lightly
sclerotized, with large lateral spine. Dufour’s mech-
anism round. Abdominal glands in segment 5
white, inconspicuous in cleared specimens. Internal
sclerite not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from Ecuador and Colombia.

WAY OF LIFE. One female was attracted to an
injured worker of Paraponera clavata.

PHYLOGENETIC RELATIONSHIPS. Based on
the pair of preapical sclerites and the round Du-
four’s mechanism, this species might be an extreme-
ly aberrant member of the A. paraponerae- series.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Greek words tany and oura,
for elongate and tail, referring to the elongate ovi-
positor.

HOLOTYPE. 9 , ECUADOR: Sucumbios: Sacha
Lodge, 0.5°S, 76.5°W, 1-3 l.xii. 1994, P. Hibbs,
Malaise trap, 270 m [LACM ENT 050785]
(LACM).

PARATYPES. COLOMBIA: Valle: Rio Raposo,
19, x.1964, V. Lee, light trap (USNM). ECUA-
DOR: Sucumbios: Anagu, 0.48°S, 76.38°W, 19,
10.ix.1997, P. DeVries, injured Paraponera clavata
(LACM), Sacha Lodge, 0.5°S, 76.5°W, 29, 10-
2 1.x. 1994, P. Hibbs, Malaise trap, 270 m (LACM,
QCAZ).

Apocephalus contracticauda new species

(Figs. 97-98)

SPECIES RECOGNITION. The narrow, elon-
gate apex of the ovipositor is diagnostic for this
species.

DESCRIPTION. Body length 1.3 mm. Frons yel-
low, anterior margin relatively straight. One pair of
supra-antennal setae present. Flagellomere 1 light
brown, round. Proboscis normal, small. Palpus yel-
low. Dorsum of thorax light brown; pleuron white.
Anterior scutellar seta small, fine, subequal to pos-
terior setulae of scutum. Legs yellowish-brown;
apex of hind femur with abrupt darkening on an-
terior face. Mean costal length 0.43 wing length.
Wing vein R2+3 present. Halter brown. Abdominal
tergites dark brown. Venter of abdomen yellow.
Abdominal tergites of normal form. Tergite 3 even-
ly colored. Tergite 6 completely divided, with short
setae at posterolateral corner. Venter of segments
3-5 bare. Abdomen without dense lateral setae.
Ventral setae of segment 6 long, consisting of sev-
eral setae in a straight row. Ovipositor (Figs. 97-
98) straight in lateral view, lightly but evenly scler-
otized dorsally. Lateral darkening thin, elongate,
margin dark, complete. Dorsoapical sclerite of ovi-
positor not differentiated. Apicodorsal margin of
ovipositor straight. Apicoventral margin of ovipos-
itor drawn out into pointed process. Ovipositor
without ventral postapical sclerite. Sternite 7 nar-
row but apically expanded. Abdominal glands in
segment 5 white, inconspicuous in cleared speci-

Brown: Revision of Apocephalus miricauda- group

mens. Internal sclerite rounded, with a short pro-
cess.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in eastern Costa Rica.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin words contractus and
cauda, for narrow and tail, referring to the narrow
apex of the ovipositor.

HOLOTYPE. COSTA RICA: Limon: 7 km SW
Bribri, 9.58°N, 82.88°W, ix-xi.1989, P. Hanson,
Malaise trap, 50 m [LACM ENT 005293]
(LACM).

Apocephalus indistinctus new species
(Fig. 99)

SPECIES RECOGNITION. This species has a
heavily sclerotized ovipositor, similar to those of
the A. funditus- subgroup species. It differs from
them by the relatively straight line of the ovipositor
in lateral view.

DESCRIPTION. Body length 1.1 mm. Frons yel-
low, anterior margin relatively straight. One pair of
supra-antennal setae present. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax light brown; pleuron yellow. An-
terior scutellar seta large, bristle-like. Legs yellow;
apex of hind femur slightly darker on anterior face.
Mean costal length 0.48 wing length. Wing vein
R2+3 present. Halter brown. Abdominal tergites
dark brown. Venter of abdomen yellow. Abdominal
tergites of normal form. Tergite 3 evenly colored.
Tergite 6 completely divided, with short setae at
posterolateral corner. Venter of segments 3-5 with
long, scattered setae. Abdomen without dense lat-
eral setae. Ventral setae of segment 6 long, consist-
ing of a complete ventral and lateral row. Ovipos-
itor (Fig. 99) slightly sinuous in lateral view, lightly
but evenly sclerotized dorsally. Lateral darkening
broadened, enlarging posteriorly, apically rounded,
margin dark, complete. Dorsoapical sclerite of ovi-
positor rounded anteriorly. Apicodorsal margin of
ovipositor straight. Apicoventral margin of ovipos-
itor straight. Ovipositor without ventral postapical
sclerite. Sternite 7 narrow but apically expanded.
Dufour’s mechanism elongate. Abdominal glands in
segment 5 white, inconspicuous in cleared speci-
mens. Internal sclerite somewhat rectangular, with
long, broad process.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in Amazonian Ecuador.

WAY OF LIFE. Unknown.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin words in and distinctus,
for not and different, referring to the lack of a dis-
tinct sternite 7.

HOLOTYPE. 9 , ECUADOR: Sucumbios: Sacha
Lodge, 0.5°S, 76.5°W, 14-24.iii.1994, P. Hibbs,
Malaise trap, 270 m [LACM ENT 036449]
(LACM).

Contributions in Science, Number 482

Apocephalus dinoponerae new species

(Figs. 100-101)

SPECIES RECOGNITION. With its broad me-
dial sclerite and black lateral darkenings, this spe-
cies is similar to A. kungae, with which it occurs.
In A. dinoponerae, however, the ovipositor is
straight in lateral view, rather than curved upward
as in A. kungae.

DESCRIPTION. Body length 2-2.4 mm. Frons
yellow, anterior margin relatively straight. One pair
of supra-antennal setae present. Flagellomere 1 yel-
low, round. Proboscis normal, small. Palpus yellow.
Dorsum of thorax light brown; pleuron white. An-
terior scutellar seta slightly enlarged. Legs yellow-
ish-brown; apex of hind femur with abrupt dark-
ening on anterior face. Mean costal length 0.53
wing length; range 0.51-0.55. Wing vein R2+3 pre-
sent. Halter brown. Abdominal tergites dark
brown. Venter of abdomen yellow. Abdominal ter-
gites of normal form. Tergite 3 evenly colored. Ter-
gite 6 anteriorly emarginate, with short setae at
posterolateral corner. Venter of segments 3-5 with
short setae, concentrated medially. Abdomen with-
out dense lateral setae. Ventral setae of segment 6
long, consisting of a complete ventral and lateral
row. Ovipositor (Figs. 100-101) straight in lateral
view, lightly but evenly sclerotized dorsally. Lateral
darkening posteriorly enlarged, inner margin ex-
tended medially, margin dark, complete. Dorsoap-
ical sclerite of ovipositor not differentiated. Api-
codorsal margin of ovipositor rounded. Apicoven-
tral margin of ovipositor drawn out into pointed
process. Ovipositor without ventral postapical
sclerite. Sternite 7 triangular. Dufour’s mechanism
elongate. Abdominal glands in segment 5 white, in-
conspicuous in cleared specimens. Internal sclerite
a simple, round loop.

GEOGRAPHICAL DISTRIBUTION. Known
only from southeastern Colombia. The host ant,
Dinoponera longipes, is known from Brazil and
Peru (Kempf, 1971).

WAY OF LIFE. Females, both egg-layers and
feeders (see Behavioral Aspects), were attracted to
injured workers of Dinoponera longipes. Four to
nine eggs (n = 3 oviposition events, x = 6.00 eggs,
sd = 2.65) were laid, usually through the suture
between the propodeum and the petiole but on at
least one occasion through the antennal suture in
the head. Larvae finished feeding and emerged from
the host after 1-7 days (n = 22 larvae, x = 4.00
days, sd = 1.95).

DERIVATION OF SPECIFIC EPITHET. The
name is based on the genus name of its host, Di-
noponera longipes.

HOLOTYPE. $, COLOMBIA: Amazonas: 22
km NW Leticia, 4.04°S, 69.99° W, 28.viii.1997, B.
Brown, G. Kung, injured Dinoponera longipes
[LACM ENT 093676] (UNCB).

PARATYPES. COLOMBIA: Amazonas: Ama-
cayacu National Park, 3.82°S, 70.26°W, 1$,
4.ix.l997, 5?, 5.ix.l997, B. Brown, G. Kung, in-

Brown: Revision of Apocephalus miricauda- group ■ 5 1

Figures 102-113. Apocephalus species. Figures 102-103. Apocephalus latinsulosus new species. 100. Dorsal. 101. Left
lateral. Figures 104-105. Heads, anterolateral view. 104. Apocephalus lopesi (Borgmeier). 105. Apocephalus brevifrons
new species. Figures 106-109. Internal, sclerotized loop. 106. Apocephalus eurydomus new species. 107. Apocephalus
flexus new species. 108. Apocephalus kungae new species. 109. Apocephalus petiolus new species. Figure 110. Tergite 3,
Apocephalus maculosus new species. Figure 111. Venter of abdomen, Apocephalus eurydomus new species. Figures 112-
113. Segments 5 and 6 of abdomen, lateral. 112. Apocephalus densepilosus Borgmeier. 113. Apocephalus comosus new
species.

jured Dinoponera longipes (LACM), 22 km NW
Leticia, 4.04°S, 69.99° W, 22$, 27.viii.1997, 3$,
28.viii.1997, 1$, 7.ix.l997, B. Brown, G. Kung,
injured Dinoponera longipes (LACM, MCZC,
MZSP, UNCB, USNM).

Apocephalus latinsulosus new species

(Figs. 102-103)

SPECIES RECOGNITION. The small, separate
lateral sclerites of the ovipositor are diagnostic for
this species.

52 ■ Contributions in Science, Number 482

DESCRIPTION. Body length 1.9-2. 5 mm. Frons
yellow, anterior margin relatively straight. One pair
of supra-antennal setae present (but one specimen
has a second small pair). Flagellomere 1 light
brown, round. Proboscis normal, small. Palpus yel-
low. Dorsum of thorax light brown; pleuron white.
Anterior scutellar seta small, fine, subequal to pos-
terior setulae of scutum. Legs yellowish-brown;
apex of hind femur with abrupt darkening on an-
terior face. Mean costal length 0.63 wing length;
range 0.61-0.67. Wing vein R2+3 present. Flalter

Brown: Revision of Apocephalus miricauda-gxoup

116.

117.

tooth-like swelling

Figures 114-118. Apocephalus species. Figures 114-117. Dufour’s mechanisms. 114. Apocephalus dichromatus Brown.
115. Apocephalus pseudocercus Brown. 116. Apocephalus sp. 116 (unnamed A. grandipalpis- group species). 117. Apo-
cephalus paraponerae Borgmeier. Figure 118. Apocephalus fuscipalpis Borgmeier. Ovipositor, lateral.

brown. Abdominal tergites dark brown. Venter of
abdomen yellow. Abdominal tergites of normal
form. Tergite 3 evenly colored. Tergite 6 completely
divided, with short setae at posterolateral corner.
Venter of segments 3-5 with a few scattered setae.
Abdomen without dense lateral setae. Ventral setae
of segment 6 long, consisting of a complete ventral
and lateral row. Ovipositor (Figs. 102-103) straight
in lateral view, lightly but evenly sclerotized dor-
sally. Lateral darkening posteriorly enlarged, trun-
cate, margin dark, complete. Dorsoapical sclerite of
ovipositor thin, triangular. Apicodorsal margin of
ovipositor straight. Apicoventral margin of ovipos-
itor straight. Ovipositor without ventral postapical
sclerite. Ovipositor with small, separate sclerites
posterolaterally. Sternite 7 narrow but apically ex-
panded. Abdominal glands in segment 5 white, in-
conspicuous in cleared specimens. Internal sclerite
not seen.

GEOGRAPHICAL DISTRIBUTION. Known
from a single mid-elevation site in Costa Rica.

WAY OF LIFE. Females of this species were at-
tracted to injured workers of the presumed host,
Pachycondyla impressa (Roger). However, we ob-
served no ovipositions; indeed, most of the females
were without mature eggs and could not possibly

Contributions in Science, Number 482

have parasitized hosts. They apparently were only
interested in feeding (see Behavioral Aspects).

PHYLOGENETIC RELATIONSHIPS. This spe-
cies is nearly a perfect intermediate between the A.
attophilus and A. miricauda- groups. It has a highly
differentiated apical sclerite that is nevertheless still
articulated with the anterior portion of the ovipos-
itor by a forked process in addition to the thin me-
dian strip. This multiple articulation is all that ex-
cludes it from the A. attophilus-group.

DERIVATION OF SPECIFIC EPITHET. The
name is based on the Latin words latus, for side,
and insulosus, for islands, referring to the many
separate lateral sclerites on the ovipositor.

HOLOTYPE. $ , COSTA RICA: Guanacaste: Es-
tacion Cacao, 10.93°N, 85.47°W, 30.vi.1997, B.
Brown, injured Pachycondyla impressa [LACM
ENT 093422] (LACM).

PARATYPES. COSTA RICA: Guanacaste: Esta-
cion Cacao, 10.93°N, 85.47°W, 2$, ii.1989, P.
Hanson, Malaise trap (LACM), 1 9 , vii.1993, R.M.
Guzman, 19, 12-17. vii.1993, F.A. Quesada
(INBC), 29, 29.vi.1997, 79, 30.vi.1997, B.
Brown, J. Paldi, E. Holscher, injured Pachycondyla
impressa (LACM, MUCR).

Brown: Revision of Apocephalus miricauda- group ■ 53

anterior process of v-shaped
darkening fused

Q apical sclerite with thin

medial connection only

Q apical sclerite of ovipositor

differentiated ventrally

Dufour’s mechanism elongate

Q- ovipositor greatly elongate

6- ovipositor with lateral darkening

119

Figure 119. Hypothesis of relationships among A. miricauda-group and A. attop hilus-gr oup taxa.

KEY TO FEMALES

Many species treated in this work are externally
similar, with few characters to separate them be-
sides those found in the female ovipositor. There-
fore, dissection of the female abdomen will possibly
be necessary to obtain an accurate identification.

I did not examine specimens of A. maculicauda
Borgmeier, a species that might belong in the A.
miricauda-group.

The fossil species, A. succineus, is not included
in this key.

1 Abdominal segments 4 and 5 with long lateral
setae (Fig. 112); ovipositor as in Fig. 70, with

small, medial preapical sclerite

A. densepilosus Borgmeier

[ Pachycondyla crassinoda ; Amazon]

- Abdominal segments 4, and usually 5, bare lat-
erally 2

2 Abdominal segment 5 with dense patch of long

setae on ventrolateral lobe (Fig. 113)

A. comosus new species

[Ectatomma tuberculatum ; Central America]

- Abdominal segment 5 without ventrolateral,
lobe-like process bearing dense, long setae ... 3

3 Frons with narrow anterior process bearing rec-
linate supra-antennal seta directly below lower

54 ■ Contributions in Science, Number 482

interfrontal seta (Fig. 104)

A. lopesi (Borgmeier)

[Odontomachus haematodus;

Brazil, Colombia,
Costa Rica, Guyana]

- Frons with straight anterior margin with nor-

mal, proclinate supra-antennal setae; if supra-
antennal setae somewhat porrect or reclinate,
then not located directly below lower interfron-
al setae (Fig. 105) 4

4 Ovipositor broad, with lateral darkenings well-
defined, long (usually comprising about one-
half of ovipositor length), subparallel, relatively
narrow (Figs. 1-9); most species with a pair of
preapical sclerites dorsally (Figs. 1, 3, 6), one
with none (Fig. 8); ventrally ovipositor with
separate, triangular, postapical sclerite with
pointed (Figs. 2, 4) or truncate (Figs. 7, 9) apex
5

- Ovipositor of most species thinner and more

elongate, with lateral darkenings less defined,
shorter, not parallel, or broader; 1-2 dorsal
preapical sclerites present in some species, oth-
ers with none; ventrally with or without sepa-
rate postapical sclerite, which — if present — is
always pointed 8

5 Lateral darkenings of ovipositor yellowish-

Brown: Revision of Apocephalus miricauda-group

A. paraponerae - series

AS

\<p

<?' ^ $
V

<r

/„

120

Figure 120. Hypothesis of relationships within A. paraponerae-subgroup.

brown in color; ovipositor posteriorly with pair
of dark, round, preapical sclerites (Fig. 6); apex
of ventral postapical sclerite truncate (Fig. 7)

A. melinus new species

[Dolichoderus attelaboides ; Amazon]

- Lateral darkenings of ovipositor black; ovipos-

itor with or without dorsal preapical sclerites;
apex of ventral postapical sclerite pointed or
truncate 6

6 Dorsally with area of moderate sclerotization

reaching to apex and without preapical sclerites
(Fig. 8); apex of ventral postapical sclerite trun-
cate (Fig. 9); one pair of supra-antennal setae
present A. roeschardae new species

[Cephalotes atratus ; Amazon]

- Dorsally with posterior portion of ovipositor

relatively unsclerotized except for pair of small
preapical sclerites; apex of ventral postapical
sclerite pointed; two pairs of supra-antennal se-
tae present 7

7 Lateral darkenings relatively thin (Fig. 1);

preapical pair of sclerites larger, darker; venter
of abdominal segment 6 usually with only 2 se-
tae (sometimes with up to four)

A. paraponerae Borgmeier

[Paraponera clavata, Ectatomma tuberculatum,

Contributions in Science, Number 482

Pacbycondyla spp.; widespread in lowlands of
Neotropical Region]

- Lateral darkenings relatively thicker (Fig. 3);
preapical pair of sclerites relatively thin, light-
colored; venter of abdominal segment 6 with
row of several setae ... A deceptus new species

[Pacbycondyla commutata ; Ecuador]

8 Ovipositor spatulate (dorsally concave), apical-
ly upturned, often with lateral, upturned apices
(Fig. 37); ovipositor usually without differenti-
ated medial preapical sclerite, sometimes with
lateral darkenings indistinct or not differenti-
ated; ovipositor of most species with long pos-
teroventral filament-like process (Fig. 40); in
most species, apical region of darker sclerotea-
tion encircles entire ovipositor (Fig. 39) ... 9

- Ovipositor apically flat or convex, often with

preapical sclerites; lateral darkenings usually
distinctive; ovipositor without narrow, thread-
like extension; ventral sclerotization not encir-
cling entire sclerite 23

9 Abdominal segments 3-5 bare ventrally . . 10

- Abdominal segments 3-5 with black setae . . .

11

10 Intersegment 6-7 with sclerotized, black, clear-
ly visible striations

Brown: Revision of Apocephalus miricauda- group ■ 5 5

,\cr \pr ^

c j£ .rtS» d?

///y

v v v v

A. striativentris new species

[host unknown; Brazil, Costa Rica, Ecuador,

Peru]

- Intersegment 6-7 without darkened striations
A. incomptus new species

[host unknown; Ecuador]

11 Abdominal setae dense, numerous, sometimes

short 12

- Abdominal setae scattered, relatively few, rela-
tively long 15

12 Ovipositor with dorsal spine (Fig. 45) ... 13

- Ovipositor without dorsal spine 14

13 Posterolateral apex of ovipositor truncate (Fig.

44) A. acanthus new species

[host unknown; Costa Rica]

- Posterolateral apex of ovipositor right-angled

(Fig. 46) A. spatulatus Borgmeier

[host unknown; Brazil]

14 Apical region of ovipositor triangular (Fig. 42)
A. triangularis new species

[host unknown; Brazil, Mexico]

- Apical region of ovipositor quadrate (Fig. 43)
A. quadratus Brown

[host unknown; Brazil, Ecuador]

56 ■ Contributions in Science, Number 482

15 Tergite 6 entire, not divided into two separate

sclerites 16

- Tergite 6 divided into two separate sclerites . .

17

16 Fateral margin of ovipositor smooth, not inter-
rupted by emarginations (Fig. 41)

A. magnicauda new species

[Camponotus sericeiventris; Ecuador, Peru]

- Lateral margin of ovipositor with emargination
and tooth-like process in lateral view (Fig. 38)
A. emargilatus new species

[( Camponotus sericeiventris ; Costa Rica]

1 7 Dorsal apex of ovipositor drawn out in long, nar-
row, parallel-sided process (Figs. 31-32) ... 18

- Dorsal apex of ovipositor without process of

this type 19

18 Dorsum of ovipositor with prominent, finger-
like process (Fig. 31) ... A. digitalis Borgmeier

[host unknown; Brazil]

- Dorsum of ovipositor without process (Fig. 32)
A. denotatus new species

[host unknown; Costa Rica, Panama]

19 Apex of ovipositor narrowed, pointed (Fig. 30);

Brown: Revision of Apocephalus miricauda- group

laterodorsal tooth-like swelling present (Fig.

118) 20

- Apex of ovipositor truncate (Figs. 33, 35, 36);
ovipositor without tooth-like swelling ... 21

20 Apex of ovipositor darkly sclerotized (Fig. 30)

A. brochus new species

[Camponotus banghaasi, C. rapax ; Brazil, Co-
lombia]

- Apex of ovipositor not darkly sclerotized ....
A. fuscipalpis Borgmeier

[host unknown; Brazil]

21 Posterolateral notch of ovipositor not promi-
nent; lateral darkening of ovipositor short but

distinct (Fig. 33)

A. pachycondylae new species

[Pacbycondyla apicalis, P. obscuricornis ; Costa

Rica]

- Ovipositor with well-differentiated posterolat-

eral, raised notch (Fig. 37); lateral darkenings
not distinct 22

22 Posterodorsal apex of ovipositor with thick,

black sclerotized margin (Fig. 35)

A. atrimarginatus new species

[Pacbycondyla unidentata ; Costa Rica]

- Posterodorsal apex of ovipositor with at most

thin dark margin (Fig. 36)

A. batillus new species

[host unknown; Panama]

23 Ventral apex of ovipositor with well-developed,

pointed, triangular postapical sclerite (Figs. 16,
18, 20, 24, 26) and lateral bars of sclerotization
that extend medially (Fig. 18); dorsally without
central preapical sclerite (but with small pair of
sclerites in one species; Fig. 5); lateral darken-
ings of some species thickened, rounded, and
projecting laterally (Figs. 6, 11, 13) 24

- Ventral apex of ovipositor usually truncate; lack-

ing postapical sclerite and lateral bars; dorsally, in
some species with differentiated central preapical
sclerite (one species with aberrant paired sclerites)
(Fig. 96); lateral darkenings otherwise 36

24 Posterodorsal apex of ovipositor concave, with

thin, medial process (Fig. 14); ovipositor dis-
tinctly constricted in middle; ventrally with
well-developed, prominent triangular pointed
apex A. dracodermus new species

[host unknown; Peru]

- Ovipositor not as above 25

25 Ovipositor broad and triangular in appearance,

broadest at apex (Fig. 21)

A. persecutor Borgmeier

[host unknown; Brazil, Costa Rica, Ecuador]

- Ovipositor more parallel-sided, usually broad-
est at midlength of lateral darkenings .... 26

26 Lateral darkenings relatively straight, not giving

ovipositor a rounded appearance (Figs. 10, 19,
23, 25, 27), although one species with small,
rounded apical lobes 27

- Lateral darkenings curved, giving posterior one-

third of ovipositor a rounded appearance (Figs.
5, 11, 13, 15, 17, 22) 31

27 Venter of segments 4-5 bare; postapical sclerite

Contributions in Science, Number 482

extremely short, triangular-shaped (Fig. 26) . . .

A. spiculus new species

[host unknown; Panama]

- Venter of segments 4-5 with black setae; post-
apical sclerite of various forms 28

28 Lateral darkenings rounded, thickened, lobe-
like (Fig. 27); venter of abdomen with row of

setae present on segments 4-6

A. torulus new species

[host unknown; Colombia, Ecuador]

- Lateral darkenings thin; ventral setation of ab-
domen various 29

29 Dorsum of ovipositor with reticulate sculpture,
medially light-colored (Fig. 19); venter of ovi-
positor with forked, anteriorly projecting pro-
cess (Fig. 20) .... A. reticulatus new species

[host unknown; Costa Rica]

- Dorsum of ovipositor smooth, without reticu-
late sculpture; ventrally without forked process
30

30 Ovipositor medially darkened; laterally paral-
lel-sided, with apical darkenings not divergent
(Fig. 10); venter of abdominal segment 6 with
row of long setae, segments 4-5 with relatively

small, black setae

A. conecitonis new species

[host unknown, but usually found with army

ants; Costa Rica]

- Ovipositor not medially darkened; laterally
with small expansion, apical darkenings slightly
divergent (Fig. 23); venter of abdominal seg-
ment 6 with row of long setae, segments 4-5

with few, short, almost invisible setae

A. secus new species

[host unknown; Costa Rica]

31 Venter of ovipositor with dark-colored, poste-
riorly expanded, triangular sclerite (Fig. 12);

dorsum of ovipositor as in Fig. 11

A. constrictus new species

[host unknown; Costa Rica]

- Venter of ovipositor without a well-defined ster-

nite 32

32 Venter of abdominal segments 3-5 with dense

long setae, especially medially; lateral darken-
ings extremely broad, black (Fig. 13)

A. crassilatus new species

[Pacbycondyla spp.; widespread in Neotropical

lowlands]

- Venter of abdominal segments 3-5 with at most

thin, short, scattered setae; lateral darkenings
various 33

33 Ovipositor dorsally with small preapical scler-

ites (Fig. 5); expanded portion of ovipositor
more than twice as broad as rest of ovipositor;
frons with two pairs of subequal supra-antennal
setae A. strongylus new species

[host unknown; Brazil]

- Ovipositor without preapical sclerites; expand-

ed portion less than twice as broad as rest of
ovipositor; frons at most with a smaller, lower
pair of supra-antennal setae (most species with
only one pair) 34

Brown: Revision of Apocephalus miricauda- group ■ 57

34 Ovipositor with medial, moderately sclerotized

triangular area dorsally (Fig. 22)

A. curtinotus new species

[host unknown; Brazil]

- Ovipositor without medial sclerotization ... 35

35 One pair of supra-antennal setae present. Ven-
trally, ovipositor with transverse, posterior
sclerites not touching (Fig. 18); postapical scler-
ite elongate; tergite 6 only slightly emarginate
anteriorly; posteriorly with large pair of setae

A. inpalpabilis new species

[host unknown; Costa Rica]

- A second, extremely small pair of supra-anten-
nal setae present. Ventrally, ovipositor with
tranverse sclerites joined to form a single struc-
ture (Fig. 16); postapical sclerite short, broad;
tergite 6 deeply emarginate anteriorly; posteri-
orly with less differentiated pair of setae ....
A. indeptus new species

[host unknown; Costa Rica]

36 Ovipositor curved dorsally at midlength (Fig.

49); sternite 7 with lateral fringe of small pro-
cesses (Fig. 48) 37

- Ovipositor relatively straight or curved ventral-

ly at midpoint in lateral view; sternite 7 without
lateral processes 40

37 Sternite 7 extremely broad, posteriorly exceed-
ing width of dorsum of ovipositor (Fig. 47); two

pairs of supra-antennal setae

A. miricauda Borgmeier

[ Dinoponera gigantea ; Brazil]

- Sternite 7 narrower, not exceeding width of

dorsum of ovipositor; one pair of supra-anten-
nal setae 38

38 Without separate dorsal preapical sclerite; in-
stead, broad, sclerotized lobe extending to apex
(Fig. 50); sternite 7 with posterior rounded ex-
pansion extending across most of segment;

sclerotized loop relatively small

A. kungae new species

[Dinoponera longipes\ Colombia]

- Ovipositor with distinct preapical sclerite (Figs.

51, 53); sternite 7 much narrower, only slightly
expanded posteriorly; sclerotized loop extreme-
ly large or small 39

39 Sclerotized loop extremely large, over one-half

width of ovipositor (Fig. 51)

A. flexus new species

[host unknown; Ecuador]

- Sclerotized loop much smaller than one-half

width of ovipositor (Fig. 53)

A. orbiculus new species

[host unknown; Costa Rica]

40 Dorsum of ovipositor with separate (i.e., not

anteriorly attached to other sclerites), single dis-
tinct preapical sclerite (e.g., Fig. 78) 41

- Sclerites of ovipositor various (Figs. 54, 55, 57,
58, 60, 62, 64, 71), but not as above .... 56

41 Lateral darkenings extremely broad, thickened

(Figs. 67, 68, 72) 42

- Lateral darkenings narrower 44

42 Lateral darkenings rounded, lobe-like (Fig. 72);

58 ■ Contributions in Science, Number 482

ventral apex of ovipositor without dark, heavily
sclerotized plate; frons subequal in length and

width A. lobicauda new species

[Ectatomma tuberculatum ; Costa Rica]

- Lateral darkenings shaped differently; ventral
apex of ovipositor with dark-colored, heavily
sclerotized median sclerite; frons short ... 43

43 Dorsal preapical sclerite narrow, much longer

than broad (Fig. 67)

A. spatulicauda Borgmeier

[host unknown; Brazil]

- Dorsal preapical sclerite broader than long (Fig.

68) A. inimicus Borgmeier

[host unknown; Brazil]

44 Preapical sclerite large, oval, shiny (Fig. 73) . .
A. globosus new species

[ Pachycondyla villosa ; Costa Rica]

- Preapical sclerite not oval, trapezoidal in most

species (Figs. 74-78) 45

45 Ovipositor, in lateral view, strongly downturned

at midlength (as in Fig. 65); ovipositor dorsally
as in Fig. 66 A. intonsus new species

[host unknown; Panama]

- Ovipositor, in lateral view, straight or only

slightly deflected at apex 46

46 Abdominal tergite 3 with lateral, round light
patch (Fig. 110); venter of abdomen with small
patch of short setae mediolaterally on segment

5, otherwise 3-5 bare

A. maculosus new species

[host unknown; Ecuador]

- Abdominal tergite 3 uniformly dark-colored;
abdominal setation various but not as above . .
47

47 Anterior margin of preapical sclerite deeply

cleft (Figs. 74-78) 48

- Anterior margin of preapical sclerite entire

(Figs. 80, 82, 83) 51

48 Venter of abdomen bare; apices of lateral dark-

ening not markedly convergent posteriorly
(Figs. 75-76); sclerotized loop with broad pro-
cess (Fig. 106) 49

- Venter of abdomen setose; apices of lateral

darkening markedly convergent posteriorly
(Figs. 77-78); sclerotized loop with long, nar-
row process (Fig. 109) 50

49 Dorsal preapical sclerite broadly and deeply

cleft (Fig. 75); apex of ovipositor flat

A. glabriventris new species

[Ectatomma tuberculatum ; Mexico]

- Dorsal preapical sclerite with narrow cleft (Fig.

76); apex of ovipositor downturned

A. minutus Borgmeier

[host unknown; Brazil]

50 Ventral setae of abdominal segments 3-5 short,
about one-half length of long ventral setae on
posterior margin of segment 6; tergite 6 divided
A. cardiacus new species

[host unknown; Costa Rica]

- Ventral setae of abdomen long, dense, giving it
a markedly “hairy” appearance; ventral setae
subequal in length to ventral setae on posterior

Brown: Revision of Apocepbalus miricauda- group

margin of segment 6; tergite 6 entire

A. petiolus new species

[host unknown; Costa Rica]
51 Preapical sclerite much longer than wide, ante-
riorly rounded (Fig. 80); sternite 7 broad, tri-
angular (Fig. 81)

A. gigantivorus new species

[Dinoponera gigantea ; Brazil]
- Preapical sclerite about as long as wide, ante-
rior margin not rounded (Figs. 82-83); sternite

7 various 52

52 Apices of lateral darkening markedly conver-
gent posteriorly (Figs. 83, 85, 87, 89) ... 53

- Apices of lateral darkening not convergent pos-
teriorly (Fig. 82) .... A. piliventris Borgmeier

[Pacbycondyla striata ; Brazil]

53 Preapical sclerite posteriorly cleft (Fig. 83); ster-
nite 7 thin triangular, with medial dark strip
that projects posteriorly from triangular base,
making apex appear three-pronged (Fig. 84);
sclerotized loop large, about one-half width of

ovipositor, similar in shape to Fig. 107

A. annulatus new species

[host unknown; Costa Rica]

- Preapical sclerite not cleft posteriorly; sternite 7

not appearing three-pronged 54

54 Sternite 7 distinctive: anteriorly broad, then
narrowed and expanded posteriorly, with nu-
merous marginal setae and with apex of medial
black thickening enlarged; posteriorly nar-
rowed and again expanded to width of ovipos-
itor (Fig. 83) ... A. contortiventris new species

[Pacbycondyla impressa; Ecuador]

- Sternite 7 not as above; medial black thicken-
ing, when present, not apically expanded . . 55

55 Sternite 7 apically expanded (Fig. 88)

A. eurydomus new species

[Pacbycondyla barpax ; Costa Rica, Panama,

USA]

- Sternite 7 not expanded apically (Fig. 90) ...
A. conformalis new species

[host unknown; Brazil]

56 Single preapical sclerite present but attached to

anterior end of lateral darkenings by a distinc-
tive sclerotized bar (Figs. 54, 55, 57, 58); most
species with lateral postapical sclerites (e.g.,
Fig. 55) 57

- Preapical sclerites, if present, not attached to
anterior end of lateral darkenings (Figs. 91-97,
99-100, 102); lateral postapical sclerites absent
61

57 Venter of abdomen with numerous long setae,

giving it a “hairy” appearance 58

- Venter of abdomen with shorter and fewer se-
tae, mostly concentrated medially 59

58 Preapical sclerite and lateral bars forming rela-
tively smooth anterior margin (Fig. 54); lateral

postapical sclerites absent

A. meniscus new species

[host unknown; Peru]

- Preapical sclerite projecting anteriorly (Fig. 55);

lateral postapical sclerites present

Contributions in Science, Number 482

A. barbiventris new species

[Odontomacbus barbiventris ; Costa Rica]

59 Preapical sclerite extremely large, filling much
of space between lateral darkenings (Fig. 57)
A. amplidiscus new species

[host unknown; Costa Rica]

- Preapical sclerite small, discrete, with anteriorly
directed bars of sclerotization attaching to an-
terior end of lateral darkenings (Figs. 58, 60)
60

60 Sternite 7 broad triangular (Fig. 59)

A. paldiae new species

[Odontomacbus cbelifer ; Costa Rica]

- Sternite 7 anteriorly rounded, narrowing pos-
teriorly (Fig. 61) . . A. cyclodiscus new species

[host unknown; Panama]

61 Ovipositor with oval, window-like clear areas

(Fig. 91) A. fenestratus new species

[host unknown; Costa Rica]

- Ovipositor without oval clear areas 62

62 Wing vein R2+3 absent; lateral darkenings of
ovipositor extremely broad, triangular (Fig.

69); frons short (Fig. 105)

A. brevifrons new species

[host unknown; Costa Rica, Mexico]

- Wing vein R2+3 present; lateral darkenings not

so broad (except A. latinsulosus (Fig. 102),
which differs by having small lateral sclerites);
frons about as long as wide 63

63 Lateral darkenings downturned, projecting pos-
terolaterally, apparently ending free of rest of

ovipositor (Fig. 92)

A. asyndetus new species

[Gnamptogenys bispinosus ; widespread in Neo-
tropical lowlands]

- Lateral darkenings not strongly downturned

and not ending freely 64

64 Lateral darkenings separated by posterior pro-
cess of ovipositor, which expands posteriorly to

fill entire space between them (Fig. 93)

A. catbolicus new species

[Pacbycondyla crassinoda, barpax , impressa,

Odontomacbus spp.]

- Ovipositor not as above 65

65 Long, thin sclerite subequal in size to lateral

darkenings extending posteriorly between them
(Figs. 94-95) 66

- Ovipositor without long, narrow sclerite ex-
tending between lateral darkenings ...... 67

66 Lateral darkenings lightly sclerotized, yellow-

ish-brown in color, extending posteriorly, so
that posterior margin of ovipositor is concave
(Fig. 94); posterior margin of tergite 6 with
three or more thick setae that are as long as
tergite A. lyratus Borgmeier

[host unknown; Brazil, Ecuador]

- Lateral darkenings heavily sclerotized, almost
black in color, not extending posteriorly past
apex of ovipositor (Fig. 95); posterior margin
of tergite 6 with one or two thinner setae that

are clearly shorter than length of tergite

A. trifidus new species

Brown: Revision of Apocephalus miricauda-group ■ 59

[Pachycondyla crassinoda ; Brazil, Colombia,

Ecuador, Peru]

67 Ovipositor elongate, with iong, thin sclerites

and a pair of preapical sclerites (Fig. 96); venter
of abdominal segment 6 with large patch of sev-
eral rows of setae that increase in length pos-
teriorly; venter of other abdominal segments
bare A. tanyurus new species

[ Paraponera clavata ; Colombia, Ecuador]

- O vipositor and ventral setation not as above . .

/ 68

68 Lateral darkenings strongly expanded posteri-
orly (Fig. 102); ovipositor with lateral, isolated

sclerites (Fig. 103)

A. latinsulosus new species

[Pachycondyla impressa ; Costa Rica]

- Lateral darkenings, if slightly expanded poste-
riorly, without lateral sclerites 69

69 Entire ovipositor, including space between lat-
eral darkenings, dark brown, shiny (Fig. 64);
ovipositor strongly downturned at midlength

(Fig. 65) A. funditus new species

[host unknown; widespread Neotropical Re-
gion]

- At least some portion of ovipositor not dark

brown and shiny (Figs. 97, 99, 100); not down-
turned 70

70 Apical portion of ovipositor narrow; lateral
darkenings thin, subparallel; space between lat-
eral darkenings subequal to thickness of lateral

darkenings (Fig. 97)

A. contracticauda new species

[host unknown; Costa Rica]

- Apical portion of ovipositor relatively broad
(Figs. 66, 100) and shaped differently ... 71

71 Lateral darkenings apically expanded, separat-

ed by rounded sclerite (Fig. 99); venter of ovi-
positor evenly sclerotized, without distinct ster-
nite A. indistinctus new species

[host unknown; Ecuador]

- Lateral darkenings not expanded apically, sep-

arated by broad, lightly sclerotized region (Fig.
100); ventrally with triangular sternite, expand-
ed posteriorly (Fig. 101)

A. dinoponerae new species

[Dinoponera longipes ; Colombia]

BEHAVIORAL ASPECTS

Females of most species had a stereotyped behavior,
similar to that described for A. paraponerae
(Brown and Feener, 1991a). They were attracted to
crushed, injured workers of their host ant species
(Table 2 is a list of crushed ants and the flies that
have been attracted to them), as were males. Indi-
viduals of both sexes approached hosts, but males
did not remain for long periods of time. A female
will approach the host, walk over it, and follow one
of two routines: either she will quickly (within
about 15 seconds) begin to probe with her ovipos-
itor and attempt to lay eggs (“layers”) or she will
feed on hemolymph from the crushing wounds of

Table 2. List of ant species that, when injured, have at-
tracted A. miricauda- group parasitoids (some species,
marked with an asterisk, are attracted to healthy, nonin-
jured hosts).

Ant host

Apocephalus parasitoid

Camponotus banghaasi

brochus *

C. rapax

brochus *

C. sericeiventris

emargilatus

C. sericeiventris

magnicauda *

Cepbalotes atratus

catholicus

C. atratus

roeschardae

Dinoponera gigantea

gigantivorus

D. gigantea

miricauda

D. longipes

dinoponerae

D. longipes

kungae

Dolichoderus attelaboides

catholicus

D. attelaboides

melinus

D. attelaboides

paraponerae

D. decollatus

melinus

Ectatomma goninion

catholicus

E. lugens

paraponerae

E. tuberculatum

comosus

E. tuberculatum

lobicauda

E. tuberculatum

paraponerae

Odontomachus bauri

catholicus

O. chelifer

catholicus

O. chelifer

paldiae

O. haematodus

lopesi

O. hastatus

catholicus

Pachycondyla apicalis

crassilatus

P. apicalis

paraponerae

P. commutata

deceptus

P. commutata

melinus

P. crassinoda

catholicus

P. crassinoda

densepilosus

P. crassinoda

paraponerae

P. crassinoda

trifidus

P. harpax

catholicus

P. impressa

catholicus

P. impressa

crassilatus

P. impressa

latinsulosus

P. striata

piliventris

P. unidentata

atrimarginatus

P. unidentata

crassilatus

P. villosa

crassilatus

P. villosa

globosus

P. villosa

paraponerae

Paraponera clavata

paraponerae

P. clavata

tanyurus

the ant (“feeders”). Feeders were not seen to ovi-
posit, and those examined closely were found to be
incapable of oviposition because they had no ma-
ture eggs in their ovaries. This is reminiscent of an
observation by Disney (1994), that most carrion-
feeding female phorids were not gravid. After ovi-
positing, layers often would quickly leave the host
but sometimes would also stay to feed.

Little work has been done on the development of

60 ■ Contributions in Science, Number 482

Brown: Revision of Apocephalus miricauda-group

sexual maturity in adult phorid flies, so it is not
known if most phorids are able to develop eggs pri-
or to feeding or if they require some sort of protein
meal to produce their eggs. Other workers have
found that carbohydrate meals (sugar) increased fe-
cundity of Megaselia halterata (Wood), but it was
not found to be an absolute necessity (Binns, 1980).
Difficulties in keeping Apocephalus females alive in
culture might make similar studies problematic, but
there is at least an indication that these flies require
a meal of hemolymph before eggs can be matured.

In contrast to other A. miricauda-group species,
the A. spatulatus-subgroup species A. brochus and
A. magnicauda attacked healthy, living hosts. In
particular, we saw A. brochus appear to be at-
tempting to oviposit in the back of the heads of the
host ants. The hosts of both of these species, Cam-
ponotus spp., are highly divergent from those of
other A. miricauda- group taxa; therefore, it is pos-
sible that the shift in behavior was accompanied by
the shift in host. The only specimen of A. emargi-
latus, the sister-species of A. magnicauda, that was
collected from a host was attracted to an injured
worker of Camponotus sericeiventris, however.
This lends support to a scenario where, in A. mag-
nicauda only, the host shift occurred first, and then
the shift to attacking uninjured ants. Much further
research is necessary to resolve these issues.

The other species of this group to have radically
shifted their hosts, A. melinus and A. roeschardae,
are possibly also each other’s closest relatives. Each
behaves in a similar manner to A. paraponerae, at-
tacking injured hosts exclusively. It would be inter-
esting to compare the chemical components of the
mandibular glands of the hosts of these two species
with those of Paraponera clavata. Apparently the
chemicals of Dolichoderus attelaboides are similar
enough to those of ponerine ants to occasionally
attract species such as A. paraponerae and A. cath-
olicus (Table 2).

ACKNOWLEDGMENTS

Jesse Cantley expertly produced the illustrations for this
paper. Technical assistance was rendered by Vladimir Ber-
ezovskiy. For help with field work in Ecuador, I thank
Jesse Cantley, Peter Hibbs, and Jacqueline Roschard; for
permission to do research there, I thank the Fundacion
Maquipucuna, Fundacion Jatun Sacha, and the Catholic
University (especially Dr. Giovanni Onore). For help with
field work in Costa Rica, I thank Jesse Cantley, Erik
Holscher, and Jill Paldi; for permission to do research
there, I thank the Organization for Tropical Studies, the
Arthropods of La Selva (ALAS) Project, and the Instituto
Nacional de Biodiversidad (especially Manuel Zumbado).
For help with field work in Colombia, I thank Giar-Ann
Kung; for help obtaining permission to do research there,
I thank Fernando Fernandez, Diego Campos, and Mike
Sharkey. For help obtaining permission to do field work
in Peru, I thank the Smithsonian BIOLAT program and
Dr. Gerardo Lamas. In Guyana I was aided by Carol Kel-
loff of the Smithsonian Institution’s Biodiversity in the
Guyanas project, Mike Tamessar, Diante Nerine, and Alex

Contributions in Science, Number 482

and Adriana Mendes. For material from Panama I thank
Diomedes Quintero and John Pickering.

This research was funded by grants from the Canadian
Natural Sciences and Engineering Research Council to B.
Brown and G.E. Ball, the Mellon Foundation, the Weiler
Foundation (to the Natural History Museum of Los An-
geles County), the Smithsonian Institution BIOLAT pro-
gram, and National Science Foundation (NSF) grant DEB-
9407190 to B. Brown. NSF grants BSR-9025024, DEB-
9401069, and DEB-9706976 funded the ALAS project;
grants DEB-9522581, DEB-96421221, and the Smithson-
ian Environmental Sciences Program supported John Pick-
ering.

LITERATURE CITED

Arnett, R.H., G.A. Samuelson, and G.M. Nishida. 1993.
The insect and spider collections of the world.
Gainesville, Florida: Sandhill Crane Press, vi + 310

pp.

Baroni Urbani, C. 1994. The identity of the Dominican
Paraponera (Amber Collection Stuttgart: Hymenop-
tera, Formicidae. V: Ponerinae, partim). Stuttgarter
Beitrdge zur Naturkunde, Serie B 197:1-9.

Baroni Urbani, C., B. Bolton, and P.S. Ward. 1992. The
internal phylogeny of ants. Systematic Entomology
17:301-329.

Binns, E.S. 1980. Mating behavior, fecundity and feeding
in the mushroom phorid, Megaselia halterata
(Wood) (Diptera). Entomologist’s Monthly Maga-
zine 116:45-57.

Bolton, B. 1995. A new general catalogue of the ants of
the world. Cambridge, MA: Harvard University
Press, 504 pp.

Borgmeier, T. 1925. Novos subsidios para o conhecimento
da familia Phoridae. Archivos do Museu Nacional,
Rio de Janeiro 25:85-281.

. 1958. Neue Beitraege zur Kenntnis der neotro-

pischen Phoriden (Diptera, Phoridae). Studia Ento-
mologica 1:305-406.

. 1961. Weitere Beitraege zur Kenntnis der neotro-

pischen Phoriden, nebst Beschreibung einiger Dohr-
niphora-Anen aus der indo-australischen Region
(Diptera, Phoridae). Studia Entomologica 4:1-112.

. 1968. A catalogue of the Phoridae of the World

(Diptera, Phoridae). Studia Entomologica 11:1-367.

. 1969. New or little-known phorid flies, mainly of

the Neotropical Region. Studia Entomologica 12:
33-132.

. 1971. Further studies on phorid flies, mainly of

the Neotropical Region (Diptera, Phoridae). Studia
Entomologica 14:1-172.

Brown, B.V. 1992. Generic revision of Phoridae of the Ne-
arctic Region and phylogenetic classification of
Phoridae, Sciadoceridae and Ironomyiidae (Diptera:
Phoridea). Memoirs of the Entomological Society of
Canada 164:1-144.

. 1993. Taxonomy and preliminary phylogeny of

the parasitic genus Apocephalus, subgenus Meso-
phora (Diptera: Phoridae). Systematic Entomology
18:191-230.

. 1994. Revision and new species of the Apoce-
phalus ( Mesophora ) truncaticercus- infragroup (Dip-
tera: Phoridae). Contributions in Science 449:1-7.

. 1996. A further species of Apocephalus, subgenus

Mesophora (Diptera: Phoridae) parasitic on stingless
bees (Hymenoptera: Apidae: Meliponinae). Studia
dipterologica 3:231-235.

. 1997a. Parasitic phorid flies: A previously unrec-

Brown: Revision of Apocephalus miricauda- group 161

ognized cost to aggregation behavior of male sting-
less bees. Biotropica 29:370-372.

— . 1997b. Revision of the Apocephalus attophilus-

group of ant-decapitating flies (Diptera: Phoridae).
Contributions in Science 468:1-60.

— . 1997c. Systematics and fossil evidence of host-

parasitoid relationships of Calamiscus Borgmeier
(Diptera: Phoridae). Journal of Natural History 31:
1253-1259.

— — — . 1999. Review of the fossil Phoridae. Journal of
Natural History 33:1561-1573.

Brown, B.V., and D.H. Feener, Jr. 1991a. Behavior and
host location cues of Apocephalus paraponerae
(Diptera: Phoridae), a parasitoid of the giant tropical
ant Paraponera clavata (Hymenoptera: Formicidae).
Biotropica 23:182-187.

— . 1991b. Life history parameters and immature

stages of Apocephalus paraponerae (Diptera: Phori-
dae), a parasitoid of the giant tropical ant Parapo-
nera clavata (Hymenoptera: Formicidae). Journal of
Natural History 25:221-231.

— . 1995. Efficiency of two mass sampling methods

for sampling phorid flies (Diptera: Phoridae) in a
tropical biodiversity survey. Contributions in Science
459:1-10.

. 1998. Parasitic phorid flies (Diptera: Phoridae) as-
sociated with army ants (Hymenoptera: Formicidae:
Ecitoninae, Dorylinae) and their conservation biol-
ogy. Biotropica 30:482-487.

Brown, W.L., Jr. 1976. Contributions toward a reclassifi-
cation of the Formicidae. Part VI. Ponerinae, tribe
Ponerini, subtribe Odontomachiti. Section A. Intro-
duction, subtribal characters. Genus Odontomachus.
Studia Entomologica 19:67-171.

Coquillett, D.W. 1901. Apocephalus Coquillett, nov. gen.
Proceedings of the Entomological Society of Wash-
ington 4:501.

Crawford, R.L. 1983. Grid systems for recording speci-
men collection localities in North America. System-
atic Zoology 32:389-402.

Disney, R.H.L. 1989. Scuttle flies — Diptera, Phoridae, Ge-
nus Megaselia. Handbooks for the Identification of
British Insects 10:1-155.

— . 1994. Scuttle flies: The Phoridae. London: Chap-
man and Hall, xii + 467 pp.

Farris, J.S. 1989. Hennig86: A PC-DOS program for phy-
logenetic analysis. Cladistics 5:163.

Feener, D.H., Jr., L.F. Jacobs, and J.O. Schmidt. 1996. Spe-
cialized parasitoid attracted to a pheromone of ants.
Animal Behavior 51:61-66.

Iturralde-Vinent, M.A., and R.D.E. MacPhee. 1996. Age
and paleogeographical origin of Dominican amber.
Science 273:1850-1852.

Kempf, W.W. 1971. A preliminary review of the ponerine
ant genus Dinoponera Roger (Hymenoptera: For-
micidae). Studia Entomologica 14:369-394.

Lattke, J.E. 1995. Revision of the ant genus Gnampto-
genys in the New World (Hymenoptera: Formici-
dae). Journal of Hymenoptera Research 4:137-193.

Morehead, S.A., and D.H. Feener. 1997. Is there cryptic
host race formation in a parasitoid of ants? Tropical
diversity: origins, maintenance and conservation, p.
87. Program and Abstracts, Association for Tropical
Biology Annual Meeting. San Jose, Costa Rica.

Peterson, B.V., and W.H. Robinson. 1976. A new North
American genus and species of the family Phoridae
(Diptera). Canadian Entomologist 108:119-121.

Ride, W.D.L., C.W. Sabrosky, G. Bernardi, and R.V. Mel-
ville. 1985. International Code of Zoological No-
menclature. London: International Trust for Zoolog-
ical Nomenclature, xx + 338 pp.

Schmitz, H. 1915. Neue Beitrage zur Kenntnis der myr-
mecophilen und termitophilen Phoriden. Wiener En-
tomologischen Zeitung 34:311-330.

Silveira-Costa, A.J., and P.R.S. Moutinho. 1996. Attract-
ing parasitic phorid flies (Diptera: Phoridae) to in-
jured workers of the giant ant Dinoponera gigantea
(Hymenoptera: Formicidae). Entomological News
107:93-98.

Thompson, F.C. 1994. Bar codes for specimen data man-
agement. Insect Collection News 9:2-4.

Ward, P.S. 1994. Adetomyrma, an enigmatic new ant ge-
nus from Madagascar (Hymenoptera: Formicidae),
and its implications for ant phylogeny. Systematic
Entomology 19:159-175.

Received 7 April 1999; accepted 2 November 1999.

Natural History Museum

of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

Ai H

Number 483
16 October 2000

Contributions
in Science

Cranial Morphology of Pterodaustro
guinazui (Pterosauria: Pterodactyloidea)
from the Lower Cretaceous of
Argentina

Luis M. Chiappe, Alexander W. A. Kellner,
David Rivarola, Sergio Davila, and
Marilyn Fox

of Los Angeles County

Natural History Museum

Serial
Publications
ol THE
Natural History
Museum ol
Los Angeles
County

Scientific

Publications

Committee

John Heyning, Deputy Director
for Research and Collections
John M. Harris, Committee Chairman
Brian V. Brown
Kenneth E. Campbell
Kirk Fitzhugh
Karen Wise

Robin A. Simpson and K. Victoria Brown,

Managing Editors

The scientific publications of the Natural History Museum
of Los Angeles County have been issued at irregular in-
tervals in three major series; the issues in each series are
numbered individually, and numbers run consecutively, re-
gardless of the subject matter.

# Contributions in Science, a miscellaneous series of tech-
nical papers describing original research in the life and
earth sciences.

# Science Bulletin, a miscellaneous series of monographs
describing original research in the life and earth sci-
ences. This series was discontinued in 1978 with the
issue of Numbers 29 and 30; monographs are now
published by the Museum in Contributions in Science.

# Science Series, long articles and collections of papers on
natural history topics.

Copies of the publications in these series are sold through
the Museum Book Shop. A catalog is available on request.

The Museum also publishes Technical Reports, a miscel-
laneous series containing information relative to scholarly
inquiry and collections but not reporting the results of
original research. Issue is authorized by the Museum’s Sci-
entific Publications Committee; however, manuscripts do
not receive anonymous peer review. Individual Technical
Reports may be obtained from the relevant Section of the
Museum.

Natural History Museum
of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

Printed at Allen Press, Inc., Lawrence, Kansas
ISSN 0459-8113

Cranial Morphology of Pterodaustro guinazui
(Pterosauria: Pterodactyloidea) from the
Lower Cretaceous of Argentina

Luis M. Chiappe,1 Alexander W. A. Kellner,2
David Rivarola,3 Sergio Davila,3
and Marilyn Fox4

ABSTRACT. With hundreds of filamentlike teeth in its mandibles and several other unique cranial features,
the Argentine Early Cretaceous Pterodaustro guinazui ranks among the most specialized of pterosaurs.
Based on newly collected specimens, this study provides a detailed description of the peculiar skull mor-
phology of Pterodaustro and discusses its phylogenetic position within pterosaurs. An overview of the
stratigraphy, sedimentology, and chronology of the Lagarcito Formation from which Pterodaustro comes
is provided, along with an interpretation of the paleoenvironment. Cranial morphology corroborates the
sister-taxon relationship between Pterodaustro and the Late Jurassic Ctenochasma proposed by most pre-
vious authors.

RESUMEN. Con cientos de dientes filamentosos en sus mandibulas y varios otros caracteres craneanos
distintivos, Pterodaustro guinazui del Cretacico temprano de Argentina, es uno de los pterosaurios mas
especializados. Sobre la base de ejemplares recientemente colectados, este estudio provee una description
detallada de la peculiar morfologia craneana de Pterodaustro y analiza sus relaciones filogeneticas dentro
de los pterosaurios. Tambien se provee una resena estratigrafica, sedimentologica, y cronologica de la
Formation Lagarcito — la formation portadora de Pterodaustro — junto con una interpretation del paleoam-
biente. La morfologia craneana corrobora la relation de grupo hermano entre Pterodaustro y Ctenochasma,
del Jurasico tardio, que fuera propuesta por la mayoria de los autores previos.

INTRODUCTION

Paleontological expeditions led by J.F. Bonaparte in
the late 1960s and early 1970s made a significant
contribution to the knowledge of pterosaur evolu-
tion with the discovery of the spectacular pterodac-
tyloid Pterodaustro guinazui Bonaparte, 1970
(original spelling emended by Wellnhofer [1978]
following the guidelines of the ICZN). Collected
from the banks of a creek cutting through beds of
the Lagarcito Formation at a site known today as
Loma del Pterodaustro, in what is now the Parque
Nacional Sierra de las Quijadas (Fig. 1), Ptero-
daustro was the first pterosaur to be found in Ar-

1. Department of Vertebrate Paleontology, Natural
History Museum of Los Angeles County, 900 Exposition
Boulevard, Los Angeles, California 90007, USA.

2. CNPq Fellow, Departamento de Geologia e Palen-
tologia, Museu Nacional/Universidade Federal do Rio de
Janeiro, Quinta da Boa Vista, Sao Cristovao, Rio de Ja-
neiro, RJ 20.940-040, Brazil.

3. Departamento de Geologia, Universidad Nacional de
San Luis, Chacabuco y Pedernera, (5700) San Luis, Ar-
gentina.

4. Department of Vertebrate Paleontology, Peabody
Museum of Natural History, Yale University, New Haven,
Connecticut 06520, USA.

Contributions in Science, Number 483, pp. 1-19
Natural History Museum of Los Angeles County, 2000

gentina. When discovered, Pterodaustro represent-
ed only the second record of pterosaurs from South
America (Price, 1971). Most importantly, Ptero-
daustro provided strong evidence that pterosaurs
had evolved a filter-feeding morphology uncommon
among tetrapods.

Twenty-five years after Bonaparte’s exploratory
expeditions to the Mesozoic deposits of the central
Argentine province of San Luis, three large-scale
excavations (in 1994, 1996, and 1998; see Chiappe
et al., 1998a, b) were conducted at the quarry in
the Loma del Pterodaustro where the first speci-
mens of Pterodaustro were collected (Fig. 1). These
excavations produced hundreds of skeletal remains
of this pterosaur, including adult, juvenile, and ne-
onate specimens (Chiappe et ah, 1998b).

Despite the availability of material of Pterodaus-
tro (even before the newly collected specimens) and
the fact that this multitoothed pterosaur has con-
sistently been used as a startling example of mor-
phological specialization within this group (e.g.,
Benton, 1990; Wellnhofer, 1991; Chiappe and
Chinsamy, 1996), the cranial anatomy of Ptero-
daustro has received little attention beyond early
descriptions by Bonaparte (1971) and Sanchez
(1973). The recent expeditions to Loma del Ptero-

0 10 20 Km

Figure 1 Geographic distribution of the Lagarcito Formation and location of the Loma del Pterodaustro fossil site

daustro collected several skulls and jaws of adult
individuals. These are described here in detail. We
also review previous anatomical data on this taxon
in light of these specimens and discuss the phylo-
genetic position of Pterodaustro among pterodac-
tyloids.

Institutions are abbreviated as follows: PVL, Sec-
cion Paleontologia de Vertebrados, Instituto Miguel
Lillo (San Miguel de Tucuman, Argentina); and
MHIN-UNSL-GEO, Museo de Plistoria Natural,
Universidad Nacional de San Luis (San Luis, Ar-
gentina).

GEOLOGICAL SETTING OF THE
LAGARCITO FORMATION

The Lower Cretaceous rocks of the Argentine Prov-
ince of San Luis form a geotectonic unit known as
the San Luis Basin (Flores and Criado Roque,
1972). This basin has been interpreted as a rift ba-
sin resulting from cortical stresses generated during

the breakdown of Gondwana (Ramos, 1990). Lith-
ologically, this basin corresponds to a typical con-
tinental sequence of red beds exceeding 1,000 me-
ters in thickness. These rocks crop out in several
ranges grouped under the name Cordon de Serran-
ias Occidentales, which extend in a north-south di-
rection over roughly 300 kilometers (Fig. 1).

Stratigraphically, the Lower Cretaceous rocks of
San Luis are divided into the Gigante Group and
the Lagarcito Formation (Flores and Criado Roque,
1972; Fig. 2), which comprise the entire duration
of two cycles of infilling of a continental basin (Ri-
varola, 1994). These cycles represent two deposi-
tional megasequences involving environments that
range from alluvial fans associated with alluvial
plains to fluvial plains and lacustrine environments.
Rocks of the Gigante Group form most of the two
megasequences, and they are composed of con-
glomerates, sandstones, claystones, and evaporites.
The Lagarcito Formation forms the top section of

2 ■ Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull

Claystones - Basalts —3 [

Siltstones J

Fine Sandstones

Medium Sandstones

Conglomerate

Claystones -1 I
Siltstones — I
Fine Sandstones

Legend

Claystone

\^r-J Siltstone
iTiMl Sandstone

llvXl Conglomerate

|v vl Basalt

Irregular
1 bedding
I I Laminated

Massive
F.i Facies 1
^ Fluid scape
Ripples
”2- Slumps
k Intraclasts
Bioturbation
Pterodaustro
Pleuropholidae
Semionotidae
• Conchostraca
~ Trace fossils
$ Plant remains
^ Anura
O Ostracoda

Figure 2 Stratigraphic section and fossil occurrences of the Loma del Pterodaustro fossil site (from Chiappe et ah, 1998b)

the second megasequence, and it is formed by fine
sediments developed under fluvio-lacustrine envi-
ronments associated with limited development of
aeolian dunes.

In the Sierra de Las Quijadas, the Lagarcito For-
mation is best exposed at Quebrada de Hualtaran
(Fig. 1). In this area, 45 meters of continuous sed-
iments of sandstones and mudstones, representing
at least three fluvio-lacustrine sequences, rest over
a basaltic flow dated between 107.4 and 109.4 Ma
(Yrigoyen, 1975).

Most fossils from the Lagarcito Formation have
been excavated at a small site (—50 m2). This site
has become known as Loma del Pterodaustro
(Chiappe et ah, 1995). The Loma del Pterodaustro
fossil site corresponds to the basal 8 meters of the
section of the Lagarcito Formation at Quebrada de
Hualtaran (Fig. 2). A detailed sedimentological
study of these deposits has been presented else-
where (Chiappe et ah, 1998a). Three lithofacies can
be recognized from base to top: Facies 1, inversely
graded, massive sandstones to massive, matrix-sup-
ported conglomerates with lenticular geometry and
disordered fabric; Facies 2, fine-grained sandstone
with a flat top and base, and asymmetric ripples;
and Facies 3, massive to laminated claystones, silt-

stones, and very fine sandstones, which represent a
thickening and coarsening upward sequence. Facies
1-3 have been interpreted as debris flows, sheet-
floods deposited in a sand-flat near the shore, and
a typical lake sequence, respectively. With very few
exceptions, all fossils from Quebrada de Hualtaran
come from Facies 3 (Chiappe et ah, 1998a, b). This
facies has been subdivided into three subfacies of
low-energy deposits (Fig. 2). One of them (F3.1;
Fig. 2) is composed of laminated, very fine sedi-
ments entombing the majority of fossils. This su-
bfacies was interpreted as forming in the offshore
portion of a lake.

BIOTA, PALEOENVIRONMENT,

AND CHRONOLOGY OF

THE LAGARCITO FORMATION AT

LOMA DEL PTERODAUSTRO

The fine sandstones and claystones of the Lagarcito
Formation at Loma del Pterodaustro have provided
abundant fossil remains (Chiappe et ah, 1995,
1998a, b). These include a diverse array of trace
fossils, plant remains, conchostracans, ostracods,
and various vertebrates. Among the vertebrate fau-
na are semionotid and pleuropholid fishes, anurans,

Contributions in Science, Number 483

Chiappe et ah: Pterodaustro Skull ■ 3

and abundant pterosaur remains, most of which are
probably of Pterodaustro. Preservation of delicate
structures such as the needlelike mandibular teeth
of Pterodaustro and the imprints of stems and re-
productive plant structures led to the classification
of these beds as a Konservat Lagerstatte sensu Sei-
lacher et al. (1985) (see also Seilacher, 1990; Chiap-
pe et ah, 1995, 1998a).

The facies association of the lower section of the
Lagarcito Formation, along with the absence of ev-
idence of subaerial exposure and evaporite levels,
suggest that rocks at Loma del Pterodaustro cor-
respond to a fluvio-lacustrine sequence of long du-
ration. Preservation of laminations in subfacies 3.1
indicates that the lake was at least periodically ther-
mally stratified, and it may have developed an an-
oxic bottom that prevented destruction of the lam-
inations and favored preservation of delicate struc-
tures.

The lower section of the Lagarcito Formation, at
Quebrada de Hualtaran, is interpreted as a com-
plete sequence of transgression and expansion of a
perennial lake over an alluvial sandy flat, followed
by its gradual infilling and shallowing during a
high-stand period of the lacustrine system. The pre-
dominant climate during deposition of the Lagar-
cito Formation is interpreted as semiarid and sea-
sonal (Chiappe et ah, 1998a).

Originally, the Lagarcito Formation was placed
in the Tertiary (Flores, 1969). After the discovery
of Pterodaustro guinazui, Bonaparte (1970) placed
this unit within the Upper Jurassic on the basis of
similarities between this pterosaur and Late Jurassic
pterodactyloids. Subsequently, with the discovery
of an Aptian-Albian palynoflora in the underlying
La Cantera Formation (Gigante Group) and the
Early Cretaceous dates for the basalts at Quebrada
de Hualtaran, Yrigoyen (1975) placed the Lagar-
cito Formation within the Upper Cretaceous, and
Bonaparte (1978) allocated those strata to the Low-
er Cretaceous. Recent interpretations, combining
sedimentological, stratigraphical, and paleontolog-
ical data, have adjusted the chronology of this lith-
ostratigraphic unit, supporting an Albian age (see
Chiappe et ah [1998a] for a more extensive discus-
sion of the age of these beds).

CRANIAL ANATOMY OF
PTERODAUSTRO GUINAZUI

This study is mostly based on recently collected ma-
terial, including a skull and jaw (MHIN-UNSL-
GEO-V-57), another skull and jaw missing their
rostral halves (MHIN-UNSL-GEO-V-135), and an
isolated, nearly complete jaw (MHIN-UNSL-GEO-
V-175). As in other known skulls of Pterodaustro,
the new crania and jaws are flattened, preserved
essentially in two dimensions. Many of the bones

are broken, thus complicating identification be-
tween sutures. In MHIN-UNSL-GEO-V-57 and
MHIN-UNSL-GEO-V-135 the skull is exposed on
its left side; only a few right bones are visible. In
MHIN-UNSL-GEO-V-57 the left mandible is ex-
posed laterally, whereas in MHIN-UNSL-GEO-V-
135 the right one is exposed. MHIN-UNSL-GEO-
V-175 exposes its left side.

SKULL

The skull of Pterodaustro is characterized by a re-
markably long, slender, and upwardly curved pre-
orbital region, which comprises more than 85 per-
cent of the skull length (Figs. 3, 4). MHIN-UNSL-
GEO-V-57 (Fig. 3) has a total cranial length of
almost 29 cm. This is approximately 20 percent
larger than that of PVL-3860, the most commonly
figured specimen of Pterodaustro (Sanchez, 1973;
Bonaparte, 1978; Wellnhofer, 1991). MHIN-
UNSL-GEO-V-135 (Fig. 5) is somewhat (—10 per-
cent) larger than MHIN-UNSL-GEO-V-57. The na-
soantorbital fenestra in this taxon is comparatively
small (Kellner, 1995), reaching between 10 and 12
percent of the total skull length. Because the skull
is flattened, no detailed information regarding the
temporal openings is available.

In lateral view, the snout curves gently upward
(Figs. 3-5). A long premaxilla and maxilla, whose
suture is not discernible in any of the available
specimens, form the snout. As in other pterosaurs
(Wellnhofer, 1978), the slender premaxilla forms
the entire dorsal margin of the snout, approaching
the rostral margin of the orbit. In MHIN-UNSL-
GEO-V-57 it is uncertain whether the premaxilla
was toothed or not because the most rostral end of
the snout is not preserved. Unfortunately, none of
the remaining known skulls clarify this issue. In
MHIN- UNSL-GEO-V-57 both premaxillae seem
to be fused to each other for most of their length.
Sanchez (1973) reported a thin, caudal tongue of
the maxilla contacting the nasal and excluding the
premaxilla from the nasoantorbital fenestra (Figs.
6 A, 7). This condition, which is different from that
of most pterosaurs (e.g., Wellnhofer, 1978; Well-
nhofer and Kellner, 1991), has not been corrobo-
rated by either of the new specimens, in which the
premaxilla forms the dorsal margin of the nasoan-
torbital fenestra (Figs. 6B, 8, 9). Also differing from
previous interpretations (Sanchez, 1973), the pre-
maxilla is extended farther caudally over the rostral
half of the orbit (cf. Figs. 6 A, B, 8).

The maxilla forms most of the lateral surface of
the snout (Figs. 3-5, 7). This bone tapers rostrally
from the rostral margin of the nasoantorbital fe-
nestra. In all available specimens the suture be-
tween this bone and the rostral end of the jugal is
not clear. In fact, these two bones appear to be

Figure 3 Skull and jaws of Pterodaustro guinazui (MHIN-UNSL-GEO-V-57) in left lateral view. Scale bar = 5 centimeters

4 ■ Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull

Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull ■ 5

6 ■ Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull

SO, supraorbital; SQ, squamosal; SR, sclerotic ring

Figure 5 Skull and jaws of Pterodaustro guinazui (MHIN-UNSL-GEO-V-135). Skull in left lateral view; mandible in
right lateral view

fused to each other. However, a thin tongue of the
maxilla projects caudally beneath the jugal, reach-
ing the caudal margin of the nasoantorbital fenestra
(Figs. 6B, 8). The maxilla bears hundreds of tiny
teeth, which in MHIN-UNSL-GEO-V-57 extend
from the rostral end of the maxilla to near the ros-
tral margin of the nasoantorbital fenestra (Fig. 4).
Thus, the caudal extension of the maxillary tooth
row of this specimen extends much more than that
illustrated by Sanchez (1973) in her skull recon-
struction (Fig. 7).

The dorsocaudal corner of the nasoantorbital fe-
nestra is formed by a subtriangular nasal (Figs. 6-
9). This element has a comparatively long, thin ros-
tral process that reaches the center of the nasoan-
torbital fenestra, underlying the premaxilla (Figs.
6B, 9). Ventrally, the nasal tapers into a thin pro-
cess that fails to reach the jugal, although it ends
very close to it (Sanchez, 1973). The contact of the

right nasal and jugal in MHIN-UNSL-GEO-V-57
(Fig. 6B) is not natural and can be regarded as a
result of deformation during crushing. The mor-
phology of the ventral nasal process is similar to
that of archaeopterodactyloids ( sensu Kellner,
1996, 1997) such as Pterodactylus Cuvier, 1809,
and Germanodactylus Young, 1964. All more de-
rived pterodactyloids, which are members of the
Dsungaripteroidea (Young, 1964; Kellner, 1996),
have either displaced this nasal process medially
(e.g., Anhanguera Campos and Kellner, 1985) or
have almost lost it (e.g., Quetzalcoatlus Lawson,
1975) (see Kellner and Langston, 1996).

The lacrimal of Pterodaustro is wedged between
the nasal, prefrontal, and jugal. This bone has three
distinct processes: a short rostral one, a longer and
wider caudal one, and a hook-shaped ventral one
(Figs. 6B, 8, 9). Although now flattened, in life the
ventral process probably extended slightly lateral to

Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull ■ 7

QJ?

Figure 6 Camera lucida drawings of the postorbital region of Pterodaustro guinazui. A, from Sanchez (1973) (PVL-
3860). B, MHIN-UNSL-GEO-V-57. Abbreviations: ITF, infratemporal fenestra; OR, orbit; PF, prefrontal; QJ, quadra-
tojugal; SA, surangular; STF, supratemporal fenestra. Other abbreviations as in Figure 4

the orbit, as seen in some other pterosaurs known
from less distorted specimens (e.g., Anbanguera). In
MHIN-UNSL-GEO-V-57 (Fig. 6B), this process is
significantly more robust than the process illustrat-
ed by Sanchez (1973; Fig. 6A), which agrees more
with that of MHIN-UNSL-GEO-V-135 (Fig. 9).
MHIN-UNSL-GEO- V- 135 clearly shows that the
bone identified as a prefrontal by Sanchez (1973)

is, in fact, the dorsal exposure of the caudal process
of the lacrimal (Fig. 9).

Rostrally, both frontals are united by a suture
that becomes indistinguishable in the caudal half.
Caudally, the frontals are wedged between the post-
frontals and parietals (Figs. 6, 8).

The parietals in both specimens are very crushed,
making their edges difficult to interpret, particular-

8 ■ Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull

Figure 7 Sanchez’s (1973) cranial reconstruction of Pterodaustro guinazui. Abbreviations as in Figures 4 and 6

ly in their caudal region. As far as they can be ob-
served, both elements are fused to each other along
the midline. Although the boundaries of this bone
were not clearly illustrated by Sanchez (1973; Fig.
7), the new specimens indicate that the parietal
forms the medial margin of the supratemporal fe-
nestra (Fig. 6B), as in most archosaurs (Romer,
1956). Interestingly, the caudal region of the pari-
etals of all available specimens, including those pre-
viously described by Bonaparte (1971) and Sanchez
(1973), is slightly extended as a thin bony lamina
(Figs. 6, 8, 9). Although some of this appearance
results from crushing, it cannot be ruled out that
the parietals projected caudally as a small crest.
Even if this interpretation is correct, this parietal
crest would have been very small and not as devel-
oped as in some other pterosaurs (e.g., Gallodac-
tylus Fabre, 1974).

Caudal to the lacrimal, there is a flattened bone,
here interpreted as the prefrontal. Based on MFilN-
UNSL-GEO-V-57, this bone would correspond to
the rostral portion of what Sanchez (1973) inter-
preted as the supraorbital (Fig. 6A). The slender
supraorbital does indeed line the dorsal margin of
the orbit and precludes the frontal from the orbital
margin (Sanchez, 1973), but it does not appear to
extend rostrally to reach the lacrimal (Fig. 6B).

In MHIN-UNSL-GEO-V-57, the postfrontal is a
suboval element that contacts the frontal medially
and the parietal caudally and is partially overlain
by the postorbital (Fig. 6B). A caudally oriented
process of the postfrontal, as illustrated by Sanchez
(1973), was not observed in the new specimens.

The postorbital is a triradiate element (Fig. 6B).
Its dorsal process is slightly expanded and abuts
against the parieto-postfrontal suture. The ventral
process overlays the dorsocaudal process of the ju-
gal, whereas the caudal process overlies the lateral
surface of the squamosal. The unusual condition of
having the postorbital completely separated from
the frontal by the postfrontal (Figs. 6B, 8), also in-
dicated by Sanchez (1973; Fig. 6A), distinguishes

Pterodaustro from other pterosaurs (Wellnhofer,
1978; Wellnhofer and Kellner, 1991).

The jugal is a conspicuous bone that forms the
ventrocaudal and ventral margins of the nasoan-
torbital fenestra and the orbit, respectively, and the
rostroventral corner of the infratemporal fenestra
(Fig. 6A). The jugal has four distinct processes
(Figs. 8, 9). The rostral process forms the ventral
margin of the nasoantorbital fenestra and, in
MHIN-UNSL-GEO-V-135, projects rostrally be-
yond the nasoantorbital fenestra for about 9 mm
(Fig. 9). This extension is shorter than interpreted
by Bonaparte (1971). The dorsal process contacts
the lacrimal, forming a bony bar that separates the
orbit from the nasoantorbital fenestra. This process
is robust and tapers to a sharp point (Fig. 9). Cau-
dally, the jugal forks into dorsocaudal and ventro-
caudal processes. The former is overlain by the
postorbital and takes part in the rostral margin of
the infratemporal fenestra. The second partially
overlies the quadrate (displaced in MF3IN-UNSL-
GEO-V-57).

The tetraradiate jugal of all known Pterodaustro
specimens (Figs. 6, 9) differs markedly from the tri-
radiate shape of the jugal of other pterodactyloids,
including Pterodactylus and Ctenocbasma Meyer,
1851, resembling more the morphology of the jugal
in some basal pterosaurs (Wellnhofer, 1991). The
difference appears to be mainly on the ventrocaudal
portion, where the jugal of Pterodaustro bears an
extended process overlying the quadrate (Figs. 6B,
8). This contact between the quadrate and the jugal
is known only for this taxon. Furthermore, Ptero-
daustro also differs from some derived pterodac-
tyloids (e.g., Anhanguera, Pteranodon Marsh,
1876) in that its jugal lacks a pronounced caudo-
dorsal ridge near the nasoantorbital fenestra.

The quadratojugal is very difficult to discern in
most specimens. In MFiIN-UNSL-GEO-V-57, a
thin and incomplete bone, displaced ventrally to the
quadrate, is tentatively identified as the left quad-
ratojugal (Fig. 6B). According to Sanchez (1973),

Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull ■ 9

Figure 8 Nasoantorbital, orbital, and postorbital regions of Pterodaustro guinazui (MHIN-UNSL-GEO-V-57), in left
lateral view. A, general view. B, detail of the jugal and surrounding bones

this element forms all of the ventral margin of the
infratemporal opening. However, such a configu-
ration has not been observed in any other ptero-
dactyloid (Wellnhofer, 1978).

The quadrate is not well preserved in any of the
new specimens. This bone is rostroventrally tilted
(more than 150° relative to the ventral margin of
the skull; Fig. 6), a feature common to all Ar-
chaeopterodactyloidea (Kellner, 1996, 1997), as
well as the Azhdarchidae (Unwin and Junchang,
1997). Rostrally, the main body of the quadrate
projects mediad as a broad pterygoid flange. The

proximal end of the quadrate appears to articulate
with the squamosal.

The squamosal has an expanded, ventrolaterally
rounded body (Figs. 6B, 8, 9). As shown by San-
chez (1973), this bone forms the entire ventral
margin of the supratemporal fenestra (Fig. 6), re-
stricting the postorbital to the rostral margin of
this arcade. The squamosal has a thin, tapering
caudomedial process that abuts the braincase (Fig.
6A). Rostroventrally, the squamosal bears a sharp,
thin process that runs through the proximal half of
the caudal margin of the quadrate.

10 ■ Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull

Figure 9 A, nasoantorbital, orbital, and postorbital regions of Pterodaustro guinazui (MHIN-UNSL-GEO-V-135), in left
lateral view. B, interpretive drawing. Abbreviations as in Figures 4 and 6; J(r) refers to the right jugal

Aside from the parietals, the remaining elements
of the braincase appear to be completely fused,
lacking discrete sutures. Yet a truncated, subtrian-
gular bone extends laterally from the braincase in
MHIN-UNSL-GEO-V-57 (Fig. 6B). This is most
likely the paraoccipital process of the opisthotic. A
small window between this bone and the squamo-
sal may be the posttemporal fenestra (Fig. 6B). Al-
though Sanchez (1973) did not identify this bone,
it is very likely that the bone positioned caudo-
ventral to the squamosal in PVL-3860 might also
be the opisthotic.

Portions of the palatal bones appear in MHIN-

UNSF-GEO-V-57 and MHIN-UNSL-GEO-V-135,
but their compression ventral to the maxilla at the
level of the nasoantorbital fenestra prevents recov-
ering any information.

The orbit encloses a sclerotic ring. The number
of plates forming the sclerotic ring is uncertain. In
MHIN-UNSL-GEO-V-57, at least eight or nine
plates can be distinguished (Fig. 6B). Pterosaur scle-
rotic rings have been reported previously, but com-
plete sclerotic rings are known only for a handful
of specimens. In Rhampborhyncbus (Meyer, 1847),
sclerotic rings are formed by 13 to 15 elements and
in Pterodactylus by about 20 elements (Wellnhofer,

Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull 111

Figure 10 Detail of the midportion of the maxilla of Pterodaustro guinazui (MHIN-UNSL-GEO-V-57), in left lateral
view

1978). In more derived forms such as Pteranodon
(Eaton, 1910; Bennett, 1991) and anhanguerids
(specimen in Iwaki Coal and Fossil Museum, Ja-
pan), 12 or 13 sclerotic plates are present.

MANDIBLE

Like the snout, the mandible is curved and slender,
tapering rostrally (Figs. 3, 4, 7). Most of the avail-
able information is from the lateral surface. The
mandible follows the extension of the preorbital re-
gion of the skull, making the lower jaw of Ptero-
daustro proportionally longer than in other ptero-
saurs. The dentary extends almost the entire length
of the mandible, and its medial dentigerous margin
is higher and thicker than the lateral one. In
MHIN-UNSL-GEO-V-57, for example, the dentary
margin lateral to the tooth row is one-third to one-
quarter shorter than the medial one.

The dentary bears teeth throughout its length
(Figs. 4, 6B). This condition differs from that of
other archaeopterodactyloids such as Ctenochas-
ma, Germanodactylus, Gallodactylus, and Ptero-
dactylus, in which the dental row ends more rostral
to the caudal end of the dentary. For example,
whereas Ctenochasma, Gnathosaurus Meyer, 1834,
and Pterodactylus have teeth along less than 65
percent of the mandibular length (Wellnhofer,
1978), Pterodaustro has teeth along nearly 90 per-
cent of its mandibular length.

The region of the surangular is not well pre-
served in any of the new specimens and the bound-
ary between this bone and the dentary is unclear.
In MHIN-UNSL-GEO-V-57, the caudal portion of
the surangular is comparatively thick, forming the
rostrodorsal part of the articular surface for the
quadrate. Toward the rostral part, the surangular
appears to form a process that contributes to the
dorsocaudal margin of the mandibular ramus. This
process seems to end before the first mandibular
tooth, where it interlocks with the dentary.

The angular is exposed laterally (Fig. 6B), but its
boundaries with the surrounding bones, particular-
ly with the articular, are difficult to interpret in the
available material. Rostrally, this bone tapers under
the ventral margin of the dentary. Based on MHIN-
UNSL-GEO-V-57 (Fig. 6B), the dorsal extension of
the angular is greater than interpreted by Sanchez
(1973; Fig. 7).

The articular forms the caudal tip of the man-
dible (Figs. 6B, 7). This bone has a long, thin wedge
underneath the angular. Unfortunately, the lateral
exposure of this bone prevents recovering infor-
mation about the articular facet of the mandible.

DENTITION

The upper dentition is formed by hundreds of mi-
nute teeth of equal size (Figs. 4, 10, 11). These teeth
have spatulate crowns and thin, conical bases. In-
terestingly, the upper teeth are not set in alveoli
( contra Sanchez, 1973), nor in a longitudinal
groove. Instead, they are joined to the lateral sur-
face of the recessed maxillary dentigerous margin
(Figs. 10, 11). Thus, their attachment to the maxilla
must have been by means of individual ligaments
or, more likely, by a supporting soft-tissue structure
that lined the dentigerous margin of the upper jaw.
Interestingly, dorsal to most maxillary teeth, a dor-
sorostrally oriented row of tiny ossicles is present,
usually composed of four elements (Figs. 10, 11).

The mandibular teeth are very close to each other
(Figs. 12-14). In cross section, they are oval to sub-
elliptical (Chiappe and Chinsamy, 1996) and con-
siderably thicker rostrally than in the middle and
caudal sections of the mandible. The teeth are set
in a groove along most of the jaw (Fig. 13), al-
though shallow individual alveoli develop in the
rostral end of the dentary (Fig. 12).

The external surface of the mandibular teeth is
smooth (Figs. 13, 14). From histological sections
we determined that the mandibular teeth are
formed by a peripheral, thin layer of nonprismatic

12 ■ Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull

1 mm

Figure 11 Camera lucida drawing of the maxillary teeth and associated rows of ossicles of Pterodaustro guinazui (MHIN-
UNSL-GEO-V-57), in left lateral view

Figure 12 Detail of the rostral end of the dentary of Pter-
odaustro guinazui (MHIN-UNSL-GEO-V-57), in left lat-
eral view. Scale bar = 1 centimeter

enamel and an inner core of dentine surrounding a
central pulp cavity (Chiappe and Chinsamy, 1996).
These sections have not shown incremental lines in
either the enamel or the dentine, contrary to teeth
of Anhanguera (Chinsamy and Kellner, 1996).

SYSTEMATIC REMARKS

The first remains of Pterodaustro guinazui were all
isolated specimens providing limited anatomical in-
formation about this pterosaur. A humerus (PVL-
2571) was established as the holotype and a few
other elements as the hypodigm (Bonaparte, 1970).
On the basis of the similarity between these ele-
ments and those of Pterodactylus, Bonaparte
(1970) assigned Pterodaustro to the Pterodactyli-
dae.

With the discovery of its peculiar skull and man-
dible, Pterodaustro was placed in its own family,
Pterodaustriidae (Bonaparte, 1971), particularly
because of its unique dentition and extreme exten-
sion of its preorbital region. Bonaparte (1971) con-
sidered that the Pterodaustriidae was related both
to the Pterodactylidae and the Ctenochasmatidae.
Further cranial material of Pterodaustro was stud-
ied by Sanchez (1973), who arrived at the same
conclusions as Bonaparte. The systematic relation-
ships proposed by Bonaparte (1971) were adopted
by later authors (e.g., Casamiquela and Chong-
Diaz, 1980; Wellnhofer, 1991), although Wellnho-
fer (1978: fig. 32) considered Pterodaustro more
closely related to Ctenochasma.

In a cladistic analysis of pterosaur cervical ver-
tebral morphology, Howse (1986: figs. 11, 12)
placed Pterodaustro in a polytomy with Pterodac-
tylus antiquus Soemmerring, 1812, Pterodactylus
longicollum Meyer, 1854, and Ctenochasma

Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull ■ 13

14 ■ Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull

(Fig. 15). Howse (1986) correctly pointed out that
all these pterosaurs have elongated midcervical ver-
tebrae with low neural spines, although he failed to
acknowledge the presence of this condition in Ger-
manodactylus and Gallodactylus suevicus (Quen-
stedt, 1855).

Bennett (1994) regarded Pterodaustro as a basal
pterodactyloid that formed a trichotomy with Pter-
odactylus kochi Wagner, 1837, and all other pter-
odactyloids (Fig. 15). This interpretation disagrees
with all previous hypotheses by setting Pterodaus-
tro apart from Ctenocbasma, a claim that Bennett
(1994) supported by citing the absence of a pre-
maxillary sagittal crest in Pterodaustro. Pterodaus-
tro does not have a premaxillary sagittal crest, but
such a crest is also absent in several other ptero-
dactyloids (e.g., Gallodactylus, Pteranodon, Nyc-
tosaurus Marsh, 1876), which Bennett (1994) in-
cluded in his grouping of all pterodactyloids other
than Pterodaustro and Pterodactylus kochi. Fur-
thermore, several characters presented by Bennett
(1994) that diagnose subsequent, more exclusive
nodes cannot be observed in Pterodaustro and are
therefore equivocal, at least relative to this taxon
(e.g., cervicalization of dorsal vertebrae and pres-
ence of helical jaw joint).

More recently, Kellner (1996) presented a com-
prehensive study of pterosaur interrelationships.
This author recognized a monophyletic group, the
Archaeopterodactyloidea, consisting of a clade
formed by Pterodactylus and Germanodactylus,
and its sister-group, an unnamed clade formed by
Ctenochasmatidae and Gallodactylidae (Fig. 15).
Kellner (1996) included Pterodaustro within the
Ctenochasmatidae, as the sister-taxon of Cteno-
chasma (Fig. 15). All the synapomorphies diagnos-
ing the Archaeopterodactyloidea (e.g., rounded
caudal end of the skull, resulting in a ventral dis-
placement of the squamosal; quadrate strongly in-
clined caudally, with an angle of about 150° relative
to the ventral margin of the skull; elongated mid-
cervical vertebrae; neural spines of the midcervical
vertebrae low and bladelike; see Kellner, 1996,
1997) are present in Pterodaustro.

In Kellner’s analysis, the close relationship of
Pterodaustro and Ctenochasma is supported by a
single synapomorphy: the presence of a large num-
ber of teeth (more than 150; Kellner, 1996, 1997).
However, the rostral extension of the surangular of
Pterodaustro and Ctenochasma is smaller than that
of other archaeopterodactyloids and may be anoth-
er synapomorphy of these taxa.

A close relationship between Pterodaustro and
Ctenochasma was also recently supported by Un-
win and Junchang (1997), who included these and
other Late Jurassic-Early Cretaceous pterosaurs

(e.g., Gnathosaurus, Huanhepterus Dong, 1982;
Cearadactylus Leonardi and Borgomanero, 1985)
within the Ctenochasmatidae (Fig. 15). However,
these authors did not mention any specific syna-
pomorphy uniting these taxa but simply pointed to
the filter-feeding specializations usually inferred for
them. In contrast to Kellner’s (1996) hypothesis,
Unwin and Junchang (1997) considered the Cten-
ochasmatidae to be the sister-group of Pterodacty-
lus, regarding Gallodactylus as a more basal mem-
ber of the Ctenochasmatoidea — the higher taxon
used by Unwin and Junchang (1997) to group
Ctenochasmatidae, Pterodactylus, and Gallodacty-
lus. Unfortunately, these authors did not provide
evidence for the proposed relationships within their
Ctenochasmatoidea.

However, support for the sister-group relation-
ship of Ctenochasmatidae and Gallodactylidae is
provided by the presence of a concave dorsal mar-
gin of the skull (Kellner, 1996, 1997). In all other
pterosaurs, the dorsal margin of the skull is either
straight or convex. According to the reconstruction
presented by Bennett (1991), the dorsal margin of
the skull in Pteranodon is also concave. However,
it must be noted that few complete skulls are
known and that in some of these the dorsal margin
is essentially straight (Bennett, 1994: fig. 1). If the
dorsal margin of the skull of Pteranodon was in-
deed concave, it is probable that this condition was
achieved independently (Kellner, 1996).

Kellner (1996) diagnosed the monophyletic
group formed by Pterodactylus and Germanodac-
tylus on the basis of their particular dentition (more
than 15 peglike teeth) and the presence of a straight
lateral process of the nasal that is not connected to
the maxilla. MHIN-UNSL-GEO-V-57 has a
straight lateral process of the nasal, suggesting that
this apomorphic character may be a synapomorphy
of a more inclusive clade, most likely the Archaeo-
pterodactyloidea. Because this process is very deli-
cate, its supposed absence in Gallodactylus and
Ctenochasma (Kellner, 1996) may result from pres-
ervational factors.

The skull of Pterodaustro exhibits numerous
characters of its own that are clear autapomorphies
of this taxon. The most obvious are in the denti-
tion. The spatulate teeth of the maxilla of Ptero-
daustro differ from those of all other pterosaurs
(Bonaparte, 1971; Sanchez, 1973). In addition, the
fact that the maxillary teeth are not set in alveoli
and that they are associated with a row of ossicles
are other autapomorphic characters. Another dis-
tinct autapomorphy of Pterodaustro is the remark-
able extension of its preorbital region, which is not
known in other pterosaurs (Bonaparte, 1971; San-
chez, 1973). As a consequence of this preorbital

Figure 13 Stereoelectromicrograph of the mandibular teeth of Pterodaustro guinazui (MHIN-UNSL-GEO-V-55) showing
their insertion in a dentary groove with lower (A) and higher (B) magnification. In A and B, scale bars equal 2 and 1
millimeters, respectively

Contributions in Science, Number 483

Chiappe et ah: Pterodaustro Skull ■ 15

Figure 14 Stereoelectromicrograph of the mandibular teeth of Pterodaustro guinazui (MHIN-UNSL-GEO-V-55). Note
the smoothness of the enamel’s surface. Scale bar = 1 millimeter

extension, the nasoantorbital fenestra of Ptero-
daustro occupies only 10-12 percent of the total
length of the skull. In contrast, the nasoantorbital
fenestra constitutes 15-16 percent of the length of
the skull in Ctenochasma and 18-19 percent in
Pteranodon (excluding the cranial crest; see Ben-
nett, 1991: fig. 2).

Furthermore, unlike any other pterosaur, in Pter-
odaustro the postorbital is separated from the fron-
tal, and the quadratojugal appears to form the ven-
tral margin of the infratemporal opening (Sanchez,
1973), which in other pterodactyloids is mostly
formed by the quadrate (e.g., Pterodactylus; Well-
nhofer, 1978). Besides, the presence of four jugal
processes in Pterodaustro may be another autapo-
morphy. In all pterodactyloids for which the jugal
is known, the caudoventral region of this bone is

rounded and does not form an independent pro-
cess. The apparent contact between the jugal and
the quadrate on the lateral surface of the skull
(Bonaparte, 1971; Sanchez, 1973) is another poten-
tial autapomorphy of Pterodaustro, because this
feature is not known for other pterodactyloids.

The mandible of Pterodaustro also displays sev-
eral autapomorphic features. The large number of
teeth (nearly 500 for each ramus) and their fila-
mentlike aspect are unique among pterosaurs. Also,
the extensive length of the mandible, which follows
the extension of the preorbital region of the skull,
is another autapomorphy of Pterodaustro. This
condition differs from that in all other pterosaurs,
including Pteranodon, in which the elongation of
the preorbital region is not matched by the lower
jaw, which is considerably shorter (Bennett, 1991).

16 ■ Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull

Howse ( 1 986)

Kellner (1996)

Bennett ( 1 994)

Unwin and Junchang (1997)

Figure 15 Different phylogenetic hypotheses of the relationships of Pterodaustro guinazui to other pterodactyloids

In sum, although the skull and jaws of Ptero-
daustro are easily distinguishable from those of
other pterosaurs by a large number of autapo-
morphic features, two derived cranial characters
(i.e., more than 150 teeth and a reduced rostral
projection of the surangular) support the close re-
lationship to Ctenochasma advocated by most au-
thors.

ACKNOWLEDGMENTS

We thank K. Padian for his comments and suggestions
about the initial manuscript. We are also grateful to L.
Meeker for preparing the photographs and to S. Orell for
editing the manuscript. L. Rhoads and W. Evans helped
us with additional assistance on the illustrations and the
manuscript. Logistic support for the field work was pro-
vided by the Departamento de Geologia (Universidad Na-
tional de San Luis, Argentina) and, during the 1994 ex-

pedition, by the Museo Argentino de Ciencias Naturales
“Bernardino Rivadavia.” Field and laboratory support
was provided by grants to L. Chiappe from the National
Geographic Society (5051-93) and the Philip McKenna
Foundation.

LITERATURE CITED

Bennett, S.C. 1991. Morphology of the Late Cretaceous
pterosaur Pteranodon and systematics of the Ptero-
dactyloidea. Doctoral Dissertation. Lawrence: Uni-
versity of Kansas, 680 pp.

. 1994. Taxonomy and systematics of the Late Cre-
taceous pterosaur Pteranodon (Pterosauria, Ptero-
dactyloidea). Occasional Papers of the Natural His-
tory Museum, University of Kansas 169:1-70.
Benton, M. 1990. The reign of the reptiles. New York:
Crescent Books, 143 pp.

Bonaparte, J. 1970. Pterodaustro guinazui gen. et sp. nov.
Pterosaurio de San Luis, Argentina, y su significado

Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull ■ 17

en la geologia regional (Pterodactylidae). Acta Geo-
logica Lilloana 10(10):207— 22 6.

— . 1971. Description del Craneo y Mandibulas de

Pterodaustro guinazui (Pterodactiloidea — Pterodaus-
triidae. nov.) de la Formacion Lagarcito, San Luis,
Argentina. Publication del Museo de Ciencias Na-
turales de Mar del Plata 1 (9):263— 272.

. 1978. El Mesozoico de America del Sur y sus Te-

trapodos. Tucuman, Argentina: Ministerio de Cul-
tura y Educacion, Fundacion Miguel Lillo, 535 pp.

Campos, D.A., and A.W.A. Kellner. 1985. Panorama of
flying reptiles study in Brazil and South America.
Anais da Academia Brasileira de Ciencias 57(4):
453-466.

Casamiquela, R.M., and G. Chong-Diaz. 1980. La pre-
sencia de Pterodaustro Bonaparte (Pterodactylo-
idea), del Neojurasico (?) de la Argentina, en los An-
des del Norte de Chile. Adas 2do. Congreso Argen-
tina Paleontologi'a Bioestratigafta 1:201-213.

Chiappe, L.M., and A. Chinsamy. 1996. Pterodaustro’s
true teeth. Nature 379:211-212.

Chiappe, L.M., D. Rivarola, A. Cione, M. Fregenal-Mar-
tinez, A. Buscalioni, H. Sozzi, L. Buatois, O. Gallego,
E. Romero, A. Lopez-Albarello, S. McGehee, C.
Marsicano, S. Adamonis, J. Laza, F. Ortega, and O.
Dilorio. 1995. Inland biota from a Lower Creta-
ceous Lagerstatte of central Argentina. 2nd Inter-
national Symposium on Lithographic Limestones
( Cuenca , Spain), Extended Abstracts, 57-60.

Chiappe, L.M., D. Rivarola, A. Cione, M. Fregenal-Mar-
tinez, H. Sozzi, L. Buatois, O. Gallego, J. Laza, E.
Romero, A. Lopez, A. Buscalioni, C. Marcicano, S.
Adamonis, P. Ortega, S. McGehee, and O. Dilorio.
1998a. Biotic association and paleoenvironmental
reconstruction of the “Loma del Pterodaustro ” fossil
site (Lagarcito Formation, Early Cretaceous, San
Luis, Argentina). Geobios 31(3):349-369.

Chiappe, L.M., D. Rivarola, E. Romero, S. Davila, and L.
Codorniu. 1998b. Recent advances in the paleontol-
ogy of the Lower Cretaceous Lagarcito Formation
(Parque Nacional Sierra de Las Quijadas, San Luis,
Argentina). In Lower and Middle Cretaceous terres-
trial ecosystems, ed. S.G. Lucas, J.L Kirkland, and
J. W. Estep. Bulletin of the New Mexico Museum of
Natural History 14:187-192.

Chinsamy, A., and A.W.A. Kellner. 1996. Unravelling the
microstructure of pterosaur teeth. Journal of Verte-
brate Paleontology 16(suppl. to 3):27-28A.

Cuvier, G. 1809. Memoire sur le squelette fossile d’un
Reptil volant des environs d’Aichstedt, que quelques
naturalistes ont pris pour un oiseau, et done nous
formons un genre de Sauriens, sous le nom de Ptero-
Dactyle. Annales du Musee d’Histoire Naturelle,
Paris 13:424.

Dong, Z. 1982. On a new Pterosauria ( Huanhepterus
quingyangensis gen. et sp. nov.) from Ordos, China.
Vertebrata PalAsiatica 20(2):1 15-121 [in Chinese].

Eaton, G.F. 1910. Osteology of Pteranodon. Memoirs of
the Connecticut Academy of Arts and Science 2:1-
38.

Fabre, J. 1974. Un nouveau Pterodactylidae sur le gise-
ment “Portlandien” de Canjuers (Var.): Gallodacty -
lus canjuersensis nov. gen., nov. sp. Annales de Pa-
leontologie (Vertebres) 62(l):35-70.

Flores, M. 1969. El Bolson de Las Salinas en la Provincia
de San Luis. Adas 4tas Jornadas Geologicas Argen-
tinas, Mendoza 1:311-327.

Flores, M., and P. Criado Roque. 1972. Cuenca de San

Luis, ler Simposio de Geologia Regional Argentina,
Cordoba, 567-580.

Howse, S.C.B. 1986. On the cervical vertebrae of the Pter-
odactyloidea (Reptilia: Archosauria). Zoological
Journal Linnean Society 88:307-328.

Kellner, A.W.A. 1995. The relationships of the Tapejaridae
(Pterodactyloidea) with comments on pterosaur phy-
logeny. In Sixth Symposium on Mesozoic Terrestrial
Ecosystems and Biota, Short Papers, ed. A. Sun and
Y. Wang, 12-11 . Beijing, China: Ocean Press.

. 1996. Description of new material of Tapejaridae

and Anhangueridae (Pterosauria, Pterodactyloidea)
and discussion of pterosaur phylogeny. Doctoral Dis-
sertation. New York: Columbia University, 347 pp.

— — - — . 1997. On the relationships of some early ptero-
dactyloid pterosaurs. In 15° Congresso Brasileiro de
Paleontologia, UNESP, Rio Claro, Sao Paulo, Bole-
tim de Resumos, 100.

Kellner, A.W.A., and W. Langston, Jr. 1996. Cranial re-
mains of Quetzalcotaltus (Pterosauria, Azhdarchi-
dae) from the Late Cretaceous sediments of Big Bend
National Park, Texas. Journal of Vertebrate Pale-
ontology 1 6(2):222— 23 1 .

Lawson, D.A. 1975. Pterosaur from the Latest Cretaceous
of west Texas. Discovery of the largest flying crea-
ture. Science 187:947-948.

Leonardi, G., and G. Borgomanero. 1985. Cearadadylus
atrox, nov. sp.: Novo Pterosauria (Pterodactyloidea)
da Chapada do Araripe, Ceara, Brasil. Departamen-
to Nacional da Produgdo Mineral, Coletanea de Tra-
balhos Paleontologicos, Seria Geologica 27:75-80.

Marsh, O.C. 1876. Principal characters of American
pterodactyls. American Journal of Science and Arts
12:479.

Meyer, H. v. 1834. Gnathosaurus subulatus, ein Saurus
aus dem lithographischen Schiefer von Solnhofen.
Museum Senckenbianum, Frankfurt 1:3.

— . 1847. Homoeosaurus maximiliani und Rham-

phorhynchus (Pterodactylus) longicaudus, zwei fos-
sile Reptilien aus dem Kalkschiefer vol Solenhofen.
Frankfurt, Germany: S. Schmerschen Buchhandlung.

— . 1851. Ctenochasma roemeri. Palaeontographica

2:82.

— . 1854. Briefliche Mitteilung An Professor Bromm.

Neues Jahrbuch fur Miner alogie 1854:47-58.

Price, L.I. 1971. A presenga de Pterosauria no Cretaceo
Inferior da chapada do Araripe, Brasil. Anais da Ac-
ademia Brasileira de Ciencias 43 (supl.):45 1-461.

Quenstedt, F.A. 1855. Uber Pterodactylus suevicus im lith-
ographischen Schiefer Wiirttembergs. Tubingen,
Germany.

Ramos, V. 1990. Field guide to geology of the Central
Andes (31°-33° SL). IUGS. ILP. Universidad de
Buenos Aires: 4.

Rivarola, D. 1994. Stratigraphy and palaeoenvironments
analysis of early cretaceous units of Sierra de Las
Quijadas, San Luis Province, Argentina. Internation-
al Congress of Sedimentology, Recife, Brasil 14:
G67-G69.

Romer, A.S. 1956. Osteology of the reptiles. Chicago, Il-
linois: University of Chicago Press, 772 pp.

Sanchez, T. 1973. Redescripcion del craneo y mandibulas
de Pterodaustro guinazui Bonaparte (Pterodaytylo-
idea, Pterodaustriidae). Ameghiniana 10:313-325.

Seilacher, A. 1990. Taphonomy of Fossil-Lagerstatten,
overview. In Paleobiology, a synthesis, ed. D.E.G.
Briggs and P.R. Crowther, 266-270. Oxford, United
Kingdom: Blackwell Scientific Publications.

18 ■ Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull

Seilacher, A., W.E. Reif, and F. Westphal. 1985. Sedimen-
tological, ecological and temporal patterns of fossil
Lagerstatten. Philosophical Transactions Royal So-
ciety, London, B 311:5-23.

Soemmerring, S.Th.V. 1812. Uber einen Ornithocephalus.
Denkschriften der Akademie der Wissenschaften
Munchen, Mathematisch-Physikalischen Classe 3:
89-158.

Unwin, D.M., and L. Junchang. 1997. On Zhejiangopte-
rus and the relationships of pterodactyloid ptero-
saurs. Historical Biology 12:199-210.

Wagner, A. 1837. Beschreibung eines neuentdeckten Or-
nithocephalus, nebst allgemeinen Bemerkungen iiber
die Organisation dieser Gattung. Abhandlungen der
Mathematisch-Physikalischen Classe der Koniglich
Bayerischen Akademie der Wissenschaften 2:165-
198.

Wellnhofer, P. 1978. Pterosauria. Handbuch der Palaeo-

herpetologie, Teil 19. Stuttgart, Germany: Gustav Fi-
scher Verlag, 82 pp.

. 1991. The illustrated encyclopedia of pterosaurs.

London: Salamander Books Ltd, 192 pp.

Wellnhofer, P., and A.W.A. Kellner. 1991. The skull of Ta-
pejara wellnm (Reptilia, Pterosauria) from the Low-
er Cretaceous Santana Formation of the Araripe Ba-
sin, northeastern Brazil. Mitteilungen der Bayerische
Staatssammlung fur Palaontologie und Historische
Geologie 31:89-10 6.

Young, C.C. 1964. On a new pterosaur from Sinkiang,
China. Vertebrata PalAsiatica 8:221-256.

Yrigoyen, M. 1975. La edad Cretacica del Grupo Gigante
(San Luis), su relacion con cuencas circunvecinas.
Acta ler. Congreso Geologico Argentino de Paleon-
tologia y Bioestratigrafia, Tucuman 2:9-56.

Received 29 October 1999; accepted 25 May 2000.

Contributions in Science, Number 483

Chiappe et al.: Pterodaustro Skull ■ 19

Natural History Museum

of Los Angeles County

900 Exposition Boulevard
Los Angeles, California 90007

LSgLX

Number 484
16 October 2000

Contributions
in Science

The Patterned-Wing Species of

CONICEROMYIA (DlPTERA: PHORIDAE)
Giar-Ann Kung and Brian V. Brown

Natural History Museum

of Los Angeles County

Serial
Publications
ol THE
Natural History
Museum ol
Los Angeles
County

Scientific

Publications

Committee

John Heyning, Deputy Director
for Research and Collections
John M. Harris, Committee Chairman
Brian V. Brown
Kenneth E. Campbell
Kirk Fitzhugh
Karen Wise

Robin A. Simpson and K. Victoria Brown,

Managing Editors

The scientific publications of the Natural History Museum
of Los Angeles County have been issued at irregular in-
tervals in three major series; the issues in each series are
numbered individually, and numbers run consecutively, re-
gardless of the subject matter.

# Contributions in Science, a miscellaneous series of tech-
nical papers describing original research in the life and
earth sciences.

# Science Bulletin, a miscellaneous series of monographs
describing original research in the life and earth sci-
ences. This series was discontinued in 1978 with the
issue of Numbers 29 and 30; monographs are now
published by the Museum in Contributions in Science.

# Science Series, long articles and collections of papers on
natural history topics.

Copies of the publications in these series are sold through
the Museum Book Shop. A catalog is available on request.

The Museum also publishes Technical Reports, a miscel-
laneous series containing information relative to scholarly
inquiry and collections but not reporting the results of
original research. Issue is authorized by the Museum’s Sci-
entific Publications Committee; however, manuscripts do
not receive anonymous peer review. Individual Technical
Reports may be obtained from the relevant Section of the
Museum.

Natural History Museum
of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007

Printed at Allen Press, Inc., Lawrence, Kansas
ISSN 0459-8113

The Patterned-Wing Species of Coniceromyia
(Diptera: Phoridae)

Giar-Ann Kung1 and Brian V. Brown1

ABSTRACT. Fifteen species of patterned-wing Coniceromyia are recognized, including the following ten
new to science: Coniceromyia apicalis, C. aurantia, C. bilineata, C. brevivena, C. globosa, C. impluvia, C.
impudica, C. leucomacula, C. setitarsalis, and C. truncata. A key to the identification of males of these
species is given.

INTRODUCTION

The genus Coniceromyia Borgmeier (1923) is a
group of 32 species confined to the New World.
Most species are tropical, with only two reaching
the southern U.S.A. (Borgmeier, 1968). Only a
small fraction of the true diversity of this genus has
been described.

One easily recognizable trait of males of some
species of this genus is the presence of patterned
wings. Patterned wings are herein defined as the
presence of definite pigment in the wing membrane,
but some of the patterning found in some species
is caused by, or enhanced by, the presence of dense,
reduced setae. There is no evidence that species
sharing wing-patterning form a monophyletic
group, but they are distinctive, attractive, conspic-
uous taxa whose recognition could inspire studies
on the function of such markings.

In this paper we describe ten new species of pat-
terned-wing Coniceromyia and present a key to all
species known to display this attribute.

METHODS

Most specimens were collected into 70% ethanol and crit-
ical-point dried using hexamethyldisilazane (Brown,
1993) or a standard critical-point drier (Gordh and Hall,
1979).

Terms used are those of Me Alpine (1981). Recently,
Stuckenberg (1999) has proposed that the large third an-
tennal segment of the Cyclorrhapha should be called the
postpedicel, rather than flagellomere 1, as it may comprise
a fusion of flagellomere 1 with other flagellomeres. The
actual segments that might have been involved in such a
fusion are still unidentified, however, and they still are
homologous to the flagellomeres of more primitive Dip-
tera. Therefore, we continue to use the term flagellomere
1 for the first visible flagellomere, regardless of its poten-
tial composition.

Tarsal ratios are length/width of the tarsal segment.

In addition to regular locality labels, all specimens have

1. Entomology Section, Natural History Museum of
Los Angeles County, 900 Exposition Boulevard, Los An-
geles, California 90007. Email: gkung@nhm.org and
bbrown@nhm.org.

Contributions in Science, Number 484, pp. 1-10
Natural History Museum of Los Angeles County, 2000

Code-49 barcoded labels, and their information is record-
ed in a database. The barcode of each holotype is reported
in brackets.

Material was deposited in the following museums (cod-
ens from Arnett et al., 1993).

EAPC Agroecological Inventory Collection, Departa-
mento de Proteccion Vegetal, Escuela Agricola
Panamericana, Apartado 93, Tegucigalpa, Hon-
duras (R. Cave)

EMUS Department of Biology, Utah State University,
Logan, Utah 84322-5305, U.S.A. (W.J. Hanson)
INBC Instituto Nacional de Biodiversidad, A.P. 22-
3100, Santo Domingo, Heredia, Costa Rica (M.
Zumbado)

LACM Entomology Section, Natural History Museum
of Los Angeles County, 900 Exposition Boule-
vard, Los Angeles, California 90007, U.S.A.
(B.V. Brown)

MCZC Museum of Comparative Zoology, Harvard
University, Cambridge, Massachusetts 02138,
U.S.A. (on indefinite loan to B.V. Brown)
MUCR Museo de Insectos, Universidad de Costa Rica,
San Pedro, San Jose, Costa Rica (P. Hanson)
MUSM Museo de Historia Natural, Universidad Na-
cional Mayor de San Marcos, Av. Arenales
1267, Apartado 14-0434, Lima-14, Peru (G. La-
mas)

MZLU Museum of Zoology, Lund University, Helgo-
nav. 3, S-223, 62 Lund, Sweden (R. Danielsson)
ROME Department of Entomology, Royal Ontario Mu-
seum, 100 Queens Park, Toronto, Ontario, Can-
ada M5S 2C6 (D.C. Darling)

USNM United States National Museum, Smithsonian
Institution, Washington, DC 20560, U.S.A. (on
indefinite loan to B.V. Brown)

SYSTEMATICS

Coniceromyia Borgmeier, 1923

Coniceromyia Borgmeier, 1923:338. Type species:

C. epicantha Borgmeier, by original designation.

DIAGNOSIS. Frons with median furrow. One
pair of reclinate supra-antennal setae present. Fla-
gellomere 1 elongate, conical in most species. Ane-
pisternum bare or setulose; anepisternal furrow ab-
sent. Foremetatarsus of male with elongate, fringed
process at apex. Tibiae with large, unpaired setae;

hind tibia without dorsal, longitudinal rows of en-
larged setulae. Costa usually darkened; wing vein
R2+3 absent or vestigial. Epandrium with fused sur-
syli shifted to left side. Hypandrium with bilobed
right process. A full list of genus-level characters
was given by Borgmeier (1963b).

PHYLOGENETIC RELATIONSHIPS. Conicer-
omyia was hypothesized to belong to a newly re-
stricted subfamily Phorinae by Brown (1992a, b),
with its closest relative being the Old World genus
Plethysmochaeta Schmitz (1924). No phylogenetic
relationships have been proposed for species within
the genus, and no taxonomic subdivisions, such as
subgenera, are currently used. The patterned-wing
species described herein do not form a monophy-
letic group but apparently are representatives of a
number of lineages that have independently evolved
this trait.

TAXONOMIC NOTES. Sexual dimorphism in
the genus is problematic, as females lack many of
the characters that are taxonomically useful in
males (including patterned wings). Therefore, most
of the 32 described species are known from male
specimens, although some have questionably asso-
ciated females. Most species are known from fewer
than five specimens, and twelve species have been
recorded from the holotype alone.

Male genitalia, often of great value in determin-
ing phorid species (e.g., Brown, 1996, 1998; Dis-
ney, 1989), are relatively uniform in Coniceromyia
and were not illustrated. A full discussion of their
structure is given by Brown (1992a).

WAY OF LIFE. The natural history of all species
is unknown, although one species was collected
with army ants. Species of related genera (Brown,
1992b) are scavengers.

The elaborately decorated males of some species,
especially C. stephensoni Peterson and C. leucom-
acula new species, possibly use their colored wings
and brightly marked forefemora in courtship dis-
plays or to defend territories against other males.
No such observations have been made on these spe-
cies, however, as specimens of both species have
only been collected by traps.

DISTRIBUTION. Most species are found in the
Neotropical Region, although two species reach the
southern U.S.A.

NOTES ON IDENTIFICATION. The most re-
cent key to species is that of Borgmeier (1963a);
since then, several further species have been de-
scribed (Borgmeier, 1969a, b; Borgmeier and Pra-
do, 1975; Peterson, 1982; Peterson and Arntfield,
1971; Prado, 1976).

Coniceromyia apicalis new species

(Fig- 1)

SPECIES RECOGNITION. This species is most
easily recognized by a thickened Mx vein and the
continuation of the wing darkening to the apical
margin. A smooth and entire wing margin and the
presence of setae on the anepisternum further dif-

2 ■ Contributions in Science, Number 484

ferentiate this species from C. vespertilio Schmitz
(1927).

DESCRIPTION. Male. Body length 1.75 to 2.40
mm. Frons brown. Flagellomere 1 brown, elongate-
conical. Arista apical and pubescent. Palpus brown,
small, with short, dark setae. Dorsum of thorax
brown. Pleural regions brown. Anepisternum with
short, fine setae. Scutellum brown. Legs yellowish-
brown. Foreleg with two strong setae anterodorsal
to posterodorsal on tibia; tarsomeres about twice
as long as wide (e.g., in one specimen, tarsal ratio
is 2.00:1.75:1.75:1.67:2.00); tarsomere 1 with an-
terobasal seta, anterior excavation, and anteroapi-
cal process, without basal triangular process; pos-
teroventral setae on tarsomere 1 without curved
tips. Midcoxa with long, erect setae; tips of setae
curved, without thick ventrolateral seta. Hind fe-
mur with sparse, tiny, blunt setae on basal half; se-
tae extend dorsally to one-third height of posterior
face of femur at basal extremity. Wing with Mx
thickened, about one-half thickness of R4+5, thicker
than M2 and CuAx; space between Mx and M2 nar-
rowed basally, so that the veins are parallel in basal
third and divergent in apical two-thirds. Wing with
line of darkened pigment and slightly denser setae
parallel to and posterior to leading edge. Apical
third of wing darkened by pigment from anterior
margin to CuAx (Fig. 1). Mean costal length 0.42
wing length; range 0.36 to 0.46. Halter white. Ter-
gites brown. Tergite 1 medially constricted, the
middle complete to partially split. Abdomen gray
ventrally. Terminalia yellowish-brown.

GEOGRAPHICAL DISTRIBUTION. Known
from two sites in Costa Rica.

DERIVATION OF SPECIFIC EPITHET. The
name is Latin for apical, referring to the continu-
ation of the wing pattern to the apical margin.

HOLOTYPE. d , COSTA RICA: Guanacaste, Es-
tacion Pitilla, 11°N, 85.43°W, vi.1989, P. Hanson,
Malaise trap, 600 m (LACM) [LACM ENT
029706].

PARATYPES. COSTA RICA: Alajuela, Penas
Blancas Valley, 10.32°N, 84.76°W, Id, 7.xi.l987,
E. Cruz, Malaise trap (LACM), 20 km S Upala,
10.73°N, 85.10°W, Id, 1-3.V.1990, ED. Parker
(EMUS); Guanacaste, Estacion Pitilla, 11°N,
85.43°W, 3d, iv.1989, 5d, v.1989, Id, vi.1989, P.
Hanson, Malaise trap, 600 m (LACM, MUCR),
2d, v.1994, P. Rios, Malaise trap, 700 m (INBC).

Coniceromyia aurantia new species

(Fig. 2)

SPECIES RECOGNITION. This species is most
easily recognized by a basal, oval, orange macula
on the anterior face of the forefemur and a poster-
oventral row of setae on the foretibia.

DESCRIPTION. Male. Body length 2.20 to 2.30
mm. Frons reddish-brown. Flagellomere 1 brown,
elongate-conical. Arista apical, pubescent. Palpus
yellow with short, dark setae. Dorsum of thorax
yellowish-brown. Pleural regions yellowish-brown.

Kung and Brown: Patterned-Wing Coniceromyia

Anepisternum with short setae. Scutellum yellow-
ish, brown medially. Legs yellowish-brown. Fore-
femur with oval, orange macula on basal half of
anterior face. Foretibia with dorsal row of short,
brown setae anterior to two strong dorsal setae;
with posteroventral row of short, thick setae. Fore-
leg with tarsomeres 2 to 5 one and one-half times
as long as wide (e.g., in one specimen, tarsal ratio
is 3.29:1.63:1.50:1.50:1.50); tarsomere 1 with an-
teroapical process and shallow excavation, without
basal triangular process; posteroventral setae on
tarsomere 1 without curved tips. Setae on midcoxa
not long, without curved tips; with thick ventral
setae. Hind femur with dense, tiny, blunt poster-
oventral setae on basal two-fifths; distribution of
setae extends dorsally to one-third height of pos-
terior face in basal half, tapers to ventral margin in
apical half. Wing (Fig. 2) with darkened pigment
along and M2, line of dense setae posterior to
anteroapical margin. Apical three-fourths of wing
darkened by dense setae. Mean costal length 0.34
wing length. Halter yellowish-white. Tergites dark
brown. Tergite 1 medially constricted, middle en-
tirely split. Abdomen yellowish-white ventrally.
Terminalia light brown.

PHYLOGENETIC RELATIONSHIPS. We con-
sider this species to be part of a monophyletic
group, along with C. leucomacula new species and
C. stepbensoni Peterson (1982), based on the pres-
ence of posteroventral setae on the foretibia and a
differentiated macula on the anterior face of the
forefemur.

GEOGRAPHIC DISTRIBUTION. Amazonian
Peru.

DERIVATION OF SPECIFIC EPITHET. The
name is Latin for orange, referring to the orange
macula on the forefemur.

HOLOTYPE. d, PERU: Madre de Dios, Manu
N.P., Cocha Cashu Station, 23-30.viii.1986, D.C.
Darling, Malaise trap, 380 m (MUSM) [LACM
ENT 137646].

PARATYPES. PERU: Madre de Dios, Manu
N.P., Cocha Cashu Station, Id, 18-22.viii.1986,
Id, 22-27.viii.1986, Id, 23-30.viii.1986, Id,
3 1 . viii— 1 .ix. 1986, D.C. Darling, Malaise trap, 380
m (LACM, ROME), Pakitza, 11.94°S, 71.28°W,
Id, 13-1 8.ii. 1992, D. Quintero, Malaise trap
(LACM), Id, 10.vi-6.vii.1993, R. Cambra, yellow
pans (LACM).

Coniceromyia bilineata new species

(Fig. 3)

SPECIES RECOGNITION. This species is most
recognizable by the two striae on the wing at the
anterior margin and along Mx. The most similar
species is C. striativena Borgmeier (Borgmeier,
1963a), which has darkening along M2.

DESCRIPTION. Male. Body length 2.15 to 2.35
mm. Frons dark brown. Flagellomere 1 dark brown
with long pubescence, elongate-conical. Arista api-
cal, pubescent. Palpus yellow. Dorsum of thorax

Contributions in Science, Number 484

dark brown. Pleural regions same color as dorsum
of thorax dorsally, lighter in color ventrally. Ane-
pisternum without setae. Scutellum dark brown.
Foreleg dark brown, yellowish-brown from apical
region of femora to apex; midleg slightly lighter
brown from tip of femur to apex; hind leg dark
brown. Foreleg with three dorsal setae on tibia; tar-
sal ratio of tarsomeres 2 to 4 subequal (e.g., in one
specimen, ratio is 1.80:0.89:1.00:1.14:2.00); tar-
somere 1 with basal triangular process, anterior ex-
cavation, and anteroapical process; posteroventral
setae on tarsomere 1 without curved tips. Foreleg
with pulvilli slightly enlarged. Midcoxa with fine
setae and thick ventrolateral seta. Hind femur with
dense, tiny, blunt posteroventral setae on basal
third, distribution of setae tapered apically toward
ventral margin. Wing with darkened pigment along
anteroapical margin and M:; subcostal cell dark-
ened by pigment at apex (Fig. 3). Mean costal
length 0.48 wing length. Halter yellow. Tergites
dark brown, almost black. Tergite 1 shortened or
not shortened medially. Abdomen gray ventrally.
Terminalia light brown.

GEOGRAPHICAL DISTRIBUTION. Known
from one site in Costa Rica.

DERIVATION OF SPECIFIC EPITHET. The

name is Latin for two-lined, referring to the two
striae on the wing.

HOLOTYPE. d, COSTA RICA: Guanacaste,
Volcan Cacao, Cerro Pedregal, 10.93°N, 85.48°W,
ii— iv. 1989, 1. Gauld, D. Janzen, Malaise trap, 1000
m (LACM) [LACM ENT 053782].

PARATYPES. COSTA RICA: Guanacaste, Santa
Rosa National Park, 10.95°N, 85.62°W, Id, 21. ii-
14.iii. 1987, I. Gauld, D. Janzen, Malaise trap
(LACM), Volcan Cacao, Cerro Pedregal, 10.93°N,
85.48°W, 4d, ii— iv. 1989, 1. Gauld, D. Janzen, Mal-
aise trap, 1000 m (INBC, LACM), Id, ii-iv.1989,
P. Hanson, Malaise trap, 1000 m (LACM).

Coniceromyia blomae Peterson and
Arntfield, 1971

Coniceromyia blomae Peterson and Arntfield,

1971:395-398, fig. 1.

NEW MATERIAL EXAMINED. MEXICO:
Chiapas, Yerba Buena, 16.35°N, 96.07°W, Id,
8.vi.l 969, W. Mason, 1760 m (LACM).

Coniceromyia brevivena new species

(Fig. 4)

SPECIES RECOGNITION. This species differs
from all other patterned-wing Coniceromyia by the
presence of the apical half of R2+3. This species is
further differentiated from C. maculipennis Borg-
meier (1969b) by the lack of darkening along the
posterior wing margin.

DESCRIPTION. Male. Body length 3.1 mm.
Frons dark brown. Flagellomere 1 dark brown,
elongate-conical. Arista apical and pubescent. Pal-
pus light brown with short, dark setae. Dorsum of

Kung and Brown: Patterned-Wing Coniceromyia H 3

thorax dark brown. Pleural regions same color as
dorsum of thorax. Anepisternum with short, fine
setae. Scutellum dark brown. Legs brown, lighter
apically. Foreleg with three dorsal setae on tibia;
tarsomeres 2 to 4 twice as long as wide, tarsal ratio
is 3.50:2.00:2.00:2.00:1.60; tarsomere 1 with three
thick anterobasal setae ventral to anterior excava-
tion, without basal triangular process; posteroven-
tral setae on tarsomere 1 without curved tips. Mid-
coxa with fine setae and thick ventrolateral seta.
Hind femur with tiny, blunt posteroventral setae on
basal half; distribution of setae tapered apically to-
ward ventral margin. Apical third of wing, except
for apex, darkened by pigment; costal and subcos-
tal cells darkened by pigment. Apical half of R2+3
present (Fig. 4). Costal length 0.43 wing length.
Halter yellow. Tergites dark brown. Tergite 1 me-
dially constricted. Abdomen dark gray ventrally.
Terminalia dark brown.

GEOGRAPHICAL DISTRIBUTION. Known
from a single site in Peru.

DERIVATION OF SPECIFIC EPITHET. The
name is Latin for short vein, referring to the pres-
ence of the apical half of R2+3.

HOLOTYPE. 6, PERU:“ Madre de Dios, Rio
Tambopata Reserve, 12.83°S, 69. 28°W, 8.xi.l983,
T. Erwin, canopy fogging, 290 m (USNM) [LACM
ENT 028082].

Coniceromyia globosa new species

(Fig. 5)

SPECIES RECOGNITION. This species is most
easily recognized by the round macula on the wing
between the anteroapical margin and M,. The most
similar species are C. setitarsalis new species and
C. impluvia new species, in which the darkening on
the wing between the anteroapical margin and M,
is not distinctly round.

DESCRIPTION. Male. Body length 1.8 to 1.9
mm. Frons brown. Flagellomere 1 brown, elongate-
conical. Arista apical, pubescent. Palpus brown.
Dorsum of thorax brown. Pleural regions same col-
or as dorsum of thorax. Anepisternum without se-
tae. Scutellum brown. Legs yellowish-brown. Fore-
leg with two anterodorsal setae on tibia; tarsal ratio
of tarsomeres 2 to 4 subequal (e.g., in one speci-
men, ratio is 2.29:1.00:1.14:1.33:2.00); tarsomere
1 with anteroventral excavation and anteroapical
process, without basal triangular process; poster-
oventral setae on tarsomere 1 without curved tips.
Midcoxa with fine setae and thick ventrolateral
seta. Hind femur with dense, tiny, blunt posterov-
entral setae in basal third; setae slightly tapered api-
cally. Wing with pigment darkened along R4+5, an-
teroapical margin, M1? M2; darkened pigment
forming round macula between Mj and anteroap-
ical margin (Fig. 5). Mean costal length 0.48 wing
length; range 0.46 to 0.50. Halter yellowish-brown.
Tergites dark brown. Tergite 1 shortened medially.
Abdomen gray ventrally. Terminalia light brown.

GEOGRAPHICAL DISTRIBUTION. Costa
Rica and Panama.

DERIVATION OF SPECIFIC EPITHET. The
name is Latin for spherical, referring to the round
macula on the wing.

HOLOTYPE. d, COSTA RICA: San Jose, Brau-
lio Carrillo NP, 10.17°N, 84.12°W, 10.iv.1985, H.
Goulet, L. Masner, 500 m (LACM) [LACM ENT
003289].

PARATYPES. COSTA RICA: San Jose, Braulio
Carrillo NP, 8.2 km NE Tunel, 10.12°N, 83.97°W,
Id, 15.V.1988, P. Hanson, 1500 m (LACM). PAN-
AMA: Panama Prov., Cerro Jefe, Id, 31 .vii. 1978,
N.E. Woodley, 975 m (MCZC).

Coniceromyia impluvia new species

(Fig. 6)

SPECIES RECOGNITION. This species is most
similar to C. setitarsalis new species but differs by
a shorter flagellomere, presence of darkening along
CuA1? and absence of curved tips on ventral setae
of tarsomere 1 on the foreleg. The clear window in
the darkening of the wing, between the apical mar-
gin and Mj, is twice as long as wide, further dif-
ferentiating this species from C. setitarsalis.

DESCRIPTION. Male. Body length 1,7 to 2.0
mm. Frons dark brown. Flagellomere 1 brown,
elongate-conical, length of tapered portion less than
length of untapered portion. Arista apical, pubes-
cent. Palpus yellow. Dorsum of thorax brown. Pleu-
ral regions same color as dorsum of thorax. Ane-
pisternum without setae. Scutellum brown. Legs
brown to yellowish-brown. Foreleg with two dorsal
setae on tibia; tarsal ratio of tarsomeres 2 to 4 sub-
equal (e.g., in one specimen, ratio is 2.29:0.88:1.00:
1.00:1.16); tarsomere 1 with anterior excavation
and anteroapical process; posteroventral setae on
tarsomere 1 without curved tips. Tarsomere 1 with-
out triangular process. Midcoxa with fine setae and
thick ventrolateral seta. Hind femur with dense,
tiny, blunt posteroventral setae in basal third; dis-
tribution of setae slightly tapered apically. Line of
pigment and dense setae parallel to and posterior
to leading edge of wing. Wing darkened by pigment
near tip of R4+5 along anteroapical margin, M1? M2,
and, faintly, along CuA^ darkened pigment present
between M, and anteroapical margin in apical half,
but not extending to apical margin (Fig. 6). Mean
costal length 0.45 wing length; range 0.42 to 0.47.
Halter white. Tergites dark brown. Abdomen gray
ventrally. Terminalia light brown.

GEOGRAPHICAL DISTRIBUTION. Lowland
Costa Rica.

DERIVATION OF SPECIFIC EPITHET. The

name is Latin for window, referring to the presence
of a clear window in the wing darkening.

HOLOTYPE. 6, COSTA RICA: Puntarenas, 24
km W Piedras Blancas, 8.77°N, 83.4°W, xii.1990,
P. Hanson, Malaise trap, 200 m (LACM) [LACM
ENT 040466].

PARATYPES. COSTA RICA: Limon, 16 km W

4 ■ Contributions in Science, Number 484

Kung and Brown: Patterned-Wing Coniceromyia

Guapiles, 10.15°N, 83.92°W, 1 6, iii-v.1990, P.
Hanson, Malaise trap, 400 m (LACM); Puntarenas,
Road to Rincon, 24 km W Piedras Blancas, 8.77°N,
83.4°W, Id, iii— iv. 1989, R Hanson, I. Gauld, Mal-
aise trap, 200 m (LACM), Id, xi.1990, P Hanson,
Malaise trap, 200 m (INBC), 3 km S Rincon,
8.68°N, 83.48°W, Id, ix-xi.1989, Id, xii.1989, P
Hanson, Malaise trap, 10 m (LACM). PANAMA:
Darien, Cana Pirre Trail, 7.72°N, 77.7°W, Id,
7.vi.l996, A. Gillogly, FIT, 1250 m (LACM).

Coniceromyia impudica new species

(Fig. 7)

SPECIES RECOGNITION. This species is most
easily recognized by its dark coloration, distinctive
wing pattern, and the presence of curved tips on
the ventral setae on tarsomere 1 of the foreleg.

DESCRIPTION. Male. Body length 2.35 to 2.85
mm. Frons blackish-brown. Flagellomere 1 dark
brown with long pubescence, elongate-conical.
Arista apical, pubescent. Palpus yellow. Dorsum of
thorax and pleural regions dark brown. Anepister-
num without setae. Scutellum dark brown. Legs
with femora dark brown, yellow-orange apically;
tibiae dark brown, yellow-orange basally and api-
cally; tarsi yellowish-brown. Foreleg with three to
four dorsal to anterodorsal setae on tibia; tarsal ra-
tio of tarsomeres 2 to 5 subequal (e.g., in one spec-
imen, ratio is 2.00:1.25:1.25:1.33:1.33); tarsomere
1 with basal triangular process, anterior excava-
tion, and anteroapical process; tarsomere 1 with
curved tips on some posteroventral setae. Foreleg
with pulvilli slightly enlarged. Midcoxa with fine
setae and thick ventrolateral seta. Hind femur with
dense, tiny, blunt posteroventral setae on basal
third, setae tapered apically toward ventral margin.
Wing with darkened pigment along, and in mem-
brane posterior to, M1} with enlarged areas at tip
of R4+5 and mid-Mj. Macula darkened by pigment
on middle of M2 present to absent, R4+5 slightly
darkened by pigment and narrow line of darkened
pigment parallel to and posterior to the leading
edge of the wing (Fig. 7). Mean costal length 0.50
wing length; range 0.47 to 0.53. Halter yellow. Ter-
gites dark brown, almost black. Tergite 1 shortened
or lightened medially. Abdomen dark, almost
black, ventrally. Terminalia dark brown.

GEOGRAPHICAL DISTRIBUTION. Honduras.

DERIVATION OF SPECIFIC EPITHET. The
name is Latin for bold, referring to the dark and
distinctive wing pattern.

HOLOTYPE. d, HONDURAS: Francisco Mor-
azan, San Antonio de Oriente, Cerra Uyuca,
14.03°N, 87.07°W, 12-1 8.ii.l990, R. Cave, Mal-
aise trap in cloud forest (LACM) [LACM ENT
061908].

PARATYPES. HONDURAS: Cortes, Parque Na-
cional Cusuco, 15.48°N, 88.22°W, 2d, 2.iii.l995,
R. Cordero, Malaise trap (LACM), 1600 m, 6d,
30.ix.1995, 3d, 15.X.1995, R. Cave, Malaise trap
in oak/pine cloud forest, 1600 m (EAPC, MZLU);

Contributions in Science, Number 484

Francisco Morazan, Parque Nacional La Tigra,
14.25°N, 87.08°W, Id, 29.iii.1995, Id,

13.vi.1995, R. Cave, Malaise trap in oak/pine
cloud forest (MZLU); San Antonio de Oriente, Cer-
ra Uyuca, 14.03°N, 87. 07°W, 2d, 12-1 8.ii. 1990,
Id, 7-13.V.1990, R. Cave, Malaise trap in cloud
forest (LACM).

Coniceromyia leucomacula new species

(Figs. 8, 9)

SPECIES RECOGNITION. This species is most
easily recognized by a large white patch on the fore-
femur and an oval macula between M2 and CuA^
A second darkening on the wing, between M4 and
M2, may be present or absent.

DESCRIPTION. Male. Body length 2.3 to 2.9
mm. Frons brown. Flagellomere 1 with various de-
grees of orange and brown, elongate-conical. Arista
apical, pubescent. Palpus yellow-orange. Dorsum
of thorax brown with yellowish margins. Pleural
regions yellowish-brown to brown. Anepisternum
with short, fine setae. Scutellum brown, darker
than pleural regions. Legs mostly yellow. Foreleg
with anterior of femur black apically with white
patch about three-quarters length of femur, poste-
rior side brown. Midfemur brown basally and yel-
lowish-brown apically. Hind femur with apical an-
teroventral black spot. Foretibia with dorsal row of
orange setae anterior to one to four black dorsal
setae, with row of posterior spine-like setae, row
extending ventrally along apical margin; tarsal seg-
ments clearly longer than wide (e.g., in one speci-
men, tarsal ratio is 5.20:3.60:4.00:3.00:2.67); tar-
somere 1 with thick basal seta, ventral to anterior
excavation, and anteroapical process, without basal
triangular process; posteroventral setae on tarso-
mere 1 without curved tips. Midcoxa with fine se-
tae and thick ventrolateral seta. Hind femur with
tiny, blunt posteroventral setae on basal half; setae
extend slightly higher at basal extremity. Wing with
costal cell slightly darkened by pigment, anteroap-
ical margin with faint darkening of pigment and
dense, fine setae. Apical half with large, oval, dark-
ening of pigment between M2 and CuA} and small-
er and lighter spot between Mt and M2 (Fig. 8).
Mean costal length 0.44 wing length; range 0.41 to
0.47. Halter white. Tergites brown. Tergite 1 me-
dially constricted; middle entirely to partially split.
Tergites 3 through 6 with lighter anterior margin.
Abdomen gray ventrally. Terminalia yellowish. Hy-
pandrium brown.

VARIATION. Specimens from Estadon Cacao
differ from the holotype and other specimens by
overall lighter coloration, including the lack of
darkening between Mj and M2 and smaller macula
between M2 and CuAi (Fig. 9).

PHYLOGENETIC RELATIONSHIPS. See C.
aurantia, above.

GEOGRAPHICAL DISTRIBUTION. Known
from three midelevation sites in Costa Rica.

DERIVATION OF SPECIFIC EPITHET. The

Kung and Brown: Patterned-Wing Coniceromyia ■ 5

name is a combination of the Greek word for
white, leukos, and Latin word macula, referring to
the white patch of the forefemur.

HOLOTYPE. d, COSTA RICA: Puntarenas,
Monteverde Biological Station, 10.33°N, 84.79°W,
9-18.iii.1995, B. V. Brown, Malaise trap, 1700 m
(LACM) [LACM ENT 051936].

PARATYPES. COSTA RICA: Cartago, La Can-
greja, 9.8°N, 83.47°W, 7c 3, vii.1991, Id, viii-
ix.1991, 2d, xi.1991, 4d, xii.1991, lOd, vi-
vii.1992, 9d, ix-xii.1992, P. Hanson, Malaise trap,
1950 m (INBC, LACM, MCZC, MUCR, USNM);
Guanacaste, Cerro Pedregal, 10.93°N, 85.48°W,
3d, ii— iv. 1989, I. Gauld, D. Janzen, Malaise trap,
1000 m (LACM), Estacion Cacao, 10.93°N,
85.47°W, 2d, ii.1989, 1. Gauld, D. Janzen, Malaise
trap, 900 m (LACM); Puntarenas, Monteverde Bi-
ological Station, 10.33°N, 84.79°W, Id, 9-
18.iii.1995, B. V. Brown, Malaise trap, 1700 m
(LACM).

Coniceromyia setitarsalis new species

(Fig. 10)

SPECIES RECOGNITION. This species is most
similar to C. impluvia but differs by a longer fla-
gellomere, absence of darkening along CuA1? and
presence of curved tips on the ventral setae of tar-
somere 1 on the foreleg. The clear window in the
darkening of the wing, between the apical margin
and Mj, is about as long as wide, further differen-
tiating this species from C. impluvia.

DESCRIPTION. Male. Body length 1.9 to 2.6
mm. Frons dark brown. Flagellomere 1 dark brown
with long pubescence, elongate-conical, with ta-
pered portion longer than untapered portion. Arista
apical, plumose. Palpus brown. Dorsum of thorax
dark brown. Pleural regions same color as dorsum
of thorax. Anepisternum without setae. Scutellum
dark brown. Legs brown to yellowish-brown, ligh-
ter apical of femora. Foretibia and tarsus with long,
fine, erect, ventral setae. Foreleg with two dorsal
setae on tibia; tarsal ratio of tarsomeres 2 to 4 sub-
equal (e.g., in one specimen, ratio is 2.67:1.11:0.89:
1.00:2.00); tarsomere 1 with anterior excavation
and anteroapical process. Excavation bordered
with setae, ventrobasal margin of excavation with
dense, fine setae. Tarsomere 1 without basal trian-
gular process; some posteroventral setae on tarso-
mere 1 long, thin, curve-tipped, and erect. Midcoxa
with fine setae and thick ventrolateral seta. Hind
femur with dense, tiny, blunt setae on basal third;
distribution of setae tapered apically toward ventral
margin. Wing darkened with pigment at apex of
costa, base of M1} with macula anterior to M! at
midpoint of M1? pigment lighter at apex; pigment
darkened along anteroapical margin and apex of
M2, M2 faintly darkened by pigment (Fig. 10).
Mean costal length 0.52 wing length; range 0.49 to
0.54. Halter yellow. Tergites dark brown. Tergite 1
shortened medially. Abdomen gray ventrally Ter-
minalia light brown.

6 ■ Contributions in Science, Number 484

GEOGRAPHICAL DISTRIBUTION. Known
from four midelevation sites in Costa Rica.

DERIVATION OF SPECIFIC EPITHET. The
name is a combination of the Latin word seta and
Greek word tarsus, referring to the setose tarso-
mere 1 of the foreleg.

HOLOTYPE. d, COSTA RICA: Cartago, La
Cangreja, 9.8°N, 83.97°W, vi-vii.1992, P. Hanson,
Malaise trap, 1950 m (LACM) [LACM ENT
062665].

PARATYPES. COSTA RICA: Cartago, La Can-
greja, 9.8°N, 83.97°W, Id, vii.1991, Id, viii-
ix.1991, 2d, xi.1991, Id, vi-vii.1992, Id, ix-
xii.1992, P. Hanson, Malaise trap, 1950 m
(LACM); Puntarenas, Monteverde, 10.10°N,
83.43°W, Id, l-5.vi.1988, B. V. Brown, Malaise
trap in stunted forest, 1700 m (LACM), Id, 1-
10.iii.1992, D. M. Wood, Malaise trap, 1500 m
(LACM); San Jose, Braulio Carrillo National Park,
8.2 km NE Tunel, 10.12°N, 83.97°W, Id,
15.V.1988, P. Hanson, Malaise trap, 1500 m
(LACM), Zurqui de Moravia, 10.05°N, 84.02°W,
Id, vii.1990, 5d, ix-x.1990, 2d, x-xii.1990, Id,
iii.1991, 2d, v.1991, Id, vi.1991, 3d, vii.1991,
Id, ix.1991, Id, v.1992, 2d, iii— iv. 1993, 2d, iv-
v.1993, 7d, l-15.vi.1993, 3d, ix-x.1993, Id,
v.1995, Id, vi.1995, Id, i.1996, P. Hanson, Mal-
aise trap, 1600 m (INBC, LACM, MUCR).

Coniceromyia stephensoni Peterson, 1982

Coniceromyia stephensoni Peterson, 1982:136-

138, figs. 1-2.

PHYLOGENETIC RELATIONSHIPS: See C.

aurantia, above.

NEW MATERIAL EXAMINED. COSTA RICA:
Puntarenas, Las Alturas, 8.95°N, 82.83°W, Id, 10-
13.vi.1998, B. Brown, V. Berezovskiy, Malaise trap
#1, 1600 m (LACM).

Coniceromyia striativena Borgmeier, 1963a
(Fig. 11)

Coniceromyia striativena Borgmeier, 1963a:457-

458, fig. 3.

EMENDED DESCRIPTION. We examined the
holotype of this species and found an error in the
original description: the halter is whitish-yellow,
not black as stated by Borgmeier.

NEW MATERIAL EXAMINED. COSTA RICA:
Alajuela, 5 km W San Ramon, 10.05°N, 84.05°W,
2d, i.1997, 3d, iv.1997, O. Castro, Malaise trap,
1200 m (LACM); Cartago, Turrialba, 9.93°N,
83.67°W, Id, 15-19.vii.1966, P. Spangler, Malaise
trap, 600 m (USNM). MEXICO: Chiapas, San
Cristobal, 16.75°N, 92.67°W, Id, 20.vii.l 969, W.
Mason, Malaise trap, 2000 m (LACM).

Coniceromyia truncata new species

(Fig. 12)

SPECIES RECOGNITION. This species is most
easily distinguished from the other patterned-wing

Kung and Brown: Patterned-Wing Coniceromyia

Figures 1-12 Wings. 1. Coniceromyia apicalis new species. 2. Coniceromyia aurantia new species. 3. Coniceromyia
bilineata new species. 4. Coniceromyia brevivena new species. 5. Coniceromyia globosa new species. 6. Coniceromyia
impluvia new species. 7. Coniceromyia impudica new species. 8. Coniceromyia leucomacula new species [La Cangreja].
9. Coniceromyia leucomacula new species [Estacion Cacao]. 10. Coniceromyia setitarsalis new species. 11. Coniceromyia
striativena Borgmeier. 12. Coniceromyia truncata new species

Contributions in Science, Number 484

Kung and Brown: Patterned-Wing Coniceromyia ■ 7

Coniceromyia by the sinuous, widely spaced M2
and CuA, and the truncate apical margin of the
wing.

DESCRIPTION. Male. Body length 2.40 to 2.75
mm. Frons dark brown. Flagellomere 1 oval,
brown. Arista subapical, pubescent. Palpus yellow
with long, dark setae. Dorsum of thorax dark
brown. Pleural regions brown. Anepisternum with-
out setae. Scutellum dark brown. Legs yellowish-
brown to dark brown. Forefemur with more than
one row of long, thin setae on anteroventral margin
and one row of long, thin setae on posteroventral
margin. Foreleg with two strong dorsal or near-dor-
sal setae on tibia; tarsomere 1 with triangular pro-
cess, ventral excavation, and apical process. Basal
margin of triangular process with dense, short, fine
setae. Posteroventral setae on tarsomere 1 without
curved tips. Tarsal segments twice as long as wide
(e.g., in one specimen, tarsal ratio is 2.00:1.87:
1.83:1.80:1.88). Midcoxa with fine setae and thick
ventrolateral seta. Hind femur with tiny, blunt pos-
teroventral setae on basal half; distribution of setae
slightly tapered apically. Wing with apical margin
truncate. M2 and CuAj slightly sinuous, with the
space between these two veins markedly large.
Dense setae and darkened pigment anteriorly and
posteriorly parallel to R4+5, between M, and M2,
extending slightly apical of Mx. Costal and subcos-
tal cells slightly darkened by pigment. Dense patch
of fine setae between M2 and CuA1? producing a
faint darkening (Fig. 12). Mean costal length 0.50
wing length; range 0.48 to 0.52. Halter yellow. Ter-
gites dark brown; tergite 1 lightened medially. Ab-
domen gray ventrally. Terminalia light brown.

GEOGRAPHICAL DISTRIBUTION. Known
from four midelevation sites in Costa Rica.

DERIVATION OF SPECIFIC EPITHET. The
name is Latin for truncate, referring to the apical
margin of the wing.

HOLOTYPE. A, COSTA RICA: San Jose, 6 km
N San Gerardo, 9.55° N, 83.8°W, xi.1992, P. Han-
son, Malaise trap, 2800 m (LACM) [LACM ENT
050221].

PARATYPES. COSTA RICA: Cartago, 4 km NE
Canon, 9.71°N, 83.94°W, Id, vi.1995, P. Hanson,
Malaise trap, 2350 m (LACM), Genesis II, 9.71°N,
83.91°W, Id, ii.1995, Id, viii.1995, Id, vii.1996,
P. Hanson, Malaise trap, 2350 m (LACM), Villa
Mills, 9.57°N, 83.73°W, Id, xi-xii.1989, Id, iii-

iv. 1990, P. Hanson, Malaise trap, 3000 m (LACM);
Puntarenas, Las Alturas, 8.95°N, 82.83°W, Id, iii-

v. 1995, P. Hanson, Malaise trap, 2100 m (LACM);
San Jose, 2 km W Empalme, 9.72°N, 83.97°W, Id,

vi. 1995, Id, vii.1995, P. Hanson, Malaise trap,
2300 m (LACM), 20 km S Empalme, 9.63°N,
83.85°W, 2d, viii.1988, Id, iii-iv.1990, P. Hanson,
Malaise trap, 2800 m (LACM), Sendero el Carbon,
Estacion Cuerici, 5 km E Villa Mills, 9.57°N,
83.73°W, Id, 16.iii.1996, A. Picado, 2600 m
(INBC), 6 km N San Gerardo, 9.55°N, 83.8°W, 3d,
vi.1992, Id, ix.1992, Id, xi.1992, P. Hanson,
Malaise trap, 2800 m (LACM, MUCR).

8 ■ Contributions in Science, Number 484

KEY TO PATTERNED-WING CONICEROMYIA
MALES

1 Anepisternum with setae 2

- Anepisternum without setae 7

2 Anterior of forefemur black with distinct

white markings 3

- Anterior of forefemur not black with distinct

white markings 4

3 Anterior of forefemur with large white patch
extending three-quarters length of femur.
Wing with oval darkening between M2 and
CuAj. Wing with or without lighter darkening

between Ma and M2 (Figs. 8, 9)

C. leucomacula new species

[Costa Rica]

Anterior of forefemur mostly black with nar-
row white markings. Wing with large double-
lobed darkening, one lobe between anterior
margin and M2, other lobe between M2 and

CuAt C. stephensoni Peterson

[Costa Rica, Panama]

4 Apical half of R2+3 present. Costal and sub-
costal cell entirely darkened. Apical third of
wing, except for apex, darkened. Darkening

not extending to CuA3 (Fig. 4)

C. brevivena new species

[Peru]

Apical half of R2+3 not present. Both costal
and subcostal cells not entirely darkened.
Apex of wing darkened to CuA1? or not dark-
ened 5

5 Forefemur with anterior, oval, orange macula
on basal half. Tarsomere 1 of foreleg without
basal seta. Foretibia with posteroventral row

of setae in apical half

C. aurantia new species

[Peru]

- Forefemur without orange macula. Tarsomere
1 of foreleg with basal seta. Foretibia without
posteroventral row of setae in apical half 6

6 Darkening of wing not extending to apical
margin. M1 not thickened. Space between M1
and M2 not narrowed basally. Apical two-
fifths of wing, except for apex, darkened . . .
C. blomae Peterson and Arntfield

[Mexico]

Darkening of wing extending to apical margin.
Mj thickened. Space between and M2 nar-
rowed basally so that veins are parallel in bas-
al third and divergent in apical two-thirds.
Apical third of wing darkened (Fig. 1) . . . . .
C. apicalis new species

[Costa Rica]

7 Flagellomere 1 oval. M2 and CuAt widely
spaced and sinuous. Apical margin of wing
truncate (Fig. 12) . . C. truncata new species

[Costa Rica]

- Flagellomere 1 elongate-conical. M2 and CuAj

not widely spaced and sinuous. Apical margin
of wing not truncate 8

Kung and Brown: Patterned-Wing Coniceromyia

8 Wing with darkening in membrane posterior

to M, 9

- Wing without darkening in membrane poste-

rior to darkening, if present, restricted to
wing veins 11

9 Foretibia without excavation with three to

four dorsal to anterodorsal setae. Wing pat-
terning along M2-enlarged midvein so that the
darkening extends into membrane posterior to
M2. Patterning does not extend posteriorly to
CuAt C. imp u die a new species

[Honduras]

- Foretibia excavate with one long seta. Wing
patterning extends posteriorly to CuAj . . 10

10 Wing margin emarginate at M2 and CuAa.
Costal and subcostal cells not darkened. Dark-
ening present in apical half of wing. Darkening

not present on posterior margin

C. vespertilio Schmitz

[Brazil]

Wing margin not emarginate; smooth and en-
tire. Costal and subcostal cells darkened. Pos-
terior half and apical two-fifths of wing dark-
ened C. maculipennis Borgmeier

[Brazil]

11 Wing pattern restricted to two lines of dark-
ening: anterior margin and Mj (Fig. 3); wing
without distinct darkening between these two
lines. Ventral setae on tarsomere 1 of foreleg

without curved tips

C. bilineata new species

[Costa Rica]

- Wing pattern not restricted to two lines of

darkening along anterior margin and Mt. Ven-
tral setae on tarsomere 1 with or without
curved tips 12

12 Anterior margin of wing darkened, but mem-
brane between anterior margin and not
darkened. Wing darkened in three striae: an-
teroapical margin, M1} and M2 (Fig. 11) ... .
C. striativena Borgmeier

[Costa Rica, Mexico]
Wing with membrane between anterior mar-
gin and Mj darkened; darkening on M2 pre-
sent or absent 13

13 Foreleg with basal triangular process on tar-
somere 1 . Wing with darkened pigment along
Mls with enlarged areas at tip of R4+5 and
mid-M,. Darkening mid-M2 present or absent.
Without darkening along CuAj (Fig. 7) ....
C. impudica new species

[Honduras]

- Foreleg without basal triangular process on

tarsomere 1 . Wing with darkened pigment be-
tween anteroapical margin and along an-
teroapical margin, Mls and M2. Darkening
along M2 sometimes faint. Darkening along
CuAi present or absent 14

14 Wing with a distinct round macula between
anterior margin and Mx; darkening present
along M2, although sometimes faint (Fig. 5)
C. globosa new species

Contributions in Science, Number 484

[Costa Rica, Panama]

- Darkening between anterior margin and Mx
not distinctly round; darkening on M2 present,
although sometimes faint (Figs. 6, 10) . . 15

15 Tapered portion of flagellomere 1 longer than
untapered portion. Tarsomere 1 of foreleg
with long, thin, curve-tipped, erect, posterior
setae. CuA! without darkening (Fig. 10) ....

C. setitarsalis new species

[Costa Rica]

- Tapered portion of flagellomere 1 shorter than
untapered portion. Setae on posterior of tar-
somere 1 of foreleg short, not curve-tipped or

erect. CuA! faintly darkened (Fig. 6)

C. impluvia new species

[Costa Rica]

ACKNOWLEDGMENTS

Figures 1 and 3 through 12 in this paper were skillfully
prepared by Jesse Cantley. We thank P. Hanson and R.
Cave for sending unsorted samples, from which we ex-
tracted phorid specimens. This work was partly funded by
National Science Foundation grant DEB-9407190 and an
NSF Research Experience for Undergraduates Supple-
ment.

LITERATURE CITED

Arnett, R. H., G. A. Samuelson, and G. M. Nishida. 1993.
The insect and spider collections of the world.
Gainesville, Florida: Sandhill Crane Press, vi + 310

pp.

Borgmeier, T. 1923. Contribui^ao para o conhecimento
dos Phorideos do Brasil. Archivos do Museu Na-
tional, Rto de Janeiro 24:323-346.

. 1963a. New or little known Coniceromyia and

some other Neotropical or Paleotropical Phoridae
(Dipt.). Studia Entomologica 6:449-480.

. 1963b. Revision of the North American phorid

flies. Part I. The Phorinae, Aenigmatiinae and Me-
topininae, except Megaselia (Diptera, Phoridae). Stu-
dia Entomologica 6:1-256.

. 1968. A catalogue of the Phoridae of the World

(Diptera, Phoridae). Studia Entomologica 11:1-367.

. 1969a. Bredin-Archbold-Smithsonian biological

survey of Dominica: The Phoridae of Dominica
(Diptera). Smithsonian Contributions to Zoology
23:1-69.

. 1969b. New or little-known phorid flies, mainly

of the Neotropical Region. Studia Entomologica 12:
33-132.

Borgmeier, T., and A. P. do Prado. 1975. New or little
known phorid flies with descriptions of eight new
genera (Dipt. Phoridae). Studia Entomologica 18:3—
90.

Brown, B. V. 1992a. Generic revision of Phoridae of the
Nearctic Region and phylogenetic classification of
Phoridae, Sciadoceridae and Ironomyiidae (Diptera:
Phoridea). Memoirs of the Entomological Society of
Canada 164:1-144.

. 1992b. Life history, immature stages and undes-
cribed male of Rhynchomicropteron (Diptera: Phor-
idae). Journal of Natural History 26:407-416.
. 1993. A further chemical alternative to critical-

Kung and Brown: Patterned-Wing Coniceromyia H 9

point-drying for preparing small (or large) flies. Fly
Times 11:10.

. 1996. Preliminary analysis of a host shift: Revi-
sion of the Neotropical species of Apocepbalus, sub-
genus Mesophora (Diptera: Phoridae). Contributions
in Science 462:1-36.

. 1998. New species and records of Gymnophora

Macquart (Diptera: Phoridae) from southeast Asia.
Contributions in Science 471:1-13.

Disney, R. H. L. 1989. Scuttle flies — Diptera, Phoridae,
Genus Megaselia. Handbooks for the Identification
of British Insects 10:1-155.

Gordh, G., and J. C. Hall. 1979. A critical point drier used
as a method of mounting insects from alcohol. En-
tomological News 90:57-59.

McAlpine, J. F. 1981. Morphology and terminology —
Adults. In Manual ofNearctic Diptera, vol. 1, ed. J.
F. McAlpine, B. V. Peterson, G. E. Shewell, H. J.
Teskey, J. R. Vockeroth, and D. M. Wood, 9-63.
Ottawa: Agriculture Canada.

Peterson, B. V. 1982. A new species of Coniceromyia (Dip-
tera: Phoridae) from Panama. Memoirs of the En-
tomological Society of Washington 10:136-138.

Peterson, B. V., and P. W. Arntfield. 1971. A new species
of Coniceromyia from Chiapas, Mexico. Studia En-
tomologica 14:395-398.

Prado, A. P. do. 1976. Records and descriptions of phorid
flies, mainly of the Neotropical Region (Diptera;
Phoridae). Studia Entomologica 19:561-609.

Schmitz, H. 1924. Mitteilungen iiber allerlei Phoriden. Nr.
9. Natuurhistorisch maandblad 13:148-150.

. 1927. Revision der Phoridengattungen, mit Bes-

chreibung neuer Gattungen und Arten. Natuurhis-
torisch maandblad 16:30-40, 45-50, 59-65, 72-79,
92-100, 110-116, 128-132, 142-148, 164, 176.

Stuckenberg, B. R. 1999. Antennal evolution in the Bra-
chycera (Diptera), with a reassessment of terminol-
ogy relating to the flagellum. Studia dipterologica 6:
33-48.

Received 12 July 1999; accepted 25 February 2000.

Natural Hi story Museum

of Los Angeles County
900 Exposition Boulevard
Los Angeles, California 90007