Content uploaded by Olga Otero
Author content
All content in this area was uploaded by Olga Otero on Oct 27, 2017
Content may be subject to copyright.
Zoological Journal of the Linnean Society
, 2006,
146
, 227– 237. With 10 figures
© 2006 The Linnean Society of London,
Zoological Journal of the Linnean Society,
2006,
146
, 227– 237
227
Blackwell Science, Ltd
Oxford, UK
ZOJZoological Journal of the Linnean Society
0024-4082The Lin-
nean Society of London, 2006? 2006
146
?
227237
Original Article
A NEW POLYPTERID FROM THE LATE MIOCENE OF CHADO. OTERO
ET AL.
*Corresponding author. E-mail: olga.otero@univ-poitiers.fr
A new polypterid fish:
Polypterus faraou
sp. nov.
(Cladistia, Polypteridae) from the Late Miocene,
Toros-Menalla, Chad
OLGA OTERO
1
*, ANDOSSA LIKIUS
2
, PATRICK VIGNAUD
1
and MICHEL BRUNET
1
1
Laboratoire de Géobiologie, Biochronologie et Paléontologie humaine, CNRS UMR 6046, Faculté des
Sciences fondamentales et appliquées, Université Poitiers, 40 av. du Recteur Pineau, F-86 022 Poitiers
cedex, France
2
Département de Paléontologie, Université de N’Djaména, N’Djaména, Tchad
Received June 2004; accepted for publication July 2005
Polypterus faraou
sp. nov.
(Cladistia, Polypteridae) from the Late Miocene of Toros-Menalla (western Djourab,
Chad) is described on the basis of a subcomplete articulated skeleton preserved in three dimensions. This is the first
time such a complete fossil polypterid skeleton has been described. It is the only verifiable fossil record for the genus
Polypterus
.
P. faraou
closely resembles
P. bichir
and
P. endlicheri
, extant fish found in the Chad–Chari system.
Intrarelationships among the polypterids are not yet resolved: however,
P. faraou
shows a primitive shape of the
body and a primitive shape of the opening of the lateral line on the scales, similar to that of three living species
(
P. bichir
,
P. endlicheri
and
P. ansorgii
). © 2006 The Linnean Society of London,
Zoological Journal of the Linnean
Society
, 2006,
146
, 227–237.
ADDITIONAL KEYWORDS: anatomy – Anthracotheriid Unit – Polypteriformes – western Djourab.
INTRODUCTION
Palaeontological expeditions have been conducted in
Chad since the early 1930s. In 1994, the MPFT
(Mission Paléoanthropologique Franco-Tchadienne)
initiated a series of field missions to the Mio-Pliocene
sites of the eastern Djurab Desert in northern Chad
(Brunet
et al
., 1995, 1998; Brunet & MPFT, 2000) and,
since 1997, to the Miocene sites of Toros-Menalla,
western Djurab. The Anthracoteriid Unit in Toros-
Menalla has yielded a rich vertebrate fauna (Vignaud
et al
., 2002), the oldest known species of hominid,
Sahelanthropus tchadensis
(Brunet
et al
., 2002), as
well as the first nearly complete and articulated skel-
eton of a fossil polypterid. This polypterid belongs to a
new species that is described here. The estimated age
of the Anthracotheriid Unit is
c
. 7 Myr (Vignaud
et al
.,
2002).
Polypterids (Cladistia) are basal actinopterygian
fishes (Rosen
et al
., 1981). They are represented by
two extant genera, i.e.
Polypterus
, with ten species and
15 subspecies (Poll, 1941a, b, 1942; Hanssens, Teugels
& Thys van der Audenaerde, 1995; Britz, 2004; there
are 11 species if
P. polli
is considered as a valid spe-
cies), and the monotypic genus
Erpetoichthys
. Nine
fossil genera have been described. Two have been
described on the basis of isolated skull bones and
scales (Meunier & Gayet, 1996; Gayet & Meunier,
1991), and six others have been described on the basis
of the articular head of pinnules (Gayet & Meunier,
1996; Werner & Gayet, 1997). Finally,
Serenoichthys
has been described on the basis of scale and finlet
counts and distribution (Dutheil, 1999).
Serenoichthys
is the only other fossil articulated skeleton known so
far, but the head is missing. These fossil genera are
Cretaceous in age and African, except for
Dagetella
and
Latinopollia
, which are of Late Cretaceous and
Early Palaeocene age, respectively, and South Ameri-
can. Post-Palaeocene polypterid history is exclusively
African. Isolated scales and skull bones attributed to
Polypterus
sp. testify to the presence of the genus in
African freshwaters since the Eocene. Moreover, two
228
O. OTERO
ET AL.
© 2006 The Linnean Society of London,
Zoological Journal of the Linnean Society,
2006,
146
, 227– 237
types of African Cretaceous pinnules were tentatively
attributed to two fossil
Polypterus
species,
P. dageti
(Gayet & Meunier, 1996) and
P. sudanensis
(Werner &
Gayet, 1997). The intrarelationships among poly-
pterids have yet to be clarified; indeed, the monophyly
and the intrarelationships among extant species have
not been established.
This paper provides the first description of a near
complete fossil polypterid skeleton including cranium,
and it is the only clearly recognizable fossil record for
the genus
Polypterus
. The new species from the Late
Miocene of Toros-Menalla (western Djurab, Chad) is
assigned to the genus
Polypterus
based on its morpho-
logical characteristics, which fall within the anatomi-
cal range of Recent species of the genus, from which it
differs by a unique combination of characters. We also
discuss the primitive or derived state of certain osteo-
logical features observed in this species.
SYSTEMATIC PALAEONTOLOGY
A
CTINOPTERYGII
C
OPE
, 1887
(
SENSU
R
OSEN
ET
AL
., 1981)
CLADISTIA
COPE
, 1871
P
OLYPTERIDAE
G
ÜNTHER
, 1870
P
OLYPTERUS
S
T
H
ILAIRE
, 1802
Type species. Polypterus bichir
St Hilaire, 1802.
P
OLYPTERUS
FARAOU
SP
.
NOV
.
Derivation of name
. In Chadian Arabian,
faraou
means flattened. The name is given in reference to the
dorso-ventrally depressed head of the fish.
Type specimen
. TM090-001-039, housed in the Centre
National d’Appui à la Recherche (CNAR, N’Djamena,
Chad).
Diagnosis
. A
Polypterus
species characterized by:
depressed head (as in
P. weeksii
and
P. endlicheri
only), head length approximately 20% of body length
(less than 20% in other
Polypterus
species except
P. ansorgii
,
P. bichir
and
P. endlicheri
) and head width
at the level of the preoperculum reaches 70% of its
length (more than in any other
Polypterus
species);
opening of the lateral line on the nasal 3 close to its
posterior border (as in
P. bichir
and
P. endlicheri
,
whereas it is central in other species); large and
rounded snout, shorter than the interorbital distance
(as in most
Polypterus
species except
P. bichir
and
P. palmas
the snout of which is longer than the
interorbital distance); superolateral position of the
orbit (as in
P. ansorgii
,
P. weeksii
,
P. endlicheri
and
P. bichir
, whereas the orbit is lateral in
P. delhezi
,
P. ornatipinnis
,
P. senegalus
,
P. retropinnis
and
P. palmas
); orbit length around the half the interor-
bital space (as in
P. bichir katangae
and possibly in
P. teugelsi
, whereas it is more than 60% in
P. ansorgii
,
and less than 45% in the other species) and orbit
smaller than suboperculum width (as in
P. bichir
and
P. endlicheri
only); prognathus lower jaw (as in
P. bichir
and
P. endlicheri
only); at least 5–7 prespi-
racular bones (the number of prespiracular bones is
possibly equal to or above seven only in
P. weeksii
, in
P. bichir bichir
and in
P. endlicheri congicus
); deep
body with approximately 23 transversal scales, 58–60
scale rows, and 14 predorsal scales (such body shape
exists elsewhere only in
P. ansorgii
,
P. bichir
and
P. endlicheri
); 15 finlets with successive pinnules of
the finlets overlapping the base of the following
one (such dorsal fin mophology elsewhere only in
P. ansorgii
,
P. bichir
and
P. endlicheri congicus
);
grooved lateral line scales with notched posterior mar-
gins (as in
P. ansorgii
,
P. bichir
and
P. endlicheri
only).
The holotype is 590 mm in standard length. Such
dimensions are also reached in
P. bichir
and
P. endlicheri
only.
Occurrence. Anthracotheriid Unit of Toros-Menalla,
Late Miocene of Chad.
MATERIAL AND METHODS
Specimen TM090-001-039 was found lying on its back.
It is preserved in three dimensions. The dorsal and
lateral body surfaces have globally kept their living
shape, whereas the ventral surface has collapsed
(Fig. 1A–C). Most of the scales of the body are present
and articulated with each other. All the bones of the
right half of the skull are in natural position, whereas
some bones of the left side are slightly displaced. This
three-dimension preservation allows us to observe the
natural position of the bones and the shape of the cra-
nium; it also allows us to measure the dimensions of
the specimen (Fig. 2) and to compare some dimensions
and meristic data between the specimen and Recent
species. Jollie’s (1984) nomenclature of the bones is
followed.
ANATOMICAL DESCRIPTION
HEAD AND BODY SHAPE (FIGS 1–3)
The standard length of the specimen is 59 cm (Fig. 2,
a). We observe a natural dorso-ventral depression of
the head. The head length (from the front of the mouth
to the posterior level of the opercular bone) is equiva-
lent to about 20% of the standard length (Fig. 2, m).
The width of the head, measured at the level of the
preoperculum, is around 70% of its length (Fig. 2, n).
The distance between the lateral border of the spira-
cular series at the level of the preoperculum is 45% of
the total length of the head (Fig. 2, o). The lower jaw is
A NEW POLYPTERID FROM THE LATE MIOCENE OF CHAD 229
© 2006 The Linnean Society of London, Zoological Journal of the Linnean Society, 2006, 146, 227 –237
prognathus. The general shape of the snout is wide
and rounded, and it is shorter than the interorbital
space is wide (Fig. 2, compare c and e). The orbit has a
superolateral position. The orbital length is 14% of the
head length (Fig. 2, p) and half the interorbital dis-
tance (Fig. 2, q). The interorbital distance is shorter
than the orbit–spiracle distance. Posteriormost nasals
(nasals 3) meet in the midline. The suboperculum is
deeper than the orbit length. The gular plates are
more than twice as long as wide. The dorsal finlets
insert on the posterior five-sixths of the body. The
axial skeleton is seen where the ventral scales are dis-
placed. The left pelvic fin is preserved, whereas only
the scaly base of the pectoral and anal fins and the
Figure 1. Photograph of specimen TM90-01-39, holotype of Polypterus faraou sp. nov., in A, dorsal, B, left lateral, and
C, ventral views.
Figure 2. Measurements in specimen TM90-01-39, holotype of Polypterus faraou sp. nov., and ratios used in the text.
230 O. OTERO ET AL.
© 2006 The Linnean Society of London, Zoological Journal of the Linnean Society, 2006, 146, 227 –237
proximal part of some caudal rays are preserved. The
position of the fins is given (Fig. 2, h–l).
SNOUT, ORBIT AND UPPER JAW (FIG. 4)
The INTERNASAL (median rostral) is the unpaired dia-
mond-shaped bone at the front of the head. It carries
the ethmoid commissure. It lies dorsal to the rostro-
premaxillae, and inserts laterally between the nasals
2, and posteriorly between the nasals 3.
The paired nasal bones and the nostrils are in place
on the left side but slightly displaced on the right.
NASAL 1 is a curved tubular bone, where the infraor-
bital canal runs from the rostro-premaxilla to a lateral
line pore posterior to the nostril. On the right side, a
small displaced piece of bone lying in the nostril open-
ing could be a fragment of the missing right nasal 1.
The rounded NASAL 2 lies lateral to the internasal and
carries the canal from the rostro-premaxillae to the
nasals 3. NASAL 3 is roughly square and joins its anti-
Figure 3. Photograph and drawing of the skull of TM90-01-39, holotype of Polypterus faraou sp. nov., in dorsal view.
Figure 4. Photograph and drawing of the snout of TM90-01-39, holotype of Polypterus faraou sp. nov., in anterolateral
dorsal view.
A NEW POLYPTERID FROM THE LATE MIOCENE OF CHAD 231
© 2006 The Linnean Society of London, Zoological Journal of the Linnean Society, 2006, 146, 227 –237
mere at the midline. There are no internasal super-
numerary plates (they develop frequently in some
living species, e.g. P. weeksii; Poll, 1942). The supraor-
bital sensory canal of either side passes posteriorly
through nasals 1, 2 and 3 to the frontals. The opening
from the canal within nasal 3 lies within the posterior
two-thirds of the bone.
On each side of the skull, the ROSTRO-PREMAXILLA
carries the infraorbital canal from the midline (inter-
nasal) to the maxilla, via the lacrymal. It extends
between the nostril and the orbit, along the ventral
margin of nasal 3, and sutures posteriorly with the
frontal. The ascending process of the rostro-premax-
illa is postero-ventrally covered by the LACRYMAL. The
lacrymal is a tubular bone that carries the infraorbital
canal from the rostro-premaxilla to the maxilla. The
lacrymal shows a preorbital process dorsally. The MAX-
ILLA reaches from the level of the nostril to the pres-
piracular series. The POSTORBITAL borders the back of
the orbit from the maxilla to the frontal, along the
leading edge of the prespiracular series. It is slightly
curved backwards in its caudal part, delimiting a free
narrow space with the lateral border of the frontal at
the postero-dorsal edge of the orbit. The infraorbital
canal runs from the lacrymal to the postorbital
through the maxilla and to the frontal through the
postorbital. On this latter bone, there is a tiny dorsal
opening whereas the ventral one, just above the max-
illa, is wide.
Nasal 3 is the only bone of the snout with a dermal
ornamentation of distinctly separated tubercles cov-
ered by ganoine. The other snout bones are smooth,
ganoine-free and with numerous small pores particu-
larly in the rostral part. Caniniform teeth are devel-
oped on the ventral face of the rostro-premaxilla in
front of the maxilla, and on the maxilla from the ante-
rior tip to the midlevel of the orbits.
DORSAL SIDE OF THE HEAD (FIGS 3 AND 4)
The FRONTALS are the largest bones of the head. The
frontal branch of the lateral canal has two openings
above the orbit (only one is seen on the right side of the
fossil). It connects with the infra-orbital branch in
front of the spiracular bones, at the level of the der-
mosphenotic. According to Jollie (1984), the canal is
shared by both the frontal and the dermosphenotic at
that point, and there is an opening on the dermosphe-
notic. Only the opening of the dermosphenotic can be
seen, but not the limits of the bone itself. It seems to be
absent or totally incorporated in the frontal. At the
postero-lateral angle of each frontal, the canal passes
to the INTERTEMPORO-SUPRATEMPORAL, where it opens
in a pore midway along the length of the bone. The
intertemporo-supratemporal is roughly rectangular
in shape. There are six EXTRASCAPULAR PLATES. Four
are aligned along the posterior border of the inter-
temporo-supratemporals, and two are lateral. The
last-mentionned are diamond-shaped, and they lie
between the post-spiracular bones, the more medial
extrascapular plates and the posttemporal. There are
openings of the lateral line between the extrascapular
plates and between the lateral extrascapular plate
and the posttemporal. The POSTTEMPORAL is roughly
rectangular in shape and shows a posterior opening
for the lateral-line canal. The posterior process is not
seen. It is probably broken on the left side where the
scales are slightly displaced (it is present in living
Polypterus but hidden by the scales). The posterior
midline scale is missing.
The PRESPIRACULAR, SPIRACULAR and POSTSPIRACU-
LAR bones border the lateral edge of the former dorsal
bones of the skull, from the postorbital to the posttem-
poral, above the opercular series. The spiracular
bones lie along the postero-lateral border of the
frontal and the lateral border of the intertemporo-
supratemporal. The spiracular formulae are, respec-
tively, 5–2−3 on the left side, and 7–2−4 on the right.
The SPIRACLE opens along the second spiracular bone,
above the DERMOHYAL, the dorsal face of which can be
seen.
All of the skull roofing bones have a dermal orna-
mentation of distinctly separated tubercles covered by
ganoine on their dorsal face.
ORBITOSPHENOID, ECTOPTERYGOID AND LATERAL
ETHMOID (FIG. 4)
Parts of these three bones and their spatial relation-
ships are seen in the left orbit of the specimen.
OPERCULAR SERIES AND CHEEK (FIG. 5)
The opercular series is well developed, with a large
SUBOPERCULUM the dorsal extremity of which reaches
the preorbital canal median opening, and an OPERCU-
LUM as large as the PREOPERCULUM. The posteroven-
tral process of the preoperculum is present. Two
SUPERNUMERARY CHEEK PLATES lie anterior to the
posteroventral process, along the preoperculum ven-
tral border. Together with the process, they totally
cover the cheek. Anteriorly, the preoperculum is
sutured with the maxilla. The preopercular sensory
canal has four openings. The first is dorsal and is
located between the spiracular and postspiracular
series. The second opening lies close to the posterior
border of the bone at one-quarter of its height. The
third is in a more ventral position, just above the ven-
tral process of the preoperculum. The fourth opening
lies at the lower extremity of the ventral process.
All of the cheek bones have their surface covered by
the ganoine tubercules.
232 O. OTERO ET AL.
© 2006 The Linnean Society of London, Zoological Journal of the Linnean Society, 2006, 146, 227 –237
LOWER JAW AND GULAR AREA (FIG. 6)
The lower jaw is prognathus. The posterior level of the
suture between the DENTOSPLENIAL and the ANGULAR
reaches the anterior level of the supernumerary cheek
plates. The angular–prearticular–articular sutures
are not seen, and nor is the quadrate–articular arti-
culation; they are hidden by the gular and the cheek
supernumerary plates. The coronoids and mento-
meckelian bones are not observable, owing to their
internal position and small dimension. Caniniform
teeth are developed on the edge of the dentosplenial.
Figure 5. Photograph and drawing of the opercular series and cheek of TM90-01-39, holotype of Polypterus faraou sp.
nov., in laterodorsal view.
Figure 6. Photograph and drawing of the skull and pectoral girdle of TM90-01-39, holotype of Polypterus faraou sp.
nov., in ventral view.
A NEW POLYPTERID FROM THE LATE MIOCENE OF CHAD 233
© 2006 The Linnean Society of London, Zoological Journal of the Linnean Society, 2006, 146, 227 –237
Ventrally, the wide and long GULAR PLATES are
mostly covered by a thin ganoin ornamentation except
in their anteriormost parts. They cover nearly the
whole ventral surface between the lower jaws.
Because they are slightly displaced, a medial bone is
observable which could be the urohyal.
PECTORAL GIRDLE AND BASE OF THE FINS (FIGS 6, 7)
The pectoral girdle is composed of stout bones. Dor-
sally and posteriorly, the SUPRACLEITHRUM borders
the operculum and it covers the first scale row. Later-
ally, the POSTCLEITHRUM is probably represented by a
bony plate broken in three fragments. It is delicately
ornamented with smooth ganoine tubercles, and it is
lying above an aggregation of small scales which is the
lateral edge of the base of the pectoral fin. Ventrally,
the CLEITHRA join in the midline. They are covered
anteriorly by large clavicles that articulate with one
another by two small processes and a notch.
SCALES AND FINLETS (FIG. 8), AXIAL SKELETON
There are 15 pinnules bearing the finlets. The first
inserts behind the 14th row of scales. Each pinnule is
separated from the following one by three scale rows.
They are long enough to cover the base of the following
one when folded in life. The pair of scales with which
each pinnule articulates are L-shaped. The distal
extremity of each pinnule is forked and the articular
base is symmetrical. On one pinnule, at least two fin
rays can be seen. The caudal fin starts dorsally behind
the last finlet at the 48th scale row, and ventrally at
the 53rd scale row. The first six dorsal and first seven
ventral caudal rays are preserved in their proximal
part. All scales have exposed surface covered by
ganoin. There are 23 scales on each transversal row
across the body on either side: most of the scales of the
lateral line bear a groove that opens posteriorly in a
notch, and some present a simple pore. There are eight
scales above and 14 below the lateral line in a trans-
versal row. There are 58–60 longitudinal scales, and
thus probably about 60 vertebrae (Daget & Desoutter,
1983). This agrees with the count of centra made
where the covering scales are displaced. The centra
are typical, rounded with one median and two lateral
deep gutters ventral to the lateral apophysis.
PELVIC FIN (FIG. 9) AND ANAL FIN
The scaly base of the left pelvic fin and the proximal
part of the rays are preserved. The fin inserts behind
the 31st transversal row of scales. There are at least
11 pelvic rays. The position of the anal fin is assumed
based on the preceding small square scales known in
living polypterids.
SYSTEMATIC AFFINITIES AND DISCUSSION
Polypterus faraou has a character unique to the cla-
distians: the dorsal fin is composed of several finlets,
each consisting of a pinnule articulated with one or
Figure 7. Photograph and drawing of the operculum, cleithrum and postcleithrum of TM90-01-39, holotype of Polypterus
faraou sp. nov., in lateral view.
Figure 8. Photograph and drawing of pinnules 1–4 in place with the surrounding scales of TM90-01-39, holotype of
Polypterus faraou sp. nov., in dorsal view.
234 O. OTERO ET AL.
© 2006 The Linnean Society of London, Zoological Journal of the Linnean Society, 2006, 146, 227 –237
more soft rays. It also shows polypteriform character-
istics such as absence of branchiostegals and a maxilla
that carries the infraorbital sensory canal. It has
polypterid features: infraorbital canals fused to the
maxilla and fusion of lateral rostral bones to the pre-
maxilla (Pehrson, 1947; Jollie, 1984), the presence of
spiracular bones, the preoperculum covering the cheek
and three separate nasal bones. The lateral-line sys-
tem on the head of Pol ypterus faraou conforms to the
common pattern described by Jollie (1984) for living
polypterids.
Polypterus faraou has symmetrical pinnules as in
Recent Polypterus species and in Erpetoichthys,
whereas a different structure of the articular head of
the pinnules characterized the fossil genera described
by Gayet & Meunier (1996) and Werner & Gayet
(1997) (see Introduction). It also differs from Ser-
enoichthys, in which the body is much shorter than
in Recent genera (Dutheil, 1999). However, it does
not show the elongation of the body as observed in
Erpetoichthys, nor the reductive characters of that
genus, e.g. no suboperculars and no supernumerary
cheek plates (Smith, 1865; Poll, 1941b; O. Otero, pers.
observ.). Moreover, at least one of the pinnule bears at
least two rays as in Recent Pol ypterus species, by con-
trast with Erpetoichthys, the pinnules of which bear
one single fin ray (Daget, 1950; O. Otero, pers. observ.).
From this evidence, we assign the new species to the
genus Polypterus.
Poll (1941a, b, 1942, 1965) described the Recent
polypterid species and defined six anatomically based
clusters of species and subspecies organized on ana-
tomical trends deriving from a central type (Fig. 10A).
The possible phylogenetic significance of these clusters
has not been established; they only have a descriptive
value. Polypterus faraou belongs to the P. bichir –
P. endlicheri cluster with: (1) head length about 20% of
body length; (2) head depressed; (3) prognathus jaw; (4)
opening of the lateral line on the nasal 3 close to its pos-
terior border; (5) orbit length smaller than the subo-
perculum width; (6) successive pinnules overlapping
the base of the following ones when folded in life; (7)
grooved lateral line scales with notched posterior mar-
gin. It also has other characteristics observed in
P. bichir, P. endlicheri and also in some other species:
(1) superolateral position of the orbit (P. ansorgii,
P. weeksii); (2) 5–7 prespiracular bones (P. ornatipin-
nis, P. weeksii); (3) 15 finlets (P. ansorgii); (4) deep body,
with 46 transversal scales and 58 scale rows
(P. ornatipinnis, P. ansorgii, P. weeksii); (5) only 14 pre-
dorsal scales (P. delhezi); (6) known maximum stan-
dard length of 590 mm (P. ornatipinnis). Moreover,
Polypterus faraou differs from P. bichir and resembles
P. endlicheri based on its depressed head, and large
and rounded snout which is shorter than the interor-
bital distance; it differs from P. endlicheri and resem-
bles P. bichir katangae based on its orbital length,
which is 50% of the interorbital space in adults (less
than 40% in adults of P. endlicheri subspecies and
P. bichir subspecies bichir and lapradei). Finally,
Polypterus faraou is also the only Polypterus species
with such a wide head: the head width at the level of
the preoperculum is about 70% of the length; this can
also be expressed by the distance between the lateral
borders of the spiracular series (at the level of the pre-
operculum), which is 45% of the head length whereas
it is around one-third in large specimens of P. bichir.
In conclusion, the new taxon is assigned to the
genus Polypterus based on its morphological charac-
teristics, which fall into the range observed in the spe-
cies of the genus, particularly the cluster P. endlicheri
– P. bichir, from which it differs by a unique combina-
tion of characters. The suggestion that the cluster
formed by P. faraou and the Recent P. endlicheri and
P. bichir (Fig. 10A) is a natural group will need to be
tested as intrarelationships among Polypterus species
are still unknown. Nevertheless, this hypothesis is in
agreement with the distribution of extant species:
among the three polypterid fishes living in the Chad–
Chari system, P. endlicheri and P. bichir are present
Figure 9. Photograph and drawing of the left pelvic fin of TM90-01-39, holotype of Polypterus faraou sp. nov., in lateral
view.
A NEW POLYPTERID FROM THE LATE MIOCENE OF CHAD 235
© 2006 The Linnean Society of London, Zoological Journal of the Linnean Society, 2006, 146, 227 –237
(Fig. 10B), as represented by P. endlicheri endlicheri
and P. bichir bichir; the third one is P. senegalus
(Gosse, 1990). However, the phylogenetic relation-
ships of Recent polypterids must be established before
we can draw any biogeographical inferences. A cladis-
tic analysis on the basis of anatomical characters will
require further anatomical review of the species,
which seem morphologically so close to each other and
differ mainly in their combination of characteristics.
Moreover, difficulty exists in choosing an outgroup
because (1) the closely related Erpetoichthys must be
included as an ingroup in such an analysis (this mono-
typic genus is characterized by reductive characters
and the elongation of the body; it thus could be more
closely related to one of the species of Polypterus), and
(2) the related fossils are only known by their pinnules
and/or scales, except Serenoichthys (see Introduction).
However, two anatomical characters have been dis-
cussed previously in terms of primitive/derived states,
and for the first time they are both observed in a fossil
attributed to the genus Polypterus:
1. The scales of the lateral line bear a groove that
opens posteriorly in a notch and this is regarded as
a primitive state for polypteriforms. The derived
state is thought to be lateral line pores piercing the
body of the scale. This assumption was made by
Poll on the basis of ontogenetic arguments.
P. faraou shows this character state, as do P. bichir,
P. endlicheri and P. ansorgii.
2. A deep and short body (low vertebrae count) is a
primitive state for polypteriforms. This assumption
(Daget & Desoutter, 1983) is congruent with the
body shape of the Cretaceous Serenoichthys (as
noted by Dutheil, 1999), and also with the general
observations made on the elongation of the body
among actinopterygians. P. faraou has a short and
deep body with 46 scales in a transversal row, as
Figure 10. The Recent polypterid species and Polypterus faraou sp. nov. (A) Position of the species in the anatomically
based clusters of species, as defined by Poll, P. teugelsi not included (see Britz, 2004, for discussion), and (B) distribution in
the African hydrographical basins; data from FishBase (Froese & Pauly, 2005) and map of the ichthyofaunic regions of
Africa from Roberts (1975).
236 O. OTERO ET AL.
© 2006 The Linnean Society of London, Zoological Journal of the Linnean Society, 2006, 146, 227 –237
in P. bichir, P. endlicheri, P. weeksii and P. ansorgii,
and with fewer than 60 scales in a longitudinal row,
as in all Recent species except P. ornatipinnis,
P. bichir and Erpetoichthys. We can also consider
the trend to anguilliform shape through the posi-
tion and the count of the dorsal finlets. On this
basis, P. faraou has no anguilliform character, the
number of pinnules being greater than 12, and the
count of predorsal scales being lower than 15 (as in
P. bichir, P. endlicheri and P. ansorgii).
The polypterid from Toros-Menalla (western
Djurab, Chad) described here is the first fossil speci-
men known by a near complete articulated skeleton
including a cranium. It is assigned to the genus
Polypterus and recognized as a new species, Polypterus
faraou. It is c. 7 Myr in age (Vignaud et al., 2002).
ACKNOWLEDGEMENTS
We thank the Chadian authorities (Ministère de
l’Education Nationale et de la Recherche, Université
de N’Djaména, CNAR). For their support, we extend
our gratitude to the French Ministère de l’Enseigne-
ment supérieur, de la Recherche et de la Technologie
(CNRS; Université de Poitiers) and the Ministère
des Affaires Etrangères (DCSUR, Paris; Coopération
SCAC, N’Djaména), to the Région Poitou-Charentes
and the Département de la Vienne. The MPFT field-
work is also supported by an NSF/RHOI Grant. We
thank Emmanuel Fara (CNRS, Poitiers) for discus-
sion, the referees for their reviews and comments, and
also all members of the MPFT. Drawings and photo-
graphs by Olga Otero.
REFERENCES
Britz R. 2004. Polypterus teugelsi, a new species of bichir from
the Upper Cross River system in Cameroon (Actinopterygii:
Cladistia: Polypteridae). Ichthyological Explorations of
Freshwaters 15: 179–186.
Brunet M, MPFT. 2000. Chad: discovery of a vertebrate
fauna close to the Mio-Pliocene boundary. Journal of Verte-
brate Paleontology 20: 205–209.
Brunet M, Beauvilain A, Coppens Y, Heintz E, Moutaye
EHA, Pilbeam D. 1995. The first australopithecine 2500
kilometers west of the Rift Valley (Chad). Nature 378: 273–
275.
Brunet M, Beauvilain A, Geraards D, Guy F, Kasser M,
Mackaye HT, MacLatchy L, Mouchelin G, Sudre J,
Vignaud P. 1998. Tchad: découverte d’une faune de
mammifères du Pliocène inférieur. Comptes Rendus de
l’Académie des Sciences, Paris, Série IIa 326: 153–158.
Brunet M, Guy F, Pilbeam D, Mackaye HT, Likius A,
Djimdoumalbaye A, Beauvilain A, Blondel C, Bocher-
ens H, Boisserie JR, de Bonis L, CoppensY, Dejax J,
Denys C, Duringer P, EisenmannV, Gongdibé F,
Fronty P, Geraads D, LehmannT, Lihoreau F,
Louchart A, Adoum M, Merceron G, Mouchelin G,
Otero O, Pelaez Campomanes P, Ponce de Leon M,
Rage JC, Sapanet M, Schuster M, Sudre J, Tassy P,
Valentin X, Vignaud P, Viriot L, Zazzo A. 2002. A new
hominid from the Upper Miocene of Chad, Central Africa.
Nature 418: 145–151.
Daget J. 1950. Révision des affinités phylogénétiques des
Polyptéridés. Mémoires de l’Institut Français d’Afrique Noire
11: 1–178.
Daget J, Desoutter M. 1983. Essais de classification cladis-
tique des polyptéridés (Pisces; Brachiopterygii). Bulletin du
Muséum National d’Histoire Naturelle, Paris, Série A 4: 661–
674.
Dutheil D. 1999. The first articulated fossil cladistian: Ser-
enoichthys kemkemensis gen. et sp. nov., from the Creta-
ceous of Morocco. Journal of Vertebrate Paleontology 19:
243–246.
Froese R, Pauly D (eds). 2005. FishBase. World Wide Web
electronic publication. www.fishbase.org.
Gayet M, Meunier F. 1991. First discovery of Polypteridae
(Pisces, Cladistia, Polypteriformes) outside of Africa. Géobios
24: 463–467.
Gayet M, Meunier F. 1996. Nouveaux Polypteriformes du
Coniacien-Sénonien d’In Becetem (Niger). Comptes Ren-
dus de l’Académie des Sciences, Paris, Série IIa 322: 701–
707.
Gosse J-P. 1990. Polypteridae. In: Lévêque C, Paugy D, Teu-
gels GG, eds. Faune des poissons d’eaux douces et saumâtres
d’Afrique de l’Ouest I. MRAC Orstom, collection faune trop-
icale, 79–87.
Hanssens MM, Teugels GG, Thys van den Audenaerde
DFE. 1995. Suspecies in the Polypterus palmas complex
(Brachiopterygii; Polypteridae) from West and Central
Africa. Copeia 1995: 694–705.
Jollie M. 1984. Development of the head and pectoral skeleton
of Polypterus with a note on scales (Pisces: Actinopterygii).
Journal of Zoology, London 204: 469–507.
Meunier F, Gayet M. 1996. A new Polypteriform from the
Late Cretaceous and the middle Paleocene of South America.
In: Arratia G, Viohl G, eds. Mesozoic Fishes, 1. Systematics
and paleoecology. Proceedings of the 1st International Meet-
ing on Mesozoic Fishes, Eichstätt, 1993. München, Germany:
Verlag Dr. F. Pfeil, 93–103.
Pehrson T. 1947. Some new interpretations of the skull in
Polypterus. Acta Zoologica, Stockholm 28: 241–258.
Poll M. 1941a. Contribution à l’étude systématique des Poly-
pteridae (première partie). Revue de Zoologie et de Botanique
Africaines 35: 141–179.
Poll M. 1941b. Les tendances évolutives des Polyptères
d’après l’étude systématique des espèces. Annales de la
Société Royale Zoologique de Belgique 72: 157–173.
Poll M. 1942. Contribution à l’étude systématique des Poly-
pteridae (suite et fin). Revue de Zoologie et de Botanique
Africaines 35: 269–308.
Poll M. 1965. Anatomie et systématique des Polyptères. Bul-
letin de l’Académie Royale de Belgique, Sciences 1965: 553–
569.
A NEW POLYPTERID FROM THE LATE MIOCENE OF CHAD 237
© 2006 The Linnean Society of London, Zoological Journal of the Linnean Society, 2006, 146, 227 –237
Roberts TR. 1975. Geographical distribution of African fresh-
water fishes. Zoological Journal of the Linnean Society 57:
249–319.
Rosen DE, Forey PL, Gardiner B, Patterson C. 1981.
Lungfishes, tetrapods, paleontology and plesiomorphy. Bul-
letin of the American Museum of Natural History 167: 159–
276.
Smith JA. 1865. Description of Erpetoichthys, a new genus of
ganoid fish, from Old Calabar, Western Africa; forming an
addition to the Family Polypterini. Proceedings of the Royal
Society of Edinburgh 1865–1866: 654–659.
Vignaud P, Duringer P, Mackaye HT, Likius A, Blondel
C, Boisserie JR, de Bonis L, Eisenmann V, Geraads
D, Guy F, Lehmann T, Lihoreau F, Lopez-Martinez
N, Mourer-Chauviré C, Otero O, Rage J-C, Schuster
M, Viriot L, Zazzo A, Brunet M. 2002. Geology and
palaeontology of the Upper Miocene Toros-Menalla fossilif-
erous area, Djurab Desert, Northern Chad. Nature 418:
152–155.
Werner C, Gayet M. 1997. New fossil Polypteridae from the
Cenomanian of Sudan. An evidence of their high diversity in
the Early Late Cretaceous. Cybium 21: 67–81.