Content uploaded by Olivier Lambert
Author content
All content in this area was uploaded by Olivier Lambert on Apr 14, 2017
Content may be subject to copyright.
BULLETIN DE L’INSTITUT ROYAL DES SCIENCES NATURELLES DE BELGIQUE
BULLETIN VAN HET KONINKLIJK BELGISCH INSTITUUT VOOR NATUURWETENSCHAPPEN
SCIENCES DE LA TERRE, 78: 277-316, 2008
AARDWETENSCHAPPEN, 78: 277-316, 2008
Sperm whales from the Miocene of the North Sea: a re-appraisal
by Olivier LAMBERT
LAMBERT, O., 2008 – Sperm whales from the Miocene of the North
Sea: a re-appraisal. In: STEURBAUT, E., JAGT, J.W.M. & JAGT-
YAZYKOVA, E.A. (Editors), Annie V. Dhondt Memorial Volume.
Bulletin de l’Institut royal des Sciences naturelles de Belgique,
Sciences de la Terre, 78: 277-316, 20 gs, 1 table, Brussels, October
31, 2008 – ISSN 0374-6291.
Abstract
A review of the sperm whales (Cetacea, Odontoceti, Physeteroidea)
from the Miocene of the southern margin of the North Sea Basin
is undertaken, mostly based on the cranial material of the area of
Antwerp (north of Belgium) described for the rst time by P.-J.
Van Beneden, B. A. L. du Bus, and O. Abel more than a century
ago. This work leads to the detailed redescription of the species
Eudelphis mortezelensis, Physeterula dubusi, Placoziphius duboisi,
and Thalassocetus antwerpiensis, the identication in the North
Sea Basin of the eastern North American species Orycterocetus
crocodilinus, and the description of a new undetermined physeterid.
The stratigraphic information associated to some of these Miocene
taxa is rened (E. mortezelensis, O. crocodilinus, and P. duboisi),
whereas a more important incertitude persists for others. These
results further emphasize the physeteroid diversity during the
Miocene.
The performed phylogenetic analysis places Eudelphis as the
most basal stem-physeteroid, displaying the most salient features of
the superfamily (supracranial basin, strong asymmetry of the bony
nares and premaxillae) but retaining enamel on teeth and a rather
conservative skull morphology (deep maxillary alveoli, large left pre-
maxillary foramen, distinct falciform process of the squamosal…).
Together with Orycterocetus, Placoziphius is provisionally kept
outside the family Physeteridae, dened as the clade grouping all
the physeteroids more closely related to Physeter than to Kogia.
The large species Physeterula dubusi is the most stemward
physeterid, retaining functional upper teeth lacking enamel. Among
the physeterids the new undetermined taxon is sister-group to the
clade Aulophyseter + Physeter, sharing with these two genera the
preorbital process distinctly lower than the lateral margin of the
rostrum base. With a sagittal crest in its supracranial basin the small
Thalassocetus antwerpiensis is conrmed as an archaic kogiid.
The evolutionary history of the supracranial basin and the oral
apparatus are discussed. The parsimony analysis suggests that the
spermaceti organ remained small in the supracranial basin of most
physeteroids, the basin probably functioning as a parabolic structure
for reecting and focusing the echolocative sounds. It is proposed
that the spermaceti organ only considerable increased in size in the
lineage of the Recent species Physeter macrocephalus, possibly as
a sexually dimorphic sound transmitting organ. Preceded by the
loss of enamel, the reduction of the upper dentition associated to a
decrease of the size of the temporal fossa occurred in parallel in the
physeterids and the kogiids, much likely related to a major change
in diet and food processing.
Keywords: Odontoceti, Physeteroidea, sperm whale, Miocene,
North Sea, phylogeny.
Résumé
La révision des cachalots (Cetacea, Odontoceti, Physeteroidea) du
Miocène du bord sud du Bassin de la Mer du Nord est entreprise,
principalement sur base du matériel crânien de la région d’Anvers
(nord de la Belgique) décrit pour la première fois il y a plus d’un
siècle par P.-J. Van Beneden, B. A. L. du Bus, et O. Abel. Ce
travail mène à la redescription des espèces Eudelphis mortezelensis,
Physeterula dubusi, Placoziphius duboisi, and Thalassocetus
antwerpiensis, à l’identication dans le Bassin de la Mer du Nord
d’Orycterocetus crocodilinus, espèce connue auparavant de la côte
est des Etats-Unis, et à la description d’un nouveau physétéridé
indéterminé. L’information stratigraphique associée à certains de ces
taxons (E. mortezelensis, O. crocodilinus, et P. duboisi) est afnée,
alors qu’une plus grande incertitude persiste pour les autres taxons
étudiés. Ces résultats soulignent la diversité des physétéroidés
durant le Miocène.
L’analyse phylogénétique entreprise positionne Eudelphis
comme le plus basal des ‘stem’-physétéroidés. Celui-ci montre
les caractères principaux de la super-famille (bassin supracrânien,
forte asymétrie des narines osseuses et des prémaxillaires), mais
il conserve de l’émail sur les dents et une morphologie crânienne
relativement archaïque (alvéoles maxillaires profondes, grand
foramen prémaxillaire gauche, processus falciforme du squamosal
développé…). Avec Orycterocetus, Placoziphius se place ici en
dehors de la famille Physeteridae, dénie comme le clade groupant
tous les physétéroidés plus proches parents de Physeter que de
Kogia. La grande espèce Physeterula dubusi constitue le plus basal
des physétéridés, conservant des dents fonctionnelles supérieures
sans émail. Parmi les physétéridés, le nouveau taxon indéterminé
est le groupe-frère du clade Aulophyseter + Physeter, partageant
avec ces derniers le processus préorbitaire nettement plus bas que
le bord latéral du rostre à sa base. Avec une crête sagittale dans son
278 Olivier LAMBERT
bassin supracrânien, le statut de kogiidé archaïque de Thalassocetus
antwerpiensis est conrmé.
L’histoire évolutive du bassin supracrânien et de la région orale
sont commentés. L’analyse de parcimonie suggère que l’organe du
spermaceti a conservé une taille réduite dans le bassin supracrânien
de la plupart des physétéroidés, le bassin fonctionnant probablement
comme une structure parabolique permettant la réexion et la
focalisation des sons d’écholocalisation. La taille de l’organe du
spermaceti n’aurait augmenté signicativement que dans la lignée de
l’espèce actuelle Physeter macrocephalus, peut-être comme organe
de transmission des sons sexuellement dimorphique. Précédée de
la perte de l’émail dentaire, la réduction de la dentition supérieure
associée à la diminution de taille de la fosse temporale se produit en
parallèle chez les physétéridés et les kogiidés, vraisemblablement
en relation avec un changement majeur dans l’alimentation et le
traitement de la nourriture.
Mots-clefs: Odontoceti, Physeteroidea, cachalot, Miocène, Mer du
Nord, phylogénie.
Introduction
The three Recent sperm whale species roughly occupy
the limits of the range of sizes for all the known
physeteroids. Physeter macrocephalus is the largest,
with adult males reaching 18 meters; the two species
of Kogia are among the smallest, with a mean body
length of only 2.7 meters in Kogia simus (CALDWELL
& CALDWELL, 1989; RICE, 1989). Besides the lack of
upper functional teeth and a diet mostly predominated
by squid, the main cranial feature grouping these
three species and differentiating them from all other
odontocetes is the large supracranial basin, containing
organs related to the production and transmission of
sounds (review of the homologies with the forehead
morphology of other odontocetes in CRANFORD et al.,
1996). The presence of a supracranial basin can be
traced in the fossil record back to the early Miocene,
with two species from Patagonia (MORENO, 1892;
LYDEKKER, 1893). Older presumably physeteroid
material, from the late Oligocene of Azerbaijan is too
fragmentary to infer the presence of a supracranial basin
(MCHEDLIDZE, 1970).
Besides the dental and fragmentary non-diagnostic
material (e.g. Scaldicetus carretti DU BUS, 1867, S.
grandis DU BUS, 1872, Prophyseter dolloi ABEL,
1905), four physeteroid species from the Miocene
of Antwerp, north of Belgium, southern margin of
the North Sea Basin, were described on the basis of
more diagnostic skull elements: the large Physeterula
dubusi VAN BENEDEN, 1877, the moderate size
Eudelphis mortezelensis DU BUS, 1872, the smaller
Placoziphius duboisi VAN BENEDEN, 1869, and the
even smaller Thalassocetus antwerpiensis ABEL, 1905.
The main specimens were found during the building
of fortications around the city of Antwerp, mostly
between 1861 and 1863 (VAN BENEDEN, 1861; DU BUS,
1867; VANDEN BROECK, 1874; ABEL, 1905) in several
localities of the south-east suburbs of the city (Fig. 1).
A. Map of the north of Belgium giving the position
of Antwerp and Kessel on the southern margin of
the North Sea Basin. B. Map of Antwerp and its
suburbs with the two fortication rings dating from
the 19th century (modied from VANDEN BROECK,
1874). The Miocene localities where physeteroid
remains were found are indicated by arrows: Ft
4. Fort n°4, Oude God (Vieux-Dieu), Mortsel
(Eudelphis mortezelensis). G. Ancien fortin n°1
(Physeterula dubusi). M. Briqueterie de la société
Pauwels, Edegem (Placoziphius duboisi). S3. 3ème
section (inner fortication ring) (Physeteridae
indet. IRSNB M.1937). V. Veldekens, Berchem
(Placoziphius duboisi). One locality (Veldekens)
yielded remains some years ago, whereas the
others were exploited during the years 1861-1863.
Fig. 1 –
279
Miocene sperm whales from the North Sea
These species have been the subject of comments
in various works dealing with other physeteroids
(KELLOGG, 1925, 1927, 1965; MUIZON, 1991; HIROTA
& BARNES, 1995; KAZÁR, 2002; BIANUCCI & LANDINI,
2006), but most authors were reluctant to use them (or
part of them) as comparison material or in phylogenetic
analyses of the superfamily Physeteroidea, probably
for two main reasons: (1) the lack of good quality
illustrations and detailed descriptions in the original
papers and (2) the relatively fragmentary state of
the type material (actually needing also some re-
preparation work). An additional weak side of these
specimens is the lack of precise stratigraphic data
associated. This last point can be partly solved with a
revision of the stratigraphy of the old localities and with
the addition of more recently discovered specimens
from new localities (Veldekens, Berchem and fort of
Kessel, 16 km south-east-east to Antwerp). Following
the re-preparation of the specimens the aims of this
paper are therefore: (1) to redescribe in detail these
specimens, with correct illustrations, (2) to compare
them with more recently described physeteroid taxa, (3)
to clarify the stratigraphy for some of the species, and
(4) to include these species in a phylogenetic analysis
of the superfamily Physeteroidea. These systematic and
phylogenetic sections will be followed by comments
on functional anatomy, mostly focused on the sound
producing area (supracranial basin) and the feeding
apparatus.
Material and methods
Institutional abbreviations. IRSNB: Institut royal des Sciences
naturelles de Belgique, Brussels, Belgium; MNHN: Muséum
national d’Histoire naturelle, Paris, France; USNM: United
States National Museum of Natural History, Washington
DC, USA; ZMA: Zoölogisch Museum Amsterdam, The
Netherlands.
Measurements. Due to the fragmentary and dislocated state of
preservation of most of the specimens the few measurements
that could be taken are placed directly in the text of the
description.
Taxonomic names. In a way to lighten the text the species
names for monospecic genera are omitted. In the genus
Orycterocetus only the species O. crocodilinus was
considered for comparison. Additionally author names
of fossil physeteroid species from localities outside
the North Sea mentioned in the text are provided here:
Aprixokogia kelloggi WHITMORE & KALTENBACH, 2008;
Aulophyseter morricei KELLOGG, 1927; ‘Aulophyseter’
rionegrensis GONDAR, 1975; Brygmophyseter shigensis
(HIROTA & BARNES, 1995); Diaphorocetus poucheti
(MORENO, 1892); Idiophyseter merriami KELLOGG, 1925;
Idiorophus patagonicus (LYDEKKER, 1893); Orycterocetus
crocodilinus COPE, 1868; Praekogia cedrosensis BARNES,
1973; Scaphokogia cochlearis MUIZON, 1988; Zygophyseter
varolai BIANUCCI & LANDINI, 2006. For further details on the
age and locality of these species, see the review in BIANUCCI
& LANDINI (2006).
Non-diagnostic material from Antwerp area. Based on three
fragments of the anterior part of a large rostrum, ABEL (1905)
erected the species Prophyseter dolloi. The author provided
two arguments to identify this species as a sperm whale: the
transverse compression of the anterior of the rostrum and the
rst step towards a reduction of the upper teeth, indicated
by the partial occlusion with bony outgrowths of the alveoli
(specially the premaxillary alveoli). Even if, mostly because
of the size, the physeteroid hypothesis is better supported, a
similar trend towards the reduction of the upper (and lower)
teeth is observed in Ziphiidae (MUIZON, 1991; LAMBERT,
2005). Furthermore several ziphiids display a transversely
compressed rostrum. As this specimen does not match the
dimensions of the rostrum and/or alveoli of other known
physeteroids from the North Sea, P. dolloi is considered
Odontoceti aff. Physeteroidea and it will not be considered in
detail farther in the text.
The status of the tooth-based species Scaldicetus caretti
DU BUS, 1867 (type species) and S. grandis (DU BUS, 1872) is
briey commented on in the section Systematic palaeontology.
More recently, a fragmentary physeteroid skeleton from
the Miocene of Antwerp has been described (PETERS &
MONTEIRO, 2005), but due to the limited information
available only rough similarities with Orycterocetus
crocodilinus and Physeterula dubusi could be noted. Even
more recently an isolated tooth from the Neogene of Antwerp
previously identied as a pinniped canine (Paleophoca nystii
VAN BENEDEN, 1859) was corrected as a physeteroid tooth
(KORETSKY & RAY, 2008).
Systematic palaeontology
Order Cetacea BRISSON, 1762
Suborder Odontoceti FLOWER, 1867
Superfamily Physeteroidea GRAY, 1821
Genus Eudelphis DU BUS, 1872
Type species: E. mortezelensis DU BUS, 1872
Diagnosis. The diagnosis is the same as for the only in-
cluded species.
280
Eudelphis mortezelensis DU BUS, 1872
Figs 2-6
1872 — Eudelphis mortezelensis DU BUS, p. 500.
1874 — Eudelphis mortselensis VAN DEN BROECK, p. 146.
1898 — Delphinus mortezelensis TROUESSART, p. 1032.
1905 — Scaldicetus mortselensis ABEL, p. 66; g. 5.
2006 — Eudelphis mortselensis BIANUCCI & LANDINI, p. 104.
Holotype
IRSNB M.523, the skull of a young individual,
associated with two teeth, missing mostly elements of
the supracranial basin.
Type locality
Fort n°4, Oude God (Vieux-Dieu), Mortsel, 6 km south-
west to Antwerp city (Fig. 1). Geographic coordinates:
51°10’25’’N-4°27’36’’E.
Type horizon
Berchem Formation, Antwerpen Sands Member. This
member is dated by dinoagellates from Langhian
(possibly latest Burdigalian) to middle Serravalian
(LOUWYE et al., 1999, 2000; LOUWYE, 2005).
Emended diagnosis
Eudelphis mortezelensis is a moderate size physeteroid
somewhat larger than Orycterocetus crocodilinus. It
differs from all the other members of the superfamily
in: the presence of a large left premaxillary foramen,
the vomer widely exposed in ventral view, and the
postglenoid process of the squamosal ventrally
elongated. It differs from all the other physeteroids
except Zygophyseter in the plate-like falciform process
of the squamosal, from all the other physeteroids
except Scaphokogia in the rostrum more progressively
narrowing forwards, from Brygmophyseter and
Zygophyseter in the proportionally smaller teeth,
from Zygophyseter in the shorter zygomatic process
of the squamosal, from Idiorophus and Scaphokogia
in the widely dorsally open mesorostral groove, from
Aulophyseter, Idiophyseter, Kogia, Orycterocetus,
Physeter, and Scaphokogia in the presence of at least
10 deep maxillary alveoli, from Aulophyseter, Kogia,
Orycterocetus, Physeter, and Physeterula in the
enamelled crown of the teeth, and from Aulophyseter,
Orycterocetus, Physeter, and Physeterula in the low
supraoccipital shield.
Comments
The species name was changed in E. mortselensis by
VANDEN BROECK (1874), following the label associated
to the skull. As there is no clear evidence of an
inadvertent mistake as dened by the ICZN (1999), this
change is not supported and the original spelling should
be preserved.
When revising the sperm whales described by DU
BUS (1867, 1872) and VAN BENEDEN (1869, 1877),
ABEL (1905) proposed to refer this species to the
genus Scaldicetus, together with the tooth-based S.
caretti (type species) and S. grandis. BIANUCCI &
LANDINI (2006) correctly restricted the genus and
species names of the two latter to the original material
of du Bus, considering isolated teeth as not diagnostic,
and re-evaluating the original genus name Eudelphis
for the species S. mortezelensis. Two separate works
(BIANUCCI & LANDINI, 2006; KIMURA et al., 2006)
suggested a change for the genus name of the Japanese
species Scaldicetus shigensis. As the revision of Kimura
et al., proposing the new genus name Brygmophyseter
Barnes, was published in the time interval between the
acceptance and the publication of the work of Bianucci
and Landini, it has the priority on the latter for this
taxonomic change. Justifying the change in genus name
by the fact that the type species of Scaldicetus, S. caretti
is based on isolated teeth, KIMURA et al. (2006) propose
that ‘Scaldicetus is probably a grade taxon, not a
natural biological grouping’. Nevertheless, in the same
work they continue to refer fragmentary physeteroid
specimens to the genus Scaldicetus.
ABEL (1905, g. 6) referred to E. mortezelensis an
isolated atlas IRSNB M.524 from the Miocene of the
area of Anwerp (possibly Fort n°1, east of Antwerp). This
complete vertebra, with a maximum width of 226 mm
and a maximum width of the anterior articular facets of
157 mm, approximately matches the size of the occipital
condyles of the holotype of E. mortezelensis, somewhat
larger with a more developed transverse process than
the atlas of the holotype of Placoziphius duboisi, and
distinctly smaller and more slender than the atlas of the
holotype of Physeterula dubusi (see below). As noted
by ABEL (1905) its general morphology is similar to
Physeter, particularly the high transverse process with
a rectilinear dorsomedially directed lateral margin. This
isolated atlas is referred here to Physeteroidea indet.
Description of the holotype of Eudelphis mortezelensis
With most of the bones of the rostrum unfused, this skull
likely belongs to a relatively young individual. Most of
the elements of the rostrum, including two teeth, and
of the basicranium are preserved, whereas the walls of
the supracranial basin and the supraorbital processes are
much more fragmentary. The total length of the skull as
preserved is 852 mm. No more than 100 mm of the apex
Olivier LAMBERT
281
Skull of Eudelphis mortezelensis IRSNB M.523 (holotype), Miocene of Antwerp. A. Dorsal view. B. Schematic
reconstruction of the skull in dorsal view. The shaded areas represent the preserved elements. Scale bar for A-B equals
200 mm.
Fig. 2 –
Miocene sperm whales from the North Sea
282
of the rostrum are probably missing (see reconstruction,
Fig. 2B). As no antorbital notch is preserved the rostrum
length is estimated to ca. two thirds of the condylobasal
length. Bizygomatic width is 592 mm. Taking account
of the distance between the glenoid fossae and the
distance between one fossa and the preserved apex
of the rostrum, the mandibular bluntness index sensu
WERTH (2006: mandibular width/length) was originally
not much larger than 0.5. The dimensions of the skull
are intermediary between Diaphorocetus poucheti /
Orycterocetus crocodilinus (smaller) and Aulophyseter
morricei (larger).
The skull’s long axis, mostly determined by the
direction of the rostrum, is roughly perpendicular to the
surface of the occipital condyles (Fig. 3), whereas in the
Recent Kogia and Physeter this axis is directed ventrally,
indicating a skull ventrally projected compared to the
body axis (CRANFORD, 1999); the primitive condition
of Eudelphis is also observed in Orycterocetus and
Zygophyseter, whereas Praekogia (considering the
orientation of the occipital condyles) and Scaphokogia
already have a rostrum ventrally projected.
Premaxilla. From the proportions of the premaxilla
and maxilla at the preserved anterior end the premaxilla
was likely anteriorly longer (Fig. 2). A thin plate of
the premaxilla nearly completely covers dorsally the
anterior part of the mesorostral groove; the mesorostral
opening widens until the level of the left premaxillary
foramen. The mesorostral groove is dorsally closed by
the joined premaxillae in Idiorophus, Scaphokogia, and
a part of the specimens of Aulophyseter morricei. On
the dorsal surface of the right premaxilla, a foramen
pierces the bone along the suture with the maxilla at
230 mm anterior to the right premaxillary foramen; on
the left premaxilla, two similar foramina are slightly
anterior. From a median slope on the anterior two thirds,
the dorsal surface of the premaxilla changes to a lateral
slope 65 mm anterior to the right premaxillary foramen;
this level might approximately indicate the anterior
limit of the supracranial basin. A premaxillary foramen
is present on both right and left premaxillae; the left
foramen is 60 mm more anterior than the larger right
foramen (width: 14 mm versus 11 mm for the left). The
right premaxillary foramen is followed posterolaterally
by a sulcus (= postero-lateral sulcus). The sulcus
laterally margins a triangular depressed and convex
surface of the premaxilla. Interestingly the anterior part
of this depression is marked by a striated surface that
might correspond to the area of origin for the nasal plug
muscle observed anterior to the premaxillary sac fossa
in other odontocetes (e.g. the ziphiids, see HEYNING,
1989). The left premaxillary foramen, not as reduced
as in most physeteroids, opens posteriorly into a wide
and concave area with a smooth surface, just anterior to
the left bony naris. The more conservative aspect of the
premaxillae in this area compared to other physeteroids
(a feature seemingly present in Diaphorocetus, Bianucci,
pers. comm., 2008) might have been accompanied by a
similarly less derived soft tissue anatomy. Therefore it
cannot be excluded that Eudelphis retained well-formed
premaxillary sacs, structures present in all the Recent
odontocetes except Kogia and Physeter, in which only
incipient or vestigial pockets are observed (HEYNING,
1989; CRANFORD et al., 1996).
Both premaxillae are nearly complete postero-
medially but their posterolateral portion is lost. The
posterior widening and elevation of the right premaxilla
is however sketched, from the anterior limit of the
corresponding bony naris. Considering the slope of the
dorsal surface of the bone at that level, the supracranial
basin could not extend much farther backwards.
Maxilla. In dorsal view, the maxilla is more robust
and wider than the premaxilla on most of the rostrum
length. It lacks the proximal abrupt narrowing seen in
Aulophyseter morricei, Diaphorocetus, Idiorophus,
Orycterocetus, Physeterula, Placoziphius, and
Zygophyseter. Several foramina pierce each maxilla
along the suture with the rostrum (two on the right
side, one on the left side) but the open sutures prevent
from providing accurate diameters (Fig. 2). The lateral
margin of the maxilla raises towards the antorbital notch,
giving a nearly vertical orientation to the lateral wall
of the supracranial basin. One large anteroposteriorly
elongated foramen (length 32 mm) pierces the right
maxilla probably anteromedial to the lost antorbital
notch. A second foramen is 85 mm backwards, at the
top of the lateral slope of the supracranial basin. On the
left maxilla at least one small foramen, three moderate
size foramina and one large longitudinally elongated
foramen (length > 57 mm) are located in the area of the
lost left antorbital notch.
Deep maxillary alveoli are only preserved on their
medial side. The rst posterior right alveolus is 150 mm
anterior to the level of the right premaxillary foramen.
However, as the surface of the maxilla is damaged
posteriorly, more posterior alveoli might have been
originally present. 10 alveoli are preserved on a length
of 272 mm for the right maxilla, and 10 on 280 mm
for the left maxilla (Fig. 3D). The total tooth count was
however certainly higher, with several anterior alveoli
missing (at least three), but it did not reach the count
of Idiorophus patagonicus (22 teeth on each side of
Olivier LAMBERT
283
Skull of Eudelphis mortezelensis IRSNB M.523 (holotype), Miocene of Antwerp. A. Right lateral view. B. Posterior
view. C. Left lateral view. D. Ventral view of the palate. Scale bar for A-D equals 200 mm.
Fig. 3 –
Miocene sperm whales from the North Sea
284
the upper jaw; LYDEKKER, 1893). Individualized upper
alveoli are also present in Diaphorocetus (13-14),
Zygophyseter (13) and possibly Brygmophyseter. The
anteroposterior diameter of the alveoli ranges from
20 to 29 mm, with short septa and a mean depth of 30
mm. The ventral surface of the maxilla is nearly at,
subhorizontal, and wide. A palatine foramen is located
80 mm anterior to the tip of the right palatine.
Vomer. The anteriorly complete vomer is shorter than
the premaxilla and maxilla, contrary to the condition in
Kogia and Scaphokogia. Ventrally the vomer widens
and attens forwards. Being no more keeled, it forms a
wide and at surface exposed between the maxillae and
premaxillae on a length of 240 mm, with a maximum
width of 45 mm (Fig. 3D). Such a wide exposure of
the vomer is possibly related to the young age of the
animal (see young Physeter macrocephalus gured in
KELLOGG, 1925, pl. 6).
Mesethmoid. The morphology of the mesethmoid is the
only indication for the asymmetry of the bony nasal
tracts. The right tract has a maximum anteroposterior
diameter of 37 mm; the left tract has a diameter larger
than 56 mm. The keel of the mesethmoid between the
two nares is distinctly tilted on the left side (30-40°
from the vertical).
Palatine. The two palatines are partly preserved as
unfused simple plates, somewhat thickened posteriorly.
Left pterygoid of Eudelphis mortezelensis IRSNB
M.523 (holotype), Miocene of Antwerp, in ventral
view. Scale bar equals 30 mm.
Fig. 4 –
Pterygoid. Both pterygoids are also detached from the
skull. The general shape of the bone is similar to Kogia
and Physeter, except for a strong reduction of the dorsal
plate usually covering the ventrolateral surface of the
palate (Fig. 4). This reduction is much likely related
to the development of the pterygoid sinus in the palate
area (see the illustration of Kogia in FRASER & PURVES,
1960, pl. 16), a trend documented, but in a lesser extent,
in other physeteroids (MUIZON, 1984; pers. obs.).
Actually, several small and deep fossae on the median
lamina of the pterygoid just anterior to the eustachian
notch indicate further development of the pterygoid
sinus. The posterior tip of the unexcavated hamular
process is medial, differing from the laterally directed
tip in Aulophyseter morricei, Kogia, Orycterocetus, and
Physeter.
Frontal. The deeply worn portions of frontal preserved
at the posterior wall of the supracranial basin only give
a faint idea of the abrupt elevation of this wall; the
dorsal extent of the basin is unknown.
Supraoccipital. The outer surface of the supraoccipital
shield is only preserved on 55 mm above the foramen
magnum, where it is slightly convex; more dorsally
the bone is abraded. Nevertheless, the slope on this
short portion, at an angle no more than 50° from the
horizontal plane, allows a comparison with other
physeteroids. The supraoccipital shield is more erected
with a concave surface in Aulophyseter morricei (see
comment in KIMURA et al., 2006), Orycterocetus,
Physeter, and Physeterula, whereas the condition in
Diaphorocetus and Brygmophyseter, with a lower slope
and a generally convex or at surface, is more similar
to Eudelphis. A small to moderate size occipital crest
might have limited posteriorly the supracranial basin,
but not on such an extent as in Orycterocetus, Physeter,
or Physeterula.
Squamosal. On both sides, the zygomatic process
of the squamosal is incomplete (Fig. 5). However,
the orientation of the ventral surface of the process,
continuous with the glenoid fossa, indicates a short
triangular apex of the zygomatic process no more
than 140 mm anterior to the anterolateral margin
of the exoccipital, shorter than in Brygmophyseter,
Diaphorocetus, Orycterocetus, Physeterula, and
Zygophyseter. The anteroposteriorly attened post-
glenoid process forms a wide plate anteroventrally
curved. The ventral tip, triangular in anterior view, is
considerably longer ventrally than the post-tympanic
process, separated from the latter by a deep and wide
Olivier LAMBERT
285
Fig. 5 – Right squamosal, exoccipital, and basioccipital of Eudelphis mortezelensis IRSNB M.523 (holotype), Miocene of
Antwerp. A. Lateral view. B. Anteroventral view. Scale bar for A-B equals 50 mm.
external auditory meatus. This condition is similar
to the archaeocete Zygorhiza TRUE, 1908 and basal
odontocetes as Agorophius COPE, 1895 and Squalodon
GRATELOUP, 1840. The postglenoid process is shorter in
all other physeteroids for which this area is preserved.
An oval sternomastoideus fossa excavates the lateral
surface of the post-tympanic process.
The triangular glenoid surface is anteromedially
and posteromedially limited by a depression (=
tympanosquamosal recess), especially deep towards
the postglenoid process. The falciform process is
a thin and moderately large plate nearly ventrally
facing, separated from the glenoid surface by a
series of fossae corresponding to the medial part
of the tympanosquamosal recess. The condition in
Zygophyseter seems similar. In Orycterocetus, Physeter,
Physeterula, Placoziphius (see below), and the kogiids
Kogia and Praekogia, the falciform process is either
reduced to a slender projection, or to a short bud, or is
even absent. In Recent physeteroids as well as in Recent
ziphiids the pterygoid sinuses are proportionally much
larger than in other odontocetes (FRASER & PURVES,
1960); more voluminous air spaces surrounding the ear
bones would be necessary for keeping an acoustically
isolating thin air layer around the ear bones during
deep diving (for binaural hearing or to allow efcient
operation of the ossicular chain), when increasing
pressure strongly squeezes the air sacs (FRASER &
PURVES, 1960; CRANFORD et al., 2008). It is proposed
here that the reduction of the falciform process of the
squamosal in the physeteroids from the condition in
Eudelphis is directly related to the enlargement of
the pterygoid sinus, this enlargement being possibly
associated to the shift to deeper foraging areas in more
crownward taxa.
The ear bones of the holotype are lost, suggesting that
these were better isolated from the rest of the basicranium
than in e.g. the archaic odontocete Simocetus FORDYCE,
2002. However, a roughly cylindrical bony element
striated with longitudinal grooves, is transversely
directed along the posterior margin of the left external
auditory meatus (not preserved on the right side), from
the spiny process of the squamosal to the lateral margin
of the post-tympanic process; it might correspond to a
part of the enlarged posterior process of the tympanic
bulla, partly fused with elements of the squamosal in
physeteroids and ziphiids (KASUYA, 1973; MUIZON,
1984). Between the posterior meatal crest and this
hypothetical element of the tympanic bulla is an
elongated fossa likely corresponding to a diverticle of
the pterygoid sinus fossa, extending posterodorsally to
the external auditory meatus.
The oor of the temporal fossa is wide (180
mm from the lateral margin of the cerebral cavity to
Miocene sperm whales from the North Sea
286
the lateral margin of the squamosal). The supraorbital
and postorbital processes of the frontal being lost, the
dimensions and shape of the temporal fossa cannot be
assessed; the fossa is posteriorly longer than in Physeter
and likely anteriorly shorter than in Brygmophyseter
and Zygophyseter.
Alisphenoid. A large foramen ovale opens laterally on
the alisphenoid, followed on a short length by a wide
and shallow sulcus (path for mandibular nerve V3).
Exoccipital. The laterally extended exoccipital is
dorsally overhung by the oor of the temporal fossa
(Fig. 3C). In ventral view, the anterior surface of
the exoccipital is excavated by a wide depression,
posterolaterally elongated along the surface of contact
between the posterior process of the tympanic bulla and
the squamosal (Fig. 5B). In a similar position FORDYCE
(1994, 2002) describes in Simocetus and Waipatia
FORDYCE, 1994 a narrow cleft that might correspond
to a ventral exit for the facial nerve. The outline of the
depression in IRSNB M.523 seems however to t better
with a posterior extension of the pterygoid sinus. In
this case, the posterior process of the tympanic bulla
would have been margined both anterolaterally (fossa
described above on the squamosal) and posteromedially
by extensions of the peribullary sinus. No clue for a
posterior sinus, identied along the anterior surface of
the paroccipital process in most Recent odontocetes but
not in physeteroids (FRASER & PURVES, 1960), could
be found.
Basioccipital. The diameter of the foramen magnum
is 67 mm and the width across the robust occipital
Maxillary tooth of Eudelphis mortezelensis
IRSNB M.523 (holotype), Miocene of Antwerp.
A. Posterolateral view. B. Anterolateral view.
Scale bar for A-B equals 30 mm.
Fig. 6 –
condyles is estimated to 164 mm. The basioccipital
basin is widely open posteroventrally. The jugular notch
is wide. The posterior lacerate foramen has a reduced
size (greatest diameter 15 mm).
Teeth. Two teeth were found associated to the skull.
However, their exact position along the alveolar groove
cannot be guaranteed. The tooth in the posterior fourth
left alveolus is the best preserved (Fig. 6). This 88 mm
high tooth is generally cylindrical, moderately curved,
with a low crown (11 mm long). The section is circular
at the base of the crown (diameter 10 mm). Apically the
root is anteroposteriorly attened with a longitudinal
groove on the anterior surface; proximally the attening
twists to a more lateromedial direction. The proximal
narrowing of the root is weak; maximum and minimum
diameters at mid-height section and at proximal section
of the root are respectively 24/17 mm and 20/14 mm.
The crown is separated from the root by a marked step
due to the thickened cement on the root, and is covered
with a smooth cap of enamel, as mentioned by ABEL
(1905). The diameter of the crown smaller than the root
diameter at their boundary, also seen on the second tooth
preserved, is rather unusual in enamel-bearing fossil
physeteroids; on most of the isolated teeth observed, the
enamel layer is either thicker than, or form a continuous
surface with, the adjoining cement of the crown. A
constriction, much shallower than the constriction at
the cement-enamel boundary, marks the apical two-
fths of the crown. The second tooth, located in the
posterior sixth right alveolus, is poorly preserved with
a crown height of 9 mm for a total height of 93 mm.
In Idiorophus the enamelled crown is proportionally
higher, approximately one third of the total height
of the tooth (LYDEKKER, 1893), whereas the teeth of
Brygmophyseter and Zygophyseter are proportionally
larger and more swollen on the distal portion of the root.
The physeteroid species from the Neogene of the North
Sea dened on the basis of enamelled teeth, Hoplocetus
ritzi HAMPE, 2006, Scaldicetus caretti, and S. grandis
(see material and methods for the status of the two latter
species), differ signicantly in the more robust, swollen
root, with a stronger proximal reduction of the section.
The pulp cavity of these two teeth is hollow,
conrming the young age of the animal. However,
substantial apical wear is already present and a groove
along the posterior surface of the left tooth corresponds
to the contact with the opposite lower tooth. Besides
the indication of a weaned animal, the wear proves
that the teeth were functional and associated to similar
mandibular teeth.
Olivier LAMBERT
287
Discussion
Besides the presence of a supracranial basin,
asymmetric bony nares, and a widened and posteriorly
longer right premaxilla, i.e. the main characters
dening the superfamily Physeteroidea, Eudelphis
mortezelensis is characterised by a set of features that
could be interpreted as primitive for a physeteroid. The
left premaxillary foramen is not as reduced as in other
physeteroids; the falciform process of the squamosal
remains well developed, a feature likely related to the
lesser development of the pterygoid sinus around the
ear bones; and the postglenoid process is ventrally
long. Furthermore, the teeth retain a cap of enamel
on the crown, as in Brygmophyseter, Idiorophus,
and Zygophyseter; more than 10 maxillary teeth are
lodged in deep alveoli, a feature shared at least with
Brygmophyseter, Diaphorocetus, Idiorophus, and
Zygophyseter; and the slope of the supraoccipital shield
is low.
In addition to these plesiomorphic characters, the
rostrum of E. mortezelensis has maxillae wider in dorsal
view than in most other physeteroids, lacking an abrupt
narrowing at mid-length. The exposed ventral surface
of the vomer is wide and not keeled, a feature possibly
age-related.
Genus Placoziphius VAN BENEDEN, 1869
Type species: P. duboisi VAN BENEDEN, 1869
Diagnosis. The diagnosis is the same as for the only in-
cluded species.
Placoziphius duboisi VAN BENEDEN, 1869
Figs 7-8
Placoziphius duboisii VAN BENEDEN, p. 11; pl. 1-2.
Placoziphius duboisii VAN BENEDEN & GERVAIS, pl.
27, gs. 11-12.
Placoziphius duboisii ABEL, p. 87.
Placoziphius duboisii KAZÁR, pl. 1, gs. 1-2.
1869 —
1880 —
1905 —
2002 —
Holotype
IRSNB M.530 (old number 1718), a fragmentary
skull including most of the rostrum, fragments of the
supraorbital processes and of the supracranial basin,
the two squamosal/exoccipital sets, and an associated
fragment of atlas.
Type locality
Edegem, briqueterie de la société Pauwels (VAN
BENEDEN, 1869), 400 m south-east to Fort n°6, 6 km
south to Antwerp (Fig. 1). Geographic coordinates:
51°09’38’’N-4°24’50’’E.
Type horizon
Berchem Formation, Edegem Sands Member. This
member is dated by means of dinoagellates to early
Miocene, late or latest Aquitanian to early Burdigalian
(LOUWYE et al., 1999, 2000; LOUWYE, 2005). Two skull
fragments tentatively referred to P. duboisi, namely
a left exoccipital and the left side of a rostrum base,
were recently found in the Antwerpen Sands Member,
in the foundations of new buildings at the locality of
Veldekens, Berchem, 3.5 km south-east to the centre
of Antwerp (geographic coordinates: 51°20’24’’N-
4°26’17’’E), possibly extending the range of the species
in the middle Miocene.
Emended diagnosis
Placoziphius duboisi has a size close to Diaphorocetus
poucheti and the smallest specimens of Orycterocetus
crocodilinus, larger than all the kogiid species. It differs
from all the other physeteroids for which the rostrum is
known, except Kogia and possibly Diaphorocetus and
‘Aulophyseter’ rionegrensis, in the premaxilla narrower
than the maxilla on the whole length of the rostrum
in dorsal view. It differs from Eudelphis in the abrupt
narrowing of the rostrum on its proximal half, the
reduced left premaxillary foramen and falciform process
of the squamosal, and the narrower and keeled vomer
in ventral view, from ‘Aulophyseter’ rionegrensis,
Diaphorocetus, Eudelphis, Idiorophus, Orycterocetus,
and Zygophyseter in the loss of the posterior maxillary
alveoli, from Aulophyseter, Orycterocetus, and
Physeter in the lower slope of the maxilla-frontal suture
on the supraorbital process, and from Aulophyseter and
Physeter in the higher preorbital process compared to
the lateral margin of the rostrum base. It differs from
the kogiids in the widened posterior portion of the right
premaxilla. It further differs from Diaphorocetus in
the more laterally located right premaxillary foramen,
from Orycterocetus in the wider dorsal surface of the
supraorbital process lateral to the supracranial basin,
and from Thalassocetus in the wider glenoid surface
and the longer plate-like postglenoid process of the
squamosal.
Remark
The species ‘Aulophyseter’ rionegrensis is based on
two skulls from the middle Miocene of Argentina, one
of them rather complete and associated to the mandible
and teeth, but poorly illustrated in the original work
(GONDAR, 1975). COZZUOL (1996) and KAZÁR (2002)
Miocene sperm whales from the North Sea
288
estimated that the species differs signicantly from
members of the genus Aulophyseter, the second author
nding more similarities for ‘A.’ rionegrensis with
Diaphorocetus poucheti. In addition the outline of the
rostrum in dorsal view and the narrow premaxilla on
the rostrum are also similar to the somewhat smaller
Placoziphius duboisi. A detailed comparison of the
three species and a re-illustration of the two Argentine
species would allow a better understanding of their
relationships.
Comments on the attribution of other specimens to
Placoziphius duboisi
ABEL (1905) referred to the species another specimen
from the area of Antwerp, still in a matrix of hardened
sand. This specimen could not be further prepared and
it will not be taken into account here.
More recently, KAZÁR (2002) referred two additional
specimens to the species: a partial skeleton from the
Miocene of Stotzing (Austria) and the fragmentary
skull IRSNB M.525 from the area of Antwerp part
of the syntype of Thalassocetus antwerpiensis sensu
ABEL, 1905.
Besides similarities at the level of the rostrum, a
few differences between the holotype of Placoziphius
duboisi and the Stotzing specimen are noted. If the
exoccipital is correctly identied on the lateral view
of the Stotzing specimen (KAZÁR, 2002, pl. 4, g.
1), either the postglenoid process of the squamosal is
missing (not mentioned by the author), or it is much
shorter ventrally than on the holotype of P. duboisi. The
squamosal of the holotype of P. duboisi is not hidden by
the exoccipital in posterior view, contrary to the Stotzing
specimen. Among the characters listed in the diagnosis
of P. duboisi by KAZÁR (2002), only two are observed
in the holotype: (1) premaxillae not exceeding maxillae
in width at any given point and (2) distal portion of the
rostrum narrow. Both characters might be present in
Diaphorocetus, the rst character is seen in Kogia, and
a narrow distal portion of the rostrum is rather common
among physeteroids [e.g., Orycterocetus and part of the
specimens of Aulophyseter morricei (see KIMURA et al.,
2006)]. The difculty to nd other comparison points
between these two fragmentary skulls, with different
preservation states, leaves open the question of the
attribution of the Stotzing specimen to P. duboisi. In this
work I prefer not to consider it in the species, pending
new data from more complete specimens.
More conspicuous differences between the holotype
of P. duboisi and IRSNB M. 525, part of the syntype
of T. antwerpiensis sensu ABEL, 1905, are noted.
As mentioned in KAZÁR (2002) IRSNB M.525 is
signicantly smaller and its supraorbital region is not
extremely thickened in lateral view. Furthermore, the
supraorbital process of IRSNB M.525 is narrower
in dorsal view; a sagittal crest covered by the right
premaxilla is identied in its supracranial basin, not
homologous to the crest covered by the left nasal in
P. duboisi; and the squamosal of IRSNB M.525 differs
considerably in the short, more nodular than plate-like
postglenoid process, higher than the ventral margin of
the exoccipital, and in the narrower glenoid surface.
IRSNB M.525 is therefore excluded from the hypodigm
of P. duboisi (see below).
Description of the holotype of Placoziphius duboisi
In addition to previous contributions (VAN BENEDEN,
1869; ABEL, 1905; KAZÁR, 2002), some elements
of description are given here, based on the holotype
specimen IRSNB M.530. The apex of the rostrum
is missing (Fig. 7B); considering the thickness and
orientation of the different elements at the preserved
extremity, the rostrum was likely somewhat shorter than
in Orycterocetus, but possibly longer than estimated
in KAZÁR (2002). The distance between the right
premaxillary foramen and the preserved apex of the
rostrum is 280 mm. At the level of the right premaxillary
foramen, the distance between the lateral margins of the
premaxillae is 129 mm. As in other physeteroids the
rostrum is wide at its base. The proximal narrowing
is abrupt. More anteriorly the concave lateral margins
indicate a more progressive narrowing.
Premaxilla. Even if the dorsal surface of both
premaxillae is missing distally, these bones did
not exceed the maxillae in width in dorsal view,
as suggested by KAZÁR (2002), contrasting with
Orycterocetus. A similar condition is present in Kogia
and suspected in ‘Aulophyseter’ rionegrensis and
Diaphorocetus. The mesorostral groove is widely open
on its whole length, with a maximum opening 37 mm
wide. Contrary to the assumption of KAZÁR (2002)
and as noted by Van BENEDEN (1869) and KELLOGG
Skull of Placoziphius duboisi IRSNB M.530
(holotype), Miocene of Antwerp. A. Anterodorsal
view. B. Dorsal view with the reconstructed outline
of the rostrum. C. Ventral view of the rostrum. D.
Right lateral view. The relative position of the
different elements in A-D is approximate. E. Left
squamosal in anteroventral view. Scale bar for A-E
equals 100 mm.
Fig. 7 –
Olivier LAMBERT
289
Miocene sperm whales from the North Sea
290
(1925), a small left premaxillary foramen (diameter 5
mm) is present in the holotype at a level slightly more
anterior than the right premaxillary foramen. The left
foramen is followed posteriorly by a narrow and short
sulcus likely corresponding to the postero-lateral sulcus
observed in most odontocetes. The right premaxillary
foramen is much larger (diameter 14 mm) and followed
by a shallow and wide postero-lateral sulcus. On both
sides, the surface of the premaxilla is regularly concave
posterior to the premaxillary foramen (Fig. 7A), as in
Orycterocetus. From the level of the right bony naris the
right premaxilla widens and raises to form the posterior
wall of the supracranial basin, crossing the sagittal plane
of the skull; the maximum width of the bone is 143 mm,
at mid-height. Posterodorsally the right premaxilla is
not vertical, more similar to O. crocodilinus USNM
14730 than to O. crocodilinus USNM 22926. Along the
median margin of the right premaxilla and posterior to
the bony nares, a plate-like fragment of the mesethmoid
is preserved.
Maxilla. No alveoli are detected along the poorly
preserved alveolar groove (Fig. 7C). However, taking
account of the missing anterior elements, the shallow
posterior alveoli in ‘Aulophyseter’ rionegrensis,
Diaphorocetus, and Orycterocetus crocodilinus USNM
22926, and the preservation similarities with O.
crocodilinus USNM 14730, the possibility that anterior
alveoli were present cannot be rejected, although this
hypothesis is not better supported than a more derived
alveolar groove as seen in Aulophyseter morricei.
60 mm anterior to the right premaxillary foramen a
partly preserved dorsal infraorbital foramen pierces the
right maxilla. Medial to the partly preserved U-shaped
right antorbital notch is a large dorsal infraorbital
foramen, partly subdivided by a lateral septum, and in
line with a smaller posterior foramen. The two foramina
are located along the lateral wall of the supracranial
basin. On the left maxilla, the condition is similar,
possibly with an additional foramen. Lateral to the
foramina, the dorsal surface of the maxilla is at and
somewhat wider than in Orycterocetus.
In lateral view, the angle of the slope of the maxilla-
frontal suture above the orbit is roughly 25-30° (Fig.
7D), probably close to ‘Aulophyseter’ rionegrensis
and Diaphorocetus, and lower than in Aulophyseter
and Orycterocetus. Because the posterior portion of
the maxilla is missing, the shape of the temporal fossa
cannot be reconstructed.
Similar to Orycterocetus crocodilinus USNM
22926, an oblique groove marks the left maxilla on
the posterolateral wall of the supracranial basin; this
groove isolates anteromedially a shelf of the maxilla
partly covered by a thin sheet of at bone, interpreted
by comparison with Orycterocetus as the left nasal.
Suture marks on the right side of this thin bone, likely
indicating the contact with the right premaxilla, further
support this identication. This shelf is supercially
similar to the sagittal crest in Kogia, Praekogia, and
possibly Thalassocetus (see below), but in the three
latter taxa the surface is concave and covered by the
right premaxilla.
Vomer. The nearly complete vomer is shorter anteriorly
than the premaxilla and maxilla, contrasting with the
kogiids. Ventrally the keeled vomer appears on a length
of 190 mm.
Squamosal-Exoccipital. The zygomatic process of
the squamosal is moderately elongated. The glenoid
fossa is wide, approximately triangular, and nearly
continuous with the shallow tympanosquamosal recess.
The latter is only somewhat deeper along the external
auditory meatus. Considering the preserved base of the
falciform process, this element consisted in a thin and
narrow projection, more reduced than in Eudelphis.
The postglenoid process is a vertical anteroposteriorly
compressed plate slightly longer ventrally than the post-
tympanic process but proportionally shorter and less
curved than in Eudelphis. As noted by KAZÁR (2002),
elements of the posterior process of the tympanic bulla
are still attached to the grooved ventral surface of
the post-tympanic process, reaching the ventrolateral
margin of the exoccipital. Differing from Eudelphis,
no fossa separates this element from the external
auditory meatus. Ventrolateral to the paroccipital
process a narrow sulcus marks the exoccipital in a
position similar to the depression mentioned above in
Fig. 8 – Atlas of Placoziphius duboisi IRSNB M.530
(holotype), Miocene of Antwerp, in anterior view.
Scale bar equals 50 mm.
Olivier LAMBERT
291
Eudelphis. The proportions of this sulcus, narrower than
in Orycterocetus, better t the description by FORDYCE
(1994) of a narrow cleft possibly corresponding to the
exit of the facial nerve in Waipatia.
Atlas. The fragmentary free atlas associated to the
holotype skull (Fig. 8) has a maximum ventral length
of 32 mm, proportionally shorter than in Zygophyseter.
The maximum width is estimated to 186 mm. The long
axis of the moderately concave anterior articular facet
is, as noted by VAN BENEDEN (1869), more oblique
from the vertical than in Physeter and Zygophyseter.
The transverse process is shorter and more developed
dorsoventrally than in Zygophyseter.
Discussion
The holotype of Placoziphius duboisi shares more
similarities with Orycterocetus, ‘Aulophyseter’
rionegrensis, and Diaphorocetus than with any other
physeteroid, at the level of the size, the rostrum shape,
the morphology of the premaxillae at the rostrum base
and in the supracranial basin, and the supraorbital area
of the maxilla.
Among the differences with Orycterocetus listed by
KAZÁR (2002), only the narrowness of the premaxilla
compared to the maxilla on the rostrum is seen in the
holotype of P. duboisi. In addition, the rostrum of the
smaller P. duboisi is likely shorter; at least the posterior
alveoli are missing; the maxilla on the supraorbital
process is wider, suggesting a narrower supracranial
basin; and the elevation of the maxilla towards the
occipital crest is less abrupt.
Considering the illustrations of Diaphorocetus in
MORENO (1892, pl. 10), I disagree with the comment
of KAZÁR (2002) about the outline of the rostrum in
this species; the lateral margin is distinctly concave,
as in Orycterocetus, Placoziphius, and several other
physeteroids. Diaphorocetus has a right premaxillary
foramen more medially located; it possibly possessed
deeper maxillary alveoli, but the elongated braincase
of the holotype might be related to the dorsoventral
crushing of the skull. Despite obvious similarities,
better preserved specimens of Diaphorocetus and
Placoziphius would certainly allow a more detailed
investigation of their systematic relationships with
Orycterocetus.
Finally, Placoziphius shares a derived character
with Kogia: the premaxilla narrower than the maxilla
on the whole length of the rostrum. This hypothetical
relationship could be tested with the description of new
rostral material of Aprixokogia or Praekogia.
Genus Orycterocetus LEIDY, 1853
Type species: O. quadratidens (LEIDY, 1853)
Orycterocetus crocodilinus COPE, 1868
Figs 9-10
Referred specimen
IRSNB M.1936, a fragmentary skull found around 1910
during the construction of the fort of Kessel.
Locality
Fort of Kessel, Kessel, 16 km south-east-east to Antwerp
(Fig. 1). Geographic coordinates: 51°09’02’’N-
4°37’44’’E.
Horizon
Berchem Formation, Antwerpen Sands. This specimen
displays the same preservation state as specimens of
the ziphiid Archaeoziphius microglenoideus and the
mysticete Isocetus depauwi from the same locality;
dinoagellate dating of sediment found around the skull
of I. depauwi produced a middle Miocene, tentatively
late Langhian to early Serravalian age (LAMBERT &
LOUWYE, 2006).
Comments
As this work is not a revision of the genus Orycterocetus,
O. crocodilinus being mostly known by more complete
specimens from the east coast of USA (KELLOGG,
1965), I will only provide a brief description of the
specimen from the North Sea and I will not discuss the
denition and content of the genus. Such a work should
indeed be centred on the American specimens (A. C.
Dooley Jr., pers. comm., 2007).
Brief description of the referred specimen IRSNB
M.1936
This skull is made of four fragments corresponding to
both supraorbital areas and both squamosal-exoccipital
sets. The size of the different elements falls within
the size interval of three specimens of Orycterocetus
crocodilinus provided by KELLOGG (1965), somewhat
smaller than USNM 22926. Partly fused sutures,
particularly on the basicranium, suggest a young
individual.
The left supraorbital process of the frontal is roughly
complete, as well as the covering maxilla. The pointed
preorbital process of the frontal is barely thickened.
A reduced space could contain a short extension
(no more than 16 mm long) of the lacrimal between
the maxilla and the frontal (Fig. 9B), more reduced
Miocene sperm whales from the North Sea
292
Skull of Orycterocetus crocodilinus IRSNB M.1936, Miocene of Kessel. A. Left supraorbital area in dorsal view. B.
Left lateral view. The relative position of the different elements is approximate. Scale bar for A-B equals 100 mm.
Fig. 9 –
than in the kogiids (see MUIZON, 1988, g. 35). The
triangular postorbital process is moderately elongated
ventroposteriorly, less than in O. crocodilinus USNM
22926 and USNM 14730; its laterodorsal surface, not
concealed by the maxilla in dorsal view, is concave.
The length of the orbit is 121 mm. The elevation of the
frontal towards the dorsal apex of the temporal fossa
is abrupt, indicating together with the lateral outline of
the exoccipital a high and anteroposteriorly compressed
temporal fossa. On top of the temporal fossa, the frontal-
maxilla suture is nearly vertical for a length of 95 mm.
The slightly medially sloping surface of the maxilla
covering the supraorbital process of the frontal is
narrower and more erected than in Placoziphius. 55
mm posterior to the partly preserved left antorbital
notch a large foramen pierces the maxilla and is
followed posterolaterally by a wide sulcus along the
outer face of the lateral wall of the supracranial basin
(Fig. 9A). On the inner wall of the basin a smooth
concave surface gives the position of the apex of the
left premaxilla. A wide longitudinal groove excavates
the dorsal surface of the maxilla at its posterior end,
less oblique than in O. crocodilinus USNM 22926. The
more fragmentarily preserved dorsal surface of the right
maxilla only displays a large foramen similar to the
foramen described on the left side and a wide concave
surface for the large plate of the right premaxilla, along
the inner wall of the supracranial basin.
Each squamosal lacks the apex of the zygomatic
process; the anterior section is high, transversely
attened (Fig. 9B). More reduced than in Placoziphius
the postglenoid process is anteroposteriorly attened,
thin and short, limiting a narrow external auditory
meatus. The glenoid fossa is barely separated from
the shallow tympanosquamosal recess (Fig. 10A). The
falciform process is reduced to a small peg anterior
to the spiny process, as in O. crocodilinus USNM
22926 and USNM 22931, smaller than in Eudelphis
and Placoziphius. Suture marks on the post-tympanic
process of the squamosal and on the anterior surface
of the exoccipital likely indicate the contact with the
elongated posterior process of the tympanic bulla.
As in Eudelphis, an elongated depression margins
this area posteromedially on the anterior surface of
the exoccipital. Medial to this depression the anterior
surface of the paroccipital process is hollowed by a
small shallow circular fossa, possibly for the posterior
sinus.
Olivier LAMBERT
293
Skull of Orycterocetus crocodilinus IRSNB M.1936, Miocene of Kessel. A. Anteroventral view of the right
squamosal. B. Posterior view of left and right portions of the basicranium. The relative position of the different
elements is approximate. Scale bar for A-B equals 50 mm.
Fig. 10 –
In posterior view the ventral margin of the
exoccipital bears a distinct concavity (Fig. 10B). The
ventral surface of the alisphenoid is concave medially;
the foramen ovale is not preserved and only a faint
groove indicates the path for the mandibular nerve V3.
Discussion
Besides slight differences, mostly at the level of the
postorbital process of the frontal, this specimen perfectly
ts the intraspecic variability within the Maryland and
Virginia species Orycterocetus crocodilinus. From a
stratigraphic point of view all the cranial material of
O. crocodilinus (excluding isolated teeth) from the east
coast of the USA was found in zones 9, 11, 12, and 13 of
the Calvert Formation (KELLOGG, 1965), dated between
latest early and early middle Miocene (VERTEUIL &
NORRIS, 1996). This matches the Antwerpen Sands
origin of IRSNB M.1936. It should be noted here that
isolated teeth and ear bones from the Miocene of the
faluns de Tourraine et Anjou, France, were tentatively
referred to Orycterocetus crocodilinus (GINSBURG &
JANVIER, 1971). The two other species of the genus
Orycterocetus are based on more fragmentary material
that cannot be compared to IRSNB M.1936 (see
comments in KAZÁR, 2002; BIANUCCI et al., 2004).
BIANUCCI et al. (2004) attribute a mandible from the
Miocene of Pietra leccese, southern Italy, to a larger
species of Orycterocetus.
Family Physeteridae
Genus Physeterula VAN BENEDEN, 1877
Type species: P. dubusi VAN BENEDEN, 1877
Diagnosis. The diagnosis is the same as for the only in-
cluded species.
Physeterula dubusi VAN BENEDEN, 1877
Figs 11-14; Table 1
1877 —
1890 —
1905 —
2008 —
Physeterula dubusii (part) VAN BENEDEN, p. 852; pl. 1,
gs. 1-3.
Kogia dubusii COPE, p. 608.
Physeterula dubusii ABEL, p. 79; gs. 11-12.
Physeterula dubusii WHITMORE & KALTENBACH, g. 89a.
Holotype
IRSNB M.527 (old number 3192), a fragmentary
skull including rostral fragments of the premaxillae,
parts of the supracranial basin and of the basicranium,
with the associated mandible and 18 mandibular teeth,
four additional detached teeth, the fragmentary atlas,
four caudal vertebrae, one chevron, six rib fragments,
and one fragment of the sternum. In the preliminary
description of the species, VAN BENEDEN (1877, p.
852) designated this specimen as the ‘type pour cette
description’. The mandible and isolated teeth are
illustrated in VAN BENEDEN (1877) and the skull in
ABEL (1905, p. 75, 77).
Miocene sperm whales from the North Sea
294
Type locality
Boundary between the communes of Deurne and Bor-
gerhout (MOURLON, 1878), eastern suburbs of Antwerp
(Fig. 1).
Type horizon
No precise stratigraphic data are available for the
holotype. The appearance of the surface of the bones,
mostly the premaxillae and the mandible, is similar
to specimens of the ziphiid Ziphirostrum marginatum
(colour, erosion, and possibly aborted invertebrate
borings); this might indicate an origin in the same
levels, possibly the late Miocene Deurne Sands Member
(LAMBERT, 2005). This member is locally present at
the boundary between the communes of Deurne and
Borgerhout (DE MEUTER et al., 1976).
Referred specimen
IRSNB M.528, fragments of the basicranium (see
comment below). The label associated to the specimen
provides the following information: nouvelle enceinte,
3ème section, vers le saillant du fossé du ravelin, face
gauche de l’ancien fortin n°1, sous Deurne (MOURLON,
1878). These indications correspond approximately
to the geographic coordinates 51°13’6’’N-4°27’7’’E.
Among the sections described in the area (DE MEUTER
et al., 1976), one mostly includes Antwerpen Sands
Member (Stenen Brug), whereas the other (Rivierenhof)
also includes the late Miocene Deurne Sands Member
and the early Pliocene Kattendijk Formation.
Emended diagnosis
Physeterula dubusi is a large physeteroid with a skull
size close to Idiorophus patagonicus and Zygophyseter
varolai. It differs from the other physeterids
Aulophyseter and Physeter in the retention of functional
maxillary teeth, the shorter anterior extension of the
supracranial basin, the preorbital process being at the
same vertical level as the dorsolateral margin of the
rostrum base, the frontal-maxilla suture less erected
in lateral view, and more numerous dorsal infraorbital
foramina. It further differs from Physeter in the maxilla
being absent from the dorsal view of the rostrum for
more than half its length, the elevated temporal fossa,
the mandibular symphysis shorter than half the length
of the mandible, and the fact that some of the caudal
vertebrae are longer than wide. Finally it differs from
Aulophyseter in the robust and shortened postorbital
process of the frontal.
Comments
Although VAN BENEDEN (1877), followed by
MOURLON (1878), refers three specimens from the
area of Antwerp to Physeterula dubusi, ABEL (1905)
only cites two specimens, IRSNB M.527 (old number
3192) and IRSNB M.528 (old number 3191), without
argument. Considering the brief description of the
elements preserved for each specimen by Mourlon
and the size difference in IRSNB M.528 between the
dorsal part of the skull and the mandible fragments on
one side and the basicranium elements on the other side
(much larger), an erroneous mix of two specimens by
Abel is proposed. Furthermore, the mandible fragments
have a completely different preservation type (lighter
colour, bone deeply worn) compared to the dorsal part
of the skull. As the basicranium and mandible elements
do not bear any date of discovery, contrary to the
dorsal elements, IRSNB M.528 is split in three. The
basicranium roughly similar to the holotype is referred
to P. dubusi whereas the referral of the mandible and
dorsal elements to the species is rejected. The mandible
is considered as Physeteridae indet. and a new
collection number (IRSNB M.1937) is attributed to the
supracranial elements, which are described farther in
the text.
Description
Skull
Because the second specimen is only fragmentarily
preserved the description is mostly based on the
holotype. All the preserved elements of this large skull
are dissociated, displaying no clear contacts, except
for the vomer and the mesethmoid at the choanae and
the exoccipital and the squamosal on both sides of
the basicranium. Combining the preserved portion of
the mandible and the basicranium, the condylobasal
length of the skull is estimated at more than 1375 mm,
probably close to Brygmophyseter shigensis, Idiorophus
patagonicus, and Zygophyseter varolai.
Premaxilla. Considering the length of the mandible
associated to the holotype, an important anterior
portion of the premaxillae is missing. On the rostrum
the premaxilla is wide (Fig. 11A), as in Aulophyseter
morricei. Even if the degree of dorsal closure of
the mesorostral cannot be estimated the separation
between the two premaxillae is certainly narrower than
illustrated in ABEL (1905, g. 12). Although there is no
rm contact between the right premaxilla and maxilla at
the rostrum base, the correspondence between the slope
of the premaxilla at the rostrum base and the slope of
the maxillary lateral wall of the supracranial basin
implies that the more anterior part of the same piece of
Olivier LAMBERT
295
Fig. 11 – Skull of Physeterula dubusi IRSNB M.527 (holotype), Miocene of Antwerp. A. Fragments of premaxillae and
right maxilla in dorsal view. B. Right maxilla, squamosal, and exoccipital in lateral view. The relative position
of the different elements in A-B is approximate. C. Posterior part of the vomer and mesethmoid in dorsal view.
D. Supraorbital area of the left frontal in lateral view. Scale bar for A-D equals 200 mm.
the premaxilla keeps a moderate median slope. This is a
rst argument supporting the idea that the supracranial
basin was not as developed anteriorly in Physeterula as
in Physeter and possibly Aulophyseter. At least three
small foramina pierce the dorsal surface of the left
premaxilla. Each foramen is followed anteriorly by a
sulcus; none of them corresponds to the more posterior
left premaxillary foramen of most odontocetes. One
somewhat larger foramen is partly preserved at the
posterior end of the fragment of right premaxilla;
it might correspond to the main right premaxillary
foramen.
The detached large posterior plate of the right
premaxilla (239 x 149 mm) displays only a slight
concavity (Fig. 12D); it ts therefore entirely on the
posterior wall of the supracranial basin, agreeing with
the abrupt and important elevation of this posterior
wall. As no rm clue of the right bony naris margin
is seen on this isolated fragment, lacking its median
margin, we cannot assess the degree of overlap of the
right premaxilla on the left side of the basin.
Maxilla. The right maxilla is preserved from 250 mm
anterior to the antorbital notch to the posterior wall of
the supracranial basin. The lateral margin of the rostrum
base diverges anteriorly for 75 mm, before a pronounced
Miocene sperm whales from the North Sea
296
narrowing (Fig. 11A) as seen in Aulophyseter. The
maxilla was probably absent from the dorsal surface
of the rostrum for more than half its length. This faster
narrowing of the rostrum in Physeterula compared to
Physeter is a second element indicating a more limited
anterior extension of the supracranial basin in the former.
Anterior to the antorbital notch, two foramina (32 and
20 mm long respectively) are partly preserved along
the suture with the premaxilla. The U-shaped antorbital
notch is deep and wide with an anteriorly pointed long
preorbital process of the maxilla. As in Orycterocetus,
the antorbital notch is not lower than the dorsolateral
margin of the rostrum, differing on this point from
Aulophyseter and Physeter. Between the antorbital
notch and the maxillary crest laterally limiting the
supracranial basin is a large dorsal infraorbital foramen
(106 mm long). A second oval foramen (maximum
diameter 49 mm) is located 68 mm posterior to the large
foramen, also lateral to the maxillary crest. In the same
area, there are three foramina in Orycterocetus, only
one major foramen (maxillary incisure) in Physeter,
located at the top of the elevated maxillary crest, and
two foramina both median to the maxillary crest in the
Fig. 12 – Skull of Physeterula dubusi IRSNB M.527 (holotype), Miocene of Antwerp. A. Frontals, exoccipitals, basioccipital,
and squamosals in posterior view. The relative position of the different elements is approximate. B. Right squamosal
and exoccipital in anteroventral view. C. Right pterygoid in ventral view. D. Posterior portion of the right premaxilla,
from the posterior wall of the supracranial basin, in anterior view. Scale bar for A-D equals 200 mm.
holotype of Aulophyseter morricei. In Physeterula the
maximum lateral extension of the supracranial basin
is behind the posterior dorsal infraorbital foramen, as
in O. crocodilinus USNM 22926 (some intraspecic
variation is observed at this level in the latter). Differing
from Orycterocetus and more similar to Aulophyseter
and Physeter, the posteromedian extremity of the right
maxilla does not reach the sagittal plane of the skull.
In lateral view of the supraorbital process, the
posterodorsal elevation of the maxilla-frontal suture
is less abrupt than in Aulophyseter and Physeter for
most of its length (around 30° from the horizontal
plane, close to Placoziphius), only curving towards the
supraoccipital crest on the posterior third (Fig. 11B).
Frontal. The supraorbital process of the left frontal is
nearly complete (Fig. 11D), only lacking a fragment
of the posteroventral surface of the postorbital
process. The preorbital process is thick and short. The
postorbital process is shorter than in Aulophyseter and
Orycterocetus, more similar to Physeter.
The frontal part of the posterior wall of the
supracranial basin is preserved, whereas the maxillae
Olivier LAMBERT
297
and the supraoccipital are nearly completely lost in this
area. The two frontals make a high and wide surface
slightly anteriorly concave. The maximum vertical
distance between the roof of the cerebral cavity and the
top of the frontal is 225 mm. The posterodorsal outline
of the left temporal fossa indicates a high and pointed
roof, as in Aulophyseter and Orycterocetus. In posterior
view, the temporal fossa deeply cuts the lateral margin
of the occipital shield (Fig. 12A); the maximum width
of the occipital shield is estimated to 490 mm and the
minimum width at the level of the temporal fossae is
310 mm.
Vomer-mesethmoid. The anterior portion of the vomer
is missing. Anterior to the choanae the vomer is robust,
laterally thickened (Fig. 11C). The ossied mesethmoid
lls the mesorostral groove on a short distance. As in
other physeteroids, the bony nasal tracts are considerably
asymmetric: the keel of the mesethmoid separating the
two tracts is tilted to the left side (approximately 30°
to the vertical) and the left tract is much larger than the
left. On the ventral surface of the rostrum the vomer is
either not or weakly keeled.
Pterygoid. The pterygoid is robust, particularly at the
base of the short posterolaterally pointed hamular
process (Fig. 12C). The dorsal plate is not as reduced
as in Eudelphis.
Squamosal. IRSNB M.528 is signicantly larger and
more robust than the holotype: the distance between
the oor of the temporal fossa and the ventral apex of
the paroccipital process of the exoccipital is 205 mm
in IRSNB M.528 versus 184 mm in the holotype. In
lateral view the zygomatic process of the squamosal
is triangular with an elongated, although incomplete,
anterior apex (Fig. 11B). The postglenoid process is a
thin vertical plate, shorter than in Eudelphis, posteriorly
limited by a well-dened external auditory meatus. The
sternomastoideus fossae are shorter than in Physeter,
restricted to the proximity of the exoccipital. In ventral
view (Fig. 12B), considering the shape of the squamosal
the lost alisphenoid was considerably less developed
outward and backward than in Aulophyseter (see
KELLOGG, 1927, pl. 2), more similar to Orycterocetus
and Physeter. The glenoid fossa is a wide and weakly
concave surface continuous with the surface of the
tympanosquamosal recess. The falciform process of the
squamosal is reduced to a tiny bud.
Exoccipital. The exoccipital is thick along its lateral
contact with the squamosal. The dorsolateral margin of
the exoccipital posteriorly limiting the temporal fossa
overhangs the ventral part of the bone. The jugular
notch is wide (Fig. 12A).
Mandible. The more complete right dentary is
preserved on a length of 1170 mm (Fig. 13A), but both
extremities are missing (condyle and anterior apex).
From the curve of the dentary in dorsal view, the length
of the unfused narrow symphysis is estimated to 400-
500 mm, less than one half of the total length of the
mandible, proportionally shorter than in Zygophyseter
and adult Physeter (a very young individual presented
by FLOWER, 1869, pl. 56, displays a symphysis less than
half the total length of the mandible). Considering the
estimated width of the basicranium of this specimen,
the mandibular bluntness index (sensu WERTH, 2006)
is no more than 0.5, indicating a relatively elongated
mandible. The maximum height of the incomplete
ramus is 220 mm; the median surface is excavated by
a large mandibular foramen, more than 390 mm long.
The minimum height of the ramus, located before the
symphysis, is 95 mm. At least six mental foramina
pierce the lateral surface of the bone. Around 19 deep
alveoli are counted on a length of 735 mm, with at
least six alveoli more posterior than the symphysis.
This count approximately matches the count for
the maxillary-premaxillary teeth of Orycterocetus
crocodilinus (20; KELLOGG, 1965), higher than in
Brygmophyseter and Zygophyseter. The diameter of
the alveoli ranges from 27 to 32 mm, with thin or only
partial septa. The anterior alveoli are strongly inclined,
whereas the posterior alveoli are roughly vertical.
On the right lateral surface of the symphysis,
along the alveolus of the fth anterior dentary tooth
is an extra bony growth (Fig. 13E). This thicker area
likely corresponds to periosteal new bone formation.
As no disruption of the outline of the mandible could
be detected ventrally and medially, this 30 mm width
oblique protuberance is not the result of a healed
fracture. Considering that it follows laterally the
curve of the fth alveolus, the most probable cause
is osteomyelitis due to a spread of bacterial infection
from the pulp of the adjacent tooth or from periodontal
tissue. The adjacent tooth is unfortunately not totally
preserved and might even have been relocated in the
wrong alveolus. Mandibular osteomyelitis has been
described in the Miocene kentriodontid Hadrodelphis
calvertense (DAWSON & GOTTFRIED, 2002), but in that
case the bony growth is more distant from the alveolar
row. Contrary to the above case, no drainage stula is
clearly detected; a mental foramen might be slightly
enlarged but its outline is not complete. A dental abscess
Miocene sperm whales from the North Sea
298
Mandible and teeth of Physeterula dubusi IRSNB M.527 (holotype), Miocene of Antwerp. A. Right dentary in lateral
(A1) and dorsal view (A2). B1-2 and C1-2. Detached upper or lower teeth in anterior/posterior views. D. Detail of the
right dentary with three teeth from the approximate level of the end of the symphysis, in lateral (D1) and medial (D2)
view. E. Extra bony growth probably due to osteomyelitis on the lateral wall of the right dentary, in dorsal (E1) and
lateral view (E2). Scale bar for A equals 200 mm and for B-E equals 50 mm.
Fig. 13 –
is similarly mentioned in the maxillary alveolar groove
of the holotype of Aulophyseter morricei (KELLOGG,
1927).
Teeth. The mandibular teeth of the holotype have a
height ranging from 119 to 135 mm; the lowest teeth
are located anteriorly. The section of these regularly
curved teeth is roughly circular, with a maximum
diameter at mid-height ranging from 24 to 27 mm (Fig.
13D). The pointed crown lacks enamel, similarly to the
teeth of Kogia, Orycterocetus, Physeter, and possibly
Aulophyseter morricei (see comment in KIMURA et al.,
2006, contradicting KELLOGG, 1927). The root-crown
separation is not clear. The surface of the root is uted.
The pulp cavity is nearly completely lled, suggesting
a relatively old specimen. A part of the teeth referred to
Orycterocetus crocodilinus and illustrated by KELLOGG
(1965, pl. 30) were found isolated; for the comparison,
only the teeth associated with skull elements are used.
These teeth are lower, more slender, and more strongly
curved than most of the teeth of Physeterula, except
a detached tooth displaying a stronger curvature (Fig.
13C). Note that occasional smaller maxillary teeth of
Physeter are sometimes more strongly curved than
dentary teeth (pers. obs.). The longitudinal grooves on
the sides of the teeth of Orycterocetus differ from the
more numerous grooves giving a uted aspect to the
teeth of Physeterula. Actually the teeth of Orycterocetus
Olivier LAMBERT
299
are more similar to two isolated physeteroid teeth from
the Neogene of Antwerp gured by ABEL (1905, gs. 9-
10), themselves close to teeth of Kogia. The description
and illustration of the teeth of Aulophyseter morricei in
KIMURA et al. (2006) are not sufciently detailed for
a comparison. The dentary teeth of Physeter are more
robust and much larger.
Most of the teeth display an elongated elliptical
anterior and/or posterior wear facet (Fig. 13D),
supporting the hypothesis of opposite functional
maxillary teeth. Apart from one tooth of the left dentary
displaying an important apical wear, the apex of the
crown of the teeth is barely worn. These elements,
combined with the elongated shape of the mandible,
suggest that Physeterula probably fed by rst grasping
the moderate size prey between its toothed jaws and
then swallowing it by suction.
Post-cranial elements
Atlas. As in Physeter, Placoziphius, and Zygophyseter
the atlas is free (Fig. 14F). Only the ventral part is
preserved; its maximum anteroposterior length is 61
mm; the distance between the lateral margins of the
anterior articular surfaces is estimated to 215 mm. The
ventral portion of the neural canal is proportionally
wider than in Physeter.
Caudal vertebrae. Four caudal vertebrae are preserved
with the holotype, displaying the typical articulation
facets for the corresponding chevron (one chevron is
associated). Considering the development of the partly
worn transverse process, the presence of a vertebrarterial
canal, and the general size of each vertebra (Table
1), these vertebrae are anteroposteriorly sequenced
as shown in Figure 14A-D. By comparison with the
Post-cranial elements of Physeterula dubusi IRSNB M.527 (holotype), Miocene of Antwerp. A, B, C, D. Four
caudal vertebrae in left lateral (1) and dorsal (2) view. E. Chevron in anterior/posterior (E1) and lateral (E2) view.
F. Atlas fragment in anterior view. G-K. Rib fragments in anterior view. L. Right part of the presternum and
ankylosed second segment. All the scale bars equal 50 mm.
Fig. 14 –
Miocene sperm whales from the North Sea
300
vertebral column of Physeter (e.g., FLOWER, 1869)
and taking account of the slight transverse compression
of the centrum of vertebra D, these vertebrae can be
positioned at the central torso-caudal tail stock transition
sensu BUCHHOLTZ (2001). In Physeter all the vertebrae
of the torso have a relative length of the centrum < 1
(BUCHHOLTZ, 2001), whereas in this specimen at least
the centra of three caudals are longer than high or wide,
possibly a primitive condition. This difference suggests
that, given the relationship between relative length of
the vertebra and axial exibility (BUCHHOLTZ, 2001),
this peculiar area of the column was more exible
in Physeterula than in Physeter, allowing a wider
amplitude of movement.
Interestingly, one vertebra (Fig. 14D) bears on
at least one of the lateral surfaces of the centrum a
supplementary thick sheet of bone (maximum width
32 mm, height 94 mm) projecting posteriorly behind
the epiphysis. The appearance of this bony outgrowth
ts characteristics of the diffuse idiopathic skeletal
hyperostosis (DISH) listed in KOMPANJE (1999):
‘dripping candle wax’ corresponding to ligamentous
calcications, preservation of the intervertebral disc
height, and vertebral foramina not affected. Similar bony
outgrowths observed in lumbar vertebrae of mosasaurs
and a fossil mysticete lead MULDER (2001) to relate
this phenomenon to the axial locomotion, accumulating
important strains in this area of the vertebral column.
Chevron. The isolated chevron (Fig. 14E) is similar to
one of the last chevrons of Physeter, in the area of the
11th-13th caudal vertebrae; its height and width are 111
and 91 mm respectively.
Ribs. The most complete associated rib fragment is
the rst left rib (Fig. 14H), with a curve more open
centrum length centrum width centrum height
A+112 130 116
B131 122 e107
C129 118 113
D119 106 106
Table 1 – Measurements (mm) of the caudal vertebrae of
Physeterula dubusi IRSNB M.527 (holotype),
Miocene of Antwerp. The letters identifying the
vertebrae correspond to Figure 14. The centrum
height does not include the prominent facet for
the chevron. ‘e’ indicates estimate, + nearly
complete.
than in Physeter and Zygophyseter. The capitulum,
the tuberculum, and the distal end are unfortunately
missing. The minimum proximal width of the neck is
71 mm. Other ribs are more fragmentary.
Sternum. The right side of the rst segment of the
sternum (presternum) ankylosed with the second
segment is preserved (Fig. 14L), not signicantly
differing from Physeter except for the much less robust
condition and the less prominent anterolateral corner
(not complete). The total length of these two fused
elements is estimated at 214 mm.
Discussion
Physeterula dubusi is a large sperm whale, with a
size close to Brygmophyseter shigensis, Idiorophus
patagonicus, and Zygophyseter varolai. The lack of
enamel on the teeth places this species in the crown-
group Physeteroidea as dened by BIANUCCI & LANDINI
(2006). The supracranial basin is posteriorly limited by
a high wall, as in Aulophyseter, Orycterocetus, and
Physeter, but the basin is not as anteriorly developed
on the rostrum as in the latter. The right maxilla does
not reach the sagittal plane posteromedially, contrary
to Orycterocetus and similarly to Aulophyseter and
Physeter. The retention of functional maxillary
teeth, at least three dorsal infraorbital foramina, and
the preorbital process positioned at a higher level
along the lateral margin of the rostrum base differ
from Aulophyseter and Physeter. The postorbital
process of the frontal is shorter than in Aulophyseter,
Diaphorocetus, and Orycterocetus. The mandibular
symphysis is shorter and at least some of the caudal
vertebrae are proportionally longer than in Physeter.
Physeteridae indet.
Fig. 15
1877 — Physeterula dubusii (part) VAN BENEDEN, p. 852.
Referred specimen
IRSNB M.1937, third specimen of Physeterula dubusi
sensu VAN BENEDEN (1877), fragmentary skull including
rostrum fragments, most of the supracranial basin, and
supraorbital processes. It was discovered December 3,
1863 in the 3ème section of the inner fortication ring
around Antwerp (MOURLON, 1878; Fig. 1).
Horizon
MOURLON (1878) provided a brief description of the
sediment and a list of bivalves discovered with the
Olivier LAMBERT
301
Skull IRSNB M.1937, Physeteridae indet., Miocene of Antwerp. A. Dorsal view. B. Schematic reconstruction of the
dorsal view. The shaded area corresponds to the posterior portion of the supracranial basin. The relative position of
the right and left side elements is approximate in A-B. The basicranium is completely lacking, reconstructed based on
other physeteroids. C. Right lateral view. D. Area of the bony nasal tracts in ventral view, displaying the asymmetry of
the tracts and the presphenoid uncovered by the vomer. Scale bar for A-C equals 200 mm and for D equals 50 mm.
Fig. 15 –
Miocene sperm whales from the North Sea
302
skull; these elements indicate a Miocene age for this
skull, possibly late Miocene (Diest Formation, Deurne
Sands Member; R. Marquet, pers. comm., 2007).
Description
The specimen IRSNB M.1937 was not gured in
the brief descriptions of Physeterula dubusi by VAN
BENEDEN (1877) and ABEL (1905). The postorbital
width of this moderate size fragmentary skull is
estimated at 470 mm, close to Diaphorocetus poucheti
and Orycterocetus crocodilinus.
Premaxilla. On the rostrum, the premaxillae are poorly
preserved. The preserved length of the right premaxilla
provides a minimum estimation for the rostrum length
of 530 mm. The dorsal surface of the premaxilla
is sloping medially for most of the rostrum length,
suggesting a supracranial basin limited to the cranium
and rostrum base. 120 mm anterior to the antorbital
notch, the surface of each premaxilla becomes concave,
progressively more excavated backwards. A left
premaxillary foramen is present (diameter 8 mm), not
as reduced as in most other physeteroids. The enlarged
right premaxillary foramen (diameter 22 mm) is slightly
anterior to the level of the antorbital notch (Fig. 15A-B).
Lateral to the right bony naris, the right premaxilla has a
V section and an elevated median margin, unfortunately
incomplete, probably originally crossing the sagittal
plane of the skull.
Maxilla. From the antorbital notch, the lateral margin of
the rostrum base is oriented longitudinally for a short
distance before an abrupt narrowing towards the lateral
margin of the premaxilla. Consequently the dorsal
surface of the rostrum was predominantly occupied
by the premaxillae on more than half its length, as in
Aulophyseter and Physeterula. Only a tiny foramen
pierces the maxilla along the suture with the premaxilla,
differing on that point from Physeterula. The U-shaped
antorbital notch is as wide as long. The wide preorbital
process of the maxilla is distinctly lower than the
dorsolateral margin of the rostrum base at the same
level, a condition observed in Aulophyseter, Physeter,
and possibly the poorly preserved Idiophyseter. Two
large foramina pierce the right maxilla lateral to the
wall of the supracranial basin. The median margin of
the anterior foramen overhangs a longitudinal groove
anteriorly directed. The posterior foramen is followed
backwards by a wide sulcus nearly reaching the occipital
crest, a feature absent in Physeterula. A small foramen
on the median wall of this posterior foramen is directed
posteroventrally. In lateral view the maxilla-frontal
suture is sigmoid above the orbit, more erected than in
Physeterula but less than in Aulophyseter and Physeter.
Frontal. The supraorbital process of the frontal is
thick, leaving no space for a projection of the lacrimal
between frontal and maxilla (Fig. 15C). The postorbital
process is robust and shorter than in Aulophyseter. The
temporal fossa is dorsally pointed and anteroposteriorly
compressed.
Supraoccipital. Combined with the shape of the
underlying frontal and maxilla, the preserved
fragments of supraoccipital indicate an erected and
slightly concave occipital shield roughly similar to
Aulophyseter, Orycterocetus, and Physeterula.
Mesethmoid/Vomer. The mesethmoid is preserved at
the level of the bony nasal tracts. The right and left
tracts have a minimum diameter of 24 and 53 mm
respectively. The dorsal keel of the mesethmoid is
tilted to the left side. As noted in Aprixokogia, Kogia,
and Physeter (WHITMORE & KALTENBACH, 2008)
the vomer of IRSNB M.1937 does not cover the
presphenoid ventrally at the level of the bony nasal
tracts (Fig. 15D), supporting the hypothesis that this
condition characterizes at least the stem-physeteroids.
Discussion
The fragmentary specimen IRSNB M.1937, previously
referred to Physeterula dubusi but actually much
smaller (skull size close to Orycterocetus crocodilinus)
displays a dorsally pointed temporal fossa and an
erected slightly concave supraoccipital shield. It
shares with Aulophyseter and Physeter the preorbital
process lower than the dorsolateral margin of the
rostrum base, and with Aulophyseter and Physeterula
the dorsal surface of the rostrum occupied by the sole
premaxillae on more than half its length. In addition
to the higher temporal fossa it differs from Physeter in
the more numerous dorsal infraorbital foramina. Only a
small foramen is present along the maxilla-premaxilla
suture at the rostrum base, differing from Physeterula.
The postorbital process is much shortened compared
to Aulophyseter, Diaphorocetus, and Orycterocetus,
closer to Physeterula and some specimens of Physeter.
This specimen differs from all the known physeterids,
but it is too fragmentary to allow the denition of a new
taxon. It is provisionally referred to Physeteridae indet.
Olivier LAMBERT
303
Family Kogiidae
Genus Thalassocetus ABEL, 1905
Type species: T. antwerpiensis ABEL, 1905
Diagnosis. The diagnosis is the same as for the only
included species.
Thalassocetus antwerpiensis ABEL, 1905
Figs 16-18
1905 — Thalassocetus antwerpiensis (part) ABEL, p. 70, g. 7.
2002 — Placoziphius duboisii (part) KAZÁR, p. 154.
2006 — Thalassocetus antwerpiensis BIANUCCI & LANDINI, p. 125.
Lectotype
IRSNB M.525 (old number 3248).
Type locality
Antwerp, no exact locality.
Type horizon
The dark colour and the preservation state of the bones
of the lectotype suggest an origin in one of the members
of the Berchem Formation, lower to middle Miocene
(LOUWYE, 2001, 2005), possibly the Antwerpen Sands.
Diagnosis
Thalassocetus antwerpiensis is a small physeteroid
species with a skull size close to Kogia breviceps.
It differs from all the other kogiids in the nodular
postglenoid process of the squamosal, and the
thin supraorbital area of the skull, from Kogia and
Scaphokogia in the U-shaped short antorbital notch,
from Aprixokogia, Kogia, and Praekogia in the right
antorbital notch being distinctly outside the supracranial
basin and in larger dorsal infraorbital foramina, from
Aprixokogia and Praekogia in the temporal fossa
being anteroposteriorly compressed, and from Kogia
and Praekogia in the absence of a wide notch in the
squamosal for the enlarged posterior process of the
tympanic bulla.
Comments
The status of the second specimen referred to T.
antwerpiensis by ABEL (1905) (IRSNB M.526) was
only briey commented in KAZÁR (2002). This right
portion of the cranium including the frontal and
maxillary parts of the orbit and temporal fossa has a
size close to IRSNB M.525. The postorbital process
of the frontal is indeed more slender, excavated on
its posterolateral surface; the lateral maxillary crest
limiting the supracranial basin might be lower; and the
elevation of the maxilla towards the posterior wall of
the supracranial basin is less abrupt. Nevertheless, the
differences between these two specimens are clearly less
important than the differences between any of them and
the holotype of P. duboisi. Considering the fragmentary
state of IRSNB M.526 it is nevertheless more careful to
refer it to Kogiidae aff. Thalassocetus.
Fig. 16 – Skull of Thalassocetus antwerpiensis IRSNB M.525 (lectotype), Miocene of Antwerp. A. Anterodorsal view. B.
Corresponding line drawing. Scale bar equals 50 mm.
Miocene sperm whales from the North Sea
304
Description of the lectotype of Thalassocetus
antwerpiensis
ABEL (1905) provided a brief description of
Thalassocetus antwerpiensis; concerning the specimen
identied here as the lectotype only the fragmentary
supracranial basin was illustrated in anterodorsal view.
The only informative elements are the right side and
the posterior wall of the supracranial basin from the
antorbital notch to the supraoccipital crest (lacking
portions of the maxilla), a part of the left supraorbital
process, the right squamosal, and the right exoccipital.
The postorbital width of this small skull is roughly
estimated at 280-290 mm, larger than the holotypes of
Praekogia cedrosensis and Scaphokogia cochlearis, in
the size interval of Kogia breviceps (sample of three
specimens in ROSS, 1984), and signicantly smaller
than all the other physeteroids from Antwerp studied
here. Because the premaxilla is not preserved on the
rostrum the description starts with the maxilla.
Maxilla. The right antorbital notch is widely open
anterolaterally (Fig. 16). Even if the preorbital process
of the maxilla is incomplete, it did not limit laterally
a slit-like notch, contrary to Kogia and Scaphokogia.
Medial to the antorbital notch the maxilla is pierced by
a major infraorbital foramen, larger than in Aprixokogia,
Kogia, and Praekogia. At least two septae, only partly
preserved, indicate a more or less complete subdivision
of this foramen in three branches. The two smaller
posterior branches extend posteriorly as grooves along
the maxillary crest laterally limiting the supracranial
basin: the rst branch on the outer wall of the crest, the
second branch on the inner wall. This condition, with
only one of the dorsal infraorbital foramina opening
inside the supracranial basin is roughly similar to
Orycterocetus crocodilinus (e.g., USNM 14730). The
maxillary crest laterally limiting the basin occupies a
more median position compared to the right antorbital
notch than in Aprixokogia, Kogia, and Praekogia. In the
two last genera the notch enters the supracranial basin.
On the preorbital process the outer margin of the maxilla
is thin in lateral view, contrasting with other kogiids.
Premaxilla. The right premaxilla is only preserved
on the posterior wall of the supracranial basin, for a
large part located on the left side of the skull beyond
the sagittal plane. As previously suggested by ABEL
(1905) and BIANUCCI & LANDINI (2006), this area
strongly reminds the sagittal crest of Aprixokogia,
Kogia, and Praekogia (the sagittal crest of Scaphokogia
is probably secondarily modied by the enlargement
of the supracranial basin; MUIZON, 1988). The
premaxilla is posteriorly pointed. Its transverse section
is triangular; the medial face is applied along the curved
median margin of the left maxilla, forming together
a crest overhanging the maxilla (Figs. 16-17A). The
surface of the premaxilla is slightly concave at its
widest preserved level. In Kogia this depression (Fig.
17B) corresponds to the seat of the nasofrontal sac
and adjoining small spermaceti organ (SCHENKKAN &
PURVES, 1973; CRANFORD et al., 1996; Fig. 20B-C).
If the hypothesized homology with the sagittal crest
of Aprixokogia, Kogia, and Praekogia is correct, a
similar soft anatomy structure might have been present
in Thalassocetus. In Physeter, the enlarged spermaceti
organ occupies most of the supracranial basin and the
A. Vertex of the skull of Thalassocetus antwerpiensis IRSNB M.525 (lectotype), Miocene of Antwerp, in dorsal
view. Scale bar equals 30 mm. B. Vertex of the skull of Recent Kogia breviceps ZMA 14.817, in dorsal view. Note
the similarities at the level of the sagittal crest.
Fig. 17 –
Olivier LAMBERT
305
Skull of Thalassocetus antwerpiensis IRSNB M.525 (lectotype), Miocene of Antwerp. A. Supraorbital area and
squamosal/exoccipital in right lateral view. The relative position of the two elements is approximate. B. Right
squamosal/exoccipital in anteroventral view. Scale bar for A-B equals 50 mm.
Fig. 18 –
right premaxilla is at and much wider than in Kogia
(see discussion below; Fig. 20A).
In non-kogiid physeteroids, the left nasal is located
median to the left maxilla, posterior to the left bony
naris. Because this area is occupied by the right
premaxilla in Thalassocetus, the left nasal is probably
absent in the latter, as in Aprixokogia, Kogia, Praekogia,
and Scaphokogia.
Frontal. The geometric relationship between the
supraorbital process of the frontal and the overlying
maxilla might be inaccurate; the past reconstruction
of the skull fragment from several elements lead to
slight mistakes and the orbit might be articially too
anterolaterally directed, also modifying somewhat the
shape of the temporal fossa. The preorbital process of
the frontal is not complete. The postorbital process is
long (73 mm) and pointed, roughly vertically oriented
(Fig. 18A). In lateral view the frontal-maxilla suture
is much less erected than proposed by ABEL (1905),
with an angle no more than 20-25° to the horizontal
plane for most of its length. Where it gets close to the
supraoccipital crest the suture curves to a nearly vertical
orientation. The preserved dorsal and anterior outlines
of the temporal fossa suggest a fossa anteroposteriorly
compressed, longitudinally shorter than in Aprixokogia
and Praekogia. On the posterior wall of the supracranial
basin, most of the covering of the frontals by the
maxillae is missing. In anterodorsal view, the outline of
the frontals on the cranium is posteromedially pointed,
more than in Aprixokogia, Kogia, and Praekogia.
Parietal. On the supraoccipital crest, between the
frontal and the supraoccipital is a narrow strip of
bone (Figs 16-17A) that probably corresponds to the
parietal, reaching the top of the crest as suggested by
ABEL (1905), medially longer than in Idiophyseter (see
comment in KELLOGG, 1927). The dorsomedian area of
the supraoccipital is slightly concave and nearly vertical
at its dorsal end. The general slope of the supraoccipital
shield was probably around 60-70°, as in Kogia (the
slope is steeper in some individuals of the latter).
Miocene sperm whales from the North Sea
306
Squamosal. The zygomatic process of the squamosal
is short (Fig. 18A). Even if no bony contact with
the dorsal part of the cranium is preserved, the
reconstructed outline of the temporal fossa indicates
that the zygomatic process was originally overhung
to some extent by the postorbital process of the
frontal, as seen in Kogia. The distance from the apex
of the zygomatic process to the ventral end of the
postglenoid process is 64 mm. The dorsal margin of
the zygomatic process elevates posteriorly and bears
a low hump at mid-length. The postglenoid process
is nodulous, not as anteroposteriorly attened as in
most other physeteroids, and short, reaching the same
ventral level as the post-tympanic process. The distance
between the postglenoid process and the exoccipital is
not as important as in Kogia and Praekogia, leaving no
space for an enlarged posterior process of the tympanic
bulla as in the two latter, more similar to Aprixokogia
on this point. However, individual variation has been
proposed at this level in the latter, based on isolated
squamosals (WHITMORE & KALTENBACH, 2008). The
glenoid fossa is narrow, surrounded anteromedially and
posteromedially by a developed tympanosquamosal
recess (Fig. 18B). The remains of falciform process are
broken at the base, probably less reduced than in Kogia,
Physeter, and Physeterula. Along the narrow external
auditory meatus, the ventral surface of the post-
tympanic process is grooved, indicating the surface of
contact with, or elements of, the posterior process of the
tympanic bulla, elongated but narrow.
Discussion
The temporal fossa of the small Thalassocetus
antwerpiensis is anteroposteriorly compressed, as
in physeteroids more derived than Brygmophyseter
and Zygophyseter (BIANUCCI & LANDINI, 2006). The
postglenoid process of the squamosal is nodular rather
than anteroposteriorly attened, differing from other
physeteroids.
As previously mentioned (ABEL, 1905; BIANUCCI &
LANDINI, 2006), Thalassocetus shares with members of
the family Kogiidae a sagittal crest between the bony
nares and the supraoccipital. The polarity of the character
discussed here is still somewhat controversial: either
the developement of a sagittal crest is plesiomorphic,
present in some basal physeteroids (not only kogiids)
and subsequently lost in derived physeterids, in relation
with the increased development of the spermaceti
organ (MUIZON, 1991), or the presence of a crest is a
synapomorphy of the kogiids (BIANUCCI & LANDINI,
2006). As until now no basal physeteroid has shown a
crest obviously homologous to the kogiid sagittal crest,
the second hypothesis is preferred.
Among the other morphological areas providing
characters dening the kogiids (MUIZON, 1988, 1991;
BIANUCCI & LANDINI, 1999, 2006), only the right
antorbital notch is preserved in Thalassocetus; it is
clearly not slit-like, contrary to Kogia, Scaphokogia,
and possibly Praekogia. Furthermore, the notch is
located outside the supracranial basin and the dorsal
infraorbital foramina are large, differing from the
Kogiinae MUIZON, 1991 (= Aprixokogia + Kogia +
Praekogia). The supraorbital area is thinner in lateral
view than in other kogiids. The supracranial basin
is not as deep and as extended posteriorly as in the
monogeneric subfamily Scaphokogiinae MUIZON,
1988. This genus might represent the most basal
member of the kogiid lineage.
Phylogeny
The phylogenetic analysis is based on the matrix
of BIANUCCI & LANDINI (2006), to which several
characters and taxa have been added (Appendices 1-
2), with a total of 36 morphological characters and 16
taxa. Several characters were modied according to
recently published data and additional observations.
Two outgroups are chosen a priori, the archaeocete
Zygorhiza and the stem-odontocete Agorophius. All the
characters are treated as unordered.
The cladistic analysis (Paup 4.0b10; heuristic search;
character-state optimisation acctran) produced seven
most parsimonious trees (tree length 69 steps; CI 0.72;
RI 0.76). The consensus tree is shown in Figure 19, on a
stratigraphic framework. Besides the addition/removal
of some taxa this tree differs from the consensus
tree of BIANUCCI & LANDINI (2006) in the more
crownward position of Diaphorocetus (depending on
the interpretation of the damaged supraoccipital shield),
the more stemward position of Orycterocetus and
Placoziphius, and the nested position of Aulophyseter
among the Physeteridae (based on the new information
published by KIMURA et al., 2006). Keeping in mind
the low resolution at the level of Diaphorocetus,
Orycterocetus, and Placoziphius, a new denition and
content of the families Kogiidae and Physeteridae is
proposed here. The Kogiidae group all the physeteroids
more closely related to Kogia than to Physeter. The
main characters dening the clade are: small size;
development of a sagittal crest; loss of the two nasals;
maxillae, premaxillae, and vomer reaching the tip of
the rostrum; and posterior process of the periotic not
ventrally oriented (the two last synapomorphies are
Olivier LAMBERT
307
Fig. 19 – Consensus tree of the phylogenetic analysis, presented on a stratigraphic frame. The length of the branches is
therefore not proportional to the number of steps towards a given taxon. Tree length 69 steps; CI 0.72; RI 0.76. 1
Physeteroidea; 2 Physeteridae; 3 Kogiidae. The thick bars indicate the stratigraphic record of the species; question
marks associated to the bars indicate stronger uncertainty on dates. Data from areas other than the North Sea are
taken from BARNES, 1973; BIANUCCI & LANDINI, 1999, 2006; COZZUOL, 1996; HIROTA & BARNES, 1995; KELLOGG,
1927, 1965; MUIZON, 1988. The schematic jaws illustrate the presence/absence of functional maxillary teeth in
the different taxa as proposed by the parsimony analysis; question marks associated to the jaws indicate equivocal
states.
unknown in Thalassocetus). The narrower posterior
portion of the right premaxilla possibly constitutes
another synapomorphy of the Kogiidae.
The Physeteridae group all the physeteroids more
closely related to Physeter than to Kogia. This clade
is only poorly dened, with the dorsal exposure of
the maxilla on the rostrum limited to the posterior
portion (reversion in Physeter). In this family there is
also a trend in the increase of size, but a similar trend
characterizes the clade Brygmophyseter + Zygophyseter.
With better preserved specimens of Diaphorocetus and
Placoziphius changes in the content and denition of
the Physeteridae can be expected.
Focusing on the Miocene North Sea species,
Eudelphis is the most stemward physeteroid in
the proposed consensus tree, possessing a distinct
supracranial basin, asymmetric bony nares, a widened
and posteriorly longer right premaxilla, and a triangular
zygomatic process of the squamosal in lateral view,
but retaining a large left premaxillary foramen and a
long postglenoid process of the squamosal. This basal
position is further supported by the rather conservative
general morphology of the premaxillae at the rostrum
base. The branch of Eudelphis is followed by a clade
Miocene sperm whales from the North Sea
308
grouping the Mediterranean Zygophyseter and the
Pacic Brygmophyseter. The position of Placoziphius
is not resolved, partly due to its fragmentary state.
Nevertheless it provisionally falls outside the family
Physeteridae because of the long dorsal exposure of the
maxilla on the rostrum. Physeterula is the most basal
member of this family. The undetermined specimen
IRSNB M.1937 is more closely related to Aulophyseter
and Physeter, based on the lower level of the preorbital
process compared to the lateral margin of the rostrum
base. Finally the cladogram suggests a sister-group
relationship between Thalassocetus and all the other
known kogiids (see above for the synapomorphies of
the Kogiidae).
Considering the stratigraphic frame of the cladogram
(Fig. 19) the early Miocene age of several physeteroids
points to an Oligocene origin for the superfamily, as
proposed by the discovery of fragmentary remains from
this epoch (MCHEDLIDZE, 1970; FORDYCE & MUIZON,
2001). Several lineages branch before the appearance
of the crown-Physeteroidea (including the last common
ancestor of Kogia and Physeter, and all its descent),
including the branch towards Eudelphis and the clade
Brygmophyseter + Zygophyseter. The oldest Kogiidae
is probably Thalassocetus, but new material found in
situ would better support this hypothesis. It should be
noted that a fossil species of Kogia, K. pusilla, has been
described in the middle Pliocene of Italy (BIANUCCI
& LANDINI, 1999), partly lling the gap between
Praekogia and Kogia. The oldest Physeteridae, but
not the most stemward, is Aulophyseter, dated from
the middle Miocene. Therefore the rst Physeteridae
should have appeared during the early Miocene or
before. A long ghost lineage separates Physeter from
its sister-group Aulophyseter, although postcranial
remains from the Pliocene of Piedmont, Italy, have
been referred to Physeter (PARONA, 1930). Because
the divergence between the well dened Kogiidae and
Placoziphius/Diaphorocetus is not later than very early
in the Miocene, the origin of the crown-Physeteroidea
must probably be sought in Oligocene levels. A more
precise dating of the different cladogeneses will depend
on the improvement of stratigraphic information and a
better resolved phylogeny.
Functional anatomy and palaeoecology
In the evolutionary history of the sperm whales, two
features deserve special attention and are commented
here: the spermaceti organ and the oral apparatus.
Evolution of the spermaceti organ and associated
structures
The spermaceti organ is a fat-lled cavity of the
forehead of Recent physeteroids surrounded by a
connective tissue wall (MEAD, 1975). The proposed
homology with the right posterior dorsal bursa of
non-physeteroid odontocetes supports the hypothesis
of a similar function in the production of echolocative
sounds (see CRANFORD et al., 1996; CRANFORD, 1999;
MADSEN et al., 2002). In Physeter the spermaceti
organ is the most voluminous head structure, running
from the posterior wall of the supracranial basin to the
level of the blowhole, farther anteriorly than the tip of
the rostrum of adult males (Fig. 20A). It is separated
from the dorsal surface of the rostrum by another
large structure, the junk, made of a series of lens-like
concavoconvex lipid elements limited by connective
tissue columns (CLARKE, 1978). The junk of Physeter
has been proposed to be homologous with the melon of
non-physeteroid odontocetes, much likely involved in
the conduction of echolocative sounds (CRANFORD et
al., 1996; for an alternative function of the spermaceti
organ and junk in Physeter, namely a battering ram
for male-male ghts, see CARRIER et al., 2002). As
commented above, the spermaceti organ is considerably
smaller in Kogia, occupying a reduced volume of the
supracranial basin close to the dish-like pedestal of the
right premaxilla (part of the sagittal crest). Most of the
anterior part of the forehead of Kogia is instead lled by
the melon (CRANFORD et al., 1996, g. 8; Fig. 20B-C).
The asymmetry of the bony nares and surrounding
bony elements is already much pronounced in the
oldest, most stemward physeteroids (e.g. Eudelphis),
more than in any other odontocete. It is therefore
pertinent to propose a similar (even if to a lower degree)
asymmetry for the forehead soft anatomy, including the
nasal passages, air sacs, and sound production structures
(phonic lips). This condition supports the existence
of a spermaceti organ in all known physeteroids.
Considering the size of the spermaceti organ in the two
Recent physeteroid genera Kogia and Physeter, the size
and anterior extent of the supracranial basin in fossil
physeteroids, and the size of the posterior dorsal bursae
in other odontocetes (see CRANFORD et al., 1996), it
is more parsimonious to propose a relatively small
spermaceti organ in basal physeteroids, considerably
shorter anteriorly than in Physeter (BIANUCCI &
LANDINI, 2006), but also less voluminous, not
occupying the whole supracranial basin. In that case,
the primary function of the supracranial basin would not
be to contain the spermaceti organ, but rather to act as a
parabolic structure reecting the sounds produced in an
Olivier LAMBERT
309
Fig. 20 – Soft anatomy of the forehead in the extant Physeter
and Kogia. A. Schematic parasagittal section of
the head of Physeter (modied from CRANFORD,
1999). B-C. Tomography reconstructions of
the head of Kogia (modied from CRANFORD
et al., 1996). cr cranium; so spermaceti organ.
B. Vertical (parasagittal) section. C. Horizontal
(frontal) section. The hypothetical homologies
between cephalic structures are highlighted with
similar shading. Note the enlarged spermaceti
organ, the longer rostrum, and the anteriorly
shifted blowhole in Physeter. Drawings are not at
the same scale.
anterodorsal direction. Further supporting the primary
role of the supracranial basin as a parabolic structure, the
fact that the skull is projected ventrally to the body axis
in both Kogia and Physeter (CRANFORD, 1999) suggests
a trend towards a more efcient anterior focus of the
echolocative sounds. As is the case in Kogia, the most
voluminous structure in the supracranial basin of the
stem-physeteroids would have been the melon. And the
blowhole was probably located roughly vertical to the
bony nares, as in the latter (for an alternative hypothesis
with a larger spermaceti organ and a somewhat anteriorly
shifted blowhole in stem-physeteroids see BIANUCCI &
LANDINI, 2006). Besides the development of a small
spermaceti organ, the most basal physeteroids possibly
retained some forehead structures inherited from more
‘classical’ odontocetes, for example the premaxillary
sacs and the nasals plugs (see above, the conservative
morphology of the premaxillae around the premaxillary
foramina in Eudelphis and possibly Diaphorocetus).
In the family Physeteridae there is a trend towards
an anterior development of the supracranial basin,
occupying the posterior portion of the rostrum. But in
no fossil physeterid the basin reaches the development
seen in Physeter; in the latter the dorsal surface of the
rostrum is concave and considerably widened along its
whole length. So the spermaceti organ of known fossil
physeterids was probably not as anteriorly long and
voluminous as in Physeter. However, it should be noted
that the rather fragmentarily known middle Miocene
Idiophyseter possibly possessed an anteriorly extended
supracranial basin (see KELLOGG, 1925). At some
unknown time of the evolution of the physeterids, the
blowhole and the underlying right side sound production
structures shifted to a more anterior location, leading to
a longer nose. Such a feature might have been selected
for the production of lower frequency sound and/or for
acoustic display, as suggested by CRANFORD (1999).
Interestingly the rostrum of Physeter is proportionally
longer than in related physeteroids (e.g. Aulophyseter),
suggesting that a longer rostrum (edentulous, as
discussed below) has been selected in the former to
support the longer spermaceti organ and junk, leading
to an increased distance between the sound production
area and the posterior wall of the supracranial basin. It
should be noted that the trend towards a larger body size
is not unique to physeterids (Aulophyseter, Physeter,
and Physeterula), as some stem-physeteroids are also
large (e.g., Brygmophyseter and Zygophyseter), even if
not as large as Physeter. Additionally KELLOGG (1925)
refers anterior fragments of a rostrum, mandible, and
teeth from the late Miocene of California to a huge
enamel-bearing physeteroid, providing an estimation of
Miocene sperm whales from the North Sea
310
the skull length of 12-15 feet (3.7-4.6 m).
Besides the trend towards a reduction of the body
size, in the Kogiidae the spermaceti organ remains
small and, except in Scaphokogia, it lies on a bony
shelf (sagittal crest) somewhat isolating it from the
oor of the supracranial basin. This modication is
accompanied by the loss of the left nasal. As proposed
by MUIZON (1988) the size and the smooth surface
of the supracranial basin in Scaphokogia suggests
an important development of the spermaceti organ,
progressively lling the basin and reducing the sagittal
crest. However, contrary to the lineage towards Physeter
the enlarged supracranial basin is limited to the cranium
by an abrupt step in Scaphokogia, not extending on the
rostrum base. There is therefore no clue for an anterior
shift of the narial passages and blowhole in kogiids.
Evolution of the feeding apparatus
Both the large Physeter and the considerably smaller
Kogia lack functional upper teeth; in these two genera
only vestigial teeth sometimes erupt on the maxilla
(CALDWELL & CALDWELL, 1989; RICE, 1989).
Furthermore the crown of the lower teeth does not
bear enamel any more, the temporal fossa is relatively
reduced, and in Physeter a common anomaly of the
mandible leading to a strong lateral curvature does
not lead to the death of the individual (RICE, 1989).
Together with behavioural and anatomical observations
(WERTH, 2004; BLOODWORTH & MARSHALL, 2005) all
these elements support the idea that Recent physeteroids
mostly rely on suction for catching and swallowing their
prey (mostly squid). In Kogia, the strongly shortened
rostrum and mandible (amblygnathy, K. breviceps and
K. sima are the two odontocete species with the highest
bluntness index; WERTH, 2004) reduce the length of
the oral opening and correlate well with other suction
feeding clues, whereas Physeter retains long jaws and
sucks the prey directly into the oropharyngeal opening
(WERTH, 2004).
Freed from their feeding role the robust lower
teeth of Physeter could be used for combat, as seen in
other odontocete lineages, e.g. the ziphiids (WERTH,
2006); tooth scars inicted by other large males are
frequent on the head of Physeter males (KATO, 1984;
RICE, 1989), suggesting that beside vocalization ghts
might be occasionally determinant for breeding success
(CARRIER et al., 2002; WHITEHEAD, 2003). Differing
from Physeter macrocephalus, with adult males several
meters larger than females, no obvious size dimorphism
is detected in Kogia breviceps and K. sima (ROSS,
1984; CALDWELL & CALDWELL, 1989). Parallel rows
of scars have been observed in Kogia too, but on both
sexes, presumably related to ghts during the mating
season (GASKIN, 1972). However, the lower teeth of
Kogia are much thinner and sharper than in Physeter,
borne on a slender and short underslung mandible not
best tted for combat. These elements, together with
the body size, the prominent dorsal n, and the ‘false
gill’ white coloration between the ipper and the eye,
give Kogia the appearance of a bottom feeding shark
(CALDWELL & CALDWELL, 1989; WERTH, 2006). This
mimicry could have been selected to repel potential
predators. Additionally, the elongated lower teeth
might be used by Kogia to retain the food after having
suctioned it, during expulsion of water (BLOODWORTH
& MARSHALL, 2005).
The most stemward physeteroids retain robust
upper and lower teeth (especially Brygmophyseter and
Zygophyeter) (Fig. 19), bearing enamel, and associated
with a large temporal fossa. These taxa used their teeth
for catching large preys, possibly cutting them with
the posterior teeth (BIANUCCI & LANDINI, 2006). The
relatively more slender teeth of the smaller Eudelphis
probably allowed the capture of somewhat smaller
preys. Among the physeterids, at least Physeterula
retains functional upper teeth, but its teeth lack enamel,
meaning that the loss of enamel occurred before the loss
of functional maxillary teeth.
There is little information available on the oral
apparatus of fossil kogiids (BARNES, 1973; BIANUCCI
& LANDINI, 1999). In Aprixokogia the alveolar groove
retains at least seven large and deep alveoli (WHITMORE
& KALTENBACH, 2008) and in Scaphokogia a shallow
alveolar groove containing some relict aveoli is
described along the moderately shortened rostrum
(MUIZON, 1988). No teeth were found associated
with any fossil kogiid. With the elements in hand it is
more parsimonious to propose that the loss of enamel
occurred one time in the evolutionary history of the
physeteroids, before the node grouping kogiids and
physeterids (as in the tree of BIANUCCI & LANDINI,
2006, but differing from MUIZON, 1991). The temporal
fossa of Thalassocetus and Praekogia is reduced,
indicating a jaw musculature less voluminous than in
stem-physeteroids, closer to the Recent Kogia.
The presence of functional upper teeth in at least one
physeterid suggests that the reduction of the dentition
occurred in parallel in both lineages: kogiids and
physeterids, possibly related to a similar diet change
(major food items of Physeter are cephalopods, but the
diet of Kogia is somewhat more varied; RICE, 1989;
CALDWELL & CALDWELL, 1989). This convergence
supports the idea of an independent improvement of
the suction feeding function in multiple odontocete
Olivier LAMBERT
311
Conclusions
1. Based on the cranial material of the Antwerp area, the
systematic revision of the sperm whale species from the
Miocene of the southern margin of the North Sea lead
to the detailed redescription of four species: the stem-
physeteroids Eudelphis mortezelensis and Placoziphius
duboisi, the physeterid Physeterula dubusi, and the
kogiid Thalassocetus antwerpiensis, the identication
of the eastern North American species Orycterocetus
crocodilinus in the North Sea, and the description of a
new undetermined physeterid.
2. The precision of the stratigraphic information
associated to several specimens has been improved.
All the specimens originate from Miocene levels.
Placoziphius duboisi is dated from early Miocene,
Eudelphis mortezelensis from latest early to middle
Miocene, and Orycterocetus crocodilinus from
the middle Miocene. The incertitude is higher for
Thalassocetus antwerpiensis (much likely early to
middle Miocene), Physeterula dubusi, and the new
undetermined physeterid (probably late Miocene).
The fossil record in the North Sea further stresses
the relatively high physeteroid diversity during the
Miocene, compared to the two relict Recent genera
Kogia and Physeter. The Pliocene gap in the fossil
record of the physeteroids is likely partly due to changes
in deposition conditions; indeed isolated physeteroid
teeth, often bearing enamel, are not rare in Pliocene
shallow marine deposits.
3. The cladistic analysis produced a consensus tree
with Eudelphis as the most stemward physeteroid
known by cranial material. Eudelphis displays a distinct
supracranial basin and a strong asymmetry of the bony
nares and premaxillae, but it retains enamel on teeth and
its morphology is conservative in several areas of the
skull (rostrum shape, maxillary alveoli, premaxillary
foramina, falciform process of the squamosal…). The
family Physeteridae is dened as the clade grouping all
the physeteroids more closely related to Physeter than to
Kogia. It includes among others the large Physeterula,
retaining functional upper teeth, and the moderate
size IRSNB M.1937. Placoziphius is provisionally
placed outside the family Physeteridae, together with
Orycterocetus. Displaying a sagittal crest the small
Thalassocetus is conrmed as a kogiid, sister-group to
all the other members of the family.
4. Based on the anatomy of the different taxa and on the
framework of the proposed phylogeny, the evolutionary
history of the supracranial basin and the oral apparatus
are commented. Already developed on the cranium of
Eudelphis the supracranial basin probably originates
during the Oligocene. Parsimony suggests that the
basin contained a relatively small spermaceti organ
in early physeteroids, leading to the proposal that the
basin mostly acted as a parabolic structure for reection
and anterior focusing of the echolocative sounds. In
the physeterid clade the spermaceti organ became
considerably enlarged towards Physeter, eventually
growing anteriorly on the dorsal surface of the rostrum
to produce the enormous nose of Physeter. A strong
sexual dimorphism observed in this area for Physeter
might indicate a function in acoustic sexual display.
In kogiids the spermaceti organ remained small, being
more distinctly separated from the bottom of the
supracranial basin by the development of the sagittal
crest.
All the stem-physeteroids for which this area is
known bear functional upper teeth with enamel on the
crown. The enamel is probably lost before the clade
Kogiidae + Physeteridae. A reduction of the upper
teeth associated to the decreasing size of the temporal
fossa is a trend observed in parallel in both lineages of
crown-physeteroids: the small kogiids and the larger
physeterids, likely associated to a major change in diet.
The lower teeth are secondarily used as weapons in
Physeter, whereas in Kogia they keep a function in food
processing.
5. Even if fragmentary the cranial physeteroid material
of the North Sea taken as a whole proved to be a useful
source of information for investigating different aspects
of the evolutionary history of the sperm whales during
the Miocene. Besides the promising current anatomical,
ecological, and experimental works on Recent sperm
whales, new fossils from different times will be crucial
to better understand the origin and development of
the various adaptations (deep diving, echolocation,
food processing, sexual dimorphism…) of the Recent
members of this fascinating cetacean group.
Acknowledgements
I wish to thank S. Berton and M. Bosselaers for the preparation
of most of the specimens described here, M. Bosselaers for the
discovery of new specimens in Berchem, W. Miseur for his long
Miocene sperm whales from the North Sea
lineages, progressively replacing the grasping function
of the jaws by a more efcient suction apparatus
(WERTH, 2006).
312
and patient photographic work, R. Marquet for indications on the
stratigraphy and associated molluscan faunas in the area of Antwerp,
D. J. Bohaska (USNM), J. G. Mead and C. W. Potter (USNM), C.
de Muizon (MNHN), G. Lenglet (IRSNB), and A. Rol (ZMA) for
kindly providing access to the specimens under their care, and the
reviewers G. Bianucci and C. de Muizon for their constructive
comments. My work at the IRSNB is nancially supported by the
Research Project MO/36/016 of the Belgian Federal Science Policy
Ofce. This paper is dedicated to Annie Dhondt (1942-2006), whose
advice and support proved to be crucial during my rst years at the
IRSNB.
References
ABEL, O., 1905. Les Odontocètes du Boldérien (Miocène
supérieur) des environs d’Anvers. Mémoires du Musée Royal
d’Histoire Naturelle de Belgique, 3: 1-155.
BARNES, L.G., 1973. Praekogia cedrosensis, a new genus and
species of fossil pygmy sperm whale from Isla Cedros, Baja
California, Mexico. Contributions in Science Los Angeles
County Museum, 247: 1-20.
BIANUCCI, G. & LANDINI, W., 1999. Kogia pusilla from
the Middle Pliocene of Tuscany (Italy) and a phylogenetic
analysis of the family Kogiidae (Odontoceti, Cetacea). Rivista
Italiana di Paleontologia e Stratigraa, 105 (3): 445-453.
BIANUCCI, G. & LANDINI, W., 2006. Killer sperm whale: a
new basal physeteroid (Mammalia, Cetacea) from the Late
Miocene of Italy. Zoological Journal of the Linnean Society,
148: 103-131.
BIANUCCI, G., LANDINI, W. & VAROLA, A., 2004. First
discovery of the Miocene northern Atlantic sperm whale
Orycterocetus in the Mediterranean. Geobios, 37: 569-573.
BLOODWORTH, B. & MARSHALL, C.D., 2005. Feeding
kinematics of Kogia and Tursiops (Odontoceti: Cetacea):
characterization of suction and ram feeding. The Journal of
Experimental Biology, 208: 3721-3730.
BUCHHOLTZ, E.A., 2001. Vertebral osteology and swimming
style in living and fossil whales (Order Cetacea). Journal of
Zoology, 253 (2): 175-190.
BRISSON, M.-J., 1762. Regnum Animale in classes IX
distributum, sine synopsis methodical, 296 pp. Theodorum
Haak, Paris.
CALDWELL, D.K. & CALDWELL, M.C., 1989. Pygmy sperm
whale Kogia breviceps (de Blainville, 1838): Dwarf sperm
whale Kogia simus Owen, 1866. In: RIDGWAY, S. H. &
HARRISON, R. (Editors), Handbook of Marine Mammals, vol.
4: River dolphins and the larger toothed whales, pp. 235-260.
Academic Press, London.
CARRIER, D.R., DEBAN, S.M. & OTTERSTROM, J., 2002. The
face that sank the Essex: potential function of the spermaceti
organ in aggression. Journal of Experimental Biology, 205:
1755-1763.
CLARKE, M.R., 1978. Structure and proportions of the
spermaceti organ in the sperm whale. Journal of the Marine
Biological Association of the United Kingdom, 58: 1-17.
COPE, E.D., 1868. An addition to the vertebrate fauna of the
Miocene period, with a synopsis of the extinct Cetacea of
the United States. Proceedings of the Academy of Natural
Sciences of Philadelphia, 19 (4): 138-157.
COPE, E.D., 1890. The Cetacea. The American Naturalist, 24:
599-616.
COPE, E.D., 1895. Fourth contribution to the marine fauna of
the Miocene period of the United States. Proceedings of the
American Philosphical Society, 34: 135-155.
COZZUOL, M.A., 1996. The record of the aquatic mammals
in southern South America. Münchner Geowissenschaftliche
Abhandlungen (A), 30: 321-342.
CRANFORD, T.W., 1999. The sperm whale’s nose: sexual
selection on a grand scale? Marine Mammal Science, 15 (4):
1133-1157.
CRANFORD, T.W., AMUNDIN, M. & NORRIS, K.S., 1996.
Functional morphology and homology in the Odontocete
nasal complex: implications for sound generation. Journal of
Morphology, 228: 223-285.
CRANFORD, T.W., MCKENNA, M.F., SOLDEVILLA, M.S.,
WIGGINS, S.M., GOLDBOGEN, J.A., SHADWICK, R.E., KRYSL,
P., ST. LEGER, J.A. & HILDEBRAND, J.A., 2008. Anatomic
geometry of sound transmission and reception in Cuvier’s
beaked whale (Ziphius cavirostris). The Anatomical Record,
291 (4): 353-378.
DAWSON, S.D. & GOTTFRIED, M.D., 2002. Paleopathology
in a Miocene kentriodontid dolphin (Cetacea: Odontoceti).
Smithsonian Contributions to Paleobiology, 93: 263-270.
DE MEUTER, F.J., WOUTERS, K. & RINGELE, A., 1976.
Lithostratigraphy of Miocene sediments from temporary
outcrops in the Antwerp city area. Service Géologique de
Belgique, Professional Paper, 3: 1-19.
DU BUS, B.A.L., 1867. Sur quelques Mammifères du Crag
d’Anvers. Bulletin de l’Académie Royale des Sciences, des
Lettres et des Beaux-Arts de Belgique, 24: 562-577.
DU BUS, B.A.L., 1872. Mammifères nouveaux du Crag
d’Anvers. Bulletin de l’Académie Royale des Sciences, des
Lettres et des Beaux-Arts de Belgique, 34: 491-509.
FLOWER, W.H., 1867. Description of the skeleton of Inia
geoffrensis and the skull of Pontoporia blainvillii, with
remarks on the systematic position of these animals in the
Order Cetacea. Transactions of the Zoological Society of
London, 6: 87-116.
FLOWER, W.H., 1869. On the osteology of the cachalot or
sperm-whale (Physeter macrocephalus). Transactions of the
Zoological Society of London, 6: 309-372.
FORDYCE, R.E., 1994. Waipatia maerewhenua, new genus
Olivier LAMBERT
313
and new species (Waipatiidae, new family), an archaic
late Oligocene dolphin from New Zealand. In: Berta, A. &
Deméré, T. A. (Editors), Contributions in marine mammal
paleontology honoring Frank C. Whitmore, Jr. Proceedings
of the San Diego Society of Natural History, 29: 147-178.
FORDYCE, R.E., 2002. Simocetus rayi (Odontoceti:
Simocetidae, new family): a bizarre new archaic Oligocene
dolphin from the eastern North Pacic. Smithsonian
Contributions to Paleobiology, 93: 185-222.
FORDYCE, R.E. & MUIZON, C. DE, 2001. Evolutionary history
of cetaceans: a review. In: MAZIN, J.-M. & DE BUFFRÉNIL, V.
(Editors), Secondary adaptation of tetrapods to life in water,
pp. 169-233. Verlag Dr. Friedrich Pfeil, München.
FRASER, F.C. & PURVES, P.E., 1960. Hearing in cetaceans:
Evolution of the accessory air sacs and the structure of the
outer and middle ear in recent cetaceans. Bulletin of the
British Museum (Natural History), Zoology, 7: 1-140.
GASKIN, D.E., 1972. Whales, dolphins and seals; with special
reference to the New Zealand region, 200 pp. Heinemann
Educational Books, London.
GINSBURG, L. & JANVIER, P., 1971. Les mammifères marins
des faluns miocènes de la Touraine et de l’Anjou. Bulletin du
Muséum National d’Histoire Naturelle, Paris, 22: 161-195.
GONDAR, D., 1975. La presencia de cetaceos Physeteridae
en el Terciario Suprior (‘Rionegrense’) de la Provincia
de Rio Negro. Actas del Primier Congreso Argentino de
Paleontologia y Biostratigraa Tucuman, Argentina, 2:
349–354.
GRATELOUP, J.P.S., 1840. Considérations générales sur la
géologie et la zoologie fossile de la commune de Léognan,
près Bordeaux. Actes de la Société Linnéenne de Bordeaux,
11: 335-346.
GRAY, J.E., 1821. On the natural arrangement of vertebrose
animals. London Medical Repository, 15 (1): 296–310.
HAMPE, O., 2006. Middle/late Miocene hoplocetine sperm
whale remains (Odontoceti: Physeteridae) of North Germany
with an emended classication of the Hoplocetinae. Fossil
Record, 9 (1): 61-86.
HEYNING, J.E., 1989. Comparative facial anatomy of beaked
whales (Ziphiidae) and a systematic revision among the
families of extant Odontoceti. Contributions in Science,
Natural History Museum of Los Angeles County, 405: 1-64.
HEYNING, J.E. & MEAD, J.G., 1996. Suction feeding in
beaked whales: morphological and observational evidence.
Contributions in Science, Natural History Museum of Los
Angeles County, 464: 1-12.
HIROTA, K. & BARNES, L.G., 1995. A new species of Middle
Miocene sperm whale of the genus Scaldicetus (Cetacea;
Physeteridae) from Shiga-Mura, Japan. The Island Arc, 3:
453-472.
ICZN, 1999. International Code of Zoological Nomenclature.
4th ed. 306 pp. International Trust for Zoological
Nomenclature, London.
KASUYA, T., 1973. Systematic consideration of recent toothed
whales based on the morphology of tympano-periotic bone.
Scientic Reports of the Whales Research Institute, Tokyo,
25: 1-103.
KATO, H. Observation of tooth scars on the head of male
sperm whale, as an indication of intra-sexual ghting.
Scientic Reports of the Whales Research Institute, 35: 39-
46.
KAZÁR, E., 2002. Revised phylogeny of the Physeteridae
(Mammalia: Cetacea) in the light of Placoziphius Van
Beneden, 1869 and Aulophyseter Kellogg, 1927. Bulletin
de L’Institut Royal des Sciences Naturelles de Belgique,
Sciences de la Terre, 72: 151-170.
KELLOGG, R., 1925. Two fossil physeteroid whales from
California. Contributions to Palaeontology from the Carnegie
Institution of Washington, 348: 1-34.
KELLOGG, R., 1927. Study of the skull of a fossil sperm-
whale from the Temblor Miocene of southern California.
Publications of the Carnegie Institution of Washington, 346:
1-23.
KELLOGG, R., 1965. Fossil marine mammals from the
Miocene Calvert Formation of Maryland and Virginia. The
Miocene Calvert sperm whale Orycterocetus. Bulletin of the
United States National Museum, 247 (2): 47-63.
KIMURA, T., HASEGAWA, Y. & BARNES, L.G., 2006. Fossil
sperm whales (Cetacea, Physeteridae) from Gunma and
Ibaraki prefectures, Japan; with observations on the Miocene
fossil sperm whale Scaldicetus shigenis Hirota and Barnes,
1995. Bulletin of the Gunma Museum of Natural History, 10:
1-23.
KOMPANJE, E.J.O., 1999. Considerations on the comparative
pathology of the vertebrae in Mysticeti and Odontoceti;
evidence for the occurrence of discarthrosis, zygarthrosis,
infectious spondylitis and spondyloarthritis. Zoologische
Mededelingen, 73: 1-11.
KORETSKY, I.A. & RAY, C.E., 2008. Phocidae of the Pliocene
of eastern USA. Virginia Museum of Natural History Special
Publication, 14: 81-139
LAMBERT, O., 2005. Systematics and phylogeny of the fossil
beaked whales Ziphirostrum du Bus, 1868 and Choneziphius
Duvernoy, 1851 (Cetacea, Odontoceti), from the Neogene of
Antwerp (North of Belgium). Geodiversitas, 27: 443-497.
LAMBERT, O. & LOUWYE, S., 2006. Archaeoziphius
microglenoideus, a new primitive beaked whale (Mammalia,
Cetacea, Odontoceti) from the Middle Miocene of Belgium.
Journal of Vertebrate Paleontology, 26 (1): 182-191.
LEIDY, J., 1853. Observations on extinct Cetacea. Proceedings
of the Academy of Natural Sciences of Philadelphia, 6:
377– 378.
Miocene sperm whales from the North Sea
314
LOUWYE, S., 2001. New species of dinoagellates cysts
from the Berchem Formation, Miocene, northern Belgium
(southern North Sea Basin). Geobios, 34 (2): 121-130.
LOUWYE, S., 2005. The Early and Middle Miocene
transgression at the southern border of the North Sea Basin
(northern Belgium). Geological Journal, 40: 441-456.
LOUWYE, S., DE CONINCK, J. & VERNIERS, J., 1999.
Dinoagellate cyst stratigraphy and depositional history of
Miocene and lower Pliocene formations of northern Belgium
(southern North Sea Basin). Geologie en Mijnbouw, 78: 31-
46.
LOUWYE, S., DE CONINCK, J. & VERNIERS, J., 2000.
Shallow marine Lower and Middle Miocene deposits at the
southern margin of the North Sea Basin (northern Belgium):
dinoagellate cyst biostratigraphy and depositional history.
Geological Magazine, 137 (4): 381-394.
LYDEKKER, R., 1893. Contribution to the knowledge of the
fossil vertebrates of Argentina. Part II. Cetacean skulls from
Patagonia. Anales del Museo de La Plata, 1893: 1-14.
MADSEN, P.T., PAYNE, R., KRISTIANSEN, N.U., WAHLBERG,
M., KERR, I. & MØHL, B., 2002. Sperm whale sound
production studied with ultrasound time/depth-recording
tags. The Journal of Experimental Biology, 205: 1899-1906.
MCHEDLIDZE, G.A., 1970. Some general characteristics of
the evolution of cetaceans, part 1, 112 pp. Akademia Nauk
Gruzinskoi S.S.R., Institut Paleobiologii, Tbilisi.
MEAD, J.G., 1975. Anatomy of the external nasal passages
and facial complex in the Delphinidae (Mammalia: Cetacea).
Smithsonian Contributions to Zoology, 207: 1-72.
MORENO, F.P., 1892. Noticias sobre algunos cetáceos fósiles
y actuales de la República Argentina. Revista del Museo de
La Plata, 3: 383-400.
MOURLON, M., 1878. Sur le gisement du Cachalot nain
(Physeterula dubusii, Van Beneden). Bulletin de l’Académie
Royale des Sciences, des Lettres et des Beaux-Arts de
Belgique, 45: 178-182.
MUIZON, C. DE, 1984. Les Vertébrés de la Formation Pisco
(Pérou). Deuxième partie: Les Odontocètes (Cetacea,
Mammalia) du Pliocène inférieur du Sud-Sacaco. Travaux de
l’Institut Français d’Etudes Andines 27: 1-188.
MUIZON, C. DE, 1988. Les Vertébrés fossiles de la Formation
Pisco (Pérou). Troisième partie: Les Odontocètes (Cetacea,
Mammalia) du Miocène. Travaux de l’Institut Français
d’Etudes Andines, 42: 1-244.
MUIZON, C. DE, 1991. A new Ziphiidae (Cetacea) from the
Early Miocene of Washington State (USA) and phylogenetic
analysis of the major groups of odontocetes. Bulletin du
Muséum National d’Histoire Naturelle, Paris (4) 12: 279-
326.
MULDER, E.W.A., 2001. Co-ossied vertebrae of mosasaurs
and cetaceans: implications for the mode of locomotion of
extinct marine reptiles. Paleobiology, 27 (4): 724-734.
PARONA, C.F., 1930. Resti di un ‘»Physeter» scoperti
nel Pliocene (astiano) di Vigliano d’Asti, in frazione Val
Montasca. Atti della Reale Accademia delle Scienze di
Torino, 65: 240-245.
PETERS, A.M.M. & MONTEIRO, H.J., 2005. A small sperm
whale (Cetacea: Odontoceti, Physeteridae) from the Miocene
of Antwerp. Deinsea, 11: 87-101.
RICE, D.W., 1989. Sperm whale Physeter macrocephalus
Linnaeus, 1758. In: Ridgway, S. H. & Harrison, R. (Editors),
Handbook of Marine Mammals, vol. 4: River dolphins and
the larger toothed whales, pp. 177-233. Academic Press,
London.
ROSS, G.J.B., 1984. The smaller cetaceans of the south east
coast of southern Africa. Annals of the Cape Provincial
Museums of Natural History, 15 (2): 173-410.
SCHENKKAN, E.J. & PURVES, P.E., 1973. The comparative
anatomy of the nasal tract and the function of the spermaceti
organ in the Physeteridae (Mammalia, Odontoceti). Bijdragen
tot de Dierkunde, 43 (1): 93-112.
TROUESSART, E., 1898. Catalogus mammalium tam viventium
quam fossilium, 5: 665-1264.
TRUE, F.W., 1908. The fossil cetacean, Dorudon serratus
Gibbes. Bulletin of the Museum of Comparative Zoology, 52
(4): 5-78.
VAN BENEDEN, P.-J., 1859. Rapport sur la découverte
d’ossements fossiles, faite à Saint-Nicolas. Bulletin de
l’Académie Royale des sciences de Belgique, 8: 123-146.
VAN BENEDEN, P.-J., 1861. La côte d’Ostende et les fouilles
d’Anvers. Bulletin de l’Académie Royale des sciences, des
Lettres et des Beaux-Arts de Belgique, 12: 453-483.
VAN BENEDEN, P.-J., 1869. Sur un nouveau genre de ziphioide
fossile (Placoziphius), trouvé à Edeghem, près d’Anvers.
Mémoires de l’Académie Royale des Sciences, des Lettres et
des Beaux-Arts de Belgique, 37: 1-12.
VAN BENEDEN, P.-J., 1877. Note sur un Cachalot nain du
crag d’Anvers (Physeterula dubusii). Bulletin de l’Académie
Royale des Sciences, des Lettres et des Beaux-Arts de
Belgique, 44: 851-856.
VAN BENEDEN, P. J. & GERVAIS, P., 1880. Ostéographie des
cétacés vivants et fossiles, 634 pp. Arthus Bertrand, Paris.
VANDEN BROECK, E., 1874. Esquisse géologique et
paléontologique des dépôts pliocènes des environs d’Anvers,
première partie. Annales de la Société Malacologique de
Belgique 9 (2): 87-374.
VERTEUIL, L. DE & NORRIS, G., 1996. Miocene dinoagellate
stratigraphy and systematics of Maryland and Virginia.
Micropaleontology, suppl. 42: 1-172.
WERTH, A.J., 2004. Functional morphology of the sperm
whale tongue, with reference to suction feeding. Aquatic
Olivier LAMBERT
315
Mammals, 30: 405-418.
WERTH, A. J., 2006. Mandibular and dental variation and
the evolution of suction feeding in Odontoceti. Journal of
Mammalogy, 87 (3): 579-588.
WHITEHEAD, H., 2003. Sperm whales: social evolution in the
ocean, 431 pp. The University of Chicago Press, Chicago.
WHITMORE, F.C. & KALTENBACH, J.A., 2008. Neogene
Cetacea of the Lee Creek Phosphate Mine, North Carolina.
Virginia Museum of Natural History Special Publication, 14:
181-269.
Zygorhiza 0000- 00--0 02000 00000 00002 -0--- --0000
Agorophius ?01?0 00000 00?00 ????0 ?0?00 000-0 000?00
Eudelphis 1110? 1?01? ????1 ????0 00?0? ?100? ??1100
Zygophyseter 21b01 11111 0?101 22?10 10100 2121? 000110
Brygmophyseter 211?1 1111? 01?01 ????0 101?? ?120? ?01?10
Diaphorocetus 011?1 1111? 0?111 ????? 00??1 21?0? 001??0
Placoziphius 01101 11111 0???1 ????? ????1 1??0? 001210
Orycterocetus 01101 11111 02?21 22011 00?00 11101 001210
Physeterula 211?1 111?? 02??1 ????1 00??2 ?1200 ?0?210
Aulophyseter 21102 11111 02121 220?1 01??2 21200 a11?10
Physeter 22102 11111 02121 22011 01a11 21300 011210
IRSNB M.1937 ?11?1 ??01? ?2?2? ????? ????2 1120? 01????
Thalassocetus 3?1?1 10??2 11?21 ????? ????? ??101 ?01?20
Scaphokogia 31211 10102 1?1?? 221?? ?1?10 2100? 11????
Praekogia ?12?1 10??2 11?01 ????? ???1? ??011 1?1201
Kogia 31212 10102 1b121 22111 01a11 b1011 101201
Appendix 1
Data-matrix of 36 characters for two
outgoups Zygorhiza and Agorophius,
and 14 analysed taxa. All characters
with multiple states are treated as
unordered. 0 primitive state; 1, 2, 3
derived states; a variable between
0 and 1; b variable between 1 and
2; ? missing character; - irrelevant
character.
Miocene sperm whales from the North Sea
Olivier LAMBERT
Département de Paléontologie
Institut royal des Sciences naturelles de Belgique
Rue Vautier 29, B-1000 Bruxelles, Belgique
e-mail: Olivier.Lambert@naturalsciences.be
Typescript submitted: April 8, 2008
Revised typescript accepted: May 27, 2008
Appendix 2
List of characters used in the cladistic analysis. The 23
rst characters are taken from BIANUCCI & LANDINI, 2006
(modied where noted).
1. Size of skull (expressed as condylobasal length) (BIANUCCI
& LANDINI, 2006): 0, 50-90 cm; 1, 90-120 cm; 2, >120
cm; 3, <50 cm.
2. Supracranial basin of the skull (BIANUCCI & LANDINI,
2006): 0, absent; 1, present; 2, extended onto the whole
dorsal surface of the rostrum.
3. Antorbital notch (modied from BIANUCCI & LANDINI,
2006): 0, absent; 1, present; 2, transformed into a very
narrow slit.
4. Maxillae, premaxillae and vomer, all reaching the tip of
the rostrum which is not formed only by the premaxillae
(BIANUCCI & LANDINI, 2006): 0, no; 1, yes.
5. Frontal-maxilla suture, with skull in lateral view (modied
from BIANUCCI & LANDINI, 2006): 0, forming an angle
< 15° from the axis of the rostrum; 1, 15-35°; 2, > 35°.
It should be noted that this character varies with age in
Physeter (see KELLOGG, 1925).
6. Right premaxilla (BIANUCCI & LANDINI, 2006): 0,
5 10 15 20 25 30 36
316 Olivier LAMBERT
posteriorly extended as the left premaxilla; 1, more
posteriorly extended than the left premaxilla.
7. Right premaxilla (modied from BIANUCCI & LANDINI,
2006): 0, not widened posteriorly; 1, posterior extremity
of the right premaxilla laterally widened, occupying at
least one third of the width of the supracranial basin.
8. Left premaxillary foramen very small or absent (BIANUCCI
& LANDINI, 2006): 0, no; 1, yes.
9. Increase in size of the right premaxillary foramen (BIA-
NUCCI & LANDINI, 2006): 0, no; 1, yes.
10. Lack of nasals (BIANUCCI & LANDINI, 2006): 0, both
nasals present; 1, one nasal absent; 2, both nasals
absent.
11. Presence of a sagittal crest (BIANUCCI & LANDINI, 2006):
0, absent; 1, present as a shelf covered by the pointed
right premaxilla.
12. Occipital shield (BIANUCCI & LANDINI, 2007): 0, convex
and forming an angle of about 40° from the axis of the
rostrum; 1, as state 0 with an angle of about 60°; 2, at
or concave forming an angle of about 90°. Most of the
dorsal elements of the supracranial basin are damaged
in Diaphorocetus poucheti; therefore the supraoccipital
shield might have been originally more erected. For
Aulophyseter morricei, a second specimen displays a
supraoccipital shield more elevated than in the holotype;
the latter might have undergone some abrasion (KIMURA
et al., 2006).
13. Fusion of lacrimal and jugal (BIANUCCI & LANDINI,
2006): 0, no; 1, yes.
14. Temporal fossa (BIANUCCI & LANDINI, 2006): 0,
anteroposteriorly elongated (width/height > 1); 1, not
anteroposteriorly elongated (width/height = 1); 2,
anteroposteriorly compressed (width/height < 1).
15. Zygomatic process of squamosal in lateral view
(BIANUCCI & LANDINI, 2006): 0, ‘L’-shaped with dorsal
margin ventrally bending in its posterior portion; 1,
triangular, with dorsal margin dorsally bending in its
posterior portion.
16. Anterior bullar facet of the periotic (BIANUCCI & LANDINI,
2006): 0, very anteroposteriorly elongated; 1, reduced; 2,
absent or very small.
17. Accessory ossicle of the tympanic bulla (BIANUCCI &
LANDINI, 2006): 0, absent; 1, present; 2, present and
partially fused with the anterior process.
18. Posterior extension of the posterior process of the periotic
parallel to the general plane of the bone and not ventrally
orientated (BIANUCCI & LANDINI, 2006): 0, no; 1, yes.
19. Involucrum of the tympanic bulla with an evident central
concavity, visible in ventral and medial views, due to the
marked pachyostosis of its anterior and posterior portion
(BIANUCCI & LANDINI, 2006): 0, no; 1, yes.
20. Loss of dental enamel (BIANUCCI AND LANDINI, 2006): 0,
no; 1, yes. KIMURA et al. (2006) suggest that enamel is
absent in Aulophyseter morricei, contradicting KELLOGG
(1927).
21. Size of teeth (greatest diameter of root expressed
as percentage of the condylobasal length of skull)
(BIANUCCI & LANDINI, 2006): 0, < 3%; 1, > 3%.
Considering the strong heterodonty in Zygorhiza this
character is restricted to single-rooted teeth.
22. Upper tooth row (BIANUCCI & LANDINI, 2006): 0, deep
alveoli; 1 alveoli shallow or absent.
23. Ventral position of the mandibular condyle (BIANUCCI &
LANDINI, 2006): 0, no; 1, yes.
24. Long axis of the skull (CRANFORD, 1999): 0, roughly
parallel to the long axis of the body (perpendicular to the
surface of the occipital condyles); 1, projected ventrally.
25. Dorsal exposure of the maxilla on the rostrum (modied
from KAZÁR, 2002): 0, exposed on most of the length of
the rostrum, narrower than the premaxilla at some levels;
1, wider than the premaxilla all along; 2, exposure
limited to less than half the rostrum length.
26. Anteroposterior level of right premaxillary foramen:
0, distinctly anterior to antorbital notch; 1, slightly
anterior to antorbital notch; 2, same level or posterior to
antorbital notch.
27. Asymmetry of the bony nares: 0, absent or reduced; 1,
strong, left bony naris signicantly larger than right
naris.
28. Number and size of dorsal infraorbital foramina, in the
area of the right antorbital notch and posteriorly: 0,
small to moderate size foramina, at least three-four; 1,
three large foramina; 2, two large foramina; 3, one large
foramen (maxillary incisure).
29. Transverse level of right maxillary crest laterally limiting
the supracranial basin: 0, median to the antorbital notch;
1, at the level of the antorbital notch or lateral.
30. Right maxilla reaching or crossing the sagittal plane of
the skull on the posterior wall of the supracranial basin:
0, no; 1, yes.
31. Projection of lacrimal-jugal between frontal and maxilla
on preorbital process (seen in lateral view): 0, short or
absent; 1, long.
32. Preorbital process considerably lower than the elevated
dorsolateral margin of the rostrum base: 0, no; 1, yes.
33. Zygomatic process of the squamosal: 0, elongated, length
more than two times higher than height at mid-length;
1, more robust and/or shorter, length no more than two
times higher than height at mid-length.
34. Falciform process of the squamosal: 0, contacting the
corresponding pterygoid; 1, forming a thin isolated plate;
2, reduced to a simple peg or absent.
35. Postglenoid process of the squamosal: 0, signicantly
ventrally longer than post-tympanic process; 1, roughly
same ventral extent as post-tympanic process.
36. In lateral view of the skull, wide notch posterior to the
postglenoid process of the squamosal for the enlarged
posterior process of the tympanic: 0, no; 1, yes.
