taxonID	type	description	language	source
03E387DFDB5CAA35B13410DF192AF8FD.taxon	type_taxon	TYPE SPECIES. — Thalassocetus antwerpiensis Abel, 1905.	en	Alfsen, Apolline, Bosselaers, Mark, Lambert, Olivier (2021): New sperm whale remains from the late Miocene of the North Sea and a revised family attribution for the small crown physeteroid Thalassocetus Abel, 1905. Comptes Rendus Palevol 20 (39): 807-822, DOI: 10.5852/cr-palevol2021v20a39, URL: https://doi.org/10.5852/cr-palevol2021v20a39
03E387DFDB5CAA35B13410DF192AF8FD.taxon	diagnosis	DIAGNOSIS. — Same as for the sole currently described species T. antwerpiensis.	en	Alfsen, Apolline, Bosselaers, Mark, Lambert, Olivier (2021): New sperm whale remains from the late Miocene of the North Sea and a revised family attribution for the small crown physeteroid Thalassocetus Abel, 1905. Comptes Rendus Palevol 20 (39): 807-822, DOI: 10.5852/cr-palevol2021v20a39, URL: https://doi.org/10.5852/cr-palevol2021v20a39
03E387DFDB5CAA34B0FB11DC1A59FE7D.taxon	materials_examined	LECTOTYPE. — IRSNB M. 525, a fragmentary cranium including part of the supracranial basin, the right orbit, and the right part of the basicranium. TYPE HORIZON AND AGE. — An origin in the Berchem Formation, possibly the Antwerpen Sands Member, dated from the late early to middle Miocene (late Burdigalian to Langhian; Louwye 2005; Louwye et al. 2010), has been tentatively proposed by Lambert (2008), based on color and preservation state. TYPE LOCALITY. — Antwerp area (Fig. 1). No precise locality recorded.	en	Alfsen, Apolline, Bosselaers, Mark, Lambert, Olivier (2021): New sperm whale remains from the late Miocene of the North Sea and a revised family attribution for the small crown physeteroid Thalassocetus Abel, 1905. Comptes Rendus Palevol 20 (39): 807-822, DOI: 10.5852/cr-palevol2021v20a39, URL: https://doi.org/10.5852/cr-palevol2021v20a39
03E387DFDB5CAA34B0FB11DC1A59FE7D.taxon	diagnosis	EMENDED DIAGNOSIS. — This small size physeterid (postorbital width lower than 300 mm in the juvenile to subadult lectotype, within the range of adults of Kogia spp.) can be distinguished from all other physeteroids (stem physeteroids, physeterids, and kogiids) by the following unique combination of cranial morphological features: antorbital notch being located outside the proportionally narrow supracranial basin (differing from most kogiids); at least one right dorsal infraorbital foramen being located inside the supracranial basin; retention of the left nasal along the left posterolateral wall of the supracranial basin (differing from kogiids for which the postnarial region is known); left nasal being posteriorly pointed, with the apex nearly reaching the nuchal crest and being close to the sagittal plane of the cranium (differing, among others, from Orycterocetus crocodilinus Cope, 1868); absence of a sagittal facial crest in the supracranial basin (differing from most kogiids); abrupt dorsal elevation of the posterior part of the maxilla towards the nuchal crest; short, triangular zygomatic process of the squamosal (ratio between distance from anterior tip of zygomatic process to exoccipital and postorbital width lower than 0.25); nodular, somewhat anteroposteriorly thickened postglenoid process of the squamosal; anteroposteriorly short posttympanic process of the squamosal, lacking a broad notch for the posterior process of the tympanic (differing from most kogiids); and anterodorsal portion of the surface of the occipital shield being dorsoventrally concave and transversely convex.	en	Alfsen, Apolline, Bosselaers, Mark, Lambert, Olivier (2021): New sperm whale remains from the late Miocene of the North Sea and a revised family attribution for the small crown physeteroid Thalassocetus Abel, 1905. Comptes Rendus Palevol 20 (39): 807-822, DOI: 10.5852/cr-palevol2021v20a39, URL: https://doi.org/10.5852/cr-palevol2021v20a39
03E387DFDB5DAA3CB17A145B19F2FA0C.taxon	materials_examined	REFERRED SPECIMEN. — IRSNB M. 2329, a fragmentary cranium including the rostrum base, most of the facial region, and the right part of the basicranium. HORIZON AND AGE. — IRSNB M. 2329 originates from a yet unnamed lithological unit (layer V ofHoedemakers & Dufraing 2015; but see Goolaerts et al. 2020 for further details) made of fine, bluegrey glauconitic sand, that is currently interpreted as being intercalated between the upper layers of the Antwerpen Sands Member (Berchem Formation) and the base of the Deurne Sands Member (Diest Formation). No biostratigraphic analysis of this unit has been published yet, but the upper age limit of the Antwerpen Sands and the lower age limit of the Diest Formation may constrain it to an interval ranging from about 11.3 to 9 Ma (early Tortonian, earliest late Miocene; Louwye 2005; Louwye et al. 2007). It is for now not possible to completely exclude the possibility that this unit corresponds to an unknown lower part of the Diest Formation (as later confirmed in Goolaerts et al. 2020; in this case its upper age limit may fall in the interval ranging from 9 to 7.5 Ma (late Tortonian; Louwye et al. 2007). This unit matches well the sedimentological and palaeontological features of a coarse grey-green glauconitic sand level temporarily exposed during construction work at the hospital AZ Monica, campus Deurne, located 2.3 km north to the R 11 tunnel site (Bosselaers et al. 2004; level f; M. B., pers. obs.; Fig. 1). Tentatively interpreted in that earlier work as corresponding to a lower portion of the Deurne Sands Member, this level yielded fossil cetacean remains including the articulated skeleton of a large cetotheriid (identified as Plesiocetus sp. in Bosselaers et al. 2004) and the cranium of a ziphiid (identified as Ziphirostrum marginatum in Lambert 2005).	en	Alfsen, Apolline, Bosselaers, Mark, Lambert, Olivier (2021): New sperm whale remains from the late Miocene of the North Sea and a revised family attribution for the small crown physeteroid Thalassocetus Abel, 1905. Comptes Rendus Palevol 20 (39): 807-822, DOI: 10.5852/cr-palevol2021v20a39, URL: https://doi.org/10.5852/cr-palevol2021v20a39
03E387DFDB5DAA3CB17A145B19F2FA0C.taxon	description	LOCALITY. — IRSNB M. 2329 was discovered in 2014 by Leo Dufraing during excavations for a tunnel on the road R 11 between Mortsel and Borsbeek, along the Antwerp International airport (Fig. 1). Geographic coordinates: 51 ° 11 ’ 08 ” N, 4 ° 28 ’ 18 ” E.	en	Alfsen, Apolline, Bosselaers, Mark, Lambert, Olivier (2021): New sperm whale remains from the late Miocene of the North Sea and a revised family attribution for the small crown physeteroid Thalassocetus Abel, 1905. Comptes Rendus Palevol 20 (39): 807-822, DOI: 10.5852/cr-palevol2021v20a39, URL: https://doi.org/10.5852/cr-palevol2021v20a39
03E387DFDB5DAA3CB17A145B19F2FA0C.taxon	discussion	COMMENTS This specimen shares all the diagnostic features of Thalassocetus antwerpiensis as listed above. However, because of: 1) a series of minor morphological differences (see below; width of right maxilla between antorbital notch and largest dorsal infraorbital foramen, aspect of anterodorsal surface of left nasal, shape of lateral surface of postorbital process of frontal, and degree of concavity of posterior surface of exoccipital); 2) the fragmentary state of the lectotype of Thalassocetus antwerpiensis; and 3) a possibly older geological age for the latter, we choose to provisionally keep an open taxonomic attribution for IRSNB M. 2329. It may either belong to the species T. antwerpiensis, or to a new, closely related taxon. The second option would mean that the diagnosis proposed above would apply to the genus Thalassocetus, whereas a new diagnosis would have to be proposed for each species in that genus.	en	Alfsen, Apolline, Bosselaers, Mark, Lambert, Olivier (2021): New sperm whale remains from the late Miocene of the North Sea and a revised family attribution for the small crown physeteroid Thalassocetus Abel, 1905. Comptes Rendus Palevol 20 (39): 807-822, DOI: 10.5852/cr-palevol2021v20a39, URL: https://doi.org/10.5852/cr-palevol2021v20a39
03E387DFDB5DAA3CB17A145B19F2FA0C.taxon	description	DESCRIPTION OF THALASSOCETUS SP. IRSNB M. 2329 General morphology and ontogenetic stage From the supraoccipital to the anterior edge of the truncated vomer the cranium has a preserved length of 314 mm and a postorbital width of 281 mm. The small size of the cranium and the unfused sutures between all the bones suggest that this specimen was a juvenile. It is relatively small for a physeteroid, close to cranial dimensions of extant dwarf and pygmy sperm whales (Kogia spp.), in which adults can reach a cranium width of 245 mm for Kogia sima (Owen, 1868) and 378 mm for Kogia breviceps (Blainville, 1838) (Ross 1984). However, based on the preserved parts we can assume that the rostrum length of this specimen was greater than in Kogia spp. Also, the supracranial basin is markedly smaller than in the latter: the crests that laterally delimit the basin have a more medial position, not reaching the antorbital region and not including the right and left antorbital notches inside the basin (Fig. 2), as opposed to Kogia spp. (e. g. Velez-Juarbe et al. 2015). The supracranial basin is delimited on the right side by the right maxilla, on the posterior side by the right premaxilla, and on the left side by the left maxilla (posteriorly) and left premaxilla (anteriorly). With a maximum width of 141 mm between the right and left margins (approximately equivalent to half the postorbital width), the basin is about as wide as long, not extending anteriorly beyond the right premaxillary foramen, its anterior boundary being rather defined by a dorsomedial elevation of the right premaxilla and vomer (see below). The floor of the supracranial basin is mainly made by the right premaxilla. Between the highly asymmetrical bony nares (the left naris being much larger than the right one), the presphenoid closes posteriorly the broad mesorostral groove. In the reconstructed lateral view (Fig. 3), the temporal fossa appears as slightly higher than anteroposteriorly long, in a way similar to Orycterocetus. Premaxilla The strong asymmetry of this cranium is especially expressed in the organization of the premaxillae in the supracranial basin. Only the posterior part of the left premaxilla is preserved, displaying strong similarities with Orycterocetus crocodilinus (Kellogg 1965). The premaxilla forms the anterior part of the left lateral wall of the supracranial basin, where it wedges into a groove of the maxilla as in Orycterocetus. On the other hand, the right premaxilla has a much larger posterior extent, as a roughly transversely flat and anteroposteriorly concave broad plate that reaches the posterodorsal margin of the supracranial basin and the base of both the lateral walls of the basin, in a way similar to Orycterocetus. The surface of the supracranial basin is therefore smooth, lacking any indication of a sagittal facial crest, a feature described in all kogiids (Velez-Juarbe et al. 2015; Collareta et al. 2017 a). In lateral view, from the anterior edge of the right bony naris to the posterodorsal edge of the basin, the right premaxilla raises posterodorsally with an approximate angle of 110 ° with respect to the long axis of the rostrum. Only the right premaxilla is preserved anterior to the antorbital notch. Transversely narrower than the right maxilla, it is pierced by a large premaxillary foramen that is located about 30 mm anterior to the level of the antorbital notch, anterior to the anteriormost right dorsal infraorbital foramen (as in Orycterocetus). Maxilla Lateral to the crests defining the supracranial basin, both maxillae become gradually dorsoventrally thinner towards the lateral edges of the cranium. They are asymmetrical at this level: this lateral part is slightly wider on the left maxilla (62 mm vs 56.5 mm on the right side, posterior to the posteriormost dorsal infraorbital foramina) and the dorsal surface is more transversely concave on the right side. The right and left maxillae almost contact each other along the posterior wall of the supracranial basin, behind the right premaxilla. The right maxilla displays four dorsal infraorbital foramina. The small, posteriormost of these foramina is located just lateral to the maxillary crest defining the supracranial basin and it is followed posteriorly by a long groove along the basin’s margin. The second foramen is larger and located on the wall of the basin, just posterior to the level of the corresponding antorbital notch. Placed at the level of the antorbital notch, the third foramen is the largest. The narrower fourth foramen is located anterior to the notch and preceded anteriorly by two grooves at the rostrum base. The preserved portion of the left maxilla exhibits three dorsal infraorbital foramina, all located outside the supracranial basin. The posteriormost of these foramina is also followed posteriorly by a groove, which is shorter than that on the right side. Just anteromedial, the second foramen is markedly smaller, whereas the third is only partly preserved, but was originally the largest, located at the level of the antorbital notch. Lateral to the right dorsal infraorbital foramina a series of small depressions / fossae likely indicate insertions of facial muscles (red dotted lines in Figure 3). The short, broadly open antorbital notch is followed posteriorly by a short sulcus; it differs markedly from the “ slit-like ” notch of Kogia and several extinct kogiids (Koristocetus Collareta, Lambert, Muizon, Urbina & Bianucci, 2017, Nanokogia Velez-Juarbe, Wood, Gracia & Hendy, 2015, Pliokogia Collareta, Cigala Fulgosi & Bianucci, 2019, and Scaphokogiinae Muizon, 1988), and it is more similar to the notch seen, for example, in Orycterocetus and Physeter. In lateral view, the maxillae get gradually thinner along their oblique posterodorsal ascent from the antorbital notch to the nuchal crest, where their long axis almost reaches a vertical orientation. In ventral view, the maxillae are only partly preserved along the rostrum base, where they cover the vomer ventrally (Fig. 4). Anteriorly, the maxillae are too damaged for the presence of an alveolar groove and of alveoli for functional teeth to be assessed. Other bones of the palate (palatines and pterygoids) are not preserved. Vomer At the rostrum base, the thick vomer makes the ventral and lateral walls of a broad, U-shaped mesorostral groove. In dorsal view the raised medial edge of the right premaxilla covers the lateral wall of the groove. Bones that originally covered the vomer ventrally are missing but a medial crest is present on the latter, suggesting that this ventralmost part of the vomer was exposed ventrally between the right and the left maxillae (as often occurs in odontocetes). Frontal In dorsal view, frontal bones are only visible at the lateral edge of the neurocranium, in the supraorbital area and anterior to the nuchal crest. At the level of the nuchal crest the right and left frontals contact each other along the sagittal plane. Only the right frontal has the preorbital and postorbital processes preserved. Moderately thickened dorsoventrally, the short preorbital process is separated from the partly preserved antorbital process of the maxilla by a narrow notch, at least 12 mm deep. The latter was originally occupied by the lacrimal, which is missing on this specimen. Somewhat laterally and ventrally truncated, the postorbital process is directed ventrally. Posterior to this process the frontal gets dorsoventrally thinner while raising posterodorsally above the temporal fossa. In lateral view of the supraorbital region the maxilla-frontal suture draws an angle of about 30 ° with the horizontal plane of the rostrum. In the posterodorsal quarter of the frontal, where the latter is sandwiched between maxilla and supraoccipital, this suture reaches an almost vertical orientation. In ventral view, the frontals outline the cerebral cavity anterodorsally. This cavity has a cordiform shape, with a maximum width of 177 mm and a length of 124 mm. Ventral to the right frontal a part of the parietal bone is preserved (Fig. 4). Nasal This specimen has two bony nares but only one nasal (the left one), as reported for part, but not all non-kogiid physeteroids (two nasals are recorded in part of the stem physeteroids and no nasal is observed in kogiids; Flower 1867 b; Kellogg 1965; Velez-Juarbe et al. 2015; Lambert et al. 2017; Collareta et al. 2017 a). This nasal being fully accessible thanks to the unfused, removable plate-like posterior part of the right premaxilla and posteromedial part of the right maxilla in the supracranial basin (Fig. 5), we could clearly see that this bone is located between the sagittal plane and the left margin of the supracranial basin, posterior to the left bony naris and the presphenoid. Extending to the posterior edge of the basin, this dorsoventrally thin bone thickens slightly medially and, to a greater extent, anterolaterally towards the posterior margin of the left bony naris. The nasal is 118 mm long and 54 mm wide; it contributes to approximately 77 % of the length of the supracranial basin. In dorsal view, with the overlying right maxilla and premaxilla in place, the nasal is only visible in its anterolateral and posterolateral regions, which poke out under the right premaxilla. The thin left edge of the nasal covers the left maxilla along the lateral wall of the supracranial basin, while its right edge is sutured in its anterior half with the right frontal and in its posterior half with the right maxilla. Ventrally the nasal bone mainly rests on the left frontal. Supraoccipital With a partly abraded outer surface, the supraoccipital is mostly preserved on the upper part of the occipital shield. This region is dorsoventrally concave and slightly transversely convex, drawing an angle of approximately 70 ° with respect to the long axis of the rostrum in lateral view. Along with the frontals and maxillae the upper edge of the supraoccipital constitutes a thick nuchal crest whose dorsomedial portion projects posterodorsally. Squamosal Only a part of the right side of the basicranium is preserved, detached from the rest of the cranium. Its position and orientation relative to the main dorsal fragment has been interpreted based on the orientation of the surfaces of the squamosal and parietal on both sides of the fracture zone (Fig. 3). It is mostly comprised of the squamosal and exoccipital, with a small fragment of parietal preserved in the posteroventrolateral corner of the cerebral cavity. In lateral view the zygomatic process has a triangular shape and its apex is anteriorly directed. The zygomatic process is proportionally short, with a distance from the anterior tip to the squamosal exoccipital suture of 66 mm. The supramastoid crest gradually raises posterodorsally towards the temporal crest, as in many other physeteroids (Bianucci & Landini 2006). Medially, the narrow squamosal fossa is anteroposteriorly and transversely concave. The short and slender postglenoid process is an anteroposteriorly thin plate that anteriorly defines a deep and narrow external auditory meatus. Between this meatus and the anteroventral margin of the exoccipital, the posttympanic process is short (maximum anteroposterior length in ventral view equals 15 mm), leaving a space for the posterior process of the tympanic ventral to the posttympanic process that is much more limited than in Kogia spp. and related taxa (e. g. Velez-Juarbe et al. 2015). The posttympanic process was reaching approximately the same dorsoventral level as the postglenoid process. In ventral view, the mandibular fossa is moderately concave dorsoventrally and poorly separated from the tympanosquamosal recess (Fig. 6). The latter displays two oblique, anterolaterally elongated fossae separated by a thick and low crest. The deepest and broadest fossa is the posterolateral one, located along the anterior meatal crest. Most of the falciform process of the squamosal and the alisphenoid are missing. Exoccipital In lateral view the ventralmost region of the preserved basicranium is made by the exoccipital. The preserved part of this bone is lateral to the missing right occipital condyle. Posteriorly and slightly laterally, it covers the squamosal as an anteroposteriorly thin plate with a roughly flat, slightly anteriorly tilted posterior surface. In ventral view the paroccipital process is weakly thickened. Potential bite marks Two deep and broad, subparallel oblique grooves running posterolaterally on the dorsal surface of the right frontal above the orbit (Figs 3; 7) may correspond to healed bite marks, possibly by a large shark (for examples of shark bites on fossil marine mammal bones, seeBianucci et al. 2010; Collareta et al. 2017 b; for putative healing of a bite on cetacean bone, see Kallal et al. 2012). The flanks of these grooves with a V-shaped section are indeed made of compact bone with a smooth surface, contrasting with other, postmortem damage in the frontal, revealing more spongy bone, and suggesting therefore that some post-bite bone repair may have occurred. The anterior groove extends on a short distance on the lateral edge of the maxilla, where it appears as two narrower, parallel grooves (Fig. 7). Ventrally, the preorbital region of the right frontal is also deeply cut by a groove that is similarly directed posterolaterally (Fig. 4); this groove is thus interpreted as resulting from the same biting event (opposite tooth).	en	Alfsen, Apolline, Bosselaers, Mark, Lambert, Olivier (2021): New sperm whale remains from the late Miocene of the North Sea and a revised family attribution for the small crown physeteroid Thalassocetus Abel, 1905. Comptes Rendus Palevol 20 (39): 807-822, DOI: 10.5852/cr-palevol2021v20a39, URL: https://doi.org/10.5852/cr-palevol2021v20a39
03E387DFDB5DAA3CB17A145B19F2FA0C.taxon	discussion	COMPARISON WITH THALASSOCETUS Originating from the same geographic region (southern North Sea Basin) as the lectotype of Thalassocetus antwerpiensis IRSNB M. 525, and possibly from a slightly younger horizon, IRSNB M. 2329 shares with the latter a series of morphological features. First of all, their size is very similar (postorbital width estimated to 280 - 290 mm and 280 mm for T. antwerpiensis and IRSNB M. 2329, respectively; Lambert 2008; this work). In addition, based on the degree of fusion of the cranial sutures (e. g. frontals, maxillae, and premaxillae not fused) they most likely do not correspond to markedly different ontogenetic stages. More specifically, the right lateral side of the small supracranial basin, comprised of the right maxilla, is nearly identical in terms of: 1) the extent of the lateral crest on the maxilla; 2) the size and position of the dorsal infraorbital foramina and their associated grooves; and 3) the shape of the antorbital notch (Fig. 8). In contrast with the interpretation proposed by Lambert (2008: fig. 16, but see fig. 18), the right maxilla of the lectotype of T. antwerpiensis is posteriorly incomplete, but the surface of the underlying frontal indicates a strong elevation of the posteriormost part of the maxilla towards the nuchal crest that is highly similar to the condition observed in IRSNB M. 2329. The short zygomatic process of the squamosal, the supramastoid crest, and the squamosal fossa are also nearly identical in these two specimens, and the same can be said for the frontals and supraoccipital along the nuchal crest. All these strong anatomical similarities point to close relationships between IRSNB M. 525 and IRSNB M. 2329 (see also the phylogenetic analysis below) and allow for a reinterpretation of the bones preserved in the supracranial basin of the lectotype of T. antwerpiensis. The feature that has been interpreted as a sagittal crest in the lectotype of T. antwerpiensis (Bianucci & Landini 2006; Lambert 2008) corresponds very well to the left posterolateral edge of the supracranial basin of IRSNB M. 2329. It actually appears somewhat more prominent in the lectotype of T. antwerpiensis due to the loss of most of the left maxilla outside of the basin. With such a reinterpretation of the sagittal crest as a whole, different sub-parts can be discussed. The region that was identified as a shallow fossa in the right premaxilla of the lectotype of T. antwerpiensis (corresponding to the premaxillary fossa of Barnes 1973) is actually very similar in outline and position to the triangular posterodorsal end of the left nasal of IRSNB M. 2329, whereas the upturned median plate of the left maxilla in the lectotype of T. antwerpiensis matches the orientation of the crest on the left maxilla defining the posterolateral margin of the supracranial basin. Medial to the bone reinterpreted here as the left nasal, the dorsal surface of the right frontal of the lectotype of T. antwerpiensis displays suture marks for the attachment of the right maxilla, in a way similar to IRSNB M. 2329. Because the two specimens differ in a series of minor morphological features (for example the width of the right maxilla between the antorbital notch and the largest dorsal infraorbital foramen, the aspect of the anterodorsal surface of the left nasal, the shape of the lateral surface of the postorbital process of the frontal, and the degree of concavity of the posterior surface of the exoccipital), and considering also their possibly different geological age, as well as their relatively fragmentary state of preservation, they are provisionally kept in separate taxa. That said, the aforementioned new interpretation of the bones of the facial region of the lectotype of T. antwerpiensis leads to a number of fundamental changes in the codings of this taxon in the character-taxon matrix (see below). Phylogeny The main goal of our phylogenetic analysis is to test for the phylogenetic affinities of Thalassocetus, based on the new morphological interpretation of its neurocranium as provided above. A more in-depth investigation of relationships within Physeteroidea and of physeteroids with the other main odontocete clades is beyond the scopes of this work. Taking into account the herein morphological reinterpretation of Thalassocetus antwerpiensis, seven changes were made for the codings of the latter in the character-taxon matrix published by Collareta et al. (2019), namely: 1) character 3 was changed from “? ” to “ 1 & 2 ”, since a supracranial basin was identified in T. antwerpiensis, although it is not possible to exclude the possibility that it might extend onto the whole dorsal surface of the rostrum; 2) since the right premaxilla is no longer identified on the cranium, character 13 was changed from “ 1 ” to “? ”; 3) with the lack of a sagittal crest, character 14 was changed from “ 1 ” to “ 0 ”; 4) the identification of one nasal bone changed character 19 from “ 2 ” to “ 1 ”; 5) the truncated right maxilla could not allow any supposition on its posterior extent, and character 21 was changed from “ 1 ” to “? ”; 6) for character 26 the anteroposterior extent of the temporal fossa was assessed to be longer or approximately the same length as the distance between antorbital process of the maxilla and anterior wall of the temporal fossa (changed from “? ” to “ 1 & 2 ”); and 7) character 30 dealing with the occipital shield was corrected from “ 1 ” to “ 1 & 2 ” because the shield draws an angle between 60 and 90 ° with the reconstructed long axis of the rostrum and it has a flat surface. For the other, unchanged codings of characters see the matrix in the Appendix 2. With these corrections for T. antwerpiensis and the addition of the new specimen IRSNB M. 2329, the heuristic search resulted in 72 most parsimonious trees with 144 steps, a consistency index (CI) of 0.542, and a retention index (RI) of 0.735. The strict consensus tree is shown in Figure 9 with bootstrap support values. It is important to note that, as in previous physeteroid phylogenies, bootstrap values are generally low (often under 50) for most nodes in all major physeteroid clades, including Physeteridae. Only Kogiidae gets a significantly higher value (which is relevant for the discussion of the status of Thalassocetus, our main point of interest, see below). Therefore, relationships as obtained here have to be considered with caution. More complete specimens will allow for the addition of new characters, which may ultimately lead to a stronger support for the main nodes of the physeteroid tree. Differing from the strict consensus tree of Collareta et al. (2019), our analysis recovers Eudelphis du Bus, 1872 as sister-group of a clade including macroraptorial physeteroid genera (i. e., Acrophyseter Lambert, Bianucci & Muizon, 2008, Brygmophyseter Barnes in Kimura, Hasegawa & Barnes, 2006, Livyatan Lambert, Bianucci, Post, Muizon, Salas-Gismondi, Urbina & Reumer, 2010, and Zygophyseter Bianucci & Landini, 2006) and ‘ Aulophyseter’ rionegrensis Gondar, 1974 + the crown Physeteroidea (Fig. 9). Among the latter, the family Physeteridae is rearranged in two clades (see below), whereas relationships within the family Kogiidae are less resolved. Thalassocetus is no longer inside the Kogiidae, being instead found inside the Physeteridae (defined here by three characters: temporal fossa approximately as long as the distance between the antorbital process of the maxilla and the anterior wall of the temporal fossa; number of mandibular teeth> 14, unknown in Thalassocetus; and dorsal process of the periotic anteroposteriorly shorter, but dorsally extended beyond the medial margin of the internal acoustic meatus, unknown in Thalassocetus, reversion), as the sister-group of the new specimen IRSNB M. 2329 (sharing two characters: maximum width of skull <40 cm, reversion, and right premaxillary foramen distinctly anterior to the level of the antorbital notch, reversion), in a clade also including Diaphorocetus Ameghino, 1894, Orycterocetus, and Placoziphius Van Beneden, 1869 (defined by three characters: three large foramina in the area of the right antorbital notch and posteriorly, reversion; right premaxillary foramen slightly anterior to the level of the antorbital notch, reversion; and postorbital process of the frontal much ventrally extended, with a vertical length of the process equal to or greater than the horizontal length of the orbit).	en	Alfsen, Apolline, Bosselaers, Mark, Lambert, Olivier (2021): New sperm whale remains from the late Miocene of the North Sea and a revised family attribution for the small crown physeteroid Thalassocetus Abel, 1905. Comptes Rendus Palevol 20 (39): 807-822, DOI: 10.5852/cr-palevol2021v20a39, URL: https://doi.org/10.5852/cr-palevol2021v20a39
