Beneziphius brevirostris, Lambert O., 2005

Lambert O., 2005, Systematics and phylogeny of the fossil beaked whales Ziphirostrum du Bus, 1868 and Choneziphius Duvernoy, 1851 (Mammalia, Cetacea, Odontoceti), from the Neogene of Antwerp (North of Belgium), Geodiversitas 27 (3), pp. 443-497 : 474-482

publication ID

https://doi.org/ 10.5281/zenodo.5376445

persistent identifier

https://treatment.plazi.org/id/03FED57F-FFF0-9F51-8397-FD1AFC4EFC8C

treatment provided by

Marcus

scientific name

Beneziphius brevirostris
status

sp. nov.

Beneziphius brevirostris n. sp.

HOLOTYPE. — IRSNB ED002 - M.1885, a partial skull including the anterior part of the vertex, the anterior of the cranium, and the rostrum.

ETYMOLOGY. — “ brevi ” from Latin brevis: short, “ rostris ” from Latin rostrum: bill, beak. “ brevirostris ” because of the short and pointed rostrum characterizing this new species.

REFERRED SPECIMEN. — IRSNB 3782-M.1886, a partial skull including the anterior of the cranium and the nearly complete rostrum (syntype of Ziphiopsis phymatodes du Bus, 1868 , and referred to Choneziphius planirostris by Abel [1905]).

TYPE HORIZON. — No data available, Neogene. Judging from the worn aspect of the holotype, it might have been reworked and found in a base gravel (for example the gravel at the base of the Kattendijk Formation, lower Pliocene). The referred specimen IRSNB 3782- M.1886 was found close to the first specimen of Ziphiopsis phymatodes sensu du Bus, 1868 , IRSNB 3781- M.543, probably in the same stratigraphic unit ( du Bus 1868).

TYPE LOCALITY. — Antwerp area, Belgium, exact locality uncertain.

DIAGNOSIS. — Small ziphiid species with a skull small- er than Ziphirostrum marginatum or Choneziphius planirostris , differing from:

Choneziphius in: a more pointed rostrum in lateral and dorsal views; a distinct prenarial basin; a flat surface of the premaxillary sac fossae; a space between the less asymmetrical premaxillary sac fossae (ratio between maximum widths of left and right premaxillae at the level of the fossae of 0.69 on the holotype, Fig. 25 View FIG ); the less abrupt elevation of the premaxillae towards the vertex, not reaching a vertical position; less anteriorly directed transverse premaxillary crests;

Ziphirostrum and Messapicetus in: a shorter and more pointed rostrum; the premaxillae shorter than the maxillae on the apex of the rostrum; joined thickened premaxillae anteriorly limiting the prenarial basin; the subhorizontal surface of the maxilla lateral to the prenarial basin covered by small excrescences;

Aporotus in: fusion of the premaxillae above the mesorostral groove; flat maxilla medially to the antorbital notch;

– all the other known ziphiids in the dorsal roofing of the mesorostral groove by the premaxillae.

DESCRIPTION ( FIGS 22-24 View FIG View FIG View FIG )

The fragmentary skulls both lack the supraorbital processes and the basicranium. The length of the rostrum is estimated at 295 mm for the holotype IRSNB ED002-M.1885, close to the smallest individual of Choneziphius planirostris , and much shorter than in Ziphirostrum marginatum . The second rostrum IRSNB 3782-M.1886 was probably slightly shorter than the holotype. By comparing the width of the rostrum at its base, IRSNB 3782-M.1886 is smaller than Z. marginatum ( Table 4).

Premaxillae

The pachy-osteosclerotic premaxillae dorsally close the mesorostral groove until 60 mm before the preserved apex of the rostrum on the holotype ( Fig. 22 View FIG ). The joined premaxillae are spindle-shaped, with a median maximum width, and strong forwards and backwards tapering. Between the flattened maxillae, the premaxillae are distinctly depressed, as a fossa pierced by a pair of premaxillary foramina. This area is homologous to the prenarial basin present in Ziphirostrum marginatum , but less excavated, with a convex floor only 7-8 mm deep.

In lateral view, the rostrum slopes steeply from 20 mm anterior to the palatine to the apex, with a short dorsal elevation of the premaxillae in front of the prenarial basin ( Fig. 23 View FIG ).

The premaxillary sac fossae are not excavated, in contrast to Choneziphius planirostris and as in Ziphirostrum sp. The right fossa is wider than the left (respective maximum widths of the premaxillae at that level are 56 and 39 mm on the holotype), and there is no median contact between their rounded median margins, similarly to Z. marginatum , with a space between them of 5 mm for the holotype. The ascending process of

The Neogene ziphiids Ziphirostrum and Choneziphius ( Cetacea, Odontoceti )

the premaxilla is nearly vertical, less elevated than in Choneziphius planirostris . The transverse premaxillary crest is closer to Ziphirostrum marginatum , more laterally directed than in Choneziphius planirostris or Ziphius ( Fig. 22C View FIG ).

Maxillae

The extremity of the rostrum is likely formed solely by the maxillae, with shorter premaxillae and a mesorostral groove dorsally open at its apex. However, the bone is too dense, worn, and incomplete to be certain of structure. The progressive posterior widening of the maxilla, with the development of excrescences on its horizontal s u r f a c e, i s c o m p a r a b l e t o t h e c o n d i t i o n i n Choneziphius planirostris , although the bone is relatively shorter. This surface of the maxilla bearing excrescences is anteriorly preceded by a foramen followed by a series of sulci forwards. A second foramen pierces the lateral margin of the surface at the level of the apex of the premaxillary sac fossa. The maxilla is excavated laterally to the premaxillary sac fossa; this excavation, not as deep as in Choneziphius planirostris , is slightly overhung by the premaxilla medially. The transverse compression of the vertical medial plates of the maxillae behind the vertex is pronounced, and the now-lost frontals were narrower than the nasals.

In lateral view, the alveolar groove is shallow with weakly marked alveoli on less than 100 mm. These rounded excavations (less than 1-2 mm deep) have an approximate diameter of 3 mm and interalveolar septa of 2 mm. Fourteen alveoli are counted on the left side of the holotype. A posteriorly directed foramen pierces the maxilla just above the alveolar groove, at the same level in the two skulls, 72 mm before the preserved apex of the rostrum on the right side of the holotype.

Nasal

The nasals form an anteromedian rounded projection overhanging the bony nares, only differing from Ziphirostrum marginatum in more concave anterolateral margins, resembling the condition of Aporotus recurvirostris (see below). The suture between nasals and frontals slightly projects anteromedially.

Vomer-palatine

The features of the palate and the ventral face of the rostrum are similar to Choneziphius planirostris and Ziphirostrum marginatum : large palatines and pairs of foramina along the median maxillary suture. The vomer is ventrally visible between the maxillae for 100 mm, until 30 mm before the apex of the rostrum.

SYSTEMATIC DISCUSSION

The referred specimen IRSNB 3782-M.1886 was attributed by du Bus (1868) to the species Ziphiopsis phymatodes (second individual),

together with the skull IRSNB 3781-M.543 (see discussion below), and by Abel (1905) to Choneziphius planirostris , probably because of the presence of moderately developed excrescences on the platform of the maxillae at the base of the rostrum. Those irregularities on the maxillae were shown above as highly variable in C. planirostris . Other similarities of Beneziphius brevirostris n. gen., n. sp. with that species are: dense, dorsally fused premaxillae, forming a prominence anteromedially to the platform of the maxillae; and excavation of the maxillae along the lateral margins of the premaxillary sac fossae. However, several putative derived characters of Choneziphius planirostris are absent in Beneziphius brevirostris n. gen., n. sp.: the premaxillary sac fossae are unexcavated, less asymmetrical (see Fig. 25 View FIG ) and medially separated; the premaxillary crests are more laterally directed and, in lateral view, the rostrum quickly narrows anteriorly, a feature different from the roughly parallel dorsal and ventral margins of the rostrum on most of its length in C. planirostris . Furthermore, the nasals of Beneziphius brevirostris n. gen., n. sp. are preserved in a way different from the dorsally broken nasals of Choneziphius planirostris . Perhaps, as above, the nasals in C. planirostris were anterodorsally elongated as in Ziphius cavirostris . If this hypothesis is correct, the morphology of the nasals of Beneziphius brevirostris n. gen., n. sp. might also be considered as primitive.

These differences with Choneziphius planirostris are also observed in Ziphirostrum marginatum . Another common character between Z. marginatum and Beneziphius brevirostris n. gen., n. sp. is the presence of a prenarial basin: it is shallower in B. brevirostris n. gen., n. sp., but likely homologous. Indeed, the basin is formed in both taxa by the deepening of the joined posteromedial and posterolateral sulci, and it is laterally margined by thick and wide posterolaterally curving strips of the maxillae. This condition is different from the completely filled space between the maxillae in Choneziphius planirostris (infilled by the premaxillae) and might be used as a synapomorphy of Ziphirostrum marginatum and Beneziphius brevirostris n. gen., n. sp. However, it is also possible that the prenarial basin was secondarily lost in C. planirostris .

REVISION OF ZIPHIOPSIS PHYMATODES DU BUS, 1868 The specimen IRSNB 3781-M.543, a partial rostrum with the left premaxillary sac fossa ( Fig. 26 View FIG ), syntype of Ziphiopsis phymatodes by du Bus (1868, first of two individuals), and figured by Van Beneden & Gervais (1880: pl. 27bis, fig. 1, reversed), was referred to Choneziphius planirostris by Abel (1905), here again probably because of the presence of high excrescences on the dorsal surface of the maxillae at the base of the rostrum. As discussed above, the second individual of Ziphiopsis phymatodes sensu du Bus, 1868 , IRSNB 3782- M.1886, is referred to a new genus and species, Beneziphius brevirostris n. gen., n. sp. du Bus (1868) already noticed several differences at the level of the premaxillae, the excrescences on the maxillae and the size between these two specimens. Indeed, the thickened premaxillae are n a r r o w e r o n t h e r o s t r u m o f I R S N B 3 7 8 1- M.543, and the premaxillae diverge for more than 40 mm before the deeper prenarial basin. The excrescences on the maxilla are much developed (some of them are longer than 8 mm, see Fig. 26A View FIG ) and occupy a much wider medially sloping maxillary surface. IRSNB 3781-M.543 is also larger than IRSNB 3782-M.1886. Even if the size of the excrescences was demonstrated to be variable in Choneziphius planirostris , and in spite of the similar kind of abrupt anterior narrowing of the rostrum, this list of differences indicates that IRSNB 3781-M.543 is not referable to Beneziphius brevirostris n. gen., n. sp. Conversely, this skull lacks important apomorphies of Choneziphius : the mesorostral groove is still posteriorly open; the premaxillary sac fossae are unexcavated and probably separated; in lateral view, the rostrum is more quickly anteriorly lowering.

The morphology of the maxillae and the premaxillae around the prenarial basin is comparable to the condition in Ziphirostrum turniense , specially the holotype IRSNB 3785-M.539. The rostrum of Z. turniense is however much longer (550 mm for IRSNB 3785-M.539, and more or less 400 mm for IRSNB 3781-M.543) and less pointed.

The specimen IRSNB 3781-M.543, identified here as the holotype of Ziphiopsis phymatodes du Bus, 1868 , lacks sufficient features to diagnose the species, which is thus a nomen dubium. The holotype is therefore classified Ziphiidae incertae sedis. Similarities with Ziphirostrum turniense and Beneziphius brevirostris n. gen., n. sp. were demonstrated, but the absence of information about the vertex precludes a better systematic resolution.

Therefore, if considered separately, the character “presence of excrescences on the dorsal face of the maxillae at the base of the rostrum” appears to be poorly diagnostic at the species level, being present in three different taxa and highly variable in Choneziphius planirostris – even bilaterally on one specimen.

REVISION OF ZIPHIOPSIS SERVATUS DU BUS, 1868 The rostrum IRSNB 3806-M.540, holotype and only specimen of the species Ziphiopsis servatus sensu du Bus, 1868 , was figured by Van Beneden & Gervais (1880: pl. 27bis, fig. 8), and later referred by Abel (1905) to Mioziphius belgicus .

This incomplete rostrum is 450 mm long, but some centimetres are probably missing posteriorly ( Fig. 27 View FIG D-F). At the base of the rostrum, the lateral margin of the dorsal surface is elevated relatively to a shallow longitudinal depression, mainly marking the premaxilla. This concave area is medially followed by a slope towards a median elevation of the premaxillae, partly missing on the specimen but originally closing the mesorostral groove. Therefore, no prenarial basin could be present. Anteriorly, the margins of the broken area diverge, reaching a maximum width of 48 mm 270 mm before the apex. This morphology is probably related to a median crest of the joined premaxillae, which, because of its prominence, could have been easily broken.

Such a structure is observed on a large and robust rostrum found recently in Antwerp (no precise locality), NMB 002 ( Fig. 27 View FIG A-C). This specimen (preserved length 700 mm), includes most of the rostrum and the anterior portion of the cranium. The anterior 290 mm of the rostrum are marked by a high longitudinal median premaxillary crest, narrowing and slightly increasing in height posteriorly. The maximum height of the rostrum is near the posterior limit of the crest (120 mm high and 80 mm wide). The posterior margin of the crest is abrupt, posteriorly extended by a lower and narrower crest, nearly reaching the premaxillary sac fossae. This lower crest is margined by a shallow longitudinal depression on the premaxilla, laterally limited by a low elevation of the maxilla, similar to IRSNB 3806-M.540. This allows one to interpret the dorsal break of the premaxillae in IRSNB 3806-M.540 as the base of a lost median crest probably resembling the condition of NMB 002. The only difference between the two specimens which might be significant at a specific level is the shorter and more pointed rostrum in IRSNB 3806-M.540. However, IRSNB 3806-M.540 is too fragmentary to allow more precise taxonomic determination.

The anterior part of the cranium of NMB 002 shows interesting features, never previously observed in Belgian specimens: the closely appressed and deeply excavated premaxillary sac fossae are strongly asymmetric, with a much wider right premaxillary sac fossa nearly occupying the same surface on the left side of the medi- an plane of the skull as on the right ( Fig. 27B, C View FIG ). The fossae are anteriorly limited by a high elevation of the bone, in continuity with their lateral ridge. The right fossa is pierced in its anteromedi- an portion by three foramina; in that area, the surface is marked by irregular depressions.

Anterolaterally to the premaxillary sac fossa, the right maxilla supports a crest-like elevation, longitudinally elongated and medially overhanging the depressed suture between maxilla and premaxilla. This medially directed maxillary crest, with a maximum height of 20 mm roughly at the level of the unpreserved antorbital notch, is reminiscent, although smaller, of the crest in several platanistoids, in particular Zarhachis . The depressed surface between this crest and the premaxillary sac fossa is pierced by a large dorsal infraorbital foramen (diameter of 12 mm); this foramen sends several posterolateral, lateral and anterior sulci. A second smaller foramen (5 mm), piercing the maxilla posterolaterally to the crest, also sends posterior and anterolateral sulci, and five additional tiny foramina are present on the lateral flank of the crest, coupled with short anteriorly directed sulci. Posteriorly to the crest, the surface of the maxilla is posterolaterally sloping, and is not overhung by the lateral margin of the premaxillary sac fossa.

The morphology of the strongly asymmetric premaxillary sac fossae, anteriorly enclosed, shows similarities with Tusciziphius Bianucci, 1997 , from the Italian Pliocene, but no rostrum is known for that genus. More striking similarities are shared with Eboroziphius coelops Leidy, 1876 , from the Phosphate Beds of South Carolina (figured by Leidy 1877: pls 30, 31), again at the level of the premaxillary sac fossae, but also on the rostrum: the dorsal face of the rostrum of the holotype of E. coelops is marked by a median broken area separating two low depressions, as in IRSNB 3806-M.540 and probably related to a median longitudinal premaxillary crest. Furthermore, the maxilla of the holotype of E. coelops is elevated anterolaterally to the premaxillary sac fossa. The maxillary crest is poorly preserved, but it could be

A

Lambert O.

more dorsally and anteriorly developed than in NMB 002. A large foramen is also present between this elevation and the premaxillary sac fossa. This brief comparison between fragmentary specimens is not enough detailed to discriminate inter- and intraspecific differences. E. coelops was referred to Ziphiidae incertae sedis by Fordyce & Muizon (2001), because of a nondiagnostic holotype. NMB 002 is only slightly more complete, and is considered here, together with IRSNB 3806-M.540, as Ziphiidae aff. Eboroziphius . The interest of NMB 002 is then more anatomical than taxonomic: the elevated premaxillary crest on the anterior part of the rostrum tentatively explains the break observed in the holotype of E. coelops and in IRSNB 3806- M.540; it represents an alternative way for a fossil ziphiid to increase the mass of the apical part of its rostrum.

The extreme asymmetry of the premaxillary sac fossae and the abrupt anterior wall of the right fossa, characteristic of NMB 002, are present in USNM 13796, the holotype of Pelycorhamphus pertortus Cope, 1895 (exact locality unknown, Miocene of the Chesapeake Group). The fossil includes the right premaxillary sac fossa and a short portion of the right premaxilla on the rostrum ( Fig. 28 View FIG ). The wide and deeply excavated premaxillary sac fossa is similar to NMB 002, and although the premaxilla is more elevated anterior to the fossa, the two specimens are probably related. The fragment of skull USNM 360081 (location uncertain, North or South Carolina,?Pliocene), with a label Choneziphius trachops cf., is nearly identical to the holotype of Pelycorhamphus pertortus .

IRSNB

Institut Royal des Sciences Naturelles de Belgique

Kingdom

Animalia

Phylum

Chordata

Class

Mammalia

Order

Cetacea

Family

Hyperoodontidae

Genus

Beneziphius

Darwin Core Archive (for parent article) View in SIBiLS Plain XML RDF