Barilium dawsoni (Lydekker, 1888) Lydekker, 1888

Norman, David B., 2010, A taxonomy of iguanodontians (Dinosauria: Ornithopoda) from the lower Wealden Group (Cretaceous: Valanginian) of southern England, Zootaxa 2489, pp. 47-66 : 50-54

Barilium dawsoni (Lydekker, 1888) comb. nov.

Figs 3–4

Nomenclatural history

Iguanodon dawsoni Lydekker, 1888a: 51 .

Iguanodon dawsoni [ Lydekker 1888b: 196].

Iguanodon dawsoni [ Lydekker 1890a: 38].

Iguanodon dawsoni [ Lydekker 1890b: 259].

Iguanodon dawsoni Lydekker, 1888 [in Steel 1969: 18] Iguanodon dawsoni Lydekker, 1888 [in Norman 1987: 168]. Iguanodon dawsoni Lydekker, 1888 [in Norman & Weishampel 1990: 530]. Iguanodon dawsoni Lydekker, 1888 [in Blows 1998: 31]. Iguanodon dawsoni Lydekker, 1888 [in Norman 2004: 416]. Iguanodon dawsoni Lydekker, 1888 [in Galton 2009: 245]

Holotype. NHMUK R798, R798b, R799, R800–R806, R4771, R4742 ( Figs 3 , 4 : illustrative examples). Despite the variety of registered numbers, all refer to a single partial skeleton collected from one quarry in the village of Shornden, near Hastings, East Sussex.

Referred material. NHMUK R4746, R3788, R2848, R3789, and further specimens to be described by Norman (in preparation).

Locality and horizon. Collected from exposures of the Hastings Sub-Group, Wadhurst Clay Formation (Valanginian), in a series of quarries in the Hastings area of East Sussex, England.

Revised diagnosis based on the holotype. Iguanodontian with the following unique character combination (* indicates autapomorphies - all other characters are apomorphic within basal iguanodontians even if they occur sporadically within ornithopods more generally), summarised by element:

Ilium ( Fig. 3 ): preacetabular process (prp) with a prominent medial flange that arises from a large sacrodorsal rib/transverse process facet near the base of the process; rim of the sacrodorsal rib facet (srf) is visible, in lateral view, in the upper part of the recess between the preacetabular process and the pubic peduncle (pp)*;lateral surface of the preacetabular process twists axially to face dorsally toward its anterior end*; the principal body of the iliac blade is flat and stands vertically ( Fig. 3 B); in profile the entire dorsal margin of the iliac blade is gently convex; the dorsal edge of the acetabular blade is flattened and transversely thick* ( Fig. 3 B); a minor indent occurs along the posterodorsal margin of the postacetabular blade, shows no accompanying lateral expansion of the blade; postacetabular portion of the iliac blade is bluntly truncated and its lower half curves gently medially; very restricted brevis shelf (fossa) that is not bounded by a prominent lateral ridge*; the ventral border of the postacetabular blade lies horizontally and at the same level as the ischial peduncle (ip)*; the medial surface of the ilium is prominently scarred across its ventral half where it anchors the sacral yoke but, uniquely, bears yoke scars low down along the ventral margin of the postacetabular blade* ( Fig. 3 C, srs); acetabulum is extremely large, forming a exceptionally broad, laterallyoriented, cup-shaped depression*.

Vertebrae ( Fig. 4 A, B): posterior dorsal vertebrae have spool-shaped, cylindrical centra with thin everted articular margins and transversely rounded haemal surfaces*; proximal caudal vertebrae ( Fig. 4 C, D) have broad, dorsoventrally compressed centra* (subrectangular in anterior view: Fig. 4 D); distal caudal centra are deeply amphicoelous*.

Comparisons with closely-related iguanodontians. This taxon appears (to date) to be restricted to the Valanginian stage and is representative of a massive (estimated body length of more than 8 metres) and extremely robust iguanodontian ornithopod. Comparisons will be drawn with the earlier (Kimmeridgian– Berriasian) camptosaur-grade iguanodontians, the exactly contemporary taxon Hypselospinus dawsoni (see below), and the younger (Hauterivian–Lower Aptian) and geographically local Iguanodon bernissartensis and Mantellisaurus atherfieldensis .

Camptosaurs. The massive iliac blade of Barilium dawsoni , with its smoothly curved dorsal margin, lack of a lateral eversion on its dorsal edge posterodorsal to the ischial peduncle, the massive dorsally twisted preacetabular process, deep postacetabular blade, modest brevis shelf (fossa) and no appreciable development of a lateral ridge that bounds the brevis shelf, contrasts strikingly with the form of the ilium in all known camptosaurs ( Gilmore 1909, 1925; Carpenter & Wilson 2008). The forms and proportions of the known posterior dorsal and anterior caudal vertebrae also contrast strongly with those seen in all known camptosaurs ( Gilmore 1909, 1925).

Hypselospinus fittoni . The shape and proportions of the ilium are completely distinct (compare Figs 3 , 5 ), and these differences are discussed further below.

Iguanodon bernissartensis . This taxon is perhaps the nearest in size and overall build to the new taxon proposed here, although I. bernissartensis is from the Hauterivian–Lower Aptian and, has not (to date) been recognised in the Valanginian. The ilium of I. bernissartensis ( Norman 1980: figs 63, 64) differs from that of B. dawsoni in the following ways: it has a stout preacetabular process, but it is not twisted dorsally towards its anterior tip and is more strongly bowed ventrally; the main iliac blade is more strongly convex and sinuous along its dorsal margin and also has a thickened, but rounded, dorsal edge which is everted laterally above and behind the ischial peduncle; the postacetabular blade is more elongate, tapers and has a pronounced brevis fossa. The centra of the posterior dorsals of I. bernissartensis are ventrally keeled ( Norman 1980: fig. 38) and show thickened, everted articular margins. Neural spines of dorsal vertebrae are similarly proportioned, being broad-bladed and moderately inclined, however their posterior edges only have a short median recess immediately above the neural platform in I. bernissartensis . The anterior caudal vertebrae of I. bernissartensis ( Norman 1980: fig. 49) do not exhibit the dorsoventrally compressed outline seen in B. dawsoni and the distal caudals do not develop strongly amphicoelous articular surfaces.

Mantellisaurus atherfieldensis . This comparatively gracile Hauterivian–Lower Aptian species bears little resemblance to the new taxon B. dawsoni . The ilium of M. atherfieldensis ( Norman 1986: figs 36, 53, 54) differs in almost all its structure and proportions to those described and illustrated in B. dawsoni . The dorsal vertebrae of M. atherfieldensis have generally smaller, lower and more cylindrical centra, with a ventral keel and more strongly everted articular margins; the neural spines, though broad and blade-like, are more elevated (being at least 2.5 times higher than the height of the centrum, compared with a maximum of 2 times that ratio in the case of B. dawsoni ). The anterior caudal centra of M. atherfieldensis do not form dorsoventrally compressed (subrectangular) structures, and there are no strongly amphicoelous articular surfaces on distal caudal centra.

Anatomical discussion. Additional material from the Wadhurst Clay Formation of southern England that is referable to this taxon expands upon what is presently known of the anatomy of the holotype (Norman in preparation).

Axial skeleton. The sacrum (NHMUK R3788, and additional material) comprises six fused vertebrae (including the sacrodorsal) and, even though closely opposed, the row of tall neural spines shows no evidence of fusion to form a sacral plate. Ossified tendons were arranged on either side of the sacral spines in a lattice as seen in camptosaurs ( Gilmore 1909, 1925; Carpenter & Wilson 2008) and more derived iguanodontians ( Norman 1980, 1986).

Pelvic structure. The anterior pubic process is blade-like (dorsoventrally expanded) and appears slightly upwardly curved, as well as having a transversely thickened dorsal edge and narrow ventral margin. The ischium has a J-shaped shaft, is very robust (with a well-developed, proximally placed, flap-shaped obturator process), and the distal end of the ischial shaft terminates in a large, anteriorly expanded, ‘foot’. The acetabulum is unusually large and open laterally. Apart from their considerably larger size, the shape and general proportions of the anterior pubic blade and ischium resemble those seen in camptosaur-grade taxa ( Carpenter & Wilson 2008); however, the pubis (proper), judged by the incomplete preserved remains, was probably shorter than the shaft of the typical camptosaur ischium, and the lower pelvic bones probably more closely resembled those of I. bernissartensis (as illustrated by Blows 1998: fig. 3A).

Appendicular skeleton. A group of associated and similarly robust remains (NHMUK R2848) are referable to this taxon and include a scapula that is massive, yet the blade is notably constricted above the glenoid region, but expands toward its distal (suprascapular) edge (to a greater degree than seen in all examples of I. bernissartensis ). The coracoid bears a large, fully-enclosed coracoid foramen adjacent to the scapulocoracoid suture; by contrast, this foramen forms a notch on the scapulocoracoid in the similarly massively built I. bernissartensis ( Norman 1980: fig. 53). The femur is massive with a particularly large, spherical head and very prominent and thick anterior trochanter, the femoral shaft is curved and angular-sided, and the fourth trochanter is also large and ‘crested’ with a slightly pendant tip as seen in H. fittoni ( Fig. 7 B).

The Henfield Skull. A large, poorly preserved, skull block (NHMUK R8306) was recovered from a locality at Henfield by Stuart Baldwin and attributed to I. bernissartensis ( Norman 1977, 2004; Norman & Weishampel 1990), primarily by reference to its large size and robust proportions, allied to the general suspicion that it was collected from the Weald Clay Formation (Barremian). The Hastings Sub-Group also crops out at Henfield and this specimen may prove ultimately to be Valanginian (Norman in preparation). Maxillary crowns and other components of the skull are preserved, along with a finely prepared endocast of the neurocranial cavity ( Norman 1977, 2004, Norman & Weishampel 1990) and this may also prove to be attributable to Barilium sp. (Norman in preparation).

The Old Roar Quarry Specimen. Blows (1998) referred a partial articulated skeleton collected from Old Roar Quarry, near Hastings (NHMUK R3788) to Iguanodon dawsoni (following Norman 1977); this specimen comprises the posterior dorsal series, sacrum, both ilia and parts of the right pubis and ischium. The assignment appears to be correct, despite the unusual shape of the right ilium. A composite restoration that combined an outline of the holotype ilium, the dorsal and sacral series of NHMUK R3788, and restored some unassociated lower pelvic bones, was provided by Blows (1998: fig. 3A). The outline and general form of the right ilium of NHMUK R3788, which displays an unusually sinuous dorsal margin, reflects probable preburial desiccation and post-mortem distortion; this is clear not only from the extensively cracked and crazed appearance of its bone surface but, more crucially, from the form of the left ilium which is attached to the opposite side of the mounted specimen (Norman in preparation).

Interpretations. Norman (1977, 1987) took an essentially pragmatic decision to retain this species within the genus Iguanodon prior to being able to study inter- and intrageneric variation in ornithopods more widely (while admitting – following Ostrom 1970 – that the nomenclature of Wealden ornithopods was recognised to be “in a chaotic state”; Norman 1987: p. 132). It is now clear from the diagnosis that the anatomy of this taxon, as well as its stratigraphic age (Valanginian), separates it anatomically, biologically and chronologically from the far better established taxa of the Hauterivian–Lower Aptian (Weald Clay and Wessex Formations) of the UK ( Norman 1980, 1986) and Barremian–Lower Aptian of Belgium ( Norman 1980; Yans et al. 2006). It is interesting to note that the generic status of the latter species was the subject of debate following the preliminary description of new material by Dollo (1882) and Hulke (1882). Discussion concerning generic names involved Seeley in Hulke’s (1882) account of I. seelyi and was presented more formally in two discussion papers by Seeley (1887b). I tacitly accept the new generic assignment of I. atherfieldensis Hooley, 1925 as Mantellisaurus atherfieldensis ( Hooley, 1925) , as proposed by Paul (2007). However, additional taxonomic proposals ( Paul 2008) are not accepted (Norman in press; Norman in preparation).

Several missing skeletal components (manus, forelimb, parts of shoulder girdle such as the sternal bones) inhibit the accurate systematic placement of this taxon within the clade Iguanodontia (Norman in preparation). However, even at the cursory level of this account, it should be clear that this unique Valanginian ornithopod shares general anatomical affinities with a range of camptosaur–styracosternan grade iguanodontian taxa. However, the anatomy of this taxon contrasts strongly with that described in earlier camptosaur-grade iguanodontians as well as the contemporary and sympatric iguanodontian ( Hypselospinus fittoni ) and the younger Weald Clay Formation taxa Iguanodon and Mantellisaurus ( Norman 1980, 1986). The absolute distinction in the structure of the holotype (and referred material) confirms that it represents a new genus of basal iguanodontian: Barilium dawsoni (Lydekker, 1888) comb. nov.

Figures

FIGURE 3. Barilium dawsoni (Lydekker, 1888) comb. nov. NHMUK R 802, holotype ilium in lateral (A), dorsal (B), and medial (C) views. Abbreviations: acet – acetabulum; ip – ischial peduncle; pp – pubic peduncle; prp – preacetabular process; srf – sacral rib facet; srs – posterior sacral rib scar. Broken surfaces indicated by cross-hatching.

FIGURE 4. Barilium dawsoni (Lydekker, 1888) comb. nov. Some further examples of the holotype partial skeleton. A. Middle dorsal vertebra, NHMUK R 798. B. Posterior dorsal vertebra, NHMUK R 805. C, D. Proximal caudal vertebra, NHMUK R 805 (in lateral and anterior views respectively). Broken surfaces indicated by cross-hatching.

FIGURE 5. Hypselospinus fittoni (Lydekker, 1889) comb. nov. NHMUK R 1635, holotype ilium. A. Lateral view. B. Posterior view to show the inflected shelf, lateral ridge and brevis fossa. C. Medial view showing the form of the sacral rib scars. D. Dorsal view showing the transversely narrow dorsal margin of the ilium compared with that of B. dawsoni (Fig. 3 B). The mediodorsal edge of the ilium shows an abscess-like excavation that may be pathological in origin. Even though the dorsal margin is eroded laterally posterior to the ischial peduncle (cross-hatched area) there is little indication of a laterally everted edge. Abbreviations: acet – acetabulum; brf – brevis fossa; lr – lateral ridge that bounds the brevis fossa; nde – narrow dorsal edge to the iliac blade; path? – probable pathological structure; prp – preacetabular process; srs – sacral rib scar on medial surface of iliac blade. Broken surfaces indicated by cross-hatching.

FIGURE 7. Hypselospinus fittoni (Lydekker, 1889) comb. nov. NHMUK R 1148, referred specimen. The right femur, illustrated as part of the holotype of Iguanodon hollingtoniensis Lydekker, 1889. A. Anterior view. B. Anteromedial view. Abbreviations: aig – anterior intercondylar groove almost occluded by shoulders developed from the distal articular condyles; at – anterior trochanter; 4 t – fourth trochanter and associated muscles scars associated with the caudifemoral complex. Broken surfaces are indicated by cross-hatching.