Hadrosaurs. David A. Eberth
pieces (Fig. 3.20). The proximal end of the right femur preserves the lesser trochanter and a small part of the greater trochanter. The greater trochanter is better preserved on the left femur. The lesser trochanter forms a finger-like process that terminates ventral to the greater trochanter and is separated from it by a distinct cleft. The dorsal surface of the greater trochanter is rounded in lateral view. The femoral head is not preserved on either femur, but the preserved portion of the greater trochanter suggests it was separated from the head by a saddle-shaped sulcus in dorsal view. The mid-shaft region is present on the left femur, but extremely badly crushed and eroded. The fourth trochanter is visible and projects caudally. It takes the form of an elongate crest that is concave medially and fringed with striations. The distal condyles from the right side are preserved, although they have become offset from each other by craniocaudal crushing and much of the lateral condyle is eroded on its caudal surface. A deep intercondylar extensor groove is present cranially, and was probably almost entirely enclosed by expansion of the lateral and medial condyles. The ventral surface of the medial condyle is strongly convex ventrally, and it curves strongly caudally in medial view. Detailed comparisons with other taxa are precluded by the poor state of preservation; however, features of the greater, lesser, and fourth trochanters and the distal condyles do not appear to differ significantly from the femora of Iguanodon (Norman, 1980) or Probactrosaurus (Norman, 2002).
Other Appendicular Material Three large conjoined bone fragments appear to be parts of the appendicular skeleton, but are too poorly preserved for a reliable identification or any features of their anatomy to be seen.
DISCUSSION
Ontogenetic Age of IVPP V 12534
All the neural arches of the preserved vertebrae are fused to their respective centra, suggesting that the holotype of Equijubus normani, IVPP V 12534, is a skeletally mature individual (Brochu, 1996). There are only six vertebrae in the sacrum, a low number compared to other basal iguanodonts (see above), but the first and sixth preserved sacrals are incomplete; thus, the low sacral count might be due to incomplete preservation of the sacrum rather than lack of fusion between sacral vertebrae.
3.20. Fragments of femora of IVPP V 12534, holotype of Equijubus normani. (A) left femur in caudal view. (B–E) right femur: (B) proximal end in cranial view; (C–E) distal end in (C) cranial, (D) medial and (E) distal view. Abbreviations: 4t, fouth trochanter; gt, greater trochanter; lt, lesser trochanter. Scale bar equals 10 cm.
Changes Made to Previous Diagnoses of Equijubus
Two characters have been removed from the original diagnosis of You, Luo, et al. (2003):
1. “Very large lower temporal fenestra,” which is later described as being “twice the size of the orbit” (You, Luo, et al., 2003:349). The lower temporal (= infratemporal) fenestra is complete only on the right side of the skull. The size of the right infratemporal fenestra relative to the right orbit is exaggerated by the mediolateral compression of the skull, which has crushed the right frontal and postorbital rostroventrally into the orbit (Fig. 3.4). The shape and relative size of the right infratemporal fenestra of IVPP V 12534 are not dissimilar to those of Iguanodon (Norman, 1980:fig. 2), Altirhinus (Norman, 1998:fig. 3), and possibly Xuwulong (the only skull of this taxon has been crushed caudoventrally, as indicated by the displacement of the jugal relative to the quadrate [You et al., 2011:fig. 3]).
2. “Lacking the median primary ridge on the crown of the dentary teeth” (You, Luo, et al., 2003:349). Reinterpretation of the dentary teeth of IVPP V 12534 indicates that there is actually a distally offset primary ridge (Fig. 3.10A, B).
The emended diagnosis of Equijubus normani proposed by Paul (2008) consists of two proportional characters and a number of qualitative statements. Paul’s diagnosis includes the autapomorphic finger-like process of the jugal originally noted by You, Luo, et al. (2003) and included in the revised diagnosis herein (autapomorphy 1), as well as the distinctive morphology of the lacrimal noted by You, Luo, et al. (2003) and included herein (autapomorphy 2); Paul phrased his description of the lacrimal somewhat differently (“lacrimal long,” “anterior process wedges between premaxilla and maxilla,” “ventral edge at level of dorsal edge of maxilla” [Paul, 2008:201]). Paul (2008) also noted the lack of contact between the lacrimal and the nasal, which is related to the contact between the premaxilla and prefrontal that we included above in the unique combination of characters. Paul (2008:201) also mentioned that the shaft of the quadrate is “nearly straight,” similar to our observation that the quadrate gently curves caudally along its entire length.
However, the same general issues noted by Barrett et al. (2009) in Paul’s (2008) emended diagnosis of Jinzhousaurus yangi and by McDonald (2012a) in the diagnosis of “Dollodon bampingi” (= Mantellisaurus atherfieldensis) apply to that of Equijubus normani. The remaining characters used by Paul (2008:201) in the emended diagnosis of E. normani are ambiguous subjective statements that would be better treated as quantitative characters (e.g., premaxilla “maxillary process is shallow,” “dorsal apex of maxilla sited anteriorly,” “quadratojugal tall,” “quadrate tall,” quadrate “lateral foramen set moderately low,” quadrate “dorso-posterior buttress small,” “diastema long,” “main body of ilium deep”), possible taphonomic consequences (“primary palpebral absent”), or widely distributed among basal iguanodonts (e.g., “premaxilla projects well below level of tooth rows” [Barrett et al., 2009], tooth “battery tightly packed” [Norman, 2004]). Paul (2008:201) also noted that the antorbital fossa is “small”; however, the antorbital fossa is not visible in lateral view (Figs. 3.3, 3.4). The proportional character “premaxillary tip to anterior orbital rim/latter to paraoccipital process tip length ratio ~1.0” (Paul, 2008:201) should be employed with caution until a morphometric analysis is carried out to assess intraspecific variation in basal iguanodont cranial proportions using taxa for which multiple skulls are known (e.g., Iguanodon bernissartensis [Norman, 1980]), as has been carried out for several hadrosaurids (Dodson, 1975; Evans, 2010; Campione and Evans, 2011). Paul (2008) also mentioned that 23 tooth positions are present in the maxilla; however, the caudal end of the right maxilla is obscured by the right jugal and the coronoid process of the right dentary, and the caudal end of the left maxilla is obscured by the coronoid process of the left dentary, rendering an accurate count of alveoli impossible. Finally, the character “dentary pre-coronoid process length/minimum depth ratio under 4” is also problematic; McDonald (2012a) demonstrated that the proportions of basal iguanodont dentaries can exhibit a considerable degree of intraspecific variation, so it is entirely possible that an individual of Equijubus normani will be discovered in which this ratio is greater than four.
Phylogenetic Relationships of Equijubus
The new information on the anatomy of Equijubus presented here allows its phylogenetic placement to be more confidently tested. Previous phylogenetic analyses have recovered Equijubus in a variety of positions. The analysis by You, Luo, et al. (2003) in the original description of Equijubus found it to be the most basal hadrosauroid, more derived than Iguanodon, Ouranosaurus, Altirhinus, and Jinzhousaurus, but more basal than Probactrosaurus. In contrast, the analysis of Norman (2002) placed Equijubus (referred to as “Mazongshan sp.”) in a much more basal position – basal to Iguanodon, Mantellisaurus, Ouranosaurus, and Altirhinus, and in a polytomy with Lurdusaurus. However, Norman