‘Iguanodon’ hillii The fragmentary nature of the holotype specimen of ‘Iguanodon’ hillii precludes detailed comparison. Many features of the tooth are shared with the majority of basal hadrosauroids and hadrosaurids, including a strong primary ridge, the presence of mammillate marginal denticles, and the absence of secondary ridges (see comparative comments on “Trachodon cantabrigiensis,” above). Although the tooth crown is incomplete basally, the absence of secondary ridges in ‘I.’ hillii is probably genuine, as in those basal hadrosauroids that possess secondary ridges these normally extend onto the apical part of the tooth (e.g., Head, 1998; Norman, 2002). In addition, this clearly prevents referral of the specimen to Iguanodon or other similar iguanodontian taxa, such as Mantellisaurus. The distal offset of the primary ridge distinguishes the tooth of ‘I.’ hillii from those of hadrosaurids (e.g., Prieto-Márquez, 2010a), although the lack of distal recurvature is more similar to the hadrosaurid condition than that seen in “T. cantabrigiensis” (see above) or Telmatosaurus (Weishampel et al., 1993; Dalla Vecchia, 2006). Bifurcated rows of mammilae similar to those of ‘I.’ hillii are also present in some specimens of Telmatosaurus (Dalla Vecchia, 2006).
Character Analysis and Phylogenetic Positions of “Trachodon cantabrigiensis” and ‘Iguanodon’ hillii
In order to assess the systematic position of “Trachodon cantabrigiensis” and ‘Iguanodon’ hillii, we scored the holotype teeth for characters listed in three recent phylogenetic analyses of hadrosauriformes: Norman (2002), Sues and Averianov (2009, which includes most of the characters presented in Horner et al. [2004]), and Prieto-Márquez (2010a). Dental characters comprise approximately 6% (21 of 370 characters in Prieto-Márquez, 2010a) to 18% (12 of 67 characters in Norman, 2002) of the total number of characters in these and other hadrosauroid phylogenetic analyses. However, only a subset of the dental characters can be scored for either “T. cantabrigiensis” or ‘I.’ hillii, and therefore they cannot be given formal phylogenetic treatments here. Nevertheless, given the morphological variation and distribution of these dental characters with Hadrosauroidea, even isolated teeth can be systematically placed within the clade with a reasonable degree of confidence in reference to specific phylogenies, as below.
Using the matrix of Norman (2002), NHMUK R496 can be scored for six characters (29–32, 35, and 38), half of which represent the plesiomorphic state (characters 29, 35, and 38). Of the derived character states, the restricted distribution of the dentary tooth enamel (30[1]) occurs in the least inclusive clade formed by Protohadros and hadrosaurids. NHMUK R496 has mammillated papillae 31[1]) that occur in non-hadrosaurid hadrosauroids, but lacks the reduced papillae of hadrosaurids (31[2]). The angular morphology of the tooth root suggests a derived condition of tooth emplacement (32[2]), which occurs in Protohadros, Bactrosaurus, Probactrosaurus, and hadrosaurids. GSM 1966 can only be scored for four characters (29–31 and 35) and is identical to “T. cantabrigiensis” in all scores. It should be noted that characters 29 and 35 of Norman (2002) are somewhat problematic, as they combine several morphological features into one character (e.g., presence or absence of secondary ridges and crown shape in the case of character 29) and there are no metrics (such as a ratio for crown height vs. length) to distinguish between states for broad and narrow crowns. We have tried to make our criteria for scoring characters compatible with those used by Norman (2002), but acknowledge that others might operationalize these character descriptions differently.
“Trachodon cantabrigiensis” can be scored for four characters in the matrix of Sues and Averianov (2009). Of these, three characters, 97(0), 98(0), and 100(0), exhibit the plesiomorphic state and are phylogenetically uninformative. A single derived state is present (character 99), in which the tooth crown is dominated by a single primary ridge. This character optimizes as an unambiguous, unreversed synapomorphy of the least inclusive clade formed by (Bactrosaurus + Levnesovia) and Hadrosauridae in the four most-parsimonious trees (MPTs) recovered by this analysis. Importantly, NHMUK R496 lacks the more symmetrical (98[0]) dentary crowns that are characteristic of more derived hadrosaurids. ‘Iguanodon’ hillii has identical character scores to “T. cantabrigiensis” for this analysis.
In terms of the analysis presented by Prieto-Márquez (2010a), NHMUK R496 can be scored for characters 4–11 and 14, and some of the component character states optimize unambiguously on to the consensus tree presented in this study. NHMUK R496 possesses the following character states (those with unambiguous distributions are marked with an asterisk): 4(1)*, 5(3)*, 6(1)*, 7(0), 8(2), 9(1), 10(0), 11(1)*, and 14(1). Character 4(1) suggests placement in an Eolambia+Protohadros clade and 6(1) indicates inclusion in a Lophorhothon + Hadrosauridae clade. Character 5(3) optimizes as a hadrosaurid synapomorphy in this analysis and 11(1) suggests similarity to a clade formed by (Corythosaurus + Lambeosaurus) + Hypacrosaurus. The character scores of NHMUK R496 for characters 8, 9, and 10 imply that that it cannot be referred to a variety of derived hadrosaurid clades. The absence of a sinuous primary ridge (character 7[0]) indicates that NHMUK R496 cannot be assigned to Saurolophinae (although a secondary reversal to this primitive state does occur among derived saurolophines). ‘Iguanodon’ hillii can be scored for characters 5–7, 9–11, and 14 and is identical in all states to “T. cantabrigiensis,” except in retaining the primitive state for character 11.
Character analysis of NHMUK R496 and GSM 1966 in the context of the phylogenetic analyses presented by Norman (2002) and Sues and Averianov (2009) suggests that both “Trachodon cantabrigiensis” and ‘Iguanodon’ hillii are basal hadrosauroids that lie just outside Hadrosauridae. The characters of the teeth suggest that they are both more derived than Altirhinus, Eolambia, and Probactrosaurus, and are consistent with them being Protohadros-Bactrosaurus-Telmatosaurus grade taxa. The analysis of Prieto-Márquez (2010a) supports the inclusion of both taxa in Hadrosauroidea and provides some evidence in favor of hadrosaurid affinities in both cases. The latter placements depend almost exclusively on the absence of secondary ridges in these specimens, which optimizes as a hadrosaurid synapomorphy. However, the utility of this character is slightly problematic, given that the presence or absence of faint secondary ridges can vary ontogenetically and positionally within hadrosauroid dentitions. For example, the presence or absence of secondary ridges is variable on the maxillary teeth of Levnesovia (Sues and Averianov, 2009) and Jayewati (McDonald, Kirkland, et al., 2010), suggesting that the loss of secondary ridges is likely convergent in “T. cantabrigiensis” and derived hadrosaurids. As “T. cantabrigiensis” lacks many of the features optimizing as hadrosaurid dental characters in all three analyses, we consider that the available evidence is more consistent with a basal hadrosauroid position for this taxon.
Table 6.1. Taxa Used in Morphometric Analysis and Their Specimen Number or Literature Reference
| Taxon | Reference |
| Brachylophosaurus canadensis | FMNH PR 682 |
| Bactrosaurus johnsoni* | Godefroit et al. (1998) |
| Charonosaurus jiayinensis |
|