DIRECT EVIDENCE OF THE ROSTRAL ANATOMY OF THE AÏSTOPOD PHLEGETHONTIA, WITH A NEW CRANIAL RECONSTRUCTION

Journal of Paleontology, Mar 2007 by Anderson, Jason S

The nasals are both visible (Fig. 2). The right nasal is displaced rostrally while the left is in articulation with the frontal. The nasals are irregularly diamond-shaped. They contact each other on the midline just rostral to the medial anterior process of the frontal, and run rostrally in articulation until they are separated by the nasal process of the premaxilla. The nasal is widest laterally at a point just rostral to the lateral anterior process of the frontal. The nasal passes rostrally along the nasal process of the premaxilla, and it gradually tapers to a point about even with the position where the sharp dorsal angle of the nasal process of the premaxilla begins to become less vertical. As a result, the nasal forms the entire dorsal margin of the external naris.

The prefrontal is poorly preserved. There is a strong impression of the caudal border of this bone, but the rostral border is difficult to distinguish. This is unfortunate because there is some uncertainty about the homology of this element. Anderson (2002) suggested it was a prefrontal because the large base of the triangularshaped bone seems to be oriented dorsally in most specimens, and because of the absence of a lacrimal canal. A large impression on the ventral surface of this specimen is suggestive of a reversed triangular shape expected of a lacrimal; however, this appears to be the rostral continuation of one of the palpebral ossifications.

Discussion.-One of the reasons the nasals are poorly represented in previously known specimens is that the rostrum of aïstopods is frequently near the margins of nodules, which corresponds with a zone of dissolution (Anderson, 2002). In ROM 48445 the rostrum is well away from the edge and is preserved in its entirety. Skull length measurements place this specimen towards the small end of the growth series of Anderson (2002), although it is twice as large as the smallest specimen. However, the specimen requires further preparation (which was not attempted) before the degree of cranial ossification can be determined.

The shape of the premaxilla leading to the "beak" is evident in other figured specimens (AMNH 2564, MCZ 2038); however, it was previously considered to be an artifact of preservation (Anderson, 2002) or was misinterpreted as separate premaxilla and nasal (McGinnis, 1967). The premaxilla of most other specimens is too fragmentary to show this morphology. One exception is NMW 1896 II 34 (Anderson, 2002, fig. 3.1), which clearly shows the nasal ramus of the premaxilla rising from the tip of the rostrum continuously, without the abrupt change in angle that forms the beaklike morphology. Reconsideration of this specimen suggests that this might be due to the view of this element; since the bases of the teeth are visible, the maxillary ramus may be rotated dorsomedially, obscuring more dorsal morphology.

This new information necessitates a new cranial reconstruction, which is presented in Figure 3. This reconstruction shows the correct shape and proportions of the nasal, and its relationship with the premaxilla's nasal ramus and anterior margin of the frontal. The frontal has been amended to show the longer lateral processes, while the premaxilla has been altered to show the more sharply inclined proximal portion of the nasal ramus. The length of the dental ramus of the premaxilla as seen in dorsal and palatal view (Fig. 3.1, 3.2) has been increased to match the length seen in lateral view (Fig. 3.3), and the size of the tooth bases has been amended from the previous reconstruction.