Morganucodon, Kuhne, 1949
publication ID |
https://doi.org/10.1093/zoolinnean/zlae062 |
DOI |
https://doi.org/10.5281/zenodo.14981613 |
persistent identifier |
https://treatment.plazi.org/id/375EAA21-FF82-CB4E-FEEF-A7090A7242E8 |
treatment provided by |
Plazi |
scientific name |
Morganucodon |
status |
|
In this section, we revise information on the endocast of the inner ear and stapedial morphology and provide a detailed description of the intrapetrosal circum-promontorial plexus of Morganucodon . Graybeal et al. (1989) first reported on the bony labyrinth of the inner ear of Morganucodon based on serial sectioning, but several features can now be described in greater detail and/or corrected based on high-resolution µCT scanning. Graybeal et al. (1989) noted that the petrosals were donated by F. R. Parrington; therefore, they were almost certainly found in the Pontalun 3 fissure. In addition to the description of the internal morphology, we provide high-resolution images and three-dimensional (3D) surface files of 19 petrosals that are particularly well preserved ( Figs 1–6 View Figure 1 View Figure 2 View Figure 3 View Figure 4 View Figure 5 View Figure 6 ; Supporting Information, Fig. S1 View Figure 1 ). The surface anatomy of the petrosal of Morganucodon has previously been described in detail by Kermack et al. (1981), and our description focuses mainly on the bony labyrinth of the inner ear and its inferred neurovasculature. In this section, we refer to Graybeal et al. (1989) and Kermack et al. (1981) when relevant, but any broader comparison with other Mesozoic mammaliaforms can be found in the Discussion.
Inner ear endocast
Endocasts are from specimens NHM UK PV M21105 ( Figs 1A–G View Figure 1 , 2C, D View Figure 2 ), NHM UK PV M21200 ( Fig. 2E, F View Figure 2 ), NHM UK PV M21208 ( Fig. 3C, D View Figure 3 ), and NHM UK PV M27144 View Materials ( Fig. 4C, D View Figure 4 ). None of the endocasts is complete, and most frequently parts of the semicircular canals are missing. Except for NHM UK PV M21220 View Materials ( Fig. 6A–D View Figure 6 ), which is represented only by the pars canalicularis that houses the semicircular canals, all specimens preserve the cochlear canal, and many preserve the endocasts of the vestibule and at least one semicircular canal. A detailed list of the specimens and their preservation and completeness is provided in Table 1 View Table 1 . The following cochlear canal description relies most heavily on NHM UK PV M19112 ( Fig. 2A, B View Figure 2 ), NHM UK PV M21105 ( Fig. 2C, D View Figure 2 ), NHM UK PV M21200 ( Fig. 2E, F View Figure 2 ), and NHM UK PV M21208 ( Fig. 3C, D View Figure 3 ), all from Pant 2 fissure, 1955 pocket.
The cochlear canal is gently curved, with the greater curvature facing ventrolaterally or abneural and the lesser curvature facing dorsomedially or neural. The ‘slightly curved’ nature of the cochlear canal has already been noted by Graybeal et al. (1989: 112) based on serial-sectioned specimens. The cochlear canal ranges between 1.63 and 1.84 mm in length from the anterior aspect of the fenestra vestibuli to the apex and between 2.43 and 2.57 mm in length from the posterior border of the fenestra vestibuli to the apex ( Table 2 View Table 2 ), which is at the lower limit of measurements noted by Graybeal et al. (1989: 2.5–3.1 mm). The apex of the cochlear canal is expanded along its neural margin, possibly to accommodate the lagenar macula ( Figs 2B, D, F View Figure 2 , 3B, D, F View Figure 3 ). The expansion is particularly visible in lateral or medial view but is also recognizable when comparing cochlear canal diameter. The diameter increases between 5% and 15% from anterior to the cochlear foramen to the midpoint of the apical expansion in most of the specimens. In contrast, Graybeal et al. (1989) noted that the cochlear canal ‘tapers smoothly to its anterior tip with no indication for […] a specialized surface for the lagena macula’. Indeed, the very tip of the cochlear canal anterior to the apical expansion tapers to a narrow point, but an expansion and distinct oval-shaped bulge along the neural aspect of the cochlear apex is present in all specimens that were µCT scanned ( Figs 2B, D, F View Figure 2 , 3B, D, F View Figure 3 , 5B, D View Figure 5 ). The serial sections used by Graybeal et al. (1989) are at much lower resolution, with 25–76 µm intervals between slices, in comparison to <3 µm on average in our data set. The lower resolution is likely to have hindered recognition of the extent of the apical expansion. A very gentle expansion appears to be visible in fig. 2C of Graybeal et al. (1989), but it is less clear in the other views of this figure. A separate canal for the lagenar nerve is not preserved in any of the specimens; however, 9 of the 32 specimens that preserve the cochlear foramen well enough have a very small foramen along the lateroposterior edge of the cochlear foramen (e.g. see NHM UK PV 27691 in Fig. 1H View Figure 1 ). In 13 specimens there is a notch in the same location that, probably, in life was represented by a foramen. The bar that separates the small foramen from the cochlear nerve is very thin and could easily break. The foramen or notch is absent in 8 of the 32 specimens. It is possible that the lagenar nerve or vessels supplying the cochlear duct passed through this foramen in life. Interestingly, all of the six specimens from Pontalun 3 have a foramen (five) or notch (one), whereas it is absent in the two Pontalun 2 specimens in which we could observe the morphology. The picture is more mixed in the Pant 2 sample, with 4 specimens preserving a clear foramen, 12 preserving a notch, and 6 lacking a foramen.
Along the ventral aspect of the cochlear canal is a gently elevated ridge and groove that extends from the perilymphatic foramen posteriorly to the apex of the cochlear canal anteriorly. The ridge and groove are best visible in NHM UK PV M19112 ( Fig. 2A View Figure 2 ), NHM UK PV M21105 ( Fig. 2C View Figure 2 ), M21200 ( Fig. 2E View Figure 2 ), and NHM UK PV M27690 View Materials ( Fig. 4A View Figure 4 ). The groove is reminiscent of, although much shallower than, the base of the secondary osseous lamina of other early mammaliaforms ( Schultz et al. 2017, 2022a, b, Hoffmann et al. 2023). A proper primary or secondary osseous lamina, as in extant therians, is absent in Morganucodon , but the gentle elevation and groove possibly marks the attachment site of the basilar membrane, and we refer to it here as the secondary lamina base. Next to the ridge and groove are numerous openings for the circum-promontorial plexus ( Fig. 1G View Figure 1 ), which is described in more detail below.
The semicircular canals have not been covered previously by Kermack et al. (1981) or Graybeal et al. (1989). The pars canalicularis is preserved most completely in NHM UK PV M21105 and NHM UK PV M21220 View Materials ( Fig. 6 View Figure 6 ), with all three semicircular canals nearly intact. NHM UK PV M21208 ( Fig. 3C, D View Figure 3 ), NHM UK PV M21210 ( Fig. 3E, F View Figure 3 ), NHM UK PV M27135 ( Fig. 4A, B View Figure 4 ), and NHM UK PV M27144 View Materials ( Fig. 4C, D View Figure 4 ) preserve the lateral and posterior semicircular canals, and NHM UK PV M21173 ( Fig. 3A, B View Figure 3 ), NHM UK PV M21200 ( Fig. 2E, F View Figure 2 ), NHM UK PV M21165 (Supporting Information, Fig. S1A, B View Figure 1 ), NHM UK PV M21178 (Supporting Information, Fig. S1C, D View Figure 1 ), NHM UK PV M21192 View Materials (Supporting Information, Fig. S1E, F View Figure 1 ), and NHM UK PV M21207 (Supporting Information, Fig. View Figure 1 S1G, H) preserve only the lateral semicircular canal. The lateral semicircular canal is the shortest in length, ranging between 1.1 and 1.2 mm in the 12 specimens ( Table 2 View Table 2 ). The posterior semicircular is more than twice the length, ranging between 2.1 and 2.7 mm in the six specimens that preserve it ( Table 2 View Table 2 ), and the anterior semicircular canal is the longest, being 3.1 mm in NHM UK PV M21105 and 2.7 mm in NHM UK PV M21220 View Materials ( Table 2 View Table 2 ). In the two specimens that preserve all semicircular canals, the radius of curvature is similar between the lateral and posterior canal, being 0.30 and 0.31 mm in NHM UK PV M21105 and 0.27 and 0.29 mm in NHM UK PV M21220 View Materials , respectively. In contrast, the radius of curvature is much larger in the anterior semicircular canal, being 0.57 mm in NHM UK PV M21105 and 0.55 mm in NHM UK PV M21220 View Materials . The anterior semicircular canal is not only larger, but it is also more elliptical than either the lateral or posterior semicircular canal ( Table 2 View Table 2 ; Fig. 6 View Figure 6 ). The semicircular canals are stout, with an average diameter of 0.24 mm for the anterior semicircular canal, 0.26 mm for the posterior semicircular canal, and 0.34 mm for the lateral semicircular canal. The anterior and posterior semicircular canals meet to form a crus commune ( Fig. 6A, E View Figure 6 ). The semicircular canals lead to distinct and well-rounded ampullae ( Fig. 6 View Figure 6 ). The vestibule, containing the membranous saccule and utricle in life, is not inflated.
Vasculature
An extensive network of canals surrounds the cochlear labyrinth of Morganucodon ( Figs 1–5 View Figure 1 View Figure 2 View Figure 3 View Figure 4 View Figure 5 ). This dense network of venous channels has already been recognized by Kermack et al. (1981), who termed it the circum-promontorial plexus. Here, we document the 3D morphology of the circum-promontorial plexus and variations in vascularization.
The circum-promontorial plexus consists of two main venous canals, the inferior petrosal sinus medially (‘ventral vessel’ of Graybeal et al. 1989) and the prootic sinus laterally (‘dorsal vessel’ of Graybeal et al. 1989). The inferior petrosal sinus and prootic sinus are connected within the petrosal by several canals that pass transversally, ventral (hypocochlear sinus after Harper and Rougier 2019) and dorsal (epicochlear sinus after Harper and Rougier 2019) to the cochlear labyrinth. In addition, the inferior petrosal sinus and prootic sinus meet at the apex of the cochlear canal, where, in some specimens, they expand into a large sinus rather than forming distinct vessels ( Figs 1B, D View Figure 1 , 2C, D View Figure 2 ). The number and size of the inferior petrosal sinus, prootic sinus, and epicochlear and hypocochlear sinuses are highly variable in our sample.
The inferior petrosal sinus passes along the medial aspect of the cochlear canal between the petrosal and basisphenoid/ basioccipital. It exits the promontorium extracranially through the inferior petrosal foramen along the medial edge of the promontorium anterior to the perilymphatic foramen (labelled ‘ips’ in Figs 1A, E View Figure 1 , 2A, C View Figure 2 , 3A, C View Figure 3 , 4A, C View Figure 4 ). This foramen has been labelled either ‘circum-promontorial plexus of sinus canals’ (pr. s. cnl; fig. 75, 81 82) or ‘foramen for circum-promontorial plexus’ (for. Pr.; fig. 83) in the paper by Kermack et al. (1981). The inferior petrosal sinus ranges from multiple broadly anastomosing canals, as in NHM UK PV M21105 ( Figs 1B View Figure 1 , 2C View Figure 2 ) and NHM UK PV M21200 ( Fig. 2E View Figure 2 ), to a single and clearly delineated canal, as in NHM UK PV M27135 ( Fig. 4A View Figure 4 ), NHM UK PV M27144 View Materials ( Fig. 4C View Figure 4 ), and NHM UK PV M27690 View Materials ( Fig. 5E View Figure 5 ). The diameter of the canals varies greatly. We assume that the inferior petrosal sinus would have drained into the jugular vein with the lateral head vein in life ( Wible and Hopson 1995, Rougier and Wible 2006).
Along the lateral aspect of the cochlear canal extends the prootic sinus; likewise ranging from a single well-delineated canal, as in NHM UK PV M19112 ( Fig. 2B View Figure 2 ) and NHM UK PV M27144 View Materials ( Fig. 4D View Figure 4 ), to multiple canals, as in NHM UK PV M21200 ( Fig. 2F View Figure 2 ). The prootic sinus opens into the lateral trough through the prootic foramen where it would have joined the posttrigeminal vein (‘vena capitis lateralis’ of Kermack et al. 1981) to form the lateral head vein (‘main trunk of vena capitalis lateralis’ of Kermack et al. 1981) ( Wible and Hopson 1995, Rougier and Wible 2006). In most specimens there are additional smaller openings from the circum-promontorial plexus into the lateral trough that are likely to have transmitted emissary veins to join the extracranial posttrigeminal vein in the lateral trough ( Figs 1A, C View Figure 1 , 2A, C, E View Figure 2 , 3A, C View Figure 3 , 4A View Figure 4 , 5A, C, G View Figure 5 ). Ventral to the cochlear canal and connecting the inferior petrosal sinus and prootic sinus are the hypocochlear sinuses. This part of the circum-promontorial plexus is the most variable in our sample. It can consist of several well-delineated canals along the posterior aspect of the cochlear labyrinth, as in NHM UK PV M21210 ( Fig. 3E View Figure 3 ), NHM UK PV M27144 View Materials ( Fig. 4C View Figure 4 ), or NHM UK PV M27690 View Materials ( Fig. 5A View Figure 5 ), or a dense honeycomb-like anastomosis of canals, as in NHM UK PV M19112 ( Fig. 2A View Figure 2 ), NHM UK PV M21208 ( Fig. 3C View Figure 3 ), or NHM UK PV M27702 ( Fig. 4G View Figure 4 ). Regardless of the shape of the hypocochlear sinuses, all specimens have in common that several canals extend from the hypocochlear sinus anteriorly along the shallow base of the secondary osseous lamina or attachment point of the hearing membrane. Numerous small canals connect the internal aspect of the cochlear labyrinth to the vessels that pass within the secondary osseous lamina base, indicating that those vascular channels might have drained the cochlea itself ( Fig. 1G View Figure 1 ). Graybeal et al. (1989) missed the connections between the cochlea and the venous plexus, possibly because they were too small to be detected in the serial sectioning. Preservational variation might also have influenced the reconstructions by Graybeal et al. (1989); although some specimens are very well preserved, others might suffer from diagenetic alteration and fungal tunnelling. In some specimens, the vascular canals are infilled and difficult to distinguish from the surrounding bone in the µCT scans; this is particularly true for the small vascular canals that open into the cochlear cavity.
Connecting the inferior petrosal sinus and prootic sinus along the dorsal aspect of the cochlear canal are the epicochlear sinuses. All specimens in our sample have between one and three posterior epicochlear sinuses that pass posterior to the cochlear nerve. Only a handful of specimens also have an anterior epicochlear sinus that passes anterior to the cochlear nerve (e.g. Figs 3B View Figure 3 , 5B, D, F, H View Figure 5 ; Supporting Information, Fig. S2B View Figure 2 ). Interestingly, the anterior epicochlear sinus appears to be particularly prevalent in the Pontalun 3 specimens ( Fig. 5 View Figure 5 ). Five of six Pontalun 3 specimens have an anterior epicochlear sinus vs. only 4 of 24 specimens from Pant 2.
Stapes
Here, we describe the first stapes found in association with the petrosal for Morganucodon ( Fig. 7 View Figure 7 ).
The stapes of NHM UK PV M21179 and NHM UK PV M27691 were displaced into and preserved in the cochlear cavity of the petrosal ( Fig. 7A, H View Figure 7 ). We recognized the presence of the stapes only after scanning during the segmentation process and were able to reconstruct the morphology digitally. Unfortunately, the stapes of NHM UK PV M27691 must have fallen out of the petrosal at some point after µCT scanning the specimen; as of August 2023 it is not associated with the specimen anymore and appears to be lost. Both stapes are fragmentary; the stapedial head is missing, and the crura are preserved in part, but the stapedial footplate appears nearly complete ( Fig. 7E, L View Figure 7 ). The stapedial footplate is oval in outline, with curved edges around the periphery (also referred to as bullate stapedial footplate). The dimensions of the footplate are nearly identical in the two specimens, with a length of 0.66 mm and width of 0.48 mm in NHM UK PV M27691 and a length of 0.66 mm and width of 0.47 mm in NHM UK PV M21179 ( Table 3 View Table 3 ). The footplate tapers anteriorly, which is best seen in internal or external view ( Fig. 7B, E, I, L View Figure 7 ). Most of the footplate is relatively flat, neither strongly convex nor concave, but the tapered anterior edge is bent and points externally ( Fig. 7C, F, J, M View Figure 7 ). The bend is best seen in lateral or medial view ( Fig. 7C, F, J, M View Figure 7 ), and, although it might be slightly exaggerated by small cracks in NHM UK PV M27691, the overall shape of the anterior edge is very similar in the two specimens. Part of the posterior and anterior crura are preserved in NHM UK PV M21179 ( Fig. 7C, F View Figure 7 ) and NHM UK PV M27691 ( Fig. 7J, M View Figure 7 ). The posterior crus is nearly circular in cross-section at its base but then flattens towards the top. In NHM UK PV M27691, the most apically preserved portion of the crus is oval, with its long axis twice the length of the short axis (0.16 vs. 0.08 mm). However, the crus is fragmentary, and the preserved cross-section, at most, represents a mid-crus section. The posterior crus is even less well preserved in NHM UK PV M21179 and is not informative on the shape of the crus beyond the base. The posterior crus is placed along the posteromedial edge of the stapedial footplate. The anterior crus is circular in cross-section in NHM UK PV M21179 (with a diameter of 0.10 mm). It is less well preserved in NHM UK PV M27691, where only the posterior edge of the crus is present, and the crus appears rather flat. However, at the base of the footplate is a slightly elevated circular ridge expanding from the crus, which we believe to be representative of the real outline of the crus. The circle is 0.1mm in diameter, which matches the dimension of NHM UK PV M21179. The anterior crus arises along the anterolateral margin of the footplate, slightly offset from the curved anterior margin. The two crura are widely separated and do not merge towards the stapedial footplate, and they appear to be coursing in parallel. Those in NHM UK PV M21179 are angled slightly posteriorly.
Kermack et al. (1981) previously attributed several isolated and fragmentary stapes to Morganucodon , which were presumably all recovered through screen washing of the Pant 2 fissure in 1955. It is noteworthy that the two new stapes described here, NHM UK PV M21179 and NHM UK PV M27691, come from separate fissure fillings (Pant 2 1955 and Pontalun 3), are both associated with the petrosal of Morganucodon , and are very similar in morphology and dimensions, but they differ substantially from the isolated stapes described by Kermack et al. (1981); see Discussion.
No known copyright restrictions apply. See Agosti, D., Egloff, W., 2009. Taxonomic information exchange and copyright: the Plazi approach. BMC Research Notes 2009, 2:53 for further explanation.