Attachment of the ectotympanic bone to the otic region and incorporation of the malleus-incus complex in the ossicular chain are two key events in the evolution of the Mammalian middle ear; the former holds the tympanic membrane and the latter forms a functionally semi-independent unit that resulted in an increased bandwidth of hearing, particularly of high-frequency sounds. Because the incus (quadrate) is recessed at the periotic in the cranium, it served as the anchor point for the gradual evolutionary shift of the malleus (articular), the gonial (prearticular), and the ectotympanic (angular) away from the dentary bone and relocated at the base of the cranium. Thus, in addition to understanding the homology of the auditory bones, as reviewed by several authors, a focused subject in recent paleontological and developmental studies is how these jaw bones were detached from the dentary, which primarily concerns on the role played by the Meckel’s cartilage and the developmental genetic mechanisms regulating these processes. In contrast, the ancestral phenotypes of the Mammalian middle ear remain little known due to rareness of fossils. The middle ear of the American Opossum, Didelphis, was considered as the ancestral ear type for Therians, but the saddle-shaped incudomalleolar joint is already specialized for Mammals as a whole. Similarly, the abutting contact of the malleus and incus in Monotremes is also peculiar, even though a similar pattern was claimed to be present in the Mesozoic Multituberculate Jeholbaatar and the Eutriconodontan Yanoconodon. A morphological gap exists between the primary quadroarticular synovial jaw joint of non-Mammalian Cynodonts and the middle ear of extant Mammals.
In a paper published in the journal National Science Review on 25 August 2020, Fangyuan Mao of the Key Laboratory of Evolutionary Systematics of Vertebrates at the Institute of Vertebrate Paleontology and Paleoanthropology of the Chinese Academy of Sciences, the Center for Excellence in Life and Paleoenvironment, and the Division of Paleontology at the American Museum of Natural History, Cunyu Liu of the Beipiao Pterosaur Museum of China, Morgan Hill Chase and Andrew Smith of the Microscopy and Imaging Facility at the American Museum of Natural History, and Jin Meng, also of the Division of Paleontology at the American Museum of Natural History, and of Earth and Environmental Sciences at the City University of New York, report a new Multituberculate Mammal from the Early Cretaceous Jehol Biota.
The holotype specimen of the new species preserves the ectotympanic, malleus-surangular unit, incus, and stapes, together with hyoid bones. The 3D morphologies of these elements were revealed by high-resolution CT-scan, which provide the first detailed structures of these auditory bones in Multituberculates. Along with those of Eutriconodontan Liaoconodon and stem Therian Origolestes, by far the most unequivocal middle ears known in Mesozoic Mammals, the ancestral phenotypes of the Mammalian middle ear can be explored. For comparison, we also present high-resolution 3D reconstructions of the middle ear of the Short-beaked Echidna, Tachyglossus, the Opossum, Didelphis and the Hedgehog, Erinaceus, as representatives of extant Monotremes and Therians (Marsupials and Placentals). Although the latter have been known for many decades, the 3D morphology of the ossicular chain in anatomical articulation has not been portraited the way Mao et al. present in this study.
It has been a common view that the Definitive Mammalian Middle Ear evolved independently in Monotremes, Therians, and Multituberculates. Although they differ considerably in morphology, the same homologous elements make up the middle ear in Monotremes and Therians, respectively (the stapes, incus, malleus, and ectotympanic). The independent origin of the middle ear in Mammals thus only refers to the process and perhaps timing of detachment of the middle ear bones from the dentary. Under this assumption we further specify osteological changes of the auditory bones after their detachment in three lineages of Mesozoic Mammals, of which the new Multituberculate is of particular interest because it provides new morphological evidence of the Definitive Mammalian Middle Ear from an extinct group that is distantly related to Monotremes and Therians. Our goal is to focus on the transitional morphology that could represent the ancestral phenotypes for Mammalian middle ears so that the morphological gap between the mandibular middle ear of non-Mammalian Cynodonts and the Definitive Mammalian Middle Ear in extant Mammals can be bridged. It is also important to have an additional example that shows convergent evolution, thus plasticity in evolutionary development, of the Definitive Mammalian Middle Ear in Mammals. Moreover, these ancestral phenotypes provide direct evidence to address the relation of development and evolution in the Mammalian middle ear. Mao et al. show that many, but not all, of the primitive features are recapitulated in embryological morphogenesis of the middle ear in extant Mammals and that future researches in palaeontology and development biology are needed to answer questions raised in this study.
The new species is placed in the genus Sinobaatar and given the specific name pani, in honour of the Junyi Pan, the collector of specimen. The species is described from a single disarticulated skeleton from the Early Cretaceous (Aptian) Jiufotang Formation of Liaoning Province, China. This is in the collection of the Beipiao Pterosaur Museum of China.
Sinobaatar pani differs from other multituberculates in having the following combination of features: a gracile skeleton with a long tail (possibly arboreal); a strong zygomatic process of the maxilla; a single infraorbital foramen; tooth formula 3-0-5-2/1?-0?-?-2; tooth cusp formula P1-3 (1:2), P4 (3:4), P5 (1:4:2?), M1 (3:4), M2 (Ri:2:3)/m1 (3:2), and m2 (2:2); I2 robust with three cusps (a main mesial one and two minor distal ones); P1-3 cusps showing the trend of coalescing; P3 greatly reduced in size; distal cusp of P4 and mesial one of P5 the highest cusps and forming the peak (in lateral view) in the middle of the cheek tooth row; P5 proportionally not so enlarged relative to P4 and M1; molar cusps slim; distolingual cusp of M1 transversely orientated (mesiodistally short) with labial and lingual ridges; M2 considerably shorter than M1; M2 cusps increasing size distally.
Sinobaatar pani is represented by a disarticulated skeleton and
split skull with some teeth. The hyloid apparatus is similar to some
Eutherians in having seven rod-like elements (the basihyal, ceratohyal,
stylohyal, and thyrohyal) in which the stylohyal is long and slender.
The epihyal was not preserved; it could be fused with either the
ceratohyal or the stylohyal. The auditory bones include both pairs of
the mallei and ectotympanic, one stapes, and one incus.
The ectotympanic is sickle-shaped, consisting of the curved ventral and a relatively straight dorsal limbs whose ends are separated by a wide tympanic notch. The ventral limb, presumably homologous to the reflected lamina of the angular, is better developed than that of Liaoconodon and Origolestes, but is less so than the horseshoe-shaped ectotympanic in extant Mammals. The dorsal limb is plate-like, uncommon in Mammals but reminiscent of the plate-like ectotympanic of Arboroharamiya; its lateral side bears an extensive contact facet for the malleus, similar to that of Monotremes. On the medial side, the crista tympanica is weak on the dorsal limb but distinct in the ventral one so that the tympanic sulcus is shallow on the dorsal limb but deep in the ventral one. The sulcus accommodates the annulus fibrosus, a thickened circumferential rim of the pars tensa of the tympanic membrane that attaches the membrane to the sulcus. The ectotympanic and the malleus form an incomplete oval frame for supporting the tympanic membrane, which gives an estimated area of 4.24 mm² for the membrane.
Mao et al. interpret that the bodies of the malleus and surangular are fused but their anterior processes are separated by a suture. The combined malleus-surangular unit is robust compared to the malleus of extant mammals. The surangular part is posterolateral to the malleus, with its anterior process inserting between the ectotympanic and the malleus and gradually tapering anteriorly. On the medial side of the malleus, a groove leads to the foramen for the chorda tympani nerve that pierces the anterior process, echoing the view that the anterior process is homologous to the prearticular in non-Mammalian Cynodonts. The canal does not penetrate the anterior process of the surangular but exits dorsally from a slit between the anterior processes of the surangular and malleus; this serves as an evidence for the identification of the surangular. The articular facet for the incus is a crescent concavity at the posteromedial end of the malleus-surangular unit; it is primarily within the malleus and dorsally bounded by the surangular, similar to that in Liaoconodon and Origolestes. A narrow bony lip extends along the lateral edge of the facet, which braces the articular facet. Between the articular facet and the manubrial base is the neck; its posterior border is shallowly concave so that the posterior border of the malleus shows a double-concavities, similar to that of Didelphis and Monotremes. The manubrial base thickens notably so that a step-like boundary is formed between it and the neck and the manubrium. Because of this configuration, the manubrial base on the lateral side was misidentified as the incus in Jeholbaatar. The manubrium is a thin and short prong, parallel to the anterior process and tapering distally.
The incus is proportionally small and quite flat but differs from the platelet-like incus of Monotremes. It has a body with a convex articular surface for the malleus, a short process (broken), and a long process that has an angle with the body. The distal end of the long process flares to give the shape of a lenticular process, but the latter does not show a bending from the long process. The stapes has an oval footplate with an estimated area of 0.215 mm²; it is convex medially toward the fenestra vestibuli and concave on the lateral surface, similar to the Jurassic Multituberculate Pseudobolodon and the stem Therian Origolestes. Judged from the crushed segments, the stapes is most likely bicrural, with the anterior crus being near the center and the posterior one at the edge of the footplate, as in Pseudobolodon and Origolestes. There should be a sizable stapedial foramen, but a meaningful reconstruction of the complete stapes is difficult.
Based on the frame formed by the ectotympanic and malleus, the estimated area of the tympanic membrane (4.24 mm²) gives an effective area of 2.827 mm²; the transformer ratio between the effective area and the stapedial footplate (0.215 mm²) is 13.15, larger than that of Morganucodon (10.71), The tympanic membrane to the stapedial footplate area ratio is 19.7, similar to some neonatal Mammals but larger than that of Morganucodon (16.0); this ratio increases during ontogeny in extant Mammals. The transformer ratio of the levers is difficult to determine because of the displaced ossicles, but development of the manubrial neck and the stapedial process of the incus would affect the ratio. The transversely narrow ectotympanic is tightly bound with the malleus so that both (plus the surangular in the fossil forms) will vibrate as a unit. Such a unit would have a relatively heavy mass so that these Mesozoic Animals could hear only relatively lower frequency airborne sounds in a narrower range of frequency, as in Monotremes. Interestingly, the inner ear of all these forms has only developed a curved cochlea, perhaps a receiving system (inner ear) that matches the delivering system (middle ear).
Recent palaeontological and developmental studies have converged on detachment of the auditory bones from the dentary during the evolution and development of the Mammalian middle ear. With the new evidence, Mao et al. further argue that auditory bone features also reflect the relation between development and evolution and largely endorse that 'portions of the ossicles that are phylogenetically older develop earlier than portions representing more recent evolutionary inventions'. These evolutionary changes are best preserved in the 3D auditory bones from Mesozoic representatives of three major Mammalian clades: the Eutriconodontan Liaoconodon, the stem Therian Origolestes, and the Multituberculate Sinobaatar pani. Mao et al. view these forms as representing ancestral phenotypes of the mammalian middle ear at different evolutionary stages. While the auditory bones already detached from the dentary in the three phenotypes, the transitional middle ear of Liaoconodon is most primitive in that the malleus and ectotympanic have long anterior processes that are still in contact with the ossified Meckel’s cartilage; thus, hearing and chewing functions were not completely separated. Origolestes is more derived in having lost the bony contact of the auditory bones to the ossified Meckel’s cartilage so that hearing and chewing functions were decoupled. The definitive Mammalian middle ear of Sinobaatar pani is further derived, in having some features similar to those of extant Mammals but still more primitive than the latter in several aspects; it would not be a surprise if a similar middle ear is found in a basal Therian in the phylogenetic tree between Origolestes and Didelphis or in a species basal to Monotremes.
In Mammalian ontogeny the developing ectotympanic starts as a tri-pronged structure with an anterior limb; the rest of the ectotympanic forms a partial circle. This configuration is similar to the ectotympanic of Liaoconodon; the latter is the closest approximation of the angular bone of Morganucodon. In a sequential way, the anterior limb is resorbed and the dorsal limb (relatively straight) and the ventral one (more curved) gradually elongate and become more complete as a horseshoe-shaped frame with a small tympanic notch in later stages of ossicular development. This trend is well reflected in the gradual evolutionary changes of the ectotympanic in Origolestes, Sinobaatar, and the Cretaceous Eutherians, such as Uchkudukodon and Ambolestes. The developing ectotympanic is transversely narrow at early stage, but gradually expand laterally to form the bulla or external auditory meatus; the early developmental stage of extant Mammals, again, is echoed by the evolutionary pattern of the fossils.
It has been shown that a part of the manubrial base of the malleus, either termed the orbicular apophysis or the processus brevis, is homologous to the retroarticular process of the articular because it arises from the second pharyngeal arch; this finding lends support to the view that the manubrium is a neomorph. The manubrium has been interpreted as being absent in Liaoconodon but present in Origolestes. In the light of Sinobaatar pani, it is most probable that the manubrial neck and a true manubrium were not yet developed in both Liaoconodon and Origolestes but did evolve in Sinobaatar pani and Uchkudukodon. Origolestes has a short process that tapers anteriorly, similar to Liaoconodon and differing from the indentation of the manubrial neck and expansion of the manubrial base in Sinobaatar pani. What previously identified as the manubrium in Origolestes appears to be the ventral extension of the malleus. Mao et al.'s data suggest that the manubrium probably evolved along with formation of the manubrial neck; both would increase the lever ratio for sound transmit of the ossicular chain. The formation of the manubrium is also coordinated with development of the ectotympanic, regulated by various genes and developmental mechanisms. Mao et al. infer that the more complete ectotympanic in Sinobaatar pani may have played a role in the evolutionary development of the manubrium. These auditory bones show evolutionary and developmental consistency in detailed morphologies. For instance, in the primitive middle ear of Liaoconodon the ectotympanic with a long anterior limb but poorly developed dorsal and ventral ones is associated with the malleus that lacks the manubrium. In the relatively derived middle ear of Sinobaatar pani, however, the ectotympanic without the anterior limb but with better developed dorsal and ventral ones is associated with the malleus that has developed the neck, base and manubrium. Morphologically, the auditory bones in Origolestes seem to be intermediate between those of Liaoconodon and Sinobaatar pani. These configurations are comparable to the morphogenesis during the embryonical development of the middle ear in extant Mammals, such as that the neck and manubrium are developed and ossified in later stages of the malleus and the ectotympanic develops from a tri-pronged form to a horseshoe-shaped frame.
The malleus-surangular unit of Sinobaatar pani adds to the increasing evidence that the surangular did not disappear abruptly during the evolution of the Mammalian middle ear but had persisted in basal mammals as a primitive character. In these basal forms the surangular occupies a similar position in relation to the malleus and contributes to the articular facet for the incus. Such a pattern is comparable to the quadroarticular articulation of Morganucodon in which the surangular forms a considerable part of the articular facet. Its disappearance in extant Mammals may be attributed to the evolutionary reduction of the ear ossicle mass for efficient hearing of high frequency sounds, but whether the surangular survived as a remnant in embryonic stage of extant Mammals remains unclear.
The incudomalleolar hinge joint differs in shape and position from those of Monotremes and Therians. Nonetheless, it retains the convex-to-concave articulation between the incus (quadrate) and malleus (articular), the basic pattern of the primary synovial jaw joint that persisted throughout the evolutionary radiation of the Mammalian malleus-incus complex, except for Monotremes. Functionally, because the lateral lip braced the articular facet, airborne sounds coming from the lateral side of the tympanic membrane could be efficiently transmitted to the incus and stapes; at the same time, the crescent facet allows some rotation of the incus relative to the malleus. This joint appears to be a conservative feature present in at least the three Mesozoic Mammals discussed in Mao et al.'s study but has not been documented in known embryonic morphogenesis of extant Mammals. Because the incus anchors at the periotic in the cranium, roughly retaining the position of its precursor (quadrate) in non-Mammalian Cynodonts; the gradual shift of the malleus (articular), gonial (prearticular), and ectotympanic (angular) was basically a rearward retreating from the dentary during the evolution of Mammals; thus, changes and reorientation of the auditory bones were more significant at their anterior and ventral side than at the incudomalleolar articulation. In contrast to the conservative braced hinge joint, these changes suggest that developmental heterochrony has played a role in the evolutionary development of the Mammalian middle ear. Given its shape and composition, the braced hinge joint is derivable from the quadroarticular jaw articulation in non-Mammalian Cynodont. On the other hand, the saddle-shape joint of Therians could be derived from the braced hinge joint by shift of the incus to the caudal side of the malleus. Still, it remains unclear how the abutting condition in monotremes evolved because the phylogenetic components and middle ear fossils in the lineage toward Monotremes are poorly known. It is possible that the monotreme condition may have also derived independently from a similar braced hinge pattern by migration of the incus to the dorsal side of the malleus. Developmental study may prove to be indicative for this issue, given that new observations continue to become available, such as that the ectotympanic and malleus of the echidna are originally in a vertical position in early ontogeny, similar to Therians, before flipping to the horizontal condition in adult.
With the assumption that the Definitive Mammalian Middle Ear evolved independently in Monotremes, Therians, and Multituberculates, there should be no common ancestral phenotype of the middle ear for these clades. However, the auditory bones of Liaoconodon, Origolestes and Sinobaatar pani display some shared primitive features, such as the braced hinge incudomalleolar joint, an incomplete ectotympanic, and presence of the surangular, that potentially illustrate the ancestral phenotype of the Mammalian middle ear in each lineage. These forms narrow the morphological gap between the mandibular middle ear of non-mammalian cynodonts and the Definitive Mammalian Middle Ear of extant Mammals. The differences of these phenotypes, such as the degree of development of the ectotympanic (its anterior, dorsal and ventral limbs), the morphology of the malleus (development, or not, of the manubrial neck, base, and the manubrium), and the fusion or separation of the malleus and the surangular, are interpreted as representing various evolutionary stages in different lineages. These phenotypes show comparable pattern with the morphogenesis of the middle ear in extant Mammals and to some degree support the relation between evolution and development. The derived features of Sinobaatar pani, such as development of the manubrium, must be interpreted as a result of convergent evolution to those of extant Mammals, which suggests plasticity in the evolutionary development of the middle ear; it illustrates that evolutionary ‘experiments’ for better hearing had taken place in various clades during Mammalian evolution. Future palaeontological and developmental studies are needed to test the issues raised by the discoveries of the Mesozoic Mammals.
See also...
Online courses in Palaeontology.
Follow Sciency Thoughts on Facebook.
Follow Sciency Thoughts on Twitter.