A commentary was recently published as a piece of Critiques & Debates at National Science Review (NSR) (hereinafter referred to as "Commentary"), which raised issues about the article (referred to as "Original") published in Nature. The title of the "Original" is "A monotreme-like auditory apparatus in a Middle Jurassic haramiyidan"¹. The Commentary is titled "Monotreme middle ear is not primitive for Mammalia" ². The authors of the Original gave a response to the Commentary, referred to as "Response" ³. 

Based on CT data, the Original work segmented and reconstructed the malleus and incus in a new specimen of Vilevolodon⁴ and interpreted that the middle ear is similar to that of monotremes in having a flat incus and fully-overlapped incus- malleus joint. Five features related to the monotreme middle ear have been optimized in the phylogeny to reach to the conclusion that such a fully-overlapped malleus-incus joint is the primitive condition of the mammalian middle ear. 

The Review first questioned the identification of the ear bones for several reasons: 1) The incus-malleus joint reported in the Original is significantly different from that in the holotype specimen of Vilevolodon⁴. 2) The Original mentioned that the reported specimen is similar to the multituberculate Jeholbaatar⁵ in the malleus-incus morphology, although noting that new evidence has shown the "incus" of Jeholbaatar to be the raised base of the malleus body, not the incus⁶. 3) The Original reinterpreted the ectotympanic of the euharamihyidan Qishou⁷ as two overlapped bones, the malleus and incus, and used it to support their identification. However, using images of high-resolution CT slices (Figure 1k-m) the Review convincingly demonstrated that the element in question is only one thin bone, rejecting the reinterpretation of the Original. 4) Among known euharamiyidans, the undisputed incus is only known in Arboroharamiya, which has a convex and spherical body with stapedial and lenticular processes, therefore similar to the incus of therians⁷. Thus, the ear ossicles identified in the Original are not only different from the holotype specimen, but also very different from those of closely related euharamiyidans; instead, they are very similar to those of the distantly related monotremes. 5) The length of the platelet “incus” in the Original is around one millimeter and its thickness is much smaller; thus, whether or not the CT scan with the resolution of 0.0327 mm used in the Original can detect the suture between the “incus” and malleus is skeptical. 

The Review argued that, as in Qishou and Jeholbaatar, the so-called incus of Vilevolodon identified in the Original, may well be a raised part of another bony element; this also explains why the "incus" and malleus from both sides are still tightly stacked in their "anatomical position" after they were distantly displaced during the preservation. The Review also noted that in the 509 characters used in the phylogenetic analysis of the Original, five characters related to the monotreme middle ear with fully overlapping incus- malleus have been emphasized. Among the 130 taxa used for the analysis, only two extant monotremes and three euharamiyidans can be fully coded with the five characters. Although the three euharamiyidans possess highly inconsistent ossicular conditions, as currently interpreted, they were nonetheless coded identical to those of the monotremes. For Jeholbaatar that has the ossicles preserved, the five characters were coded as "?". Using an unstated method, the five characters were "optimized" in the strict consensus tree, which let the Original to draw the conclusion that the monotreme type middle ear with the fully overlapped incus-malleus joint is primitive for mammals, from which evolved other types of mammalian middle ears.  

The Review believed that the Original has overlooked an alternative hypothesis: the braced hinge joint⁶ (=partially overlapping joint) can be interpreted as the primitive type of the mammalian middle ear, as mapped on the identical phylogenetic tree of the Original (Fig. 1o). Under this hypothesis, during the evolution from mammals-like reptiles to mammals, the ear ossicles do not make abrupt changes in morphology and positional relationship and there is no need for simultaneous detachment of the ossicles from the dentary (it allows a transitional stage with the connection of the ossified Meckel’s cartilage). There is also no need to regain the ossified Meckel’s cartilage in adult individuals within mammals, nor is there a reversal in which the separated hearing and chewing functions were reassociated. This evolutionary process represented by the alternative is more consistent with fossil and developmental biological evidence than the one in the Original. Moreover, within mammals, the evolution from the braced hinge joint to the fully overlapped joint of monotremes and the saddle-shaped joint of therians requires only three independent evolutionary steps, which is more parsimonious than the hypothesis in the Original that needs at least four evolutionary steps (Fig. 1n). 

The Response acknowledged some issues raised by the Review, modified some character coding for relevant taxa used for the phylogenetic analysis, and obtained the new optimized result that abandoned the conclusion reached in the Original; it argued instead that the primitive condition of the mammalian middle ear remains uncertain. The Response did not respond to the suspicion on the specimen identification. Given that the digital data are not in public domain, the suspicion remains to be verified. 

Fig. 1. Evolutionary hypotheses of the incus-malleus joint of the mammalian middle ear and related characters. n, Hypothesis from the Original work. o, Hypothesis presented in the Commentary. The figure is from ref. 2. 

References： 

1. Wang, J. et al. A monotreme-like auditory apparatus in a Middle Jurassic haramiyidan. Nature 590, 279-283 (2021). https://www.nature.com/articles/s41586-020-03137-z 

2. Meng, J. and Mao, F.Y. Monotreme middle ear is not primitive for Mammalia. Natl. Sci Rev, (2021). https://doi.org/10.1093/nsr/nwab131) 

3. Wible, J. et al. Response to “Monotreme middle ear is not primitive for Mammalia.” Natl. Sci Rev, https://doi.org/10.1093/nsr/nwab132.  

4. Luo, Z.-X. et al. New evidence for mammaliaform ear evolution and feeding adaptation in a Jurassic ecosystem. Nature 548, 326–329 (2017). https://www.nature.com/articles/nature23483 

5. Wang, H.-B., Meng, J. & Wang, Y.-Q. Cretaceous fossil reveals a new pattern in mammalian middle ear evolution. Nature 576, 102–105 (2019). https://www.nature.com/articles/s41586-019-1792-0 

6. Mao, F.-Y. et al. Exploring ancestral phenotypes and evolutionary development of the mammalian middle ear based on Early Cretaceous Jehol mammals. Natl Sci Rev, 8, nwaa188 (2021). https://academic.oup.com/nsr/article/8/5/nwaa188/5896965?login=true 

7. Meng, J. et al. A comparative study on auditory and hyoid bones of Jurassic euharamiyidans and contrasting evidence for mammalian middle ear evolution. J Anat 236, 50-71 (2020). https://onlinelibrary.wiley.com/doi/abs/10.1111/joa.13083