Is there an exemplar taxon for modelling the evolution of early tetrapod hearing?

Christensen et al. recently published a study of hearing in neotenic and experimentally metamorphosed axolotls (Ambystoma mexicanum) and a larval and adult tiger salamander (A. tigrinum), which contributes greatly to our understanding of salamander sound perception in water and air [1]. They demonstrate that premetamorphic, atympanic aquatic salamanders are capable of perceiving high frequency airborne sounds similar to terrestrial adults, demonstrating an interesting capacity for an atympanic ear to function for sound perception in a variety of media. This important research, in combination with other recent studies on the sound perception capabilities of lungfish [2], is helping to better clarify issues related to the water-to-land transition. However, the implication of their experimental design demands a consideration of the evolutionary history of their selected exemplar taxa so these valuable data can be interpreted in a broader context. Our criticisms focus on two major points: the use of salamanders as a proxy for hearing ability intermediate between aquatic and terrestrial vertebrates, and the use of microsaurs as exemplars of the early tetrapod condition.

The currently accepted hypothesis for the origin of a tympanic ear is as a derivation from the spiracle of sarcopterygian fish [3]. In ‘fish’ the spiracle is associated with gill breathing: water passes through the spiracular notch in the rear of the skull, past the braincase and the bracing hyomandibula, to the oropharynx and gill arches [4]. With the loss of gills, the persisting spiracular lumen is co-opted into a middle ear chamber, and the cranial support role of the hyomandibula (now termed stapes) is eliminated, leaving this bone to extend freely into the middle ear chamber (figure 1). In derived tetrapods, the spiracular opening is expanded and covered by a membrane (the tympanum) that collects sound. Vibrations are transmitted to the inner ear via the stapes, which becomes more delicate and better able to propagate fine vibrations [7–10].

Figure 1.

Phylogenetic context of middle ear evolution in early tetrapods. Phylogeny modified from [5,6]. TE, tympanic ear. Illustrations not to scale.

1. Salamanders are not evolutionary intermediates

Christensen et al. state ([1], p. 2), ‘Equally important, [the urodele auditory system] can be regarded as a potential model for the intermediate evolutionary developmental stage’ between lungfish and frogs. However, there is strong evidence that salamanders have secondarily lost a tympanic ear, which dramatically changes the interpretation of the results of this study. Most palaeontologists agree that modern amphibians (Lissamphibia) are a monophyletic group with origins among a group of temnospondyls, the Amphibamidae [5,11–17] (but see [18] for an alternative view). Amphibamids are considered to have had a tympanic hearing system identical to that of frogs [9,10,19] (figure 1). This hearing system was probably present in the first temnospondyls [20], dating well into the Carboniferous. Interestingly, a middle ear similar to salamanders is also seen in some secondarily atympanic frogs [21].

This secondary loss of tympanic hearing is also evident physiologically. All lissamphibians possess two epithelia: one associated with low-frequency environmental stimulations and one with high-frequency aerial sound. Salamanders show a phylogenetically correlated reduction in innervation of the epithelium associated with high frequency sound perception [22], indicating the presence of a tympanic hearing system in their evolutionary past. A similar trend is found in the atympanic caecilians [23,24]. This point is critical because Christensen et al. tested sensitivity to sound vibrations at the auditory nerve, which directly reflects the innervation of the sensory epithelia detecting the stimulus. Because salamanders have remnants of the greater innervation of the fully tympanic ear, it is expected that they can perceive sound better than animals without any evolutionary history of terrestrial hearing. The fact that secondarily aquatic salamanders, such as axolotl, do not differ in their ability to hear aerial sound from terrestrial salamanders reflects the conservation of a degree of terrestrial hearing capability. Even under the heterodox ‘lepospondyl hypothesis’ [18] this loss is recognized.

2. Microsaurs are not representative of early tetrapods

Christensen et al. [1] assert that microsaurs are morphological exemplars for the early tetrapod condition (a rationale for this was not provided) and that similarities of the middle ear can be used to associate the functional data derived from salamanders with the early evolution of frog-like hearing. This is problematic for two reasons. First, phylogenetically microsaurs are more closely related to the amniote crown group than any early tetrapod [5,11,24,25]. They are derived in both lacking a spiracular embayment and in possessing a short, robust stapes, unlike all taxa that span the water-to-land transition up to, and including, frogs (figure 1). Rather than representing a transition between aquatic and terrestrial living, microsaurs are highly specialized terrestrial species adapted to burrowing.

Second, whereas Christensen et al. are correct in recognizing superficial similarities between salamanders and microsaurs, including the lack of an otic notch and stapes with broad footplates and short columellae, microsaurs differ in several important anatomical details [26–28]. For example, microsaurs have the stapes completely filling the fenestra vestibularis, whereas salamanders also have an operculum [22]. All lissamphibians [24,29] and some temnospondyls [19] possess a posterior pressure relief pathway via the perilymphatic foramen, whereas microsaurs possess an ossified crista interfenestralis that prevents this posterior perilymphatic flow, strongly suggesting they possess a more amniote-like anterior pattern [26,30].

3. Conclusion and recommendations

Experimental studies of modern taxa have the potential to inform our understanding of major transitions in evolutionary history otherwise preserved only in the fossil record, but the selection of exemplary taxa for such studies is non-trivial to their applicability. Researchers must consider the evolutionary histories of taxa to ensure that the model organisms are appropriate analogues so that conclusions can be drawn with confidence. The middle ear of Polypterus is more anatomically similar to that seen in stem tetrapods and lungfish near the tetrapodomorph divergence and could be an appropriate analogue to test in future studies of this question.

Given the above, we assert that the conclusion that ‘early tetrapods also may have been able to detect aerial sound before the appearance of the tympanic middle ear’ ([1], p. 8) drawn from observations of salamanders and microsaurs is unfounded. Instead, the results of this study provide interesting insights into the strength of the conservation of an auditory apparatus adapted to terrestrial hearing in a group of aquatic salamanders.

Authors' contributions

All authors contributed equally.

Competing interests

We have no competing interests.

Funding

This study was supported by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada to J.S.A.