We thank Anderson et al. [1] for engaging in our paper on hearing in salamanders [2]. We have written our paper primarily as experimental physiologists basing hypotheses on recent animals and with no research expertise in palaeontology. Therefore, discussions with palaeontologists are especially welcome. In many ways, palaeontology should constrain our hypotheses and the finding of a fossil tympanate urodele, for example, could change our views, as could more information on the ear region of the early tetrapods. It is also true that comparative physiology has erred in its ways previously-by comparing disparate living exemplar species and inferring that such a lineup reflects evolutionary history (i.e. naming recent species as 'lower', 'primitive' or 'ancestral'). A stellar example is the older view of the tympanate ear originating in the tetrapod ancestors with the recent groups-anurans, lizards, archosaurs and mammals exemplifying different stages in the evolution of the middle ear. This view has been abandoned, based on the thorough palaeontological investigations by Clack [3,4] and others showing independent origins of the tympanate ear in all the major groups.When this is said, we find the criticism to be misdirected concerning two major points, the first being the use of model animals. Anderson et al. state that 'Salamanders are not evolutionary intermediates' and with that we fully agree; we certainly do not regard salamanders or any other living creatures as evolutionary intermediates, 'missing links' or even as 'exemplar species'. Furthermore, to designate any living species as 'exemplar' seems to us to revert to the practices described above-any living species will only be superficially similar to the ancient species and may, in the best case, illustrate some aspects of its physiology. Rather, the solid comparative approach would be to investigate many recent species (i.e. to compare the physiology of salamanders, caecilians and earless frogs and infer ancestral physiology from common features). Another strategy in comparative physiology is to use living organisms to investigate biological solutions to problems common to fossils and (unrelated) recent animals-for example, to use large mammals as models for large dinosaurs. Anderson et al. claim that salamanders are an inappropriate model, because 'there is strong evidence that salamanders have secondarily lost a tympanic ear' and this 'dramatically changes the interpretation of the results of' our study [2]. Now, it is probable that salamanders are secondarily reduced, also in their middle ear configuration, but it is by no means certain that their ancestors were tympanate. This is most likely under the temnospondyl hypothesis; under the lepospondyl hypothesis, the divergence-dating by Pyron [5] suggests a caudate -anuran split at 292 Ma, long before the atympanate proanuran Triadobatrachus, and so making a tympanate ancestry of salamanders much less likely. However, even if the salamanders are secondarily reduced from tympanate ancestors they could sti...