- Email: [email protected]
The Origin of High-Frequency Hearing in Whales
Report
The Origin of High-Frequency Hearing in Whales Highlights
Authors
d
An ancient toothed whale is described, which possesses a well-preserved inner ear
d
Features associated with ultrasonic hearing are preserved in the ear of this whale
Morgan Churchill, Manuel Martinez-Caceres, Christian de Muizon, Jessica Mnieckowski, Jonathan H. Geisler
d
Ultrasonic hearing evolved with echolocation in the first toothed whales
Correspondence
d
Hearing at higher frequencies began in the ancestors of toothed whales
[email protected]
In Brief Churchill et al. describe a new species of early toothed whale (Echovenator sandersi), which possesses one of the oldest known well-preserved inner ears. Comparison with other fossil whales reveals that the ultrasonic hearing evolved at the base of the toothed whale radiation, and high-frequency hearing evolved before echolocation.
Churchill et al., 2016, Current Biology 26, 1–6 August 22, 2016 ª 2016 Elsevier Ltd. http://dx.doi.org/10.1016/j.cub.2016.06.004
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
Current Biology
Report The Origin of High-Frequency Hearing in Whales Morgan Churchill,1,* Manuel Martinez-Caceres,2 Christian de Muizon,2 Jessica Mnieckowski,3 and Jonathan H. Geisler1 1Department
of Anatomy, College of Osteopathic Medicine, New York Institute of Technology, Old Westbury, NY 11568, USA of Earth History, National Museum of Natural History, Paris 75005, France 3Ecological Planning Group, Savannah, GA 31401, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cub.2016.06.004 2Department
SUMMARY
Odontocetes (toothed whales) rely upon echoes of their own vocalizations to navigate and find prey underwater [1]. This sensory adaptation, known as echolocation, operates most effectively when using high frequencies, and odontocetes are rivaled only by bats in their ability to perceive ultrasonic sound greater than 100 kHz [2]. Although features indicative of ultrasonic hearing are present in the oldest known odontocetes [3], the significance of this finding is limited by the methods employed and taxa sampled. In this report, we describe a new xenorophid whale (Echovenator sandersi, gen. et sp. nov.) from the Oligocene of South Carolina that, as a member of the most basal clade of odontocetes, sheds considerable light on the evolution of ultrasonic hearing. By placing high-resolution CT data from Echovenator sandersi, 2 hippos, and 23 fossil and extant whales in a phylogenetic context, we conclude that ultrasonic hearing, albeit in a less specialized form, evolved at the base of the odontocete radiation. Contrary to the hypothesis that odontocetes evolved from low-frequency specialists [4], we find evidence that stem cetaceans, the archaeocetes, were more sensitive to high-frequency sound than their terrestrial ancestors. This indicates that selection for high-frequency hearing predates the emergence of Odontoceti and the evolution of echolocation. RESULTS Description Systematics Order Cetacea Brisson 1762. Suborder Odontoceti Flower 1867. Family Xenorophidae Uhen 2008. Echovenator sandersi gen. et sp. nov. Etymology From Latin for ‘‘echo hunter,’’ referring to its use of echolocation while foraging. Species name is in honor of Albert Sanders, former curator of The Charleston Museum, for his contributions to our knowledge of Oligocene Cetacea. Holotype Georgia Southern Museum (GSM) 1098, a nearly complete skull with 39 teeth, mandible, and atlas, presumed to represent a sin-
gle individual (Figures 1 and 2; see also Figures S1 and S2, Table S1, and Supplemental Experimental Procedures, part I). Locality and Age Drainage ditch associated with Limehouse Branch Creek, Berkeley County, South Carolina. Approximate coordinates are: N 33 010 5300 ; W 80 060 0600 . Basal bed of the Chandler Bridge Formation, late Oligocene (ca. 24–27 Ma) in age [5]. This bed is a probable equivalent of the marine or marginal marine lithofacies, which was deposited in a nearshore marine or restricted bay environment [5]. Diagnosis Echovenator has the following synapomorphies of Xenorophidae: premaxilla underlies the ascending process of the maxilla, a frontal window that exposes the maxilla and premaxilla, and an elongate ventrolateral tuberosity of the periotic [6, 7]. Echovenator differs from all xenorophids in having fused, or partially fused, fronto-nasal and maxillo-premaxillary sutures and in having a paranaris fossa; it differs from Xenorophus sloanii and Albertocetus meffordorum in having a postnarial fossa and a depressed internasal suture; and it differs from Cotylocara macei in having an undivided postnarial fossa and in having larger dorsal exposures of parietals. Phylogenetic Analysis Our phylogenetic analysis recovered a single most parsimonious tree of 41201.26 steps (Figures 3 and S3). Within this phylogeny, Echovenator is the sister taxon of a clade comprising Cotylocara and an undescribed xenorophid in The Charleston Museum (ChM) PV2758; both taxa are deeply embedded within Xenorophidae. Xenorophidae was found to be comprised solely of taxa, both described and undescribed, from the Oligocene of the southeastern United States. Inner Ear Morphology Our study found that Echovenator shares many features of the inner ear associated with ultrasonic hearing (Table S2; Figure S4). The secondary bony lamina of the cochlea (SLA) is 72.21% of the total length of the cochlear canal, on the low end of the range exhibited by extant odontocetes (71%–92%), but far above that seen in other mammals [8]. The laminar gap at the first quarter turn, related to size of the basilar membrane, is 0.20 mm. This gap is far smaller than that measured for any non-odontocete taxa in this study (i.e., 0.38–0.90) but larger than most odontocetes examined, although similar in size to ziphiid whales (0.21–0.22). The basal ratio (height of cochlea/maximum width of basal turn; [9]) is 0.4, within the range of modern odontocetes (0.36–0.58). In addition to the above characteristics, Echovenator possesses a cochlea with two widely spaced turns (maximum distance, 1.55 mm); this spacing is within the range of other odontocetes (1.04–4.04), although the number of turns Current Biology 26, 1–6, August 22, 2016 ª 2016 Elsevier Ltd. 1
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
Figure 1. Holotype Skull, Mandible, and Right Bulla of Echovenator sandersi (A–N) Holotype mandible (A), atlas (B and C), cranium (D–K), and right tympanic bullae (L–N) of Echovenator sandersi. Cranium presented in dorsal (D and E), ventral (F and G), lateral (H and I), and anterior (J and K) views, with anatomical features labeled on the left and cranial bones indicated by color on the diagrams to the right. Abbreviations are as follows: afm, antorbital foramina of the maxilla; ang, angular process of the mandible; asl, ascending process of the lacrimal; asm, ascending process of the maxilla; con, occipital condyle; cp, coronoid process of the mandible; en, external nares; ep, embrasure pits; fps, falciform process of the squamosal; frf, frontal fossa; gf, glenoid fossa; ic, exposed incisivomaxillary canal; icn, intercondyloid notch; mf, mental foramina; mpc, medial prenarial crest; mr, mandibular ramus; ms, mandibular symphysis; nc, nuchal crest; npp, narial process of the premaxilla; paf, palatine foramen; pgp, postglenoid process; plf, posterior lacerate foramen (cranial hiatus); pnf, paranaris fossa; ppe, paraoccipital process of the exoccipital; prpf, preorbital process of the frontal; ps; posterior sinus; pspf, postorbital process of the frontal; ptc, parasagittal temporal crest; sopf, supraorbital process of the frontal; sqf, squamosal fossa; stc, subtemporal crest; sul, supraorbital sulcus of the maxilla; VIII, exit of the mandibular branch of the trigeminal nerve VIII; zyg, zygomatic process of the squamosal. Colors used to indicate different bones are as follows: cream, frontal; red, nasals; teal green, parietals; light purple, alisphenoid; light blue, basioccipital; blue green, mandible; dark blue, exoccipital and supraoccipital; dark orange, jugal; orange, maxilla; yellow, palatine; dark green, lacrimal; light green, premaxilla; dark pink, presphenoid and basisphenoid; dark purple, squamosal; dark red, vomer; brown, basioccipital. Dark gray-shaded regions indicate reconstructed parts with patching putty. Dashed lines and hatched regions indicate missing parts of the specimen. See also Supplemental Experimental Procedures, Figures S1 and S2, and Table S1.
is rather high, only surpassed by Lipotes (2.04) and Pontoporia (2.03) among the living odontocetes we examined. Principal-component analysis (PCA) performed on measurements of the inner ear separated taxa with known infrasonic hearing ranges (hippopotamids and mysticetes; [10, 11]) from those with ultrasonic hearing abilities (extant odontocetes; [11]), with no overlap (Figure 4). Approximately 56% of the variation in the inner ear measurements is explained by variation in body size, with larger taxa having more positive scores. Taxa with infrasonic and ultrasonic hearing are separated on the PC2 axis, which explained 29% of the variation in the dataset. The principle drivers of variation along this axis are increasing interturn distance, % SLA, and spiral ganglion canal size, as well as decreasing fenestra cochlearis area, cochlear length, and cochlear height. Whales with ultrasonic hearing possess more positive scores on this axis, while infrasonic hearing specialists have more negative scores. Most importantly, Echovenator is situated on the lower edge of the ultrasonic hearing morphospace, which includes all other extant and fossil odontocetes. Basilosaurids occupy an intermediate position between the ultrasonic and infrasonic specialists on PC2. Discriminant function analysis 2 Current Biology 26, 1–6, August 22, 2016
(DFA) classifies Echovenator as having high-frequency hearing with a posterior probability (pp) of 1.0. Basilosaurids are classified as having high-frequency hearing with a pp of 1.0; however, if hippopotamids were assigned a separate category (‘‘terrestrial’’) from mysticetes, then basilosaurids are classified as belonging to the terrestrial hearing category with a pp of 0.99. DISCUSSION The skull of Echovenator possesses rostral basins, a postnarial fossa, and expanded ascending processes of the maxillae. These are osteological correlates for facial air sacs and hypertrophied facial musculature, suggesting that Echovenator, like extant odontocetes, was capable of echolocation [6]. However, there is no way, at present, to infer the frequency range of odontocete vocalizations from facial osteology alone. By contrast, there are numerous anatomical correlates of ultrasonic hearing in the inner ear of odontocetes [3, 9, 11–14]. Perhaps the most important correlate is the length, width, and thickness of the basilar membrane [12, 15, 16]. This soft tissue structure is not preserved in fossils; however, the bony support
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
Figure 2. Right Periotic of Echovenator sandersi (A–F) Right periotic of Echovenator sandersi in ventrolateral (A and D), dorsal (B and E) and medial (C and F) views. Abbreviations are as follows: aes, anteroexternal sulcus; app, anterior process of the periotic; cc, cochlear crest (posterolateral crest of Basilosauridae); cn, foramen for cochlear nerve (branch of the vestibulo-cochlear nerve, VIII) and tractus spiralis foraminosus; dc, dorsal crest (superior process or tegmen tympani); elf, foramen for the endolymphatic canal; eth, epitympanic hiatus; fc, fenestra cochleae (round window); fg, groove for the facial nerve; fi, fossa incudi; fn, internal foramen for the facial nerve (VII); fm, fossa for the malleus head; fst, fossa for the stapedial muscle; gtt, groove for the tensor tympani muscle; iam, internal auditory meatus; plf, foramen for the cochlear aqueduct (perilymphatic canal); ppp, posterior process of the periotic; pr, promontorium; pyr, pyramidal process; smf, suprameatal fossa; st, stapes; vlt, ventrolateral tuberosity of the periotic; vn, foramen for the vestibular nerve (branch of the vestibulo-cochlear nerve, VIII). Internal morphology of the inner ear shown in Figure S1.
structures for the basilar membrane, the primary and secondary bony lamina, are preserved as ridges along the inner and outer walls of the cochlear canal, respectively, and are discernable using HRXCT data. A long SLA relative to the length of the cochlear canal (% SLA) is a clear indicator of ultrasonic hearing and is found in all extant odontocetes [17] as well as echolocating bats, but not most other mammals [8]. The gap between the bony lamina, which approximates the width of the basilar membrane, can also be used to distinguish infrasonic and ultrasonic hearing [18], although with less certainty due to greater susceptibility to post-mortem alteration [4]. Other morphological correlates for ultrasonic hearing include a high basal ratio [9] and a higher number of ganglion cells [17, 19]. The basal ratio largely reflects expansion of the basal turn, where high-frequency sounds are detected, and a larger spiral ganglion canal may correlate with more ganglion cells and increased innervation and signal processing capability [20]. Our study found that Echovenator had many of the above listed features, strongly supporting our interpretation that it could perceive ultrasonic sound. Furthermore, inclusion of Echovenator in a comprehensive phylogenetic analysis reveals the unambiguous early evolution of these traits within Odontoceti (Figure 3). Overall, the inner ear morphology of Echovenator is similar to that observed in the probable xenorophid ear described by Park et al. [3]. However, in contrast with their study, we found that the SLA is similar in length to those of extant odontocetes, and not intermediate between archaeocetes and crown odontocetes, as described in Park et al. [3]. This difference is largely methodological; we calculated % SLA using lengths in mm,
whereas Park et al. [3] used number of cochlear turns. When using their method, the % SLA in Echovenator drops to 56%, but we prefer our approach because theirs does not account for the fact that apical turns are of different diameters and lengths than basal turns. Echovenator retains primitive features lost in later diverging odontocetes but present in basilosaurids. We found that the size of the spiral ganglion canal in Echovenator is proportionally similar to that of basilosaurid archaeocete whales, which is in turn larger than the canal of mysticetes or hippopotamids. This implies that while Echovenator could hear ultrasonic frequencies, its ability to process these sounds was reduced compared to extant odontocetes. The length of the cochlea in Echovenator is only slightly smaller than that of basilosaurids (87%–93%) despite the fact that its body length is estimated to be one-fifth as large (1.48 m in Echovenator based on equations in [21] compared to 5.2 m in Zygorhiza; [22]). A relatively long cochlea, as compared to body size, has been previously associated with ultrasonic hearing in mammals [23], and body size does decrease across basal branches of the odontocete stem [24], even though cochlear length appears relatively static. This may suggest that a decoupling of body size and cochlear length was involved in the early evolution of ultrasonic hearing in odontocetes. Our study of Echovenator strongly supports the hypothesis that the first known odontocetes had ultrasonic hearing, but could this ability have evolved even earlier? The answer to this question lies among the archaeocetes, the paraphyletic stem group that gave rise to all extant cetaceans. The hearing abilities of archaeocetes have been difficult to infer [13, 17], with some studies supporting high-frequency hearing [12, 25] and others supporting low-frequency hearing [3, 4, 26]. Based on their Current Biology 26, 1–6, August 22, 2016 3
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
Figure 3. Simplified Phylogeny Indicating the Phylogenetic Relationships for Echovenator sandersi, Showing the Evolution of HearingRelated Traits Whales represented in color text, with archaeocetes in green, mysticetes in red, and odontocetes in blue, with Echovenator highlighted in bold. Horizontal bars near inner ears are scale bars (5 mm), with hearing ranges indicated in red text, while black bars and text indicate the origin of specific inner ear features based on ancestral character state reconstruction. Dashed lines for characters represent regions of the tree where the ancestral character state reconstruction is ambiguous for a given trait. Approximate maximum apical extent of SLA from base of the first cochlear turn delimited by yellow arc, with an arrow indicating this feature in Piscobalaena. CL refers to cochlear length, while FC refers to area of the fenestra cochlearis. Isosurface renderings of inner ears presented in apical (top) and medial (middle) views, with the cochlea shaded blue and other inner ear structures shaded red. Piscobalaena and Phocoena are represented by left ears that have been flipped to ease comparison with the other renderings. Complete phylogeny is presented in Figure S3, and character matrix is included within the supplemental material.
position within the PCA (Figure 4), basilosaurids show greater capabilities for hearing higher frequencies than their closest terrestrial relatives, hippos. This result is largely driven by two observations: their secondary bony lamina is longer, and their spiral ganglion canal is wider than those of hippopotamids. When viewed in a phylogenetic context (Figure 3), this indicates that Eocene cetaceans could hear higher frequencies than their terrestrial ancestors. This suggests that adaptations for hearing higher frequencies preceded the evolution of Odontoceti and 4 Current Biology 26, 1–6, August 22, 2016
that this capability was further developed and then co-opted for echolocation by early members of this clade. EXPERIMENTAL PROCEDURES Phylogenetic Analysis To determine the phylogenetic relationships of Echovenator sandersi, we added four Oligocene odontocetes to a previously published supermatrix [6]: Papahu taitapu, Otekaikea huata, Otekaikea marplesi, and Mirocetus
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
classified with 100% accuracy into each category, and separate analyses were used to determine which category fossil whale taxa belonged to. Laminar gap measurements were excluded from PCA and DFA because of concerns over accuracy of the measurements due to postmortem dislocation or damage to the laminae, while turns were excluded due to concerns of potential interdependence of this feature with cochlear length. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, four figures, two tables, and one supplemental data file and can be found with this article online at http://dx.doi.org/10.1016/j.cub.2016.06.004. AUTHOR CONTRIBUTIONS M.C., M.M.-C., C.d.M., and J.H.G. wrote the paper. M.M.-C. collected the CT data, and M.C. and J.H.G. performed data analyses. J.M. assisted in the naming of the genus and species and in the phylogenetic analyses. ACKNOWLEDGMENTS
Figure 4. PCA of Nine Measurements of the Inner Ear of Whales and Their Close Relatives PC 1 represents variation in body size, while PC 2 represents variation in morphological features associated with high- and low-frequency hearing. Ultrasonic hearing morphospace is indicated in light gray and infrasonic in dark gray. Symbols represent hippopotamids (white squares), archaeocetes (black triangles), fossil mysticetes (black diamonds), extant mysticetes (white diamonds), fossil odontocetes (black circles), extant odontocetes (white circles), and Echovenator sandersi (star). See also Figure S4 and Table S2.
riabinini (Data S1). The computer application TNT [27] was used to find the most parsimonious tree using ‘‘New Technology Search’’ until the single shortest tree was found 1,000 times (Figure S3). This phylogeny was then used to perform ancestral character state reconstruction on morphological features associated with ultrasonic hearing. All ancestral character state reconstruction was carried out in Mesquite 3.03 [28] using parsimony, equal branch lengths, and continuous characters. Character matrix is included as supplemental data (Data S1). Inner Ear Morphology The inner ear of Echovenator was scanned at the Nikon Metrology X-ray facilities in Tring, UK at a cubic voxel size of 0.05 mm. Ears of an additional 25 whale and outgroup taxa were also scanned at University of Texas Austin and National Museum of Natural History (Paris, France). PCA and DFA were performed on eight measurements of the cochlea and relevant structures taken from isosurface renderings of inner ears (Table S2). Measurements were selected based on their importance to hearing at different frequencies, as inferred from former studies of hearing in whales [4, 5, 12]. Taxa sampled include two genera of Hippopotamidae, the closest extant relatives to Cetacea [27], two basilosaurids, three mysticetes, and a range of odontocetes, comprising almost the entire extant diversity of Odontoceti, as well as several fossil odontocetes taxa including Squalodon calvertensis, Zarhachis flagellator, Phocageneus sp., and Kentriodon pernix. Measurements used included length of cochlea (CL), length of the SLA, widths of the basal turn (W1 and W2), height of the cochlea (H), distance between turns of the cochlea (ITD), radius of the spiral ganglion canal (GAN), and area of the opening of the fenestra cochlearis (FC). Detailed descriptions of how we measured these features can be found in Supplemental Experimental Procedures, part II. All PCA and DFA were performed in R 3.2.2. The R module MissMDA was used to calculate missing values in the analysis. Discriminant function analyses were used to generate classification rates for extant taxa using two (low and high frequency) and three (low frequency, high frequency, and terrestrial) categories. Extant whales were classified according to hearing data from the literature [11], while hippopotamids were treated as either low frequency [10] or terrestrial, depending on the number of categories used. Extant taxa were
First and foremost we thank Billy Palmer, Sr. for collecting and preparing the holotype of Echovenator sandersi and donating it to the Georgia Southern Museum. We thank K. Smith and B. Tharp (GSM), M. Brown and J. Carew (Mace Brown Museum of Natural History, College of Charleston), N. Pyenson (USNM), and N. Simmons (AMNH) for the opportunity to study and scan specimens housed in their respective institutions. Several individuals and institutions helped with the acquisition of microCT scans: G. Cle´ment, D. GeffardKuriyama, and F. Goussard (Muse´um national d’Histoire naturelle, including the AST-RX facility, Paris, France), J. Maisano and M. Colbert (The University of Texas High-Resolution X-ray Computed Tomography Facility), G. Bever (NYITCOM), Nikon Metrology X-ray Facilities, and Viscom France. F. Goussard and D. Geffard-Kuriyama also provided crucial training in virtual reconstruction methods using CT scan data. We thank D. Patel for CT segmenting of the inner ear of Choeropsis. During this project, we benefitted from discussions with B. Beatty, A. Sanders, D. Ketten, and R. David. The manuscript was improved with helpful comments by R. Racicot, Z.-X. Luo, and an anonymous reviewer. Financial support from the project was provided by the National Science Foundation (NSF-DEB 0640361 and NSF-EAR 1349607 to J.H.G.). Received: April 9, 2016 Revised: May 31, 2016 Accepted: June 1, 2016 Published: August 4, 2016 REFERENCES 1. Au, W.W.L. (1993). The Sonar of Dolphins (Spring-Verlag). 2. Jones, G. (2005). Echolocation. Curr. Biol. 15, R484–R488. 3. Park, T., Fitzgerald, E.M.G., and Evans, A.R. (2016). Ultrasonic hearing and echolocation in the earliest toothed whales. Biol. Lett. 12, 20160060. 4. Ekdale, E.G., and Racicot, R.A. (2015). Anatomical evidence for low frequency sensitivity in an archaeocete whale: comparison of the inner ear of Zygorhiza kochii with that of crown Mysticeti. J. Anat. 226, 22–39. 5. Katuna, M.P., Geisler, J.H., and Colquhoun, D.J. (1997). Stratigraphic correlation of Oligocene marginal marine and fluvial deposits across the middle and lower coastal plain, South Carolina. Sediment. Geol. 108, 181–194. 6. Geisler, J.H., Colbert, M.W., and Carew, J.L. (2014). A new fossil species supports an early origin for toothed whale echolocation. Nature 508, 383–386. 7. Sanders, A.E., and Geisler, J.H. (2015). A new basal odontocete from the upper Rupelian of South Carolina, U.S.A., with contributions to the systematics of Xenorophus and Mirocetus (Mammalia, Cetacea). J. Vert. Paleontol. 35, e890107.
Current Biology 26, 1–6, August 22, 2016 5
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
8. Ekdale, E.G. (2013). Comparative anatomy of the bony labyrinth (Inner Ear) of Placental Mammals. PLoS ONE 8, e66624. 9. Ketten, D.R., and Wartzok, D. (1990). Three-dimensional reconstructions of the dolphin ear. In Sensory Abilities of Cetaceans, J.A. Thomas, and R.A. Kastelein, eds. (Plenum Press), pp. 81–105. 10. Barklow, W.E. (2004). Low-frequency sounds and amphibious communication in Hippopotamus amphibious. J. Acoust. Soc. Am. 68, 1125–1132. 11. Ketten, D.R. (1994). Functional analyses of whale ears: adaptations for underwater hearing. IEEE Proc. Underwater Acoust. 1, 264–270. 12. Ketten, D.R. (1992). The cetacean ear: form, frequency, and evolution. In Marine Mammal Sensory Systems, J.A. Thomas, R.A. Kastelein, and A.Y. Supin, eds. (Plenum Press), pp. 53–75. 13. Fleischer, G. (1976). Hearing in extinct cetaceans as determined by cochlear structure. J. Paleontol. 50, 133–152. 14. Fraser, F.C., and Purves, P.E. (1960). Hearing in cetaceans: evolution of the accessory air sacs and the structure and function of the outer and middle ear in recent cetaceans. Bull. br. Mus. Nat. Hist. Zool. 7, 1–141. 15. von Be´ke´sy, G. (1970). Travelling waves as frequency analysers in the cochlea. Nature 225, 1207–1209. 16. West, C.D. (1985). The relationship of the spiral turns of the cochlea and the length of the basilar membrane to the range of audible frequencies in ground dwelling mammals. J. Acoust. Soc. Am. 77, 1091–1101. 17. Ketten, D.R. (1992). The marine mammal ear: specializations for aquatic audition and echolocation. In Evolutionary Biology of Hearing, D.B. Webster, R.R. Fay, and A.N. Popper, eds. (Springer-Verlag), pp. 717–750. 18. Geisler, J.H., and Luo, Z. (1996). The petrosal and inner ear of Herpetocetus sp. (Mammalia: Cetacea) and their implication for the phylogeny and hearing or archaic mysticetes. J. Paleontol. 70, 1045–1066.
6 Current Biology 26, 1–6, August 22, 2016
19. Wever, E.G., McCormick, J.G., Palin, J., and Ridgway, S.H. (1971). The cochlea of the dolphin, Tursiops truncatus: hair cells and ganglion cells. Proc. Natl. Acad. Sci. USA 68, 2908–2912. 20. Luo, Z., and Marsh, K. (1996). Petrosal (periotic) and inner ear of a Pliocene kogiine whale (Kogiinae, Odontoceti): implications on relationships and hearing evolution of toothed whales. J. Vert. Paleontol 16, 328–348. 21. Pyenson, N.D., and Sponberg, S.N. (2011). Reconstructing body size in extinct crown Cetacea (Neoceti) using allometry, phylogenetic methods, and tests from the fossil record. J. Mamm. Evol. 18, 269–288. 22. Kellogg, R. (1936). A review of the Archaeoceti. Publ. Carnegie Instit. Wash. 482, 1–366. 23. Davies, K.T.J., Maryanto, I., and Rossiter, S.J. (2013). Evolutionary origins of ultrasonic hearing and laryngeal echolocation in bats inferred from morphological analyses of the inner ear. Front. Zool. 10, 2. 24. Montgomery, S.H., Geisler, J.H., McGowen, M.R., Fox, C., Marino, L., and Gatesy, J. (2013). The evolutionary history of cetacean brain and body size. Evolution 67, 3339–3353. 25. Fahlke, J.M., Gingerich, P.D., Welsh, R.C., and Wood, A.R. (2011). Cranial asymmetry in Eocene archaeocete whales and the evolution of directional hearing in water. Proc. Natl. Acad. Sci. USA 108, 14545–14548. 26. Uhen, M.D. (2004). Form, function, and anatomy of Dorudon atrox (Mammalia, Cetacea): an archaeocete from the Middle to Late Eocene of Egypt. Univ. Michigan Paper Paleontol. 34, 1–222. 27. Goloboff, P.A., Farris, J.S., and Nixon, K.C. (2008). TNT, a free program for phylogenetic analysis. Cladistics 24, 774–786. 28. Maddison, W.P., and Maddison, D.R. (2011). Mesquite: a modular system for evolutionary analysis, 2.72.
The Origin of High-Frequency Hearing in Whales Highlights
Authors
d
An ancient toothed whale is described, which possesses a well-preserved inner ear
d
Features associated with ultrasonic hearing are preserved in the ear of this whale
Morgan Churchill, Manuel Martinez-Caceres, Christian de Muizon, Jessica Mnieckowski, Jonathan H. Geisler
d
Ultrasonic hearing evolved with echolocation in the first toothed whales
Correspondence
d
Hearing at higher frequencies began in the ancestors of toothed whales
[email protected]
In Brief Churchill et al. describe a new species of early toothed whale (Echovenator sandersi), which possesses one of the oldest known well-preserved inner ears. Comparison with other fossil whales reveals that the ultrasonic hearing evolved at the base of the toothed whale radiation, and high-frequency hearing evolved before echolocation.
Churchill et al., 2016, Current Biology 26, 1–6 August 22, 2016 ª 2016 Elsevier Ltd. http://dx.doi.org/10.1016/j.cub.2016.06.004
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
Current Biology
Report The Origin of High-Frequency Hearing in Whales Morgan Churchill,1,* Manuel Martinez-Caceres,2 Christian de Muizon,2 Jessica Mnieckowski,3 and Jonathan H. Geisler1 1Department
of Anatomy, College of Osteopathic Medicine, New York Institute of Technology, Old Westbury, NY 11568, USA of Earth History, National Museum of Natural History, Paris 75005, France 3Ecological Planning Group, Savannah, GA 31401, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cub.2016.06.004 2Department
SUMMARY
Odontocetes (toothed whales) rely upon echoes of their own vocalizations to navigate and find prey underwater [1]. This sensory adaptation, known as echolocation, operates most effectively when using high frequencies, and odontocetes are rivaled only by bats in their ability to perceive ultrasonic sound greater than 100 kHz [2]. Although features indicative of ultrasonic hearing are present in the oldest known odontocetes [3], the significance of this finding is limited by the methods employed and taxa sampled. In this report, we describe a new xenorophid whale (Echovenator sandersi, gen. et sp. nov.) from the Oligocene of South Carolina that, as a member of the most basal clade of odontocetes, sheds considerable light on the evolution of ultrasonic hearing. By placing high-resolution CT data from Echovenator sandersi, 2 hippos, and 23 fossil and extant whales in a phylogenetic context, we conclude that ultrasonic hearing, albeit in a less specialized form, evolved at the base of the odontocete radiation. Contrary to the hypothesis that odontocetes evolved from low-frequency specialists [4], we find evidence that stem cetaceans, the archaeocetes, were more sensitive to high-frequency sound than their terrestrial ancestors. This indicates that selection for high-frequency hearing predates the emergence of Odontoceti and the evolution of echolocation. RESULTS Description Systematics Order Cetacea Brisson 1762. Suborder Odontoceti Flower 1867. Family Xenorophidae Uhen 2008. Echovenator sandersi gen. et sp. nov. Etymology From Latin for ‘‘echo hunter,’’ referring to its use of echolocation while foraging. Species name is in honor of Albert Sanders, former curator of The Charleston Museum, for his contributions to our knowledge of Oligocene Cetacea. Holotype Georgia Southern Museum (GSM) 1098, a nearly complete skull with 39 teeth, mandible, and atlas, presumed to represent a sin-
gle individual (Figures 1 and 2; see also Figures S1 and S2, Table S1, and Supplemental Experimental Procedures, part I). Locality and Age Drainage ditch associated with Limehouse Branch Creek, Berkeley County, South Carolina. Approximate coordinates are: N 33 010 5300 ; W 80 060 0600 . Basal bed of the Chandler Bridge Formation, late Oligocene (ca. 24–27 Ma) in age [5]. This bed is a probable equivalent of the marine or marginal marine lithofacies, which was deposited in a nearshore marine or restricted bay environment [5]. Diagnosis Echovenator has the following synapomorphies of Xenorophidae: premaxilla underlies the ascending process of the maxilla, a frontal window that exposes the maxilla and premaxilla, and an elongate ventrolateral tuberosity of the periotic [6, 7]. Echovenator differs from all xenorophids in having fused, or partially fused, fronto-nasal and maxillo-premaxillary sutures and in having a paranaris fossa; it differs from Xenorophus sloanii and Albertocetus meffordorum in having a postnarial fossa and a depressed internasal suture; and it differs from Cotylocara macei in having an undivided postnarial fossa and in having larger dorsal exposures of parietals. Phylogenetic Analysis Our phylogenetic analysis recovered a single most parsimonious tree of 41201.26 steps (Figures 3 and S3). Within this phylogeny, Echovenator is the sister taxon of a clade comprising Cotylocara and an undescribed xenorophid in The Charleston Museum (ChM) PV2758; both taxa are deeply embedded within Xenorophidae. Xenorophidae was found to be comprised solely of taxa, both described and undescribed, from the Oligocene of the southeastern United States. Inner Ear Morphology Our study found that Echovenator shares many features of the inner ear associated with ultrasonic hearing (Table S2; Figure S4). The secondary bony lamina of the cochlea (SLA) is 72.21% of the total length of the cochlear canal, on the low end of the range exhibited by extant odontocetes (71%–92%), but far above that seen in other mammals [8]. The laminar gap at the first quarter turn, related to size of the basilar membrane, is 0.20 mm. This gap is far smaller than that measured for any non-odontocete taxa in this study (i.e., 0.38–0.90) but larger than most odontocetes examined, although similar in size to ziphiid whales (0.21–0.22). The basal ratio (height of cochlea/maximum width of basal turn; [9]) is 0.4, within the range of modern odontocetes (0.36–0.58). In addition to the above characteristics, Echovenator possesses a cochlea with two widely spaced turns (maximum distance, 1.55 mm); this spacing is within the range of other odontocetes (1.04–4.04), although the number of turns Current Biology 26, 1–6, August 22, 2016 ª 2016 Elsevier Ltd. 1
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
Figure 1. Holotype Skull, Mandible, and Right Bulla of Echovenator sandersi (A–N) Holotype mandible (A), atlas (B and C), cranium (D–K), and right tympanic bullae (L–N) of Echovenator sandersi. Cranium presented in dorsal (D and E), ventral (F and G), lateral (H and I), and anterior (J and K) views, with anatomical features labeled on the left and cranial bones indicated by color on the diagrams to the right. Abbreviations are as follows: afm, antorbital foramina of the maxilla; ang, angular process of the mandible; asl, ascending process of the lacrimal; asm, ascending process of the maxilla; con, occipital condyle; cp, coronoid process of the mandible; en, external nares; ep, embrasure pits; fps, falciform process of the squamosal; frf, frontal fossa; gf, glenoid fossa; ic, exposed incisivomaxillary canal; icn, intercondyloid notch; mf, mental foramina; mpc, medial prenarial crest; mr, mandibular ramus; ms, mandibular symphysis; nc, nuchal crest; npp, narial process of the premaxilla; paf, palatine foramen; pgp, postglenoid process; plf, posterior lacerate foramen (cranial hiatus); pnf, paranaris fossa; ppe, paraoccipital process of the exoccipital; prpf, preorbital process of the frontal; ps; posterior sinus; pspf, postorbital process of the frontal; ptc, parasagittal temporal crest; sopf, supraorbital process of the frontal; sqf, squamosal fossa; stc, subtemporal crest; sul, supraorbital sulcus of the maxilla; VIII, exit of the mandibular branch of the trigeminal nerve VIII; zyg, zygomatic process of the squamosal. Colors used to indicate different bones are as follows: cream, frontal; red, nasals; teal green, parietals; light purple, alisphenoid; light blue, basioccipital; blue green, mandible; dark blue, exoccipital and supraoccipital; dark orange, jugal; orange, maxilla; yellow, palatine; dark green, lacrimal; light green, premaxilla; dark pink, presphenoid and basisphenoid; dark purple, squamosal; dark red, vomer; brown, basioccipital. Dark gray-shaded regions indicate reconstructed parts with patching putty. Dashed lines and hatched regions indicate missing parts of the specimen. See also Supplemental Experimental Procedures, Figures S1 and S2, and Table S1.
is rather high, only surpassed by Lipotes (2.04) and Pontoporia (2.03) among the living odontocetes we examined. Principal-component analysis (PCA) performed on measurements of the inner ear separated taxa with known infrasonic hearing ranges (hippopotamids and mysticetes; [10, 11]) from those with ultrasonic hearing abilities (extant odontocetes; [11]), with no overlap (Figure 4). Approximately 56% of the variation in the inner ear measurements is explained by variation in body size, with larger taxa having more positive scores. Taxa with infrasonic and ultrasonic hearing are separated on the PC2 axis, which explained 29% of the variation in the dataset. The principle drivers of variation along this axis are increasing interturn distance, % SLA, and spiral ganglion canal size, as well as decreasing fenestra cochlearis area, cochlear length, and cochlear height. Whales with ultrasonic hearing possess more positive scores on this axis, while infrasonic hearing specialists have more negative scores. Most importantly, Echovenator is situated on the lower edge of the ultrasonic hearing morphospace, which includes all other extant and fossil odontocetes. Basilosaurids occupy an intermediate position between the ultrasonic and infrasonic specialists on PC2. Discriminant function analysis 2 Current Biology 26, 1–6, August 22, 2016
(DFA) classifies Echovenator as having high-frequency hearing with a posterior probability (pp) of 1.0. Basilosaurids are classified as having high-frequency hearing with a pp of 1.0; however, if hippopotamids were assigned a separate category (‘‘terrestrial’’) from mysticetes, then basilosaurids are classified as belonging to the terrestrial hearing category with a pp of 0.99. DISCUSSION The skull of Echovenator possesses rostral basins, a postnarial fossa, and expanded ascending processes of the maxillae. These are osteological correlates for facial air sacs and hypertrophied facial musculature, suggesting that Echovenator, like extant odontocetes, was capable of echolocation [6]. However, there is no way, at present, to infer the frequency range of odontocete vocalizations from facial osteology alone. By contrast, there are numerous anatomical correlates of ultrasonic hearing in the inner ear of odontocetes [3, 9, 11–14]. Perhaps the most important correlate is the length, width, and thickness of the basilar membrane [12, 15, 16]. This soft tissue structure is not preserved in fossils; however, the bony support
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
Figure 2. Right Periotic of Echovenator sandersi (A–F) Right periotic of Echovenator sandersi in ventrolateral (A and D), dorsal (B and E) and medial (C and F) views. Abbreviations are as follows: aes, anteroexternal sulcus; app, anterior process of the periotic; cc, cochlear crest (posterolateral crest of Basilosauridae); cn, foramen for cochlear nerve (branch of the vestibulo-cochlear nerve, VIII) and tractus spiralis foraminosus; dc, dorsal crest (superior process or tegmen tympani); elf, foramen for the endolymphatic canal; eth, epitympanic hiatus; fc, fenestra cochleae (round window); fg, groove for the facial nerve; fi, fossa incudi; fn, internal foramen for the facial nerve (VII); fm, fossa for the malleus head; fst, fossa for the stapedial muscle; gtt, groove for the tensor tympani muscle; iam, internal auditory meatus; plf, foramen for the cochlear aqueduct (perilymphatic canal); ppp, posterior process of the periotic; pr, promontorium; pyr, pyramidal process; smf, suprameatal fossa; st, stapes; vlt, ventrolateral tuberosity of the periotic; vn, foramen for the vestibular nerve (branch of the vestibulo-cochlear nerve, VIII). Internal morphology of the inner ear shown in Figure S1.
structures for the basilar membrane, the primary and secondary bony lamina, are preserved as ridges along the inner and outer walls of the cochlear canal, respectively, and are discernable using HRXCT data. A long SLA relative to the length of the cochlear canal (% SLA) is a clear indicator of ultrasonic hearing and is found in all extant odontocetes [17] as well as echolocating bats, but not most other mammals [8]. The gap between the bony lamina, which approximates the width of the basilar membrane, can also be used to distinguish infrasonic and ultrasonic hearing [18], although with less certainty due to greater susceptibility to post-mortem alteration [4]. Other morphological correlates for ultrasonic hearing include a high basal ratio [9] and a higher number of ganglion cells [17, 19]. The basal ratio largely reflects expansion of the basal turn, where high-frequency sounds are detected, and a larger spiral ganglion canal may correlate with more ganglion cells and increased innervation and signal processing capability [20]. Our study found that Echovenator had many of the above listed features, strongly supporting our interpretation that it could perceive ultrasonic sound. Furthermore, inclusion of Echovenator in a comprehensive phylogenetic analysis reveals the unambiguous early evolution of these traits within Odontoceti (Figure 3). Overall, the inner ear morphology of Echovenator is similar to that observed in the probable xenorophid ear described by Park et al. [3]. However, in contrast with their study, we found that the SLA is similar in length to those of extant odontocetes, and not intermediate between archaeocetes and crown odontocetes, as described in Park et al. [3]. This difference is largely methodological; we calculated % SLA using lengths in mm,
whereas Park et al. [3] used number of cochlear turns. When using their method, the % SLA in Echovenator drops to 56%, but we prefer our approach because theirs does not account for the fact that apical turns are of different diameters and lengths than basal turns. Echovenator retains primitive features lost in later diverging odontocetes but present in basilosaurids. We found that the size of the spiral ganglion canal in Echovenator is proportionally similar to that of basilosaurid archaeocete whales, which is in turn larger than the canal of mysticetes or hippopotamids. This implies that while Echovenator could hear ultrasonic frequencies, its ability to process these sounds was reduced compared to extant odontocetes. The length of the cochlea in Echovenator is only slightly smaller than that of basilosaurids (87%–93%) despite the fact that its body length is estimated to be one-fifth as large (1.48 m in Echovenator based on equations in [21] compared to 5.2 m in Zygorhiza; [22]). A relatively long cochlea, as compared to body size, has been previously associated with ultrasonic hearing in mammals [23], and body size does decrease across basal branches of the odontocete stem [24], even though cochlear length appears relatively static. This may suggest that a decoupling of body size and cochlear length was involved in the early evolution of ultrasonic hearing in odontocetes. Our study of Echovenator strongly supports the hypothesis that the first known odontocetes had ultrasonic hearing, but could this ability have evolved even earlier? The answer to this question lies among the archaeocetes, the paraphyletic stem group that gave rise to all extant cetaceans. The hearing abilities of archaeocetes have been difficult to infer [13, 17], with some studies supporting high-frequency hearing [12, 25] and others supporting low-frequency hearing [3, 4, 26]. Based on their Current Biology 26, 1–6, August 22, 2016 3
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
Figure 3. Simplified Phylogeny Indicating the Phylogenetic Relationships for Echovenator sandersi, Showing the Evolution of HearingRelated Traits Whales represented in color text, with archaeocetes in green, mysticetes in red, and odontocetes in blue, with Echovenator highlighted in bold. Horizontal bars near inner ears are scale bars (5 mm), with hearing ranges indicated in red text, while black bars and text indicate the origin of specific inner ear features based on ancestral character state reconstruction. Dashed lines for characters represent regions of the tree where the ancestral character state reconstruction is ambiguous for a given trait. Approximate maximum apical extent of SLA from base of the first cochlear turn delimited by yellow arc, with an arrow indicating this feature in Piscobalaena. CL refers to cochlear length, while FC refers to area of the fenestra cochlearis. Isosurface renderings of inner ears presented in apical (top) and medial (middle) views, with the cochlea shaded blue and other inner ear structures shaded red. Piscobalaena and Phocoena are represented by left ears that have been flipped to ease comparison with the other renderings. Complete phylogeny is presented in Figure S3, and character matrix is included within the supplemental material.
position within the PCA (Figure 4), basilosaurids show greater capabilities for hearing higher frequencies than their closest terrestrial relatives, hippos. This result is largely driven by two observations: their secondary bony lamina is longer, and their spiral ganglion canal is wider than those of hippopotamids. When viewed in a phylogenetic context (Figure 3), this indicates that Eocene cetaceans could hear higher frequencies than their terrestrial ancestors. This suggests that adaptations for hearing higher frequencies preceded the evolution of Odontoceti and 4 Current Biology 26, 1–6, August 22, 2016
that this capability was further developed and then co-opted for echolocation by early members of this clade. EXPERIMENTAL PROCEDURES Phylogenetic Analysis To determine the phylogenetic relationships of Echovenator sandersi, we added four Oligocene odontocetes to a previously published supermatrix [6]: Papahu taitapu, Otekaikea huata, Otekaikea marplesi, and Mirocetus
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
classified with 100% accuracy into each category, and separate analyses were used to determine which category fossil whale taxa belonged to. Laminar gap measurements were excluded from PCA and DFA because of concerns over accuracy of the measurements due to postmortem dislocation or damage to the laminae, while turns were excluded due to concerns of potential interdependence of this feature with cochlear length. SUPPLEMENTAL INFORMATION Supplemental Information includes Supplemental Experimental Procedures, four figures, two tables, and one supplemental data file and can be found with this article online at http://dx.doi.org/10.1016/j.cub.2016.06.004. AUTHOR CONTRIBUTIONS M.C., M.M.-C., C.d.M., and J.H.G. wrote the paper. M.M.-C. collected the CT data, and M.C. and J.H.G. performed data analyses. J.M. assisted in the naming of the genus and species and in the phylogenetic analyses. ACKNOWLEDGMENTS
Figure 4. PCA of Nine Measurements of the Inner Ear of Whales and Their Close Relatives PC 1 represents variation in body size, while PC 2 represents variation in morphological features associated with high- and low-frequency hearing. Ultrasonic hearing morphospace is indicated in light gray and infrasonic in dark gray. Symbols represent hippopotamids (white squares), archaeocetes (black triangles), fossil mysticetes (black diamonds), extant mysticetes (white diamonds), fossil odontocetes (black circles), extant odontocetes (white circles), and Echovenator sandersi (star). See also Figure S4 and Table S2.
riabinini (Data S1). The computer application TNT [27] was used to find the most parsimonious tree using ‘‘New Technology Search’’ until the single shortest tree was found 1,000 times (Figure S3). This phylogeny was then used to perform ancestral character state reconstruction on morphological features associated with ultrasonic hearing. All ancestral character state reconstruction was carried out in Mesquite 3.03 [28] using parsimony, equal branch lengths, and continuous characters. Character matrix is included as supplemental data (Data S1). Inner Ear Morphology The inner ear of Echovenator was scanned at the Nikon Metrology X-ray facilities in Tring, UK at a cubic voxel size of 0.05 mm. Ears of an additional 25 whale and outgroup taxa were also scanned at University of Texas Austin and National Museum of Natural History (Paris, France). PCA and DFA were performed on eight measurements of the cochlea and relevant structures taken from isosurface renderings of inner ears (Table S2). Measurements were selected based on their importance to hearing at different frequencies, as inferred from former studies of hearing in whales [4, 5, 12]. Taxa sampled include two genera of Hippopotamidae, the closest extant relatives to Cetacea [27], two basilosaurids, three mysticetes, and a range of odontocetes, comprising almost the entire extant diversity of Odontoceti, as well as several fossil odontocetes taxa including Squalodon calvertensis, Zarhachis flagellator, Phocageneus sp., and Kentriodon pernix. Measurements used included length of cochlea (CL), length of the SLA, widths of the basal turn (W1 and W2), height of the cochlea (H), distance between turns of the cochlea (ITD), radius of the spiral ganglion canal (GAN), and area of the opening of the fenestra cochlearis (FC). Detailed descriptions of how we measured these features can be found in Supplemental Experimental Procedures, part II. All PCA and DFA were performed in R 3.2.2. The R module MissMDA was used to calculate missing values in the analysis. Discriminant function analyses were used to generate classification rates for extant taxa using two (low and high frequency) and three (low frequency, high frequency, and terrestrial) categories. Extant whales were classified according to hearing data from the literature [11], while hippopotamids were treated as either low frequency [10] or terrestrial, depending on the number of categories used. Extant taxa were
First and foremost we thank Billy Palmer, Sr. for collecting and preparing the holotype of Echovenator sandersi and donating it to the Georgia Southern Museum. We thank K. Smith and B. Tharp (GSM), M. Brown and J. Carew (Mace Brown Museum of Natural History, College of Charleston), N. Pyenson (USNM), and N. Simmons (AMNH) for the opportunity to study and scan specimens housed in their respective institutions. Several individuals and institutions helped with the acquisition of microCT scans: G. Cle´ment, D. GeffardKuriyama, and F. Goussard (Muse´um national d’Histoire naturelle, including the AST-RX facility, Paris, France), J. Maisano and M. Colbert (The University of Texas High-Resolution X-ray Computed Tomography Facility), G. Bever (NYITCOM), Nikon Metrology X-ray Facilities, and Viscom France. F. Goussard and D. Geffard-Kuriyama also provided crucial training in virtual reconstruction methods using CT scan data. We thank D. Patel for CT segmenting of the inner ear of Choeropsis. During this project, we benefitted from discussions with B. Beatty, A. Sanders, D. Ketten, and R. David. The manuscript was improved with helpful comments by R. Racicot, Z.-X. Luo, and an anonymous reviewer. Financial support from the project was provided by the National Science Foundation (NSF-DEB 0640361 and NSF-EAR 1349607 to J.H.G.). Received: April 9, 2016 Revised: May 31, 2016 Accepted: June 1, 2016 Published: August 4, 2016 REFERENCES 1. Au, W.W.L. (1993). The Sonar of Dolphins (Spring-Verlag). 2. Jones, G. (2005). Echolocation. Curr. Biol. 15, R484–R488. 3. Park, T., Fitzgerald, E.M.G., and Evans, A.R. (2016). Ultrasonic hearing and echolocation in the earliest toothed whales. Biol. Lett. 12, 20160060. 4. Ekdale, E.G., and Racicot, R.A. (2015). Anatomical evidence for low frequency sensitivity in an archaeocete whale: comparison of the inner ear of Zygorhiza kochii with that of crown Mysticeti. J. Anat. 226, 22–39. 5. Katuna, M.P., Geisler, J.H., and Colquhoun, D.J. (1997). Stratigraphic correlation of Oligocene marginal marine and fluvial deposits across the middle and lower coastal plain, South Carolina. Sediment. Geol. 108, 181–194. 6. Geisler, J.H., Colbert, M.W., and Carew, J.L. (2014). A new fossil species supports an early origin for toothed whale echolocation. Nature 508, 383–386. 7. Sanders, A.E., and Geisler, J.H. (2015). A new basal odontocete from the upper Rupelian of South Carolina, U.S.A., with contributions to the systematics of Xenorophus and Mirocetus (Mammalia, Cetacea). J. Vert. Paleontol. 35, e890107.
Current Biology 26, 1–6, August 22, 2016 5
Please cite this article in press as: Churchill et al., The Origin of High-Frequency Hearing in Whales, Current Biology (2016), http://dx.doi.org/10.1016/ j.cub.2016.06.004
8. Ekdale, E.G. (2013). Comparative anatomy of the bony labyrinth (Inner Ear) of Placental Mammals. PLoS ONE 8, e66624. 9. Ketten, D.R., and Wartzok, D. (1990). Three-dimensional reconstructions of the dolphin ear. In Sensory Abilities of Cetaceans, J.A. Thomas, and R.A. Kastelein, eds. (Plenum Press), pp. 81–105. 10. Barklow, W.E. (2004). Low-frequency sounds and amphibious communication in Hippopotamus amphibious. J. Acoust. Soc. Am. 68, 1125–1132. 11. Ketten, D.R. (1994). Functional analyses of whale ears: adaptations for underwater hearing. IEEE Proc. Underwater Acoust. 1, 264–270. 12. Ketten, D.R. (1992). The cetacean ear: form, frequency, and evolution. In Marine Mammal Sensory Systems, J.A. Thomas, R.A. Kastelein, and A.Y. Supin, eds. (Plenum Press), pp. 53–75. 13. Fleischer, G. (1976). Hearing in extinct cetaceans as determined by cochlear structure. J. Paleontol. 50, 133–152. 14. Fraser, F.C., and Purves, P.E. (1960). Hearing in cetaceans: evolution of the accessory air sacs and the structure and function of the outer and middle ear in recent cetaceans. Bull. br. Mus. Nat. Hist. Zool. 7, 1–141. 15. von Be´ke´sy, G. (1970). Travelling waves as frequency analysers in the cochlea. Nature 225, 1207–1209. 16. West, C.D. (1985). The relationship of the spiral turns of the cochlea and the length of the basilar membrane to the range of audible frequencies in ground dwelling mammals. J. Acoust. Soc. Am. 77, 1091–1101. 17. Ketten, D.R. (1992). The marine mammal ear: specializations for aquatic audition and echolocation. In Evolutionary Biology of Hearing, D.B. Webster, R.R. Fay, and A.N. Popper, eds. (Springer-Verlag), pp. 717–750. 18. Geisler, J.H., and Luo, Z. (1996). The petrosal and inner ear of Herpetocetus sp. (Mammalia: Cetacea) and their implication for the phylogeny and hearing or archaic mysticetes. J. Paleontol. 70, 1045–1066.
6 Current Biology 26, 1–6, August 22, 2016
19. Wever, E.G., McCormick, J.G., Palin, J., and Ridgway, S.H. (1971). The cochlea of the dolphin, Tursiops truncatus: hair cells and ganglion cells. Proc. Natl. Acad. Sci. USA 68, 2908–2912. 20. Luo, Z., and Marsh, K. (1996). Petrosal (periotic) and inner ear of a Pliocene kogiine whale (Kogiinae, Odontoceti): implications on relationships and hearing evolution of toothed whales. J. Vert. Paleontol 16, 328–348. 21. Pyenson, N.D., and Sponberg, S.N. (2011). Reconstructing body size in extinct crown Cetacea (Neoceti) using allometry, phylogenetic methods, and tests from the fossil record. J. Mamm. Evol. 18, 269–288. 22. Kellogg, R. (1936). A review of the Archaeoceti. Publ. Carnegie Instit. Wash. 482, 1–366. 23. Davies, K.T.J., Maryanto, I., and Rossiter, S.J. (2013). Evolutionary origins of ultrasonic hearing and laryngeal echolocation in bats inferred from morphological analyses of the inner ear. Front. Zool. 10, 2. 24. Montgomery, S.H., Geisler, J.H., McGowen, M.R., Fox, C., Marino, L., and Gatesy, J. (2013). The evolutionary history of cetacean brain and body size. Evolution 67, 3339–3353. 25. Fahlke, J.M., Gingerich, P.D., Welsh, R.C., and Wood, A.R. (2011). Cranial asymmetry in Eocene archaeocete whales and the evolution of directional hearing in water. Proc. Natl. Acad. Sci. USA 108, 14545–14548. 26. Uhen, M.D. (2004). Form, function, and anatomy of Dorudon atrox (Mammalia, Cetacea): an archaeocete from the Middle to Late Eocene of Egypt. Univ. Michigan Paper Paleontol. 34, 1–222. 27. Goloboff, P.A., Farris, J.S., and Nixon, K.C. (2008). TNT, a free program for phylogenetic analysis. Cladistics 24, 774–786. 28. Maddison, W.P., and Maddison, D.R. (2011). Mesquite: a modular system for evolutionary analysis, 2.72.