Report

Unexpected Convergent E

volution of Nasal Domesbetween Pleistocene Bovids and CretaceousHadrosaur Dinosaurs

Highlights

d Pleistocene Rusingoryx atopocranion are first known

mammals with hollow nasal crests

d Rusingoryx ontogeny and evolution are broadly similar to

lambeosaurine hadrosaurs

d The best-supported nasal crest function is phonic

modification

d Combination of convergent ontogeny, evolution, and function

may explain crest rarity

O’Brien et al., 2016, Current Biology 26, 503–508February 22, 2016 ª2016 Elsevier Ltd All rights reservedhttp://dx.doi.org/10.1016/j.cub.2015.12.050

Authors

Haley D. O’Brien, J. Tyler Faith, Kirsten

E. Jenkins, ..., Gilbert Price, Yue-xing

Feng, Christian A. Tryon

Correspondencehaley.d.obrien@gmail.com (H.D.O.),j.faith@uq.edu.au (J.T.F.)

In Brief

O’Brien et al. analyze first known

mammalian osseous nasal crests from a

new Rusingoryx assemblage. Analogous

to those of lambeosaurine hadrosaurs,

these crests may facilitate vocalization.

This morphology is an outstanding

example of convergent evolution

between dinosaurs and bovids, driven by

ontogenetic, evolutionary, and

environmental pressures.

Current Biology

Report

Unexpected Convergent Evolution of Nasal Domesbetween Pleistocene Bovidsand Cretaceous Hadrosaur DinosaursHaley D. O’Brien,1,* J. Tyler Faith,2,* Kirsten E. Jenkins,3 Daniel J. Peppe,4 Thomas W. Plummer,5 Zenobia L. Jacobs,6

Bo Li,6 Renaud Joannes-Boyau,7 Gilbert Price,8 Yue-xing Feng,8 and Christian A. Tryon91Department of Biological Sciences, Graduate Program in Ecology and Evolutionary Biology, Ohio University, Athens, OH 45701, USA2Department of Archaeology, University of Queensland, Brisbane, St Lucia, QLD 4072, Australia3Department of Anthropology, University of Minnesota, 395 Humphrey Center, 301 19th Avenue S, Minneapolis, MN 55455, USA4Department of Geology, Baylor University, One Bear Place #97354, Waco, TX 76798, USA5Department of Anthropology, Queens College, City University of New York, 314 Powdermaker Hall, 65-30 Kissena Boulevard, Flushing,

NY 11367, USA6Center for Archaeological Science, University of Wollongong, Room 268, Building 41, Northfields Avenue,Wollongong, NSW 2522, Australia7Department of GeoScience, Southern Cross University, PO Box 157, Lismore, NSW 2480, Australia8School of Earth Sciences, University of Queensland, Brisbane, St Lucia, QLD 4072, Australia9Department of Anthropology, Harvard University, Peabody Museum, 11 Divinity Avenue, Cambridge, MA 02138, USA*Correspondence: haley.d.obrien@gmail.com (H.D.O.), j.faith@uq.edu.au (J.T.F.)

http://dx.doi.org/10.1016/j.cub.2015.12.050

SUMMARY

The fossil recordprovides tangible,historical evidencefor the mode and operation of evolution across deeptime. Striking patterns of convergence are some ofthe strongest examples of these operations, whereby,over time, similar environmental and/or behavioralpressures precipitate similarity in form and functionbetweendisparately related taxa.Herewepresent fos-sil evidence for an unexpected convergence betweengregarious plant-eating mammals and dinosaurs.Recent excavations of Late Pleistocene deposits onRusinga Island,Kenya, haveuncoveredacatastrophicassemblage of the wildebeest-like bovid Rusingoryxatopocranion. Previously known from fragmentarymaterial, these new specimens reveal large, hollow,osseous nasal crests: a craniofacial novelty for mam-mals that is remarkably comparable to thenasal crestsof lambeosaurine hadrosaur dinosaurs. Using adultand juvenile material from this assemblage, as wellas computed tomographic imaging, we investigatethis convergence frommorphological, developmental,functional, and paleoenvironmental perspectives. Ourdetailed analyses reveal broad parallels betweenR. atopocranion and basal Lambeosaurinae, suggest-ing that osseous nasal crests may require a highlyspecificcombination of ontogeny, evolution, andenvi-ronmental pressures in order to develop.

RESULTS AND DISCUSSION

Abbreviated DescriptionRusingoryx atopocranion was first described from a partial skull

(National Museum of Kenya [KNM]-RU-10553A) collected from

Current Biology 26, 503

the Wasiriya Beds at the Wakondo locality of Rusinga Island,

Kenya (Figure 1; [1]). The holotype consists of basicranial, oc-

cipital, and frontal regions, as well as a right horncore [1]; how-

ever, the facial region is not preserved. Due to 3 cm of upwardly

angled nasal bones, it was initially postulated that Rusingoryx

may have had a proboscis or an enlarged, domed nasal region

[1]. Due in part to the incompleteness of the type specimen, this

reconstruction was dismissed and the taxonomic validity of Ru-

singoryx was questioned [2] until a more recent reexamination

and expanded sample of horn cores and dental material [3].

Here we present six new Rusingoryx cranial specimens exca-

vated en masse from a channel deposit within the type locality

[4]. These new specimens reveal the presence of an extreme,

osseous nasal dome and associated anatomical features

unparalleled among extant vertebrates. Because there are

numerous specimens across a spectrum of ages, from old to

young, this dataset facilitates ontogenetic, evolutionary, and

functional analyses of this bizarre Late Pleistocene bovid

morphology. Our analyses focus on KNM-RU-52572, a com-

plete adult skull that includes all craniofacial bones, maxillary

dentition, and the right horncore. For within-species compari-

son, we refer to skulls of an �12- to 16-month-old juvenile

(KNM-RU-59434) from the Wakondo assemblage and a puta-

tive female referable to the genus Rusingoryx (KNM-HA-

54933) with a diminutive nasal crest collected from the nearby

Pleistocene (�285,000–40,000 years ago) Luanda West locality

on the Homa Peninsula. Interspecific comparisons were also

made across extinct and extant Alcelaphinae, with intracranial

morphology also studied for Megalotragus isaaci (paratype:

KNM-ER-2000).

KNM-RU-52572 is post-depositionally compressed in the

sagittal plane. The caudal portion of the cranium matches

the holotype, and the dentition follows recent descriptions of

fragmentary material [3]. In mature individuals, the frontal,

nasal, and premaxillary elements form a prominent, osseous

nasal dome. Palatal morphology, best exemplified in compar-

ative material from the Homa Peninsula (KNM-HA-54933;

–508, February 22, 2016 ª2016 Elsevier Ltd All rights reserved 503

Figure 1. Rusingoryx atopocranion

(A–D) KNM-HA-54933, putative Rusingoryx female from Homa Peninsula.

(A) Skull, right lateral view.

(B) Digital rendering, with reconstruction of the supralaryngeal vocal tract (SVT) in yellow.

(C and D) Coronal and sagittal CT sections demonstrating enlarged maxillary sinuses and the course of the nasal tract.

(E–H) KNM-RU-52571, adult R. atopocranion from Wakondo.

(E) Skull, right lateral view.

(F) The SVT ascends parallel to the outer contour of the skull, departing from the hard palate near the caudal border of the premaxilla. After reaching its apex, the

SVT descends toward the internal nares in an ‘‘S’’-shaped curve.

(G and H) Coronal and sagittal CT sections.

(I–L) KNM-RU-59434, juvenile R. atopocranion excavated from Wakondo.

(I) Skull and mandible in right lateral view.

(J–L) Digital reconstruction, coronal, and sagittal CT sections (respectively) of the SVT demonstrate ontogenetic differences between adult (KNM-RU-52571) and

juvenile R. atopocranion. The nasal portion of the SVT is deeper in the juvenile, and the caudal portion is relatively more voluminous.

Scale bars, 4 cm. Abbreviations: aPSep, anterior palatine sinus septum;ms,maxillary sinus; MSep, maxillary sinus septum; NFL, nasal floor; NSep, nasal septum;

pPSep, posterior palatine sinus septum (anterior wall of airway); ps, palatine sinus. See also Figures S1 and S2.

Figures 1A–1D and S1), is likewise exaggerated. The hard pal-

ate projects deeply caudal to the lateral palatal indentations

and bears deep grooves corresponding to attachment sites

for palatal and supralaryngeal musculature (Figure S1). The

horizontal palatine bone is completely closed and does not

allow for the passage of the internal nares. Instead, the airway

passes over inflated maxillopalatine paranasal sinuses, result-

ing in a nasal passage that enters the pharynx from the dorsal

aspect (Figure 1; Movie S1). For a detailed description of

R. atopocranion cranial morphology, see Supplemental Exper-

imental Procedures.

Internal morphology reveals further peculiarities. The floor of

the nasal tract significantly departs from the hard palate,

504 Current Biology 26, 503–508, February 22, 2016 ª2016 Elsevier

ascending parallel to the outer surface of the nasal dome and

serving as a roof to the extraordinarily voluminous maxillopala-

tine paranasal sinuses. These sinuses comprise the largest pro-

portion of cranial volume and extend to the midline, where they

are separated by a median septum (Figures 1C and 1G). Based

on careful examination of computed tomography (CT) scans, the

paranasal sinuses and the airway are not connected by ostia

larger than 2 mm, rendering them dead airspace (Movie S1).

The nasal tract is straight through the rostral portion of the cra-

nium but deviates ventrally in the frontal region, forming an

‘‘S’’-shaped curve. This morphology has not been observed pre-

viously in mammalian taxa. The closest analog is, instead, the

nasal tract of basal lambeosaurine hadrosaur dinosaurs [4–10].

Ltd All rights reserved

Figure 2. Ontogenetic Similarities between

Nasal-Domed Alcelaphine Bovids and Lam-

beosaurine Hadrosaur Dinosaurs

(A and B) Juvenile R. atopocranion, KNM-RU-

59434 (A). The approximate age of the specimen is

12–16 months. The developing dome is comprised

predominantly of nasal bones, and its caudal

border is anterior to the orbit. This is in contrast to

adult R. atopocranion (B), where the dome is largely

comprised of the frontal and nasal bones dorsally,

and expanded lacrimal and maxillary bones later-

ally. The caudal border of the dome has migrated

caudally, situated above the posterior border of the

orbit. The frontal becomes angulated.

(C and D) A grossly similar pattern is seen in the

ontogeny of Hypacrosaurus altispinus. The caudal

border of the incipient juvenile crest (C) (Canadian

Museum of Nature CMN-2247) migrates from

anterior to posterior relative to the orbit. In the adult

(D) (CMN-8501), the crest is largely supported by

caudolateral portions of the premaxilla, and pos-

terior migration is contingent on flexion of the pre-

frontal and frontal bones. The airway, highlighted in

yellow, is presented for comparison with Figure 1;

redrawn from [6].

Scale bars, 4 cm. Abbreviations: Fr, frontal; Jg,

jugal; Lac, lacrimal; Mx, maxillary; Na, nasal; Occ,

occipital; PMx, premaxilla; PrF, prefrontal; Q,

quadrate; Sq, squamosal.

In addition to gross morphology, the following investigations of

ontogeny, evolution, and function of the Rusingoryx nasal

dome also reveal broad-scale similarities between Alcelaphinae

and Lambeosaurinae.

Juvenile MorphologyIn addition to mature individuals, the assemblage contains the

complete skull of a juvenile (KNM-RU-59434), ontogenetic age

approximately 12–16 months (Supplemental Experimental Pro-

cedures). This skull bears an incipient crest with a slight degree

of angulation rostral to the frontonasal suture. The nasal bones

are slightly elevated, and the premaxilla contacts the maxilla

and nasals on its caudal border. This suggests that the crest

shifts caudally throughout maturation, such that in adults it initi-

ates posterior to the caudal margin of the orbit (Figure 2). The

maxilla heightens throughout development, ultimately forming

the lateral wall of the crest. Internally, the angle between the floor

of the nasal cavity and hard palate becomes more obtuse, and

the portion of the chamber enclosing maxilloturbinates in the ju-

venile is relatively larger than that of adults. In these respects, the

suite of apparent ontogenetic changes in Rusingoryx show

remarkable similarities to lambeosaurine hadrosaur crest devel-

opment ([7]; Figure 2). In both groups, the crest enlarges by (1)

increased angulation of the analogous prefrontal and frontal

bones (hadrosaur and bovid, respectively; herein abbreviated

as the (pre)frontal elements) (2), simultaneous elongation and

caudal migration of the rostral facial bones, and (3) distension

of lateral facial elements [7].

Current Biology 26, 503

EvolutionTo evaluate the evolutionary trajectory of nasal dome acquisi-

tion, we incorporated new Rusingoryx and referred Homa Penin-

sula specimens into a phylogenetic analysis of extinct and extant

alcelaphines (Figure S2; Supplemental Experimental Proce-

dures). Subsequent character-mapping nasal morphology illu-

minates the stepwise morphological acquisition of the nasal

dome. The most primitive nasal-domed alcelaphine in our phy-

logeny, Megalotragus isaaci, possesses a diminutive nasal

dome with internal nares that enter the cranium anteriorly (Fig-

ure 3). Next in succession,M. kattwinkeli demonstrates that crest

elevation is preceded by a dorsal migration of the nasal pas-

sages (Figures 2 and 3). Subsequent elaboration of crest

morphology is predominately accomplished by inflation of lateral

facial bones and sinuses. Aswith ontogeny, thesemorphological

changes mirror the evolutionary transition from primitive to

derived lambeosaurine hadrosaurs, wherein dorsal translation

of the internal nares occurs prior to the heightening and elonga-

tion of the rostral and lateral facial bones ([8]; Figure 3).

FunctionBecause the nasal dome and tortuous nasal tract are novel

structures within Mammalia, direct functional comparisons with

living taxa cannot be made. Plausible functional interpretations

were therefore based on morphological inferences and mathe-

matical models. As Rusingoryx is an open-grassland species

[11], we prioritized thermoregulation [12], visual display [12],

and vocalization [9, 10] as functions particularly relevant to

–508, February 22, 2016 ª2016 Elsevier Ltd All rights reserved 505

Figure 3. Comparative Evolution of Nasal Crests between Lambeosaurinae and Alcelaphinae

Lambeosaurinae are in black text and Alcelaphinae in purple text. Trait acquisition of the nasal dome in both groups follows grossly similar patterns. In the earliest

members of both groups, the nasal passages enter the cranium horizontally, and the (pre)frontal elements reflect the primitive condition. Frontal elements then

begin to telescope as the nasal passages migrate to enter the pharynx dorsally. This sequence initiates crest formation. Lateral facial elements, including the

premaxilla of hadrosaurs and the maxilla of alcelaphines, are then recruited to form the lateral walls of the nasal vestibule. Crest elaboration is at its most derived

when the caudal margin of the crest rotates, forming an acute angle with the skull roof. Internally, the nasal airway must contort to follow the roof of the nasal

passage. Black elements represent structures that are not preserved. Parasaurolophini presented for consistency with source; redrawn from [8]. See also

Figure S3.

increased fitness in climatologically arid and behaviorally gregar-

ious contexts. These functions have also been hypothesized for

the nasal domes of lambeosaurine hadrosaurs [9, 10, 12],

enabling another means of comparison between these groups.

We first looked for evidence that the Rusingoryx nasal dome

houses exceptional variants of osseous structures related to

thermoregulation. In mammals, cranial thermoregulation is

achieved via evaporative cooling across the mucosae of the

maxilloturbinates [13]. Elaboration of the maxilloturbinates is

positively correlated with an increased demand to conserve res-

piratory water [13, 14], such that filling its enlarged nasal cham-

ber with a higher surface area of maxilloturbinates could be

beneficial for the thermal and hydrological budgets of Rusin-

goryx. However, because the elevated nasal cavity floor restricts

the volume of the dome available for turbinates (Figure 1; Movie

S1), the intracranial morphology of Rusingoryx is not consistent

with a hypothesis of enhanced turbinate function. In addition to

turbinates, circulating air through the large maxillary sinuses

may also dissipate heat; however, there is little evidence for

the requisite airflow between the sinuses and nasal passages

that would facilitate evaporative cooling [15]. Like all mammals,

the maxilloturbinates of R. atopocranion would have assisted

506 Current Biology 26, 503–508, February 22, 2016 ª2016 Elsevier

with thermoregulation, but there is little conclusive evidence

that the elaborate nasal dome enhanced these capacities

beyond those of other bovids.

We next evaluated the potential for visual display, which

among bovids is often realized in the form of male-male honest

or combative signaling ornaments, like horns. Given the propen-

sity for head-striking behavior among artiodactyls [16, 17], the

dome could be used for combative display. Nasal dome

morphology, however, is less congruent with such a hypothesis:

the narrow depth of the skull bones and voluminous, non-

buttressed cranial cavities (Figure 1; Movie S1) yield a skull

unlikely to have withstood substantial battery. Along with the

peculiar, posterior positioning of the horns (Figures 1 and S1),

it is unreasonable to implicate the nasal dome as having an effec-

tive role in aggressive male-male combat.

Competition among males, as well as intra- and interspecific

signaling, can also be accomplished vocally, and there are

numerousmorphological details that substantiate the hypothesis

that the nasal domemay play a role in long-distance vocalization

and/or sonic display. Externally, the hard palate possesses

deep, bilateral grooves for attachment of the muscles that

shorten the pharynx and elevate the soft palate and larynx

Ltd All rights reserved

Table 1. Reconstructed Resonant Frequencies

Taxon Specimen SVT Length n = 1 n = 2 n = 3 Note Range

Rusingoryx atopocranion KNM-RU-52572 0.74026 m 248.872 Hz 497.744 Hz 746.616 Hz B3–F4

cf. Rusingoryx female KNM-HA-54933 0.42371 m 390.597 Hz 781.194 Hz 1,171.791 Hz G4–D5

Megalotragus isaaci KNM-ER-2000 0.64147 m 258.001 Hz 516.002 Hz 774.003 Hz C3–G5

Values are based on the length of the osseous supralaryngeal vocal tract (SVT) andwere calculated using equations described in Supplemental Exper-

imental Procedures.

(Figure S1;Movie S1). In addition, the basicranium houses robust

muscular tubercles for insertion of the rostral pharyngeal con-

strictors (Figures S1 and S2; Movie S1). This muscle configura-

tion is suggestive of an exceptional ability to lift the larynx toward

the nasal tract. Such laryngeal elevation is also observed in

bovids that vocalize through fleshy nasal vestibules [18–21].

Furthermore, the large maxillary and palatal sinuses may have

amplified vocal output by acting as resonating chambers [20, 22].

Digital renderings of the osseous supralaryngeal vocal tract

(SVT; Figure 1) enabled quantitative assessment of the potential

for phonic modification using simple bioacoustic models [9, 10,

20, 23]. In mammals, the pharynx and nasal passages, together

with the SVT, modify the sound waves produced in the larynx by

vibrating preferentially at certain resonant frequencies that are

inversely proportional to vocal tract length [18–25]. Acoustic

models for harmonic wave production [9, 10, 24] suggest that

the Rusingoryx nasal dome could be capable of propagating

low-pitched sounds, between 248 and 746 Hz (Table 1; Supple-

mental Experimental Procedures). Because the soft-tissue vocal

tract length cannot be reconstructed, these estimated fre-

quencies are conservatively high. An addition of only 20 cm to

the vocal tract length would result in vocalizations below

200 Hz, closer to the range of infrasound. Comparatively, the

vocal ranges of many other mammals initiate around

�1,100 Hz [23], suggesting that Rusingoryx vocalizations are

low and potentially below the range detectable by predators.

Perhapsmost surprising is the almost complete overlap between

Rusingoryx resonant frequency reconstructions and those

calculated for lambeosaurine hadrosaurs [9, 10].

Environmentally, vocalization is a practical function for the

crest ofRusingoryx. In animals that form large or diffuse aggrega-

tions, efficient communication over long distances, particularly

through infrasonic vocalizations, may aid in predator avoidance

and intraspecific signaling [26]. Within contemporary Alcelaphi-

nae, wildebeests (Connochaetes spp.) and hartebeest (Alcela-

phus buselaphus) are found in large herds across semi-arid,

short-grass savannas of Sub-Saharan Africa. Social behavior is

integral in alcelaphines to the extent that elaborate horns, loud

vocalizations, and conspicuous coloration in adults suggest

that selection may favor species-specific social signals over

predator selection against conspicuousness [27]. Nasal vocali-

zations of Rusingoryx may similarly fall under this evolutionary

pretext. Paleoenvironmental evidence from the Wasiriya Beds

[3, 11, 28, 29] indicates widespread semi-arid grasslands, and

a markedly high degree of hypsodonty in Rusingoryx molars im-

plies a preference for such habitats [3]. Dry environments are

ideal for low-frequency vocalizations, as these sound waves

can travel over 10 km in such environments [26]. Potential selec-

tive pressures may also involve mitigation of the trade-offs

between intraspecific communication and predator avoidance

Current Biology 26, 503

because very low-frequency vocalizations may be outside of

the potential hearing range of many East African carnivores. It

has been hypothesized that similar habitat pressures may have

influenced vocal potential in lambeosaurine hadrosaurs [9],

with low-frequency vocalizations allowing these gregarious ani-

mals to communicate intraspecifically in open and closed

habitats.

Prior to this discovery, completely ossified nasal crests have

only been observed in Archosauria. The large nasal dome of Ru-

singoryx presents a new, mammalian context for studying these

elaborate structures and sheds light on the gross morphological

and developmental changes that predicate elevation of the nasal

skeleton. In two distantly related vertebrate groups, lambeosaur-

ine hadrosaurs and alcelaphine bovids, crest formation follows

homoplastic increases in maxillary height and dorsal migration

of the internal nares during both ontogeny and evolution. In

both groups, the crest is interpreted to produce vocalizations

across the analogous life history contexts of herd forming and

herbivory. Although similar behaviors are prevalent throughout

modern vertebrates, they are accomplished largely through

soft tissue structures. Just as telescoping of the nasal bones

and reducing the anterior facial skeleton is common to verte-

brates that have developed fleshy trunks, the comparable se-

quences of evolutionary and ontogenetic changes observed for

hollow dome formation (in conjunction with similar selective

pressures) suggests that there are potentially few sequences

of stepwise transformations that result in the formation of large,

hollow, osseous nasal crests. Expanding comparative studies to

additional taxa may begin to explain the rarity of such skull

morphology in other vertebrates.

SUPPLEMENTAL INFORMATION

Supplemental Information includes three figures, one table, Supplemental

Data, Supplemental Experimental Procedures, and one movie and can be

found with this article online at http://dx.doi.org/10.1016/j.cub.2015.12.050.

AUTHOR CONTRIBUTIONS

Conceptualization, H.D.O. and J.T.F.; Methodology, H.D.O.; Validation, K.E.J.,

D.J.P., C.A.T., T.W.P., and G.P.; Formal Analysis, H.D.O., J.T.F., and K.E.J.;

Writing – Original Draft, H.D.O.; Writing – Reviewing and Editing, J.T.F.,

K.E.J., D.J.P., C.A.T., T.W.P., G.P., and Z.L.J.; Visualization, H.D.O.; Supervi-

sion, C.A.T.; Project Administration, J.T.F., D.J.P., and C.A.T.; Funding Acqui-

sition, J.T.F., C.A.T., D.J.P., T.W.P.; Fieldwork, Y.-x.F., K.E.J., J.T.F., D.J.P.,

C.A.T., Z.L.J., R.J.-B., and G.P.

ACKNOWLEDGMENTS

This research was conducted under Republic of Kenya research permits

NCST/5/002/R/576 and NCST/RCD/12B/012/07. Funding was provided by

the National Geographic Society (CRE), R. Potts and the Human Origins

–508, February 22, 2016 ª2016 Elsevier Ltd All rights reserved 507

Program at the Smithsonian Institution, National Science Foundation (BCS-

1327047, BCS-1013199, BCS-1013108), Australian Research Council

(DP1092843, DP140100919, DP120101752, DE120101533), L.S.B. Leakey

Foundation, Paleontological Society, American Society of Mammalogists,

Geological Society of America, Society for Integrative and Comparative

Biology, American Museum of Natural History (AMNH), Baylor University, Har-

vard University, New York University, Ohio University, City University of New

York, the American School for Prehistoric Research, and University of Wollon-

gong. Excavation assistance: M.E. Macharwas, B. Ngeneo, J. Olelo, S. Od-

hiambo, L. Mbogori, M. Laird, D.O. Miriga, J.O. Oyugi, R. Onkondoyo, I.

Onyango, S. Odoyo, W.O. Oyoro, and J. Siembo. Specimen preparation:

E.S. Longoria. CT data collection: the radiology department at the Upper Hill

Medical Centre, Nairobi. Museum collection access: F.K. Manthi, E. Mbua,

P. Kiura, M. Munguu, I.O. Farah (National Museum of Kenya); P. Brewer (Nat-

ural History Museum of London); R.D.E. MacPhee, E. Westwig, and J. Galkin

(AMNH); and C. Argot (Museum National d’Histoire Naturelle). Sincere thanks

are given to the editor and two anonymous reviewers, as well as L.M. Witmer,

P. O’Connor, N.J. Stevens, D. Lieberman, E. Gorscak, A. Prieto-Marquez, C.

VanBuren, S. Reilly, G. O’Brien, and P. Gignac for input and editing assistance.

Received: September 1, 2015

Revised: November 18, 2015

Accepted: December 14, 2015

Published: February 4, 2016; corrected online February 22, 2016

REFERENCES

1. Pickford, M., and Thomas, H. (1984). An aberrant new bovid (Mammalia) in

subrecent deposits from Rusinga Island, Kenya. P. K. Ned. Akad.

Wetensc. B87, 441–452.

2. Harris, J.M. (1991). Koobi Fora Research Project, Volume 3 (Oxford

University Press).

3. Faith, J.T., Choiniere, J.N., Tryon, C.A., Peppe, D.J., and Fox, D.L. (2011).

Taxonomic status and paleoecology of Rusingoryx atopocranion

(Mammalia, Artiodactyla), an extinct Pleistocene bovid from Rusinga

Island, Kenya. Quat. Res. 75, 697–707.

4. Jenkins, K., Faith, J.T., Tryon, C.A., Peppe, D.J., Nightingale, S., Ogondo,

J., Johnson, C.R., and Driese, S. (2012). New excavations of a Late

Pleistocene bonebed and associated MSA artifacts from Rusinga Island,

Kenya. PaleoAnthropology 2012, A17.

5. Evans, D.C. (2006). Nasal cavity homologies and cranial crest function in

lambeosaurine dinosaurs. Paleobiology 32, 109–125.

6. Evans, D.C., Ridgely, R., andWitmer, L.M. (2009). Endocranial anatomy of

lambeosaurine hadrosaurids (Dinosauria: Ornithischia): a sensorineural

perspective on cranial crest function. Anat. Rec. (Hoboken) 292, 1315–

1337.

7. Evans, D.C. (2010). Cranial anatomy and systematics ofHypacrosaurus al-

tispinus, and a comparative analysis of skull growth in lambeosaurine ha-

drosaurids (Dinosauria: Ornithischia). Zool. J. Linn. Soc. 159, 398–434.

8. Prieto-Marquez, A., and Wagner, J.R. (2013). The ‘Unicorn’ dinosaur that

wasn’t: a new reconstruction of the crest of Tsintaosaurus and the early

evolution of the lambeosaurine crest and rostrum. PLoS ONE 8, e82268.

9. Weishampel, D.B. (1981). Acoustic analyses of potential vocalization in

lambeosaurine dinosaurs (Reptilia: Ornithischia). Paleobiology 7,

252–261.

10. Weishampel, D.B. (1997). Dinosaur cacophony: inferring function in extinct

organisms. Bioscience 47, 150–159.

508 Current Biology 26, 503–508, February 22, 2016 ª2016 Elsevier

11. Tryon, C.A., Faith, J.T., Peppe, D.J., Keegan, W.F., Keegan, K.N., Jenkins,

K.H., Nightingale, S., Patterson, D., Van Plantinga, A., Driese, S., et al.

(2014). Sites on the landscape: paleoenvironmental context of late

Pleistocene archaeological sites from the Lake Victoria Basin, equatorial

East Africa. Quat. Int. 331, 20–30.

12. Hopson, J.A. (1975). The evolution of cranial display structures in hadro-

saurian dinosaurs. Paleobiology 1, 21–43.

13. Negus, V. (1958). Comparative Anatomy and Physiology of the Nose and

Paranasal Sinuses (Williams & Wilkins).

14. Schmidt-Nielsen, K., Schroter, R.C., and Shkolnik, A. (1981). Desaturation

of exhaled air in camels. Proc. R. Soc. Lond. B Biol. Sci. 211, 305–319.

15. Badlangana, N.L., Adams, J.W., and Manger, P.R. (2011). A comparative

assessment of the size of the frontal air sinus in the giraffe (Giraffa camel-

opardalis). Anat. Rec. (Hoboken) 294, 931–940.

16. Skinner, J.D., and Chimimba, C.T. (2005). The Mammals of the Southern

African Subregion (Cambridge University Press).

17. Kingdon, J., and Hoffman, M. (2013). Pigs, Hippopotamuses, Chevrotain,

Giraffes, Deer and Bovids. Mammals of Africa, Volume VI (Bloomsbury

Publishing).

18. Frey, R., Volodin, I., and Volodina, E. (2007). A nose that roars: anatomical

specializations and behavioural features of rutting male saiga. J. Anat.

211, 717–736.

19. Kingswood, S.C., and Kumamoto, A.T. (1996). Madoqua guentheri.

Mamm. Species 539, 1–10.

20. Fitch, W.T. (2000). The phonetic potential of nonhuman vocal tracts:

comparative cineradiographic observations of vocalizing animals.

Phonetica 57, 205–218.

21. Efremova, K.O., Volodin, I.A., Volodina, E.V., Frey, R., Lapshina, E.N., and

Soldatova, N.V. (2011). Developmental changes of nasal and oral calls in

the goitred gazelle Gazella subgutturosa, a nonhuman mammal with a

sexually dimorphic and descended larynx. Naturwissenschaften 98,

919–931.

22. Alexander, G.D., Houston, D.C., and Campbell, M. (1994). A possible

acoustic function for the casque structure in hornbills (Aves:

Bucerotidae). J. Zool. (Lond.) 233, 57–67.

23. Fitch, W.T. (1997). Vocal tract length and formant frequency dispersion

correlate with body size in rhesus macaques. J. Acoust. Soc. Am. 102,

1213–1222.

24. Titze, I.R. (1994). Principles of Voice Production (Prentice Hall).

25. Rendall, D., Kollias, S., Ney, C., and Lloyd, P. (2005). Pitch (F0) and formant

profiles of human vowels and vowel-like baboon grunts: the role of vocal-

izer body size and voice-acoustic allometry. J. Acoust. Soc. Am. 117,

944–955.

26. Garstang, M., Larom, D., Raspet, R., and Lindeque, M. (1995).

Atmospheric controls on elephant communication. J. Exp. Biol. 198,

939–951.

27. Estes, R.D. (1991). Behavior Guide to African Mammals (University of

California Press).

28. Tryon, C.A., Tyler Faith, J., Peppe, D.J., Fox, D.L., McNulty, K.P., Jenkins,

K., Dunsworth, H., and Harcourt-Smith, W. (2010). The Pleistocene

archaeology and environments of the Wasiriya Beds, Rusinga Island,

Kenya. J. Hum. Evol. 59, 657–671.

29. Tryon, C.A., Peppe, D.J., Faith, J.T., Van Plantinga, A., Nightengale, S.,

and Ogondo, J. (2012). Late Pleistocene artefacts and fauna from

Rusinga and Mfangano islands, Lake Victoria, Kenya. Azania 47, 14–38.

Ltd All rights reserved