Maiabalaena nesbittae is 33 million year old fossil baleen whale from Oregon
•
Maiabalaena has neither teeth, nor baleen
•
Early whales lost teeth entirely before the evolutionary origin of baleen
•
Despite no teeth or baleen, these whales were effective suction feeders
Summary
Whales
use baleen, a novel integumentary structure, to filter feed; filter
feeding itself evolved at least five times in tetrapod history but
demonstrably only once in mammals [
].
Living baleen whales (mysticetes) are born without teeth,
but paleontological and embryological evidence demonstrate that they
evolved from toothed ancestors that lacked baleen entirely [
]. The mechanisms driving the origin of filter feeding in tetrapods remain obscure. Here we report Maiabalaena nesbittae
gen. et sp. nov., a new fossil whale from early Oligocene rocks of
Washington State, USA, lacking evidence of both teeth and baleen. The
holotype possesses a nearly complete skull with ear bones, both
mandibles, and associated postcrania. Phylogenetic analysis shows Maiabalaena
as crownward of all toothed mysticetes, demonstrating that tooth loss
preceded the evolution of baleen. The functional transition from
teeth to baleen in mysticetes has remained enigmatic because baleen
decays rapidly and leaves osteological correlates with unclear homology;
the oldest direct evidence for fossil baleen is ∼25 million years
younger [
] are inconsistent with the morphology and phylogenetic position of Maiabalaena. The absence of both teeth and baleen in Maiabalaena
is consistent with recent evidence that the evolutionary loss of teeth
and origin of baleen are decoupled evolutionary transformations, each
with a separate morphological and genetic basis [
].
Understanding these macroevolutionary patterns in baleen whales is akin
to other macroevolutionary transformations in tetrapods such as scales
to feathers in birds.
Cetacea; Pelagiceti; Neoceti; Mysticeti; Maiabalaena nesbittae gen. et sp. nov.
Etymology
Maiabalaena combines Maia-, meaning mother, and -balaena, meaning whale. Named for its phylogenetic position as basal to baleen-bearing mysticetes. The specific epithet nesbittae
honors Dr. Elizabeth A. Nesbitt for her lifetime of contribution to
paleontology of the Pacific Northwest and her mentorship and
collegiality at the Burke Museum of Natural History and Culture in
Seattle, Washington, USA.
Holotype
USNM
314627. Partial skeleton including a nearly complete cranium including
ear bones, mandibles, and hyoid elements; vertebrae; partial left and
right forelimbs; and manubrium (Figure 1, Figure 2, Figure 3, Figure 4 and S1–S4).
Figure 1Cranial Elements of the Holotype of Maiabalaena nesbittae, USNM 314627
Maiabalaena nesbittae
is diagnosed by the following combination of character states:
frontal-parietal sutures that converge posteriorly with the frontals
penetrating between the parietals; apex of the occipital shield
represents the dorsally highest part of the cranium; supramastoid crest
extending past the posterior margin of the temporal fossa but not to the
distal tip of the zygomatic process; length of the squamosal fossa is
less than three-fourths the length of the temporal fossa; triangular
coronoid process of the mandible that is anteroposteriorly longer than
it is dorsoventrally tall; and a spear shaped distal mandibular terminus
in lateral view.
Discussion
We recover Maiabalaena as the sister taxon to Sitsqwayk cornishorum, another edentulous mysticete from the Pacific Northwest [
].
This unnamed clade is united by the following combination of
synapomorphies: a pterygoid hamulus that is expanded into a
dorsoventrally flattened plate that partially floors the pterygoid sinus
fossa; an outer posterior prominence of the tympanic bulla that extends
posterior to the inner posterior prominence with the two separated by a
deep interprominential notch; a tympanic bulla with an inner posterior
pedicle present as a thin flange; a horizontal crest on the posterior
surface of the medial lobe of the tympanic bulla; a mandible with a deep
groove separating the mandibular condyle from the angular process; a
humeral head that is vertical in lateral; and a radius that is equal to
or longer than the ulna in proximodistal length.
Our phylogenetic analysis does not include Llanocetus or Mystacodon, both of which were described only recently and have not been observed by the authors. It also does not include Coronodon,
although it does include other toothed stem mysticetes from South
Carolina of similar morphology. The most recent phylogenetic analyses [
]
recover these three taxa in various positions relative to other toothed
mysticetes such as aetiocetids and mammalodontids, but no analysis
recovers them with edentulous mysticetes.
Our phylogenetic analysis recovers the clade of Maiabalaena + Sitsqwayk as the most basal branching lineage of toothless mysticetes (Figures 2 and S4).
The lack of adult mineralized teeth is interpreted from several
morphological features. First, the articulation of the mandible with the
cranium demonstrates that -Maiabalaena preserves a nearly
complete right palatal margin; notably, this palatal margin shows no
alveoli. Second, transverse CT scans corroborate an edentulous
interpretation by showing that the palatal margin lacks alveolar bone,
resembling that of edentulous mysticetes rather than toothed cetaceans (Figure 3). Third, Maiabalaena preserves a complete right mandible that also lacks alveoli along its dorsal border. The mandible of Maiabalaena
resembles those of edentulous mysticetes in lacking alveolar bone and
having a dorsally elevated mandibular canal in the body of the mandible [
Previous
hypotheses for the origin of baleen have attempted to infer the
presence of baleen in fossils from osteological correlates. In crown
mysticetes, deep palatal sulci on the ventral surfaces of the maxillae
accommodate structures that innervate and vascularize the tissue
overlying the baleen; identical sulci are absent in stem mysticetes,
although much smaller foramina in the same area have been proposed as
homologs, concurrent with the presence of multicusped, adult teeth on
the lateral margins [
].
However, these foramina are not present in all taxa within the relevant
clades, and they differ from the sulci of baleen-bearing mysticetes in
size, orientation, and overall morphology [
] (a stem cetacean). Here we identify multiple palatal foramina on the maxilla of two other basilosaurids, Basilotritus wardii and Zygorhiza kochii, and the stem odontocete Simocetus rayi [
].
The presence of these foramina in basilosaurids and a stem odontocete
demonstrate that the structures extend outside mysticetes altogether,
further casting doubt on their use for inferring baleen.
In extant mysticetes, the superficial sulci communicate internally with the superior alveolar canal (SAC) [
]
does communicate with the SAC. This connection is unsurprising given
that the SAC supplies the gingiva and upper dentition in all toothed
mammals, as well as baleen in extant mysticetes. The palatal foramina of
Maiabalaena do not visibly communicate with the SAC; instead,
they are shallow, superficial, and penetrate less than 5 mm into the
rostral bone (Figure 3).
Our observations may be limited by CT resolution, or it may be
attributed to the loss of alveolar bone and subsequent remodeling of the
palatal margin. The presence of palatal foramina in stem cetaceans and
odontocetes suggests that they supply the gingiva, as suggested by
previous authors [
]
and as seen in toothed mammals. Therefore, there is no evidence for
using palatal foramina to exclusively infer the presence of baleen.
Because all extant mysticetes possess baleen, phylogenetic bracketing [
]
provides a strong basis for inferring baleen in fossil taxa within
crown Mysticeti. However, there is insufficient evidence for inferring
baleen in stem mysticetes based solely on the absence of teeth [
]
outlined four independent, non-exclusive hypotheses for the origin of
baleen: dental filtration, medial baleen, posterior baleen, and suction
feeding. The dental filtration hypothesis was recently proposed for
another stem mysticete, Coronodon havensteini [
].
Other studies have called into question both the morphological
similarity to known dental filter feeders (e.g., crabeater seals, Lobodon) and the biomechanical viability of dental filtration in cetaceans [
]. Although the morphology observed in Maiabalaena
does not explicitly contradict the dental filtration hypotheses, its
lack of teeth more strongly supports other hypotheses instead.
The lack of evidence for both adult teeth and baleen in Maiabalaena is incompatible with the medial baleen [
],
both of which argue for an evolutionary stage during which teeth and
baleen are present at the same time. Each of the latter two hypotheses
have been criticized because they lack a clear functional basis for a
feeding mode that uses both structures simultaneously [
]. The age and phylogenetic position of Maiabalaena
suggests that the loss of teeth precedes the origin of baleen and
provides further reason to doubt both the medial and posterior baleen
hypotheses as transitional feeding modes along the lineage leading to
living mysticetes.
The absence of both teeth and baleen in Maiabalaena
is consistent with the hypothesis that tooth loss precedes the origin
of baleen using suction feeding as a transitional feeding mode [
]. In addition to the lack of a specialized feeding structure, Maiabalaena preserves a large and robust hyoid apparatus (Table S1), a structure that has been correlated with suction feeding specialization in all marine mammals [
] (and perhaps in an extant mysticete lineage, Eschrichtius robustus).
Suction feeding in odontocetes is often associated with short, broad
rostra and mandibles, a reduction in tooth number (or function), a
limited gape, a robust and expanded basihyoid bone (in terms of surface
area for muscle attachment) and teuthophagy (squid-eating [
However,
there are important exceptions to this cetacean ecomorph theme. Several
beaked whales (Ziphiidae), which are virtually edentulous, are able to
generate significant subambient pressure [
] despite relatively long narrow skull and jaws. Furthermore, there are numerous accounts of individual sperm whales (Physeter macrocephalus) that thrive as adults despite possessing a twisted, non-functional lower jaw [
]
is critical for producing subambient pressures at the rear of the
(albeit small) oral cavity. The orofacial morphology of pygmy sperm
whales (Kogia spp.), beluga whales (Delphinapterus),
and other odontocetes assists in generating significant subambient
pressures by occluding lateral gape and producing a rounded pipette-like
mouth opening [
].
For beaked whales, this specific orofacial morphology favors prey
capture via suction, which may be overlooked based on osteological
morphology alone.
We propose that Maiabalaena
used suction feeding as a transitional feeding mode, subsequent to
tooth loss and a raptorial biting prey capture mode—but prior to the
origin of baleen for filtering. Suction feeding was likely successful
via a combination of a robust hyoid (in similar size and shape to other
suction feeding cetaceans; see Table S1 and Data S1) and an orofacial morphology that occluded lateral gape similar to extant balaenids [
]. Size-corrected surface area measurements of fused cetacean basihyoid and thyrohyoid bones demonstrate that the hyoid of Maiabalaena
is substantially more robust than stem cetaceans; its surface area is
also greater than those of extant mysticetes and comparable to suction
feeding cetaceans (Table S1 and Data S1). Therefore, Maiabalaena was likely a capable suction feeder, if not a suction feeding specialist.
Notably,
this ecomorph (functional edentulism and suction feeding) has evolved
repeatedly in odontocetes; at least seven distinct lineages of
odontocetes have evolved to feed without the aid of any specialized
feeding structure (i.e., neither teeth nor baleen). This list includes
both stem odontocetes (Inermorostrum), as well as members of several distinct crown lineages including beaked whales, sperm whales, narwhals (Monodon), Risso’s dolphin (Grampus), the extinct walrus-convergent odontocete (Odobenocetops), and an extinct ziphiid-convergent delphinid (Australodelphis).
This repeated convergence on functional edentulism across multiple
lineages, each with distinct cranial and mandibular morphologies,
suggests that tooth loss is not only viable, but advantageous for
suction feeding.
At least three distinct lineages of toothed mysticetes, stemward from Maiabalaena, show evidence for some degree of suction feeding specialization (Mystacodon selenensis, Mammalodon colliveri, and an unnamed aetiocetid) [
]
categorized all stem mysticetes into two broad categories: toothed
forms employing suction-assisted raptorial feeding, and edentulous forms
filter feeding with baleen (Figure 4 in [
]). We recover Maiabalaena exactly at the phylogenetic juncture between these two categories; its position and our interpretation of Maiabalaena as a suction-feeder lacking both teeth and baleen evince a hypothesis proposed by Marx et al. [
Collectively, the clade of Maiabalaena + Sitsqwayk lies crownward of all toothed mysticetes but is stemward of all other edentulous mysticetes. Data from Maiabalaena directly informs three basic stages in the transition from teeth to baleen: (1) toothed mysticetes including Coronodon, llanocetids, mammalodontids, and aetiocetids; (2) functionally edentulous mysticetes also lacking baleen, including Maiabalaena and Sitsqwayk,
and potentially including more crownward, stem mysticetes such as
eomysticetids; and (3) edentulous mysticetes filter feeding with baleen,
likely including all crown mysticetes. Given that Maiabalaena forms a clade with Sitsqwayk, we tentatively infer Sitsqwayk as lacking both teeth and baleen, as well. Although Sitsqwayk lacks a rostral margin, the mandibles are well preserved and show no evidence of teeth [
Crownward of Maiabalaena + Sitsqwayk,
edentulous mysticetes include the extinct Eomysticetidae and crown
Mysticeti. Eomysticetids have traditionally been inferred as
baleen-bearing based on their phylogenetic position and the presence of
palatal foramina. However, palatal foramina are poor indicators of
baleen [
]. In the case of Yamatocetus,
teeth are inferred based on a scalloped palatal margin. However, this
scalloping is not clearly homologous to dental alveoli, nor does the
mandible preserve any evidence of a dentition, together showing no basis
for interpreting teeth in Yamatocetus. Two other eomysticetids, Tokarahia and Waharoa,
are more convincing: the former preserves an isolated tooth root
assigned to the genus, and the latter preserves apparent dental alveoli
at the distal tips of the mandible and rostrum. However, given that
neither taxon had teeth in situ, these authors leave open the possibility that neither taxon had an adult dentition [
].
The presence of teeth at the distal tip of eomysticetids is not
inconsistent with our hypothesis; eomysticetids clearly lacked a
functional dentition, reinforcing the hypothesis that the loss of a
functional dentition preceded the origin of baleen. Instead, the
presence of teeth in eomysticetids strengthens the comparison of Maiabalaena to beaked whales [
The
evolution of cetaceans is widely recognized as a textbook example of
macroevolutionary change documented by the fossil record; few other
vertebrate groups preserve such episodes of major evolutionary change.
In cetacean evolution, these phases include the transition from land to
sea in stem cetaceans, and evolutionary innovations associated with
crown cetaceans such as echolocation in odontocetes, and filter feeding
in mysticetes. In particular, filter feeding in baleen whales represents
an innovation without precedent among any other extant or extinct
mammalian group; explaining the origin of this complex feeding mode has
been a long-standing question since Darwin [
The
origin of filter feeding is an ecological shift that is documented by
macroevolutionary transformations, akin to transition from scales to
feathers in dinosaurs [
].
In each case, major morphological transformations are linked to
ecological transitions that fundamentally alter the natural history of
the groups in question. Our study demonstrates that suction feeding in
mysticetes occurred in functionally edentulous forms by the early
Oligocene. This loss of teeth likely paved the way for the subsequent
origin of baleen near the Oligo–Miocene boundary. The results of this
study support the decoupling of tooth loss from the origin of baleen in
whales; each represents a unique morphological transformation associated
with a distinct change in feeding ecology.
Further
information and requests for resources and reagents should be directed
to and will be fulfilled by the Lead Contact, Carlos Mauricio Peredo (cperedo@masonlive.gmu.edu).
Experimental Model and Subject Details
The description of Maiabalaena nesbittae (Data S1) is based on the holotype specimen, USNM 314627. Comparative material observed includes Aetiocetus cotylalveus (USNM25210), Aetiocetus polydentatus (cast of AMP 12), Aetiocetus weltoni (UCMP 122900), Chonecetus yabukii (cast of AMP 1), Chonecetus sookensis (NMC VP12095), Chonecetus tomitai (cast of AMP 2), Coronodon havensteini (3D model of CCNHM 108), Fucaia buelli (UWBM 84024), Fucaia goedertorum (LACM 131146), Janjucetus hunderi (NMV P216929), Mammalodon colliveri (NMV P199986), Sitsqwayk cornishorum (UWBM 82916), UWBM 82941, and UWBM 87135.
Institutional Abbreviations
AMP,
Ashoro Museum of Paleontology; CCNHM, Mace Brown Museum of Natural
History, College of Charleston, Charleston; LACM, Natural History Museum
of Los Angeles County; NMC, National Museum of Canada; NMV, National
Museum Victoria; UCMP, University of California Museum of Paleontology;
UWBM, Burke Museum of Natural History and Culture; USNM, Smithsonian
National Museum of Natural History.
Method Details
Digital Methods
We scanned the holotype skull and mandibles of Maiabalaena nesbittae
using Nikon Metrology’s combined 225/450ckV microfocus X-ray and
computed tomography (CT) walk-in vault system at National Technical
Systems in Belcamp, Maryland, USA, with a slice thickness of 0.03cmm.
Both the holotype skull and mandibles were scanned in storage cradles
mounted vertically with the posterior side oriented down to minimize
scanning width. The holotype bullae of Maiabalaena nesbittae
were scanned using Nikon Metrology’s 225ckV microfocus X-ray CT cabinet
system, also at National Technical Systems, with a slice thickness of
0.03cmm. We processed DICOM files from these scans in Mimics
(Materialise NV, Leuven, Belgium) to create three dimensional models of
the skull, right mandible, and left bulla. High density in the
basicranium hindered X-ray penetration, thus we also scanned the
holotype skull using an Artec Eva structured light scanner (Artec Group,
Palo Alto, California), scanning at 8 frames per second. These models
are available for viewing and downloading on Zenodo at the following https://doi.org/10.5281/zenodo.1415491
Phylogenetic Analysis
We tested the phylogenetic position of Maiabalaena nesbittae, using the same matrix as Peredo and Pyenson [
],
which already included USNM 314627 in the analysis. The final matrix
includes 86 operational taxonomic units and 363 total characters. We
performed a cladistic analysis using in TNT∗ [
]
using unordered and equally weighted characters. This analysis used the
‘traditional search’ option including 10,000 random addition sequences,
saving 10 trees per replicate. The analysis resulted in 610 most
parsimonious trees with a best score of 1587 steps. The final version of
this matrix is available as a separate file in the Supplemental
Information. A strict consensus tree showing all operational taxonomic
units is also available in the Supplemental Information (Figure S4).
This phylogenetic position of Maiabalaena relative to other stem mysticetes is critical to the results presented in this study. Our analysis recovers Maiabalaena as sister to Sitsqwayk, crownward of aetiocetids (and all toothed mysticetes) and basal to eomysticetids and other toothless mysticetes. Maiabalaena and Sitsqwayk
are united as a clade based on seven synapomorphies presented in the
main text. Here, we expand on the combination of character traits that
further distinguish these taxa from other stem mysticetes, namely,
eomysticetids.
The Maiabalaena + Sitsqwayk clade is sister to a clade that includes eomysticetids and all other edentulous mysticetes, including Horopeta
and crown Mysticeti. This clade is united in this analysis by the
following combination of characters: a premaxilla that is exposed in the
palate only anterior to the maxilla (character 11, state 1); contact
between the frontal and maxilla is loose with a developed groove
(character 44, state 1); maxilla-premaxilla contact is not sutured
(character 51, state 2); lacrimal is unsutured where it contacts the
maxilla and frontal (character 57, state 1); a roughly straight or
slightly concave dorsal edge of the orbit in dorsal aspect (character
69, state 0); optic groove positioned in the posterior third of the
supraorbital process (character 85, state 1); frontal positioned at the
same height as the nasals (character 86, state 1); stylomastoid fossa of
the periotic developed on much of the posterior “base” of pars
cochlearis (character 217, state 1); coronoid process of the mandible is
a triangular process with convergent anterior and posterior margins
that is higher than or equal to its length (character 265, state 2);
present gingival foramina on the mandible (character 268, state 1); an
absent ventral tubercle/hypophysis on the atlas and axis (character 297,
state 1); a humerus with an absent lesser tuberosity (character 329,
state 1).
Hyoid Surface Area Measurements
We measured the 2D surface area (in relative, scaled cm2)
of the fused basihyoid and thyrohyoids of select cetaceans in dorsal
view. Hyoids were photographed in dorsal view and proportionally scaled
to the same transverse width to standardize for size. 2D Surface area of
the scaled hyoids was measured in ImageJ. The stylohyoids were not
measured. Each specimen was measured three separate times; table S1
reports the mean values and their standard deviation, as well as each
specimen as a percentage of the total size of the largest hyoid in the
dataset (USNM 504345).
Data and Software Availability
The 3D models associated with this study are available for viewing and download on Zenodo at the following https://doi.org/10.5281/zenodo.1415491.
The final matrix used in the phylogenetic analysis for this study is
included with the supplementary materials associated with this article,
available online at https://doi.org/10.1016/j.cub.2018.10.047.
Acknowledgments
We
thank C.A. Sidor, R.C. Eng, and M.S. Rivin for coordinating access to
UWBM specimens and D.J. Bohaska, J.J. Ososky, M.R. McGowen and D.P.
Lunde for access to USNM specimens. We thank C. Peitsch, R. Peitsch, and
C. Schueler at National Technical Systems (Belcamp, Maryland), and S.B.
Sholts at the SIBIR for access to resources for CT scanning. We thank
E.A. Nesbitt for assistance with the stratigraphic position and age of
the holotype specimen. We thank J.H. Geisler, R.W. Boessenecker, M.
Brown, B.L. Beatty, and S. Boessenecker for access to the Coronodon havensteini 3D dataset, and the Imaging and Analysis Centre, Natural History Museum, London, for access to the Balaenoptera musculus 3D dataset. We thank NMNH Imaging for the photographs used throughout. Finally, we thank Alex Boersma (https://www.alexboersma.com) for the illustrations in Figure 2.
Author Contributions
All
authors contributed to the project planning. M.D.U. led preliminary
efforts at phylogenetic analysis and comparative and systematic work.
C.M.P. and N.D.P. coordinated CT scanning and modeling of digital data.
C.M.P. conducted the final phylogenetic analysis and led the comparative
and descriptive paleontology. C.M.P. and N.D.P. contributed to
discussions of marine mammal functional feeding modes. All authors
contributed to manuscript and figure preparation.
Anatomy,
feeding ecology, and ontogeny of a transitional baleen whale: a new
genus and species of Eomysticetidae (Mammalia: Cetacea) from the
Oligocene of New Zealand.
A functional comparison of the hyolingual complex in pygmy and dwarf sperm whales (Kogia breviceps and K. sima), and bottlenose dolphins (Tursiops truncatus).
A new genus and species of eomysticetid (Cetacea: Mysticeti) and a reinterpretation of ‘Mauicetus’ lophocephalus Marples, 1956: Transitional baleen whales from the Upper Oligocene of New Zealand.
Salishicetus meadi,
a new aetiocetid from the late Oligocene of Washington State and
implications for feeding transitions in early mysticete evolution.
31414-3&id=gr1.jpg){kind=link}
31414-3&id=gr2.jpg){kind=link}
31414-3&id=gr3.jpg){kind=link}
31414-3&id=gr4.jpg){kind=link}
Nenhum comentário:
Postar um comentário
Observação: somente um membro deste blog pode postar um comentário.