University of Southern California Dissertations and Theses

Dental evolution and tooth cycling in Mesozoic birds

(USC Thesis Other)

Dental evolution and tooth cycling in Mesozoic birds

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content DENTAL EVOLUTION AND TOOTH CYCLING IN MESOZOIC BIRDS
by
Yun-Hsin Wu
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(GEOLOGICAL SCIENCES )
May 2022
Copyright 2022 Yun-Hsin Wu
ii

Acknowledgements
This dissertation would not have been possible without the help from my funding agencies, colleagues,
friends, and mentors. I am sincerely grateful for Luis M. Chiappe’s trust, mentorship, encouragement,
and infinite support. I had dreamed about studying Mesozoic birds for years, and he made that dream
come true. I look forward to our continue collaboration in years to come. I would like to thank my
committee members, Cheng-Min Chuong, for his help with the Taiwan-USC Fellowship and generously
opening his lab for me ̶ the onset of the Evo-Devo trajectory of my research has greatly benefited from
his visionary perspectives; David J. Bottjer, for his invaluable academic and career advice and
encouragement over the past five years; Adam K. Huttenlocker, for the tremendous help and guidance
that laid the solid foundation of this project, and for generously sharing all the lab resources, which are
essential to this project. In addition, I would like to express my deepest gratitude to Ping Wu for
enhancing my knowledge and skills in conducting molecular works. He is a pioneer in alligator tooth
cycle research, and this project could not have been as successful without our continuous collaboration.
I would also like to thank Frank Corsetti for his helpful feedback and support throughout the years. The
exceptional academic support and the insightful comments, suggestions, and discussions from my
committee have been pillars that build up this dissertation. I am excited to further collaborating with
them to advance our knowledge about dental evolution with both paleontological and developmental
approaches.
I owe a special thanks to Hank Woolley, my dearest friend and colleague, for the delightful discussions,
brilliant insight, and for the support and encouragement through all the ups and downs in these years. I
would also like to thank Jeffery R. Thompson for the inspiring discussions and suggestions that helped
me pursue the multidisciplinary avenue and laid the phylogenetic foundation of my research. They have
been the source of great joy of my USC days, and I look forward to continuing working with them. I am
iii

furthermore grateful for my peers at the Department of Earth Sciences for being nothing but supportive
and loving. I have learnt so much from them.
I would like to thank the staff and curators at the NHMLA, especially Xiaoming Wang for sharing his
resource to help with the data analysis of this project, and Nate Smith for his guidance and suggestions
on phylogenetic analysis. I would also like to thank the past and present staff in the Department of Earth
Sciences for their support and facilitation of my PhD study. I am also indebted to Tea Jashashvili, Ivetta
Vorobyova, and Tautis Skorka from the Molecular Imaging Center at USC for their help with the CT scans
and data analyses.
Great thanks to all the curators and collection managers who made their museum collections accessible
for this research. These institutes include the Institute of Vertebrate Paleontology and
Paleoanthropology, the Beijing Museum of Natural History, the Sternberg Museum of Natural History,
and the Royal Tyrrell Museum. This research has also utilized data shared by Alida Bailleul, Bhart-Anjan
Bhullar, Daniel J. Field, Louchart Antoine, and the ESRF. Thanks are due to them for kindly sharing their
scan data, and Philip Donoghue and Chun-Chieh Wang for their help with the synchrotron scans.
I would like to thank the Ministry of Education of Taiwan and the Graduate School of USC. The Taiwan-
USC Fellowship by these two agencies secured a major financial source for my PhD study. Other financial
support includes the Department of Earth Sciences, the Dinosaur Institute at NHMLA, the Jurassic
Foundation, the Society of Vertebrate Paleontology, and the Society of Avian Paleontology and
Evolution. I am truly grateful for their contribution.
Finally, I would like to thank my beloved sister, Lily Tien-Jung Wu Wecker, for her love and support
throughout the years. She has pulled me out of the abyss for countless times, celebrated all the
highlights in my life, and helped me communicate with the rest of the world. I could not have gone this
far without her. This dissertation is dedicated to her.
iv

Table of Contents
Acknowledgements ....................................................................................................................................... ii
List of Tables ................................................................................................................................................. v
List of Figures ............................................................................................................................................... vi
Institutional Abbreviation ........................................................................................................................... viii
Abstract .........................................................................................................................................................ix
Chapter 1: Introduction ................................................................................................................................ 1
Chapter 2: Dentition Evolution of Mesozoic Birds: Character Mapping of the Number of Teeth on Each
Tooth Bearing Bone of the Stem Birds ........................................................................................ 7
Chapter 3: Dental Replacement Patterns and Tooth Cycles in Mesozoic Birds.......................................... 29
Chapter 4: Tooth Cycling Control in American Alligators (Alligator mississippiensis) ................................ 58
Chapter 5: Conclusion ................................................................................................................................. 83
References .................................................................................................................................................. 85
Appendices .................................................................................................................................................. 94
Appendix A: Character Matrix ....................................................................................................... 94

v

List of Tables
Table 2-1. Numbers of teeth on each tooth bearing bone of Mesozoic birds ........................................... 16
Table 3-1. Prenarial length (mm) of MPM-90 and MPM-373 ..................................................................... 47
Table 3-2. Total heights and erupted crown heights (mm) of teeth preserved in the three studied
Brazilian specimens .................................................................................................................... 48
Table 4-1. µCT schedule for Alli_2018_1 and Alli_2018_4 ......................................................................... 74

vi

List of Figures
Fig. 1-1. Tooth developmental stages and tooth cycle of the American alligator (A. mississippiensis) ....... 5
Fig. 1-2. Tooth growth tracking through µCT imaging .................................................................................. 6
Fig. 2-1. Maximum parsimony-based strict consensus tree of Aves .......................................................... 18
Fig. 2-2. Frequency distribution of the highest maxillary tooth count of the avian taxa included in the
present study .............................................................................................................................. 19
Fig. 2-3. Frequency distribution of the highest dentary tooth count of the avian taxa included in the
present study .............................................................................................................................. 20
Fig. 2-4. Parsimony reconstruction of changes in the highest numbers of teeth on premaxillae mapped
onto the phylogenetic topology of Chiappe et al. (2019) with dental characters excluded (Fig.
2-1b) ............................................................................................................................................ 21
Fig. 2-5. Parsimony reconstruction of changes of the teeth distribution on premaxillae mapped onto the
phylogenetic topology of Chiappe et al. (2019) with dental characters excluded (Fig. 2-1b).... 22
Fig. 2-6. Parsimony reconstruction of changes in the highest numbers of maxillary teeth mapped onto
the phylogenetic topology of Chiappe et al. (2019) with dental characters excluded (Fig. 2-1b)
.................................................................................................................................................... 23
Fig. 2-7. Parsimony reconstruction of changes in the highest numbers of dentary teeth mapped onto the
phylogenetic topology of Chiappe et al. (2019) with dental characters excluded (Fig. 2-1b).... 24
Fig. 2-8. Maximum likelihood reconstruction of changes in the highest numbers of premaxillary teeth
mapped onto the phylogenetic topology of Chiappe et al. (2019) with dental characters
excluded (Fig. 2-1b) .................................................................................................................... 25
Fig. 2-9. Maximum likelihood reconstruction of changes of the teeth distribution on premaxillae mapped
onto the phylogenetic topology of Chiappe et al. (2019) with dental characters excluded (Fig.
2-1b) ............................................................................................................................................ 26
Fig. 2-10. Maximum likelihood reconstruction of changes in the highest numbers of maxillary teeth
mapped onto the phylogenetic topology of Chiappe et al. (2019) with dental characters
excluded (Fig. 2-1b) .................................................................................................................... 27
Fig. 2-11. Maximum likelihood reconstruction of changes in the highest numbers of dentary teeth
mapped onto the phylogenetic topology of Chiappe et al. (2019) with dental characters
excluded (Fig. 2-1b) .................................................................................................................... 28
Fig. 3-1. Photographs of the enantiornithine specimens MPM-90, MPM-373, and MPM-351, and a
simplified cladogram highlighting the stem avian taxa discussed in this study ......................... 49
Fig. 3-2. Photographs and closeup of enantiornithine specimen MPM-373 .............................................. 50
Fig. 3-3. Angle measurements of MPM-373 and MPM-351 ....................................................................... 51
Fig. 3-4. Ventral views of MPM-90 (a) and MPM-373 (b), and cross sections (a’-a”, b’-b”) of the
premaxillae ................................................................................................................................. 52
Fig. 3-5. Dentition of MPM-90, MPM-373, and MPM-351 visualized through µCT imagining ................... 53
vii

Fig. 3-6. Dentition of MPM-90, MPM-373, and MPM-351 visualized through µCT imagining ................... 54
Fig. 3-7. Photographs of FHM 2503 ............................................................................................................ 55
Fig. 3-8. Tooth cycle of Archaeopteryx (SNSB BSPG VN-2010/1), Yanornis (IVPP V13358), and Ichthyornis
(FHM 2503) visualized through the cross section of their µCT data .......................................... 56
Fig. 3-9. µCT images showing teeth that have grown in alveoli and porous interdental bone .................. 57
Fig. 4-1. Schematic dentary of an enantiornithine contrasting Edmund’s wave of stimuli hypothesis (a)
vs. Osborn’s zone of inhibition hypothesis (b) for control mechanisms of reptilian tooth
replacement patterns ................................................................................................................. 75
Fig. 4-2. Juvenile alligator being anesthetized and scanned with Rigaku CT Lab 90 for growth tracking of
different tooth generations ........................................................................................................ 77
Fig. 4-3. 3D reconstruction from the µCT scans to track the growth of teeth at the positions rostral,
between, and caudal to the dental lamina ablation sites of Alli_2018_1 .................................. 78
Fig. 4-4. Tooth growth tracking records of the lower jaws of Alli_2018_1 at tooth position rostral,
between, and caudal to dental lamina ablation sites ................................................................ 79
Fig. 4-5. 3D reconstruction from the µCT scans to track the growth of teeth at the positions rostral,
between, and caudal to the dental lamina ablation sites of Alli_2018_4 .................................. 80
Fig. 4-6. Tooth growth tracking records of the lower jaws of Alli_2018_4 at tooth position rostral,
between, and caudal to dental lamina ablation sites ................................................................ 81
Fig. 4-7. Immunohistochemistry staining of sfrp1, β-catenin, and DAPI in dental lamina and tooth buds
during the pre-initiation (a-f), initiation (g-i), and growth (j-l) stages of the tooth cycle in A.
mississippiensis ........................................................................................................................... 82

viii

Institutional Abbreviation

BSPG: Bayerische Staatssammlung für Paläontologie und Geologie
BMNH: Beijing Museum of Natural History
FHM: Fort Hays State University’s Sternberg Museum of Natural History
IVPP: Institute of Vertebrate Paleontology and Paleoanthropology
MPM: Museu de Paleontologia de Marília

ix

Abstract

A diversity of toothed birds has been discovered since the 19
th
century and these have been shown to
have multiple generations of teeth (polyphyodonty). Despite detailed descriptions of the osteology of
toothed birds, no rigorous investigation of their dental evolution and tooth cycles has been undertaken.
In the present study, I have investigated and mapped dental characters onto an avian phylogenetic tree
for ancestral state reconstructions using parsimony and maximum likelihood methods. The results of the
character mapping support the modularity of dentition distribution and variation of tooth counts on
each tooth bearing bone. Previous studies on avian tooth replacement have been limited to sparse data
from tooth roots revealed through broken jawbones and detached teeth. However, detailed
descriptions of their tooth cycles are lacking, and the specifics of their replacement patterns remain
largely unknown. Here I present unprecedented data of the reconstructed tooth cycles from specimens
of Archaeopteryx, enantiornithine birds, Ichthyornis, and Yanornis through high-resolution X-ray imaging
methods. The cross sections of the teeth of these birds show that they migrated labially toward the
functional teeth in their early developmental stages. This result deviates from the known patterns of
non-avian theropods and show a homoplastic similarity with crocodilians. The high-resolution data also
reveal an alternating dental replacement pattern in the studied enantiornithine specimens, which is
consistent with previous studies in Archaeopteryx. These results imply a conserved control mechanism
underlying the tooth cycle and dental replacement in archosaurs. Hence, developmental experiments on
modern alligators were used to test two hypotheses attempting to explain these mechanisms: Edmund’s
wave of stimuli hypothesis and Osborn’s zone of inhibition hypothesis. The results of these
experiments—namely dental lamina ablation and growth tracking—do not support either of these two
x

long-standing hypotheses. This study is the first to adapt Evo-Devo approaches to investigate the
evolution and mechanism of tooth cycling in fossil archosaurs.
1

Chapter 1: Introduction

Modern birds (Neornithines) are all toothless, possessing a wide variety of keratin sheaths
(rhamphotheca) covering their edentulous beaks. However, since the 19
th
century it has been known
that different lineages of stem birds from the Mesozoic Era were toothed (Dames, 1884; Evans, 1865;
Marsh, 1880). Resorption pits on the roots of the teeth of these birds have also been reported (Dames,
1884; Howgate, 1984; Martin et al., 1980; Martin and Stewart, 1999). These are known for several
specimens of the Late Jurassic Archaeopteryx (Berlin and London specimens) (Howgate, 1984; Martin et
al., 1980), thus indicating that the earliest known birds had tooth replacement. Similar discoveries have
also been done for more advanced toothed birds. Marsh (1880) not only reported the morphology,
including histological structures, of the dentition of Late Cretaceous Hesperornis and Ichthyornis, but
also described their tooth replacement and compared it to other reptiles. Along with these historical
findings, increasing numbers of dentulous stem birds from the Cretaceous have been made in all major
continents except Antarctica (Chiappe and Witmer, 2002). Since the early 1990s, the majority of these
discoveries have come from the Early Cretaceous Jehol Biota, a series of remarkable lacustrine
Lagerstätten in Northeast China (Zhou and Wang, 2010; O’Connor and Chiappe, 2011; Chiappe and
Meng, 2016; Wang et al., 2017). However, while the range of fossil bird species and their skeletal
morphology have been studied in detail, the evolution of their dental morphology, the replacement
process and pattern of their dentition, and its underlying mechanism have been hugely neglected.
The phenomenon of having multiple generations of teeth (and tooth replacement) throughout the
toothed birds’ life is called polyphyodonty (Bertin et al., 2018). Polyphyodonty involves continuous
regeneration of new teeth and repeated tooth cycles, which requires maintaining an active dental
lamina. Such dental lamina contains a cluster of stem cells (dental lamina bulge) that are collectively the
2

source of tooth renewal (Järvinen et al., 2009; Handrigan et al., 2010; Wu et al. 2013). The dental
developmental stages, based on morphological changes of the dental lamina throughout an
odontogenesis process, have been studied in mammals and alligators. Crocodilians, including alligators,
are the closest living relatives to birds, having a similar tooth implantation to the toothed birds of the
Mesozoic Era. Both of them have thecodont implantation, in which teeth are set in tooth sockets, or
alveoli (Leblanc et al., 2017; Dumont et al., 2016). These features, and similarities to toothed birds, make
crocodilians the ideal model for shedding light on the mechanisms and controls of tooth cycle and
dental replacement of these extinct toothed birds.

Summary of tooth developmental stages from studies of modern archosaurs and mammals
• Tooth Cycle in Alligator mississippiensis (Fig. 1-1)
- Pre-initiation: equals to the bud stages defined from mammalian studies. The distal end of the dental
lamina forms a dental lamina bulge. Stem cells for odontogenesis are concentrated here. In this stage,
the dental lamina is undifferentiated.
- Initiation: equals to the cap stages defined from mammalian studies. The dental lamina starts to
differentiate and folds to form the enamel organ and dental papilla of a tooth germ.
- Early growth: equals to the bell stages defined from mammalian studies. The tooth germ further
differentiates, and the enamel organ and dental papilla form a structure of multiple layers.
- Growth: equals to the apposition stages defined from mammalian studies. Biomineralization of
enamel, dentine, and cementum starts to accumulate, forming layered structures. Resorption by
odontoclasts on the root of functional teeth is significant at this stage.
3

- Maturation: The last stage of odontogenesis. The new tooth is formed and can later replace the
functional tooth. The dental lamina separates from the new tooth, completing a tooth cycle. The dental
lamina returns to the pre-initiation (bud) stage, resuming a new tooth cycle.
The above is a summary of the tooth developmental stages of alligators and mammals, as they share a
similar process of odontogenesis (Fehrenbach and Popowics, 2016; Wu et al., 2013). Wu et al. (2013) did
not define the detailed growth stage in alligators beyond its equivalence with the bell stage. Although
they did use the term early growth and growth stages separately in their study, they did not provide a
definition for each of these two stages. The description provided here is derived from comparisons with
the contextual information of these stages in Wu et al. (2013) and descriptions of mammalian tooth
development (Fehrenbach and Popowics, 2016). Maturation stage is also not specifically described for
alligators in previous studies, so the observations from the HE staining is also compared to mammalian
studies and summarized here (Fig. 1-1). At each tooth position, there is a functional tooth, a
replacement tooth, and a dental lamina. These three structures form a tooth family. After the
maturation stage, for the polyphyodonty animals, the tooth family can go into another tooth cycle
multiple times. For mammals, the tooth regeneration ends after the second cycle.

Conclusions
µCT scans of living alligators provide unprecedented data of a complete tooth cycle from a single tooth
position of a single animal (Fig. 1-2). Given that the secretion of enamel and dentine does not start until
the growth, or apposition, stage, this is likely the earliest stage we can identify the new forming tooth
from the CT scans in fossils. The toothed birds are all extinct, but we can still access their tooth
replacement information through high-resolution X-ray imaging methods (e.g., CT and synchrotron). The
tooth cycle record of living alligator permits comparison and reconstruction of tooth developmental
4

stages in extinct toothed birds. Results from such comparisons, along with studies of tooth numbers in
Mesozoic birds, provide the basis for the research presented in this dissertation.

5

Fig. 1-1. Tooth developmental stages and tooth cycle of the American alligator (A. mississippiensis).
Stage names in parenthesis are those used in mammalian odontogenesis. In the maturation stage,
dental lamina separates from the new tooth, returning to the bud stage (cyan arrow). aw: alveolar wall;
den: dentine; dl: dental lamina; dp: dental papilla; eo: enamel organ; FT: functional tooth; RT:
replacement tooth; T: tooth.

6

Fig. 1-2. Tooth growth tracking through µCT imaging. This figure shows a full tooth cycle over a 10-
month period in juvenile A. mississippiensis.

7

Chapter 2: Dental Evolution of Mesozoic Birds: Character Mapping
of the Number of Teeth on Each Tooth Bearing Bone of the Stem
Birds

Introduction
Modern birds (Neornithines) lack teeth. Edentulism is common to all neornithine birds, including fossils
dating back to the Late Cretaceous (Field et al., 2020). Some modern anatids have serrations along the
jaw tomia of their rhamphotheca (Feduccia, 1996); Odontopterygiformes, an extinct clade of
neornithines known from the Paleocene to the Pliocene, had bony pseudoteeth in their jaws that consist
of bony cores surrounded by a keratinized sheath instead of true dental tissue (Louchart et al., 2018).
Nonetheless, these neornithines all lack true dentition even if they possess structures that may have a
similar ecological function as teeth. There are currently two reliable records of neornithine birds from
the Late Cretaceous: Vegavis iaai (Clarke et al., 2005) and Asteriornis maastrichtensis (Field et al., 2020).
The specimen of Vegavis only preserved the postcranial parts, but the specimen of Asteriornis clearly
shows a toothless beak.
Chiappe et al. (1999) pointed out that edentulism has evolved at least three times in Aves and a few
other times in non-avian theropods. Louchart and Viriot (2011) investigated tooth loss in Aves and
suggested six independent edentulism events throughout the bird phylogenetic tree. They have also
hypothesized that partial reduction of dentition in Enantiornithines may have started from maxilla or
posterior end of premaxilla, and it may have started from the rostral portion of premaxilla in
Ornithuromorpha. It has become a consensus that tooth loss evolved independently numerous times
among stem birds, and no long-term selection toward edentulism has been found (Brocklehurst and
8

Field, 2021). The development of embryonic teeth in talpid
2
mutant chicken has implied that the
formation of rhamphotheca may be correlated to tooth loss (Harris et al. 2006). A survey of embryonic
dentitions also shows that the development of null-generation dentition is inhibited in all extant
tetrapods with occlusal keratinization (Wang et al. 2017). The rhamphotheca has been hypothesized as
homologous to caruncles in amniotes (Hieronymus and Witmer 2010), which are formed from the
epithelium on the rostral end of premaxilla (Wang et al. 2017). Field et al. (2018) inferred the coverage
of highly keratinized rhamphotheca from the neurovascular foramina of Ichthyornis premaxilla and
hypothesized the keratinized coverage on the beaks of other ornithuromorph birds. If the formation of
rhamphotheca ceases the development of dentitions, the birds with keratinized coverage on their beaks
shall have tooth loss starting from the rostral portion of their premaxilla.
The fact that Mesozoic birds had diverse dentition has been noticed (i.e., Hu et al., 2015; O'Connor and
Chiappe, 2011; O'Connor et al., 2013), but there has been no comprehensive study to assess the range
of the dentition distribution of these fossil birds. With the growing avian fossil record being discovered
and reported, the range of avian dentition is beyond typical descriptions. It has been pointed out that it
is typical for enantiornithines to have four premaxillary teeth (O’Connor and Chiappe 2011; Li et al.
2014). However, this assertion has not been examined under a phylogenetic scope. Also, there is no
holistic survey of the tooth number in other tooth-bearing bones and in other clades of birds, such as
Ornithuromorpha and the groups more basal than Ornithothoraces (“the basal groups”). I would like to
investigate the evolutionary history of tooth number in the jaws of Mesozoic birds, and their partial
reduction of dentitions, by reconstructing ancestral states for tooth numbers in the premaxilla, maxilla,
and dentary through parsimony and maximum likelihood approaches.

9

Materials and Methods
I. Selection of Tree Topology
The phylogenetic tree topology of Aves—including terminal taxa and character matrix—are based on
Chiappe et al. (2019). 50 taxa of Mesozoic birds are included in this analysis. The following characters
related to dentition have been excluded from the matrix:
Character 4: Premaxillary teeth: present throughout (0); present but rostral tip edentulous (1); present
but restricted to rostral portion (2); absent (3).
Character 7: Maxillary teeth: present (0); absent (1).
Character 26: Dentary teeth: present (0); absent (1).
Character 27: Robustness of teeth relative to dentary: anteroposterior width of largest tooth crowns
(measured at thickest portion of crown’s base) far less than half of dentary dorsoventral
depth (0); close to half (or more) of dentary depth, i.e., 45%, or more (1). Taxa near the
cutoff point are coded 0/1.
Character 28: Dentary tooth implantation: teeth in individual sockets (0); teeth in a communal groove
(1).
These dental characters are excluded to avoid the risk of circular reasoning and the logic fallacy of
assuming the conclusion in premises of the argument (de Queiroz 1996, 2000). Often when studying
character evolution, the character mapping methods would be applied to separately built phylogenetic
tree topologies (e.g. Barrett et al. 2015; Huttenlocker and Farmer 2017; Schluter et al. 1997; Smith 2012;
Tullberg et al. 2002). The remaining characters are then used to conduct another phylogenetic tree
topology using the same settings as indicated in Chiappe et al. (2019) for parsimony framework using
PAUP 4.0.

10

II. Character Mapping and Ancestral State Reconstruction
Among the 50 taxa included in this analysis, the skulls or jaws of the currently available specimens of
Concornis, Elsornis, Eoalulavis, Cathayornis, Neuquenornis, Patagopteryx, Gansus yumenensis, Vegavis,
and Bellulornis are either missing or too poorly preserved to determine whether these taxa were
toothed or not. The numbers of teeth in the premaxilla, maxilla, and dentary, and the distribution of
teeth in the premaxilla (i.e., character 4 in Chiappe et al. 2019), of the remaining 41 taxa were collected
through examination of the literature and/or direct observation of the specimens. The data and source
are summarized in Table 2-1. Based on this data matrix, the frequency distribution of the maxillary tooth
count was used to parse the numbers of the maxillary teeth into bins of character states (Fig. 2-2), and
the frequency distribution of the dentary tooth count was used to determine the bins of character states
in the numbers of the dentary teeth (Fig. 2-3). The following four dental characters are individually
mapped onto the phylogenetic topology built from non-dental characters (Fig. 2-1b) to trace their
character changes and reconstruct their ancestral states using both parsimony and maximum likelihood
schemes (Mk1 model). The ancestral states were reconstructed using Mesquite 3.61.
The four new dental characters used for character mapping and ancestral state reconstructions are
defined as follows:
Character 1: Number of premaxillary teeth (or alveoli): toothless premaxilla (0); 1-3 (1); 4 (2); more than
4 (3).
Character 2: Number of maxillary teeth (or alveoli): toothless maxilla (0); 1-6 (1); 7-10 (2); 11-12 (3); 13-
15 (4); 16 or more (5).
Character 3: Number of dentary teeth (or alveoli): toothless dentary (0); 1-5 (1); 6-10 (2); 11-19 (3); 20-
25 (4); 26 or more (5)
11

Character 4: Premaxillary teeth distribution: present throughout (0); present but rostral tip edentulous
(1); present but restricted to rostral portion (2); absent (3).
The character states of character 4 here follow the 4
th
character in Chiappe et al. (2019) but were
redefined as premaxillary teeth distribution. Most scores of the character states also follow Chiappe
(2019), with two modifications. Ichthyornis and Gansus zheni were scored as uncertain (?) in the
character matrix of Chiappe et al. (2019). Based on the observation from BMNH Ph-1342 and Ph-1392,
the premaxilla of Gansus zheni is considered toothless. The premaxilla of Ichthyornis is also considered
toothless following Field et al. (2018). For taxa that do not have data on a definite number of teeth for
each tooth bearing bones, the character states of character 1-3 were entered as uncertain (?) in the
character matrix. The character matrix is presented in appendix A.

Results
The topology of the strict consensus tree built from non-dental characters remains the same as the
topology of the parsimonious tree presented in Chiappe et al. (2019) (Fig. 2-1). It is typical that the
premaxillae of dentulous enantiornithines bear four teeth, as in non-avian theropods or basal birds like
Archaeopteryx and Sapeornis (O’Connor and Chiappe 2011; Li et al. 2014). Among the enantiornithine
birds, Longirostravis appears to have five premaxillary teeth which is the highest number of premaxillary
teeth (Table 2-1). Gobipteryx is the only enantiornithine taxon with edentulous premaxilla among the
taxa examined here. The results of both the parsimony and maximum likelihood reconstructions of the
highest numbers of premaxillary teeth and the premaxillary teeth distribution show that most
enantiornithine lineages have four premaxillary teeth and these teeth are presented throughout the
tooth bearing bone (Figs. 2-4, 2-5, 2-8, 2-9). These results indicate that the most recent common
ancestor of the Enantiornithines had premaxillary teeth distributed throughout the premaxilla (Figs. 2-5,
12

2-9) but that it may not have four premaxillary teeth as do most of its descendants (Figs. 2-4, 2-8). This
uncertainty is due to the uncertain numbers of teeth on premaxilla in Protopteryx. In both the
parsimony and maximum likelihood reconstructions, the clade including the dentulous Longipterygidae
enantiornithines (e.g., Longirostravis, Rapaxavis, Longipteryx, and Shanweiniao) and Concornis and
Elsornis stand out in the reconstruction of teeth distribution on premaxilla as the character state with
the teeth restricted on the rostral portion only happens in this clade. This clade also stands out in both
the parsimony and likelihood reconstruction of the highest number of the maxillary teeth for having
edentulous maxillae. Among the enantiornithine birds, this is the only group that has edentulous
maxillae other than the edentulous Gobipteryx (Figs. 2-6, 2-10). The clade with the highest maxillary
tooth count is Pengornis. The second highest is Eopengornis (Figs. 2-6, 2-10). These two birds were
considered to belong to the monophyletic clade, Pengornithidae (Hu et al. 2015). However, on the tree
topology used in this analysis (Fig. 2-1b) the typical taxa of Pengornithidae are not clustered to form a
monophyletic group. Only Eopengornis and Parapengornis are clustered in a monophyletic clade.
Maxillary teeth are not well-preserved on Parapengornis specimens, so the current available number of
maxillary teeth of this bird is underestimated. Even though Eopengornis and Pengornis both have high
numbers of maxillary teeth, Pengornis is grouped with other birds that either lack records of maxillary
tooth numbers or have edentulous maxillae. The Eopengornis + Parapengornis clade has the most
abundant dentary teeth (1-5 teeth) in the parsimony reconstruction of the highest number of dentary
teeth. This result is also consistent with the Likelihood reconstruction (Figs. 2-7, 2-11).
Among the ornithuromorph birds, Yanornis and Yixianornis both have five premaxillary teeth. This clade
of (Yanornis + Yixianornis) has the highest numbers of premaxillary teeth among all the 50 avian taxa in
both the parsimony and likelihood reconstructions (Figs. 2-4, 2-8). Unlike the clade that contains the
Longipterygidae taxa, the Yanornis + Yixianornis clade has an edentulous rostral tip (F-g. 2-5, 2-9).
Longicrusavis is another ornithuromorph bird that has premaxillary teeth, but the tooth count is
13

uncertain. Most ornithuromorph lineages are toothless on their premaxilla. In contrast to
Enantiornithines, the parsimony and maximum likelihood reconstruction of the most recent common
ancestor of ornithuromorph birds is most likely toothless on its premaxilla (Figs. 2-4, 2-8). This split
happens at the ancestor of Ornithothoraces, which is more likely to have toothed premaxilla (Figs. 2-4,
2-8). All the toothed ornithuromorph birds have a relatively high maxillary tooth count compared to
Enantiornithines or the birds ancestral to Ornithothoraces, especially the toothed members of
Ornithurine (Figs. 2-6, 2-10). Toothed ornithurine birds also have the highest numbers of the dentary
teeth in the reconstructions (Figs. 2-7, 2-11). In both the parsimony and maximum likelihood
reconstructions, the maxilla and dentary teeth focus on the same two lineages among the
ornithuromorphs: toothed ornithurine birds and the clade of Yanornis and Yixianornis.
In this study, lineages more basal than Ornithothoraces (Enantiornithes + Ornithuromorpha) include the
toothless Confuciusornis, Changchengornis, and Eoconfuciusornis (all belong to Confuciusornithidae)
and the toothed Archaeopteryx, Jeholornis, and Sapeornis. Both Archaeopteryx and Sapeornis have four
teeth on their premaxillae, and the teeth distributed throughout the bone. Jeholornis has been found to
have no teeth on its premaxilla (Figs. 2-4, 2-8). Because of their basal positions, this also affects
reconstruction of the ancestral state of the ancestor of the entire bird clade. The parsimony
reconstructions of character 1 and 4 shows that this ancestor can have either a toothed or edentulous
premaxilla, and while the maximum likelihood reconstructions also agree with that, it shows that this
ancestor is more likely to be edentulous than toothed on its premaxilla. The reconstructions of the
numbers of maxillary and dentary teeth both show that this ancestor is more likely to be toothed on
these tooth bearing bones (Figs. 2-6, 2-7, 2-10, 2-11). The likelihood of a dentulous dentary is especially
high (Fig. 2-11).

14

Discussion
This is the first phylogenetic analysis to assess the changes of tooth numbers and teeth distribution in
toothed birds through character mapping. The jaws were separated into different tooth bearing bones,
and the ancestral state reconstruction was run separately for each region. This modularity among jaw
regions may also underly patterns of tooth loss in birds (Brocklehurst and Field, 2021). It has also been
found that tooth loss in birds, including regional tooth loss, seems to follow Dollo’s law and has not been
reversed in their evolution (Brocklehurst and Field, 2021). Tooth loss is associated with tooth number,
and regional tooth loss can be represented in regional tooth counts (e.g., numbers of teeth on each
tooth bearing bone) and teeth distribution. The results of the ancestral state reconstruction of the four
dental characters in the present study also support the modularity of dental distribution of the avian
jaws. The results of the present study are consistent between the parsimony and maximum likelihood
framework, which indicates a more plausible and robust analysis.
Although it is most common for enantiornithines to have four premaxillary teeth, the tooth number in
the premaxilla appears to be variable in this group of birds, and the ancestor of Enantiornithines may
have had less than four premaxillary teeth (Figs. 2-4, 2-8), which disagrees with O’Connor and Chiappe
(2011) and Li et al. (2014). However, this result may be due to the uncertainty introduced by multiple
clades with uncertain numbers of teeth in the premaxilla. The condition in which the premaxillary
dentition is limited to the rostral portion of this bone is limited to longipterygids, which is a
synapomorphy of the clade (O’Connor et al., 2011). The premaxillae of Ornithuromorpha are either
edentulous or have edentulous rostral ends, which can be a feature supporting the formation of
rhamphotheca from the rostral tip of the premaxilla that may inhibit tooth formation (Wang et al.,
2017).
15

Tooth number mapping of each tooth-bearing bone was conducted under both parsimony and
maximum likelihood ancestral state reconstruction and using the strict consensus most parsimonious
tree. Future studies will continue to investigate the evolution of these dental characters while taking
branch lengths into consideration. Divergence time estimation will also be conducted using the available
geological time of the fossils (Wang and Lloyd, 2016). Bayesian ancestral state reconstruction will also be
applied to this analysis to provide character mapping with a different framework (Pagel and Barker
2004). In addition to the number of teeth, other characters to depict dental morphology will be studied
and mapped onto the Aves phylogenetic trees to provide a holistic view of the morphological disparity.

16

Table 2-1. Numbers of teeth on each tooth bearing bone of Mesozoic birds. Pmx: premaxilla; Mx:
maxilla; D: dentary. Pmx Dentition: Character 4 in Luis et al. 2019. The data for Gansus and Ichthyornis
was updated using the source specified. Specimens listed here are from the collections of the Institute
of Vertebrate Paleontology and Paleoanthropology (IVPP V) and the Beijing Museum of Natural History
(BMNH Ph). Data without specified source are from Chiappe et al. (2019). Concornis, Elsornis, Eoalulavis,
Cathayornis, Neuquenornis, Patagopteryx, Gansus yumenensis, Vegavis, and Bellulornis are not listed in
this table because their skulls or jaws do not have the necessary preservation to determine whether the
taxa have dentition. Pmx: premaxilla; Mx: maxilla; D: dentary.
Taxon\Character
Pmx
Tooth #
Mx
Tooth #
D
Tooth #
Source
Archaeopteryx 4 8(9) 11 Howgate 1984
Jeholornis 0 >1 2-3 Zhou and Zhang 2003
Sapeornis 4 3 2 Wang et al. 2017
Confuciusornis 0 0 0

Changchengornis 0 0 0

Eoconfuciusornis 0 0 0

Protopteryx >1 >3 >1 BMNH Ph-1060
Orienantius 4 >=3 >=6 Liu et al. 2019
Eopengornis 4 11 12-14 Wang et al. 2014
Parapengornis 4 >2 9 Hu et al. 2015
Junornis 4/5 2 >=6 BMNH Ph-919
Eoenantiornis 4 >=2 6/7 Zhou et al. 2005
Gobipteryx 0 0 0

Longipteryx 4 0 3 BMNH Ph-826
Longirostravis 5? 0 6 IVPP V11309
Pengornis 4 >=16 >=12 IVPP V15336
Rapaxavis 3 0 2 O'Connor et al. 2011
Shanweiniao ? 0 1 O'Connor et al. 2009
Schizooura 0 0 0

Jianchangornis ? >=3 16 IVPP V16708
Archaeorhynchus 0 0 0

Longicrusavis ? ? ? O'Connor et al. 2010
Apsaravis ? ? 0

Hongshanornis ? ? ? O'Connor et al. 2010
Yanornis 5 11 18/24 BMNH Ph-1043; IVPP V12558
Yixianornis 5 ? ? IVPP V12631
Gansus zheni 0 >=11 >=15 BMNH Ph-1342; BMNH Ph-1392
Ichthyornis 0 18 21-22
Reconstruction from Field at al.
2018; Marsh 1880
Hesperornis 0 14 33 Marsh 1880
Parahesperornis 0 14 ?
Reconstruction from Martin
1984

17

Table 2-1. Numbers of teeth on each tooth bearing bone of Mesozoic birds. Continued.
Taxon\Character
Tooth #
Pmx
Tooth #
Mx
Tooth #
D
Source
Fortunguavis ? ? ? IVPP V18631
Shenqiornis 4 >=4 >=7 Wang et al. 2010
Sulcavis 4 4 7
BMNH Ph-805; O'Connor et al.
2013
Bohaiornis 4 2 6 Li et al. 2014
Parabohaiornis 4 4 >=6 IVPP V18691
Longusunguis >=3 3 6 IVPP V17964; Wang et al. 2014
Zhouornis 4 4 7 BMNH Ph-1204
CUGB P1202 4 4 >=3 Peteya et al. 2017
Gretcheniao >1 ? >=4 Chiappe et al. 2019
Anas 0 0 0
Gallus 0 0 0

18

Fig. 2-1. Maximum parsimony-based strict consensus tree of Aves. a, dental characters included (Chiappe et al., 2019). b, dental
character excluded. The excluded dental characters are specified in materials and methods session.

19

Fig. 2-2. Frequency distribution of the highest maxillary tooth count of the avian taxa included in the
present study.

20

Fig. 2-3. Frequency distribution of the highest dentary tooth count of the avian taxa included in the
present study.

21

Fig. 2-4. Parsimony reconstruction of changes in the numbers of teeth on premaxillae (character 1,
Appendix A) mapped onto the phylogenetic topology of Chiappe et al. (2019) with dental characters
excluded (Fig. 2-1b). Green taxon names belong to Enantiornithines. Purple taxon names belong to
Ornithuromorpha. Black taxon names are considered “the basal group”.

22

Fig. 2-5. Parsimony reconstruction of changes of the teeth distribution on premaxillae (character 4,
Appendix A) mapped onto the phylogenetic topology of Chiappe et al. (2019) with dental characters
excluded (Fig. 2-1b). Legends of the branch colors represent the character states (see materials and
methods). Green taxon names belong to Enantiornithines. Purple taxon names belong to
Ornithuromorpha. Black taxon names are considered “the basal group”.

23

Fig. 2-6. Parsimony reconstruction of changes in the numbers of maxillary teeth (character 2, Appendix
A) mapped onto the phylogenetic topology of Chiappe et al. (2019) with dental characters excluded (Fig.
2-1b). Green taxon names belong to Enantiornithines. Purple taxon names belong to Ornithuromorpha.
Black taxon names are considered “the basal group”.

24

Fig. 2-7. Parsimony reconstruction of changes in the numbers of dentary teeth (character 3, Appendix A)
mapped onto the phylogenetic topology of Chiappe et al. (2019) with dental characters excluded (Fig. 2-
1b). Green taxon names belong to Enantiornithines. Purple taxon names belong to Ornithuromorpha.
Black taxon names are considered “the basal group”.

25

Fig. 2-8. Maximum likelihood reconstruction of changes in the numbers of premaxillary teeth (character
1, Appendix A) mapped onto the phylogenetic topology of Chiappe et al. (2019) with dental characters
excluded (Fig. 2-1b). Green taxon names belong to Enantiornithines. Purple taxon names belong to
Ornithuromorpha. Black taxon names are considered “the basal group”.

26

Fig. 2-9. Maximum likelihood reconstruction of changes of the teeth distribution on premaxillae
(character 4, Appendix A) mapped onto the phylogenetic topology of Chiappe et al. (2019) with dental
characters excluded (Fig. 2-1b). Legends of the branch colors represent the character states (see
materials and methods). Green taxon names belong to Enantiornithines. Purple taxon names belong to
Ornithuromorpha. Black taxon names are considered “the basal group”.

27

Fig. 2-10. Maximum likelihood reconstruction of changes in the numbers of maxillary teeth (character 2,
Appendix A) mapped onto the phylogenetic topology of Chiappe et al. (2019) with dental characters
excluded (Fig. 2-1b). Green taxon names belong to Enantiornithines. Purple taxon names belong to
Ornithuromorpha. Black taxon names are considered “the basal group”.

28

Fig. 2-11. Maximum likelihood reconstruction of changes in the numbers of dentary teeth (character 3,
Appendix A) mapped onto the phylogenetic topology of Chiappe et al. (2019) with dental characters
excluded (Fig. 2-1b). Green taxon names belong to Enantiornithines. Purple taxon names belong to
Ornithuromorpha. Black taxon names are considered “the basal group”.

29

Chapter 3: Dental Replacement Patterns and Tooth Cycles in
Mesozoic Birds

Introduction
Tooth replacement is a common phenomenon among vertebrates (Edmund, 1960). Unlike the
diphyodonty of mammals, polyphyodonty (i.e., multiple generations of teeth) (Buchtová et al., 2012;
Rasch et al., 2016) has been observed among fish (Bloomquist et al., 2019; Rasch et al., 2016),
amphibians (Davit‐Béal et al., 2007), and reptiles (Richman and Handrigan, 2011). In particular, studies
have clearly shown that reptiles have alternating patterns of tooth replacement, in which when the
even-numbered teeth are under replacement, the odd-numbered teeth are not (Edmund, 1960). While
teeth are replaced on a schedule that is independent of wear or damage (Whitlock and Richman, 2013),
the underlying regulatory mechanism of the replacement process has been difficult to understand.
Within reptiles, modern studies of crocodilians—the only present-day toothed archosaurs—have
demonstrated that the dental laminae within the alveoli (tooth sockets) form a niche of stem cells for
the multiple generations of teeth, and that this mechanism is regulated by expression of a network of
regulatory genes (Tsai et al., 2016; Wu et al., 2013). Even though our understanding of tooth
development in extinct archosaurs, and its underlying regulatory mechanism, is still in its infancy,
polyphyodonty has been documented for a range of extinct archosaurs (D’Emic et al., 2013; Edmund
1962; Fastnacht, 2008), including the toothed stem birds of the Mesozoic Era (Dumont et al., 2016;
Martin et al., 1980; Marsh, 1880).
The discovery of polyphyodonty and dental replacement in toothed stem birds dates back to the 19th
century. Marsh (1880) reported replacement teeth inside resorption pits in the Late Cretaceous
30

Hesperornis and Ichthyornis. In an attempt to understand the pattern of dental replacement of these
birds, Edmund (1960) reviewed previous findings in avian dentition and concluded that the evidence
was consistent with the presence of “replacement waves” in which stimuli moving in an anterior-
posterior direction, generate waves of tooth replacement. Subsequently, based on the position of
replacement teeth preserved within the roots of functional teeth, Martin et al. (1980) (see also Martin
and Stewart, 1999) concluded that the replacement teeth of Archaeopteryx, hesperornithiforms, and
Ichthyornis developed vertically (albeit initially migrating from the lingual side) within the roots of
functional teeth. Howgate (1984) confirmed the presence of an oval resorption pit on the root of a fully
exposed tooth of the London specimen of Archaeopteryx. He also briefly mentioned that an alternating
replacement pattern may have existed in the Berlin specimen but did not provide further description of
this pattern. These early studies of avian dental replacement relied primarily on the external
morphology of the fossils. However, we now know that tooth replacement and cycling are best
understood through analysis of the jaws’ internal morphology, which requires X-ray imaging including
computed tomography (CT) imaging among other methods.
Such methods have been applied to more recent studies. A synchrotron scan of the jawbones of the
Daiting specimen of Archaeopteryx revealed replacement teeth and traces of resorption on the roots of
functional teeth, confirming an alternating tooth replacement pattern for this Late Jurassic bird (Kundrát
et al., 2019). Moreover, Dumont et al. (2016) utilized CT imaging techniques to make inferences on the
implantation, attachment, and formation time of the teeth of Hesperornis and Ichthyornis. Based on
these results, Dumont et al. (2016) suggested that in these stem birds the replacement teeth first
formed at the lingual side of functional teeth, invading them through lingual resorption pits, and in the
end expelling and replacing them. These authors provided important data on aspects of dental
replacement in some of the most immediate toothed outgroups (the ornithurines Hesperornis and
Ichthyornis) of living birds but the preservation of the scanned specimens limited the data available on
31

the tooth cycling and replacement pattern throughout the jaw. Furthermore, to date no other study has
examined these details in avian clades outside of Archaeopteryx and ornithurines (Fig. 3-1). In this
chapter, through micro-computed tomography (µCT) imaging, I examine three enantiornithine tooth-
bearing bones from the Upper Cretaceous William’s Quarry, which was discovered in 2004 in the city of
Presidente Prudente (São Paulo State) of southeastern Brazil, a locality of the Adamantina Formation
(Bauru Group) containing hundreds of enantiornithine remains (Chiappe et al., 2018). Moreover, I also
acquired the synchrotron data of the Daiting specimen of Archaeopteryx (SNSB BSPG VN-2010/1) from
Kundrát et al. (2019) and the µCT data of Yanornis (IVPP V13358) from Bailleul et al. (2019), along with
µCT scans of Ichthyornis (FHM 2503). The exquisite preservation of these fossils allows examination of
the emergence of replacement teeth within the jawbones and investigation of dental replacement of
these stem birds in three dimensions throughout tooth-bearing bones. For the first time, these data
allow testing specific hypotheses of dental replacement patterns and process in stem birds.

Methods
The three Brazilian Enantiornithine specimens, MPM-90, MPM-373, and MPM-351, were scanned using
a GE Phoenix Nanotom M at the Molecular Imaging Center of the University of Southern California
(USC). MPM-90 and MPM-351 were scanned at 2.73 µm voxel size, 80 kV, 100 mA, exposure time 1250
ms, averaging 2 frames and skip 1 frame, 360-degree rotation 1000 frames, no filter. All three specimens
were scanned at 10 µm voxel size, 100 kV, 200 mA, exposure time 750 ms, averaging 3 frames and skip 1
frame, 360-degree rotation 1440 frames, 0.1 mm Cu filter. The scans were reconstructed through GE
phoenix datos|x 2 reconstruction 2.3.3.160. The three-dimensional visualization and analyses are
conducted using Avizo Lite (9.2). Segmentations and measurements of the dentitions were also
performed with tools of this software. The segmentation of MPM-90 and MPM-351 was conducted
32

using the scans at 2.73µm resolution. The enantiornithine specimens used in this study are deposited at
the Museu de Paleontologia de Marília under collection number MPM-90, MPM-351, and MPM-373.
Digital models of the dentitions of these enantifornithines and the µCT scan data are available at Zenodo
(https://doi.org/10.5281/zenodo.5502305).
The Ichthyornis specimen FHM 2503, contains two fragments of a dentary (Fig. 3-7). Both were scanned
with the same scanner as the Brazilian enantiornithine specimens. The anterior piece was scanned at 3
µm voxel size, 60 kV, 200 mA, exposure time 1000 ms, averaging 2 frames and skip 1 frame, 360-degree
rotation 1080 frames, no filter. The posterior piece was scanned at 10 µm voxel size, 80 kV, 140 mA,
exposure time 1000 ms, averaging 2 frames and skip 0 frames, 360-degree rotation 1441 frames, no
filter. The cross sections of FHM 2503 were visualized using both scans from the anterior and posterior
pieces of the specimen.
Scan parameters of the Archaeopteryx (Daiting specimen, SNSB BSPG VN-2010/1) and the Yanornis (IVPP
V13358) specimens can be found in Kundrat et al. (2019) and Bailleul et al. (2019), respectively. The
synchrotron scan data of the Archaeopteryx specimen are available through Kundrat et al. (2019). The
cross sections of the Archaeopteryx specimen were visualized using data with 14.92 µm pixel size and
7.3 mm Z FOV. The Yanornis scan data were acquired through direct contact with the authors.

Results
• Tooth replacement in the Brazilian enantiornithines
The studied specimens consist of two sets of premaxillae (MPM-90 and MPM-373) and an incomplete
left dentary (MPM-351) exquisitely preserved in three dimensions (Fig. 3-1). The two sets of premaxillae
are very similar to one another, although MPM-373 is significantly smaller (prenarial length is 85.5% that
33

of MPM-90. Table 3-1) and slightly more compressed laterally (Fig. 3-2). The left and right premaxillary
corpi are fused to one another in each of the specimens, although a faint line delineating the inter-
premaxillary contact is visible in MPM-90 (Figs. 3-1d, 3-2). While fusion of the premaxillary corpi is
known for some enantiornithines (e.g., Gobipteryx (Chiappe et al., 2001), Longusunguis (Hu et al.,
2020)), such fusion is absent in many species within this group (Wang and Zhou, 2019). The premaxillae
have four tooth positions, as is typical of enantiornithines and other toothed birds (O’Connor and
Chiappe 2011), with most of their teeth—conical and slightly recurved—preserved in place. The general
size of the erupted crowns is larger in the 3
rd
and 4
th
teeth than in the first two positions (Table 3-2; Fig.
3-1e). The smaller size of the first two premaxillary teeth is congruent with the morphology seen in a
number of enantiornithines of different inferred lifestyles, including Sulcavis (O’Connor et al., 2013),
Shenqiornis (Wang et al., 2010), and Longipteryx (Wang et al., 2015). The frontal processes of the
premaxillae, preserved in MPM-373, are long, slender, and fully fused with each other. These processes
are typically unfused to one another in enantiornithines (Chiappe et al., 2001; Hu et al., 2020; O’Connor
and Chiappe 2011; Wang and Zhou, 2019), although their complete fusion was reported for the Early
Cretaceous Shangyang graciles (Wang and Zhou, 2019). This process has approximately twice the length
of the premaxillary corpus, as in the enantiornithines Shangyang, Longusunguis, and Eoenantiornis. The
lateral sides of the frontal processes are recessed by a shallow, longitudinal groove (Figs. 3-1b, 3-2c).
The angle formed between the longitudinal axes of the frontal and maxillary processes of MPM-373 is
between 32 (right element) and 35 (left element) degrees (Figs. 3-3a, 3-3b), which indicates that the
rostrum gently increased in depth caudally as in many enantiornithines including Zhouornis (Zhang et al.,
2014), Eoenantiornis (Zhou et al., 2005), Sulcavis (O’Connor et al., 2013), and Pengornis (Zhou et al.,
2009). The lateral and dorsal surfaces of the premaxillary bodies of both specimens are scarred by
mental foramina, although their distribution is not uniform, being less dense dorsally (Fig. 3-1). Similar
scarring of the premaxillary’s surface is known for a variety of enantiornithines (Chiappe et al., 2001;
34

Wang et al., 2015; Wang and Zhou, 2019; Zhang et al., 2014; Zhou et al., 2005). The lateral dentigerous
margin of MPM-90 is wavy but this is likely a preservational artifact as shown by the straight dentigerous
margin of the left side of MPM-373 (Fig. 3-1c). Nonetheless, it is clear that the lateral (labial) edge of the
dentigerous margin projected ventrally more than its medial (lingual) counterpart, thus leaving more of
the crowns exposed lingually. In palatal view, only fully exposed in MPM-90, the left and right
premaxillae are separated by a distinct, rostral groove that runs sagittally for almost half of the prenarial
length of these bones (Figs. 3-4a, 3-4a’). Caudally, this groove gives way to a central ridge that is
separated by parallel recesses on each side. The cranial ends of these recesses are further excavated by
a small pit (Fig. 3-4a). This palatal ridge is also visible through µCT imaging in MPM-373 (Figs. 3-4b, 3-
4b’). The overall morphology of the ventral premaxillae differs significantly from the shallow vaulted
shape of this region in the enantiornithine Gobipteryx (Chiappe et al., 2001).
Only the rostral portion of MPM-351 (left dentary) is preserved (Fig. 3-1f); this bone is broken at the
level of its fourth alveolus, which is partially preserved and lacks evidence of a replacement tooth (it
cannot be determined if a replacement tooth was absent or simply not preserved). The dorsal and
ventral margins of this bone are parallel to one another, and its rostral end is slightly convex, angling
caudally at approximately 43 degrees (Figs. 3-1f, 3-3c). The overall morphology of the dentary is
remarkably similar to that of many enantiornithines including Shenqiornis (Wang et al., 2010), Zhouornis
(Zhang et al., 2014), and Piscivorenantiornis (Wang and Zhou, 2017). The lateral surface of the dentary is
scarred by small, irregularly organized mental foramina. MPM-351 preserves the first four teeth, with
the fourth somewhat broken. The dentary teeth are similar in shape and size, and evenly spaced. Like
those in the premaxillae, these teeth are conical and slightly recurved, with heights ranging from 2.18 to
2.37 mm (Table 3-2).
35

Hundreds of partially articulated and isolated bird bones have been found at William’s Quarry, and all
diagnosable ones can be assigned to enantiornithines (Chiappe et al., 2018). This evidence suggests that
the isolated cranial material described here also belongs to enantiornithines. This identification is
further supported by the overall morphology of these specimens including the fact that functional teeth
at the 1
st
and 2
nd
positions of the premaxillae are significantly smaller than those at the 3
rd
and 4
th

positions (Fig. 3-1e; Table 3-2), the long and slender frontal process of the premaxillae (Figs. 3-1a, 3-1b),
the parallel nature of the dorsal and ventral margins of the dentary and angulation of its rostral end (Fig.
3-1f), and the clear absence of a mandibular symphysis (as evidenced by the morphology of MPM 351)
(Fig. 3-1f). While some of these traits are also known for non-enantiornithine toothed birds (e.g.,
absence of mandibular symphysis in Archaeopteryx (Wellnhofer, 2009), subparallel dorsal and ventral
margin of dentary in hesperornithiforms (Marsh, 1880)), their combined presence strongly supports the
proposed identification of these specimens. Furthermore, as the record of Late Cretaceous birds is
limited to enantiornithines and ornithuromorphs (including modern birds), the assignation of this
material as belonging to Enantiornithes is congruent with the fact that the premaxilla of all toothed Late
Cretaceous ornithuromorphs (e.g., Ichthyornis (Field et al., 2018), hesperornithiforms (Bell and Chiappe,
2020)) is devoid of teeth. The overall morphological similarity between the two sets of premaxillae—
notwithstanding their size differences—suggest that these two specimens are likely to belong to the
same species or to very closely related taxa.
µCT scans of premaxilla MPM-90 show replacement teeth forming at the right 1
st
and 4
th
tooth positions
(Fig. 3-5a-a’’); the 2
nd
and 3
rd
tooth positions of the right element have no signs of root resorption or
replacement tooth formation. Functional teeth are preserved in situ on the left side of MPM-90 at the
first two positions. The lack of resorption on these functional teeth and the absence of replacement
teeth visible within their alveoli, indicates that these positions were not being replaced at the time of
death. µCT scans reveal that all teeth in MPM-90 have deep roots set in deep alveoli (Figs. 3-5a, 3-6a),
36

even in the 1
st
and 4
th
functional teeth of the right element despite the resorption resulting from the
formation of their replacement teeth. The replacement tooth forming in the lingual side of the right 1
st

position appears to be at its early stage because it is still small compared to the size of the functional
tooth (Figs. 3-5a’, 3-6a). This replacement tooth may have been slightly displaced during postmortem, as
the tooth is positioned lingual-labially with the apex pointing towards the labial side of the alveolus (Fig.
3-5a’). The root of the functional tooth at the 1
st
position exhibits significant root resorption (i.e., half of
the height of the functional tooth) only on the lingual side (Fig. 3-5a’). At the right 4
th
position of MPM-
90, the replacement tooth has grown to about one third of the height of the functional tooth (Table 3-
2). The root of the later is resorbed significantly on both sides but more labially than lingually (Fig. 3-
5a’’), a condition rare in reptiles because of the lingually positioned dental lamina and lingual tooth
replacement (LeBlanc et al., 2017; Richman and Handrigan, 2011; Whitlock and Richman). This rare
resorption pattern may indicate either a labial replacement (labial to lingual; contra (LeBlanc et al.,
2017)) or most likely, an unusual odontoclast behavior.
MPM-373’s right premaxilla bears exposed teeth at the 2
nd
, 3
rd
, and 4
th
positions (Fig. 3-1b); µCT imaging
reveals that the 1
st
position lacks a tooth. The one exposed at the 2
nd
position is a replacement tooth,
which tip is exposed only because the dentigerous margin is incompletely preserved (Fig. 3-6b); those at
the 3
rd
and 4
th
positions correspond to functional teeth. The left premaxilla of MPM-373 preserves a
replacement tooth at the 1
st
position (Figs. 3-5b, 3-6c). µCT imaging shows that there are no teeth
preserved at the left’s 2
nd
position. The teeth at the 3
rd
and 4
th
positions of the left premaxilla are
functional teeth. Replacement teeth of similar sizes are also observed at the 4
th
positions (left and right)
through µCT images, which also reveal the absence of replacement teeth at the 3
rd
positions, indicating
symmetrical tooth cycles in at least these two positions (Fig. 3-6b-c). As in MPM-90, the functional teeth
at the 3
rd
and 4
th
positions possess large roots set deep into the premaxilla. The functional teeth at the
4
th
positions of MPM-373 are more resorbed on the lingual side, showing typical reptilian lingual
37

resorption that leads to a lingual to labial migration of the replacement tooth (Fig. 3-5b’’). The
replacement tooth at the 1
st
position and the functional tooth at the 3
rd
on the left side show evidence
of possible dislocation. The former is preserved with its curvature facing forward; the latter is tilted
some 30 degrees (Fig. 3-6c).
µCT images of dentary MPM-351 show replacement teeth forming in the 1
st
and 2
nd
positions (Figs. 3-5c-
c’’, 3-6d). The replacement tooth at the 1
st
position is significantly smaller than that at the 2
nd
position.
Considering the similar size of the functional teeth, the replacement tooth in the 1
st
position is
interpreted as being at an earlier stage of development than its counterpart in the 2
nd
position. The
functional tooth at the 3
rd
position has a deep root with no sign of resorption and no replacement tooth
is visible within the alveolus (Fig. 3-6d). The fact that the replacement tooth of the 2
nd
position is at a
later stage of development than those of the 1
st
and 3
rd
positions suggests an alternating pattern of
dental replacement for MPM-351. However, this interpretation cannot be corroborated without the
dental developmental stage of the 4
th
position, which is uncertain.
• Tooth cycle in Archaeopteryx, Brazilian enantiornithines, Yanornis, and Ichthyornis
The Archaeopteryx specimen, SNSB BSPG VN-2010/1, is highly compressed and only the right maxillary
and dentary dentitions are preserved (Kundrat, 2019). High resolution synchrotron scans show in situ (or
minimally compressed) replacement teeth preserved in the 6
th
and 8
th
position of the dentary (Fig. 3-
8a1-a2). The cross sections of these two tooth positions revealed that the newly forming teeth initiate
from the lingual side of the alveolus, start resorption of the lingual root of the functional teeth at their
early stages, and move into the pulp cavity of the functional teeth while they are still very small, but
leave the labial root of the teeth intact (Fig. 3-8a1). The new teeth then continue to grow and resorb the
bottom of both the lingual and labial root of the functional teeth (Fig. 3-8a2).
38

The cross sections of the three Brazilian enantiornithine specimens revealed multiple developmental
stages of tooth replacement of this group of birds. The earliest stage of their tooth cycle preserved in
these specimens is at the first tooth position of the right premaxilla of MPM-90. Although the new
forming tooth is a little dislocated, it clearly shows the resorption on the lingual root of its predecessor
(Fig. 3-5a’). From the 4
th
position on the same premaxilla and the left premaxilla of MPM-373, both the
lingual and labial sides of the roots of the functional teeth were resorbed away when the replacement
teeth were in very late stages, and the apical crown of the replacement teeth approached the occlusal
margin of the alveolus. No replacement teeth in stages between these two are preserved in the
premaxillae, thus their tooth cycle remains unclear; however, the first position of the dentary MPM-351
may be at a slightly later stage than the replacement tooth at the first position of MPM-90 (Fig. 3-5a’)
based on its relative size to the functional tooth (Table 3-2; Fig. 3-5c’). The replacement tooth in this
alveolus had resorbed the bottom of the lingual root of the functional tooth and had partially moved
into the pulp cavity of its predecessor (Fig. 3-5c’). MPM-351 also preserved another tooth at a similar
stage as the late mature replacement teeth in both premaxillae (Figs. 3-5a’’, 3-5b’’, 3-5c’’). As its crown
approaches the margin of the alveolus, it is in close contact with the functional tooth and resorbed both
the lingual and labial roots to the functional tooth.
The Yanornis specimen, IVPP V13358, has the unprecedently complete tooth cycle record in Aves.
Multiple tooth positions in the dentaries preserving replacement teeth in different stages of its tooth
cycle allow its tooth cycle to be pieced together from the µCT data. Cross sections presented in this
study (Fig. 3-8b1-b6) show the process of Yanornis’s tooth replacement from intact functional teeth to
their complete replacement and the emergence of new teeth. Based on observations of American
alligators’s tooth replacement, in which the biomineralization happens at the apposition (growth) stage
in tooth development and the odontoclasts mainly start to resorb the functional tooth at this stage too,
the earliest stage of the tooth cycle observed through µCT scanning of IVPP V13358 is likely the
39

apposition (growth) stage (see chapter 1). In Yanornis IVPP V13358, replacement teeth started to resorb
the lingual sides of the roots of functional teeth during early stages of development (the faint outline of
replacement teeth may indicate that they had just started biomineralization and were less dense). In
these teeth, the resorbed part of the root is almost like being carved away by the new tooth, and the
bottom of the lingual side of the root remains not resorbed (Fig. 3-8b2). As the replacement tooth
continues to grow, the lingual root of the functional tooth was further resorbed away and vacated a
space and path for the replacement tooth to enter the pulp cavity. During this time, the very bottom of
the lingual root of the functional teeth stayed without being resorbed (Fig. 3-8b3). The replacement
tooth further developed and migrated labially, and eventually crossed the pulp cavity and started
resorbing the labial side of the root of the functional tooth (Fig. 3-8b4-b5). Part of the labial side of the
root remained deep inside the alveolus when the replacement tooth was in late stage of the tooth cycle
and had resorbed most of the lingual side of the functional tooth, taking over most of the available
space inside the alveolus (Fig. 3-8b5). The replacement tooth eventually completely occupied the
alveolus and expelled the functional tooth (Fig. 3-8b6).
The outer morphology of the Ichthyornis specimen, FHM 2503 was described by Clarke (2004) and
Martin and Stewart (1977) (Fig. 3-7). Its µCT data reveals that multiple replacement teeth in different
developmental stages are preserved in this dentary. Although not as complete as the record in Yanornis
IVPP V13358, the cross sections of FHM 2503 present a rather holistic view of the dental replacement
process in Ichthyornis (Fig. 3-8c1-c4). The replacement tooth started to resorb the lingual side of the
root of the functional tooth at its early stage. Similar to Yanornis, at this stage the resorption only
happened at the part of the functional tooth immediately next to the new forming tooth. The lingual
side of the root was still deep inside the alveolus and its bottom portion—located at the deepest part of
the alveolus—remains intact (Fig. 3-8c1). As the replacement tooth grew bigger—to approximately half
of the height of the alveolus—the lingual side of the root of the functional tooth had been significantly
40

resorbed, and the replacement tooth entered the pulp cavity (Fig. 3-8c2). At this time, the bottom-most
part of the lingual side of the root had not been resorbed away, and the resorption of the labial side had
not started yet (Fig. 3-8c2). The replacement tooth continued to grow bigger and gradually occupied the
alveolus (Fig. 3-8c3-c4). However, the resorption of the labial side of the root of the functional tooth
does not seem to be very aggressive: when the crown of the new tooth reached near the occlusal
margin of the alveolus, the labial side of the root remained deep inside the alveolus (despite partial
resorption) and the replacement tooth was in close contact with the root of the functional tooth after it
occupied the pulp cavity. The early position of the replacement tooth suggests that the dental lamina of
Ichthyornis may be at the middle part of the alveolus and was more outwardly placed than in
Archaeopteryx and enantiornithines.

Discussion
Our observations strongly suggest the presence of an alternating pattern of dental replacement for
these enantiornithines. Comparative studies of such patterns often use a relation of replacement stages
known as Zahnreihen, which consists of anterior to posterior diagonal tooth rows consisting of
neighboring tooth families at sequential developmental stages (Edmund 1960; Whitlock and Richman,
2013; DeMar and Bolt, 1981). When the distance in tooth position between these diagonal tooth rows
(i.e., z-spacing) equals 2.0, it results in an alternating tooth replacement pattern in which all the odd-
numbered teeth are under replacement and synchronized at the same development stages while the
even-numbered teeth remain intact. However, if the z-spacing deviates from 2.0, the tooth row may still
present an alternating tooth replacement pattern, but one in which different tooth positions are under
different stages of their tooth cycle (Fastnacht, 2008; Hanai and Tsuihiji, 2019). Because the specimens
studied here only have three to four tooth positions preserved in each dentigerous bone, it is difficult to
41

estimate an average z-spacing. Nevertheless, the tooth rows of these enantiornithine jawbones appear
to present alternating replacement patterns with different z-spacing that deviate from 2.0. While the 1
st

and the 2
nd
functional teeth of MPM-373 are not preserved, we would expect these teeth to have been
smaller than the 3
rd
and 4
th
functional teeth (this is based on MPM-90 and other enantiornithines; see
Results). Given this expected size difference between these tooth positions, the slightly larger
replacement tooth at the 2
nd
position than that at the 4
th
position of the right premaxilla (Table 3-2)
indicates that these teeth were at a late and relatively early developmental stage, respectively. Likewise,
MPM-373’s left premaxilla has replacement teeth erupting at a similar height at the 1
st
and 4
th
tooth
positions (Figs. 3-5b’-b’’, 3-6c; Table 3-2), once again indicating a relatively late growth stage for the
anterior position versus the posterior position. The teeth at the right 1
st
position and the left 2
nd
position
are missing, but if MPM-373 were to have a symmetrical tooth cycle (as suggested by the condition in
the 3
rd
and 4
th
positions), the late stages of replacement teeth of the anterior tooth positions would
suggest a z-spacing smaller than 2.0, with an anterior to posterior tooth replacement (Fastnacht, 2008;
Hanai and Tsuihiji, 2019). In MPM-90’s right premaxilla, two erupting replacement teeth are observed at
very early and late developmental stages in the 1
st
and the 4
th
tooth positions, respectively (Fig. 3-5a’-a’’,
3-6a; Table 3-2). Dentary MPM-351 also has replacement teeth at different stages in the 1
st
and 2
nd

tooth positions (Figs. 3-5c’-c’’, 3-6d; Table 3-2). MPM-90 has two consecutive positions with intact
functional teeth, and MPM-351 has two consecutive positions with replacement teeth. They all have
neighboring tooth positions undergoing different stages in the tooth cycle, which can be the result of
alternating tooth replacement with z-spacing larger or smaller than two. Adding to this evidence the
report of an alternating tooth replacement pattern in Archaeopteryx (Kundrat et al., 2019), I regard the
control of the dental replacement of enantiornithines and Archaeopteryx to be conserved.
Teeth in the three enantiornithine specimens are all forming within the tooth sockets separated by
porous interdental bone (Fig. 3-9). The tooth replacement observed in these specimens indicates a labial
42

migration of the new forming teeth. The position of the replacement teeth within the alveoli
demonstrates that dental laminae were positioned lingually. Martin et al. (1980; also, Martin and
Stewart, 1999) described the replacement teeth of Archaeopteryx, and the hesperornithiforms
Hesperornis and Parahesperornis, as developing in circular to oval resorption pits in the lingual side of
their predecessors. They also suggested that the crown of the replacement tooth may have tilted labially
when it entered the root of its predecessor, thus inferring a labial migration similar to their observations
in crocodilians. By comparing this phenomenon of tooth replacement in birds and crocodilians, Martin
et al. (1980; also, Martin and Stewart, 1999) argued that their similarity in dental replacement
underscored a closer phylogenetic relationship (contra hypotheses supporting a dinosaurian origin of
birds). Martin et al.’s description of the replacement teeth tilting labially implies that the initiation of
replacement teeth starts lingual to the root of the functional teeth in the taxa studied by these authors.
The X-ray imaging data of the stem group birds in the present study show that all replacement teeth
preserved at very early stages resorbed the lingual side of the functional teeth’s roots; this indicates that
the initiation of tooth formation occurred lingual to the functional teeth, thus confirming Martin et al.’s
observations. Our observations indicate that during early development, the replacement teeth migrate
into the pulp cavity of their functional teeth, subsequently growing within the space formerly occupied
by the functional teeth.
Although the general process of the dental replacement may be similar among these birds, the position
of the dental lamina may not be identical. The dental lamina consists of a cluster of stem cells, so it is
unlikely to be preserved in fossils or to be observed in X-ray images, but its position may be inferred by
the position of the replacement teeth. In Archeopteryx, based on the position of the early-stage
replacement teeth (Fig. 3-8a1), the position of its dental lamina is inferred to be near the bottom of the
functional tooth at its bud or pre-initiation stage. The observation that the bottom of the roots of
functional teeth was resorbed away early during the development of new teeth also supports this
43

inference. The earliest developmental stage preserved in the three enantiornithine specimens is the first
tooth position of MPM-90. The position of the small replacement tooth in MPM-90 is about halfway in
the middle of the root of the functional tooth (Fig. 3-5a’). Given that no lingual root remains inward to
this new tooth in the alveolus, it is possible that the dental lamina was deep inside the alveolus and near
the bottom of the root of the functional tooth. When the new tooth formed, it quickly moved outward
and resorbed the significant part of the functional tooth. However, because this replacement tooth is
not entirely in situ, I cannot rule out that the tooth was formed near its current position, and the bottom
of the functional tooth was just lost in preservation. Nonetheless, the later-stage tooth in MPM-351 (Fig.
3-5c’) may provide some insight. This replacement tooth is in situ and positioned at the bottom of the
functional tooth, and the resorption of the lingual root also seems to start from the lower part of the
functional tooth. It may indicate that the dental lamina was deep inside the alveolus and near the
bottom of the root. The deep position of the dental lamina may also provide an explanation for the
seemingly lingual migration of the 4
th
position in MPM-90 (i.e., more resorption on the labial side than
lingual side) (Fig. 3-5a’’). If the new tooth was initiated at the bottom of the alveolus, it may have
entered the pulp cavity more easily as the root of the functional tooth generally tapers thinner near the
bottom. Consequently, if it tilted labial-ward in the pulp cavity, it could have subsequentially more
resorption on the labial side of the functional tooth. The earliest stage of the replacement tooth
observed in IVPP V13358 (Fig. 3-8b2) implies that the dental lamina of Yanornis was located at the
middle position of the root of the functional tooth and more outward in the alveoli compared to where
it was inferred for Archaeopteryx and enantiornithines. The remaining bottom of the lingual root of the
functional tooth also supports this inference. Similar to Yanornis, the early position of the replacement
tooth of Ichthyornis (Fig. 3-8c1) suggests that the dental lamina of this ornithuromorph bird may lay at
the middle part of the alveolus and was more outward than it is in Archaeopteryx and enantiornithines.
44

Our data also show that the early labial migration of the teeth of these stem birds differs from the
known pattern of dental replacement in non-avian theropods; for example, the replacement teeth of
Gorgosaurus, Allosaurus, and Coelophysis started by resorbing into the lingual wall of their alveoli and
remained parallel in coronal view without occupying the pulp cavity of the functional teeth until later in
their development (Edmund, 1969; Fong et al., 2016; LeBalnc et al., 2017). None of the birds examined
in the present study experienced any alveolar wall resorption during their tooth cycle. This shows that
the process of tooth replacement in birds could be more similar to that in crocodilians than to that in
non-avian theropods, even if data from the latter derives from taxa that are not particularly close to the
origin of birds. In addition, in the later stages of the tooth cycle of the two ornithuromorph birds
examined here (Yanornis and Ichthyornis), the labial side of the roots of the functional teeth stayed deep
in the alveoli even when the replacement tooth reached a very late stage and was ready to emerge from
the alveoli (Figs. 3-8b4-b5, 3-8c3-c4). Compared to the µCT record of the tooth cycle of alligators (see
chapter 1; Fig. 1-2), the resorption of both sides of the root of the functional tooth starts early, and the
labial side of the root does not remain deep in the alveolus. When the replacement tooth is
approximately half of the alveolus, the bottom of both sides of the root has been resorbed (Fig. 1-2).
Also, the replacement tooth in late stage is not in close contact with the root of the functional tooth like
that seen in Yanornis and Ichthyornis. Therefore, although the current data may show that the toothed
birds studied here shared some similarities in tooth cycle with crocodilians during their early dental
development, at least the ornithuromorph birds had a different resorption pattern in late stages. Future
studies of non-avian theropods closer to the divergence of birds (e.g., non-avian paravians) may
elucidate the evolution of the dental replacement pattern of stem birds but given the voluminous
evidence in support of the theropod origin of birds, the similarities between crocodilians and toothed
birds are likely homoplastic (contra Martin et al. (1980; also, Martin and Stewart, 1999).
45

Erickson (1996) proposed to assess tooth replacement rates by counting the difference between the
incremental lines of von Ebner of a functional tooth and those of its subsequent replacement tooth of
the same tooth family. This method has since been widely used to study the replacement rates and
frequencies in multiple dinosaur groups (D’Emic et al, 2013; 2019). Other than this measurement of
replacement rates, studies have tried to infer general relative rates of replacement through the
proportion of all teeth undergoing replacement at a given time (Whitney and Sidor, 2019). The number
of successional generations in a tooth family has also been considered as replacement frequency (Bertin
et al., 2018; Luo et al., 2004; Sander and Faber, 2003). Based on their high-resolution synchrotron scans,
Dumont et al. (2016) only observed a maximum of one replacement tooth forming lingual to the
functional teeth of Hesperornis and Ichthyornis. Because other non-avian archosaurs had been observed
as having multiple generations of replacement teeth in their tooth families (e.g., D’Emic et al., 2013;
Hanai and Tsuihiji, 2019), Dumont et al. suggested that these toothed birds may have had
oligophyodonty (i.e., reduced frequency of tooth replacement). Our observation of the toothed bird
specimens studied here is consistent with those of Dumont et al.’s. All the tooth positions undergoing
tooth replacement in Archaeopteryx, enantiornithine birds, Yanornis, and Ichthyornis have only one
replacement tooth in each. Hence, our data suggests that these stem birds only had a maximum of two
generations (one functional and one replacement tooth) for each tooth family at a time. Although these
birds share a lower number of dental generations at a given time when compared to their archosaurian
relatives, it is unclear whether they also had a lower rate of replacement (i.e., frequency). Based on
dentine incremental lines revealed by synchrotron scans, if those lines represent daily records, tooth
formation duration in Hesperornis was estimated to be 66 days (Dumont et al., 2016), with the onset
between two tooth generations being shorter. Nevertheless, the onset between two tooth generations
and the number of tooth generations present at a time may also vary due to the growth of the skull
through ontogenetic stages (Luo et al., 2004). Histological and ontogenetic data of toothed birds will
46

provide additional insights to the evolution of replacement rates and frequency, but unfortunately
growth series of these birds are exceedingly rare.
The µCT and synchrotron scans yield unprecedented data on avian tooth replacement. Although an
alternating tooth replacement pattern was reported for Archaeopteryx (Kundrat et al., 2019), previous
to this study no other data on tooth cycling was known for other stem birds. Our data strongly suggest
an alternating pattern of dental replacement for enantiornithines, one with z-spacing differing from two.
The data presented here also show that the replacement teeth in these birds developed lingually and
migrated labially at an early stage of the tooth cycle. We regard this condition as evolving independently
from that seen in living crocodilians and expect that a similar pattern will be documented amongst non-
avian paravian theropods. CT imaging of well-preserved avian specimens and their closest theropod
relatives, like those from the Jehol Biota (Chiappe and Meng, 2016), is likely to provide additional
evidence for understanding the tooth cycle of stem birds and the deep dental homologies within
archosaurs.

47

Table 3-1. Prenarial length (mm) of MPM-90 and MPM-373. Measurements were taken through CT
segmentation using Avizo Lite 9.2.

MPM-90 MPM-373
right 5.03 right 4.2
left 4.90 left 4.29

48

Table 3-2. Total heights and erupted crown heights (mm) of teeth preserved in the three studied
Brazilian specimens. Total heights of both functional and replacement teeth are measured from 3D
segmentation with Avizo Lite 9.2. Erupted crown heights are measured from photographs of the
specimens following Smith et al. (2005).
MPM-90 (Premaxilla)
Position Functional Replacement
Erupted Crown
Height
1(r) 1.85 0.14 0.54
2(r) 1.95 --- 0.36
3(r) 3.02 --- 1.18
4(r) 2.07 0.64 1.05
MPM-373 (Premaxilla)
Position Functional Replacement
Erupted Crown
Height
2(r) --- 0.82 ---
3(r) 2.52 --- 1.08
4(r) 2.59 0.56 1.08
1(l) --- 0.63 ---
3(l) 2.47 --- 0.90
4(l) 2.40 0.66 1.16
MPM-351 (Dentary)
Position Functional Replacement
Erupted Crown
Height
1(l) 2.18 0.41 1.10
2(l) 2.25 0.98 1.07
3(l) 2.37 --- 1.05
4(l) 1.40+ --- 1.03+

49

Fig. 3-1. Photographs of the enantiornithine specimens MPM-90, MPM-373, and MPM-351, and a
simplified cladogram highlighting the stem avian taxa discussed in this study. MPM-373: a, dorsal view;
b, right lateral view; c, left lateral view. MPM-90: d, dorsal view; e, right lateral view. MPM-351: f, left
lateral view. En: External nares; Fp: Frontal process.
50

Fig. 3-2. Photographs and closeup of enantiornithine specimen MPM-373. a, dorsal view. b, anterior
view. c, closeup of the frontal process showing the groove (white arrows). a. and b. show the lateral
compression of the specimen. Fp: Frontal process.

51

Fig. 3-3. Angle measurements of MPM-373 and MPM-351. a, angle between the longitudinal axes of the
frontal and maxillary processes on the right side of MPM-373. b, angle between the longitudinal axes of
the frontal and maxillary processes on the left side of MPM-373. c, angle of the rostral end of MPM-351.

52

Fig. 3-4. Ventral views of MPM-90 (a) and MPM-373 (b), and cross sections (a’-a”, b’-b”) of the
premaxillae. Orientation of the cross sections is noted at the upper right corner of a’. D, dorsal; En,
external nares; Fp, frontal process; L, left; R, right; V, ventral. Yellow arrows (a, a’) indicate the rostral
palatal groove of MPM-90 (see text; this groove is not observed in the 3D reconstruction of MPM-373).
Orange arrows (a, a”, b, b”) indicate the median palatal ridge developed in the posterior half of the pre-
narial portion of the premaxillae (see text). The white arrow (a) indicates the small pits at the cranial end
of the recesses, one on each side of the palatal ridge. MPM-373 (b) shown as a 3D reconstruction from
the µCT scan.
53

Fig. 3-5. Dentition of MPM-90, MPM-373, and MPM-351 visualized through µCT imagining. a-c, labial
views; a’-a”, b’-b”, c’-c”, cross-sections of replacement teeth at the 1st (a’-c’), 2nd (c”) and 4th (a”-b”)
positions. This figure was created to demonstrate the tooth distribution within jawbones and the
resorption between the functional and replacement teeth. a, right lateral view of MPM-90, digitally
flipped for comparison. b, left lateral view of MPM-373. c, left lateral view of MPM-351. Functional
teeth in cyan; replacement teeth in magenta. Note that the functional teeth generally have more
resorption on their lingual sides, except at the 4th tooth position of the right premaxilla of MPM-90 (a”).
La, labial; Li, lingual; Rt, replacement tooth.
54

Fig. 3-6. Dentition of MPM-90, MPM-373, and MPM-351 visualized through µCT imagining. a, right
lingual view of MPM-90. b, right lingual view of MPM-373. c, left lingual view of MPM-373. d, lingual
view of MPM-351. This figure was created to show the preserved tooth rows and families in the three
specimens. Functional teeth in cyan; replacement teeth in magenta. White dashed lines (b, c) mark the
positions of two missing teeth in MPM-373. Numbers denote tooth positions from mesial to distal.
Silhouettes in gray provide schematic outlines of the tooth-bearing bones.
55

Fig. 3-7. Photographs of FHM 2503. The reconstruction of the Ichthyornis skull was modified from Field
et al. (2018). The light grey rostral tip of the lower jaw indicates the position of a possible predentary
ossification. Parts of the dentary in the reconstruction have been marked to show the sections
represented by FHM 2503. a, lingual view of the partial anterior part of FHM 2503. b, lingual view of the
posterior part.
56

Fig. 3-8. Tooth cycle of Archaeopteryx (SNSB BSPG VN-2010/1), Yanornis (IVPP V13358), and Ichthyornis
(FHM 2503) visualized through the cross section of their µCT data. a1-a2, two stages of the
Archaeopteryx tooth cycle. a1 is slightly earlier than a2. b1-b6, tooth cycle of the Yanornis, from the very
early stage when the dental lamina may still be at pre-initiation (bud) stage, and the functional tooth
was intact (b1) to the late maturation stage when the new tooth expelled the functional tooth and
completely occupied the alveolus (b6). c1-c4, tooth cycle of the Ichthyornis, from the early growth stage
when the lingual root of the functional tooth still remained (c1) to the late maturation stage when the
new tooth approaching the occlusal margin of the alveolus (c4). c1-c3 was from the anterior piece of
FHM 2503; c4 is from the posterior piece of the specimen. Ft: Functional tooth; La, labial; Li, lingual; Rt,
replacement tooth.
57

Fig. 3-9. µCT images showing teeth that have grown in alveoli and porous interdental bone. a, left
premaxilla of MPM-90. b, MPM-351 (left dentary). Ft: functional tooth; idb: interdental bone; pc: pulp
cavity; Rt: replacement tooth.

58

Chapter 4: Tooth Cycling Control in American Alligators (Alligator
mississippiensis)

Introduction
As elucidated in previous chapters, Mesozoic toothed birds, along with other reptiles, undertake
multiple tooth replacements throughout their lifetime (polyphyodonty). This temporal repetitive tooth
cycling requires a niche of pleuripotent cells to enable regeneration of organs (Whitlock and Richman,
2013). The mechanisms underlying the polyphyodonty replacement have long been intriguing questions
for studies of the developmental and the evolutionary history of these toothed animals—not only these
birds have multiple generations of teeth, but their alternating pattern of tooth replacement is very
common among reptiles (D’Emic et al., 2013; Edmund, 1960; 1962; Handrigan et al., 2010; Wu et al.,
2021). An alternating tooth replacement pattern would have the neighboring tooth families under
different developmental stages in their tooth cycles. In a typical scenario, the even-numbered tooth
positions would be under replacement when the odd-numbered position are intact. However, the
underlying mechanisms for how to form such alternating pattern of tooth replacement have not been
fully explained, and such, they continue to be a critical question in both paleontology and
developmental biology (Edmund, 1960, 1962; Osborn, 1974, 1998). The controls to initiate repetitive
tooth cycle and form a replacement pattern throughout a tooth row need to be taken into consideration
when studying the underlying mechanisms for tooth replacement, because particular replacement
patterns are determined by individual tooth positions at certain stages of their tooth cycle. Modern
crocodilians are the closest toothed living relatives to birds, sharing a thecodont tooth implantation and
an alternating tooth replacement pattern with Mesozoic toothed birds (and non-avian dinosaurs).
59

Because American alligators (A. mississippiensis)—like all other modern crocodilians—are likely to share
the conserved underlying mechanisms that control archosaurian tooth replacement, these animals
provide an ideal model for examining tooth cycling in archosaurians, including the extinct toothed birds
of the Mesozoic Era.

Molecular Controls for Dental Replacement
In this study, I also use immunohistochemical staining to visualize the distribution of a Wnt antagonist,
secreted frizzle-related protein 1 (sfrp1), in the dental lamina. Previous studies have tried to reveal the
molecular controls for the dental replacement in reptiles and have discovered several molecular
pathways potentially involved in this regeneration process. Wu et al. (2013) used BrdU, a mitotic
labeling, to map out the label retaining cells and to identify the niche of odontogenic stem cells in the
lingual-side dental lamina bulge. Based on the morphology of these stem cells, Wu et al. (2013)
separated the tooth cycle into three stages: pre-initiation, initiation, and growth. The Wnt signaling was
found to be highly related to tooth formation. Wnt signaling pathway is essential for both tooth
initiation and replacement, and sfrp1 was found to be present in the dental lamina bulge during the pre-
initiation stage. However, sfrp1 was absent during the initiation stage and growth stage (Wu et al.,
2013). Another Wnt antagonist, secreted frizzled-related protein (sfrp2), was found to express around
the tooth but not within the tooth itself in python and lizard (Whitlock and Richman, 2013). β-catenin
may be related to the maintenance of stem cells. In mice, overexpressing β-catenin in epithelium shows
multiple enamel knot formation (Järvinen et al., 2006) and in mesenchyme, it is associated with
regulating Bmp4 and Shh expression in dental ectoderm (Fujimori et al., 2010). Bmp and Shh signaling
have also been found to be related to the dental lamina development and tooth formation in pythons
(Buchtová et al., 2008; Handrigan and Richman, 2010). Additionally, Sfrp1 competes with the frizzled
60

receptors to bind to Wnts and results in reductions of intracellular β-catenin. These results suggest that
the tooth cycling in alligator is associated with the disappearance of sfrp1 and β-catenin activation in the
stem cell niche (Wu et al., 2013).

Hypotheses for the Control Mechanisms of Tooth Replacement Patterns
The underlying mechanism for the alternating tooth replacement pattern of reptiles has been explained
by two main hypotheses, which independently have tried to explain the observed pattern of teeth
arising and being replaced in “waves”, or Zahnriehen (a German term meaning a “toothed belt”): 1)
wave stimulus and 2) zone of inhibition (Whitlock and Richman, 2013).
1) Wave stimulus. Edmund (1960, 1962) hypothesized that for reptilian tooth replacement, the series of
teeth is generated by a triggering impulse passing along the dental laminae caudally (Fig. 4-1a).
2) Zone of inhibition. Osborn (1974, 1998) proposed that a zone of inhibition formed around a
developing tooth releases inhibition signals into the surrounding mesenchyme, which in turn inhibit the
initiation of neighboring teeth (Fig. 4-1b).
These two long standing models have been used to explain the mechanisms of alternating tooth
replacement pattern for the past several decades without closer examination. For example, previous
studies of RNA-seq of different tooth developmental stages in alligators have shown that Wnt, Bmp, and
Fgf pathways are activated when a new replacement tooth enters the initiation (cap) stage (Tsai et al.,
2016). Also, an integrated network of positive and negative feedback of Bmp, Shh, Wnt, Fgf, and Eda
signaling pathways has been proposed for mammalian tooth development (Lan et al., 2014). Molecules
involved in these pathways can be potential candidates for the triggering impulses in Edmund’s
hypothesis or the inhibition signals in Osborn’s hypothesis, yet they have not been rigorously tested for
61

their roles in the formation of dental replacement patterns. Molecular studies of the controlling
mechanisms underlying the pattern of tooth cycling in living reptiles have the potential for testing both
Edmund’s and Osborn’s hypotheses, in turn clarifying key aspects of the development of the alternating
tooth replacement pattern inferred from fossils of stem birds. Wu et al. (2013) have shown that dental
lamina is connected between tooth families throughout the tooth row in American alligators. Given the
expression of β-catenin and sfrp1 in the alligator dental lamina, this tissue is likely to provide a channel
for the signals involved in the tooth cycle. My focus is on the dental lamina. I will use dental lamina
ablation and growth tracking µCT to test whether either of Edmund’s or Osborn’s hypothesis is
plausible. In addition, I will apply immunohistochemistry methods to examine sfrp1 expression in the
dental lamina of tooth families in different tooth cycling stages and in inter-family positions, with the
purpose of understanding possible routes of inhibition signal transmission. Previous studies on the
molecular networks of the ondontogenic process (Handrigan and Richchan, 2010; Wu et al., 2013;
Whitlock and Richman, 2013) have shown that molecules involved in the Wnt/β-catenin pathway may
be promising candidates for the chemical or hormonal impulse in Edmund’s hypothesis, and sfrp1
appears to be a promising candidate as an inhibition signal in Osborn’s hypothesis (Wu et al., 2013);
consequently, my proposed approach promises to shed light onto the molecular pathways that regulate
dental tooth replacement, providing a new avenue for testing these decades-old hypotheses.

Methods
I have obtained fertilized American alligator (A. mississippiensis) eggs from the Rockefeller Wildlife
Refuge in Louisiana and incubate them in the lab environment at 30°C following works by Ferguson
(1985). Alligator hatchlings to 2-year-old individuals were kept at the University of Southern California
62

(USC). All procedures were reviewed and approved by the Institutional Animal Care and Use Committee
at USC.

I. Dental Lamina Ablation and Growth Tracking
Two American alligators (A. mississippiensis), Alli_2018_1 and Alli_2018_4, hatched in September 2018
and were µCT scanned at the Molecular Imaging Center of USC following the schedule in Table 3-1. The
live animals were anesthetized with isoflurane 2-3% delivered through inhalation and then scanned with
Rigaku CT Lab 90 at 70-90 kV, 88-114 µA, voxel size 90 µm (Fig. 4-2). The µCT data were examined and
analyzed in three-dimensional visualization using Avizo Lite 9.2. The segmentation and volume
estimation of the teeth were also conducted with tools of this software. Alli_2018_1 and Alli_2018_4
were both sacrificed on October 15, 2020.
Dental lamina ablation surgery was performed on Alli_2018_1 on February 7, 2020 and Alli_2018_4 on
August 23, 2019 respectively (Table 4-1).
• Alli_2018_1
After anesthetized with isoflurane 2-3%, both functional and replacement teeth were removed
from the right lower jaw of Alli_2019_1 at tooth position 11, 14, and 15. Gauges immersed with
15.5% ferric sulfate by Astringedent were then inserted into the tooth sockets to stop bleeding
and further ablate the dental laminae. The tooth row on the left lower jaw remained intact as the
control.
• Alli_2018_4
After being anesthetized with isoflurane 2-3%, both functional and replacement teeth were
removed from the right lower jaw of Alli_2019_4 at tooth position 7, 9, and 10. Gauges immersed
63

with bleach were then inserted into the tooth sockets to stop bleeding and further ablate the
dental laminae. The tooth row on the left lower jaw remained intact as the control.
The tooth generations and growth at positions rostral, between, and caudal to the dental lamina
ablation sites were tracked with the µCT scans. Segmentation and volume estimation of these teeth
throughout multiple tooth generation were conducted using µCT data scanned on the dates highlighted
in bold in Table 3.1. The result of the dental lamina ablation surgery was also verified by post-surgery
scans confirming that the surgery sites no longer had new tooth formation.

II. Immunohistochemistry staining of sfrp1
One-year old alligator samples collected from 2018 were fixed with 4% (wt/vol) paraformaldehyde
solution and then decalcified with 0.5M EDTA for two weeks. The Immunohistochemistry staining was
performed according to Wu et al. (2013). The sfrp1 antibody used in Wu et al. (2013) corresponded to
human sfrp1 protein 161-250. The alligator sfrp1 protein (GenBank: KYO28472.1) has been aligned with
human sfrp1 protein sequence, and it has been found that the alligator sfrp1 protein 151-240 is
homologous to human sfrp1 161-250. Hence, an alligator-specific sfrp1 antibody has been made by
GenScript to bind to the following sfrp1 protein sequence at loci 151-240 from GenBank: KYO28472.1.
SFRP1:
PQDDVCIAMTTPNATEVARPKGSGVCPPCDNEMKSEAIIEHLCASEFALKMTIKEVKRENGDRKIIPKKRKALKMGPIRK
KNLKKLVLYL
Immunostaining of β-catenin is also used to mark the dental lamina and dental lamina bulb in the thin
section. DAPI staining was used for the nuclei.

64

Results
I. Dental Lamina Ablation and Growth Tracking
• Alli_2018_1 (Figs. 4-3; 4-4)
Growth tracking records of Alli_2018_1 span from 281 days before dental lamina ablation to 251 days
after the surgery. These records span up to 5 generations of tooth cycles and show at least the growth
of one full cycle. The dental lamina was ablated at tooth positions 11, 14, and 15 of the right lower jaw,
so the segmentation and volume calculation were focused on tooth positions 9, 10, 12, 16, 17, 18, 19,
and 20. The teeth from the left lower jaw were used as controls to compare tooth growth. American
alligators start their tooth cycles in ovo and can go through several generations of teeth before hatching
(Westergaard and Ferguson, 1990). Because all of the CT scans are conducted postnatal, I do not have
the record of the true first-generation teeth. Here the number of the generation only refers to the
ordinal of the tooth generations appearing in the span of the growth tracking record.
At tooth position 9, the growth curves of four tooth generations were recorded in the 281-day period
prior to ablation; all four curves show an almost complete overlap between the right (surgery side) and
left (control side) of the lower jaw. Among the four generations, two complete tooth cycles, generations
2 and 3, were recorded. While generation 2 was nearly symmetrical between the two sides of the jaw,
the left side entered the second tooth cycle somewhat earlier than the right side. However, after
entering the tooth cycle, the growth between the two sides became symmetrical. It seems that the
second-generation of right tooth 9 was damaged during the surgery, and that the volume of the tooth
declined significantly after the surgery as a result of this damage. Generation 3 shows overlapping
curves between the two sides before and after the dental lamina ablation. Position 9 on both sides also
entered generation 4 about the same time before euthanasia.
65

At tooth position 10, the growth pattern was symmetrical in generation 2. In generation 3 and 4, the
right side appears to have grown more slowly than the left side. However, this phenomenon may not be
the result of dental lamina ablation as the third-generation tooth on both sides already started the tooth
cycle 169 days before the surgery and recorded a slower growth rate on the right side. The right side
only grew faster than the left side after than the 27
th
day post-surgery. Both left and right sides entered
the next tooth cycle, generation 4, at the same time, but the right side grew more slowly than the left
side. Before euthanasia, generation 5 was observed on the left dentary, and not on the right side.
Tooth 12 and 13 are two tooth positions between dental lamina ablation sites. The general growth
patterns across generations of tooth 12 are symmetrical. Among the five generations recorded, only
generation 3 shows slower growth on the right side than on the left side. The growth curves of all the
other generations overlap between the two sides, and they also entered the tooth cycle at the same
time. The slower growth of generation 3 may not be the result of dental lamina ablation, as the growth
curves of both sides follow the same trend of slope across the surgery. The sudden drop of the volume
of the third-generation right tooth 12 is cause by damage to the tooth. Generation 4 is especially
important because its cycle started right before the surgery and has a complete growth record of the
cycle across the surgery date. Generation 4 shows that tooth growth is symmetrical and not affected by
dental lamina ablation. Tooth 13 also has symmetrical growth across all four generations. All tooth
generations in position 13 entered their tooth cycles at the same time, and they have overlapping curves
between left and right sides. The sudden drop of the right side in generation 2 and the lower slope in
generation 3 are both due to tooth damage happening when these teeth became functional. The tooth
cycles of these two positions, placed between dental lamina ablation sites and with growth straddling
the surgery, show no difference between the control and the surgery sides.
Tooth 16 to tooth 20 occupy positions posterior to the dental lamina ablation sites. Tooth 16 has similar
growth pattern between the left and right sides in the tooth cycle before the surgery. The third
66

generation started 169 days before the surgery and has a complete tooth cycle recorded across the
surgery. It appears that in this tooth cycle, the tooth grew faster on the right side than on the left side
before the surgery, but the slope of the growth curve of the right side tapered after the surgery. This
may be caused by injury of the tooth during the surgery given that tooth 16 is adjacent to the surgery
site at the 15
th
position. At 151 days after the surgery, although the tooth still had its occlusal apex
preserved, as it can be observed from the CT reconstruction, the tooth had a damaged root part in the
alveolus (Fig. 4-3). In the next tooth generation after the surgery, generation 4, both sides entered the
tooth cycle at the same time, yet the right side grew significantly faster than the left side. Whether this
is caused by dental lamina ablation is unclear given that the tooth cycle started 151 days after the
surgery.
Tooth 17 through tooth 19 show faster growth on the surgery side than on the control side in tooth
generations straddling the dental lamina ablation surgery. Tooth 17 has four tooth generations in the
record. The two generations that developed before the surgery show overlapping curves between the
surgery and control sides. The third generation started 29 days before the surgery and recorded a
complete cycle across the surgery; however, unlike the two tooth generations before the surgery, the
growth on the right side was significantly faster than that on the left side. The right side also entered the
next cycle post-surgery faster, as the fourth generation has been observed on the right side before the
euthanasia without its left counterpart.
The right tooth 18 also grew faster than its left counterpart in the tooth cycle across the surgery. Only
three generations are recorded during this period. In the tooth cycle of generation 1, the growth curves
are generally similar between the two sides when the teeth reached their maximum volume. Generation
2 completed the tooth cycle, and the tooth grew across the surgery. The right tooth 18 already grew
faster than the left side during this cycle prior to the surgery, it seems that the slopes of the growth
curves of the two sides follow the same trend of the curve before and after the surgery, and the right
67

side consistently grew faster than the left side. The right side also entered the next post-surgery tooth
cycle earlier than the left side. The start of generation 3 on the right appeared 151 days after the
surgery, while the same generation did not show up on the left until 251 days after the surgery.
The growth pattern of tooth 19 is similar than that of tooth 17. At this position, there are also four
generations recorded. Generations 1 and 2 grew before the surgery and had symmetrical growth
between the left and right sides. Generation 3 started shortly before the surgery and had a faster
growth rate on the right side than on the left side throughout its cycle. The right side also entered the
next cycle earlier than the left side as generation 4 was observed on the right jaw prior to euthanasia
and no new tooth development was recorded on the left side.
The general growth pattern is similar between the left and right side for tooth position 20. Among the
three generations recorded, the first tooth cycle shows almost overlapping curves between the two
sides. Generation 2 grew across the surgery, and it did not seem to be affected by dental lamina
ablation. The growth curves between the two sides were similar and remained on the same course
before and after the surgery. The only difference between these growth curves started with the new
tooth cycle after the surgery. Although both sides started generation 3 at the same time, the tooth on
the right side grew significantly faster than the one on the left. However, just like at tooth 16, whether
this faster growth was induced by the dental lamina ablation is unclear because it happened 151 days
after the surgery.
The odd-numbered positions in my records, namely tooth 9, 13, 17, and 19, show a very good example
of alternating tooth replacement pattern, as they all have similar growth curve patterns, and the
beginning of their tooth cycles are also synchronized among positions.
• Alli_2018_4 (Figs. 4-5; 4-6)
68

The dental lamina ablation surgery of Alli_2018_4 was performed about 5.5 month before the surgery of
Alli_2018_1. Growth tracking records of Alli_2018_4 spanning from 253 days before the dental lamina
ablation surgery to 319 days after the surgery is demonstrated in Fig. 4-6. These records span up to 5
generations of tooth cycles (Fig. 4-5) and show at least the growth of one full cycle. The dental lamina
was ablated at tooth position 7, 9, and 10 of the right lower jaw, so the segmentation and volume
calculation were focused on tooth position 5, 6, 8, 11, 12, 13, 14, and 15. As in Alli_2018_1, the teeth
from the left lower jaw were used as a control to compare tooth growth.
At tooth position 5 and 6, the two positions mesial to the surgery site, the growth rates between left
(control) and right (surgery) lower jaw do not appear to have a significant difference. Tooth 5 and 6 may
have grown slightly faster on the left side, but the curves are generally the same. The left tooth 5
entered two cycles, generation 3 and 4, earlier than the right tooth 5. Once entered a new cycle, the
growth rate (slope of the curve) was very similar between the two sides. Because these two cycles
started before the surgery, the slight differences can be regarded as an event of asymmetry happening
naturally between tooth generations, and not the result of dental lamina ablation surgery. Tooth 5 was
back to a synchronized cycle (comparing left and right sides) after these two generations and forming
generation 5 together. The growth curves of tooth 6 were also very similar. Only generation 2 shows an
asymmetrical cycle, with the right side entering the cycle earlier than the left.
The growth record of tooth position 8 is particularly important because it happened between two dental
lamina ablation loci. This position shows no significant difference in the growth pattern between the
operation (right) side and the control (left). The tooth on the left side grew slightly faster than that on
the right side, after the surgery, but both sides reached maximum volume in their tooth cycle in 319
days. Also, tooth growth on both sides resumed synchrony during the next tooth cycle as shown by the
overlapping curves in generation four.
69

Tooth positions 11 through 15 are posterior to dental lamina ablation sites. Generation 3 for position 11
was broken on 9/19/2019, shortly after the surgery, but generation 4 in this tooth position recorded the
growth pattern straddling the surgery. Generation 4 was still at its early stage a day before the surgery
and its growth curves after the surgery show a near-overlap between the surgery (right) side and the
control (left) side. This is consistent with the growth pattern of previous tooth cycles (e.g., generation 1
and 2). In our last record, the left side has entered the next tooth cycle while the right side showed no
evidence of a new cycle.
Tooth positions 12 and 13 both show symmetrical growth throughout multiple tooth cycles before and
after the surgery. The right side of generation 3 in position 12 grew slightly faster than the left side and
caused a small difference at their maximum volumes. Generation 4 completed the cycle across the
surgery, and it shows overlapping curves between left and right sides. The following generation of both
sides started 139 days after the surgery and was highly synchronized between the left and right. Tooth
generations of position 13 were also synchronized between the left and right jaws. Only in generation 3
did the left side grow slightly faster than the right side, but the difference is quite small. The tooth cycle
of generation 4 started one day before the surgery and it shows a highly symmetrical overlap between
the curves of the left and right sides throughout the records.
Tooth position 14 is the only tooth position of Alli_2018_4 that recorded a significantly different growth
pattern in post-surgery. The tooth cycle of generation 4 started 57 days before the surgery. The left and
right sides entered the tooth cycle at the same time and symmetrically. However, in later records of the
tooth cycle, after the surgery, the right side grew faster than the left side, although both sides reached
similar maximum volumes in the last record. This asymmetrical growth is also reflected on the
asymmetrical volume differences of generation 3. The earlier part of the tooth cycle was symmetrical
between left and right before the surgery. Around the time of the surgery, generation 3 reached its
highest size and started to be resorbed by the newly formed generation 4. The tooth cycle initiated after
70

the surgery (generation 5) had also started earlier on the right side than on the left. The growth curve of
this generation also shows that the tooth on the right side grew faster than the tooth on the left.
Further away from the surgery sites, tooth position 15 showed similar growth curves between the left
and right. Four tooth generations of this position were recorded in the CT data, with two complete tooth
cycles (i.e., generation 2 and 3). Tooth generations on both side of the jaws entered the tooth cycle at
the same time and have very similar curves between left and right sides, with slightly faster growth and
larger teeth on the right side. This difference is rather small, and a symmetrical growth pattern can be
inferred at this position.

II. Immunohistochemistry staining of sfrp1
The alligator specific sfrp1 antibody (asfrp1) was tested on tooth families with pre-initiation, initiation,
and early growth tooth cycling stages. β-catenin was also stained to mark dental lamina and tooth buds,
along with DAPI to visualize the nuclei in fluorescent. The signal of β-catenin presents strong staining
results throughout all three tooth cycling stages (Fig. 4-7). In one of the pre-initiation samples (Fig. 4-7a)
the sfrp1 was present in the dental lamina, but this result was not duplicated in another pre-initiation
sample (Fig. 4-7d) as the stained signal was not any stronger than the background. The sfrp1 signal was
rather weak in both the initiation and growth stages (Fig. 4-7g, j). The lack of sfrp1 signal in initiation and
growth stage is similar to the results presented by Wu et al. (2013). This test result for the effectiveness
of the custom-made alligator specific asfrp1 shows that it can be used to detect sfrp1 in the pre-
initiation stage of the dental lamina, but the experimental condition may need to be optimized because
same samples in pre-initiation stage do not show sfrp1. Further experiments need to be carried out to
optimize the conditions required to fine-tune the staining results, and to compare sfrp1 expression in
dental lamina between tooth families undergoing different tooth cycling stages.
71

Discussion
Mechanisms for the alternating tooth replacement pattern of reptiles have been discussed over the
years. Edmund (1960, 1962) explained this pattern as a result of “waves” of teeth arising and being
replaced. He proposed that these tooth replacement waves are triggered by chemical or hormonal
impulses traveling caudally from the anterior portion of the jaw (Fig. 4-1a). Subsequently, Osborn (1974,
1998) proposed the “zone of inhibition” hypothesis. According to this hypothesis, developing teeth
release inhibition signals into their surrounding areas and form zones of inhibition that inhibit tooth
growth in neighboring tooth families (Fig. 4-1b). These two hypotheses are both feasible scenarios for
explaining the formation of an alternating tooth replacement pattern. However, my results of dental
lamina ablation and growth tracking experiments suggest that Osborn’s zone of inhibition hypothesis is
less likely to be plausible. Based on this hypothesis, the reason that the odd-numbered teeth are not
under replacement while the even-numbered teeth have replacement teeth growing underneath is
because the odd-numbered tooth families are suppressed by inhibition signals coming from neighboring
developing teeth (i.e., the even-numbered). However, because the dental lamina has been shown to be
connected through the tooth families in each quarter of the jaw (Wu et al., 2013), if the inhibition
signals originate or transmit between tooth families through their dental lamina, the dental lamina
ablation of Alli_2018_1 and Alli_2018_4 would have cut off the transmission path for possible inhibition
signals. Based on Osborn’s hypothesis, the dental lamina ablation should have eliminated the inhibition
signals, thus allowing the tooth families between dental lamina ablation sites to grow freely, presumable
leading to a different dental replacement pattern than that experienced by the control jaw. However,
while the dental lamina was ablated at tooth positions 11, 14, and 15 in Alli_2018_1, the tooth positions
between these dental lamina ablation sites (tooth 12 and 13) showed no significant differences with
respect to the control side. Moreover, position 13 shows the same trend on its growth curves
throughout multiple tooth generations as the other odd-numbered tooth positions. The results of the
72

experiments on Alli_2018_4 provide an even clearer picture. In this specimen, the dental lamina was
ablated on both sides of tooth position 8; consequently, this position should not have been affected by
inhibition signals coming from either cranial or caudal direction. However, the growth curves of tooth
position 8 show no difference to that of the control side, and it is well-synchronized with other even-
numbered tooth positions (tooth 6 and 12). These results provide evidence that even when no inhibition
signals originate in, or transmit through, the dental lamina, the initiation of tooth cycle and the
replacement pattern remain uninterrupted.
Some tooth positions caudal to the dental lamina ablation sites show higher growth on the surgery sides
on than the control side. The surgery (right) side of these positions had either a tooth generation
straddling the surgery and growing faster than the control side (e.g., Alli_2018_1 tooth 17 and 18;
Alli_2018_4 tooth 14) or had the next tooth cycle starting earlier than the control side after the surgery
(e.g., Alli_2018_1 tooth 17, 18, 19). Although these results may seem to support those changes in
signaling, induced by dental lamina ablation and occurring from rostral part, affected tooth growth at
posterior positions in the jaw, these results do not support Edmund’s hypothesis either. According to
Edmund’s hypothesis, the triggering impulses for tooth replacement travel from anterior to posterior of
the jaw. If these impulses travel through the dental lamina that connects the tooth families, when the
dental lamina at three tooth positions was ablated, the tooth families posterior to these positions
should not have received the triggering impulses for tooth replacement. However, the result of this
experiment favors the opposite, as the faster growth happens only in positions posterior to the surgery
site. Also, the fast-growing positions are not the ones right next to dental lamina ablation sites but two
to three positions down the tooth row. It is possible that these positions were affected by growth
factors induced by the wound during the surgery (e.g., Greenhalgh, 1996; Barrientos et al., 2008).
Further experiments are needed to test the triggering of fast growth in positions posterior to dental
lamina ablation sites. Given that current results do not show faster growth at the positions anterior to,
73

or between, the ablation sites, this evidence may indicate that the signal was transmitted from anterior
to posterior instead of being diffused equally in every direction.
Current experimental design and interpretation regarding the dental lamina ablation and growth
tracking experiments were based on the premise that the dental lamina is the main tissue responsible
for the generation and transmission of signals for tooth initiation and replacement controls. However,
the control of tooth cycling is far more complicated than one tissue in charge for all. Just the transition
from bud to cap stages (equivalent to pre-initiation to initiation) involves Bmp, Fgf, Shh, Wnt, and Eda
signaling pathways that form a regulatory network between the epithelium, the tooth bud, and the
surrounding mesenchyme (Lan et al., 2014). Further studies of the molecular basis for tooth cycle are
necessary to understand the underlying control of tooth replacement patterns. For example, mapping
out the distribution of certain promoters or antagonists in the dental tissues at different stages of tooth
cycle would help us understand the regulation of tooth development and replacement.

74

Table 4-1. µCT schedule for Alli_2018_1 and Alli_2018_4. Alligator numbers in bold and italics mark the
scan data used for the three-dimensional segmentation and volume estimation for the growth tracking
of multiple tooth generations. The 3D reconstruction and the growth tracking records are presented in
Figs. 4-3, 4-4, 4-5, and 4-6. Red arrows mark the dental lamina ablation surgery date for both animals.
Both animals were sacrificed after the last scan on October 15, 2020.

75

Fig. 4-1. Schematic dentary of an enantiornithine contrasting Edmund’s wave of stimuli hypothesis (a)
vs. Osborn’s zone of inhibition hypothesis (b) for control mechanisms of reptilian tooth replacement
patterns. Modified from Whitlock and Richman (2013). In Edmund’s hypothesis (Edmund 1960, 1962),
waves of chemical or hormonal stimulus pass caudally. When the first wave travels through the 3
rd
tooth
position, it induces the development of the first three teeth (a’). As the second wave reaches the 1
st

76

tooth position (a’), a new tooth initiates from the dentigerous anlage (a foundation for cell proliferation
and tooth development). The waves continue to travel caudally, and a new wave initiates a new
generation of tooth formation (a’’). In Osborn’s hypothesis (1974, 1998), when the first tooth develops,
it generates an inhibition signal and suppress the tooth formation in the neighboring area of anlage (b’).
As the jaw and the anlage grow bigger, the distance between the teeth increases, and hence increases
the space between the “zone of inhibition”, allowing new teeth to form from the anlage outside of the
zone of inhibition (b’’ to b’’’). As the jaw and space between the teeth continue to expand
craniocaudally, new teeth initiate in the neighboring tooth positions of the emerged teeth (b’’’’), and
thus forming an alternating pattern. Numbers in b denote tooth positions from mesial to distal.
77

Fig. 4-2. Juvenile alligator being anesthetized and scanned with Rigaku CT Lab 90 for growth tracking of
different tooth generations.

78

Fig. 4-3. 3D reconstruction from the µCT scans to track the growth of teeth at the positions rostral,
between, and caudal to the dental lamina ablation sites of Alli_2018_1. The dental lamina was ablated
at position 11, 14, and 15 on the right lower jaw (red). The left lower jaw (blue) was left intact as the
control. The rendered skulls at the top were viewed from the lingual side at the mid-sagittal plane as a
schematic. Color of the tooth reconstructions follow the colors in Fig. 4-4; the deeper color of the teeth
the later the tooth generation. Scale bar is 5 mm.

79

Fig. 4-4. Tooth growth tracking records of the lower jaws of Alli_2018_1 at tooth position rostral,
between, and caudal to dental lamina ablation sites. Dental lamina ablation was performed at position
11, 14, and 15. Orange box marks the tooth positions between dental lamina ablation surgery sites. L./R.
Gen. #: Left/Right tooth generation #. Blue colors are left jaw records; red colors are right jaw records.
80

Fig. 4-5. 3D reconstruction from the µCT scans to track the growth of teeth at the positions rostral,
between, and caudal to the dental lamina ablation sites of Alli_2018_4. The dental lamina was ablated
at position 7, 9, and 10 on the right lower jaw (red). The left lower jaw (blue) was left intact as the
control. The rendered skulls at the top were viewed from the lingual side at the mid-sagittal plane as a
schematic. Color of the tooth reconstructions follow the colors in Fig. 4-6; the deeper the color of the
teeth the later the tooth generations. Scale bar is 5 mm.
81

Fig. 4-6. Tooth growth tracking records of the lower jaws of Alli_2018_4 at tooth position rostral,
between, and caudal to dental lamina ablation sites. Dental lamina ablation was performed at position
7, 9, and 10. Orange box marks the tooth positions between dental lamina ablation surgery sites. L./R.
Gen. #: Left/Right tooth generation #. Blue colors are left jaw records; red colors are right jaw records.

82

Fig. 4-7. Immunohistochemistry staining of sfrp1, β-catenin, and DAPI in dental lamina and tooth buds
during the pre-initiation (a-f), initiation (g-i), and growth (j-l) stages of the tooth cycle in A.
mississippiensis. Developmental stages follow Wu et al. (2013) (also, see chapter 1). dl: dental lamina
(while arrows); dlb; dental lamina bulge (yellow arrows). Scale bar = 100 µm.

83

Chapter 5: Conclusions

Through high resolution µCT scans, this study presents unprecedented data on dental replacement
patterns and tooth cycle reconstructions of toothed stem birds from the Mesozoic Era. In addition to the
reported alternating pattern of the dental replacement of Archaeopteryx (Kundrat et al., 2019), our data
suggests that enantiornithine birds may have also shared this type of replacement. An alternating tooth
replacement pattern is consistent with the widespread distribution of this pattern among living and
extinct reptiles. This also suggests a conserved underlying control for the formation of dental
replacement in reptiles. Results of the present study also show that unlike some non-avian theropods,
Archaeopteryx and members of enantiornithine and ornithuromorph birds all have replacement teeth
that migrate labially and invade the space of functional teeth at their early tooth cycling stages. This
process leads to a resorption pattern during early stages of the tooth cycle that resembles that of
crocodilians, even if phylogenetically such similarity is considered to be homoplastic. The
ornithuromorph birds examined here present a unique resorption pattern. Not only that their new teeth
migrate labially at very early tooth developmental stages, but they also have the labial roots of the
functional teeth minimally resorbed up to very late stages. This resorption pattern deviates from the
non-avian theropods in their early tooth developmental stages and is inconsistent with the crocodilians
in their late stages. Further research on tooth cycling data from the non-avian theropod lineages closer
to Aves would provide critical insight regarding the evolution of tooth migration and resorption during
tooth development in archosaurians. In addition, the positions of these birds’ dental laminae have been
inferred through the position of their replacement teeth and the resorption of the functional tooth in
early stages of their tooth cycles. The dental laminae of Archaeopteryx and the Brazilian
84

enantiornithines are more likely to be positioned deeper in the alveoli, while this structure would be
positioned closer to the middle of the alveoli of Yanornis and Ichthyornis.
Edmund’s wave of stimuli hypothesis (Edmund, 1960, 1962) and Osborn’s zone of inhibition hypothesis
(Osborn, 1974, 1998) are the two major hypotheses for explaining the mechanisms of alternating tooth
replacement patterns (Whitlock and Richman, 2013). These hypotheses were tested using dental lamina
ablation and µCT growth tracking on American alligators. The results of these experiments do not
support either of these hypotheses. Alligator specific sfrp1 has also been tested for future molecular
work aimed at identifying the distribution of this Wnt antagonist in between tooth families. Along with
the molecular data, future work should examine the possible scenario of an intrinsic control for tooth
replacement, in contrast to the more global scenario proposed by Edmund and Osborn.
Through parsimony and maximum likelihood character mapping methods, the ancestral states of the
numbers of teeth on each tooth bearing bone were reconstructed. Character evolution of the
distribution of the teeth in the premaxilla has also been investigated. Results of these studies indicate
that toothed ornithuromorph lineages have higher maxillary and dentary tooth counts than their
enantiornithine relatives. The enantiornithine clade generally has premaxillary teeth presented
throughout the premaxilla, except the longipterigids in which teeth are concentrated on the rostral end
of this bone. On the contrary, toothed ornithurines have premaxillary teeth concentrated on the
posterior part of the premaxilla. These character mapping methods can be further applied to future
studies of other dental morphologies of stem birds, including traits associated with their tooth cycle.
The research from this dissertation is the first to adapt Evo-Devo approaches to investigate the
evolution and mechanism of tooth cycling in archosaurs. It provides the basis for future studies to
further understand the complexities of tooth formation and development in Mesozoic birds.

85

References
Barrett, P.M., Evans, D.C., and Campione, N.E., 2015, Evolution of dinosaur epidermal
structures: Biology Letters (2005), v. 11, no. 6, p. 20150229.
Barrientos, S., Stojadinovic, O., Golinko, M.S., Brem, H., and Tomic-Canic, M., 2008, Growth
factors and cytokines in wound healing: Wound Repair and Regeneration: Official
Publication of the Wound Healing Society [and] the European Tissue Repair Society, v. 16,
no. 5, p. 585-601.
Bell, A., and Chiappe, L.M., 2020, Anatomy of Parahesperornis: evolutionary mosaicism in the
Cretaceous Hesperornithiformes (Aves): Life (Basel, Switzerland), v. 10, no. 5, p. 62.
Bertin, T.J.C., Thivichon-Prince, B., LeBlanc, A.R.H., Caldwell, M.W., and Viriot, L., 2018, Current
perspectives on tooth implantation, attachment, and replacement in Amniota: Frontiers in
Physiology, v. 9, p. 1630.
Bloomquist, R.F., Fowler, T.E., An, Z., et al., 2019, Developmental plasticity of epithelial stem
cells in tooth and taste bud renewal: Proceedings of the National Academy of Sciences, v.
116, no. 36, p. 17858-17866.
Brocklehurst, N., and Field, D.J., 2021, Macroevolutionary dynamics of dentition in Mesozoic
birds reveal no long-term selection towards tooth loss: iScience, v. 24, no. 3, p. 102243.
Buchtová, M., Stembírek, J., Glocová, K., Matalová, E., and Tucker, A.S., 2012, Early regression
of the dental lamina underlies the development of diphyodont dentitions: Journal of
Dental Research, v. 91, no. 5, p. 491-498.
Buchtová, M., Handrigan, G.R., Tucker, A.S., et al., 2008, Initiation and patterning of the snake
dentition are dependent on Sonic hedgehog signaling: Developmental Biology, v. 319, no.
1, p. 132-145.
Chiappe, L.M., Qingjin, M., Serrano, F., Sigurdsen, T., Min, W., Bell, A., and Di, L., 2019, New
Bohaiornis-like bird from the Early Cretaceous of China: enantiornithine interrelationships
and flight performance: PeerJ, v. 7, p. e7846.
Chiappe, L.M., and Meng, Q., 2016, Birds of stone: Chinese avian fossils from the age of
dinosaurs, Johns Hopkins University Press, 304 p.
Chiappe, L.M., Norell, M.A., and Clark, J.M., 2001, A new skull of Gobipteryx minuta (Aves:
Enantiornithes) from the Cretaceous of the Gobi Desert: American Museum Novitates, v.
3346, p. 1-15.
86

Chiappe, L.M., and Witmer, L.M., 2002, Mesozoic birds: above the heads of dinosaurs: Berkeley,
California (USA), University of California Press, 520 p.
Chiappe, L., Nava, W., Martinelli, A., Tucker, R., and Alvarenga, H., 2018, An exceptional bone
bed of Enantiornithine birds in the late Cretaceous of Brazil, in The 5th International
Palaeontological Congress, p. 170.
Clarke, J.A., 2004, Morphology, phylogenetic taxonomy, and systematics of Ichthyornis and
Apatornis: Bulletin of the American Museum of Natural History, no. 286, p. 1-179.
Clarke, J.A., Tambussi, C.P., Noriega, J.I., Erickson, G.M., and Ketcham, R.A., 2005, Definitive
fossil evidence for the extant avian radiation in the Cretaceous: Nature, v. 433, no. 7023, p.
305-308.
D’Emic, M.D., O’Connor, P.M., Pascucci, T.R., Gavras, J.N., Mardakhayava, E., and Lund, E.K.,
2019, Evolution of high tooth replacement rates in theropod dinosaurs: PLOS ONE, v. 14,
no. 11, p. e0224734.
D’Emic, M.D., Whitlock, J.A., Smith, K.M., Fisher, D.C., and Wilson, J.A., 2013, Evolution of high
tooth replacement rates in sauropod dinosaurs: PLOS ONE, v. 8, no. 7, p. e69235.
Dames, W., 1884, The Berlin Archaeopteryx: Geological Magazine, v. 3, no. 1, p. 418-424.
Davit‐Béal, T., Chisaka, H., Delgado, S., and Sire, J., 2007, Amphibian teeth: current knowledge,
unanswered questions, and some directions for future research: Biological Reviews, v. 82,
no. 1, p. 49-81.
de Queiroz, K., 1996, Including the characters of interest during tree reconstruction and the
problems of circularity and bias in studies of character evolution: The American Naturalist,
v. 148, no. 4, p. 700-708.
de Queiroz, K., 2000, Logical problems associated with including and excluding characters
during tree reconstruction and their implications for the study of morphological character
evolution: Phylogenetic Analysis of Morphological Data, p. 192-212.
DeMar, R., 1972, Evolutionary implications of Zahnreihen: Evolution, v. 26, no. 3, p. 435-450.
DeMar, R., and Bolt, J.R., 1981, Dentitional organization and function in a Triassic reptile:
Journal of Paleontology, v. 55, no. 5, p. 967-984.
Dumont, M., Tafforeau, P., Bertin, T., et al., 2016, Synchrotron imaging of dentition provides
insights into the biology of Hesperornis and Ichthyornis, the “last” toothed birds: BMC
Evolutionary Biology, v. 16, no. 1, p. 178.
87

Edmund, A.G., 1969, Dentition. in Gans, C., Bellairs, A.d. and Parsons, T.S., eds., Biology of the
Reptilia. Volume 1. Morphology A.: London and New York, Academic Press, p.117-200.
Edmund, A.G., 1962, Sequence and rate of tooth replacement in the Crocodilia, Royal Ontario
Museum, Life Sciences Division.
Edmund, A.G., 1960, Tooth replacement phenomena in the lower vertebrates: Toronto, Life
Sciences Division, Royal Ontario Museum.
Erickson, G., 1996, Incremental lines of von Ebner in dinosaurs and the assessment of tooth
replacement rates using growth line counts: Proceedings of the National Academy of
Sciences of the United States of America, v. 93, no. 25, p. 14623.
Evans, J., 1865, On portions of a cranium and a jaw, in the slab containing the fossil remains of
the Archaeopteryx: Natural History Review, no. 5, p. 415-421.
Fastnacht, M., 2008, Tooth replacement pattern of Coloborhynchus robustus (Pterosauria) from
the Lower Cretaceous of Brazil: Journal of Morphology, v. 269, no. 3, p. 332-348.
Feduccia, A., 1996, The origin and evolution of birds: New Haven, Yale University Press, 420 p.
Fehrenbach, M.J., and Popowics, T., 2016, Illustrated dental embryology, histology, and
anatomy, Saunders, 352 p.
Ferguson, M.W.J., 1985, Reproductive biology and embryology of the crocodilians. in Gans, C.,
Billett, F. and Maderson, P.F.A., eds., Biology of the Reptilia. Volume 14. Development A.:
New York, Chichester, Brisbane, Toronto, and Singapore, John Wiley & Sons, p.451-460.
Field, D.J., Benito, J., Chen, A., Jagt, J.W.M., and Ksepka, D.T., 2020, Late Cretaceous neornithine
from Europe illuminates the origins of crown birds: Nature, v. 579, no. 7799, p. 397-401.
Field, D.J., Hanson, M., Burnham, D., et al., 2018, Complete Ichthyornis skull illuminates mosaic
assembly of the avian head: Nature (London), v. 557, no. 7703, p. 96-100.
Fong, R.K.M., LeBlanc, A.R.H., Berman, D.S., and Reisz, R.R., 2016, Dental histology of
Coelophysis bauri and the evolution of tooth attachment tissues in early dinosaurs: Journal
of Morphology, v. 277, no. 7, p. 916-924.
Fujimori, S., Novak, H., Weissenböck, M., et al., 2010, Wnt/β-catenin signaling in the dental
mesenchyme regulates incisor development by regulating Bmp4: Developmental Biology,
v. 348, no. 1, p. 97-106.
Greenhalgh, D.G., 1996, The role of growth factors in wound healing: The Journal of Trauma, v.
41, no. 1, p. 159-167.
88

Hanai, T., and Tsuihiji, T., 2019, Description of tooth ontogeny and replacement patterns in a
juvenile Tarbosaurus bataar (Dinosauria: Theropoda) using CT-scan data: The Anatomical
Record, v. 302, no. 7, p. 1210-1225.
Handrigan, G.R., Leung, K.J., and Richman, J.M., 2010, Identification of putative dental epithelial
stem cells in a lizard with life-long tooth replacement: Development, v. 137, no. 21, p.
3545-3549.
Handrigan, G.R., and Richman, J.M., 2010, A network of Wnt, hedgehog and BMP signaling
pathways regulates tooth replacement in snakes: Developmental Biology, v. 348, no. 1, p.
130-141.
Harris, M.P., Hasso, S.M., Ferguson, M.W.J., and Fallon, J.F., 2006, The development of
archosaurian first-generation teeth in a chicken mutant: Current biology, v. 16, no. 4, p.
371-377.
Hieronymus, T.L., and Witmer, L.M., 2010, Homology and evolution of avian compound
rhamphothecae - Homologie et évolution du revêtement corné complexe des oiseaux: The
Auk, v. 127, no. 3, p. 590-604.
Howgate, M.E., 1984, The teeth of Archaeopteryx and a reinterpretation of the Eichstätt
specimen: Zoological Journal of the Linnean Society, v. 82, no. 1-2, p. 159-175.
Hu, H., O’Connor, J.K., Wang, M., Wroe, S., and McDonald, P.G., 2020, New anatomical
information on the bohaiornithid Longusunguis and the presence of a plesiomorphic
diapsid skull in Enantiornithes: Journal of Systematic Palaeontology, v. 18, no. 18, p. 1481-
1495.
Hu, H., O'Connor, J.K., and Zhou, Z., 2015, A new species of Pengornithidae (Aves:
Enantiornithes) from the Lower Cretaceous of China suggests a specialized scansorial
habitat previously unknown in early birds: PloS one, v. 10, no. 6, p. e0126791.
Huttenlocker, A.K., and Farmer, C.G., 2017, Bone microvasculature tracks red blood cell size
diminution in Triassic mammal and dinosaur forerunners: Current biology, v. 27, no. 1, p.
48-54.
Järvinen, E., Salazar-Ciudad, I., Birchmeier, W., Taketo, M.M., Jernvall, J., and Thesleff, I., 2006,
Continuous tooth generation in mouse is induced by activated epithelial Wnt/beta-catenin
signaling: Proceedings of the National Academy of Sciences of the United States of
America, v. 103, no. 49, p. 18627-18632.
Järvinen, E., Tummers, M., and Thesleff, I., 2009, The role of the dental lamina in mammalian
tooth replacement: Journal of Experimental Zoology Part B: Molecular and Developmental
Evolution, v. 312B, no. 4, p. 281-291.
89

Kundrát, M., Nudds, J., Kear, B.P., Lü, J., and Ahlberg, P., 2019, The first specimen of
Archaeopteryx from the Upper Jurassic Mörnsheim Formation of Germany: Historical
Biology, v. 31, no. 1, p. 3-63.
Lan, Y., Jia, S., and Jiang, R., 2014, Molecular patterning of the mammalian dentition: Seminars
in Cell & Developmental Biology, v. 25-26, p. 61-70.
LeBlanc, A.R.H., Brink, K.S., Cullen, T.M., and Reisz, R.R., 2017, Evolutionary implications of
tooth attachment versus tooth implantation: A case study using dinosaur, crocodilian, and
mammal teeth: Journal of Vertebrate Paleontology, v. 37, no. 5, p. e1354006.
Li, Z., Zhou, Z., Wang, M., and Clarke, J.A., 2014, A new specimen of large-bodied basal
enantiornithine Bohaiornis from the Early Cretaceous of China and the inference of feeding
ecology in Mesozoic birds: Journal of paleontology, v. 88, no. 1, p. 99-108.
Liu, D., Chiappe, L.M., Zhang, Y., Serrano, F.J., and Meng, Q., 2019, Soft tissue preservation in
two new enantiornithine specimens (Aves) from the Lower Cretaceous Huajiying Formation
of Hebei Province, China: Cretaceous Research, v. 95, p. 191-207.
Louchart, A., Buffrénil, V.d., Bourdon, E., Dumont, M., Viriot, L., and Sire, J., 2018, Bony
pseudoteeth of extinct pelagic birds (Aves, Odontopterygiformes) formed through a
response of bone cells to tooth-specific epithelial signals under unique conditions:
Scientific reports, v. 8, no. 1, p. 12952-9.
Louchart, A., and Viriot, L., 2011, From snout to beak: the loss of teeth in birds: Trends in
Ecology & Evolution, v. 26, no. 12, p. 663-673.
Luo, Z., Kielan-Jaworowska, Z., and Cifelli, R.L., 2004, Evolution of dental replacement in
mammals: Bulletin of Carnegie Museum of Natural History, v. 2004, no. 36, p. 159-175.
Marsh, O.C., 1880, Odontornithes: a monograph on the extinct toothed birds of North America:
Washington, U.S., U.S. Government Printing Office, 201 p.
Martin, L.D., Stewart, J.D., and Whetstone, K.N., 1980, The origin of birds: structure of the
tarsus and teeth: The Auk, v. 97, no. 1, p. 86-93.
Martin, L.D., 1984, A new hesperornithid and the relationships of the Mesozoic birds:
Transactions of the Kansas Academy of Science (1903-), v. 87, no. 3/4, p. 141-150.
Martin, L.D., and Stewart, J.D., 1977, Teeth in Ichthyornis (Class: Aves): Science, v. 195, no.
4284, p. 1331-1332.
90

Martin, L.D., and Stewart, J.D., 1999, Implantation and replacement of bird teeth, in Avian
Paleontology at the Close of the 20th Century, Smithsonian Contributions to Paleobiology
89, p. 295–300.
O’Connor, J.K., and Chiappe, L.M., 2011, A revision of enantiornithine (Aves: Ornithothoraces)
skull morphology: Journal of Systematic Palaeontology, v. 9, no. 1, p. 135-157.
O’Connor, J.K., Zhang, Y., Chiappe, L.M., Meng, Q., Quanguo, L., and Di, L., 2013, A new
enantiornithine from the Yixian Formation with the first recognized avian enamel
specialization: Journal of Vertebrate Paleontology, v. 33, no. 1, p. 1-12.
O’Connor, J.K., Zhou, Z., and Zhang, F., 2011, A reappraisal of Boluochia zhengi (Aves:
Enantiornithes) and a discussion of intraclade diversity in the Jehol avifauna, China: Journal
of Systematic Palaeontology, v. 9, no. 1, p. 51-63.
O'Connor, J.K., Gao, K., and Chiappe, L.M., 2010, A new ornithuromorph (Aves:
Ornithothoraces) bird from the Jehol Group indicative of higher-level diversity: Journal of
Vertebrate Paleontology, v. 30, no. 2, p. 311-321.
O'Connor, J.K., Wang, X., Chiappe, L.M., Gao, C., Meng, Q., Cheng, X., and Liu, J., 2009,
Phylogenetic support for a specialized clade of Cretaceous enantiornithine birds with
information from a new species: Journal of Vertebrate Paleontology, v. 29, no. 1, p. 188-
204.
O'Connor, J.K., Chiappe, L.M., Gao, C., and Zhao, B., 2011, Anatomy of the Early Cretaceous
Enantiornithine Bird Rapaxavis pani: Acta Palaeontologica Polonica, v. 56, no. 3, p. 463-
475.
Osborn, J.W., 1998, Relationship between growth and the pattern of tooth initiation in alligator
embryos: Journal of Dental Research, v. 77, no. 9, p. 1730-1738.
Osborn, J.W., 1974, On the control of tooth replacement in reptiles and its relationship to
growth: Journal of Theoretical Biology, v. 46, no. 2, p. 509-527.
Pagel, M., Meade, A., and Barker, D., 2004, Bayesian estimation of ancestral character states on
phylogenies: Systematic Biology, v. 53, no. 5, p. 673-684.
Peteya, J.A., Clarke, J.A., Li, Q., Gao, K., Shawkey, M.D., and Gabbott, S., 2017, The plumage and
colouration of an enantiornithine bird from the Early Cretaceous of China: Palaeontology,
v. 60, no. 1, p. 55-71.
Rasch, L.J., Martin, K.J., Cooper, R.L., Metscher, B.D., Underwood, C.J., and Fraser, G.J., 2016, An
ancient dental gene set governs development and continuous regeneration of teeth in
sharks: Developmental Biology, v. 415, no. 2, p. 347-370.
91

Richman, J.M., and Handrigan, G.R., 2011, Reptilian tooth development: Genesis (New York,
N.Y.: 2000), v. 49, no. 4, p. 247-260.
Sander, P.M., and Faber, C., 2003, The Triassic marine reptile Omphalosaurus: osteology, jaw
anatomy, and evidence for ichthyosaurian affinities: Journal of Vertebrate Paleontology, v.
23, no. 4, p. 799-816.
Schluter, D., Price, T., Mooers, A.O., and Ludwig, D., 1997, Likelihood of ancestor states in
adaptive radiation: Evolution, v. 51, no. 6, p. 1699-1711.
Smith, J.B., Vann, D.R., and Dodson, P., 2005, Dental morphology and variation in theropod
dinosaurs: Implications for the taxonomic identification of isolated teeth: The Anatomical
Record Part A: Discoveries in Molecular, Cellular, and Evolutionary Biology, v. 285A, no. 2,
p. 699-736.
Smith, N.D., 2012, Body mass and foraging ecology predict evolutionary patterns of skeletal
pneumaticity in the diverse "waterbird" clade: Evolution, v. 66, no. 4, p. 1059-1078.
Tsai, S., Abdelhamid, A., Khan, M.K., Elkarargy, A., Widelitz, R.B., Chuong, C.M., and Wu, P.,
2016, The Molecular Circuit Regulating Tooth Development in Crocodilians: Journal of
Dental Research, v. 95, no. 13, p. 1501-1510.
Tullberg, B.S., Ah-King, M., and Temrin, H., 2002, Phylogenetic reconstruction of parental–care
systems in the ancestors of birds: Philosophical Transitions of the Royal Society B, v. 357,
no. 1419, p. 251-257.
Wang, M., Zhou, Z., O'Connor, J.K., and Zelenkov, N., 2014, A new diverse enantiornithine
family (Bohaiornithidae fam. nov.) from the Lower Cretaceous of China with information
from two new species: Vertebrata PalAsiatica, v. 52, no. 1, p. 31-76.
Wang, M., and Lloyd, G.T., 2016, Rates of morphological evolution are heterogeneous in Early
Cretaceous birds: Proceedings of the Royal Society B: Biological Sciences, v. 283, no. 1828,
p. 20160214.
Wang, M., and Zhou, Z., 2019, A new enantiornithine (Aves: Ornithothoraces) with completely
fused premaxillae from the Early Cretaceous of China: Journal of Systematic Palaeontology,
v. 17, no. 15, p. 1299-1312.
Wang, M., and Zhou, Z., 2017, A morphological study of the first known piscivorous
enantiornithine bird from the Early Cretaceous of China: Journal of Vertebrate
Paleontology, v. 37, no. 2, p. e1278702.
Wang, S., Stiegler, J., Wu, P., et al., 2017, Heterochronic truncation of odontogenesis in
theropod dinosaurs provides insight into the macroevolution of avian beaks: Proceedings
92

of the National Academy of Sciences of the United States of America, v. 114, no. 41, p.
10930-10935.
Wang, X., O'Connor, J.K., Zheng, X., Wang, M., Hu, H., and Zhou, Z., 2014a, Insights into the
evolution of rachis dominated tail feathers from a new basal enantiornithine (Aves:
Ornithothoraces): Biological journal of the Linnean Society, v. 113, no. 3, p. 805-819.
Wang, X., O'connor, J.K., Zhao, B., Chiappe, L.M., Gao, C., and Cheng, X., 2010, New species of
enantiornithes (Aves: Ornithothoraces) from the Qiaotou Formation in Northern Hebei,
China: Acta Geologica Sinica - English Edition, v. 84, no. 2, p. 247-256.
Wang, X., Zhao, B., Shen, C., Liu, S., Gao, C., Cheng, X., and Zhang, F., 2015, New material of
Longipteryx (Aves: Enantiornithes) from the Lower Cretaceous Yixian Formation of China
with the first recognized avian tooth crenulations: Zootaxa, v. 3941, no. 4, p. 565-578.
Wang, Y., Hu, H., O'Connor, J.K., et al., 2017, A previously undescribed specimen reveals new
information on the dentition of Sapeornis chaoyangensis: Cretaceous Research, v. 74, p. 1-
10.
Wellnhofer, P., 2009, Archaeopteryx: The Icon of Evolution: Munich, Verlag Dr. Friedrich Pfeil,
208 p.
Westergaard, B., and Ferguson, M.W., 1990, Development of the dentition in Alligator
mississippiensis: upper jaw dental and craniofacial development in embryos, hatchlings,
and young juveniles, with a comparison to lower jaw development: The American Journal
of Anatomy, v. 187, no. 4, p. 393-421.
Whitlock, J.A., and Richman, J.M., 2013, Biology of tooth replacement in amniotes:
International Journal of Oral Science, v. 5, no. 2, p. 66-70.
Whitney, M.R., and Sidor, C.A., 2019, Histological and developmental insights into the
herbivorous dentition of tapinocephalid therapsids: PLOS ONE, v. 14, no. 10, p. e0223860.
Wu, P., Wu, X., Jiang, T., et al., 2013, Specialized stem cell niche enables repetitive renewal of
alligator teeth: Proceedings of the National Academy of Sciences, v. 110, no. 22, p. E2009-
E2018.
Wu, Y., Chiappe, L.M., Bottjer, D.J., Nava, W., and Martinelli, A.G., 2021, Dental replacement in
Mesozoic birds: evidence from newly discovered Brazilian enantiornithines: Scientific
Reports, v. 11, p. 19349.
Zhang, Y., O’Connor, J., Di, L., Qingjin, M., Sigurdsen, T., and Chiappe, L.M., 2014, New
information on the anatomy of the Chinese Early Cretaceous Bohaiornithidae (Aves:
Enantiornithes) from a subadult specimen of Zhouornis hani: PeerJ, v. 2, p. e407.
93

Zhou, Z., Chiappe, L.M., and Zhang, F., 2005, Anatomy of the Early Cretaceous bird
Eoenantiornis buhleri (Aves, Enantiornithes) from China: Canadian Journal of Earth
Sciences, v. 42, no. 7, p. 1331-1338.
Zhou, Z., Clarke, J., and Zhang, F., 2008, Insight into diversity, body size and morphological
evolution from the largest Early Cretaceous enantiornithine bird: Journal of Anatomy, v.
212, no. 5, p. 565-577.
Zhou, Z., and Wang, Y., 2010, Vertebrate diversity of the Jehol Biota as compared with other
lagerstätten: Science China. Earth sciences, v. 53, no. 12, p. 1894-1907.

94

Appendix A: Character Matrix
The following is the character matrix used for dental characters ancestral state reconstruction. Please
see the materials and methods session in chapter 2 for the list of characters and character states.
Taxa/Characters 1 2 3 4
Archaeopteryx 2 2 3 0
Jeholornis 0 ? 1 3
Sapeornis 2 1 1 0
Protopteryx ? ? ? 0
Orienantius 2 ? ? 0
Junornis 2/3 1 ? 0/2
Eoalulavis ? ? ? ?
Cathayornis ? ? ? ?
Gobipteryx 0 0 0 3
Eoenantiornis 2 ? 2 0/1
Neuquenornis ? ? ? ?
Sulcavis 2 1 2 0
CUGB_P1202 2 1 ? 0
Zhouornis 2 1 2 0
Bohaiornis 2 1 2 0
Parabohaiornis 2 1 ? 0
Longusunguis ? 1 2 0
Gretcheniao ? ? ? 0/2
Concornis ? ? ? ?
Elsornis ? ? ? ?
Longipteryx 2 0 1 2
Longirostravis 3 0 2 2
Rapaxavis 1 0 1 2
Shanweiniao ? 0 1 1/2
Pengornis 2 5 ? 0
Fortunguavis ? ? ? ?
Shenqiornis 2 ? ? 0
Eopengornis 2 3 3 0
Parapengornis 2 ? 2 0
Schizooura 0 0 0 3
Jianchangornis ? ? 3 ?
Longicrusavis ? ? ? 0
Apsaravis ? ? 0 ?
Hongshanornis ? ? ? ?
Gansus_yumenensis ? ? ? ?
Gansus_zheni 0 ? ? 3
Ichthyornis 0 5 4 3
Hesperornis 0 4 5 3
95

Appendix A: Character matrix (continued).
Taxa/Characters 1 2 3 4
Parahesperornis 0 4 ? 3
Vegavis ? ? ? ?
Anas 0 0 0 3
Gallus 0 0 0 3
Yanornis 3 3 3/4 1
Yixianornis 3 ? ? 1
Archaeorhynchus 0 0 0 3
Patagopteryx ? ? ? ?
Bellulornis ? ? ? ?
Confuciusornis 0 0 0 3
Changchengornis 0 0 0 3
Eoconfuciusornis 0 0 0 3

Abstract (if available)

Abstract A diversity of toothed birds has been discovered since the 19th century and these have been shown to have multiple generations of teeth (polyphyodonty). Despite detailed descriptions of the osteology of toothed birds, no rigorous investigation of their dental evolution and tooth cycles has been undertaken. In the present study, I have investigated and mapped dental characters onto an avian phylogenetic tree for ancestral state reconstructions using parsimony and maximum likelihood methods. The results of the character mapping support the modularity of dentition distribution and variation of tooth counts on each tooth bearing bone. Previous studies on avian tooth replacement have been limited to sparse data from tooth roots revealed through broken jawbones and detached teeth. However, detailed descriptions of their tooth cycles are lacking, and the specifics of their replacement patterns remain largely unknown. Here I present unprecedented data of the reconstructed tooth cycles from specimens of Archaeopteryx, enantiornithine birds, Ichthyornis, and Yanornis through high-resolution X-ray imaging methods. The cross sections of the teeth of these birds show that they migrated labially toward the functional teeth in their early developmental stages. This result deviates from the known patterns of non-avian theropods and show a homoplastic similarity with crocodilians. The high-resolution data also reveal an alternating dental replacement pattern in the studied enantiornithine specimens, which is consistent with previous studies in Archaeopteryx. These results imply a conserved control mechanism underlying the tooth cycle and dental replacement in archosaurs. Hence, developmental experiments on modern alligators were used to test two hypotheses attempting to explain these mechanisms: Edmund’s wave of stimuli hypothesis and Osborn’s zone of inhibition hypothesis. The results of these experiments—namely dental lamina ablation and growth tracking—do not support either of these two long-standing hypotheses. This study is the first to adapt Evo-Devo approaches to investigate the evolution and mechanism of tooth cycling in fossil archosaurs.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A systematic review of Enantiornithes (Aves: Ornithothoraces)

PDF

Evolution & ecology of Mesozoic birds: a case study of the derived Hesperornithiformes and the use of morphometric data in quantifying avian paleoecology

PDF

Evolution of the dinosaur flight feather: insights from 3-dimensional fossil feathers

PDF

Exploring the molecular and cellular underpinnings of organ polarization using feather as the model system

PDF

Assessing the quality of the fossil record using a phylogenetic approach

PDF

Shifting topology of stem cells and molecular expression during feather regenerative cycling

PDF

The early Triassic recovery period: exploring ecology and evolution in marine benthic communities following the Permian-Triassic mass extinction

PDF

Roles of Hox B genes in the development of chicken skin

PDF

Characterization of the progenitor cell zone in feather follicles

PDF

Paleoenvironmental and paleoecological trends leading up to the end-Triassic mass extinction event

PDF

Adaptive patterning during skin formation: assembly of vasculature and adipose tissue

PDF

Benthic and pelagic marine ecology following the Triassic/Jurassic mass extinction

PDF

Development of an “in cubo” culture system for the study of early avian development using dynamic imaging

PDF

Volumetric interactions between major ruptures and fault zones illuminated by small earthquake properties

PDF

Characterization of new stem/progenitor cells in skin appendages

PDF

Concentration and size partitioning of trace metals in surface waters of the global ocean and storm runoff

PDF

Characterizing the role of Dnmt1 and its downstream factors in progressive alopecia

PDF

Anaerobic iron cycling in an oxygen deficient zone

PDF

The geobiology of fluvial, lacustrine, and marginal marine carbonate microbialites (Pleistocene, Miocene, and Late Triassic) and their environmental significance

PDF

Spatial and temporal evolution of magmatic systems in continental arcs: a case study of dynamic arc behaviors in the Mesozoic Sierra Nevada, California

Asset Metadata

Creator Wu, Yun-Hsin (author)

Core Title Dental evolution and tooth cycling in Mesozoic birds

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Geological Sciences

Degree Conferral Date 2022-05

Publication Date 07/15/2022

Defense Date 12/10/2021

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag alligator,archosaur,Enantiornithines,Evo-Devo,Mesozoic birds,OAI-PMH Harvest,Ornithuramorpha,tooth cycle

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Bottjer, David J. (committee chair), Chiappe, Luis M. (committee member), Chuong, Cheng-Ming (committee member), Huttenlocker, Adam K. (committee member)

Creator Email ilyuyuall@gmail.com,yunhsinw@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC110520236

Unique identifier UC110520236

Legacy Identifier etd-WuYunHsin-10343

Document Type Dissertation

Format application/pdf (imt)

Rights Wu, Yun-Hsin

Type texts

Source 20220124-usctheses-batch-908 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu