Amphibian teeth: Current knowledge, unanswered questions, and some directions for future research

Abstract

Elucidation of the mechanisms controlling early development and organogenesis is currently progressing in several model species and a new field of research, evolutionary developmental biology, which integrates developmental and comparative approaches, has emerged. Although the expression pattern of many genes during tooth development in mammals is known, data on other lineages are virtually non-existent. Comparison of tooth development, and particularly of gene expression (and function) during tooth morphogenesis and differentiation, in representative species of various vertebrate lineages is a prerequisite to understand what makes one tooth different from another. Amphibians appear to be good candidates for such research for several reasons: tooth structure is similar to that in mammals, teeth are renewed continuously during life (=polyphyodonty), some species are easy to breed in the laboratory, and a large amount of morphological data are already available on diverse aspects of tooth biology in various species. The aim of this review is to evaluate current knowledge on amphibian teeth, principally concerning tooth development and replacement (including resorption), and changes in morphology and structure during ontogeny and metamorphosis. Throughout this review we highlight important questions which remain to be answered and that could be addressed using comparative morphological studies and molecular techniques. We illustrate several aspects of amphibian tooth biology using data obtained for the caudate Pleurodeles waltl. This salamander has been used extensively in experimental embryology research during the past century and appears to be one of the most favourable amphibian species to use as a model in studies of tooth development.

Full text

Biol. Rev. (2007), 82, pp. 49–81 49
doi:10.1111/j.1469-185X.2006.00003.x
Amphibian teeth: current knowledge,
unanswered questions, and some
directions for future research
Tiphaine Davit-Be
´al, Hideki Chisaka*, Sidney Delgado and Jean-Yves Sire
†
UMR 7138-‘‘Syste
´
matique, Adaptation, Evolution’’, Universite
´Pierre & Marie Curie-Paris 6Case 7077,7Quai St-Bernard,
Paris 75005, France
(Received 12 December 2005; revised 31 August 2006; accepted 11 September 2006)
ABSTRACT
Elucidation of the mechanisms controlling early development and organogenesis is currently progressing in
several model species and a new field of research, evolutionary developmental biology, which integrates
developmental and comparative approaches, has emerged. Although the expression pattern of many genes
during tooth development in mammals is known, data on other lineages are virtually non-existent. Comparison
of tooth development, and particularly of gene expression (and function) during tooth morphogenesis and
differentiation, in representative species of various vertebrate lineages is a prerequisite to understand what makes
one tooth different from another. Amphibians appear to be good candidates for such research for several reasons:
tooth structure is similar to that in mammals, teeth are renewed continuously during life ( ¼polyphyodonty),
some species are easy to breed in the laboratory, and a large amount of morphological data are already available
on diverse aspects of tooth biology in various species. The aim of this review is to evaluate current knowledge on
amphibian teeth, principally concerning tooth development and replacement (including resorption), and changes
in morphology and structure during ontogeny and metamorphosis. Throughout this review we highlight
important questions which remain to be answered and that could be addressed using comparative morphological
studies and molecular techniques. We illustrate several aspects of amphibian tooth biology using data obtained
for the caudate Pleurodeles waltl. This salamander has been used extensively in experimental embryology research
during the past century and appears to be one of the most favourable amphibian species to use as a model in
studies of tooth development.
Key words: lissamphibians, Anura, Caudata, Gymnophiona, tooth, odontogenesis.
CONTENTS
I. Introduction ...................................................................................................................................... 50
II. Critical evaluation of the use of teeth in amphibian phylogeny ..................................................... 51
(1) The origin of the lissamphibians ................................................................................................ 51
(2) The significance of teeth for lissamphibian phylogeny ............................................................. 51
(3) The significance of teeth for lissamphibian systematics ............................................................ 53
III. Lissamphibians in the laboratory ..................................................................................................... 53
IV. Overview of tooth morphology and structure in lissamphibians .................................................... 54
(1) Enameloid .................................................................................................................................. 55
(2) Dividing zone .............................................................................................................................. 56
* Present address: Department of Anatomy, Nihon University School of Dentistry at Matsudo, 870-1, Sakaecho, Nishi-2, Matsudo,
Chiba 271-8587, Japan
†
Address for correspondence: (Tel/Fax: 33-1-44-27-35-72; E-mail: sire@ccr.jussieu.fr).
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

(3) Pedicel ......................................................................................................................................... 56
V. Tooth development and replacement .............................................................................................. 60
(1) Tooth morphogenesis and differentiation .................................................................................. 60
(a) Caudata ................................................................................................................................. 60
(b) Gymnophiona ........................................................................................................................ 64
(c) Anura (excluding Pipidae) ..................................................................................................... 64
(d) Pipidae ................................................................................................................................... 65
(2) Relationships between tooth and bone support development .................................................. 66
(3) Tooth replacement and resorption ............................................................................................. 67
(4) Tooth replacement pattern ......................................................................................................... 70
VI. Tooth changes ................................................................................................................................... 71
(1) Monocuspid to bicuspid: the role of thyroxine at metamorphosis ........................................... 71
(2) Bicuspid to monocuspid: the role of androgens ........................................................................ 71
VII. Tooth regeneration ........................................................................................................................... 71
VIII. Directions for future research ........................................................................................................... 72
(1) Are pedicellate teeth homologous among lissamphibians? ....................................................... 72
(2) Are the dentition pattern and development of the dental lamina important features for
lissamphibian systematics? .......................................................................................................... 72
(3) Do ameloblasts participate in enameloid formation in lissamphibian larvae? ......................... 72
(4) How does the enameloid-enamel transition proceed through caudate ontogeny? .................. 73
(5) How do the dividing zone and the pedicel appear during lissamphibian ontogeny? .............. 73
(6) What mechanisms control the initiation of a replacement tooth in lissamphibians? ............... 73
(7) Which mechanisms control the initiation of tooth resorption? ................................................. 73
(8) What is the fate of the tooth tip in adult lissamphibians? ........................................................ 74
(9) What mechanism controls the periodicity of lissamphibian tooth replacement? ..................... 74
(10) How do thyroxine levels affect tooth shape in lissamphibian teeth? ........................................ 74
IX. Conclusions ....................................................................................................................................... 74
X. Acknowledgements ............................................................................................................................ 75
XI. References ......................................................................................................................................... 75
I. INTRODUCTION
Rapid recent progress in molecular biology and develop-
mental genetics has allowed investigators in odontology to
re-open doors that have remained closed since the end of the
1970s. Over the last two to three decades new tools have
allowed investigations to extend from tissue and cellular
integration to the molecular level, and understanding of
mechanisms controlling tooth development is progressing
rapidly using the mouse as a model species. The expression
pattern of more than 120 genes during mammalian tooth
patterning and development has been described: see http://
bite-it.helsinki.fi (Nieminen et al., 1998). We know that all
animals share many of the same molecular processes,
including regulatory genetic pathways. However, how these
commonalities are used to make one tooth different from
another is far from understood. This question can only be
answered through comparison of gene expression (and
function) during odontogenesis in various lineages or within
multiple taxa in the same lineage. Answering this question
would lead to an understanding of how teeth have changed
during evolution in terms of initiation (time), position
(space), type (morphology), mode of replacement, etc.In
fact, studies of the evolutionary developmental biology of
teeth are virtually non-existent, with the exception of some
recent evolutionary work on rodent teeth (Jernvall, Keranen
& Thesleff, 2000; Kangas et al., 2004). In particular, Kangas
et al. (2004) show correlated changes in dental characters as
a function of quantitative changes in intercellular signalling,
and conclude that most aspects of tooth shape could have
the potential for independent changes during evolution.
Although tooth diversity (e.g. shape, location, structure) is
well known in numerous species (including extinct ones)
from the main vertebrate lineages (Huysseune & Sire, 1998),
and tooth development has been compared in selected
species (Sire et al., 2002), research suffers from a lack of
comparison of the genes involved (and of their function) with
nonmammalian lineages. In parallel with detailed odonto-
genetic studies in the mouse, it is important to compare
tooth development in species representative of other lineages
(e.g. reptiles, amphibians, actinopterygian fishes, sharks) or
in a lineage that includes taxa with variants, so that tooth
evolution can be assessed in a phylogenetic framework.
Among toothed vertebrates, mammals have either a single
or two tooth generations (mono- or diphyodonty), while
nonmammalian species renew their teeth continuously
(polyphyodonty). The study of tooth development in poly-
phyodonts would seem to have several advantages for the
biologist, and molecular studies have just begun. The first
data are available for the zebrafish Danio rerio (Laurenti et al.,
2004; Jackman, Draper & Stock, 2004; Borday-Birraux et al.,
2006) but studies on tooth development are far from easy in
this species which possesses teeth in the pharyngeal region
only (Huysseune, Van der heyden & Sire, 1998; Van der
Tiphaine Davit-Be
´al and others50
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

heyden, Huysseune & Sire, 2000). In addition, the zebrafish
belongs to the actinopterygian lineage, which means more
than 420 million years of evolution separate this teleost fish
from the mouse; the tooth structure and late odontogenic
phases in fish differ from those of mammals (Huysseune et al.,
1998; Laurenti et al., 2004; Borday-Birraux et al., 2006).
Therefore, to address some of the still unanswered questions
in tooth biology, studies of species belonging to the tetrapod
lineage, such as amphibians and reptiles, may be useful.
In the context of tooth evolutionary developmental
biology, amphibians have several features of interest: (i)
their teeth have a similar structure to mammals, (ii) several
species (e.g. Pleurodeles waltl,Ambystoma mexicanum,Xenopus
laevis, Silurana tropicalis, Eleutherodactylus coqui, and Bombina
variegatus) are easy to breed in the laboratory (in contrast to
many reptiles), and (iii) for almost a century a large amount
of data has accumulated on various aspects of amphibian
tooth biology. Numerous molecular tools (a large number of
sequenced genes, possibility of transgenesis) are now
available for X. laevis and S. tropicalis (its genome is currently
being sequenced),which are used as models in embryolo-
gical and developmental studies. Numerous genes have also
been sequenced for various other species (e.g. A. mexicanum,
A. tigrinum,E. coqui; see http://www.ncbi.nlm.nih.gov).
The aims of this review are (i) to evaluate the knowledge
accumulated during the past century on amphibian teeth with
respect to development, replacement (including resorption),
changes in morphology and structure in relation to growth and
metamorphosis, and (ii) to highlight unanswered questions in
amphibian tooth biology and tooth development.
II. CRITICAL EVALUATION OF THE USE OF
TEETH IN AMPHIBIAN PHYLOGENY
(1) The origin of the lissamphibians
Amphibians appeared by the end of the Devonian or the
early Carboniferous [approximately 300 million years ago
(mya)], when the two tetrapod lineages, reptiliomorphs
(which include amniotes) and amphibians, separated from
a tetrapod ancestor (Laurin, 1998a,b; Carroll, 1988). They
comprise both living species and their extinct relatives,
grouped into the lissamphibian clade (frogs, salamanders
and caecilians), and several extinct lineages that have been
grouped either into a large group including lepospondyls
and temnospondyls (Trueb & Cloutier, 1991; Lombard &
Sumida, 1992; Ahlberg & Milner, 1994), or into lepospondyls
only (Laurin & Reisz, 1997; Laurin, 1998a,b). Lissamphi-
bians are supposed to have originated at the onset of the
Triassic period (approximately 250 mya), probably from
a lepospondyl ancestor (Laurin & Reisz, 1997, 1999;
Laurin, 2002). However, the fossil record has provided
little evidence on the evolutionary origin of lissamphibians,
and it is difficult to postulate which group among the
Paleozoic lepospondyls is most closely related to them
(Laurin, 1998a,b; Anderson, 2001). This explains why the
question of the origin of the lissamphibians has been long
debated in the literature (Romer, 1945; Holmgren, 1952;
Eaton, 1959; Jarvik, 1960; Bolt, 1969, 1977, 1979, 1991),
and why debate continues (e.g. Milner, 1988, 1993, 2000;
Trueb & Cloutier, 1991; Laurin, 1998a, 1998b, 2002;
Schoch & Carroll, 2003). After re-examination of a number
of characters in extant and extinct amphibian species
(including skeleton and soft anatomy), the hypothesis of
a common ancestry for the lissamphibians has nevertheless
been retained (Szarski, 1962; Parsons & Williams, 1963;
Laurin, 1998a,b, 2002; Schoch & Carroll, 2003). This hy-
pothesis is also supported by molecular phylogenies showing
the monophyly of lissamphibians: caecilians and salaman-
ders being sister taxa, with frogs their outgroup (Hedges,
Moberg & Maxson, 1990; Hedges & Maxson, 1993; Hay
et al., 1995; Feller & Hedges, 1998). Fossil records indicate
that the crown-group lissamphibians started diversifying by
the end of the Permian (approximately 250 mya), before the
breakup of Pangaea, and their diversity increased greatly
during the Jurassic and Early Cretaceous periods (approx-
imately 200-150 mya) (Schoch & Carroll, 2003). This was
recently confirmed using a molecular phylogeny (San
Mauro et al., 2005). Putative ancestors of salamanders are
recognized from the Carboniferous-Permian boundary
(Schoch & Carroll, 2003), a fossil caecilian possessing
reduced limbs, Eocaecilia micropodia, has been discovered
from the Jurassic period in Arizona (Jenkins & Walsh, 1993),
and frogs are also known from the Triassic and Jurassic
(Estes & Rieg, 1973; Roelants & Bossuyt, 2005).
Today (AmphibiaWeb database, Nov. 2005), Lissamphi-
bia contains 5953 species distributed into three orders:
Gymnophiona (caecilians) with 171 species; Anura (frogs,
including pipids, and toads) with 5230 species; and Caudata
(salamanders and newts) with 552 species. Note that we use
the current standard taxonomic reference for the amphibian
orders, i.e. node-based names defined on the basis of Recent
taxa instead of stem-based names which include the fossil
taxa: Anura instead of Salientia for frogs, Caudata instead of
Urodela for salamanders, and Gymnophiona instead of
Apoda for caecilians (e.g. Trueb & Cloutier, 1991; Canna-
tella & Hillis, 1993, 2004; Ford & Cannatella, 1993; Frost,
2004]. Fewer than ten species from each order have been
examined so far with respect to tooth development (Fig. 1).
(2) The significance of teeth for lissamphibian
phylogeny
In most stem-tetrapods and in extinct amphibians, teeth
were haplodont (i.e. simple: conical and unicuspid) with
some heterodonty (i.e., differing in general appearance
throughout the mouth but mainly in size) in a few species.
Tooth attachment to the bone support was in general
subthecodont (i.e. partially set in a socket), and sometimes
pleurodont (i.e. attached to the labial side). Tooth structure
has been studied in a few early tetrapods (e.g. temnospond-
yls) and lepospondyls (microsaurs, nectrideans) (e.g. Owen,
1842; Bystrow, 1938; Parsons & Williams, 1962; Peyer,
1968; Bolt, 1969, 1979). In general, the teeth were conical
with a large base. The dentine shaft surrounded a pulp
cavity and was covered by a thin enamel layer. Tooth
structure was characterised by a typical folded arrangement
of the dentine, called plicidentine (Fig. 2). Plicidentine is,
however, not a typical feature of early amphibians, and
Amphibian tooth morphology and development 51
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

cannot be used as a strong phylogenetic argument: it is
absent in several Paleozoic amphibians, in particular among
microsaurs (Peyer, 1968), and it is widespread in basal
sarcopterygian taxa (Schultze, 1969). Some temnospondyls
possessed a branchial apparatus, in which small tooth-
bearing plates occurred in the throat region (Hook, 1983;
Coates, 1996), a location which is similar to the pharyngeal
teeth described in a number of actinopterygians.
The value of tooth characters as evidence of lissamphi-
bian phylogeny has been investigated in depth by Parsons &
Williams (1962, 1963). Although the three lissamphibian
orders possess relatively few distinguishing characters
(which explains the current debate on their relationships),
the presence of bicuspid and pedicellate teeth has been
widely accepted as strong support for their monophyly (see
discussion in Laurin, 1998a; Schoch & Carroll, 2003;
Schoch & Milner, 2004). In a series of investigations on the
morphology of the mouth cavity of caudates, H. Greven,
G. Clemen and others (see, e.g. Greven & Clemen, 1979,
1980, 1985; Clemen & Greven, 1977, 1979, 1980, 1988,
2000) have shown that the number and course of dental
laminae are also of phylogenetic importance. Lissamphibian
teeth are characterised by the division of the dentine shaft
into a relatively short crown and a long pedicel, separated
by an uncalcified (or poorly calcified) region resembling a
ligament, called the dividing zone. Pedicellate teeth are
present in fossil representatives of caudate, gymnophione
and anuran lineages. However, in a few lissamphibian
species, teeth lack a dividing zone, but this feature is
considered a derived rather than a plesiomorphic character
(Parsons & Williams, 1962; Parker & Dunn, 1964; Means,
1972). The presence of bicuspid teeth in adults also has
been tentatively used to support close lissamphibian
relationships, but such a character is not restricted to
amphibians (Bolt, 1969). Bicuspid teeth are not primitive for
tetrapods and originated more than once in early tetrapods,
which may or may not be true of pedicellate teeth (Bolt,
1980). Pedicellate teeth is probably a more primitive
condition because it has been encountered in various
stem-tetrapod lineages. In addition, some taxa have only
monocuspid teeth in adults, such as pipids (e.g. Xenopus laevis:
Cambray, 1976) or several gymnophione genera (e.g.
Fig. 1. Amphibian relationships with particular focus on the taxa investigated with respect to tooth development. †indicates
extinct taxa. After Larson & Dimmick (1993), Laurin & Reisz (1997), Feller & Hedges (1998), San Mauro et al. (2004a,b).
Fig. 2. (A) Tooth of Palaeogyrinus, an extinct Embolomeri,
a stem tetrapod sensu Laurin (1998a). (B) Transverse section of
the crown showing the enamel. (C) Transverse section of the
mid shaft showing the typical folded dentine, plicidentine.
Modified from Miles & Poole (1967).
Tiphaine Davit-Be
´al and others52
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

Dermophis, Gymnopis, Caecilia, Gegeneophis, Typhlonectes: Wake,
2003). The monocuspid condition in adults is, however,
considered to be phylogenetically different from the
monocuspid condition in larvae (Wake & Wurst, 1979;
Greven, 1984; Beneski & Larsen, 1989a,b). Pedicellate teeth
(in general bicuspid) are therefore the only dental character
interpreted as evidence of close amphibian relationships
(e.g. Schultze, 1969, 1970; Bolt, 1980; Lombard & Sumida,
1992). Nevertheless, pedicellate teeth with a distinct
separation between the crown and the pedicel have been
described in extinct temnospondyls (Doleserpeton from the
Lower Permian: Bolt, 1969; Apateon: Schoch & Carroll,
2003; and other branchiosaurids: Boy, 1978). In contrast to
the condition observed in mature stages, in Apateon larvae
the teeth do not have a gap between the base and the crown
(Schoch & Carroll, 2003). In modern salamanders, the first-
generation teeth in larvae do not possess a pedicel, while
pedicels are well formed in juvenile specimens (Wistuba,
Greven & Clemen, 2002).
(3) The significance of teeth for lissamphibian
systematics
Dentition pattern, dental lamina development, and crown
morphology have been suggested to be important features to
establish relationships within the lissamphibian orders,
mostly Caudata (e.g. Laurent, 1947; Regal, 1966; Clemen,
1978a,b) and Gymnophiona (e.g. Wake & Wurst, 1979;
Clemen & Opolka, 1990; Wilkinson, 1991). However,
several studies have revealed intraspecific and ontogenetic
variations in tooth morphology (e.g. Wake, 1980; Beneski &
Larsen, 1989a,b). During the last 25 years, G. Clemen,
H. Greven and colleagues have published a series of detailed
descriptions of the mouth cavity and the dentition pattern in
numerous caudate species (Clemen, 1979a-c, 1988; Greven
& Clemen, 1985, 1990; Clemen & Greven, 1988; Ehmcke &
Clemen, 2000a). Such a large amount of data allows
comparison of the development of the dentition pattern,
the organisation of the dental lamina, and variations in tooth
shape in relation to the location of the teeth in the oral cavity,
among species and between sexes. It is beyond the scope of
this review to summarise all these descriptions, but some
interesting points are highlighted below.
In the plethodontid salamander Bolitoglossa subpalmata
(Boulenger, 1896),a direct developer, teeth are absent on
the upper jaw in young individuals, but present in adults.
This feature has been correlated to the different diet in
juveniles and adults. Juveniles and young adults use their
well-developed tongue to transport the small prey deep into
the mouth, towards the vomerine dentition, while adults
feed on larger prey (Wake & Deban, 2000). The teeth of the
upper jaw are, therefore, of very little importance in young
individuals (Ehmcke & Clemen, 2000a). Sirenids lack teeth
on the upper jaw (Clemen & Greven, 1988), but this loss
of upper jaw dentition may be secondary when considering
the condition in a fossil sirenid (Habrosaurus dilatus), which
bears teeth on the premaxillae and the maxillae (Estes,
1965).
During the breeding season, in some plethodontids the
males have a few, long (300 mmversus 200 mm in the interim
period), monocuspid teeth protruding from the upper lip
(Noble, 1929; Stewart, 1958; Clemen & Greven, 2000;
Ehmcke & Clemen, 2000a; Ehmcke et al., 2003). The males
use such teeth to stimulate the female during courtship. The
temporary monocuspidity (versus bicuspid teeth during the
interim period) of these particular teeth in males is under
the influence of androgens (Stewart, 1958). This suggests
that the premaxillary dental lamina only reacts to the rising
androgen levels at the beginning of the breeding season
(Ehmcke & Clemen, 2000b). Because tooth shape can only
be changed through tooth replacement this implies that the
bicuspid teeth located in this region of the upper jaw are
lost and replaced by monocuspid teeth during the breeding
season (see also Section VI).
It is known that metamorphosed caudates have bicuspid
teeth, while the teeth are monocuspid in the larvae;
bicuspidity being established during, or immediately after,
metamorphosis (Kerr, 1960; Chibon, 1972; Clemen &
Greven, 1974, 1977, 1979). As a consequence, monocus-
pidity in larvae must be regarded as a plesiomorphic
condition as reported for first-generation teeth in actino-
pterygians (Sire et al., 2002). However, monocuspid teeth
have been reported in some metamorphosed lissamphibians
such as pipid anurans (Katow, 1979; Greven & Laumeier,
1987), some salamanders such as the plethodontid Aneides
lugubris (Wake, Wake & Wake 1983) and several gymno-
phione genera (Taylor, 1968; Wake & Wurst, 1979; Greven
& Clemen, 1980; Wake, 2003). Do they express a less
derived condition in these species than in other lissamphi-
bians? Greven (1984) points out that the monocuspid (spike-
like) teeth in adult caudates are morphologically different
from those (with sharp edges) observed in larvae, the former
type being regarded as less derived. Such a careful
distinction may be useful in understanding lissamphibian
relationships. Wilkinson (1991) discussed whether mono-
cuspid teeth are derived or not within Gymnophiona.
III. LISSAMPHIBIANS IN THE LABORATORY
For more than a century (e.g. Owen, 1845) investigators
have taken advantage of the relative ease with which
lissamphibians can be reared in captivity from eggs or
larvae caught in the wild to study the dentition of numerous
species, especially frogs (e.g. Rana pipiens Schreber, 1782),
and salamanders [Cynops pyrrhogaster Boie, 1826; Necturus
maculosus (Rafinesque, 1818), Ambystoma mexicanum (Shaw &
Nodder, 1798) and Pleurodeles waltl Michahelles, 1830].
Around the middle of the 20
th
Century experimental work
concentrated on species from which numerous eggs could
be obtained in the laboratory, either naturally (e.g.
Ambystoma mexicanum and Pleurodeles waltl) or by artificial
induction [Xenopus laevis (Daudin, 1802) and, recently,
Silurana tropicalis (Gray, 1864)]. Appropriate breeding
conditions and developmental Tables were published:
Taylor & Kollros (1946) for R. pipiens; Nieuwkoop & Faber
(1956) for X. laevis; Gallien & Durocher (1957) for P. waltl;
and Bordzilovskaya & Dettlaff (1979) for A. mexicanum.
Thanks to these experimental model species major
Amphibian tooth morphology and development 53
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

advances were obtained in the understanding of lissamphi-
bian tooth development in larvae, mainly in A. mexicanum
and P. waltl, and/or in juvenile and adult specimens
(caudates and various anurans). Although studies dealing
with tooth morphogenesis and differentiation in lissamphi-
bians have declined since the end of the 1970s, the number
of species bred in laboratories is increasing, and three model
species (X. laevis,A. mexicanum and, to a lesser degree, P. waltl)
are still used in laboratories studying early developmental
processes, reproduction biology, and many other topics.
Unquestionably, the most studied species is X. laevis, for
which numerous developmental genes have been cloned.
However, in X. laevis the teeth form late, at the end of the
larval period, i.e. 2-3 months after hatching (Cambray,
1976; Shaw, 1979). By contrast, in Caudata the first teeth
start to develop by the end of the embryonic period, similar
to the situation in actinopterygian fish (Sire et al., 2002).
A. mexicanum and P. waltl are, therefore, more appropriate
model species to study tooth development in lissamphibians
at the molecular level, particularly in an evolutionary
perspective. Of these, P. waltl is preferred because
A. mexicanum is generally neotenic and has a large genome.
Recently, Silurana tropicalis has become a widespread
model species in the lab. It has the advantages of being
diploid (versus tetraploid in Xenopus laevis), growing more
rapidly (25°C), having shorter generation time, and genome
sequencing is already well advanced (see www.xenbase.org).
In the near future this species will substitute X. laevis for
most developmental studies, including odontogenesis.
IV. OVERVIEW OF TOOTH MORPHOLOGY
AND STRUCTURE IN LISSAMPHIBIANS
Adult lissamphibians possess an haplodont dentition, with
conical or cylindrico-conical, generally homodont teeth, but
some caudate and gymnophione species have an heterodont
dentition (Greven, 1984, 1986; Wake, 1980; Wake et al.,
1983). Teeth are restricted to the oral cavity. Lissamphi-
bians, as most nonmammalian taxa, replace their teeth
continuously during life, i.e. they are polyphyodont.
Caudate and gymnophione teeth have a large diversity of
size, shape (mono-, bi-, pluricuspid) (Fig.3), and mode of
attachment (pleurodont for most teeth, except for the
palatal teeth, which are acrodont). This diversity contrasts
with a number of well-conserved features, such as tooth
structure (a pulp cavity surrounded by a dentine cone
covered by enamel; a crown and a pedicel separated by
a dividing zone) (Fig. 4), orientation (often lingual), and
Fig. 3. Examples of tooth morphology in lissamphibians. (A, B, C) Tooth shape throughout ontogeny in the caudate, Pleurodeles
waltl. (A) First-generation tooth in a larva, stage 44. The tooth is monocuspid and the dividing zone is lacking. (B) Third- (left) and
fourth- (right) generation tooth in a five-month-old, postmetamorphosed specimen. The teeth are bicuspid and the dividing zone
is visible. (C) Detail of the tooth tip in an adult showing the two cusps. The main cusp is lingually oriented. (D, E, F) Teeth
in Gymniophona. (D) Typical tooth morphology in an embryo of Geotrypetes seraphini (left) and in a foetus of Nectocaecilia petersi (right).
(E) Adult tooth in Hypogeophis rostratus. (F) Adult tooth in Geotrypetes seraphini. (G) Adult tooth in the anuran Bombina bombina (Linnaeus,
1761). D modified from Parker & Dunn (1964); E, F from Wake & Wurst (1979): G from Clemen & Greven (1980). Scale bars: A, B,
D-G ¼100 mm; C ¼10 mm.
Tiphaine Davit-Be
´al and others54
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

replacement (lingual). In lissamphibians, the ameloblasts do
not form Tomes’ processes (which are supposed to play
a role in enamel crystal orientation: Carlson, 1990) and
amphibian enamel differs from mammalian in being non-
prismatic (Zaki, Yaeger & Gilllette, 1970; Zaki & MacRae,
1977, 1978; Kogaya, 1994). In post-metamorphic salaman-
ders the presence of enamel matrix proteins has been
identified using immunocytochemistry, especially revealing
amelogenin-like proteins (Herold, Rosenbloom & Granovsky,
1989). The first deposited enamel matrix forms globular
patches within which the enamel crystals are mostly radially
arranged (Kallenbach & Piesco, 1978). In later stages of
amelogenesis, a thick enamel layer is formed in which the
enamel crystals are oriented perpendicularly to the tooth
surface (Kogaya et al., 1992; Kogaya, 1994).
Three particular regions of lissamphibian teeth have
been the subject of many discussions in the past, and have
raised questions, still unanswered, with respect to the
enameloid/enamel transition, the formation of the dividing
zone and the nature of the pedicel.
(1) Enameloid versus enamel
The nature of the enamel-like material covering the teeth in
larval and adult caudates has long been debated since the
pioneering studies of Owen (1845), Leydig (1867) and
Hertwig (1874). Levi (1940), Kvam (1946, 1953, 1960) and
Kerr (1960) believed that the external covering of the
monocuspid teeth in larvae was a ‘‘mesodermal enamel’’,
i.e. a highly mineralised dentine, called durodentine,
exclusively deposited by the dental papilla cells. Using
polarized light, Schmidt (1957, 1958) considered the outer
surface of the adult teeth to be durodentine. Later, he
changed his view and suggested that this layer is an
‘‘ectodermal enamel’’, i.e. deposited by the enamel organ
(Schmidt, 1970). In a first attempt to study amelogenesis in
the caudate Pleurodeles waltl, Chibon, Roux & Spinelli (1971)
did not find enamel covering the teeth until metamorphosis.
In fact, in larval teeth the thin enamel layer is hardly visible
at the light microscopical level, and only transmission
electron microscopic (TEM) observations have revealed its
presence (Smith & Miles, 1971; Chibon, 1972; Roux &
Chibon, 1973; Roux, 1973). In larvae, the dental papilla
cells, the odontoblasts, deposit first a layer of a ‘‘particular
dentine’’ that mineralises more strongly than regular
dentine. This layer is now called enameloid, a term
introduced by Poole (1967) and Ørvig (1967) to replace
the confusing terms ‘‘mesodermal enamel’’, ‘‘vitrodentine’’
and ‘‘durodentine’’. In fact, the difficulty of recognising
enameloid in larval teeth resides in the fact that the first
Fig. 4. Schematic drawings showing the tooth structure and the relations to the supporting bone in adult lissamphibians. (A)
Generalised lissamphibian tooth. (B) Proteid (caudate) Necturus maculosus. (C) Salamandrid (caudate) Salamandra salamandra. (D) Ranid
(anuran) Rana pipiens. (E) Hylid (anuran) Hyla cinerea. (F) Gymnophione Hypogeophis rostratus. A modified from Casey & Lawson
(1981); B from Kerr (1960); C-F from Lawson (1966). db: dentary bone; de: dentine; dl: dental lamina; dz: dividing zone; en:
enamel; Hs: Hertwig’s sheath; mb: maxillary bone; n: nerve; oe: oral epithelium; pc: pulp cavity; pde: predentine; pe: pedicel.
Amphibian tooth morphology and development 55
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

matrix deposited by odontoblasts at the tooth tip resembles
predentine matrix in that it contains a relatively dense
collagen network (Fig. 5). The ‘‘pre-dentine-like’’ matrix is
secondarily converted into enameloid during the mineral-
isation and maturation process, which could be under the
influence of the inner dental epithelial cells, the ameloblasts.
These eventually deposit a thin layer of true enamel at the
enameloid surface. The presence of a thin layer of true
enamel covering enameloid in larval teeth has been
confirmed in the first-generation teeth of the caudate
Ambystoma mexicanum using scanning electron microscopy
(SEM) (Bolte & Clemen, 1992).
The layer of enameloid does not exist (or is extremely
reduced and located at the dentine-enamel junction) in
adult teeth, in which a thick layer of enamel covers the
dentine directly (Chibon et al., 1971; Smith & Miles, 1971).
In adults of some lissamphibian species, enamel is orange
due to the presence of iron ions, which are concentrated
within ferritin patches in the secretory ameloblasts (Randall,
1966). The formation of either enamel or enameloid is
thought to be related to heterochrony in the secretion of
ameloblasts and odontoblasts (Smith, 1995).
(2) Dividing zone
In the three orders of lissamphibians, adult teeth are usually
composed of two distinct regions: a proximal (basal) pedicel
(or pedestal) and a distal crown, separated by a well-marked,
transverse zone of weakness (Leydig, 1867; Gillette, 1955;
Parsons & Williams, 1962; Means, 1972). This is in contrast to
the presence of undivided teeth in most larval stages. In most
gymnophione genera, however, foetal teeth have a discrete
pedicel (Wake, 1976, 1980). In their extensive study of
lissamphibian tooth structure (42 Caudata, 8 Gymnophiona,
118 Anura), Parsons & Williams (1962) noted that this
separation into a crown and a pedicel is present in most adult
lissamphibians. There are, however, a few exceptions, in
which the division within the teeth has been reported as
absent, such as in the caudate Siren lacertina O
¨sterdam, 1766
and in the anuran Xenopus laevis (e.g. Parsons & Williams,
1962; Means, 1972). In these cases the teeth are calcified from
the crown to the base and anchored to the jawbone by
attachment bone (Katow, 1979; Shaw, 1979). Tesche &
Greven (1989) also report that the first-generation teeth in
anurans are not pedicellate. There are no reports of any adult
gymnophionans lacking the dividing zone.
The dividing zone most commonly appears as a well-
defined transverse region, resembling a suture between two
bones (Fig.6). This zone of weakness is revealed by the
tendency for the crowns to fall off in jaws that have been
vigorously cleaned. In such cases, because they are firmly
fused to the bone support, the pedicels are left as hollow
cylinders. Alizarin red staining has revealed that this zone of
division is either not, or is only slightly, mineralised (Gillette,
1955). This low level of mineralisation of the collagen matrix
was further confirmed by TEM studies (Wistuba et al., 2002).
Although the functional significance of this zone is not
known with certainty, it has been suggested that, in addition
to providing a certain degree of flexibility as a ligament, it
allows the tip of the tooth to break off without damage to the
underlying bone (Larsen & Guthrie, 1975; Moury, Curtis &
Pav, 1985, 1987). Working on Plethodon cinereus, Moury et al.
(1985) concluded that this uncalcified region allows tooth
flexion only in a posterior direction. In Pleurodeles waltl, most
of the dividing zone looks like a ligament linking the crown
to the pedicel (Fig. 6 C, D).InAmbystoma mexicanum,itis
noteworthy that the odontoblasts facing the dividing zone
lack cytoplasmic processes and that, therefore, this region is
devoid of dentine tubules (Wistuba et al., 2002). In P. waltl,
indeed, odontoblast processes are hardly visible in this zone
(Fig. 6E-G). The pedicel also lacks dentine tubules. The
dividing zone appears, therefore, to be a transition between
the tubular dentine of the crown and the atubular dentine of
the pedicel.
(3) Pedicel
The crown is universally agreed to be dentine, but the
nature of the pedicel has long been the subject of discussion.
Indeed, the lack of dentine tubules and the presence of
some enclosed cells in this region have led authors to
consider the pedicel either as composed of cement (e.g.
Hertwig, 1874) or of bone (e.g. Oltmanns, 1952; Schmidt,
1957). Sirena (1872) considered that the pedicel was part of
the jawbone. Hertwig (1874) was the first to assert that since
the pedicel is formed within the epithelial sheath and is
formed anew each time a tooth is replaced, it should be
considered part of the tooth rather than a bony projection
of the jaw. He also assigned the term ‘‘cementum’’ to the
substance of the pedicel since it contains cell bodies. Cell
bodies were never observed in the pedicel of young P. waltl
we have studied (Fig. 7). Studying tooth replacement in the
anuran Rana pipiens, Gillette (1955) was the first to conclude
that the pedicel is composed of dentine laid down by
odontoblasts, and that the cementum is located only at the
base of the pedicel and at its outer surface. This finding was
confirmed by Parsons & Williams (1962) in gymnophione
teeth, and by Kerr (1960) and by us herein in caudate teeth
(Fig. 7). However, despite their common origin, in adult
caudates the crown and pedicel possess a different Ca/P
ratio (Clemen, Greven & Schro
¨der, 1980; Bolte, Krefting &
Clemen, 1996). In Ambystoma mexicanum this ratio, which
determines the hardness of the tooth region, is 45.9% in the
central crown versus 38.4% in the central pedicel. This
indicates that the pedicel is less hard than the crown (the Ca/
P ratio is 53.9% in enamel and 34.9% in the bone support).
It is now well established that the pedicel is part of the
tooth. However, the term ‘‘pedicel’’ is restricted to the
mineralised cylinder of dentine which is located below
the dividing zone, while the term ‘‘tooth base’’ includes the
bone of attachment that links the pedicel to the bone
support (Moury et al., 1987). The dentine of the pedicel
lacks tubules, but it is clearly distinct from the bone of the
jaw (Howes, 1978) (Fig. 7). The same conclusion that the
pedicel/attachment bone is part of the tooth was reached
for the bone of attachment of the teeth in lower vertebrates
(Peyer, 1968; Sire & Huysseune, 2003). In fact, the base of
the pedicel is linked to the bone support by a particular
zone, which could be considered to be bone of attachment
(Fig. 7A, D). The cement-like tissue is deposited on the
Tiphaine Davit-Be
´al and others56
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

Fig. 5. Enameloid: a peculiar feature of the teeth in caudate larvae. (A) Schematic drawing of a developing tooth indicating the
location of enameloid between dentine and enamel. (B) In this first-generation tooth of Pleurodeles waltl, the enameloid matrix has
been recently deposited by the odontoblasts. Some cytoplasmic prolongations of the odontoblasts are visible in the enameloid
matrix (arrowheads). No basement membrane is visible between the ameloblast surface and the enameloid (arrows). The cytoplasm
of the ameloblasts shows large, dilated vacuoles and numerous small vesicles, but a rough endoplasmic reticulum network is hardly
visible. The enameloid matrix is composed of thin collagen fibrils loosely organised, except along the tooth surface, where they run
parallel to the cell surface. The first elements of the predentine matrix have been deposited below the enameloid. (C) Mineralisation
stage. This sample was decalcified using ethylenediaminetetraacetic acid (EDTA); the narrow, empty space located between the
enameloid and the ameloblast surface indicates that a thin layer of enamel was present at the enameloid surface, but removed
during the decalcification process. Around the tooth tip the ameloblasts show numerous cell membrane folds, which characterize
the postsecretory phase. Asterisks indicate cell prolongations from odontoblasts. (D) Enlargement of the tip of the tooth in C
showing the ameloblasts located at the enameloid surface and their prominent folds. The foamy aspect of the enameloid matrix
indicates that the mineralisation process has started. A modified from Smith & Miles, 1971; B, C, D original micrographs. Scale
bars: B, C ¼1mm; D ¼250 nm. am: ameloblasts; en: enamel; ena: enameloid; de: dentine; od: odontoblasts; pde: predentine.
Amphibian tooth morphology and development 57
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

Fig. 6. Dividing zone in the teeth of Pleurodeles waltl. (A, B) Scanning electron micrographs showing the dividing zone in
a developing (A) and a functional (B) tooth in an adult. (C) Five-month-old specimen. One mm-thick, vertical section of a functional
tooth showing the dividing zone (arrows) separating the crown from the pedicel. The matrix is thicker at the level of the dividing
zone than elsewhere along the tooth shaft, and a large part of this matrix is not mineralised. (D) Detail of the dividing zone of the
tooth in C. The mineralisation front is irregular. (E, F, G) Transmission electron micrographs. (E) General view of the structure of
the dividing zone (lingual side). The arrows indicate the mineralisation front. (F, G) Detail of the dividing zone in the region facing
the pulp (F) and facing the mesenchyme (G). (F) The surface of the dividing zone is irregular and covered by large, active
odontoblasts, which deposit a matrix composed of thin, unmineralised collagen fibrils. (G) Flattened cells of the retracting enamel
organ cover the tooth matrix, from which they are separated by a thin, barely visible basement membrane (arrow). Facing the cell
the matrix is composed of a loose network of thin unmineralised collagen fibrils, which mostly run parallel to the tooth surface. At
a distance from the cell, the collagen fibrils are thicker. Scale bars in A, B ¼100 mm; C ¼50 mm; D ¼10 mm; E ¼2mm; F, G ¼
1mm. de: dentine; dz: dividing zone; eo: enamel organ; od: odontoblasts; oe: oral epithelium; pc: pulp cavity; pe: pedicel.
Tiphaine Davit-Be
´al and others58
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

Fig. 7. The pedicel, pulp cavity and cementum of the teeth of Pleurodeles waltl. (A-D) One-mm thick, transverse sections of the teeth
in a larval stage 42 (A), and in three-month- (B), five-month- (C) and eight-month- (D) old specimen. (A) In larval teeth, the pulp
cavity contains only a few cells. It was secondarily invaded by blood vessels. The dividing zone is hardly visible. Note that the tooth
on the left is attached on one side onto the bone support and, on the other, to the attachment bone region of the adjacent tooth; in
both locations attachment bone is deposited on each surface (arrows). (B) In juveniles the pulp cavity of the developing teeth
contains a large number of more or less organised cells. (C) During growth the odontoblasts that deposit the predentine matrix are
well organized and polarised, while the centre of the pulp contains blood vessels and undifferentiated cells. (D) On the pulpal and
mesenchymal side the pedicel surface of the functional teeth is lined by odontoblast- and osteoblast-like cells, respectively. At the
pulp side, the odontoblasts are depositing predentine at the dentine surface (arrow). At the mesenchymal side the enamel organ (the
so-called cervical loop) has retracted, and a reversal line is visible (arrowheads), delimiting the dentine matrix from a thin layer of
cement, which has been secondarily deposited on the pedicel surface by osteoblast-like cells. The latter are more active at the
pedicel base, where they deposit the attachment bone matrix. (E, F) Electron micrographs of the attachment zone. (E) 12-month-old
specimen. Odontoblasts depositing predentine along the pulpal side of the pedicel of a functional tooth (arrow in D). (F) Larva,
stage 55. Osteoblast-like cells (¼cementoblasts) depositing a thin collagenous matrix on the outer surface of the pedicel. Scale bars:
A, B, D ¼10 mm; C ¼50 mm; E, F ¼1mm. ab: attachment bone; bv: blood vessel; ce: cement; db: dentary bone; de: dentine; eo:
enamel organ; ob: osteoblast; od: odontoblast; oe: oral epithelium; pc: pulp cavity; pde: predentine; pe: pedicel.
Amphibian tooth morphology and development 59
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

dentine surface of the pedicel by osteoblast-like cells, in the
region where Hertwig’s sheath has retracted after the tooth
has become functional. In Pleurodeles waltl the cementum is
not deposited in the first-generation teeth.
V. TOOTH DEVELOPMENT AND
REPLACEMENT
Since Hertwig (1874) established the first bases of
knowledge on lissamphibian teeth, developmental events
have been well documented in many species. Here, we
review generalised lissamphibian tooth development (mor-
phogenesis and differentiation), resorption and replacement
patterns, and then discuss in detail the three orders,
Caudata, Gymnophiona and Anura. Within the anurans
we will describe the Pipidae separately. We pay particular
attention to the salamander Pleurodeles waltl, a model species
widely used in tooth development.
A characteristic feature resulting from polyphyodonty in
lissamphibians is that, at a given time, several replacement
teeth can be found in a single specimen and, especially, in
juvenile stages. This is a considerable advantage when
studying tooth morphogenesis and differentiation. Indeed,
all stages of tooth development can be found on a jaw, but
early stages are barely visible at the light microscopical
level.
(1) Tooth morphogenesis and differentiation
Tooth development in a generalised lissamphibian is
schematically illustrated in figure 8 and micrographs of
P. waltl detail specific stages in a caudate (Fig. 9).
The initiation of the first-generation teeth, in which the
dental lamina develops directly from the oral epithelium,
begins at stage 34 (11 dpf). The dental lamina consists of an
epithelial invagination, two cell layers wide, into the
mesenchyme. Then, in particular regions of the dental
lamina and facing the mesenchyme, the basal epithelial cells
differentiate into placodes. Mesenchymal cells, originating
from the neural crest (e.g. Wagner, 1949; Chibon, 1966)
concentrate at the level of the placodes. The basal layer cells
of the dental lamina invaginate more or less deeply into the
mesenchyme and develop into a cap. The dental epithelium
differentiates into an enamel organ composed of two cell
layers, the inner and the outer dental epithelium, while the
facing mesenchymal cells differentiate into a dental papilla
(Figs. 8A, 9A,B). The cells of the inner dental epithelium
differentiate into ameloblasts and the dental papilla cells
into odontoblasts. The first tooth matrix is produced by the
odontoblasts at stage 35 (12 dpf). It consists of enameloid,
a dentine-like matrix composed of thin collagen fibrils
(20 nm in diameter) (see Section IV.1, and Fig.5). The
enameloid is secondarily modified by the activity of the
facing ameloblasts, which deposit a thin layer of enamel
matrix on the outer surface of the enameloid. The
ameloblasts next participate in the maturation process of
both the enameloid and enamel matrices, resulting in the
presence of a highly mineralised tooth cap. The organic
matrix of the mineralised cap is entirely removed during the
maturation process. This is clearly revealed by decalcifica-
tion which leaves an empty space between the dentine and
the ameloblast surface (see Fig. 5C). Predentine matrix is
next deposited by the odontoblasts (Figs. 8B, 9B).All
features indicate that these cells are the same as those that
previously deposited the enameloid matrix, suggesting
a switch in the functioning of odontoblasts from enameloid
to dentine matrix production. The first deposited (imma-
ture) collagen fibrils of the predentine measure between
6 and 12 nm in diameter. Their diameter then increases to
30 nm (mature fibrils) reaching 60-80 nm prior to minerali-
sation. All the odontoblasts located along the dentine shaft
possess long cytoplasmic processes, which penetrate the
predentine matrix. The dentine shaft elongates towards the
surface of the developing bone support, to which the tooth
base will eventually fuse (Figs 8C-E, 9D-F). This primary
type of tooth attachment in first-generation teeth differs
from the secondary type of attachment of replacement
teeth, in which the tooth base attaches to a pre-existing
bone surface as indicated by the presence of a cementing
line. The tooth pierces the oral epithelium and becomes
functional. In the developing first-generation teeth, the pulp
cavity is entirely occupied by odontoblasts. Most of them
degenerate when the tooth becomes functional, while some
remain active along the dentine surface. The pulp cavity is
secondarily penetrated by a capillary blood vessel through
a large pore located lingually at the interface between the
dentine shaft and the bone surface.
The development of subsequent generations of teeth is
similar to that of first-generation teeth, except for those
features described in Section IV: enamel progressively
covers enameloid which disappears at metamorphosis
(Fig. 5), there is a dividing zone separating the dentine
shaft into a crown and a pedicel (Fig. 6), and cementum is
deposited at the outer surface of the pedicel base (Fig. 7).
Another difference concerns the initiation process of the
replacement teeth; this is discussed in Section V.3.
(a)Caudata
Most caudates are oviparous, with rare exceptions such as
some salamanders of the family Salamandridae (e.g. genera
Salamandra,Lyciasalamandra,Mertensiella) which are live-
bearing (viviparous). Some species are neotenic (e.g.
Ambystoma mexicanum, Necturus maculosus, some Triturus species,
and some plethodontids such as some Gyrinophilus and
Eurycea species). These do not metamorphose, and conserve
most larval features during their life, but they can reproduce.
Tooth development was described, in greater or lesser detail,
in Triturus alpestris (Laurenti, 1768) by Wagner (1954),
N. maculosus by Kerr (1960), Pleurodeles waltl by Chibon
(1966, 1967, 1970), T. vulgaris (Linnaeus, 1758) by Smith &
Miles (1971), and A. mexicanum by Smith & Miles (1971) and
by Wistuba et al. (2002). Most of our knowledge of tooth
development in caudates comes from studies at the light
microscope and TEM level in larvae and adults of, mainly,
P. waltl and A. mexicanum. These studies have focused either
on general features of tooth development (Wistuba et al.,
2002), or on various aspects of odontogenesis such as the
Tiphaine Davit-Be
´al and others60
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

Fig. 8. Schematic representation of the development of a tooth and its successor in a generalised lissamphibian. Anterior is to the
left. (A) Morphogenesis and early cytodifferentiation. Originating from the basal epidermal layer of the oral epithelium, the primary
dental lamina extends into the subjacent mesenchyme. The distal region of the dental lamina interacts with mesenchymal cells and
forms a cup. The epithelial cells differentiate into an enamel organ, which further differentiates into an inner and an outer dental
epithelium. (B) Late cytodifferentiation. Mesenchymal cells have differentiated into odontoblasts, which deposit an unmineralised
matrix, predentine. The latter mineralises to become dentine. Facing the latter the inner dental epithelium cells differentiate into
preameloblasts. (C) The preameloblasts differentiate into ameloblasts and deposit the enamel matrix on the dentine surface. The
dentine cone elongates due to the activity of the odontoblasts and the pulp cavity starts to form. A secondary dental lamina,
originating from the upper region of the outer dental epithelium of the enamel organ at the posterior side of the tooth, extends into
the mesenchyme. (D) The tooth has elongated and its tip is close to the oral epithelium. The pedicel has started to form at the base
of the crown. The pedicel is separated from the dentine cone by an unmineralised region, the dividing zone. The secondary dental
lamina has extended deeply into the mesenchyme. (E) The tooth has attached to the supporting bone through its pedicel and its tip
has pierced the oral epithelium. The tooth is now functional and its replacement tooth has started to form. Note that in caudate
larvae the development of the first-generation tooth differs in that enameloid is the first matrix deposited by the odontoblasts, before
dentine. Modified from Kerr (1960) and Casey & Lawson (1981). am: ameloblast; de: dentine; dl: dental lamina; dz: dividing zone;
en: enamel; ide: inner dental epithelium; od: odontoblast; ode: outer dental epithelium; oe: oral epithelium; pc: pulp cavity; pde:
predentine; pe: pedicel; rt: replacement tooth; sb: supporting bone.
Amphibian tooth morphology and development 61
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

Fig. 9. Development and attachment of the first-generation teeth in Pleurodeles waltl larvae, from stage 36 to stage 55. (A) Initiation,
stage 36. The cells of the basal oral epithelium have differentiated into a dental organ. Facing them, some mesenchymal cells have
formed a small dental papilla: this is the bud stage. (B) Early cytodifferentiation, stage 36. The ameloblasts, i.e. the inner dental
epithelium cells, and the odontoblasts, i.e. the dental papilla cells, have differentiated. Tooth matrix has begun to be deposited by
the odontoblasts. (C) Late cytodifferentiation, stage 36. The crown has elongated and the enamel organ forms a typical bell shape.
Tooth matrix has started to mineralise, while predentine is deposited. (D) Tooth growth, stage 36. The pedicel has started to form as
a prolongation of the dentine shaft towards the surface of the supporting bone. (E) Stage 40. The tooth base is anchored to the
dentary bone by means of attachment bone. A blood vessel has penetrated the pulp cavity and the odontoblasts have slowed down
their activity. (F) Stage 55. Tooth recently attached to the dentary bone. The dentine crown, the pedicel and the attachment bone
are clearly visible, and the dividing zone is distinct. Scale bars: A-E ¼10 mm; F ¼50 mm. ab: attachment bone; am: ameloblast;
bv: blood vessel; db: dentary bone; de: dentine; dp: dental papilla; dz: dividing zone; eo: enamel organ; ide: inner dental epithelium;
ob: osteoblast; od: odontoblast; oe: oral epithelium; pc: pulp cavity; pe: pedicel.
Tiphaine Davit-Be
´al and others62
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

differentiation of the dental epithelium (Smith & Miles,
1971), dentinogenesis (Roux, 1973), amelogenesis (Roux &
Chibon, 1973), or on the characteristics of a particular
region such as the enameloid or enamel cap (Kogaya et al.,
1992; Kogaya, 1999), the tooth base (Moury et al., 1987), and
the dividing zone (Moury et al., 1985; Greven & Clemen,
1990; Wistuba et al., 2002).
The comparative analysis of this large amount of data
leads to the conclusion that (i) tooth morphogenesis and
differentiation in caudates are similar to that described in
mammals, and (ii) in all species examined larvae and adults
share similar features, with only a few differences, which
relate mainly to tooth size and to the presence of enameloid
in larvae versus enamel only in adults (see Section IV.1).
In larvae, teeth are attached to the paired bones of the
upper jaw (premaxillaries, maxillaries, prevomers and
palatines) and the lower jaw (dentaries and coronoids)
(Signoret, 1960) (Fig. 10). These bones form between stage
37 (Gallien & Durocher, 1957), i.e. 15 days post-fertilisation
(dpf) and stage 55 (90 dpf). The last bones to be formed are
the maxillaries, which ossify shortly before metamorphosis
(see Section V.2 for comments on the relationships between
teeth and bone supports). During metamorphosis, the
palatines disappear from the upper jaw and are replaced
by the extension of the vomers. In the lower jaw, the
coronoids disappear and only the dentaries remain (Corsin,
1966; Reilly, 1986) (Fig. 10).
The first-generation teeth start to form in embryos from
stage 33a (initiation, 9 dpf specimens). The first matrix is
deposited at stage 35 (12 dpf). The teeth grow, attach to the
bone support, pierce the buccal epithelium and become
functional when the mouth opens, at hatching (stage 37, 15
dpf). At this stage, there are on average 23 teeth on the
upper jaw: a row of eight teeth on the premaxillaries and
two rows of seven and eight teeth on the vomers and the
palatines, respectively. A row of 25 mandibular teeth is
present on the lower jaw, supported by the dentaries and
coronoids (Roux & Chibon, 1973). The teeth on the
dentaries face the premaxillary teeth, while those on the
coronoids face the vomerine and palatine teeth. Two or
three tooth generations succeed the first during larval life
and the number of tooth positions increases in each row.
Although new teeth are added at each position, tooth
resorption starts at stage 48 (50 dpf) only, suggesting the
retention of previous-generation teeth at a particular locus.
This results in the presence of two rows, with the teeth of
Fig. 10. Tooth location and bone changes in the oral cavity of Pleurodeles waltl during ontogeny. (A, B) Larva, lower and upper jaws,
respectively; (C, D) adult, lower and upper jaws, respectively. A, B are modified from Signoret (1960).
Amphibian tooth morphology and development 63
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

the second row considered replacement teeth (Roux &
Chibon, 1973). The teeth in larvae are conical and
monocuspid. The first-generation teeth are 100-150 mm
tall. The height increases in replacement teeth to reach
500 mm after metamorphosis, during which bicuspid teeth
replace monocuspid teeth (see Section VI.1). In direct
developing species (e.g. many plethodontids) bicuspid teeth
are also found in some prehatching ‘larvae’ (Ehmcke &
Clemen, 2000a).
In P. waltl, the rate of growth of larval teeth was
calculated using tritiated proline labelling (Chibon, 1977).
Five days are needed in young larvae to form a tooth, eight
in old larvae and 16 in post-metamorphosed specimens.
These experiments also indicated that some phases of
odontogenesis (initiation, morphogenesis, early differentia-
tion) proceed slowly while others are rapid (late cytodiffer-
entiation, growth and eruption).
(b)Gymnophiona
Caecilians possess numerous pedicellate teeth on the lower
and upper jaw, which are usually arrayed in two rows. The
dentition is generally homodont but exceptions exist, for
instance in foetuses (Wake, 1980) and in adults of some
species which have different degrees of bicuspidality on the
upper jaw and monocuspid teeth on the lower jaw (e.g.
Gegeneophis ramaswamii Taylor, 1964 (Greven, 1984). The
tooth structure is known at the light microscopic level
(Wake, 1976; Clemen & Opolka, 1990) and is similar to that
described in caudates. Tooth morphology differs depending
on whether the species are viviparous or oviparous, and on
developmental stage (Fig. 3).
In viviparous species, tooth development has been
studied in Dermophis mexicanus (Dume
´ril & Bibron, 1841)
by Wake (1976, 1980), and described from a single stage in
Geotrypetes seraphini (Dume
´ril, 1859) by Parker (1956), Parker
& Dunn (1964) and Wake (1976), and in Gymnopis multiplicata
Peters, 1874 and Typhlonectes compressicauda (Dume
´ril &
Bibron, 1841) by Wake (1976) and Hraoui-Bloquet &
Exbrayat (1996). The growth of embryos continues in utero
after the egg yolk has been exhausted; the foetuses develop
through metamorphosis in the oviducts and possess
particular teeth called foetal teeth. For Parker & Dunn
(1964) foetal teeth are functionless (a relictual retention of
a fish-like character). By contrast, Wake (1976, 1980, 1993)
considers they function to aid ingestion of intra-oviductal
nutrient material and to scrape the oviduct wall to stimulate
secretion during gestation. Indeed, the highly specialised
(spatula-like) shape of the foetal teeth strongly suggests not
only that they serve a purpose in food uptake, but that they
are specialised for scraping. This contrasts with the
condition found in other vertebrate larvae in which the
first-generation teeth are invariably conical and monocus-
pid, and considered an ancestral state for vertebrates (Sire
et al., 2002). In gymnophione embryos, tooth buds appear
first on the lower jaw at an early developmental stage, when
the bone support is not yet formed (Wake, 1976). Teeth
appear earlier in G. multiplicata than in T. compressicauda
(Wake, 1976). In G. multiplicata foetuses the teeth possess two
major cusps (a primary labial and a secondary lingual) and
several spike-like, minor cusps. The teeth are pedicellate
and only the top of the crown pierces the buccal epithelium.
In T. compressicauda, the foetal teeth first have two cusps, then
other cusps appear during ontogeny, indicating that
replacement has occurred during foetal life. The foetal
teeth are arranged in several rows (in contrast to a single
row in adults). This is explained by the retention of the
replacement teeth at every position, as in caudate larvae. In
foetuses, the height of the crown varies from 140 to 300 mm
depending on species (and, probably, on generation time)
and the pedicel is short (one third of the crown height) and
fused to the bone support. The foetal teeth are entirely
resorbed (or shed) at birth and replaced by bicuspid or
monocuspid teeth such as found in adults (Wake & Wurst,
1979). This sudden transition is presumed to be a response
to hormonal induction (Wake, 1993).
In oviparous species, tooth development has been studied in
Hypogeophis rostratus by Marcus (1920), Reuther (1931) and
Lawson (1965a,b)andinIchthyophis glutinosus by Clemen &
Opolka (1990). In the latter species, the dental lamina of
embyros is different from that in adults, which suggests that the
dental lamina divides later in ontogeny (Clemen & Opolka,
1990). In these species, the embryonic teeth are arranged in
a single row on the dentigerous bones and are monocuspid
(Clemen & Opolka, 1990), while they are bicuspid and
pedicellate in larvae and juveniles (Parker & Dunn, 1964).
In both viviparous and oviparous species, the odontoge-
netic processes are broadly similar to those described above
(Lawson, 1965a,b; Wake, 1976) (Figs 8, 9).InHypogeophis
rostratus, a thin cone of predentine is laid down by
odontoblasts and major and minor cusps are formed.
Mineralisation starts in the predentine when the first cusp is
formed. A thin layer of enamel is deposited by the
ameloblasts over the cusps (Lawson, 1965a). It seems that
the enamel matrix mineralises rapidly after its deposition
because no pre-enamel matrix was found in undecalcified
specimens. Enamel is thicker over the cusps than elsewhere.
Early during its development the tooth germ lies more or
less at right angles to the position occupied by the functional
tooth, and is attached to the buccal epithelium by the dental
lamina. Then it rotates approximately 90°and the dentine
cone elongates. A second mass of dentine is produced at
a short distance from the base of the dentine cone and
eventually forms the pedicel (Lawson, 1965a). The
calcification of the pedicel starts when it has reached the
base of the crown. The lingual side of the pedicel develops
more quickly than the labial side, leading to a pleurodont
type of attachment by ankylosis to the jawbone. During the
functional life of the tooth the crown dentine thickens and
the pulp cavity is reduced.
As in caudates, some phases of odontogenesis in
gymnophiones take place slowly while others are rapid.
(c)Anura (excluding Pipidae)
Bufonids (toads) do not possess teeth. In pipids and other
frogs teeth are restricted to the upper jaw, with the ex-
ception of some hylids, in which teeth are also found on the
dentary (Goin & Hester, 1961). On the upper jaw, the teeth
are arranged in a single row on the paired premaxillaries,
Tiphaine Davit-Be
´al and others64
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

the paired maxillaries and the paired vomers. In some
species the dentary is ornamented by odontoid (tooth-like)
structures, which are bony elements (Trueb, 1993). Teeth
are small (less than 1 mm long in adult Rana pipiens), but
very numerous, and their number and size varies with frog
size. For instance, Gillette (1955) counted 89-95 functional
teeth on each half of the upper jaw in large adult R. pipiens.
Each tooth position is occupied by a functional and
a replacement tooth, giving such a specimen on average
368 teeth in its dentition (small specimens have approxi-
mately 140 teeth). Teeth are bicuspid and the dentition is
homodont.
Anurans are mostly oviparous, but several species are
ovoviviparous and others are viviparous (Duellman &
Trueb, 1986; McDiarmid & Altig, 1999). Their aquatic
larvae, the tadpoles, are mostly herbivorous and detritivo-
rous, and their dentition is not composed of true,
mineralised teeth. Instead, there are horny labial teeth in
the upper and lower beak, which are formed by keratinised
epithelial cells organised into columns (Kaung, 1975;
Takahama et al., 1987). The cell located at the upper
extremity of the column forms the top of the tooth. The
number of teeth and the size of the beak increase during
larval life. At metamorphosis, the keratinized epithelial cells
are destroyed by autolysis, a process similar to that observed
in the tail.
Hertwig (1874) was first to describe tooth development in
frogs. Subsequent detailed accounts mainly confirmed his
findings. To date, descriptions of tooth development are
available for: Rana pipiens by Gillette (1955), Hyla cinerea
(Schneider, 1799) by Goin & Hester (1961), R. temporaria
Linnaeus, 1758 and R. esculenta Linnaeus, 1758 by Spinelli
& Chibon (1973) and by Chibon (1977), and H. arborea
(Linnaeus, 1758) and R. nigromaculata Hallowell, 1861 by
Sato et al. (1986a,b).
The first tooth germs appear at metamorphosis, by the
end of hindlimb organogenesis, and the first teeth are
functional after 25-26 days of growth. The developmental
features are similar to those described above for a general-
ised lissamphibian. In R. pipiens, there are six separate dental
laminae corresponding to the six dentigerous bones, and
extending lingually to the dental process of each bone
(Gillette, 1955). Predentine first calcifies in the lingual and
then in the labial cusp. Concomitant with the first
calcification of the dentine the ameloblasts differentiate,
then deposit enamel matrix first on the labial sides of the
cusps. The enamel mineralises rapidly, but slower than in
caudates (Spinelli & Chibon, 1973). During pedicel
formation the tooth progressively changes its orientation.
The calcifying process stops abruptly at the crown-pedicel
junction. Ankylosis of the pedicel to the bone is ensured by
cellular cementum, which completely bridges the gap
between the pedicel and the bone surface. Eruption is
arrested when the cementum matrix calcifies. During the
functional period more dentine is added to the crown and
to the pedicel on the pulp side. Cellular cementum is
deposited on the outer surface of the pedicel, when the
epithelial root sheath (Hertwig’s sheath) has retreated.
In Hyla cinerea teeth develop similarly to those of Rana
pipiens with only a slight difference in the mode of
attachment, which is related to the form of the dental
process of the maxillary (Goin & Hester, 1961).
(d)Pipidae
The Pipidae are a well-known anuran family thanks to
Xenopus laevis, which is used as a model in developmental
biology. X. laevis may live as long as 23 years in the laboratory
(Deuchar, 1975). In their Developmental Table for X. laevis
Nieuwkoop & Faber (1956) briefly commented on the
development of the dentition. Cambray (1976) provided the
first description of tooth development in larvae and adults,
but his doctoral thesis remained unpublished. Subsequently,
Shaw (1979, 1985, 1986, 1989) published reference data on
tooth development and replacement in X. laevis.
X. laevis differs from the other anuran taxa in that true
teeth begin to develop on the upper jaw in tadpoles during
the last stages of larval life (Fig. 11). However, teeth do not
erupt until the end of metamorphosis; they are 225-250 mm
tall (Shaw, 1979). In larvae, newly metamorphosed and adult
specimens the teeth are morphologically similar, only
differing in size. The dentition is homodont and the teeth
form a single row on the upper jaw. The teeth are conical,
monocuspid and slightly curved at the tip. In adults, only 50-
100 mm of the tooth tip projects into the mouth so that it is
hard to ascribe an important function to them (Shaw, 1979).
Tooth structure is known at the light microscopic level
only (Shaw, 1979). Unlike in many other lissamphibians, the
enamel organ is composed of three layers, a stellate retic-
ulum being present between the inner and the outer dental
epithelium. In addition, teeth are not pedicellate (there is no
dividing zone) and there is no indication of a layer of ce-
mentum along their base (Katow, 1979). They are ankylosed
to the premaxillaries and maxillaries by a short, ring-shaped
bony pedicel (bone of attachment), which is continuous with
the jaw bone in newly metamorphosed specimens but
distinct from the underlying bone in the adults. All these
features are typically those of first-generation teeth in
nonmammalian vertebrates (Sire et al., 2002).
The first-generation teeth develop at stage 55, approx-
imately 40 days before metamorphosis (Shaw, 1979)
(Fig. 11). Teeth appear in alternating (even and odd)
positions, starting from the mid-line, to reach 22 positions
on each side of the jaw at metamorphosis: eight on the
premaxillary and 14 on the maxillary. Teeth at even
positions develop first, 8-9 days before those at odd
positions. From the second day after germ initiation,
dentine deposition commences, followed shortly by enamel
formation. We have not been able to identify enameloid
unequivocally during early tooth development in Xenopus
laevis (Fig. 11 and H. Chisaka, unpublished results), but this
needs to be investigated further. Enamel formation and
mineralisation encompass a fairly short period (eight days).
Dentinogenesis is relatively slow during the first 20 days,
then the tooth germs at even positions start a period of
rapid dentinogenesis, re-orientate to a more vertical
position, and develop a dental lamina for their successor
tooth. Attachment lasts from day 23 to 25. From day 27
after first tooth germ initiation, the tooth germs at odd
positions begin their period of rapid growth. Osteoclastic
Amphibian tooth morphology and development 65
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

resorption of the first even-numbered teeth begins at 32-33
days (i.e., 4-6 days post stage 65) (Shaw, 1979). The second
generation teeth erupt five days later, i.e. slightly before the
end of metamorphosis. In summary, in larvae a complete
tooth cycle takes 33 days, with teeth being functional for
only seven days, versus 59-71 days for the complete cycle
and 24-29 days functional in adults.
(2) Relationships between tooth and bone
support development
In nonmammalian species studied so far, the development of
a first-generation tooth ends with the anchoring of the tooth
base (the pedicel or attachment bone) to a bony support (pre-
maxillaries, maxillaries, dentaries, vomers, palatines, pha-
ryngeal bones, etc.). However, the matrix of the supporting
bone is not present when the tooth is initiated (Sire et al.,
2002). Osteogenesis and odontogenesis progress approxi-
mately simultaneously. This is achieved such that, in the
dentigerous region, both the bone and the tooth matrix seem
to converge towards each other. Eventually, both matrices
(bone surface and base of the pedicel) merge, forming the so-
called primary tooth attachment (versus secondary tooth
attachment, which occurs when the bone support is already
present when the tooth attaches). Such a process suggests
the existence of coordination between the odontoblasts at
the base of the pedicel and the osteoblasts at the bone surface
facing the developing tooth (mediated by signalling mole-
cules?), at least during the final stages of development of
the two elements. Further (molecular) studies will be
necessary to understand the interactions between these cell
populations. Although these observations suggest that teeth
need the presence (or concomitant development) of a bony
support in order to develop, experimental studies do not
support this conclusion. With the aim of perturbing
organogenesis in embryonic P. waltl, Signoret (1960) applied
various concentrations of lithium chloride, a molecule known
to induce morphogenetic perturbations. He found organ
reduction (hypomorphy) in these embryos. In addition, tooth
development was found to be very sensitive to lithium
chloride: some bones which are normally dentigerous
(dentaries, premaxillaries, palatines), developed normally,
but without teeth. In addition, (i) teeth were found in regions
devoid of bone support, and (ii) some bones which were
normally not dentigerous (the angular and the parasphenoid)
were found to bear teeth. These experiments demonstrated
clearly that tooth development does not depend on the nature
and location of the bone support. The relationship between
a tooth and its surrounding bone may therefore be secondary,
resulting from topographic conditions only.
In the frog Rana pipiens, Howes (1977) transplanted teeth
during the crown formation phase to an ectopic site (either in
the anterior eye chamber or in a dorsal subcutaneous site).
these transplanted teeth grew normally and formed a nor-
mal-sized pedicel area demonstrating that (i) once initiated
the genetic programme is able to support complete odonto-
genesis, and (ii) the pedicel is purely odontogenetic in origin.
Similarly, the ablation of part of the premaxilla in this species
Fig. 11. Tooth development in an anuran, the pipid Xenopus laevis. In contrast to most anuran species, the teeth develop long before
metamorphosis. (A, B) Tadpole, stage 59, ventral and lateral views, respectively. The first-generation teeth are already well developed
at this stage. (C) Tooth pattern in a tadpole, stage 59. As in most anuran species teeth are only present on the upper jaw. (D-G) One
mm-thick vertical sections of the upper jaw of tadpoles (stages 54, 58, 63 and 65, respectively) showing various developmental stages
of first-generation teeth, from initiation (D) to attachment (G). The arrow in E indicates to the first deposition of the tooth matrix.
The arrows in F and G indicate the previous location of the enamel in these samples which were decalcified with EDTA.
Developmental stages are as in Nieuwkoop & Faber (1956). A, B modified from Nieuwkoop & Faber (1956), C from Cambray (1976).
Scale bars: C ¼250 mm; D ¼10 mm; E-G ¼50 mm. de: dentine; do: dental organ; dp: dental papilla; eo: enamel organ; ide: inner
dental epithelium; mb: maxillary bone; od: odontoblast; oe: oral epithelium; pc: pulp cavity; pm: premaxillary bone.
Tiphaine Davit-Be
´al and others66
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

did not prevent teeth from growing in the regenerated jaw
(see Section VII) in the absence of a bone support (Howes,
1978). These teeth developed to normal size and shape, and
were even replaced by their successors although the
supporting bone, which regenerated slowly, was still not
formed. Such results confirm that complete tooth develop-
ment does not require the presence of a bone support.
Observations made in the armoured catfish Corydoras spp.
and Hoplosternum littorale confirm the generality of these
findings. On the upper jaw, the first-generation teeth
develop long before the development of the maxillary. Each
tooth forms a pedicel of attachment bone, and the pedicels
of adjacent teeth merge to constitute a kind of dentigerous
bone, which will later be connected to the developing
maxillary (Huysseune & Sire, 1997b). Also, in some zebra-
fish (Danio rerio) mutants that do not form pharyngeal arches,
teeth develop in the absence of the pharyngeal bone support
(Schilling, Walker & Kimmel, 1996).
(3) Tooth replacement and resorption
The processes of tooth replacement and resorption occur in
a similar manner in anurans (including Xenopus laevis),
gymnophiones and caudates. Resorption of a functional
tooth is always related to the close presence of a growing
replacement tooth, as illustrated in Pleurodeles waltl (Fig. 12).
A replacement tooth is first seen lingually as a bud from
the region located at the limit between the dental organ and
the dental lamina of the functional tooth (Fig. 12). The bud
extends as a new dental lamina into the mesenchyme, with
which it interacts to give rise to a new tooth. The
replacement tooth grows and, once fairly well developed,
its enamel organ generally contacts the lingual side of the
functional tooth (Fig. 12C). This contact induces the
recruitment of osteoclasts as a probable reaction to pressure
forces acting on the external wall of the tooth. Most
generations of replacement teeth in lissamphibians exhibit
these general features, although the first generation of
replacement teeth does not provoke the resorption of the
first-generation teeth, resulting in their retention and hence
the presence of two tooth rows on the upper rand lower
jaws of young larvae. The first-generation teeth are very
small (20-30 mm wide) thus there is probably enough space
to accommodate two teeth from the same family. Such
a condition has also been described in the zebrafish (Van
der heyden & Huysseune, 2000).
In P. waltl, the first signs indicating imminent resorption of
the first-generation teeth are identified at larval stage 44,
long after the first-replacement teeth have been functional.
The pulp becomes more loosely organised due to a decrease
in cell number. Numerous cells are degenerative and some
macrophages are present (Roux & Chibon, 1974). The latter
Fig. 12. Tooth replacement in Pleurodeles waltl. (A) Larva, stage 56. A secondary dental lamina, originating from the upper region (*)
of the outer dental epithelium of the previous tooth, has extended into the mesenchyme. This dental lamina is composed of two
layers, the cells of which are differently arranged: flat and elongated at the posterior side and tall and polarized at the anterior side.
(B) Larva, stage 56. The cells located at the anterior side and at the extremity of the dental lamina have proliferated and have
formed a cup, which surrounds mesenchymal cells (arrow). The asterisk indicates the origin of the secondary dental lamina.
(C) Larva, stage 49. A replacement tooth is well formed, but still attached to the functional first-generation tooth by means of
the secondary dental lamina (*). Scale bars: A ¼10 mm; B, C ¼20 mm. db: dentary bone; de: dentine; dl: dental lamina; ide:
inner dental epithelium; ode: outer dental epithelium; oe: oral epithelium; pc: pulp cavity; pe: pedicel.
Amphibian tooth morphology and development 67
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

have secondarily invaded the pulp cavity and are involved in
phagocytosis of necrotic cells. Macrophages could also be
responsible for the destruction of Hertwig’s sheath and the
basal region of the dental lamina (Chibon, 1977). The first
resorption features are observed at stage 48 (50 dpf),
approximately two months before metamorphosis, when
two tooth generations have formed. The first tooth is
completely resorbed at stage 52 (64 dpf). The cells
responsible for tooth resorption are typically multinucleated
osteoclasts. There are also some mononucleated macro-
phages removing cell debris. First, the external surface of the
dentine wall at the base of the pedicel is attacked by
osteoclasts at the lingual side (Fig. 13A, E). Then, osteoclasts
(also called odontoclasts) penetrate the pulp cavity and start
to resorb the opposite wall, while the resorption extends
labially and to the top of the tooth shaft (Fig. 13B, C, F).
Simultaneously, cell necrosis is observed in the pulp cavity
and Hertwig’s sheath retracts. The dentine shaft is entirely
resorbed as well as part of the adjacent supporting bone.
The large, multinucleated cells responsible for the
resorption of lissamphibian teeth share similar features
with osteoclasts described in many vertebrate species
(Fig. 13). However, some authors have called them
odontoclasts with reference to their involvement in dentine
resorption (e.g. Clemen & Greven, 1974, 1977; Bouvet &
Chibon, 1976; Chibon & Bouvet, 1976; Wistuba, Bolte &
Clemen, 2000). The first TEM description of these cells was
for frogs (Yaeger & Kraucunas, 1969). Wistuba et al. (2000),
working on Ambystoma mexicanum, provided additional
details. Interestingly, activity of tartrate-resistant acid
phosphatase (TRAP) has not been detected during the first
stages of tooth resorption, whereas it was shown to reveal
osteoclastic activity in other vertebrates, such as teleosts
(Witten, 1997) and mammals (Sahara et al., 1998). The
presence of clastic cells in A. mexicanum, a neotenic caudate,
which lacks parathyroids, indicates that the production of
parathyroid hormone (PTH) is not a prerequisite for the
regulation of these cells. Instead, regulation through
a pituitary factor has been suggested (Pang et al., 1980).
In Pleurodeles waltl tooth replacement has been shown to be
under the influence of the thyroid hormone, thyroxine.
When this hormone is absent (or inhibited), or at low
concentration, teeth are replaced more slowly than in nor mal
specimens (Dournon & Chibon, 1974). This is probably
related to the role played by thyroxine in cell proliferation.
The fate of the tooth tip during tooth resorption has
given rise to some controversy. Most authors either state, or
assume, that lissamphibian teeth become loose at their bases
and are shed into the mouth (e.g. Gillette, 1955; Lawson,
1965b; Lawson, Wake & Beck, 1971; Clemen & Greven,
1980). For instance, it is supposed that the weak,
unmineralised dividing zone is destroyed more rapidly than
the pedicel, provoking shedding of the crown, and explain-
ing why most crowns are absent in teeth undergoing
resorption, while pedicels remain, at least partially (Casey &
Lawson, 1981). However, in their illustrations there is no
histological section showing the absence of crown resorp-
tion. Contrary to this view, Shaw (1986, 1989) found it
logical that the major part of a tooth in Xenopus laevis is
absorbed rather than lost by shedding. Chibon (1977)
reached a similar conclusion with P. waltl larvae where the
tips of teeth in an advanced state of resorption were found
entirely embedded in the oral epithelium. We confirm that
the teeth in P. waltl are entirely resorbed, at least in larvae
(Fig. 13H, I). However, in lissamphibians osteoclastic
resorption of the enamel cap has not been reported to date
in the literature, in contrast to some mammals, in which
enamel resorption has been reported (Sahara et al., 1998).
In adult caecilians [e.g. Hypogeophis rostratus (Cuvier, 1829),
Grandisonia diminutiva Taylor, 1968] tooth replacement
occurs as described in P. waltl. The process of resorption is
supposed to be rapid because partially resorbed teeth are
uncommon (Casey & Lawson, 1981). The pedicel is entirely
removed by resorption. However, it is not clear how much
of the crown is resorbed before shedding. The loss of a large
number of crowns may represent a considerable long-term
drain on calcium reserves. Indeed, in addition to its function
in permitting tooth replacement to occur, resorption is
considered a conservative process making tooth constituents
(minerals and organic matrix) available for re-use.
In the anuran Rana pipiens, Gillette (1955) found that the
tooth resorption process is similar to that described in
Pleurodeles waltl.InHemiphractus proboscideus (Jime
´nez de la
Espada, 1871), Shaw (1983) calculated that the volume of
Fig. 13. Tooth resorption in Pleurodeles waltl. (A-D) Scanning electron micrographs of teeth subjected to resorption, viewed from the
lingual side. Note the numerous, well-delimited lacunae at the resorption sites, revealing the location of the osteoclasts. (A) Adult.
Resorption has started at the level of the pedicel. (B) Adult. Resorption has extended to the whole surface of the pedicel. (C) Ten-
month-old specimen. The pedicel surface is highly resorbed as well as the base of the crown, where the pulp cavity has been
opened. (D) Adult. The tooth has been entirely resorbed, but most of the pedicel remains. (E-I) One mm-thick, vertical sections of
teeth subjected to resorption. (E) Larva, stage 51. The surface of the pedicel located close to the enamel organ of the replacement
tooth is subjected to resorption. (F) 12-month-old specimen. Most of the pedicel has been resorbed and two large, multinucleated
osteoclasts are attacking the base of the dentine crown (arrows). (G) Eight-month-old specimen. An osteoclast has penetrated the
pulp cavity and a large part of the dentine crown is resorbed. Note the decalcification of the dentine matrix prior to resorption.
Another osteoclast is apposed onto the surface of the dentary bone. The cells of the enamel organ of the resorbed tooth have not
retracted (arrow), while the dentine crown they were covering has been resorbed. (H) Larva, stage 48. The resorption of this first-
generation tooth is well advanced. Note that a single osteoclast is involved in the resorption of the dentine cone and of the
attachment bone, simultaneously. (I) Larva, stage 52. This first-generation tooth has been resorbed, but its tooth tip is still visible
(arrow), entirely surrounded by an osteoclast. Scale bars: A, B, C ¼100 mm; D, H ¼20 mm; E, I ¼10 mm; F, G ¼50 mm. ab:
attachment bone; bv: blood vessel; db: dentary bone; de: dentine; eo: enamel organ; oc: osteoclast; oe: oral epithelium; pc: pulp
cavity; pe: pedicel; rt: replacement tooth.
Tiphaine Davit-Be
´al and others68
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

the dental tissues (excluding enamel) represented 3.64% of
the volume of the supporting bone, and that 96.36% of the
volume of the dental tissues is resorbed (and thus re-used)
rather than shed. In Xenopus laevis, multinucleated osteo-
clasts are first observed at 4-6 days post-stage 65 (i.e. by the
end of metamorphosis), on the outer surface of the first-
generation teeth, along the dentine wall and along the bone
of attachment (Shaw, 1983; H. Chisaka, personal observa-
tions). Then, osteoclasts penetrate the pulp cavity, where
they start to break down the dentine, from base to tip. Most
first-generation teeth are lost (they are apparently entirely
resorbed, but the fate of the tooth tip remains uncertain) at
9-11 days post-stage 65. Therefore, the duration of the
resorption process is five days (the dentine being destroyed
Amphibian tooth morphology and development 69
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

in about two days), representing only 15% of the total
lifetime (33 days) of a first-generation tooth. Numerous
lymphocytes and macrophages are observed at the tooth
base during the resorption process. In adult X. laevis, the
dentition (upper jaw) is maintained through the resorption
and replacement of several hundred teeth per annum
(Shaw, 1985). Shaw (1989) reported that the mean volume
of the dentine of all teeth is about 23.5% of that of the
supporting bones (premaxillaries and maxillaries), and that
during tooth replacement the osteoclasts resorb up to 98%
of the dentine.
(4) Tooth replacement pattern
Intuitively, the maintenance of both the number and
sharpness of the teeth might be of great importance to
toothed lissamphibians, which catch moving live prey.
Indeed, if all the teeth were to be replaced simultaneously,
there would be phases of toothlessness, which would repre-
sent a serious disadvantage for the animal. Therefore, logi-
cally the pattern of replacement along the tooth rows ensures
that no toothless phases occur and that an efficient dentition
is always present. However, toads are toothless and they are
able to catch moving live prey. Data on tooth replacement
are available in so few species (see below) that they do not yet
permit conclusions for the entire class Lissamphibia.
The pattern of tooth replacement has attracted attention
for well over a century, but its complexity (i.e. how the
developmental cycles of the many individual teeth are
interwoven to produce a continuously functioning dentition)
defied understanding until the studies of Edmund (1960,
1962, 1969) in reptiles pointed out underlying principles (the
Zahnreihen model). Edmund postulated that waves of stimuli
pass along the jaw from front to back at regular intervals
and initiate tooth replacement in alternate tooth positions.
This model was found to apply to most common tooth
replacement patterns in reptiles, but increasing numbers of
examples exist (notably in teleost fish and amphibians) where
such replacement patterns cannot be identified (see review in
Berkovitz, 2000), and evidence for the Zahnreihen model, at
least for tooth replacement, is becoming less convincing
(Smith, 2003; Huysseune & Witten, 2006).
Among the numerous studies devoted to lissamphibian
teeth, a number include data on the pattern of tooth
replacement (e.g. Lawson, 1965b; Lawson et al., 1971; Miller
& Rowe, 1973; Wake, 1976, 1980; Chibon, 1977; see also
Berkowitz, 2000). Some of these studies were undertaken with
the aim of discovering whether or not the regular patterns of
tooth replacement proposed by Edmund (1960, 1962, 1969)
exist in lissamphibians. In fact, the reptilian condition in
which functional teeth alternate with non-functional teeth
rarely applies to lissamphibians. The pattern of tooth
replacement in lissamphibians tends to be obscured by the
fact that, in general, each tooth locus possesses a mature,
functional tooth (less than 25% are non-functional) compared
to only 50% in reptiles. This means that the teeth are retained
functionally for a long period compared to the rapid phases of
development and resorption.
In caudates, Lawson et al. (1971) reported the pattern of
tooth replacement in Plethodon cinereus (Green, 1818) using
various techniques (Alizarin Red, radiographs, wax impres-
sions). Miller & Rowe (1973) discussed that of Necturus
maculosus, and Chibon (1977) that of Pleurodeles waltl.In
Plethodon cinereus the tooth replacement pattern is consistent
with the Zahnreihen theory, but with some variations. By
contrast, in N. maculosus the replacement pattern is very
irregular, varying seasonally and with respect to the bone
support, and does not follow the Zahnreihen model. During
the winter period of hibernation tooth replacement ceases
while it is most active in late summer and autumn. In P. waltl,
tooth replacement again does not seem to be regular either.
In gymnophiones, the pattern of tooth replacement was
studied in adult Hypogeophis rostratus by Lawson (1965b)andin
foetuses and adult Gymnopis multiplicata, Typhlonectes compressi-
cauda and Dermophis mexicanus by Wake (1976, 1980). By
dividing the life cycle of teeth into a number of histological
stages on the basis of morphological characters, Lawson
(1965b) found that some developmental phases take place
slowly and others more rapidly: the stages leading from tooth
initiation to the first deposit of enamel as well as the
resorption of the functional tooth are extremely rapid,
whereas phases of tooth growth and dentine calcification,
and of pedicel calcification and attachment are long. In the
two jaws, a functional and a replacement tooth occupy each
locus. In H. rostratus, the sequence of tooth replacement
occurs, similarly to that described in reptiles by Edmund
(1960): a comparison of the developmental stages in alternate
tooth loci showed (i) that tooth replacement occurs in waves,
which run craniad along the jaw and (ii) that a full
replacement wave occupies six tooth positions (Lawson,
1965b). In embryos and foetuses of G. multiplicata, T.
compressicauda and D. mexicanus new tooth loci are added both
posteriorly along the jaw and between existing loci (Wake,
1976, 1980). Using three developmental stages in foetal teeth
and by studying tooth replacement pattern on the lower and
upper jaws (four rows), Wake (1980) demonstrated that tooth
replacement proceeds alternately. During the transition from
foetal to adult dentition, along the jaw tooth replacement
along the jaw follows a pattern similar to that predicted by
the Zahnreihen model, as in most adult lissamphibians.
In anurans, the dynamics of continuous tooth succession
was studied in Rana pipiens by Gillette (1955), in Hyla cinerea
by Goin & Hester (1961), and in R. temporaria by Lawson
(1966) and by Chibon (1977). Using Alizarin Red staining of
tooth-bearing bones, Gillette (1955) was able to divide the
developmental cycle of individual teeth in R. pipiens into
eight equal time periods. A correlation was found between
the developmental stage of (i) a given tooth and that of its
successor, (ii) a given tooth and that of adjacent teeth, and
(iii) teeth in alternate positions. The existence of this regular
pattern indicates that the initiation of new teeth is not
a random process. This is in conflict with the idea that tooth
replacement occurs only when its functioning predecessor is
lost or strenuously used (Hertwig, 1874). In fact the
initiation of a new germ is due to a condition intrinsic to
the dental lamina, and its development causes the
resorption of the predecessor, clearing the way for its
eruption. Gillette (1955) calculated that a tooth predecessor
is 45 days ahead of its successor and that the entire cycle of
a tooth takes 90 days, but this can vary with temperature,
Tiphaine Davit-Be
´al and others70
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

season and age (size) of the frogs. One can question whether
indefinite tooth succession is possible and the discovery of
edentulous individuals in some frog species could be related
to old age (Smith, 1883). However, working on several
anuran species, and in particular H. cinerea, Goin & Hester
(1961) could not define a regular pattern of replacement,
even though waves of tooth replacement seem to exist in
alternate tooth positions. They concluded that tooth
succession is a haphazard process, which cannot be
explained by Edmund’s (1960) theory, but they probably
misinterpreted the Zahnreihen concept of waves stimulating
tooth replacement in alternate tooth positions. Indeed,
Lawson (1966) showed in R. temporaria that teeth in alternate
loci are usually at the same stage of development, suggesting
that tooth replacement involves teeth in alternate loci.
In Xenopus laevis, two or three new positions are added
posteriorly to the row during metamorphosis. New positions
are subsequently added in post-metamorphosed specimens
and in adults to reach 30 positions on each side. The
timescale of tooth replacement has been studied for first-
generation teeth in newly metamorphosed X. laevis (Shaw,
1979) and in adults using impressions of the upper jaws on
dental gold-casting wax (Shaw, 1985). In adults, the tooth
replacement cycle was 38-42 days, functional life was 24-29
days and the gap period (between loss of a tooth and
attachment of its successor) was 9-17 days. The dentition
develops in alternating series, the teeth at even-numbered
loci developing before the odd-numbered loci.
VI. TOOTH CHANGES
(1) Monocuspid to bicuspid: the role of
thyroxine at metamorphosis
Most metamorphosed caudates have bicuspid teeth, while
teeth are monocuspid in larvae. It is known that
bicuspidality is established during or immediately after
metamorphosis, when the monocuspid larval teeth are
replaced (Kerr, 1960; Chibon, 1972; Clemen & Greven,
1974, 1977, 1979; Beneski & Larsen, 1989a). Therefore,
a clear relationship has been identified between tooth
morphology and the increase of thyroid activity during
metamorphosis. Two dental features which are substantially
modified during metamorphosis seem to be sensitive to
thyroxine: cusp morphology and enamel thickness. Both
cuspidality and enamel production are controlled by
ameloblasts in the dental epithelium. In the newt Triturus
helveticus (Razoumovski, 1789), Chibon (1972) and Gabrion
& Chibon (1972) showed that cusp modification and
enamel thickness are under the control of thyroid hormonal
secretion during the entire life. Indeed, in neotenic newts
(i.e. specimens retaining larval characters beyond larval life)
most teeth were of the larval type (monocuspid with a thin
enamel layer). Experimental hypophysectomy or thyroid
dysfunction in Pleurodeles waltl also resulted in the conser-
vation of larval characters, even in old specimens. In P. waltl,
enamel thickness depends on thyroid hormone levels:
increased levels resulted in thicker enamel (Roux & Chibon,
1973). In Ambystoma mexicanum, a neotenic salamander, the
teeth are partially transformed, i.e. bicuspid on the upper
jaw, monocuspid on the palatine and splenial, and of
a mixed type on the dentaries, indicating differences in
sensitivity of the dental laminae to metamorphic hormones
(Clemen & Greven, 1977; Greven & Clemen, 1990; Bolte &
Clemen, 1991, 1992). In general, paedomorphic species
possess a mosaic of larval and metamorphic tooth features,
which reflect their degree of paedomorphosis (Greven &
Clemen, 1980; Clemen & Greven, 1988). Experiments where
complete metamorphosis was induced led to fully trans-
formed teeth (Clemen, 1988; Greven & Clemen, 1990). The
cells of the dental organ could react differently (for example
due to the presence or absence of appropriate receptors)
because jaw teeth could be more sensitive to thyroxine than
palatine teeth (Clemen & Greven, 1977; Clemen, 1988; Mutz
& Clemen, 1992). Although many studies are currently
devoted to elucidating the genetic mechanisms of thyroid
hormone action through activation of its receptors during
lissamphibian development and metamorphosis (mostly in
Xenopus laevis), none are specifically related to tooth
development (but see Rose, 1999 and Section VIII below).
(2) Bicuspid to monocuspid: the role of
androgens
In some species of plethodontid salamanders, instead of the
typical bicuspid teeth the males bear monocuspid teeth on
the jaws during the mating season (Means, 1972; Ehmcke &
Clemen, 2000b). These teeth play an important role during
courtship (Duellman & Trueb, 1986). In Desmognathus fuscus
(Rafinesque, 1820),castration revealed that the appearance
of these monocuspid teeth is controlled by androgens
(Noble, 1926; Noble & Davis, 1928; Noble & Pope, 1929).
Recently, working on Bolitoglossa schizodactyla Wake & Brame,
1966, a plethodontid in which monocuspid teeth are
restricted to the premaxillary bone, Ehmcke et al. (2003)
showed that this was related to the restriction of androgen
receptors to the cells of the premaxillary dental lamina,
explaining the selective response of these cells to androgens.
VII. TOOTH REGENERATION
Most lissamphibians, and especially caudates, are able to
regenerate large parts of the body such as the tail, limbs, etc.
in larvae as well as in adults. This regeneration potential has
long been known but it is less well known that jaws also can
regenerate completely in larvae and adults (Goss & Stagg,
1958a,b; Goss, 1969). Lauga-Reyrel (1974) used this ability
to study ‘‘tooth regeneration’’ in Pleurodeles waltl: after partial
amputation the lower jaw regenerates and new teeth form. It
is noteworthy that the teeth that form on these regenerated
jaws start to develop in the buccal epithelium as soon as
wound healing is finished, before the new cartilage and,
therefore, long before the bone support appears. Tooth
regeneration does not require the presence of the regen-
erated supporting cartilage, which needs the formation of
a blastema. This shows that tooth regeneration depends only
on the regeneration of the basal layer of the buccal
Amphibian tooth morphology and development 71
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

epithelium, from which the regenerated dental lamina will
develop. However, this phenomenon is rather closer to
a tooth replacement process than to de novo tooth
ontogenesis, because of the presence of the regenerated
dental lamina. Working on jaw regeneration in the newt
Notophthalmus viridescens (Rafinesque, 1820), Graver (1973,
1974) reported that the dental lamina did not arise de novo
from the oral epithelium, but that it could only regenerate
from the residual lamina. Clemen (1979a) reached a similar
conclusion for the vomerine dental lamina in Salamandra
salamandra (Linnaeus, 1758). The complete amputation of
the dental lamina of the vomer or of the palatine gives rise to
toothless bones (Clemen, 1979b). Experiments dealing with
repeated regeneration of the jaw have shown that the events,
especially those involved in the repeated regeneration of the
dental lamina, are the same as in normal regeneration but
that the regrowth is more rapid (Graver, 1978).
In the frog Rana pipiens, Howes (1978) has shown that the
bones of the upper jaw regenerate slowly compared to the
jaw tissues in which teeth form normally.
VIII. DIRECTIONS FOR FUTURE RESEARCH
Here we identify 10 questions that remain unanswered on
various aspects of amphibian tooth biology, and we propose
research avenues that could try to answer them.
(1) Are pedicellate teeth homologous among
lissamphibians?
The presence of pedicellate teeth in most species of the three
modern lissamphibian orders could indicate that this is an
ancestral character although this can not be assessed from the
palaeontological data due to the scarcity of fossil temno-
spondyls and Paleozoic tetrapods. However, in some teleost
fish there is an uncalcified (or slowly mineralising) region,
similar to the dividing zone of lissamphibian teeth, separating
the dentine shaft from an elongated bone of attachment,
similar to a pedicel. The length of this ‘‘pedicel’’ varies from
short, as in armoured catfishes and cichlids, to long and
morphologically similar to that found in lissamphibians, as in
some osteoglossiforms (Huysseune & Sire, 1997a,b; Sire et al.,
2002). The pedicel of lissamphibian teeth and this pedicellate
structure in teleosts could have appeared independently in
both sarcopterygian and actinopterygian lineages rather than
being homologous, i.e. derived from pedicellate teeth in
a common ancestral osteichthyan. Indeed, in teleosts the
pedicellate bone of attachment is never covered by the dental
epithelium (enamel organ), in contrast to the pedicel of
lissamphibian teeth. Does the extent of the enamel organ
along the tooth shaft define the boundary between the tooth
proper (i.e. dentine shaft ]pedicel in lissamphibians) and the
attachment bone?
Comparative studies of tooth development in lissamphi-
bians and actinopterygians could reveal typical features that
would indicate whether the dividing zones in these lineages
are homologous or have been acquired independently.
(2) Are the dentition pattern and development
of the dental lamina important features for
lissamphibian systematics?
Several dental characters may represent useful tools to es-
tablish lissamphibian relationships, but it must be verified
that these characters are independent and not develop-
mentally correlated. Cusp morphology is unlikely to be of
use because examples of convergent evolution are already
known.
A comprehensive analysis of the large body of data
available on the dentition pattern and organisation of the
dental lamina in lissamphibians (see Section V.4) may prove
useful in resolving relationships within Lissamphibia.
(3) Do ameloblasts participate in enameloid
formation in lissamphibian larvae?
During the deposition of the enameloid matrix by
odontoblasts, the ameloblasts are differentiated and show
characteristic features, such as a polarized organisation,
thepresenceofaruffledborder,andtheabsenceofthe
basement membrane (Fig. 5). These features do not prove
that ameloblasts are involved in the deposition of enamel
proteins, but suggest that they are at least involved in
the maturation process of the enameloid (Smith & Miles,
1971).
It would be interesting to determine whether enamel
proteins are synthesized by ameloblasts prior to, during and
after enameloid deposition. This could be achieved using
in situ hybridisation techniques. Among the candidate pro-
teins, amelogenin may be the best choice as it is the major
enamel protein. The amelogenin gene is well known in
numerous mammals (Delgado, Girondot & Sire, 2005), in
three reptiles (Ishiyama et al., 1998; Toyosawa et al., 1998;
Delgado et al., 2006) and two lissamphibians (Toyosawa
et al., 1998; Wang et al., 2005). Amelogenin could be cloned
in a caudate such as P. waltl, and its expression studied in
early larvae during enameloid formation in first-generation
teeth. Using the same technique, the presence of protei-
nases (principally metalloproteinases MMP-20, enamelysin,
and KLK4, kallikrein 4) that are involved in enamel
maturation could be investigated. However, to date genes
coding for such proteinases are only known in mammals
(Bartlett et al., 1997; Fukae et al., 1998; Bartlett & Simmer,
1999).
In the anuran Xenopus laevis, dentine is described as
forming first in the first-generation teeth, but it is not clear
whether enameloid is present, even though it is known that
these teeth are subsequently covered by a layer of enamel
(Shaw, 1979). If enameloid is found, tooth development in
this species would be similar to that in caudates.
Observations of appropriate stages at the TEM level are
therefore required and, particularly, a more detailed study
of ameloblast functioning. In situ hybridization performed
on larvae developing first-generation teeth should reveal
whether or not ameloblasts are synthesizing amelogenin
when the odontoblasts are depositing dentine-like material
in this species.
Tiphaine Davit-Be
´al and others72
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

(4) How does the enameloid-enamel transition
proceed through caudate ontogeny?
The enameloid layer, which is present in the first-generation
teeth in caudate larvae (and covered secondarily by a thin
layer of enamel) is absent in the teeth of post-meta-
morphosed specimens, in which a thick enamel layer is
present.
More detailed study of tooth development in a growth
series of a salamander like P. waltl using light microscopy
and TEM is needed to understand the morphological
changes and the fate of the enameloid through ontogeny.
Does this layer disappear within one replacement cycle
during metamorphosis or is there a gradual reduction over
successive tooth generations during ontogeny? Answering
this question would require careful sectioning to pass
through the upper region of the 30-50 mm diameter tooth,
and the collection of data for each tooth generation, from
the first to the fourth or fifth, i.e. after metamorphosis. Such
a study could also generate information on the formation
and origin of the dentine-enamel junction.
(5) How do the dividing zone and the pedicel
appear during lissamphibian ontogeny?
In Pleurodeles waltl and Ambystoma mexicanum, it has been
reported that the first-generation teeth are undivided, whilst
a dividing zone and a pedicel are present in subsequent
teeth (Wistuba et al., 2002).
The formation of the dividing zone is related to change in
functioning of the odontoblasts: the absence of cytoplasmic
prolongations, which are known to play an important role in
dentinogenesis (Sasaki & Garant, 1986), is an indication of
this chcange in behaviour. A detailed morphological
comparison of the features of the odontoblasts lining the
crown with those lining the dividing zone in a growth series
could determine whether the dividing zone forms pro-
gressively through successive tooth generations or whether it
appears suddenly in second-generation teeth. Such a com-
parison would also clarify how can a low-mineralised zone
can be produced and maintained when surrounded by
conventional, calcified dentine, and lined by a continuous
layer of odontoblasts. It should be possible to identify the
collagen type composing the fibrils, using both immunocy-
tochemistry and in situ hybridization techniques. Indeed, it
has been reported that the fibrils associated with this zone are
smaller in diameter than those of the adjacent dentine regions
in the crown and pedicel, which are mostly composed of type I
collagen (Zaki et al., 1970; Smith & Miles, 1971). Type V
collagen fibrils are known to be thinner than type I fibrils but
it is not known whether they are present in the dividing zone.
It would also be of interest to look for non-collagen proteins,
such as proteoglycans, which are known to be involved in the
mineralisation process.
A detailed study of pedicel formation in teeth of successive
generations should also provide data on the formation of
cementum. Are the cementoblasts simply osteoblasts or are
they cells originating from the dental sac? The role of Hertwig’s
sheath during the last stages of tooth development and after
teeth have attached to the bone support also might be
investigated to determine the relationship between retraction
of Hertwig’s sheath and formation of the cementum.
(6) What mechanisms control the initiation of
a replacement tooth in lissamphibians?
Our observations on replacement tooth initiation in
numerous growth series of Pleurodeles waltl have shown that:
(1) differentiation of the secondary dental lamina starts
long before the previous tooth becomes functional, and the
replacement tooth is well developed when the predecessor
tooth attaches to the bone support. Thus, the initiation
process is correlated neither with tooth attachment nor with
eruption but presumably is initiated for all teeth in relation
to a particular developmental step of the previous tooth. In
reptiles, the regular pattern of tooth replacement strongly
suggests this possibility (Delgado, Davit-Be
´al & Sire, 2003),
however, the genetic mechanisms controlling this process
have yet to be understood.
(2) The secondary dental lamina forms from the upper
region of the dental organ of the previous tooth and is always
located lingually and slightly posterior to the latter. This
suggests that the cells of this particular region of the outer
dental epithelium have conserved the ability to differentiate
into a dental lamina. In teleost fish, in which the replacement
process is very similar to that in lissamphibians, Huysseune &
Thesleff (2004) suggested that this region of the dental organ
could be an epithelial stem cell niche. Further investigations
could extend to the genetic level, and particularly to the
molecular cascades underlying the regulation of such
a putative stem cell niche. The role of nerves in tooth
replacement also needs further study. In a cichlid fish (Tilapia
mariae) subjected to unilateral denervation of the lower jaw
through neurectomy of the ramus alveolaris trigemini, replace-
ment teeth did not form; the tooth germs already initiated at
the time of denervation continued to grow but no new germ
was initiated when a functional tooth was lost (Tuisku &
Hildebrand, 1994). This suggests that the secondary dental
lamina is no longer initiated when the nerve is disrupted;
similar studies on Lissamphibians are lacking.
(7) Which mechanisms control the initiation of
tooth resorption?
It is assumed that a functional tooth is subjected to
resorption as a reaction to the presence of a developing
replacement tooth, hence, the absence of replacement teeth
would lead to unresorbed functional teeth. In cichlid fish in
which replacement teeth were no longer initiated after
unilateral denervation, some teeth were lost in the absence
of replacement teeth but there were no data on the
functional life of these teeth (Tuisku & Hildebrand, 1994).
The first-generation teeth in Pleurodeles waltl are not resorbed
when their successors are growing, while in all other
generations the replacement tooth provokes tooth resorption
and loss (Davit-Be
´al, Allizard & Sire, in press), suggesting
that resorption occurs only when the replacement tooth
develops close to the previous tooth. Again, detailed
comparative work is required to investigate this further.
Amphibian tooth morphology and development 73
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

(8) What is the fate of the tooth tip in adult
lissamphibians?
While most authors agree that teeth are almost completely
resorbed, the fate of the enamel at the tooth tip is less clear.
Are osteoclasts able to resorb enamel? It is possible that
teeth are entirely resorbed in larvae, but not in adults,
where a large part of the crown can be shed when the
dividing zone is undergoing resorption. Detailed studies of
tooth resorption in a growth series, based on serial sections
and on observations at the TEM level is necessary to
understand this process.
(9) What mechanism controls the periodicity of
lissamphibian tooth replacement?
To date, no experimental data are vailable on the control of
tooth replacement patterns in nonmammalian species.
Some authors, however, have speculated that the tissues
capable of forming teeth could be programmed to produce
buds in a given sequence as the field becomes progressively
differentiated (e.g. Osborn, 1970, 1971, 1973; reviewed by
Berkovitz, 2000). Once the initial sequence has been estab-
lished it could be maintained by an intra-oral mechanism(s)
including repression of the developing tooth by the func-
tional tooth until a particular point. Such ‘‘local’’ control
over replacement tooth formation is in conflict with earlier
conclusions about a Zahnreihen model of waves of re-
placement (see Section V.4).
A detailed and comparative study performed on various
developmental stages could reveal whether there is
a correlation between the initiation of a new replacement
tooth and the developmental stage of the preceding tooth in
the same family. In Pleurodeles waltl, Davit-Be
´al et al. (2006)
found a correlation between the time the tooth becomes
functional (i.e. eruption]attachment) and initiation of its
successor. Study of the expression pattern of genes known to
be involved in tooth initiation in mammals may help to
clarify control mechanisms of tooth replacement in P. waltl.
(10) How do thyroxine levels affect tooth shape
in lissamphibian teeth?
It is clear that thyroxine affects both the entire enamel
organ (inner and outer dental epithelium), which deter-
mines the shape of the cusps when the first tooth tissues are
deposited, and stimulates the ameloblasts, which are
responsible for enamel formation. Thyroxine also stimulates
the osteoclasts responsible for skeletal remodelling during
metamorphosis.
Thyroxine (T
4
and its deiodinated form T
3
, which is more
active) passes through the cell membrane, binds to nuclear
receptors in immediate contact with DNA and triggers a shift
in gene transcription causing, in competent tissues, a set of
major changes through either cell alteration, proliferation,
differentiation or migration (Jacobs, Michielsen & Kuhn,
1988). Thyroid hormone receptor genes (TR alpha and beta)
have been sequenced in lissamphibians (Safi et al.,1997,
2004; Sachs et al.,2002).InXenopus laevis, the gene sonic
hedgehog (Xshh), which is known to be involved in tooth
morphogenesis in mammals, is activated as a direct response
to thyroid hormone (Stolow & Shi, 1995).
In lissamphibians, the larval thyroid gland contains
thyroxine at hatching (Hanaoka et al., 1973), but its plasma
concentration is below detectable levels: teeth are monocuspid
and the enamel is thin. During metamorphosis, plasma T
4
levels increase to reach 25-30 nmol l
[1
(Larras-Regard,
Taurog & Dorris, 1981): teeth become bicuspid and the
enamel thickens. Understanding the expression of TR and
shh genes in the enamel organ of larvae and meta-
morphosing Pleurodeles waltl could clarify the developmental
action of thyroid hormone and its relationship with enamel
thickness and cusp formation.
That the enamel organ is also sensitive to androgens is
shown by the transformation of bicuspid into monocuspid
teeth during the breeding season in males of many
plethodontid species (Stewart, 1958; Ehmcke & Clemen,
2000b; Ehmcke et al., 2003). Here, the effects of androgens
are opposite (loss of a cusp) to those of thyroxine (gain of
a cusp). It would be interesting to determine (i) how
competition between T
4
receptors and androgen receptors
is regulated in tooth-forming cells, and particularly in
ameloblasts, of these animals, and (ii) how this competition
leads to the monocuspid condition typically a feature of
larvae, in which levels of T
4
and androgens are below
detection limits detectable.
To our knowledge, there are no data on the effects of
thyroxine on tooth shape in mammals. For example, is the
enamel knot a target of T
4
or T
3
? This should be
investigated in light of the involvement of these hormones
in tooth shape changes in lissamphibians.
IX. CONCLUSIONS
(1) This review is the first major summary of current
knowledge on teeth in extinct and extant amphibians
(Caudata, Gymnophiona, and Anura). A large amount of
morphological data are already available on diverse aspects
of tooth biology in various lissamphibian species, some of
them being potentially good model animals given their easy
breeding at the laboratory and their amenability to
experimental studies.
(2) To date, research has principally focused on tooth
development and replacement, and on changes in mor-
phology and structure during ontogeny and metamorpho-
sis. Taken together these studies have established solid basis
for further research aiming to elucidate tooth developmen-
tal mechanisms in lissamphibians, and particularly in
looking for gene expression (and function) during tooth
morphogenesis and differentiation. These future studies will
prove to be important for understanding evolutionary
developmental biology of teeth. Indeed, dental research in
vertebrates suffers from a lack of comparison of genes
involved in representative species of various nonmammalian
lineages. Because all animals share many of the same
molecular processes, such comparative studies are a pre-
requisite to understand, for instance, what makes one tooth
different from another.
Tiphaine Davit-Be
´al and others74
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

(3) This review highlights important questions which
remain to be answered. They should be adressed using
comparative morphological studies at the tissue and cell
levels, and molecular techniques. Most questions which
could be addressed in lissamphibians concern more
generally tooth evolution and development in vertebrates.
For instance, the enameloid-enamel transition, which has
occurred in several lineages during vertebrate tooth
evolution, could be studied in caudates using molecular
markers for enamel proteins. Indeed, such a transition exists
in caudates, in which larval teeth possess both enameloid
and enamel, while teeth are only covered by enamel in post-
metamorphosed animals.
(4) Future research could also deal with the genetical
mechanisms that control replacement tooth initiation,
probably from the possible activation of stem cells. This is
a hot subject when considering current investigations on
tooth regeneration in mammals. Indeed, dissecting the
genetic pathway(s) involved in periodically replacing teeth
in lissamphibians could help much in understanding why
teeth can no longer be continuously replaced in mammals.
Similarly, the clear correlation that exists between the
increase of thyroxine levels at metamorphosis and tooth
shape change (monocuspid to bicuspid) seems to be a good
opportunity to study in detail the role played by thyroid
hormones in shaping the enamel organ.
(5) Lissamphibians appear to be good candidates for
such research because tooth structure is similar to that in
mammals and teeth are renewed continuously during life.
This allows to perform developmental studies not only in
larvae but also in later stages, in larger specimens. The
salamander Pleurodeles waltl, which has been used extensively
in experimental embryology research during the past
century, appears to be one of the most favourable
lissamphibian to be used as a model species in further
studies of tooth development.
X. ACKNOWLEDGEMENTS
We are grateful to Ann Huysseune (Ghent University,
Belgium), and Marvalee and David Wake (University of
California, Berkeley, USA) for critical reading of previous
drafts of the manuscript, and to Michel Laurin (CNRS,
Paris) for fruitful discussions. We thank Miss Francxoise
Allizard for excellent technical assistance. SEM and TEM
work was carried out at the ‘‘Service de Microscopie
Electronique de l’IFR de Biologie Inte
´grative – CNRS/
Universite
´Paris VI.
XI. REFERENCES
AHLBERG,P.E.&MILNER, A. R. (1994). The origin and early
diversification of tetrapods. Nature (Lond.) 368, 507–514.
ANDERSON, J. S. (2001). The phylogenetic trunk: Maximal inclusion
of taxa with missing data in an analysis of the Lepospondyli
(Vertebrata, Tetrapoda). Syst. Biol.50, 170–193.
AMPHIBIAWEB (2005). Information on amphibian biology and
conservation. [web application]. Berkeley, California: Amphib-
iaWeb. http://amphibiaweb.org.
BARTLETT,J.D.&SIMMER, J. P. (1999). Proteinases in developing
dental enamel. Crit. Rev. Oral Biol. Med.10, 425–441.
BARTLETT, J. D., SIMMER, J. P., XUE, J., MARGOLIS,H.C.&
MORENO, E. C. (1997). Molecular cloning and mRNA tissue
distribution of a novel matrix metalloproteinase isolated from
porcine enamel organ. Gene 183, 123–128.
BENESKI,J.T.J.&LARSEN, J. H. (1989a). Interspecific, ontogenetic and
life history variation in the tooth morphology of mole salamanders
(Amphibia, Urodela, Ambystomatidae). J. Morphol.199, 53–69.
BENESKI,J.T.J.&LARSEN, J. H. (1989b). Ontogenetic alterations
in the gross tooth morphology of Dicamptodon and Rhyacotriton
(Amphibian, Urodela, Dicamptodontidae). J. Morphol.199,
165–174.
BERKOVITZ, B. K. (2000). Tooth replacement patterns in non-
mammalian vertebrates. In Development, function and evolution of
teeth (eds M. F. Teaford, M. M. Smith & M. W. J. Ferguson), pp.
186–200. Cambridge University Press, London.
BOLT, J. R. (1969). Lissamphibian origins: possible Protolissam-
phibian from the Lower Permian of Oklahoma. Science 166,
888–891.
BOLT, J. R. (1977). Dissophoroid relrationships and ontogeny, and
the origin of the Lisamphibia. J. Paleont.51, 235–249.
BOLT, J. R. (1979). Amphibamus grandiceps as a juvenile dissophorid:
evidence and implications. In Mazon Creek fossils (ed. M. Nitecki),
pp. 529–563. Academic press, London.
BOLT, J. R. (1980). New tetrapods with bicuspid teeth from the Fort
Sill Locality (Lower Permian, Oklahoma). N. Jb. Geol. Pala
¨ont.
Mh.8, 449–459.
BOLT, J. R. (1991). Lissamphibian origins. In Origin of the higher groups
of tetrapods (eds. H.-P. Schultze & L. Trueb), pp. 194–222.
Comstock, Ithaca.
BOLTE,M.&CLEMEN, G. (1991). Die Zahnformen im Unterkiefer
des adulten Ambystoma mexicanum Shaw (Urodela, Amphibia).
Acta. Biol. Benrodis 3, 171–177.
BOLTE,M.&CLEMEN, G. (1992). The enamel of larval and adult
teeth of Ambystoma mexicanum Shaw (Urodela: Ambystomatidae) -
a SEM study. Zool. Anz.228, 167–173.
BOLTE, M., KREFTING,E.&CLEMEN, G. (1996). Hard tissue and
their calcium and phosphate content in Ambystoma mexicanum
(Urodela: Amphibia). Ann. Anat.178, 71–80.
BORDAY-BIRRAUX, V., VAN DER HEYDEN, C., DEBIAIS-THIBAUD, M.,
VERREIJDT, L., STOCK,D.W.,HUYSSEUNE, A., & SIRE, J.-Y.
(2006). Expression of Dlx genes during the development of the
zebrafish pharyngeal dentition: evolutionary implications. Evol.
Dev., 8, 130–141.
BORDZILOVSKAYA,N.P.&DETTLAFF, T. A. (1979). Table of stages of
the normal development of axolotl embryos and the prognos-
tication of timing of successive developmental stages at various
temperatures. Axolotl Newslett.7, 2–22.
BOUVET,J.&CHIBON, P. (1976). Destruction des tissus dentaires dans
les dents a
ˆge
´es d’Amphibiens. Bull. Soc. Zool. Fr.101,138.
BOY, J. A. (1978). Die Tetrapodenfauna (Amphibia, reptilia) des
Saarpfa
¨lzischcen Rotliegenden (Unter-Perm; SW-Deutschland).
1. Branchiosaurus. Mainzer geowiss. Mitt.7, 27–76.
BYSTROW, A. P. (1938). Dvinosaurus, als neotenische Form der
Stegocephalan. Acta Zool.19, 209–295.
CAMBRAY, J. C. (1976). Recherche sur la morphogene
`se de l’appareil
maxillo-dentaire chez Xenopus laevis Daudin (Amphibien
Anoure). The
`se 3e
`me cycle, Universite
´Paris VII.
Amphibian tooth morphology and development 75
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

CANNATELLA,D.C.&HILLIS, D. M. (1993). Amphibian phylogeny:
phylogenetic analysis of morphology and molecules. Herpetol.
Monogr.7, 1–7.
CANNATELLA,D.C.&HILLIS, D. M. (2004). Amphibians: Living a
life of slime. In Assembling the Tree of Life (eds. J. Cracraft and M. J.
Donoghue, eds). Oxford University Press, New York.
CARLSON, S. J. (1990). Vertebrate dental structures. In Vertebrate
biomineralisation: Patterns, processes and evolutionary trends (ed. J. G.
Carter), pp. 531–556. Van Nostrand Reinhold, New York.
CARROLL, R. L. (1988). Vertebrate paleontolog y and evolution. W.H.
Freeman and Company, New York.
CASEY,J.&LAWSON, R. (1981). A histological and scanning
electron microscope study of the teeth of caecilian amphibians.
Arch. Oral Biol.26, 49–58.
CHIBON, P. (1966). Analyse expe
´rimentale de la re
´gionalisation et
des capacite
´s morphoge
´ne
´tiques de la cre
ˆte neurale chez
l’amphibien urode
`le Pleurodeles waltlii Michah. Me
´
m. Soc. Zool.
36, 1–107.
CHIBON, P. (1967). Etude expe
´rimentale par ablations, greffes
et autoradiographie, de l’origine des dents chez l’amphibien
urode
`le Pleurodeles waltlii Michah. Arch. Oral Biol.12, 745–753.
CHIBON, P. (1970). L’origine de l’organe admantin des dents. Etude
au moyen du marquage nucle
´aire de l’ectoderme stomode
´al.
Ann. Embryol. Morphol.3, 203–213.
CHIBON, P. (1972). Etude ultrastructurale et autoradiographique
des dents chez les amphibiens. Relations entre la morphoge
´ne
`se
dentaire et l’activite
´thyroı
¨dienne. Bull. Soc. Zool. Fr.97, 437–
448.
CHIBON, P. (1977). Vitesse de croissance et renouvellement des dents
chez les amphibiens. J. Embryol. Exp. Morphol.42, 43–63.
CHIBON,P.&BOUVET, J. (1976). Me
´canisme cellulaire de la
destruction des dents a
ˆge
´es chez les Amphibiens. Bull. Soc. Zool.
Fr.101, 928.
CHIBON, P., ROUX, J.-P. & SPINELLI, M. (1971). Pre
´sence d’e
´mail
dans les dents d’amphibiens Urode
`les et Anoures. Etude
autoradiographique et ultrastructurale avant et apre
`slame
´ta-
morphose. C. R. Acad. Sci. (Paris) 272, 466–468.
CLEMEN, G. (1978a). Aufbau und Vera
¨nderungen der Gaumen-
zahnleisten beim larvalen und metamorphosierenden Salamandra
salamandra (L.) (Salamandridae: Amphibia). Zoomorphologie 90,
135–150.
CLEMEN, G. (1978b). Beziehungen zwischen Gaumenknochen und
ihren Zahnleisten bei Salamandra salamandra (L.) wa
¨hrend der
Metamorphose. Wilhelm Roux’s Arch.185, 19–36.
CLEMEN, G. (1979a). Die Bedeutung des Ramus palatinus fu
¨r die
Vomerspangenbildung bei Salamandra salamandra.Wilhelm Roux’s
Arch.187, 219–230.
CLEMEN, G. (1979b). Experimentelle Vera
¨nderungen am kno
¨cher-
nen Gaumenbogen der Axolotl-Larve und ihre Auswirkungen
wa
¨hrend der Metamorphose. Zool. Anz.203, 23–34.
CLEMEN, G. (1979c). Untersuchungen zur Bildung der Vomerspange
bei Salamandra salamandra (L.). Wilhelm Roux’s Arch.185, 305–321.
CLEMEN, G. (1988). Competence and reactions of early- and
late-larval dental laminae in original and not-original dental
systems of Ambystoma mexicanum Shaw. Arch. Biol. (Bruxelles) 99,
307–324.
CLEMEN,G.&GREVEN, H. (1974). Morphologische Untersuchun-
gen an der Mundho
¨hle von Urodelen. I. Die Gaumenza
¨hne von
Salamandra salamandra (L.) (Salamandridae: Amphibia). For m a et
Functio 7, 249–280.
CLEMEN,G.&GREVEN, H. (1977). Morphologische Untersuchun-
gen an der Mundho
¨hle von Urodelen. III. Die Munddachbe-
zahnung von Ambystoma mexicanum Cope (Ambystomatidae:
Amphibia). Zool. Jb. Anat.98, 95–136.
CLEMEN,G.&GREVEN, H. (1979). Morphologishe Untersuchun-
gen an der Mundho
¨hle von Urodelen. V. Die Munddachbe-
zahnung von Triturus vulgaris (L.) (Salamandridae: Amphibia).
Zool. Jb. Anat.102, 170–186.
CLEMEN,G.&GREVEN, H. (1980). Morphologische Untersuchun-
gen an der Mundho
¨hle von Urodelen. VII. Die Munddachbe-
zahnung von Amphiuma (Amphiumidae: Amphibia). Bonn. Zool.
Beitr.31, 357–362.
CLEMEN,G.&GREVEN, H. (1988). Morphological studies on the
mouth cavity of Urodela. IX. Teeth of the palate and the
splenials in Siren and Pseudobranchus (Sirenidae: Amphibia). Z.
Zool. Syst. Evol.-forsch.26, 135–143.
CLEMEN,G.&GREVEN, H. (2000). Dentigerous bones and
dentition in the paedomorphic plethodontid salamander Eurycea
neotenes.Alytes 18, 51–61.
CLEMEN, G., GREVEN,H.&SCHRO
¨DER, B. (1980). Ca and P
concentrations of a urodele tooth as revealed by electron
microprobe analysis. Anat. Anz.148, 422–427.
CLEMEN,G.&OPOLKA, A. (1990). Dental laminae and teeth of
embryonic Ichthyophis glutinosus (L.) (Amphibia: Gymnophiona).
Anat. Anz.170, 111–117.
COATES, M. I. (1996). The Devonian tetrapod Acanthostega gunnari
Jarvik: postcranial anatomy, basal tetrapod interrelationships
and patterns of skeletal evolution. Trans. R. Soc. Edinburgh 87,
363–421.
CORSIN, J. (1966). The development of the osteocranium of
Pleurodeles waltlii Michahelles. J. Morphol.119, 209–216.
DAVI T-BE
´AL, T., ALLIZARD,F.&SIRE, J.-Y. (2006). Morphological
variations in a tooth family through ontogeny in Pleurodeles waltl
(Lissamphibia, Caudata). J. Morphol.267, 1048–1065.
DELGADO, S., DAVI T-BE
´AL,T.&SIRE, J. Y. (2003). The dentition and
tooth replacement pattern in Chalcides (Squamata; Scincidae).
J. Morphol.256, 146–159.
DELGADO, S., GIRONDOT,M.&SIRE, J. Y. (2005). Molecular
evolution of amelogenin in mammals. J. Mol. Evol.60, 12–30.
DELGADO S., COUBLE M.-L., MAGLOIRE H., SIRE J.-Y. (2006).
Cloning, sequencing and expression of the amelogenin gene in
two scincid lizards. J. Dent. Res., 85, 138–143.
DEUCHAR, E. M. (1975). Xenopus: the South African clawed toad.
J. Wiley & Sons, London.
DOURNON,C.&CHIBON, P. (1974). Influence de la tempe
´rature, de
l’a
ˆge et des conditions hormonales (thyroxine) sur la pro-
life
´ration cellulaire chez la jeune larve et pendant la me
´ta-
morphose du crapaud Bufo bufo L. (Amphibien Anoure). Wilhem
Roux’s Arch. EntwMech. Org.175, 27–47.
DUELLMAN,W.E.&TRUEB, L. (1986). Biology of Amphibia. McGraw-
Hill, New York.
EATON, T. H. J. (1959). The ancestryof the moder nA mphibia:A r eview
of the evidence. Univ. Kansas Publ., Mus. Natl. Hist.12, 155–180.
EDMUND, A. G. (1960). Tooth replacement phenomena in lower
vertebrates. Contrib. R. Ont. Mus., Life Sci. Div.52, 1–190.
EDMUND, A. G. (1962). Sequence and rate of tooth replacement
in the Crocodilia. Contrib. R. Ontario Mus., Life Sci. Div.56,
1–42.
EDMUND, A. G. (1969). Dentition. In Biolog y of of the reptiles (eds
C. Gans, A. d’A. Bellairs, T. S. Parsons), pp. 117–200. New York:
Academic Press.
EHMCKE,J.&CLEMEN, G. (2000a). Development of the pattern
of dentition and dental laminae in Costa Rican plethodon-
tid salamanders (Amphibia, Urodela). Ann. Anat.182, 327–338.
Tiphaine Davit-Be
´al and others76
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

EHMCKE,J.&CLEMEN, G. (2000b). Teeth and their sex-
dependent dimorphic shape in three species of Costa Rican
plethodontid salamanders (Amphibia: Urodela). Ann. Anat.182,
403–414.
EHMCKE, J., WISTUBA, J., CLEMEN,G.&SCHLATT, S. (2003).
Targeted expression of androgen receptors in tooth-forming
tissues of a neotropical salamander (Bolitoglossa schizodactyla)
enables region-specific formation of dimorphic types of teeth in
the male. Gen. Comp. Endocrinol.134, 26–35.
ESTES, R. (1965). Fossil salamanders and salamander origins. Am.
Zool.5, 319–334.
ESTES,R.&RIEG, O. A. (1973). The early fossil record of frogs: A
review of the evidence. In Evolutionary biology of the anurans:
Contemporary research on major problems (ed. J. L. Vial), pp. 11–63.
University of Missouri Press, Columbia.
FELLER,A.&HEDGES, S. B. (1998). Molecular evidence for the
early history of living amphibians. Mol. Phyl. Evol.9, 509–516.
FORD,L.S.&CANNATELLA, D. C. (1993). The major clades of frogs.
Herpetol. Monogr.7, 94–117.
FROST D. (2004). Amphibian species of the world 3.0. An online reference.
(http://research.amnh.org/herpetology/amphibia/index.php).
Ó1998-2004, Darrel Frost and The American Museum of
Natural History.
FUKAE, M., TANABE, T., UCHIDA, T., LEE, S. K., RYU,O.H.,
MURAKAMI, C., WAKIDA, K., SIMMER, J. P., YAMADA,Y.&
BARTLETT, J. D. (1998). Enamelysin (matrix metalloproteinase-
20): Localization in the developing tooth and effects of pH and
calcium on amelogenin hydrolysis. J. Dent. Res.77, 1580–1588.
GABRION,J.&CHIBON, P. (1972). Morphologie dentaire et
disposition des dents chez les tritons ne
´ote
´niques (Triturus
helveticus Raz.). C. r. Se
´
anc. Soc. Biol.167, 50–54.
GALLIEN,L.&DUROCHER, M. (1957). Table chronologique du
de
´veloppement chez Pleurodeles waltlii Michah. Bull. Biol.2, 1–18.
GILLETTE, R. (1955). The dynamics of continuous succession of
teeth in the frog (Rana pipiens). Am. J. Anat.96, 1–36.
GOIN,C.&HESTER, M. (1961). Studies on the development,
succession and replacement of teeth in the frog Hyla cinerea.J.
Morphol.109, 279–287.
GOSS,R.J.(1969).Principles of regeneration. Academic Press, New York.
GOSS,R.J.&STAGG, M. W. (1958a). Regeneration of lower jaws in
adult newts. J. Morphol.102, 289–309.
GOSS,R.J.&STAGG, M. W. (1958). Regeneration in lower jaws of
newts after excision of the intermandibular region. J. Exp. Zool.
137, 1–11.
GRAVER, H. T. (1973). The polarity of the dental lamina in the
regenerating salamander jaw. J. Embryol. Exp. Morphol.30, 635–646.
GRAVER, H. T. (1974). Origin of the dental lamina in the
regenerating salamander jaw. J. Exp. Zool.189, 73–84.
GRAVER, H. (1978). Re-regeneration of lower jaws and the dental
lamina in adult urodeles. J. Morphol.157, 269–80.
GREVEN,H.(1984).ThedentitionofGegeneophis ramaswamii Ta yl or,
1964 (Amphibia, Gymnophiona), with comments on monocuspid
teeth in the Amphibia. Z. Zool. Syst. Evolut. Forsch.22, 342–348.
GREVEN, H. (1986). On the diversity of the tooth crowns in
Gymnophiona. Bull. Soc. Zool. Fr.43, 85–90.
GREVEN,H.&CLEMEN, G. (1979). Morphological studies on the
mouth cavity of urodeles. IV. The teeth of the upper jaw and the
palatein Necturus maculosus (Rafinesque) (Proteidae: Amphibia).
Arch. Histol. Jpn.42, 445–457.
GREVEN,H.&CLEMEN, G. (1980). Beobachtungen an den Za
¨hnen
und Zahnleisten des Munddaches von Siphonops annulatus
(Mikan) (Amphibia: Gymnophiona). Anat. Anz.147, 270–279.
GREVEN,H.&CLEMEN, G. (1985). Morphological studies on the
mouth cavity of Urodela. VIII. The teeth of the upper jaw and
the palate in two Hynobius species (Hynobiidae: Amphibia).
Z. Zool. Syst. Evol. Forsch.23, 136–147.
GREVEN,H.&CLEMEN, G. (1990). Effect of hypophysectomy on
the structure of normal and ectopically transplanted teeth in
larval and adult urodeles. Acta Embryol. Morphol. Exp.11 (n. sp),
33–43.
GREVEN,H.&LAUMEIER, I. (1987). A comparative SEM-study on
the teeth of 10 anuran species. Anat. Anz.164, 103–116.
HANAOKA, Y., KOYA, S. M., KONDO, Y., KOBAYASHI,Y.&
YAMAMOTO, K. (1973). Morphological and functional matura-
tion of the thyroid during early development of anuran larvae.
Gen. Comp. Endocrinol.21, 410–423.
HAY, J. M., RUVINSKY, I., HEDGES,S.B.&MAXSON, L. R. (1995).
Phylogenetic relationships of amphibian families inferred from
DNA sequences of mitochondrial 12S and 16S ribosomal RNA
genes. Mol. Biol. Evol.12, 928–937.
HEDGES,S.B.&MAXSON, L. R. (1993). A molecular perspective on
lissamphibian phylogeny. Herpetol. Monogr.7, 27–42.
HEDGES, S. B., MOBERG,K.D.&MAXSON, L. R. (1990). Tetrapod
phylogeny inferred from 18S and 28S ribosomal RNA
sequences and a review of the evidence for amniote relation-
ships. Mol. Biol. Evol.7, 607–633.
HEROLD, R., ROSENBLOOM,J.&GRANOVSKY, M. (1989). Phyloge-
netic distribution of enamel proteins: immunohistochemical
localization with monoclonal antibodies indicates the evolution-
ary appearance of enamelins prior to amelogenins. Calcif. Tissue
Int.45, 88–94.
HERTWIG, O. (1874). Ueber das Zahnsystem der Amphibien und
seine Bedeutung fu
¨r die Genese des Skelets der Mundho
¨hle.
Arch. mikr. Anat.11, 1–208.
HOLMGREN, N. (1952). An embryological analysis of the mamma-
lian carpus and its bearing upon the question of the origin of the
tetrapod limb. Acta Zool.33, 1–115.
HOOK, R. W. (1983). Colosteus scutellatus (Newberry), a primitive
temnospondyl amphibian from the middle Pensylvanian of
Linton, Ohio. Amer. Mus. Novit.2770, 1–41.
HOWES, R. (1977). Root formation in ectopically transplanted teeth
of the frog, Rana pipiens. I. Tooth morphogenesis. Acta Anat.97,
151–65.
HOWES, R. (1978). Regeneration of ankylosed teeth in the adult
frog premaxilla. Acta anat.101, 179–86.
HRAOUI-BLOQUET,S.&EXBRAYAT, J.-M. (1996). Les dents de
Typhlonectes compressicaudatus (Amphibia, Gymnophiona) au cours
du de
´veloppement. Ann. Sci. Nat. Zool., Paris 17, 11–23.
HUYSSEUNE,A.&SIRE, J.-Y. (1997a). Structure and development of
first-generation teeth in the cichlid Hemichromis bimaculatus
(Teleostei, Cichlidae). Tissue Cell 29, 679–697.
HUYSSEUNE,A.&SIRE, J.-Y. (1997b). Structure and development of
teeth in three armoured catfish Corydoras arcuatus,C. aeneus and
Hoplosternum littorale (Siluriformes, Callichthyidae). Acta Zool.
(Stockh.) 78, 69–84.
HUYSSEUNE,A.&SIRE, J.-Y. (1998). Evolution of patterns and
processes in teeth and tooth-related tissues in non-mammalian
vertebrates. Eur. J. Oral Sci.106, 437–481.
HUYSSEUNE,A.&THESLEFF, I. (2004). Continuous tooth replace-
ment: The possible involvement of epithelial stem cells. BioEssays
26, 665–671.
HUYSSEUNE, A., VAN DER HEYDEN,C.&SIRE, J.-Y. (1998). Early
development of the zebrafish (Danio rerio) pharyngeal dentition
(Teleostei, Cyprinidae). Anat. Embryol.198, 289–305.
Amphibian tooth morphology and development 77
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

HUYSSEUNE,A.&WITTEN, P. E. (2006). Developmental mecha-
nisms underlying tooth patterning in continuously replacing
osteichthyan dentitions. J. Exp. Zool.306B, 204–215.
ISHIYAMA, M., MIKAMI, M., SHIMOKAWA,H.&OIDA, S. (1998).
Amelogenin protein in tooth germs of the snake Elaphe
quadrivirgata, immunohistochemistry, cloning and cDNA
sequence. Arch. Histol. Cytol.61, 467–474.
JACKMAN, W. R., DRAPER,B.D.&STOCK, D. W. (2004). Fgf
signaling is required for zebrafish tooth development. Dev. Biol.
274, 139–157.
JACOBS,G.F.,MICHIELSEN,R.P.&KUHN, E. R. (1988). Thyroxine
and triiodothyronine in plasma and thyroids of the neotenic and
metamorphosed axolotl Ambystoma mexicanum: influence of TRH
injections. Gen. Comp. Endocrinol.70, 145–151.
JARVIK, E. (1960). The
´
ories de l’Evolution des Verte
´
bre
´
s reconside
´
re
´
es a
`la
lumie
`
re des re
´
centes de
´
couvertes sur les verte
´
bre
´
s infe
´
rieurs. Masson, Paris.
JENKINS,F.A.Jr&WALSH, D. M. (1993). An early Jurassic caecilian
with limbs. Nature 365, 246–249.
JERNVALL, J., KERANEN,S.V.E.&THESLEFF I. (2000). Evolutionary
modification of development in mammalian teeth: quantifying
gene expression patterns and topography. Proc. Nat. Acad. Sci.
USA 97, 14444–14448.
KALLENBACH,E.&PIESCO, N. P. (1978). The changing morphology
of the epithelium-mesenchyme interface in the differentiation
zone of growing teeth in selected vertebrates and its relation-
ship to possible mechanism of differentiation. J. Biol. Bucc.6,
229–240.
KANGAS, A. T., EVANS, A. R., THESLEFF,I.&JERNVALL, J. (2004).
Nonindependence of mammalian dental characters. Nature 432,
211–214.
KATOW, H. (1979). Structure and formation of ankylosis in Xenopus
laevis.J. Morphol.162, 327–342.
KAUNG, H. (1975). Development of beaks of Rana pipiens larvae.
Anat. Rec.182, 404–14.
KERR, T. (1960). Development and structure of some actino-
pterygian and urodele teeth. Proc. Zool. Soc. Lond.133, 401–
422.
KOGAYA, Y. (1994). Sulfated glycoconjugates in amelogenesis. Prog.
Histochem. Cytochem.29, 1–110.
KOGAYA, Y. (1999). Immunohistochemical localisation of ameloge-
nin-like proteins and type I collagen and histochemical
demonstration of sulphated glycoconjugates in developing
enameloid and enamel matrices of the larval urodele (Triturus
pyrrhogaster) teeth. J. Anat.195, 455–464.
KOGAYA, Y., KIM, S., YOSHIDA, H., SHIGA,H.&AKISAKA, T. (1992).
True enamel matrix of the newt, Triturus pyrrhogaster, contains no
sulfated glycoconjugates. Cell Tissue Res.270, 249–256.
KVAM , T. (1946). Comparative study of the ontogenetic and
phylogenetic development of dental enamel. Norske Tannlaegeforen
56, 1–198.
KVAM , T. (1953). The phylogenetic transition from mesodermal to
ectodermal enamel. K. norske vidensk. Selsk. Forh.26, 83–84.
KVAM , T. (1960). The development of the tooth tip in Triton cristatus
Laur. Acta Odont. Scand.18, 503–519.
LARRAS-REGARD,E.,TAU ROG,A.&DORRIS, M. (1981). Plasma
thyroxine and triodothyroxine levels in Ambystoma tigrinumat various
stages of metamorphosis. Gen. Comp. Endocrinol.43, 443–450.
LARSEN,J.H.Jr&GUTHRIE, D. J. (1975). The feeding system of
terrestrial tiger salamanders (Ambystoma tigrinum melanosticum
Baird). J. Morphol.147, 137–154.
LARSON,A.&DIMMICK, W. W. (1993). Phylogenetic relationships of
the salamander families: A analysis of congruence among
morphological and molecular characters. Herpetol. Monogr.7,
77–93.
LAUGA-REYREL, F. (1974). Analyse descriptive de la re
´ge
´ne
´ration
des dents apre
`s ablation partielle de la ma
ˆchoire infe
´rieure chez
Pleurodeles waltlii Michah. C. r. Se
´
anc. Soc. Biol.278, 111–114.
LAURENT, R. (1947). La disposition des dents vome
´riennes chez les
urode
`les supe
´rieurs et son importance phyloge
´ne
´tique. Bull. Mus.
Roy. Hist. Nat. Belg.23, 1–4.
LAURENTI, P., THAERON, C., ALLIZARD, F., HUYSSEUNE,A.&SIRE,
J.-Y. (2004). The cellular expression of eve1 suggests its
requirement for initial patterning of the dentition and
differentiation of the ameloblasts in the zebrafish (Danio rerio).
Dev. Dyn.230, 727–733.
LAURIN, M. (1998a). The importance of global parsimony and
historical bias in understanding tetrapod evolution. Part I.
Systematics, middle ear evolution and jaw suspension. Ann. Sci.
Nat. Zool.1, 1–42.
LAURIN, M. (1998b). The importance of global parsimony and
historical bias in understanding tetrapod evolution. Part II.
Vertebral centrum, costal ventilation, and paedomorphosis.
Ann. Sci. Nat. Zool.2, 99–114.
LAURIN, M. (2002). Tetrapod phylogeny, amphibian origins, and
the definition of the name Tetrapoda. Syst. Biol.51, 364–369.
LAURIN,M.&REISZ, R. R. (1997). A new perspective on tetrapod
phylogeny. In Amniote origins - Completing the transition to land (eds
S. S. Sumida and K. Martin), pp. 9–59. Academic Press, London.
LAURIN,M.&REISZ, R. R. (1999). A new study of Solenodonsaurus
janenschi, and a reconsideration of amniote origins and
stegocephalian evolution. Can. J. Earth Sci.36, 1239–1255.
LAWSON, R. (1965a). The development and replacement of teeth
in Hypogeophis rostratus (Amphibia, Apoda). J. Zool.147, 352–
362.
LAWSON, R. (1965b). The teeth of Hypogeophis rostratus (Amphibia,
Apoda) and tooth structure in the Amphibia. Proc. Zool. Soc. Lond.
145, 321–325.
LAWSON, R. (1966). Tooth replacement in the frog Rana temporaria.
J. Morphol.119, 233–240.
LAWSON, R., WAKE,D.&BECK, N. (1971). Tooth replacement in
the Red-Backed Salamander, Plethodon cinereus.J. Morphol.134,
259–270.
LEVI, G. (1940). Etude sur le de
´veloppement des dents des
te
´le
´oste
´ens. III. Arch. Anat. micr. Morphol. exp.35, 415–455.
LEYDIG, F. (1867). Uber die Molche der wu
¨rttembergischen Fauna.
Arch. Naturgesch.33, 163–282.
LOMBARD,R.E.&SUMIDA, S. S. (1992). Recent progress in
understanding early tetrapods. Amer. Zool.32, 609–622.
MARCUS, H. (1920). Ueber die Za
¨hne und die Korrelation ihre mit dem
Alter, untersucht an Blindwu
¨hlen, Krokodilien und Seekuhen. George
Thieme, Leipzig.
MCDIARMID R. W. & ALTI G R. (1999). Tadpoles: the biolog y of anuran larvae.
University of Chicago Press, Chicago and London. xiv ]444 p.
MEANS, D. B. (1972). Comments on undivided teeth in urodeles.
Copeia 2, 586–588.
MILES,A.E.W.&POOLE, D. F. G. (1967). The history and general
organization of dentitions. In Structural and chemical organization of
teeth (ed. A. E. W. Miles), pp. 3–44. Academic Press, London,
New York.
MILLER,W.&ROWE, D. (1973). Preliminary investigation of
variations in tooth replacement in adult Necturus maculosus.
J. Morphol.140, 63–76.
MILNER, A. R. (1988). The relationships and origin of living
amphibians. In The phylogeny and classification of the Tetrapoda, vol. 1.
Tiphaine Davit-Be
´al and others78
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

Amphibians, reptiles, birds (ed. M. J. Benton), pp. 59–102.
Clarendon Press, Oxford.
MILNER, A. R. (1993). The Paleozoic relatives of lissamphibians.
Herpetol. Monogr.7, 8–27.
MILNER, A. R. (2000). Mesozoic and Tertiary Caudata and
Albanerpetontidae. In Amphibian biology, vol. 4. Palaeontology: the
evolutionary history of amphibians (eds H. Heatwole & R. L. Carroll),
pp. 1412–1444. Surrey Beatty and Sons, Chipping Norton,
NSW, Australia.
MOURY, J., CURTIS,S.&PAV, D. (1985). Structure of the radially
asymetrical uncalcified region of the teeth of the red-backed
salamander, Plethodon cinereus (Amphibia, Plethodontidae).
J. Morphol.185, 403–412.
MOURY,J.,CURTIS,S.&PAV, D. (1987). Structural heterogeneity in the
basal region of the teeth of the red-backed Salamander, Plethodon
cinereus (Amphibia, Plethodontidae). J. Morphol.194, 11–27.
MUTZ,T.&CLEMEN, G. (1992). Development and dynamics of the
tooth-systems in Eurycea and comparison of the definitive
dentition of the palate of Gyrinophilus (Urodela: Plethodontidae).
Z. Zool. Syst. Evol. Forsch.30, 220–223.
NIEMINEN, P., PEKKANEN, M., ABERG,T.&THESLEFF, I. (1998). A
graphical WWW-database on gene expression in tooth. Eur. J.
Oral Sci.106, 7–11.
NIEUWKOOP,P.D.&FABER, J. (1956). Normal table of Xenopus laevis
(Daudin). A systematical and chronological survey of the development from
the fertilized egg till the end of metamorphosis, 1st edition. North
Holland Publishing Co, Amsterdam.
NOBLE, G. K. (1926). The production of cloaca glands in the
adult female Desmognathus fuscus by testicular transplants; the change
of the tooth form in the adult male by castration. Anat. Rec.34, 140.
NOBLE, G. K. (1929). The relation of courtship to the secondary
sexual characters of the two-lined salamander, Eurycea bislineata
(Green). Am. Mus. Nov.362, 1–5.
NOBLE,G.K.&DAVI S, S. H. (1928). The effect on the dentition
and cloaca of testicular transplants in the adult female
salamander, Desmognathus; the effect of castration in the male.
Anat. Rec.28, 24.
NOBLE,G.K.&POPE, S. H. (1929). The modification of the cloaca
and teeth of the adult salamander, Desmognathus, by testicular
transplantation and by castration. Brit.J.Exp.Biol.6, 399–411.
OLTMANNS, E. (1952). Zur Morphologie der Za
¨hne rezenter
Amphibien. Anat. Anz.98, 369–389.
ØRV I G, T. (1967). Phylogeny of tooth tissues: evolution of some
calcified tissues in early vertebrates. In:Structural and chemical
organization of teeth (ed A. E. W. Miles), pp. 45–105. Academic
Press, London & New York.
OSBORN, J. W. (1970). New approach to Zahnreihen. Nature 225,
343–346.
OSBORN, J. W. (1971). The ontogeny of tooth succession in Lacerta
vivipara Jacquin (1787). Proc. R. Soc. Lond. B.179, 261–289.
OSBORN, J. W. (1973). The evolution of dentitions. Am. Scient.61,
548–559.
OWEN, R. (1842). On the teeth of species of the genus Labyrinthodon
(Mastodonsaurus of Jaeger), common to the German Keuper
Formation and the Lower Sandstone of Warwick and
Leamington. Trans. Geol. Soc. Lond 2, 503–514.
OWEN, R. (1845). Odontography; or a treatise on the comparative anatomy of
the teeth; their physiological relations, mode of development and microscopic
structure in the vertebrate animals. Baillie
`re, London.
PANG, P. K., YANG, M., OGURO, C., PHILLIPS,J.G.&YEE,J.A.
(1980). Hypotensive actions of parathyroid hormone prepara-
tions in vertebrates. Gen. Comp. Endocrinol.41, 135–138.
PARKER, H. W. (1956). Viviparous caecilians and amphibian
phylogeny. Nature 178, 250–252.
PARKER,H.&DUNN, E. (1964). Dentitional metamorphosis in the
Amphibia. Copeia 1, 75–86.
PARSONS,T.&WILLIAMS, E. (1962). The teeth of Amphibia and
their relation to amphibian phylogeny. J. Morphol.110, 375–389.
PARSONS,T.&WILLIAMS, E. (1963). The relationships of the
modern Amphibia : a reexamination. Quart. Res. Biol.38, 26–53.
PEYER, B. (1968). Comparative Odontology (ed. R. Zangerl). The
University of Chicago Press, Chicago.
POOLE, D. F. G. (1967). Phylogeny of tooth tissue: enameloid and
enamel in recent vertebrate, with a note on the history of
cementum. In Structural and chemical organization of teeth (ed. A. E.
W. Miles), pp. 111–149. Academic Press, London & New York.
RANDALL, M. (1966). Electron microscopical demonstration of
ferritin in the dental epithelial cells of urodeles. Nature 210,
1325–1326.
REGAL, P. (1966). Feeding specializations and the classification of
terrestrial salamanders. Evolution 20, 392–407.
REILLY, S. (1986). Ontogeny of cranial ossification in the eastern newt
Notophthalmus viridescens (Caudata: Salamandridae), and its relation-
ship to metamorphosis and neoteny. J. Morphol.188, 315–326.
REUTHER, G. (1931). Die Zahnleister von Hypogeophis. Beitrag zur
kenntnis der Gymnophionen. XIV. Morphol. Jb.68, 105–112.
ROELANTS,K.&BOSSUYT, F. (2005). Archaebatrachian paraphyly
and pangean diversification pf crown-group frogs. Syst. Biol.54,
111–126.
ROMER, A. S. (1945). Vertebrate Paleontology, 2nd ed. edition.
University Chicago Press, Chicago.
ROSE, C. S. (1999). Hormonal control in larval development and
evolution - amphibians. In Origin and evolution of larval forms (eds B. K.
Hall & M. H. Wake), pp 167–216. Academic Press, San Diego.
ROUX, J. (1973). Etude ultrastructurale de la dentinoge
´ne
`se chez la
larve du triton Pleurodeles waltlii (Amphibien Urode
`le). J. Biol.
Buccale 1, 21–32.
ROUX,J.&CHIBON, P. (1973). Etude ultrastructurale de
l’ame
´loge
´ne
`se chez la larve du triton Pleurodeles waltlii (Amphib-
ien Urode
`le). J. Biol. Buccale 1, 33–44.
ROUX,J.&CHIBON, P. (1974). Pre
´sence de cellules ne
´crotiques et
de phagocytes dans la pulpe de dents fonctionnelles d’Amphib-
iens Urode
`les. Re
´sultats pre
´liminaires. C. r. Se
´
anc. Soc. Biol.278,
1369–1372.
SACHS, L. M., JONES, P. L., HAV I S , E., ROUSE, N., DEMENEIX,B.A.
&S
HI Y. B. (2002). Nuclear receptor corepressor recruitment by
unliganded thyroid hormone receptor in gene repression during
Xenopus laevis development. Mol Cell Biol.24, 8527–8538.
SAFI R., BEGUE A., HANNI C., STEHELIN D., TATA J. R. & L AUDET V.
(1997). Thyroid hormone receptor genes of neotenic amphib-
ians. J. Mol. Evol.44, 595–604.
SAFI, R., BERTRAND, S., MARCHAND, O., DUFFRAISSE, M., DELUZE,
A., VANACKER, J. M., MARANINCHI, M., MARGOTAT, A.,
DEMENEIX,B.&LAUDET, V. (2004). The axolotl (Ambystoma
mexicanum), a neotenic amphibian, expresses functional thyroid
hormone receptors. Endocrinology 145, 760–772.
SAHARA, N., ASHIZAWA, Y., NAKAMURA, K., DEGUCHI,T.&SUZUKI,
K. (1998). Ultrastructural features of odontoclasts that resorb
enamel in human deciduous teeth prior to shedding. Anat. Rec.
252, 215–228.
SAN MAURO, D., GARCIA-PARIS,M.&ZARDOYA, R. (2004a).
Phylogenetic relationships of discoglossid frogs (Amphibia:
Anura:Discoglossidae) based on complete mitochondrial
genomes and nuclear genes. Gene 343, 357–366.
Amphibian tooth morphology and development 79
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

SAN MAURO, D., GOVER,D.J.,OOMMEN,O.V.,WILKINSON,M.&
ZARDOYA, R. (2004b). Phylogeny of caecilian amphibians
(Gymnophiona) based on complete mitochondrial genomes
and nuclear RAG1. Mol. Phyl. Evol.33, 413–427.
SAN MAURO, D., VENCES, M., ALCOBENDAS, M., ZARDOYA, R., &
MEYER, A. (2005). Initial diversification of living amphibians
predated the breakup of Pangaea. Am Nat.165, 590–599.
SASAKI,T.&GARANT, P. R. (1986). A study of post-secretory
maturation ameloblasts in the cat by transmission and freeze-
fracture electron microscopy. Arch. Oral Biol.31, 587–596.
SATO, I., KOBAYASHI, M., OGINO, K., OGAKI, A., YATSU,T.&SATO,
T. (1986a). The ultrastructure of the teeth in the amphibians:
differentiation of the dentine of Hyla arborea japonica and Rana
nigromaculata nigromaculata.Shigaku 74, 334–339.
SATO, I., KOBAYASHI, M., UENO,R.&SATO, T. (1986b). The
ultrastructure of the teeth in the Amphibia ; differentiation of
the enamel. Odontology 73, 1815–1820.
SCHILLING, T. F., WALKER,C.&KIMMEL, C. B. (1996). The chinless
mutation and neural crest cell interactions in zebrafish jaw
development. Development 122, 1417–1426.
SCHMIDT, W. (1957). Zur Durodentinbildung bei Urodelen-
Za
¨hnen. Z. Zellforsch.46, 281–285.
SCHMIDT, W. (1958). Zur Histologie und Fa
¨rbung der Za
¨hne des
Japanischen Riesensalamanders. Z. Zellforsch.49, 46–57.
SCHMIDT, W. (1970). Der Zahnschmelz urodeler und anurer
Amphibien. Z. Zellforsch.104, 295–300.
SCHOCH,R.R.&CARROLL,A.R.(2003).Ontogeneticevidence
for the Paleozoic ancestry of salamanders. Evol. Dev.5, 314–324.
SCHOCH,R.R.&MILNER, R. L. (2004). Structure anad implications
of theories on the origin of lissamphibians. In Recent advances in the
origin and early radiation of vertebrates (eds G. Arratia, M. V. H.
Wilson & R. Cloutier). Verlag Dr. Pfeil, Munich.
SCHULTZE, H. P. (1969). Die Faltenza
¨hne der Rhipidistiiden
Crossopterygier, der Tetrapoden und der Actinopterygier -
Gattung Lepisosteus.Paleontogr. Ital.65, 60–137.
SCHULTZE, H. P. (1970). Folded teeth and the monophyletic origin
of tetrapods. Am. Mus. Novitat.2408, 1–10.
SHAW, J. P. (1979). The time scale of tooth development and
replacement in Xenopus laevis (Daudin). J. Anat.129, 323–
342.
SHAW, J. P. (1983). Quantitative observations on the dental tissues
during tooth replacement in Hemiphractus proboscideus.J. Dent. Res.
62, 432.
SHAW, J. P. (1985). Tooth replacement in adult Xenopus laevis
(Amphibia: Anura). J. Zool. Lond.207, 171–179.
SHAW, J. P. (1986). A longitudinal study of tooth resorption in newly
metamorphosed Xenopus laevis with comments on tooth resorp-
tion in amphibians. J. Zool.208, 215–228.
SHAW, J. P. (1989). A morphometric study of bone and tooth
volumes in the pipid frog Xenopus laevis (Daudin), with comments
on the importance of tooth resorption during normal tooth
replacement. J. Exp. Zool.249, 99–104.
SIGNORET, J. (1960). Ce
´phaloge
´ne
`se chez le triton Pleurodeles waltlii
Michah, apre
`s traitement de la gastrula par le chlorure de
lithium. Me
´
m. Soc. Zool.32, 1–117.
SIRE, J.-Y. & HUYSSEUNE, A. (2003). Formation of skeletal and
dental tissues in fish: A comparative and evolutionary approach.
Biol. Rev.78, 219–249.
SIRE, J.-Y., DAVI T -BE
´AL, T., DELGADO, S., VAN DER HEYDEN,C.&
HUYSSEUNE, A. (2002). The first generation teeth in non-
mammalian lineages: evidence for a conserved ancestral
character? Microsc. Res. Techn.59, 408–434.
SIRENA, S. (1872). Ueber den Bau die Entwicklung der Za
¨hne bei den
Amphibien und reptilien. Verh. phys.-med. Ges. Wu
¨rzb.2, 124–144.
SMITH, M. M. (1995). Heterochrony in the evolution of enamel in
vertebrates. In Evolutionary changes and heterochrony (ed. K. J.
McNamara). J. Wiley and Sons, New York.
SMITH, M. M. (2003). Vertebrate dentitions at the origin of jaws:
when and how pattern evolved. Evol. Dev.5, 394–413.
SMITH,M.&MILES, A. (1971). The ultrastructure of odontogenesis
in larval and adult Urodeles. Differentiation of the dental
epithelial cells. Z. Zellforsch.121, 470–498.
SMITH, W. J. (1883). Beitrag zur differentiellen Diagnose der Rana
fusea s. platyrrhynus und Rana arvalis s. oxyrrhinus auf grund der an
den Gaumenza
ˆhnen nachweisbaren Unterschiede. Pflu
¨ger’s Arch.
ges. Physiol.30, 581–588.
SPINELLI,M.&CHIBON, P. (1973). Etude ultrastructurale de
l’odontoge
´ne
`se chez les Amphibiens Anoures. C. r. Se
´
anc. Soc.
Biol.277, 349–51.
STEWART, M. M. (1958). Seasonal variation in the teeth of the two-
lined salamander. Copeia, 190–197.
STOLOW,M.A.&SHI,Y.B.(1995).Xenopus sonic hedgehog as a
potential mor phogen during embryogenesis and thyroid hormone-
dependent metamorphosis. Nucleic Acid Res.23, 2555–2562.
SZARSKI, H. (1962). The origin of the amphibia. Quart. Rev. Biol.37,
189–241.
TAKAHAMA, H., SASAKI, F., TACHIBANA, T., ISEKI, H., HORIGUCHI,
T. , K INOSHITA,T.&WATANABE, K. (1987). Fine structure and
degeneration of the horny teeth and the epidermis of the labial
ridge of the anuran tadpole. Tsurumi Shigaku 13, 49–63.
TAYLOR,A.C.&KOLLROS, J. J. (1946). Stages in normal
development of Rana pipiens larvae. Anat. Rec.94, 11–22.
TAYLOR, E. H. (1968). The caecilians of the world: a taxonomic review.
University of Kansas Press, Lawrence.
TESCHE,M.&GREVEN H. (1989). Primary teeth in Anura are
not pedicellate and bladed. Z. zool. syst. Evolut.-forsch.27, 326–
329.
TOYOSAWA, S., O’HUIGIN, C., FIGUEROA, F., TICHY,H.&KLEIN,J.
(1998). Identification and characterization of amelogenin genes
in monotremes, reptiles, and amphibians. Proc. Natl. Acad. Sci.
95, 13056–13061.
TRUEB, L. (1993). Patterns of cranial diversity among lissamphibia.
In The skull - Patterns of structural and systematic diversity, vol. 2 (eds.
J. Hanken & B. K. Hall), pp. 255–343. University Chicago
Press, Chicago.
TRUEB,L.&CLOUTIER, R. (1991). A phylogenetic investigation of
the inter- and intrarelationships of the Lissamphibia (Amphibia:
Temnospondyli). In Origins of the higher groups of tetrapods.
Controversy and consensus (eds. H.-P. Schultze & L. Trueb), pp.
223–313. Comstock Publishing Associates/Cornell University
Press, Ithaca, N.Y.
TUISKU,F.&HILDEBRAND, C. (1994). Evidence for a neural
influence on tooth germ generation in a polyphyodont species.
Dev. Biol.165, 1–9.
VAN DER HEYDEN,C.&HUYSSEUNE, A. (2000). Dynamics of tooth
formation and replacement in the zebrafish (Danio rerio)
(Teleostei, Cyprinidae). Dev. Dyn.219, 486–496.
VAN DER HEYDEN, C., HUYSSEUNE,A.&SIRE, J.-Y. (2000).
Development and fine structure of pharyngeal teeth in juvenile
zebrafish (Danio rerio) (Teleostei, Cyprinidae). Cell Tissue Res.302,
205–219.
WAGNER, G. (1949). Die Bedeutung der Neuralleiste fu
¨rdie
Kopfgestaltung der Amphibienlarven Untersuchungen an Chi-
maeren von Triton und Bombinator.Rev. Suisse Zool.56, 519–620.
Tiphaine Davit-Be
´al and others80
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society

WAGNER, G. (1954). Uber die Zahnbildung in xenoplastischen
Neuralleistenchima
¨ren: Zahnanlagen aus Triton-Schmelzorgan
und Bombinator-Mesektoderm in Triton-Wirten. Rev. Suisse Zool.
61, 395–402.
WAKE,D.B.&DEBAN, S. M. (2000). Terrestrial feeding in sala-
manders. In Feeding: Form, function and evolution in tetrapod vertebrates
(ed. K. Schwenk), pp. 95–116. Academic Press, San Diego.
WAKE, M. H. (1976). The development and replacement of teeth
in viviparous caecilians. J. Morphol.148, 33–64.
WAKE, M. H. (1980). Foetal tooth development and adult
replacement in Dermophis mexicanus (Amphibia: Gymnophiona).
Fields versus clones. J. Morphol.166, 203–216.
WAKE, M. H. (1993). The evolution of oviductal gestation in
amphibians. J. Exp. Zool.266, 394–413.
WAKE, M. H. (2003). The osteology of caecilians. In Amphibian
biology, vol. 5. Osteology (eds. H. Heatwole & M. Davies), pp.
1809–1876. Surrey Beatty & Sons, NSW, Australia.
WAKE,M.H.&WURST, G. (1979). Tooth crown morphology in
Caecilians (Amphibia: Gymnophiona). J. Morphol.159, 331–342.
WAKE, T. A., WAKE,D.B.&WAKE, M. H. (1983) The ossification
sequence of Aneides lugubris, with comments on heterochrony.
J. Herpetol.17, 10–22.
WANG, X., ITO, Y., LUA N, X., YAMANE,A.&DIEKWISCH,T.G.
(2005). Amelogenin sequence and enamel biomineralization in
Rana pipiens.J. Exp. Zool. B Mol. Dev. Evol.304, 177–186.
WILKINSON, M. (1991). Adult tooth crown morphology in the
Typhlonectidae (Amphibia: Gymnophiona): a reinterpretation
of variation and its significance. Z. zool. syst. Evolut.-forsch.29,
304–311.
WISTUBA, J., BOLTE,M.&CLEMEN, G. (2000). The odontoclasts of
Ambystoma mexicanum.Ann. Anat.182, 415–422.
WISTUBA, J., GREVEN,H.&CLEMEN, G. (2002). Development of
larval and transformed teeth in Ambystoma mexicanum (Urodela,
Amphibia): an ultrastructural study. Tissue Cell 34, 14–27.
WITTEN, P. E. (1997). Enzyme histochemical characteristics of
osteoblasts and mononucleated osteoclasts in a teleost fish with
acellular bone (Oreochromis niloticus, Cichlidae). Cell Tiss. Res.287,
591–599.
YAEGER,J.&KRAUCUNAS, E. (1969). Fine structure of the resorptive
cells in the teeth of frogs. Anat. Rec.164, 1–14.
ZAKI,A.E.&MAC RAE, E. K. (1977). Fine structure of the
secreting and nonsecreting ameloblasts in the frog. I. Fine
structure of the secretory ameloblasts. Am. J. Anat.148, 161–
194.
ZAKI,A.E.&MACRAE, E. K. (1978). Fine structure of the secreting
and nonsecreting ameloblasts in the frog. II. Fine structure of
the non-secretory ameloblasts. J. Morphol.158, 181–197.
ZAKI, A. E., YAEGER,J.A.&GILLETTE, R. (1970). Fine structure of
the epithelial dental organ in the frog during early odonto-
genesis. Anat. Rec.168, 79–92.
Amphibian tooth morphology and development 81
Biological Reviews 82 (2007) 49–81 Ó2007 The Authors Journal compilation Ó2007 Cambridge Philosophical Society