Cementum on Smilodon Sabers
HOLLISTON L. RIVIERE
*AND H. TODD WHEELER
School of Dentistry, Oregon Health and Sciences University, Portland, Oregon
George C. Page Museum of Discoveries, Los Angeles, California
The maxillary canines of Smilodon californicus Bovard, 1907 have a
deeply curved cementoenamel junction. The gingiva of modern cats is at-
tached to the tooth at the cementoenamel junction and provides tactile and
other dental information to the animal. The presence of cementum at the
cervix of the maxillary canines, also called sabers, would indicate that the
gingiva in Smilodon was attached in this region. Such an attachment would
be advantageous, providing stability and sensory input for the large tooth.
Also, gingiva at the cervix would impact the manner in which the teeth were
used. Previous study using scanning electron microscopy of dental casts was
indirect. The purpose of this study was to conﬁrm by direct methods the
presence of cementum at the cervix of Smilodon californicus sabers. Parts
of three Smilodon californicus sabers were sectioned and examined with
light and scanning electron microscopy (EDS). In addition, percent weight
of calcium and phosphorus was measured in enamel, dentin, and cementum
using electron dispersive spectroscopy. Cementum was identiﬁed in the
cervical region of each saber. Spectroscopy conﬁrmed that the tissue is
calciﬁed and the mineral is hydroxyapatite. Percent calcium and percent
phosphorus of individual tissues were highly variable between specimens.
However, the ratios of calcium to phosphorus were not signiﬁcantly differ-
ent from the hydroxyapatite standard. In the future, bite models will have
to take the presence of soft tissues into account. ©2005 Wiley-Liss, Inc.
Key words: Smilodon; saber; cementum; gingiva; maxillary
The canine teeth of modern cats consist of a core of
dentin surrounding the pulp cavity. The dentin of the
crown is covered with enamel and the root is covered with
cementum. The tooth is held in the socket by ﬁbers of the
periodontal ligament, which extend from cementum to
alveolar bone or from cementum to gingiva. The junction
of the crown and root is fairly linear with a very slight
curvature on the labial and lingual aspects toward the
crown and away from the root. This line is called the
cementoenamel junction (CEJ). Cementum may overlap
the enamel for a few millimeters but no further. The
gingiva is attached to the cementum along the CEJ by a
specialized epithelium called the attachment epithelium.
This epithelium produces and maintains its attachment,
effectively separating the oral environment from the un-
derlying connective tissues and thereby protecting them
from bacterial invasion and damage. The cementum at the
CEJ is acellular, while the cementum apically on the root
is cellular (Hillson, 1986; Orsini and Hennet, 1992; Har-
vey and Emily, 1993; Schroeder and Listgarten, 1997; Ten
Cate, 1998). The alveolar bone of the tooth socket and the
periodontal ligament that holds the tooth in the socket lie
inferior to the epithelial attachment of the gingiva. The
region of the tooth just apical to the CEJ is called the neck
or cervix of the tooth
The large saber-like maxillary canines of Smilodon (sa-
ber-toothed cat) are similar in construction to maxillary
canines of modern cats. The core of dentin in the crown is
covered with enamel and the root is thought to be covered
with cementum (Riviere and Wheeler, 2001). The CEJ,
however, is not linear but deeply curved toward the crown
on the labial and lingual aspects (Figs. 1 and 2). The saber
is embedded in alveolar bone up to a line drawn from just
*Correspondence to: Holliston L. Riviere, School of Dentistry,
Oregon Health and Sciences University, 611 S.W. Campus Drive,
Portland, OR 97201. Fax: 61-7-33654522.
Received 29 November 2004; Accepted 8 February 2005
Published online 7 June 2005 in Wiley InterScience
THE ANATOMICAL RECORD PART A 285A:634 – 642 (2005)
©2005 WILEY-LISS, INC.
This article was prepared by a group
consisting of both United States government employees and non-United
States government employees, and as such is subject to 17 U.S.C. Sec. 105.
apical to the beginning of the enamel on the posterior
surface to the anterior surface (Fig. 3) (Merriam and
Stock, 1932; Akersten, 1985). A substantial amount of root
lies between the alveolar crest and the cementoenamel
junction. If the attachment of the gingiva of Smilodon is
similar to extant cats, it should have covered the cervix of
the tooth from the alveolar crest to the CEJ (Fig. 4A). If
the gingiva was attached at the CEJ, then the functional
surface, or clinical crown, of the tooth would be limited by
that attachment. If the gingiva was attached inferior to
the CEJ close to the alveolar crest, then a portion of the
saber root would be exposed to the oral environment and
the clinical crown of the tooth would be greater (Fig. 4B).
Exposure of the root of a canine tooth to the oral environ-
ment in any extant mammal is considered pathologic.
Such exposure occurs rarely in mammals in the wild and
is usually in response to gingival recession, injury, or
inﬂammation (Miles and Grigson, 1990).
Cementum is the softest of the dental tissues. Exposed
root cementum undergoes many changes, including bacte-
rial contamination and invasion (Armatige and Christie,
1973a, 1973b; Bosshardt and Selvig, 1997). When located
on the crowns of teeth, cementum is quickly worn away
from functional surfaces, surviving only in lofts and folds
of enamel not directly exposed to wear (Hillson, 1986). If
Smilodon cementum is similar to modern cementum, ex-
posed cementum on the smooth surface of the root of the
sabers would be worn away by the abrasive action of food
and the tongue. The subsequently exposed dentin would
then be vulnerable to disease and wear.
Fig. 2. Illustration of four views of a right Smilodon saber. From left
to right, labial view, anterior view, lingual view, and posterior view.
Fig. 1. Labial and lingual views of an intact Smilodon saber. The
enamel-covered crown is to the left. Arrows indicate the cementoenamel
Fig. 3. Illustration of the relationship of the alveolar margin to the
cementoenamel junction of a Smilodon saber. The root is illustrated in
black. A large portion of the root lies between the alveolar margin and the
If the gingiva was attached to the CEJ and covered the
cervix of the tooth, there should be cementum on that
surface extending up to the CEJ and possibly a short
distance onto the enamel. Previous study indicated that
cementum was indeed covering the area but the method
was indirect, using casts of the surface. Scanning EM of
the casts disclosed a material that appeared cementum-
like compared to similar preparations of bobcat and cou-
gar cementum but was probably altered by time and pres-
ervation methods. The purpose of this investigation was to
demonstrate by direct methods that cementum is present
on the cervix of the root of Smilodon sabers using sections
of sabers, light, and scanning electron microscopy and
electron dispersive spectroscopy (EDS).
MATERIALS AND METHODS
Three partial Smilodon californicus sabers were ob-
tained from the George Page Museum of Discoveries (Los
Angeles, CA; specimens 52094, 52093, and 52800). Each
fragment consisted of some enamel, a relatively complete
CEJ, and some root (Fig. 5). Each saber was soaked in
several changes of acetone to remove all preservative ma-
terials. As preservative was removed, the specimens be-
gan to fragment. The location of the pieces was docu-
mented and the pieces were numbered. Fragments from
the region of the CEJ of each saber were embedded in
Bio-Plastic (Ward’s Natural Science, San Luis Obispo, CA)
according to instructions supplied with the plastic. Frag-
ments from the distal portion of the root of each specimen
were similarly prepared. Specimens varied in size and
shape. Longitudinal sections approximately 300 thick
were cut using a Buehler Isomet saw and diamond wafer-
ing blade (Buehler, 15 HC Series 10.3 cm ⫻0.3 mm).
Sections were dehydrated in successive changes of 70%,
95%, and 100% ethyl alcohol and placed between two glass
slides held together with clothes pins. They were allowed
to dry on a slide drying table overnight. The dried sections
were then mounted on clean slides and coverslipped using
Permount mounting medium (Fischer Scientiﬁc). Slides
were viewed with a Zeiss Photoscope and a Bio-Rad Radi-
ance 2100 confocal system using a Nikon E800 micro-
Fig. 4. Illustration of the extent of two possible gingival attachments.
The gingiva follows the cementoenamel junction in A. The gingiva only
covers the alveolar crest and a small portion of the root in B, leaving a
large area of the root exposed.
Fig. 5. Views of the labial and lingual aspects of the specimens used in this study. Aand Bare of
specimen 52094, Cand Dof specimen 52093, and Eand Fof specimen 52800. Arrows indicate the
cementoenamel junction. Enamel (e) and root (r) are indicated.
636 RIVIERE AND WHEELER
scope. Separate thicker sections (500 and larger) were
dehydrated in ethyl alcohol, dried, and mounted on stubs
for scanning electron microscopy (SEM). These sections
were polished and coated with carbon and viewed with
SEM (JEOL JXA-6400, Link Analytical model eXL). Cal-
cium and phosphorus spectra were measured using en-
ergy dispersive spectroscopy (electron probe microana-
lyzer, 10 kv, 5 nanoamps, 35-sec lifetime). Probes were
made of the enamel just incisal to the CEJ. Dentin and
cementum probes were made in the tissues just inferior to
the CEJ. Each probe covered an area of approximately 3–5
to a depth of approximately 1 . Spectral readings were
converted to percent weight of calcium and phosphorus
using the Quantitative Program for the microprobe and an
The fossil specimens were very difﬁcult to cut in thin
sections. Three hundred micron sections could not be cut
without crumbling. Small areas of some sections remained
intact and could be mounted. Fragments from the CEJ
refracted light from the microscope in patterns that ob-
scured the image and made photography impossible.
There appeared to be a layer of material that could be
cementum in the area of interest but the layer was not
resolvable with the standard light microscope. The speci-
mens were viewed using confocal microscopy. The CEJ
specimens were difﬁcult to focus (Fig. 6A). Reverse image
revealed a lamellar structure (Fig. 6B). The lamellae were
parallel to the dentinocemental junction. The tissue in
sections from the apical regions of roots was typical of
mammalian cellular cementum (Fig. 7). While the speci-
men was thick, it did not refract light like the acellular
specimens. Lacunae and ﬁbers were present.
Scanning Electron Microscopy
Scanning electron microscopic images revealed clearly
identiﬁable enamel and dentin at the CEJ of all fossil
specimens (Fig. 8). Also present on the dentin surface just
inferior to the CEJ was a thin layer of acellular cementum
40–45 thick (Fig. 9A and C). Cementum overlapped
enamel in specimen 52093 (Fig. 9B). Acellular cementum
in specimen 52800 was uneven in density, some areas
appearing more electron-dense than others (Fig. 9A). This
mottled appearance did not suggest any particular inter-
nal or histologic structure. No connective tissue ﬁbers
were apparent. Cementum in specimens 52093 and 52094
were very similar microscopically (Fig. 9B and C). They
both had a ﬁbrous appearance but no individual connec-
tive tissue ﬁbers could be demonstrated. There were no
lacunae. No debris and no bacteria were present in any of
Spectral analysis of the enamel, dentin, and cementum
conﬁrmed that the tissues were calciﬁed (Fig. 10). In all
three fossil specimens, enamel contained more calcium
and phosphorus by weight than dentin, and cementum
was the least calciﬁed tissue. All measurements of percent
weight of calcium and phosphorus were lower than the
standard (Table 1). The Ca/P ratios were, however, not
signiﬁcantly different from the standard Ca/P ratio (Table
2). The average Ca/P ratio for all tissues of all specimens
was 2.1 ⫾0.2 compared to the standard ratio of 2.1.
Average percent weight calcium and phosphorus of dentin
for the three specimens was not signiﬁcantly different
from the averages for enamel. This could be due to loss of
mineral over time and/or loss of organic matrix, especially
collagen, from dentin.
This study demonstrated acellular cementum at the
cervix of Smilodon sabers. This ﬁnding supports the pres-
ence of gingiva attached to the CEJ and covering the
cervix, a condition typical of modern cats. The compara-
tive morphology of teeth has been extensively described
(Dahlberg, 1968; Butler and Joysey, 1974; Kurten, 1982;
Hillson, 1986; Teaford et al., 2000). The individual tissues
are variable in structure, as is the method of attachment
to the jaw. The evolution of tooth attachment and mor-
phology of attachment among extant and fossil species has
been addressed by several investigators (Orvig, 1967;
Shellis, 1982; Osborn, 1984; Gaengler and Metzler, 1992;
Gaengler, 2000). Cementum itself can be classiﬁed accord-
ing to composition and function (Bosshardt and Selvig,
1997; Ten Cate, 1998). Acellular extrinsic ﬁber cementum
is found from the cervix to the apical third of teeth. Its
function is anchorage of the tooth in the socket. Cellular
mixed ﬁber cementum contains trapped cells and is found
in the apical third and between the roots of multirooted
Fig. 6. Confocal micrographs of the surface of a saber near the
cementoenamel junction. Dentin is visible in Abut the surface is not
resolvable. The reverse image shown in Breveals a thin surface layer,
which appears itself to be layered (arrows). Magniﬁcation, 40⫻.
Fig. 7. Confocal microscopic views of a sample of apical root dentin
with cementum. Dentin is to the right and layers of cellular cementum to
the left of the dentinocemental junction (DCJ). Lacunae, appearing as
black dots, and extrinsic ﬁbers (arrows) are visible in A. Magniﬁcation,
10⫻. A higher-power view (B) at the DCJ reveals lacunae (short arrows)
and intrinsic ﬁbers (long arrows) in the cementum (c) and dentinal tubules
in the dentin (d). Magniﬁcation, 40⫻.
638 RIVIERE AND WHEELER
teeth. Its function is adaptation of attachment to varying
conditions. Acellular aﬁbrillar cementum is found in
patches on enamel. Its function is unknown. In addition,
all placental mammalian teeth are thecodontal or sock-
eted, meaning the roots are surrounded by and attached to
alveolar bone. The periodontal attachment of teeth of the-
Fig. 8. Scanning electron micrographs of specimen 52800 (A), 52093
(B), and 52094 (C) at the dentinoenamel junction. Enamel (e) was clearly
visible on all specimens, as was dentin (d). Magniﬁcation, 500⫻.
Fig. 9. Scanning electron micrographs of cementum (c) on the sur-
face of the three specimens near the cementoenamel junction. Speci-
mens 52800 (A), 52093 (B), and 52094 (C) all have a layer of cementum
approximately 40 – 45 thick on their surface. The cementum in speci-
men 52093 overlapped the enamel (e). Cementum overlies dentin (d) in
the other specimens. Magniﬁcation, 1,000⫻.
Fig. 10. Spectrographs of the cementum, enamel, and dentin of the three specimens. The left column
represents specimen 52800. The center column represents specimen 52093. The right column represents
specimen 52094. The top row spectra are of cementum, the middle row of enamel, and the bottom row of
640 RIVIERE AND WHEELER
codont mammals can be classiﬁed into one of six groups
(Gaengler and Metzler, 1992; Gaengler, 2000): incisors
and most canines of all omnivorous, carnivorous, and her-
bivorous dentitions are single-rooted teeth with a combi-
nation of acellular aﬁbrillar cementum, acellular extrinsic
ﬁber cementum, and cellular intrinsic ﬁber cementum.
The sabers of Smilodon californicus conform to this group.
The saber is single-rooted. The cementum at the CEJ on
the enamel is acellular and may or may not be aﬁbrillar.
The cementum on the root surface at the CEJ is acellular
and presumably extrinsic ﬁber in type, since acellular
aﬁbrillar cementum is found only on enamel (Bosshardt
and Selvig, 1997; Ten Cate, 1998).
Similarly, four types of gingival epithelial attachment
are described for thecodonts. The second type describes
the epithelial attachment as being to enamel and/or to
coronal cementum when it overlaps enamel. While this
study could not demonstrate soft tissue attachment di-
rectly, the presence of acellular cementum at the cervix of
the sabers and overlapping the enamel for a short distance
is strong evidence for the location of the gingival attach-
ment in this region.
Typical cellular mixed ﬁber cementum is located on the
apical two-thirds of the root. Confocal images demonstrate
extrinsic collagen ﬁbers (Sharpy’s ﬁbers) in apical root
cementum but not in the region of the CEJ. This may be
due to the progressive destruction of the organic compo-
nent in the thinner cervical tissues. However, the type of
cementum and location of cementum observed in the
present study are consistent with the thecodont classiﬁ-
cation of Gaengler (2000) and imply an epithelial attach-
ment on the enamel or on the cementum at the CEJ.
Spectoral analysis conﬁrmed that the tissue is cemen-
tum. It is a calciﬁed tissue and consistently less calciﬁed
than the corresponding dentin or enamel. The layer is not
museum-preservative, which contains no calcium. It is not
dental plaque. It does not contain bacteria or debris char-
acteristic of that material and it is smooth-surfaced and of
constant thickness, also not characteristic of dental
plaque. It is also not matrix. There was no indication of
organic or inorganic material typical of the matrix in
which the sabers were found. The actual percent weights
of Ca and P vary between the sabers and tissues. Sabers
52093 and 52094 came from the same pit at La Brea. The
pit origin of saber 52800 is unknown. Differences in con-
ditions in the pits and age of the sabers could account for
variation in the percent weight measurements. Water con-
tent of the specimens may have varied also. In addition,
the type of apatite and crystallographic properties of den-
tin have been shown to vary from animal to animal (Sakae
et al., 1994). This may be true for cementum as well.
However, the average Ca/P ratios of the tissues were not
signiﬁcantly different from the hydroxyapatite standard,
indicating that the measurements are accurate and that
the mineral is hydroxyapatite.
The conclusion that the gingiva was attached to the CEJ
has important implications for consideration of bite mod-
els for Smilodon. A number of studies have attempted to
describe the manner in which Smilodon used the sabers
and several bite models have been proposed (Simpson,
1911; Bohlin, 1947; Akersten, 1985; Biknevicius and van
Valkenburgh, 1996). None of these, however, considered
the location or extent of oral soft tissues and the manner
in which such tissues might affect the bite. The extensive
soft tissue attachment suggested by this study would limit
the depth to which the sabers could be inserted without
injury to the gingiva. Injury would result in gingival in-
ﬂammation and recession and possible wear in the cemen-
tum and dentin on the lingual surface at the cervix as
previously described. Such wear has been observed in
approximately 1% of 838 sabers in the Page Museum
collection (data not shown). This is consistent with the
ﬁndings of Shermis (1984), who observed very low inci-
dence of periodontally induced bone resorption in Smil-
odon maxillary splanchnocrania in the La Brea collection.
Alveolar bone resorption is a universal response to peri-
odontitis resulting from gingival injury and/or inﬂamma-
tion. A low incidence of bone resorption reﬂects a low
incidence of gingival recession due to causes other than
aging. Very few sabers should show the effects of wear
resulting from gingival recession. Indeed, this is the case.
Finally, it would be of great functional advantage to
have an extensive gingiva related to such a large tooth.
Several important functions might be served. The gingival
component of the periodontal ligament would provide
greater attachment and stability for the large canine. The
gingiva also acts as a tactile organ letting the animal
know, among other things, when the tooth has reached its
maximum functional penetration (Heyeraas et al., 1993;
Jacobs and von Steenberghe, 1994; Kondo et al., 1995;
Sugaya et al., 1995; Fristad, 1997; Shroeder and Listgar-
ten, 1997). Additionally, the gingiva and periodontium
may be involved in jaw-closing reﬂexes (Eriksson et al.,
1998, 2000; Louca et al., 1998; Zafar et al., 2000). Smil-
odon, therefore, had better stability and a vast array of
sensory input to help them use the sabers without dam-
aging them. Hopefully, consideration of the location and
extent of the gingiva in these animals will lead to a deﬁn-
itive bite model and a better understanding of how these
animals utilized those large and impressive canine teeth.
TABLE 1. Percent by weight of calcium and
phosphorus in enamel, dentin, and cementum as
determined by electron dispersive spectroscopy and
Ca/P ratios for cementum, dentin, and enamel
in each specimen
Sample WT % P WT % Ca Ca/P
52093 cementum 2.0 4.7 2.4
52093 dentin 8.5 17.5 2.1
52093 enamel 9.9 19.7 2.0
52094 cementum 3.7 8.1 2.2
52094 dentin 11.5 23.2 2.0
52094 enamel 13.4 26.0 1.9
52800 cementum 5.5 12.6 2.3
52800 dentin 11.2 25.8 2.3
52800 enamel 12.7 26.9 2.1
Apatite standard 18.6 39.6 2.1
TABLE 2. Average percent weight phosphorus and
calcium in cementum, dentin and enamel from the
three specimens and average calcium
Tissue AV WT % P AV WT % Ca AV Ca/P
Cementum 3.7 ⫾1.7 8.5 ⫾3.9 2.3 ⫾.1
Dentin 10.4 ⫾1.6 22.2 ⫾4.2 2.1 ⫾.2
Enamel 12.0 ⫾1.8 24.2 ⫾3.9 2.0 ⫾.2
Apatite standard 18.6 39.6 2.1
The authors thank the George Page Museum of Discov-
eries for their generous donation of three sabers for use in
this study. They also thank Mr. Jerome Adey for help with
the scanning electron microscopy and for the EDS mea-
surements and analysis, as well as Dr. Michael Danilchik
for the confocal microscope photographs.
Akersten WA. 1985. Canine function in Smilodon (Mammalia;
Felidae; Machairodontinae): Natural History Museum of Los Ange-
les County. Contributions Sci 356:–1–22.
Armitage GC, Christie TM. 1973a. Structural changes in exposed
human cementum: I, light microscopic observations. J Periodont
Armitage GC, Christie TM. 1973b. Structural changes in exposed
human cementum: II, electron microscopic observations. J Peri-
odont Res 8:356 –365.
Biknevicius AR, Van Valkenburgh B. 1996. Design for killing: cranio-
dental adaptations of predators. In: Gittleman JL, editor. Carnivore
behavior, ecology, and evolution, vol 2. Ithica, NY: Cornell Univer-
sity Press. p 393– 428.
Bohlin B. 1947. The saber-toothed tigers once more. Bull Geol Inst
Bosshardt DD, Selvig KA. 1997. Dental cementum: the dynamic tis-
sue covering of the root. Periodontology 13:41–75.
Bovard JF. 1907. Notes on quaternary felidae from California. Univ
Calif Publ Bull Dept Geol 5:155–170.
Butler PM, Joysey KA, editors. 1978. Development, function, and
evolution of teeth. New York: Academic Press.
Dahlberg AA. 1971. Dental morphology and evolution. Chicago: Uni-
versity of Chicago Press.
Eriksson PO, Zafar H, Nordh E. 1998. Concomitant mandibular and
head-neck movements during jaw opening-closing in man. J Oral
Rehab 25:859 – 870.
Eriksson PO, Haggman-Henrikson B, Nordh E, Zafar H. 2000. Co-
ordinated mandibular and head-neck movements during rhythmic
jaw activities in man. J Dent Res 79:1378 –1384.
Fristad I. 1997. Dental innervation: functions and plasticity after
peripheral injury. Acta Odontol Scand 55:236 –254.
Gaengler P, Metzler E. 1992. The periodontal differentiation in the
phylogeny of teeth: an overview. J Periodontal Res 27:214 –225.
Gaengler P. 2000. Evolution of tooth attachment to tetrapods. In:
Teaford MF, Smith MM, Ferguson MWJ, editors. Development,
function and evolution of teeth. New York: Cambridge University
Harvey CE, Emily PP. 1993. Small animal dentistry. St. Louis:
Heyeraas KJ, Kvinnsland I, Byers MR, Jacobsen EB. 1993. Nerve
ﬁbers immunoreactive to protein gene product 9.5, calcitonin gene-
related peptide, substance P and neuropeptide Y in the dental pulp,
periodontal ligament and gingiva in cats. Acta Odontol Scand 51:
Hillson S. 1986. Teeth. New York: Cambridge University Press.
Jacobs R, van Steenberghe D Jr. 1994. Role of periodontal ligament
receptors in the tactile function of teeth: a review. J Periodont Res
Kondo T, Kido MA, Kiyoshima T, Yamaza T, Tanaka T. 1995. An
immunohistochemical and monastral blue-vascular labeling study
on the involvement of capsaicin-sensitive sensory innervation of the
junctional epithelium in neurogenic plasma extravasation in the rat
gingiva. Archs oral Biol 40:931–940.
Kurten B, editor. 1982. Teeth: form function and evolution. New York:
Columbia University Press.
Louca C, Vidgeon SD, Cadden SW, Linden RWA. 1998. The role of
gingival mechanoreceptors in the reﬂex control of human jaw-clos-
ing muscles. Arch Oral Biol 43:55– 63.
Merriam JC, Stock C. 1932. The felidae of Rancho La Brea. Carnegie
Inst Washington Pub 422:1–232.
Miles AEW, Grigson C, editors. 1990. Colyer’s variations and diseases
of the teeth of animals. New York: Cambridge University Press.
Orsini P, Hennet P. 1992. Anatomy of the mouth and teeth of the cat.
Vet Clin North Am Small Animal Practice 22:–1265–1273.
Orvig T. 1967. Phylogeny of tooth tissues: evolution of some calciﬁed
tissues in early vertebrates. In: Miles AEW, editor. Structural and
chemical organization of teeth, vol. 1. New York: Academic Press. p
Osborn JW. 1984. From reptile to mammal: evolutionary consider-
ations of the dentition with emphasis on tooth attachment. Symp
Zool Soc London 52:549 –574.
Riviere HL, Wheeler HT. 2001. Attachment of the periodontal liga-
ment and gingiva in Smilodon californicus from Rancho La Brea. J
Vert Paleo 21:94A.
Sakea T, Mishima H, Takahashi-Kunuki Y, Suzuki K, Kozawa Y.
1994. Dentine apatites from teeth of different animals: crystallo-
graphic properties. Bull Inst Oceanograph (Monaco) 14:215–219.
Schroeder HE, Listgarten MA. 1997. The gingival tissues: the archi-
tecture of periodontal protection. Periodontology 13:91–120.
Shellis RP. 1982. Comparative anatomy of tooth attachment. In:
Berkovitz BKB, Moxham BJ, Newman HN, editors. The periodontal
ligament in health and disease. Oxford: Pergamon Press. p 3–24.
Shermis S. 1984. Alveolar osteitis and other oral diseases in Smilodon
californicus. OSSA 12:187–197.
Simpson GG. 1941. The function of saber-like canines in carnivorous
mammals. Am Museum Novitates 1130:2–12.
Sugaya A, Chudler EH, Byers MR. 1995. Axonal transport of ﬂuores-
cent carbocyanine dyes allows mapping of peripheral nerve territo-
ries in gingiva. J Periodontol 66:817– 821.
Teaford MF, Smith MM, Ferguson MWJ, editors. 2000. Development,
function and evolution of teeth. New York: Cambridge University
Ten Cate R. 1998. Oral histology, development, structure, and func-
tion, 5th ed. St. Louis, MO: Mosby.
Zafar H, Nordh E, Eriksson PO. 2000. Temporal coordination between
mandibular and head-neck movements during jaw opening tasks in
man. Arch Oral Biol 45:675– 682.
642 RIVIERE AND WHEELER