JOURNAL OF MORPHOLOGY 256:146 –159 (2003)

Dentition and Tooth Replacement Pattern in Chalcides (Squamata; Scincidae) Sidney Delgado, Tiphaine Davit-Beal, and Jean-Yves Sire* Equipe “Evolution et de´veloppement du squelette dermique,” UMR 8570, Universite´ Paris 7, Paris cedex 05, France ABSTRACT This study was undertaken as a prerequisite to investigations on tooth differentiation in a squamate, the Canarian scincid Chalcides. Our main goal was to determine whether the pattern of tooth replacement, known to be regular in lizards, could be helpful to predict accurately any stage of tooth development. A growth series of 20 laboratory-reared specimens, aged from 0.5 month after birth to about 6 years, was used. The dentition (functional and replacement teeth) was studied from radiographs of jaw quadrants. The number of tooth positions, the tooth number in relation to age and to seasons, and the size of the replacement teeth were recorded. In Chalcides, a single row of pleurodont functional teeth lies at the labial margin of the dentary, premaxillary, and maxillary. Whatever the age of the specimens, 16 tooth positions were recorded, on average, in each quadrant, suggesting that positions are maintained throughout life. Replacement teeth were numerous whatever the age and season, while the number of functional teeth was subject to variation. Symmetry of tooth development was evaluated by comparing teeth two by two from the opposite side in the four jaw quadrants of several specimens. Although the relative size of some replacement teeth fitted perfectly, the symmetry criterion was not reliable to predict the developmental stage of the opposite tooth, whether the pair of teeth compared was left–right or upper–lower. The best fit was found when comparing the size of successive replacement teeth from the front to the back of the jaw. Every replacement tooth that is 40 – 80% of its definitive size is followed, in the next position on the arcade, by a tooth that is, on average, 20% less developed. Considering teeth in alternate positions (even and odd series), each replacement tooth was a little more developed than the previous, more anterior, one (0.5–20% when the teeth are from 10 – 40% of their final size). The latter pattern showed that tooth replacement occurred in alternate positions from back to front, forming more or less regular rows (i.e., “Zahnreihen”). In Chalcides, the developmental stage of a replacement tooth in a position p can be accurately predicted provided the developmental stage of the replacement tooth in position p-1 or, to a lesser degree, in position p-2 is known. This finding will be particularly helpful when starting our structural and ultrastructural studies of tooth differentiation in this lizard. J. Morphol. 256:146 –159, 2003. © 2003 Wiley-Liss, Inc. KEY WORDS: Scincidae; Chalcides; dentition; replacement pattern

This study is part of an extensive research program devoted to the development and evolution of © 2003 WILEY-LISS, INC.

the dermal skeleton, and especially the dental tissues, which show a great diversity in vertebrates (see review in Huysseune and Sire, 1998). Currently, our interest is to trace the evolutionary history of the highly mineralized superficial tissue, enamel, and especially of its major protein, amelogenin (Girondot and Sire, 1998; Delgado et al., 2001). While mammalian amelogenesis is well known, information is lacking in other taxa (e.g., Carlson, 1990; Sander, 2001). We have chosen to start our program by studying amelogenin expression during tooth differentiation in the lineage closest to mammals, i.e., the reptiles. Our main goal is to look for temporal and/or spatial differences that could account for the structural differences in the enamel (prismatic enamel in mammals vs. enamel without prisms in most reptiles). However, a prerequisite to such a work is to know the dentition pattern in the reptilian species chosen as a model. Indeed, it is particularly advisable to know where and when the appropriate stages of enamel formation can be found along the jaws in recently born, juveniles, and adults because amelogenin expression can only be studied in differentiating teeth (enamel proteins being entirely degraded by proteases during the enamel maturation process). Among reptiles, only crocodiles and lepidosaurs (Sphenodon, snakes, and lizards) possess teeth. It is difficult to obtain growth series for reptilian species either in the wild or in the laboratory. This is why we have taken advantage of our successful breeding of two species of the scincid genus Chalcides (a lizard), C. sexlineatus and C. viridanus, for many years. In contrast to mammals that have only two sets of teeth (diphyodonty), most reptiles replace their teeth throughout life (polyphyodonty). Therefore, all stages of tooth development (and, consequently, of enamel formation) are available in juvenile and

*Correspondence to: J-Y Sire, Equipe “Evolution et de´veloppement du squelette dermique,” UMR 8570, Universite´ Paris 7, Case 7077, 2, place Jussieu, 75251 Paris cedex 05, France. E-mail: [email protected]

DOI: 10.1002/jmor.10080

DENTITION AND TOOTH REPLACEMENT IN CHALCIDES

147

TABLE 1. Age, day of death, and various measurements of the 20 individuals studied No. and species

Age in months after birth

Day of death

Snout / vent length (mm)

Tooth positions in the lower jaw. Left/right

Length of the left tooth row (mm)

Mean tooth interval in mm (range)

1. C.v. 2. C.s. 3. C.v. 4. C.v. 5. C.s. 6. C.v. 7. C.s. 8. C.v. 9. C.v. 10. C.v. 11. C.v. 12. C.s. 13. C.v. 14. C.v. 15. C.s. 16. C.s. 17. C.s. 18. C.s. 19. C.v. 20. C.s.

0.5 1 1.5 4.5 7 7.5 8 10.5 11 16 17 18 20 29 31 36* 46* 58* 59* 70*

13 Aug. 13 Jul. 28 Aug. 7 Nov. 10 Jan. 14 Feb. 20 Apr. 26 May 22 May 7 Nov. 28 Nov. 6 Dec. 14 Feb. 7 Nov. 25 Jan. 20 Apr. 7 May 4 May 28 May 10 May

38 42 40 50 42 42 31 60 55 70 51 51 78 70 49 70 80 85 82 90

15/15 16/16 17/17 14/15 16/15 15/16 16/15 17/16 16/18 16/17 16/17 16/17 17/18 17/17 16/16 16/17 16/17 15/16 18/18 16/17

2.76 3.40 5.14 3.53 3.79 2.95 2.83 3.98 3.79 4.50 3.73 3.79 4.90 4.30 3.53 5.07 4.80 5.72 4.75 7.00

0.20 (0.17–0.20) 0.24 (0.17–0.30) 0.24 (0.20–0.40) 0.24 (0.20–0.30) 0.25 (0.17–0.27) 0.23 (0.17–0.23) 0.21 (0.17–0.23) 0.24 (0.20–0.33) 0.22 (0.20–0.37) 0.29 (0.23–0.37) 0.28 (0.23–0.33) 0.27 (0.20–0.37) 0.29 (0.23–0.37) 0.28 (0.23–0.47) 0.27 (0.23–0.37) 0.31 (0.27–0.37) 0.30 (0.28–0.37) 0.31 (0.27–0.47) 0.30 (0.27–0.43) 0.46 (0.33–0.53)

C.s. ⫽ Chalcides sexlineatus; C.v. ⫽ Chalcides viridanus. *Specimens in which the age was estimated using the skeletochronology method (see text, Material and Methods).

adult Chalcides, so it is not necessary to look for tooth germs in the embryo. Moreover, teeth of scincids, as those of most squamates, are regularly replaced in waves, which sweep through alternate tooth positions (e.g., Edmund, 1960, 1969). Such a regular pattern implies that it could be possible, in principle, to predict the developmental stage of any tooth in the jaw of a Chalcides specimen, provided the replacement pattern was well known in the species. In the family Scincidae the pattern of tooth replacement has been briefly reported by Edmund (1960, 1969) for several genera, but not Chalcides, in which the dentition (i.e., the number of functional and replacement teeth) as well as the replacement pattern are still unknown. The present study has been undertaken to answer the following question: can the tooth replacement pattern in Chalcides help us to predict any stage of tooth differentiation? To ensure an accurate prediction we need to know 1) whether the tooth replacement varies with age and/or with seasons, 2) whether there is a symmetry in tooth replacement (left/right sides and/or upper/lower jaws), and 3) whether teeth are replaced regularly in waves as described in the literature (e.g., Edmund, 1960, 1969; Cooper, 1965, 1966; Osborn, 1970, 1972, 1973, 1974, 1978. With these questions in mind, we have studied the dentition and the pattern of tooth replacement in a growth series of 20 Chalcides by means of radiography. MATERIALS AND METHODS Biological Material Two species of Canarian skinks, Chalcides sexlineatus Steindachner, from Gran Canaria, and C. viridanus (Gravenhorst),

from Tenerife, have been bred in our laboratory since the spring of 1996, when 30 adults were caught in the wild and brought to the laboratory. Both Chalcides species have been bred in controlled conditions, i.e., a “winter” period of 2.5 months, from the end of November to mid-February (but with room temperature never falling below 10°C, owing to heat sources switching on at 10°C) and a “temperate” period (e.g., room temperature) for the rest of the year. The photoperiod was that of the day (windows not occulted). Males and females of each species were distributed in several tanks (80 cm long / 40 cm wide, with a sandy bottom) and fed (except during the “winter” period) twice a week with insect larvae (mostly maggots). Canarian skinks are viviparous. Two to three baby skinks per female were obtained every year since the summer of 1996, from the second half of June to the end of July. Soon after birth the baby skinks were placed into small tanks (40/20 cm) and fed with small maggots. A total of 20 specimens (11 Chalcides viridanus and 9 C. sexlineatus), ranging from 0.5 month after birth to 6 years of age were used (Table 1). All lizards were measured (from snout to vent), decapitated, and the jaws dissected. The lower jaw was sectioned through the symphysis and the upper jaw through the area of fusion of the premaxillae. For each specimen the four quadrants were fixed and conserved in 70% ethanol.

Methods Age estimation. Because most specimens were born in the laboratory, their age at death was known. However, five adults belonging to the initial wild-caught sample were of unknown age (i.e., Nos. 16 –20; Table 1). The age of these individuals was estimated using skeletochronology (Castanet et al., 1993). The left femora were dissected and the soft tissues removed. The femora were decalcified for 8 h in 5% diluted nitric acid. Ten to 15 ␮m-thick frozen cross sections (Zeiss Cryomat), from the middiaphyseal level of the femur, were stained with Ehrlich’s hematoxylin. The lines of arrested growth (LAG), which appear colored darker than the rest of the bone matrix, were counted. LAG reflects the arrest of bone matrix deposition during the winter period (i.e., 1 LAG ⫽ 1 winter). In Gran Canaria and Tenerife Chalcides are born in June. The date of death of the five specimens being known, age was accurately estimated in months after birth and included in the data (Table 1).

148

S. DELGADO ET AL.

Radiographs. Radiography was chosen because 1) it reveals all forming teeth provided their matrix is mineralized (tooth mineralization starts early, when the germs are less than 10% of their definitive height); and 2) the tooth positions are not modified. In dry material observed with the binocular microscope, small germs are often masked by the surrounding soft tissues. The position of the teeth with respect to each other can also vary depending on the angle of observation. Moreover, cleaning the jaws with commercial bleach to remove the soft tissues (e.g., for a scanning electron microscopical study) in general results in the loss of all replacement teeth (they are not yet attached to the bone support) and, sometimes, of functional teeth that are highly resorbed. For each individual used in this study, the four quadrants were dehydrated in a graded series of ethanol and dried. The quadrants were placed, lingual side up, on a Kodak SO-343 highresolution film, then exposed to X-rays (10 mA, 25 kV, 30 cm from the source, exposure time from 5– 40 min, depending on the size of the jaw). The radiographs were directly observed in an Olympus binocular-dissecting microscope using transmitted light. Drawings were made using a camera lucida. Scanning electron microscopy (SEM). SEM observations were useful for the present study to get information on the shape and orientation of the functional teeth. Some of the lower jaw quadrants that were previously radiographed were rehydrated and immersed in 10% sodium hypochlorite (i.e., commercial bleach) at room temperature for 20 – 40 min (depending on the size of the jaw). The soft tissues were delicately removed using thin forceps and small brushes. The jaw quadrants were then dehydrated in a graded series of ethanol, dried, glued on a copper support, coated with a thin layer of gold/palladium, and observed in a JEOL SEM 35 operating at 25 kV. Counts and measurements. All counts and measurements were taken from the radiographs and checked against original material when necessary. Lengths and heights were measured to the nearest 50 ␮m. The number of tooth positions was determined in each quadrant, starting anteriorly and including the positions that were lacking teeth. All functional and replacement teeth were counted. The following measurements were also taken: length of the tooth row in each quadrant (from the tip of the most anterior to that of the most posterior tooth), average length of tooth intervals (from the tip of two successive teeth along the tooth row) and the height of every functional and replacement tooth. The height of a functional tooth was taken by measuring the length of a line running perpendicular from the tooth tip to a line circumscribing the base of the tooth at the lingual side; the height of a replacement tooth was similarly measured from the tip to the most basal visible mineralized part. Replacement pattern Tooth size. The size of each replacement tooth was expressed as a percentage of its probable definitive height, i.e., either the height of the functional tooth it was replacing or, when the latter was lacking, of the closest functional tooth. Relative values were chosen to avoid errors due to the different size of the specimens and to the variation of the tooth height along the row. For each individual, one can reasonably assume that differences in relative size of replacement teeth correspond to various stages of tooth differentiation. The relative values were grouped in 19 indices of tooth “development” from 0 (⫽ replacement tooth not visible on the radiograph) to 1 (⫽ functional tooth). A diagram was obtained for each lower jaw quadrant in the 20 specimens (i.e., 40 quadrants). The relative size was also expressed as a function of age and season to look for possible effects of these factors on the number and size of the replacement teeth. In each case, the number of replacement teeth was recalculated to obtain an equal number of individuals (five) per season and in each of the four age classes (1, 2, 3, and ⬎3 years) chosen. Dentition pattern. Three stages of tooth differentiation were arbitrarily defined to study the dentition symmetry and the position of the replacement teeth in relation to their functional predecessors. To this end, the total number of replacement teeth in the 40 lower jaw quadrants, i.e., 560, was divided by 3 (⫽ 187). Three size classes (small, medium, and large) totaling a number of teeth close to 187 (i.e., 188, 173, and 199, respectively) were

then defined from the relative size of all replacement teeth as follows: small ⫽ replacement teeth, the height of which did not exceed 25% of the height of the functional teeth; medium ⫽ replacement teeth from 26 – 40% of the functional teeth; large ⫽ replacement teeth larger than 40% but not functional. A fourth class comprised all functional teeth, including the teeth that were subjected to resorption because they are still functional during most of the resorption process. The position of all functional and replacement teeth was reported in schematic drawings of the upper and lower jaw quadrants of the 20 specimens. Both the radiographs and the original material were used to define accurately the positions.

RESULTS In both Chalcides species, the dentition, shape, and number of teeth and the pattern of tooth replacement were found to be similar: SEM observations and comparisons of the tooth replacement patterns suggested that intra- and interspecific variations were similar. For these reasons the results obtained on both species were used together and for convenience, we will refer to both species by using the name Chalcides only. Dentition In Chalcides the dentition consists of a row of functional teeth located on the labial margin of the dentary, premaxilla, and maxilla, and of a row of replacement teeth on the lingual side of these bones (Fig. 1a,b). The functional teeth are located in the dental groove, a depression between the labial and the lingual wall of the jaw bone (Fig. 1c). The teeth have a pleurodont mode of attachment to the jaw, i.e., they are ankylosed to the inner side of the labial wall of the jaw, which is higher than the lingual wall, leaving the lingual side of the tooth base exposed. In each dental arcade, an increase of tooth size is observed along the row. The teeth located at the rear end (except the most posterior one) are larger (wider and a little taller) than the anterior ones, but the shape of all teeth is roughly similar (i.e., they are homodont) (Fig. 1c). The anterior teeth in positions 1–5 are tilted anteriorly while those of the following positions are roughly vertical. All crowns are recurved lingually. The dentition (functional and replacement teeth) was found to be similar in the upper and lower jaws of each specimen studied. For convenience, we chose to work on the two quadrants of the lower jaw. The upper jaw quadrants were used as comparative material to study the dentition symmetry only. Number of functional and replacement teeth. Fifteen days after birth, 15 tooth positions were found in the two lower jaw quadrants (Table 1). In the 19 other specimens from 1–70 months old, the number of tooth positions ranged from 14 –18 (mean: 16.0 on the left, 16.5 on the right). In the upper jaws (10 quadrants examined only), there were nine teeth on the premaxilla and either 13 or 14 tooth positions on the left and right maxillae. The odd number of

DENTITION AND TOOTH REPLACEMENT IN CHALCIDES

Fig. 1. Lingual view of the left lower jaw of the 70-month-old Chalcides. a: Radiograph. b: Drawing of the radiograph to highlight the replacement teeth (in black). c: SEM picture of the same jaw showing the size and orientation of the functional teeth (all replacement teeth were lost after bleach treatment). Bars ⫽ 1 mm.

premaxillary teeth is due to the presence of a single tooth in the median region. This tooth is located at the position that was occupied at birth by the egg

149

tooth (this true tooth being itself derived from one of the two first germs forming at this position in the embryo). Therefore, except for this tooth, there were 17 or 18 teeth on the two upper quadrants (mean: 17.8 on the left, 17.6 on the right). In the 0.5-month-old specimen, the length of the tooth row on the left lower jaw was 2.76 mm. In the 19 other individuals, the length of the tooth row increased from 3.4 in the 1-month-old to 7.0 mm in the 70-month old specimen. During the same time the mean tooth interval increased from 0.20 to 0.46 mm, respectively (Table 1). The teeth were counted in the radiographs of the lower jaws (40 quadrants). A total of 560 replacement and 430 functional teeth were recorded (Fig. 2). In 17 out of the 20 specimens studied more replacement than functional teeth were observed. This difference is related more to variations in the number of functional than of replacement teeth. Indeed, in the growth series the number of replacement teeth was always above 22 for the two lower jaw quadrants together (range: 23–38; mean: 28) whatever the age of the specimens and the seasons, even in the winter period. This high number suggests that there is probably one replacement tooth per tooth position when we take into account that small tooth germs cannot be seen on radiographs. In contrast, the number of functional teeth was smaller (range: 11–29; mean: 21.5) and it varied much between the individuals. The lower value, 11 functional teeth (i.e., five or six functional teeth on the left and right quadrants), was observed in the 7.5month-old specimen, which died at the end of its first winter (vs. more than 20 functional teeth in the lizards which died the next summer) (Fig. 2). How-

Fig. 2. Number of replacement (_) and functional (䊐) teeth in the lower jaw (left ⫹ right quadrants) of the 20 Chalcides studied ranked in order of increasing age (from 0.5–70-month-old). The month of death is indicated in brackets.

150

S. DELGADO ET AL.

ever, the number of individuals was insufficient to test if this difference between seasons was significant. The relative size of the 560 replacement teeth counted in the 20 lower jaws is not regularly distributed over the 19 size classes chosen (Fig. 3a). Indeed, 78.8% of the replacement teeth have a relative size below 0.51 and, among them, 34.0% have a size between 0.21 and 0.35. A similar distribution is obtained when the relative size is plotted against age (Fig. 3b). In the four age classes (1-, 2-, 3-, and ⱖ4-year-old) the replacement teeth show the same size distribution (77.3, 80.4, 93.5, and 80.9% of the replacement teeth below 0.51, respectively) with a large number of teeth in the size classes 0.21– 0.35 (31.2, 44.6, 42.3, and 31.7%, respectively). Large tooth germs ⱖ0.51 are more numerous in 1- and 2-year-old specimens (65%) than in 3-year-old and older specimens (35%). When the tooth size is plotted against the seasons a different distribution is observed due to the larger number of replacement teeth in summer than in other seasons (Fig. 3c). Above a relative size of 0.41, summer accounts for 45.5% of the number of replacement teeth vs. 13.4, 18.8 and 22.3% for fall, spring, and winter, respectively. In contrast, small tooth germs (with a relative size below 0.41) are more numerous in winter (28.0%) and in fall (27.4%) than in spring (23.0%) and in summer (21.6%). Dentition symmetry. Whether or not tooth replacement was symmetrical was studied by comparing the four quadrants two by two (left/right, upper left / lower left, etc.) in several specimens. Three specimens (0.5-, 10.5-, and 36-month-old) were chosen as representative of the growth series and the results of our analysis are shown in Figures 4 and 5. First, we compared the relative size of the teeth in each position by reporting them in diagrams (Fig. 4). The three specimens presented herein have 16 teeth on each quadrant; 64 comparisons of two opposite teeth in the same position are possible for each specimen (left/right upper jaw; left/right lower jaw; left upper / left lower jaw; right upper / right lower jaw). Thus, a total of 192 comparisons have been done for the three specimens: a symmetrical replacement tooth was absent in 39 cases (20.3%); among the 153 other possible comparisons, both teeth showed a nearly identical stage of tooth development (i.e., perfect symmetry) in 25 cases only (13.0%). For the 128 other comparisons, no strict symmetry was observed whichever sides of the jaws were compared. However, in a number of positions the relative size deviation in two opposite teeth was less than 10% (a deviation can be positive or negative). For instance, in the 0.5-month-old specimen, 14 out of the 62 comparisons (22.6%) lacked opposite replacement teeth, 10 showed a good symmetry (16.1%), 27 (43.5%) showed a size deviation that did not exceed 10%, and 11 (17.8%) showed a size deviation exceeding 10% (Fig. 4). In the three specimens, the best fit (12 out of 49 comparisons [24.5%] with a good sym-

metry) was observed when comparing left and right upper jaws (Fig. 4B). In contrast to the replacement teeth, the presence or absence of functional teeth showed a good symmetry, from 61.0 – 87.0%, depending on the specimen and on the quadrant studied (Fig. 4). However, the amount of resorption was not taken into consideration for these teeth. Second, in the same specimens we compared the size of opposite replacement teeth in each position by using schematic representations of the dentition patterns in the four quadrants (as exemplified in Fig. 5). In the 25 cases studied, although the replacement teeth were grouped in three size categories only (small, medium, and large; see Materials and Methods), no symmetry in tooth development appeared whatever the side of the jaw, the tooth position, and the age of the individuals. Replacement process. Knowledge of the precise location of the replacement teeth with regard to the functional teeth is useful to interpret further serial sections. The dentition pattern has been schematically drawn in 40 lower jaw quadrants, as for the three specimens shown in Figure 5. In Chalcides, tooth replacement occurs in a disto-lingual position in all specimens studied. However, the position of the tooth germs with regard to the teeth they will replace and, as a consequence, the resorption they provoke, differs as a function of age (Fig. 5). In the 0.5-month old specimen, the tooth germs appear slightly posterior and at some distance from the functional teeth they will replace, but they are located at the same transverse plane with regard to the bottom of the jaw groove (Fig. 5A). As they grow, the replacement teeth remain aside and come progressively closer to the functional teeth. In the 10.5and 36-month-old specimens tooth germs also appear posterior to the functional teeth but through the growth series they come up from an ever deeper level in the dental groove and they appear ever closer to the functional teeth as compared to young individuals, especially in old specimens (Fig. 5C). As the replacement teeth increase their size they rapidly come in close contact with the base of the functional tooth. In young specimens, resorption begins at the disto-lingual side of the tooth base and proceeds both laterally and upwards. While growing, the mesio-labial side of the replacement tooth partially enters the pulp cavity of its functional predecessor, while the resorption proceeds towards the tip of the latter. Meanwhile, the mesio-labial surface of the functional tooth remains unaffected and, although severely resorbed, it is still functional. Later, the resorption of the mesio-labial wall of the functional tooth proceeds from within the pulp cavity. Finally, most of the functional tooth is resorbed and only its small tip (the enamel cap) remains to be shed. In some tooth positions (only observed in young specimens), a highly resorbed but functional tooth may still be present, while the replacement tooth provok-

DENTITION AND TOOTH REPLACEMENT IN CHALCIDES

151

Fig. 3. a: Distribution of the relative size of the 560 replacement teeth in the lower jaw (left ⫹ right quadrants) of the 20 Chalcides studied. b: Size distribution of the replacement teeth in function of age. c: Size distribution of the replacement teeth in function of the seasons.

152

S. DELGADO ET AL.

Fig. 4. Diagrams of the dentition in three Chalcides specimens (0.5, 10.5, and 36 months old) showing the relative size of the replacement teeth at each position (1–18). Relations between the developmental stage of the replacement teeth in the left (䊐) and right (x) quadrants of the lower (A) and upper (B) jaws and in the lower (䊐) and upper (x) quadrants of the left (C) and right (D) sides. Functional teeth are shown on the 1-value line.

DENTITION AND TOOTH REPLACEMENT IN CHALCIDES

153

Fig. 5. Schematic representation of the dentition in the four jaw quadrants of the same Chalcides individuals as in Figure 4 (A ⫽ 0.5 month; B ⫽ 10.5 month; C ⫽ 36-month-old specimen). In the upper quadrants the premaxilla is separated from the maxilla by a line. 1–18, tooth positions; open circles, teeth in course of resorption; gray circles, functional teeth; black circles, small, medium, and large replacement teeth (see text for explanation). The replacement teeth have been precisely located with regard to the functional teeth they will replace.

154

S. DELGADO ET AL.

Fig. 6. Diagrams of the relative size of the teeth (functional and replacement teeth) in all positions of the left lower jaw of nine specimens representative of the growth series. Solid lines link the replacement teeth of diminishing size in successive positions; the dotted lines link teeth in alternate positions (odd and even series).

ing resorption is well developed and a new germ is visible at its base. In adult Chalcides the replacement teeth induce resorption at the disto-lingual side of the functional teeth, as in younger specimens. But, in contrast to what is observed in juveniles, tooth erosion proceeds from below and closer to the base of the functional tooth. When the germ grows, resorption affects the entire basal region of the functional tooth, penetrating its pulp cavity from below so that a large part of the functional tooth is lost once the mesio-labial side is resorbed. Replacement Pattern The relative tooth size (replacement and functional teeth) at each position has been transferred to a diagram for each of the 40 lower jaw quadrants radiographed. Thus, the analysis of the replacement patterns concerned 560 replacement teeth and 430 functional teeth. The diagrams obtained for nine specimens, chosen as representative of the growth series (i.e., from 0.5–70 months), are shown in Figure 6 and resemble results from all the other individuals. From front to back, two different results

were obtained depending on whether we consider the size of the replacement teeth in adjacent or in alternate positions. Replacement teeth in adjacent positions. In each specimen studied every successive replacement tooth is separated from the following one by a roughly regular interval. Detailed analysis of the 40 diagrams reveals two general trends. First, whatever the tooth position and the age of the individual, a replacement tooth in a position p is followed in position p⫹1 by a replacement tooth that is, on average, 20% less developed (0.2 less in relative size). For instance, when the replacement tooth in position p is 60% of its definitive size, the replacement tooth in p⫹1 is about 40% developed, and the next in p⫹2 is about 20%. Second, when the size of a replacement tooth at a position p is less than 40% of its definitive size, position p⫹1 contains either a small replacement tooth, approx. 20% less developed, or a well-developed replacement tooth. In the latter case, we suspect an unmineralized tooth germ is probably starting its morphogenesis in position p⫹1, but it is not visible on the radiograph. Some examples are presented below to illustrate these trends (Fig. 6).

DENTITION AND TOOTH REPLACEMENT IN CHALCIDES

In the diagram of a 0.5-month-old Chalcides, six series of replacement teeth (positions 1–2, 3– 4, 5– 6, 7– 8 –9, 10 –11, 13–14) have been recorded. The first series starts in position 1 with a replacement tooth having a relative size of 0.32. The next position, position 2, shows a tooth germ approx. 20% less developed, i.e., a relative size of 0.15. Position 3 probably contains a tooth germ that has begun morphogenesis, but is not visible on the radiograph. However, there is a well-developed replacement tooth in this position. Indeed, position 3 contains a tooth with a relative size of 0.32, followed by a tooth with a relative size of 0.15, and so on. In the 11-month-old specimen, the successive series of replacement teeth are 1–2, 3– 4, 5– 6 –7, 8 –9 – 10, 11–12–13, 14 –15. The tooth in position 1 has a relative size of 0.45, in position 2 of 0.18, and there is no smaller tooth visible (unmineralized tooth germ?) in position 3; thus, a well-developed replacement tooth is present in position 3 (of a relative size of 0.50), the next in position 4 has a relative size of 0.20, and so on. In this specimen some successive tooth positions show a size difference greater than 20%, but the general tendency still applies. Finally, in the 70-month-old specimen the 14 replacement teeth take the successive positions 1–2, 3– 4 –5, 6 –7, 8 –9, 10 –11–12, and 13–14. In each group the second tooth is 20% less developed than the previous one. These general trends were found to apply for 95% of the replacement teeth in Chalcides, i.e., only a few exceptions can be found. Replacement teeth in alternate positions. Considering the total of 40 lower jaw quadrants (560 replacement teeth), the relative size of the replacement teeth in the next, more posterior, alternate position is found to increase for 80% of the replacement teeth, to be equal in 15% and to decrease in 5% (Fig. 6). However, when considering the size increase it is not as regular as described previously for the intervals between successive replacement teeth. No general trend could be deduced from the tooth replacement patterns we have studied. However, in general (65%), when a replacement tooth is small, i.e., a relative size between 0.1 and 0.4, the tooth in position p⫹2 shows a small increase only (on average, the tooth is 10% larger). For instance, in the 4.5-month-old specimen the relative size of the replacement teeth in the odd series (1–3–5–7) is: 0.28 in position 1, 0.32 in position 3, 0.39 in position 5, but 0.6 in position 7, and a functional tooth in position 9. In the 18-month-old specimen, the even series (2– 4 – 6 – 8) the relative size is: 0.28 in position 2, 0.42 in position 4, 0.52 in position 6, 0.59 in position 8, and a functional tooth in position 10. DISCUSSION Of all the studies that have dealt with the dentition and pattern of tooth replacement in squamates,

155

this study in Chalcides is, to our knowledge, the first that 1) is based on a growth series of specimens of known age having lived under similar conditions, and 2) takes into account the precise (albeit relative) size of the replacement teeth. Previous studies in reptiles involved either i) the comparison of the dentition pattern in a wide range of families (e.g., Edmund, 1960, 1969, or ii) one species in particular but with a series of specimens of unknown age (Cooper, 1966; Anguis fragilis), or iii) mature specimens only (Rocek, 1980; Lacerta viridis). In Edmund’s works, only a few specimens (in general, one or two adults) in many species were studied. The pattern of tooth replacement in these species served to establish the well-known “Zahnreihen” theory. The specimens from museum collections used by Edmund were of unknown age. In all previous studies the size of replacement teeth was only recorded as either small, medium, or large. In the wild it is difficult to obtain a number of specimens of a species representative of a growth series and it is even more difficult to catch individuals in all seasons. However, as shown in the present study, it is possible to estimate the age of wild specimens using skeletochronology, a method that is currently used in the study of lizard populations (Castanet et al., 1993). By breeding captive squamates we have obtained a growth series of Chalcides. The laboratory-bred Chalcides has the advantage of 1) providing specimens of known age regularly distributed in a growth series, and 2) providing individuals that died in different seasons. We are aware that the breeding conditions of Chalcides in our laboratory differ from the conditions encountered in nature, but the studied specimens seemed to thrive well. This study was undertaken to determine whether the knowledge of the tooth replacement pattern in Chalcides could help us to predict stages of tooth differentiation. This question can now be answered affirmatively. In addition, a number of related data concerning the dentition and tooth replacement in this scincid need to be dealt with. Dentition and Tooth Replacement in Chalcides Number of tooth positions during ontogeny in Chalcides. The number of tooth positions in Chalcides (16 –17) has been found to be roughly similar in the 20 individuals studied, i.e., from 0.5–70month-old specimens. The maintenance of the number of tooth positions along the row from birth to old age has two possible, mutually exclusive, explanations. First, the number of tooth families remains constant through ontogeny, i.e., a tooth at position p at birth belongs to the same tooth family as the tooth at position p, 6 years later. Alternatively, some tooth families have been lost from the row during ontogeny but new positions have been acquired, either posteriorly or in between. The number of positions,

156

S. DELGADO ET AL.

therefore, is kept constant but positions are not necessarily homologous in juveniles and adults. In Chalcides, on the one hand, the length of the tooth row, tooth size, and tooth intervals increase during ontogeny and, on the other hand, tooth replacement occurs slightly posterior (disto-lingually) to the functional teeth. This suggests that an increase of tooth intervals accounts for the increase of the length of the tooth row and for the accommodation of larger teeth. This could mean that the number of tooth positions is conserved during ontogeny with a slight drift of the positions to the rear facilitated by the disto-lingual replacement. A similar observation of an unchanged number of tooth positions was made by Gans (1957) in an amphisbaenid and by Cooper (1966) in the slow worm, Anguis fragilis. However, a constancy of number of tooth positions is not the common condition in most reptiles; the number of teeth generally increases with age by addition of new positions posteriorly on each row (e.g., Edmund, 1960). For instance, the agamid Agama agama has nine teeth in a row at hatching while there are 19 in adults; the row of teeth is extended posteriorly by addition of new positions as the jaw grows (Cooper et al., 1970). Similarly, a young Iguana adds several tooth positions to the posterior end of the row in 1 year (Edmund, 1969). In squamates, reduction in tooth number (i.e., loss of tooth positions) with increasing age has been reported in Lacerta and in the slow worm (Cooper, 1965, 1966) and in an amphisbaenid (Schmidt, 1960). This reduction is apparently achieved by the occasional replacement of two adjacent teeth by one successor. In jaws that have ceased growing, it is probable that one of the teeth fails to develop to accommodate larger teeth. It is also possible that replacement ceases in old specimens, as suggested by Gans (1957) in an amphisbaenid and by Bellairs and Miles (1960, 1965) in a varanid. In some adult Chalcides the number of tooth positions was less than in juveniles, suggesting that one or two positions could have been lost. However, the oldest Chalcides studied in the present work did not show cessation of replacement, and there is no evidence that replacement ceases in most squamates examined, even in very old individuals (Edmund, 1960). These somewhat contradictory results could be related to the different modes of study employed. Tooth number in relation to age and seasons. The high number of replacement teeth observed throughout the growth series of Chalcides implies that 1) the process of tooth renewal is active whatever the age (the oldest individual studied was 6 years old), and 2) that replacement teeth are present during the autumn and winter periods. There is at least one developing tooth at each position on the row but two replacement teeth (one large, one small) were frequently observed in young specimens; also, some young germs were certainly not visible on radiographs because they were not sufficiently mineralized. Thus, the number of tooth germs must have

been underestimated. The presence of two replacement teeth at the same position in young specimens suggests that there is no direct link between the age of the functional tooth and the initiation of a new tooth. Rather, this indicates that teeth at a position are initiated at a given rhythm, even if the previous tooth is not yet functional. A genetic “clock” could be responsible for the periodical tooth initiation at every position. Is this “clock” the same for each tooth position in Chalcides? Probably, if we consider the regular “developmental interval” separating each successive replacement tooth in the row. Is this “clock” synchronized for all replacement teeth in a row? Probably not if we consider the irregular “developmental interval” separating replacement teeth in alternate positions. Such a synchronized “clock” could have been the underlining genetic mechanism of what Edmund (1960, 1969) has called “wave of replacement teeth” on which his Zahnreihen theory is based. It is known that the time needed to form a tooth increases with age. For instance, in urodelan amphibians tooth development takes only days in larvae, but requires several months in adults (Chibon, 1977). In the lizards Iguana iguana and Varanus bengalensis tooth development takes about 3 months (Edmund, 1960) and it needs 12 months in 2-year-old crocodile (Edmund, 1962). The functional period of a tooth also increases with age: from 20 days in larvae to several months in adult amphibians (Chibon, 1977); at least 12 months in 2-year-old crocodiles (Edmund, 1962); 1.5– 4 months in 3–10year-old slow worms (Cooper, 1966); 3 months in I. iguana, and 3.5 months in V. bengalensis (Edmund, 1960). This implies that the time separating tooth initiations at a single position increases with age and this explains why only one replacement tooth is found in juvenile and adult Chalcides, instead of two in younger ones. All these data suggest that the genetic “clock” controlling tooth initiation is under the control of epigenetic factors such as general metabolism. This could explain why the number of replacement teeth is higher in summer than in other seasons in Chalcides. In contrast to replacement teeth, the number of functional teeth in Chalcides varies greatly with age and season, particularly during the winter period of the first year of life. The number of individuals in each age class and in each season was too small for statistical treatment, but the results nevertheless deserve a brief comment. Chalcides are born in July with functional teeth and with replacement teeth in formation. Indeed, in reptiles teeth are first formed in embryos and it is known that even tooth replacement is initiated before birth (e.g., Ro¨se, 1894; Woerdeman, 1919; Osborn, 1971; Westergaard, 1986; Westergaard and Ferguson, 1986). In young lizards teeth are replaced within a period of a few weeks (Osborn, 1971; Westergaard, 1986). Therefore, in young Chalcides several tooth generations develop during summer and autumn, then tooth renewal

DENTITION AND TOOTH REPLACEMENT IN CHALCIDES

stops when starvation occurs in winter, as illustrated by the fewer number of functional teeth. Numerous functional teeth are lost during the first winter, suggesting that the resorption process has continued during this period. Such a phenomenon was not observed in specimens older than 1 year that died in the following winter. This could be due either to the presence of larger functional teeth and/or to a slower rate of tooth renewal in juveniles and adults compared to younger individuals. Although these findings need to be confirmed with a larger number of specimens dying during the winter period, they could point to a difference in tooth replacement between 1-year-old specimens and older ones. Similarly, 3–10-year-old Anguis fragilis do not lose their teeth during winter (Cooper, 1966). Speed of tooth differentiation. The sizes of the 560 replacement teeth studied are not regularly distributed within the different size classes. This means that the rate of development is not regular during formation of each tooth. The size distribution strongly suggests that teeth grow rapidly until they reach 20% of their definitive size. After that, the rate of development reduces (or there are periods of growth arrest) during the second and third fifth of their size acquisition (until teeth acquire approximately 50% of their definitive height), and finally accelerates during the last two-fifths until the teeth attach and erupt. No major difference was observed in size distribution between young and old individuals and this is confirmed by the diagrams of tooth size. These findings are different from the traditional view (e.g., Edmund, 1960; Cooper, 1966) that teeth increase their size at a regular rate. Indeed, we expected to find such a constant rate of development rather than two steps of rapid development separated by a period of slower growth. However, these variations in tooth growth rates in Chalcides were inferred from the measurement of a large number of replacement teeth within a growth series. The period of slow growth corresponds to the period when the young replacement tooth approaches the base of the functional tooth. Indeed, the latter has to be completely resorbed (or shed) before the replacement tooth could take its place. The resorption process probably slows down (or growth is periodically arrested) the development of the replacement teeth during the period from 20% to 40 –50% of their definitive height. When the functional tooth is shed it no longer limits the growth of the new tooth and this could explain the final period of accelerated growth. In amphibians, acceleration occurs when the replacement teeth are 45% of their final size (Chibon, 1977); this could be due to the same reason as observed in Chalcides. Tooth replacement in Chalcides. In Chalcides, depending on whether juveniles or adults are considered, the replacement teeth arise in a different position with regard to the functional teeth they replace. This point was not clearly stated in the

157

literature because of the absence of growth series and/or knowledge of the age of the specimens. In 0.5-month-old Chalcides, replacement teeth develop in a disto-lingual position to the functional tooth but lateral, whereas in adults of 2 years and more they develop in a lingual but basal position. These differences are certainly related to the growth of the jaws. In young specimens, when the jaws are growing fast and teeth are increasing their size through numerous generations, disto-lingual replacement occurs. In adults, in which the jaws grow slowly, if at all, replacement is only lingual. The lingual position protects growing tooth germs from injuries. The lateral position of replacement teeth in young specimens vs. the basal position in adults can also be discussed in terms of protection (the prey and the crushing forces are larger in adults) but it could also be due to the accommodation of larger teeth within the dental grooves that increase in depth rather than in width. Edmund (1960) recognized two methods of tooth replacement in squamates: the varanid and the iguanid methods. In the varanid method, which is found in most anguinomorph lizards, replacement teeth grow in an interdental position, i.e., posterior to the functional teeth. The iguanid method concerns non-anguinomorph lizards; the replacement tooth comes up directly lingual to the functional tooth. Rieppel (1978) found a third type for anguinoid lizards, the intermediate method, in which the replacement tooth develops distolingually to the functional tooth. Rieppel found this method easily derivable from the iguanid method, which is considered a primitive type, by a shift of the replacement teeth to a disto-lingual position. Unfortunately, these studies mostly concern adult specimens. Information from a growth series in Chalcides has enabled us to analyze the variation of tooth replacement during ontogeny and to take into consideration jaw growth. We have found that tooth replacement in young and juvenile Chalcides follows the intermediate method (as described by Rieppel, 1978), and that in adults this method is progressively changed to the iguanid method (as described by Edmund, 1960). We can speculate that the intermediate method will be found in young and juveniles of squamate species, in which the iguanid method has been described to occur in adults. In addition, this would imply that one could not assume any longer that the iguanid way of replacement is primitive. However, this might be checked in growth series of iguanids. Tooth Replacement Pattern and Prediction of Precise Stages of Tooth Development This study was undertaken because two features in the tooth replacement pattern in reptiles were, a priori, favorable for an accurate prediction of precise stages of tooth development in squamates: symme-

158

S. DELGADO ET AL.

try in the dentition (e.g., Cooper, 1966) and the “Zahnreihen” theory (Edmund, 1960, 1969). Our results, based on a growth series of Chalcides, clearly indicate that symmetry is not a useful criterion and that the successive initiation of tooth replacement in alternate loci is not as regular as supposed by the “Zahnreihen” theory. In fact, a third feature, which was not envisaged at first, i.e., the development state of replacement teeth in successive positions, has appeared to be highly reliable to predict stages of tooth development in this lizard. Symmetry in replacement. In the growth series of Chalcides there was no strict symmetry (synchrony) in tooth replacement, either bilaterally or between upper and lower jaws. Indeed, although symmetry was observed in some positions, there was too much variation from one position to the other to assume that opposite teeth (left–right or upper– lower) show the same stage of development. This absence of precise symmetry in Chalcides contrasts previous reports on the bilateral symmetry in squamates, i.e., similar replacement stages in both upper and lower jaws and in both right and left sides (Cooper, 1965, 1966). In Chalcides such symmetry appears not to be necessary to maintain a dentition that is functional at all times. “Zahnreihen” theory and the condition in Chalcides. Teeth in most reptiles (with a few exceptions in Sphenodon and agamids; Cooper et al., 1970; Cooper and Poole, 1973; Robinson, 1976; Throckmorton, 1979) are replaced throughout life and tooth replacement occurs in waves which regularly sweep through alternate tooth positions (Edmund, 1960). These findings were never questioned by the authors that have subsequently worked on the subject (e.g., Cooper, Osborn, DeMar). Edmund (1960, 1962, 1969) was the first to try to explain the organization of reptilian dentitions. His “Zahnreihen” theory is based on the successive initiation of teeth in alternate loci by the propagation of a “stimulus” along the dental lamina. Each stimulus produces a tooth row, in which replacement teeth are in successive developmental stages. Edmund called these stimuli replacement waves. However, the question of how a replacement wave is generated and/or controlled has been much debated in terms of evolutionary implications (Osborn, 1970, 1984; DeMar, 1972, 1974; Osborn and Crompton, 1973) and of embryological reality (e.g., Osborn, 1971, 1972). Some authors have used Edmund’s data (i.e., only the adult condition of the dentition) to establish mathematical models of tooth replacement pattern (Osborn, 1970, 1972, 1974; DeMar, 1972, 1973, 1974). Study of the ontogeny of tooth succession in Lacerta vivipara led Osborn (1971), to propose the tooth family theory (i.e., successive generations of teeth at a same position) as a possible explanation of the replacement waves: these waves could be the automatic sequel to the pattern of tooth development in embryos. However, Osborn’s theory has been questioned by Westergaard (1986) and by

Westergaard and Ferguson (1986), who think that the tooth family theory cannot explain early dentition development in reptiles. Our study on the Chalcides dentition seems to confirm that, during ontogeny, tooth replacement does occur in waves, which sweep through alternate tooth positions. However, this pattern of replacement is difficult to establish in young and juvenile specimens because it is not as regular as suggested by the “Zahnreihen” theory proposed by Edmund (1960). However, when looking at the oldest specimens this pattern is more regular. Therefore, it seems that the “Zahnreihen” theory applies more to adult than to juvenile dentitions. In other words, when the frequency of tooth replacement is high the pattern is less regular (yet it seems to exist) than when it is low. This difference with Edmund’s findings could be also due to the way in which we have measured the tooth height compared to Edmund and following authors. Indeed, all our patterns are deduced from the precise size of the replacement teeth (at the nearest 50 ␮m), in contrast to the four conditions only (small, medium, large, very large) used by Edmund (1960). Indeed, grouping tooth sizes into classes could reduce the range of variations and could modify the pattern a little. In summary, the study of tooth development in Chalcides could benefit from the knowledge of the replacement pattern in alternate positions, but predictions will be more accurate in adults than in juveniles. Prediction of successive stages of tooth development in Chalcides. Our results for the squamate Chalcides indicate that, from front to back, we can predict the size of a replacement tooth (and thus its developmental stage), provided we have obtained accurate information on the developmental stage of the previous replacement tooth in the tooth row. For instance, using serial sections of any jaw quadrant of Chalcides, the prediction of a developmental stage of a tooth germ could be obtained as follows. When looking at successive positions—If a replacement tooth in position p is 40% or more of its final height, the tooth in position p⫹1 will be 20% less developed. If the tooth in position p is less than 40% of its final height, p⫹1 contains a tooth 20% less developed, i.e., either a small tooth or a tooth recently initiated. In the latter case, a well-developed replacement tooth will be found in p⫹1. When looking at alternate positions—If the size of a replacement tooth in position p is below 40% of its final height, the tooth in position p⫹2 will be slightly larger and in a slightly more advanced stage of development. For the foregoing, it is clear that a comparison with adjacent positions yields the best possible prediction as to the state of development of a particular germ.

DENTITION AND TOOTH REPLACEMENT IN CHALCIDES

ACKNOWLEDGMENTS We thank Prof. A. Huysseune (Gent University, Belgium) and Dr. M. Laurin (Paris) for comments and suggestions on the manuscript. We thank Prof. J. Castanet for providing Chalcides specimens and for expertise in age estimation. We thank M.M. Loth for assistance in sectioning. LITERATURE CITED Bellairs Ad’A, Miles AEW. 1960. Apparent failure of tooth replacement in monitor lizards. Br J Herpetol 2:189 –194. Bellairs Ad’A, Miles AEW. 1965. Apparent failure of tooth replacement in monitor lizards. Addendum. Br J Herpetol 3:14 – 15. Carlson S. 1990. Vertebral dental structures. In: Carter JG, editor. Skeletal biomineralizations: patterns, processes and evolutionary trends, vol. I. New York: Van Nostrand Reinhold. p 531–556. Castanet J, Francillon-Vieillot H, Meunier FJ, Ricqle`s Ade. 1993. In: Hall BK, editor. Bone and individual aging, vol. 7. Boca Raton, FL: CRC Press. p 245–283. Chibon P. 1977. Vitesse de croissance et renouvellement des dents chez les Amphibiens. J Embryol Exp Morphol 42:43– 63. Cooper JS. 1965. Tooth replacement in amphibians and reptiles. Br J Herpetol 3:214 –217. Cooper JS. 1966. Tooth replacement in the slow worm (Anguis fragilis). J Zool Lond 150:235–248. Cooper JS, Poole DFG. 1973. The dentition and dental tissues of the agamid lizard, Uromastyx. J Zool Lond 169:85–100. Cooper JS, Poole DFG, Lawson R. 1970. The dentition of agamid lizards with special reference to tooth replacement. J Zool Lond 162:85–98. Delgado S, Casane D, Bonnaud L, Laurin M, Sire J-Y, Girondot M. 2001. Molecular evidence for Precambrian origin of amelogenin, the major protein of Vertebrate enamel. Mol Biol Evol 18:2146 –2153 DeMar RE. 1972. Evolutionary implications of Zahnreihen. Evolution 26:435– 450. DeMar RE. 1973. The functional implications of the geometrical organization of dentitions. J Paleontol 47:452– 461. DeMar RE. 1974. On the reality of Zahnreihen and the nature of reality in morphological studies. Evolution 28:328 –330. Edmund AG. 1960. Tooth replacement phenomena in the lower vertebrates. Contrib R Ont Mus Life Sci Div 52:1–190. Edmund AG. 1962. Sequence and rate of tooth replacement in the Crocodilia. Contrib R Ont Mus Life Sci Div 56:1– 42. Edmund AG. 1969. Dentition. In: Gans C, Bellairs Ad’A, Parsons TS, editors. Biology of Reptilia, vol. I. London: Academic Press. p 117–200. Gans C. 1957. “Anguimorph” tooth replacement in Amphisbaena alba Linnaeus, 1758, and A. fuliginosa Linnaeus, 1758 (Reptilia: Amphisbaenidae). Breviora 70:1–12.

159

Girondot M, Sire J-Y. 1998. Evolution of the amelogenin gene in toothed and tooth-less vertebrates. Eur J Oral Sci 106(suppl 1):501–508. 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(suppl 1):437– 481. Osborn JW. 1970. New approach to Zahnreihen. Nature 225:343– 346. Osborn JW. 1971. The ontogeny of tooth succession in Lacerta vivipara Jacquin (1787). Proc R Soc Lond B 179:261–289. Osborn JW. 1972. On the biological improbability of Zahnreihen as embryological units. Evolution 26:601– 607. Osborn JW. 1973. The evolution of dentitions. Am Scient 61:548 – 559. Osborn JW. 1974. On the control of tooth replacement in reptiles and its relationship to growth. J Theor Biol 46:509 –527. Osborn JW. 1978. Morphogenetic gradients: fields versus clones. In: Butler PM, Joysey KA, editors. Development, function and evolution of teeth. London: Academic Press. p 171–201. Osborn JW. 1984. From reptile to mammals: evolutionary considerations of the dentition with emphasis on tooth attachment. Symp Zool Soc Lond 52:549 –574. Osborn JW, Crompton AW. 1973. The evolution of mammalian from reptilian dentitions. Breviora 399:1–18. Rieppel O. 1978. Tooth replacement in Anguinomorph lizards. Zoomorphologie 91:77–90. Robinson PL. 1976. How Sphenodon and Uromastyx grow their teeth and use them. In: Bellairs Ad’A, Cox CB, editors. Morphology and biology of Reptiles. Linn Soc Symp Ser no 3, London: Academic Press. p. 43– 64. Rocek Z. 1980. Intraspecific and ontogenetic variation of the dentition in the green lizard Lacerta viridis (Reptilia, Squamata). Vest Cs Spolec Zool 44:272–278. Ro¨se C. 1894. Ueber die Zahnentwicklung der Crocodile. Morphol Arbeit 3:195–228. Sander PM. 2001. Primless enamel in amniotes: terminology, function and evolution. In: Teaford M, Ferguson MWJ, Smith MM, editors. Development, function and evolution of teeth. New York: Cambridge University Press. p 92–106. Schmidt WJ. 1960. Uber einen Zwillingzahn bein einer Eidechse (Gerrhonotus multicarinatus webbii). Dt Zahna¨rztl Z 15:1149 – 1151. Throckmorton GS. 1979. The effect of wear on the cheek teeth and associated dental tissues of the lizard Uromastix aegyptius (Agamidae). J Morphol 160:195–208. Westergaard B. 1986. The pattern of embryonic tooth initiation in reptiles. In: Russell DE, Santoro J-P, Sigogneau-Russell D, editors. Me´m Mus Natn Hist Nat Paris (se´rie C) 53:55– 63. Westergaard B, Ferguson MWJ. 1986. Development of the dentition in Alligator mississipiensis. Early embryonic development in the lower jaw. J Zool Lond 210:575–597. Woerdeman MW. 1919. Beitra¨ge zur Entwicklungsgeschichte von Za¨hnen und Gebiss der Reptilien. Beitrag I. Die Anlage und Entwicklung des embryonalen Gebisses als Ganzes und seine Beziehung zur Zahnleiste. Arch Mikrosk Anat 92:104 –192.