- Email: [email protected]
Functional implications of enamel thickness in the lower molars of red colobus (Procolobus badius) and Japanese macaque (Macaca fuscata)
Daisuke Shimizu Laboratory of Physical Anthropology, Department of Zoology, Graduate School of Science, Kyoto University, Sakyo, Kyoto 606-8502 Japan. E-mail: [email protected], [email protected] Received 19 July 2000 Revised 24 June 2002 and accepted 30 July 2002 Keywords: chewing efficiency, pQCT, red colobus, folivore, enamel rim, lower molar.
Functional implications of enamel thickness in the lower molars of red colobus (Procolobus badius) and Japanese macaque (Macaca fuscata) The purpose of this study is to determine whether teeth are likely to retain their functional efficiency throughout an individual’s life time. This was done by comparing the enamel volume, the cross-sectional enamel area and the pattern of enamel distribution on unworn M2s of folivorous (Procolobus badius: red colobus; n=8) and frugivorous (Macaca fuscata: Japanese macaque; n=6) cercopithecids. The enamel volume of M. fuscata is significantly greater than that of P. badius. As the lower molars of colobines become worn, the dentine is exposed on the buccal cusps and narrow enamel rims are formed around the dentine exposures. The buccal enamel rims are especially well-developed and sharp, a pattern that has probably been selected for as being advantageous for shredding fibrous plant materials. The results of this study demonstrate that the enamel on the lingual side of the protoconid, where dentine exposure occurs first, is much thinner in P. badius than it is in M. fuscata. In addition, the dentine is exposed and thin enamel rims are formed faster in P. badius than in M. fuscata. Also, P. badius has significantly thinner and more uniform enamel distribution on the buccal wall of the crown and a higher protoconid. The buccal flare is well-developed in M. fuscata, but poorly developed in P. badius. It is tentatively suggested that the undeveloped flare and thinner enamel of P. badius combine to enable this species to maintain narrow rims, even after dental attrition, while the high cusps may be an adaptation for providing narrow enamel rims throughout life. 2002 Elsevier Science Ltd. All rights reserved.
Journal of Human Evolution (2002) 43, 605–620 doi:10.1053/jhev.2002.0593 Available online at http://www.idealibrary.com on
Introduction Aged monkeys with heavily worn teeth can be found in the wild. This shows that even worn teeth must be able to masticate food effectively. Other folivorous or graminivorous mammals (e.g., selenodont artiodactyls or bilophodont marsupials) produce and maintain functionally viable molars through dental alterations that occur as a result of tooth wear (Lumsden & Osborn, 1977; Lanyon & Sanson, 1986; McArthur & Sanson, 1988). Preliminary research shows Present address: Department of Human Anatomy and Cell Biology, The University of Liverpool, Ashton Street, Liverpool L69 3GE, U.K. 0047–2484/02/110605+16 $35.00/0
that primates might also maintain similar tooth function after wear. For example, the occlusal surfaces of thick enamelled cercopithecine molars wear flat and become better suited for grinding and crushing (Benefit, 1987). The thin enamel of the gorilla might be advantageous for retaining sharp edges necessary for shredding and slicing (Ungar & Williamson, 2000). Previously, researchers have discussed the importance of tooth geometry and enamel distribution on tooth function (Kay & Hiiemae, 1974; Lumsden & Osborn, 1977; Kay, 1978; Kay & Hylander, 1978; Benefit, 1987; Hartman, 1988; Spears & Crompton, 1996). They have concluded that unworn 2002 Elsevier Science Ltd. All rights reserved.
606
.
teeth belonging to folivorous species tend to possess high relief molar crowns with high cusps and long crests (e.g., colobines) (Kay, 1978; Kay & Hylander, 1978; Benefit, 1987), whereas frugivorous or omnivorous species tend to possess low relief molar crowns with low, rounded cusps (e.g., macaques) (Kay & Hiiemae, 1974; Lumsden & Osborn, 1977; Hartman, 1988; Spears & Crompton, 1996). In general, thick enamelled molars are more resistant to wear than thin enamelled molars, and may also be able to resist higher loads (Macho & Spears, 1999): hard object feeders tend to possess thick enamel (Kay, 1981). Conversely, thin enameled molars that possess longer and sharper crests are better able to shear tough and/or fibrous foods (Lumsden & Osborn, 1977): leaf eaters tend to possess thin enamel and long shearing crests (Kay, 1981). If enamel thickness is indeed a good indicator of a species’ diet, predictions could be made regarding enamel thickness distribution in primates with different diets. In the 1980s various methodologies for measuring enamel thickness were discussed: Martin (1983, 1985) pointed out that a single linear measurement of the enamel thickness, especially a linear measurement taken from a worn tooth (Kay, 1981; Benefit, 1987) or naturally fractured tooth (Andrews & Tobien, 1977), was of limited value. He suggested that enamel volume was the optimum measurement for comparative purposes. However, measurements of enamel volume cannot describe the pattern of enamel thickness distribution. Furthermore, it is difficult to calculate enamel volume non-destructively, but recent developments in computing tomography (CT) have made it easier to examine enamel thickness distribution and volume. Linear enamel thickness perpendicular to the dentino-enamel junction (DEJ) on the cross-sectional CT image of a tooth can be easily obtained for comparison of the patterning of enamel
thickness distribution (Macho & Thackeray, 1992; Macho & Berner, 1993, 1994; Schwartz, 2000b). Over the last few years the distribution of enamel over the tooth crown has been studied, and extensively explored in thick-enamelled hominoids (Macho & Thackeray, 1992; Macho & Berner, 1993, 1994; Macho, 1994; Gantt, 1998; Schwartz, 2000a,b), but not in thinenamelled species. Although inspection of unworn teeth should elucidate the function of worn teeth, analyses of worn tooth function have rarely been attempted. To address this deficiency, in this study, unworn lower molars from red colobus (Procolobus badius), one of the most specialized folivores among living primates, and the more frugivorous Japanese macaque (Macaca fuscata) were studied. The aim was to determine whether teeth are likely to retain their functional efficiency as wear progresses in accordance with their dietary adaptation. CT images of unworn teeth were analysed to gain an appreciation of overall tooth geometry and enamel thickness and distribution. Specifically, this study concentrates on the functional buccal side of the lower second molar. Procolobus badius on Tiwai Island in Sierra Leone eats a diet consisting of 20% mature leaf parts, 32% young leaf parts, 16% flowers, and 31% fruits and seeds (Davies et al., 1999). In the Kibale Forest of western Uganda, leaf parts comprise over 70% of the animal’s diet (Struhsaker, 1978), including 14·5% leaf buds, 27·15% young leaf parts, 23·65% mature leaf parts, and 8% leaves of unknown age. Macaca fuscata on Yakushima Island in Japan has a diet consisting of 76% fruits, 18% leaf parts, and 3% invertebrates (Maruhashi, 1980). Hence, the diets of the two species are considered sufficiently different to test the hypothesis that enamel thickness distribution is indicative of a species’ habitual diet. This paper has five objectives: (1) to determine whether enamel volume is related
607
to preferred diet; (2) to determine whether enamel volume correlates with crosssectional enamel area; (3) to determine whether enamel distribution differs between species; (4) to determine what kind of relationship exists between patterning of enamel distribution and tooth geometry; and (5) to determine whether folivorous species are likely to retain a functional cutting edge throughout life. Materials and methods Materials Eight unworn M2s of P. badius and six unworn M2s of M. fuscata were used in this study. The P. badius sample included five females and three specimens of unknown sex. All of the M. fuscata specimens were females. These specimens are housed at the Primate Institute of Kyoto University, Inuyama, Japan. The origin/age of the P. badius specimens is unknown. The M. fuscata specimens were collected from the wild and range between 3·0 and 6·5 years of age. CT scanning The molars were scanned using a pQCT scanner (peripheral quantitative computed tomography: XCT Research SA+, Norland & Stratec Co.) at the Laboratory of Physical Anthropology, Kyoto University. Specimens were mounted on a wooden stage using adhesive tape so that the cervical plane was parallel to the plane of the line of X-ray beams (Figure 1). The specimens were scanned with a tube voltage of 49·9 kV and a tube current of 0·45 mA. The thickness of the slice and the interval of the adjacent slice were both 0·1 mm. The pixel size was 0·1 mm and each CT image had a pixel matrix of 202202. Determination of the boundary between different materials or between the air and solids can be problematic in CT images. CT values change gradually rather than intermittently from the level of one tissue to that of another at such boundaries (Figure 2). In
Figure 1. Diagram of pQCT. X-ray beams irradiated from the X-ray Source are sent through the specimen toward the Detector. The specimens were placed on a wooden stage with the cervical plane parallel to the plane of the X-ray beams.
this study, the criterion known as a half maximum height (HMH) was adopted, in which the boundary of two materials having different density is determined by the median of the CT number levels of the two materials (Koehler et al., 1979; Ullrich et al., 1980; Magnusson, 1987; Spoor et al., 1993; Ohman et al., 1997). For example the CT value of the boundary between air (CT value 0) and dentine (CT value 1800) is 900. In CT images of a tooth, the air–enamel, air–dentine and enamel–dentine boundaries must be determined. However, the lower thresholds of enamel calculated from the HMH values were different at the air– enamel and enamel–dentine boundaries. The HMH value at the air–enamel boundary is always lower than the HMH value at the enamel–dentine boundary. If the former value is adopted as the lower threshold of enamel it results in an overestimation of the enamel thickness at the dentine–enamel junction (DEJ). On the other hand, if the latter value is chosen as the lower threshold of enamel, CT values that are lower than this value and higher than the HMH value at the true air–dentine boundary are recognized as dentine (Figure 2). This results in
608
.
Figure 2. CT image, CT number profile, diagram of cross-sectional CT image of a tooth, and CT number matrix. (a) CT image of M. fuscata sliced parallel to cervix plane at mid-level of the crown. (b) CT number profile in a linear trace between M and N. (c) Diagram of cross-sectional CT image of a tooth, and CT number profile in a linear trace between O and P. The boundaries between enamel, dentine and air were determined by the method of (HMH) respectively. Enamel–air boundary is the median of enamel CT number and air CT number. Enamel–dentine boundary is the median of enamel CT number and dentine CT number. Dentine–air boundary is the median of dentine CT number and air CT number. (d) A part of the CT number matrix including the enamel–dentine boundary and the enamel–air boundary referred to a square on the cross-sectional CT image (a). Filter Program I converts CT number into four values, and filter II removes non-existent ‘‘dentine area’’ on the outer surface of enamel. In diagram (d) solid areas indicate true enamel, dentine or air area, and areas with oblique lines indicate non-existent ‘‘dentine area’’.
the operational formation of non-existent dentine, partly on the outer surface of the enamel. Thus, the lower thresholds of enamel at the air–enamel boundary and dentine–enamel boundary were determined independently (Yamanaka et al., 2001) (Figure 2). In order to determine the true boundaries, two filter programs were used. The first filter program converts CT numbers into one of four values (Figure 2): the value below the air–dentine boundary is 0, the value between the air–dentine and air–enamel boundary is 1, the value between the air–enamel and dentine–enamel boundary is 2 and the value over the dentine–enamel boundary is 3. The second filter program then converts nonexistent ‘‘dentine areas’’ on the outer surface of enamel into air and enamel, respectively. With regard to the converted CT values using the first filter program in the area of non-existent dentine, the value 1 is converted into the air value (0) and the value 2 into the enamel value (3) (Figure 2). In the case shown in Figure 2, the width of the non-existent dentine is less than three pixels. This width depends on the CT scanner and its settings (i.e., the tube voltage or the tube current). The width of true dentine area contact with enamel, however, is considerably wider. By running the second filter program this can be corrected. Measurements For the three dimensional (3-D) reconstruction of the tooth, the Application Visualization System (AVS, Stardent Computer Inc.) was used on a workstation (Silicon Graphics Octane). Using 3-D reconstructed teeth on the AVS, maximum mesio-distal length was measured. Each 3-D reconstructed tooth was resliced on the AVS to obtain linear measurements and enamel area in crosssection in the anatomically determined plane that passes through the apices of the dentine horns of the mesial cusps perpen-
609
dicular to the cervical plane. The cervical plane is determined by the three points on the cervical margin across the mesiobuccal ridge, distobuccal ridge and mesiolingual ridge. Resliced images were converted to bitmap (BMP) image, and linear measurements and cross-sectional enamel area were obtained using a computer. This study concentrates on measurements on the buccal half of the crown, since wear on P. badius teeth occurs mainly in this region (Walker & Murray, 1975; Teaford, 1983; Benefit, 1987). The following measurements were recorded: for enamel volume (EV, Table 1) the number of the pixels that displayed the enamel value (3) was counted on each CT image forming one tooth, and multiplied by 0·001 (volume of a voxel is 0·10·1 0·1 mm). Cross-sectional enamel area was measured on the bucco-lingual cross-section (EA, Table 1). The number of pixels that displayed the enamel volume (3) was counted on the cross-sectional. In order to obtain an overall approximation of tooth size and shape, three measurements were taken: maximum mesio-distal length [ML, Table 1, Figure 3(b)], mesial width along the cervical plane [MCW, Table 1, Figure 3(c)], and height of the mesio-buccal cusp or protoconid [MBCH, Table 1, Figure 3(c)], size (CS) being the product of these measurements: CS=MLMCWMBCH
(1)
Equation 1 is used to obtain on overall approximation of tooth size. Such measurements as ML, MCW and MBCH are also preferred over tooth volume, as it has been used in several previous studies (Gould, 1975; Kay, 1975; Gingerich, 1977; Gingerich et al., 1982; Janis, 1988). Furthermore, tooth roots are not always fully formed when the tooth is unworn, and the cervical margins are not straight (RamirezRozzi, 1993). This confounds calculation of
.
610
Table 1 List of measurements (see Figure 3) Dimension EV EA
Total enamel volume Cross-sectional enamel area
ML
Maximum mesio-distal length
MCW
Mesial width
MBCH
Height of the protoconid
BSA
Buccal surface angle
ETBP
Enamel thickness on the buccal slope of the protoconid
ETCP
Enamel thickness on the cusp tip of the protoconid Enamel thickness on the lingual slope of the protoconid
ETLP
ET ERW
Enamel thickness along the lateral wall of the crown Enamel rim width
dentine area underneath the enamel cap. The flare of the protoconid was measured as the buccal surface angle [BSA, Table 1, Figure 3(a)] formed by the intersection of two lines. One line connects the lowest point of the DEJ on the buccal side and the enamel tip of the protoconid, and the other line passes through the lowest points of the DEJ on the buccal and lingual sides (the cervical line). Three enamel thicknesses of the protoconid are defined [Table 1, Figure 3(d)] as (1) the enamel thickness on the buccal slope of the protoconid (ETBP), which is a linear thickness perpendicular to the DEJ on the buccal slope of the protoconid situated 0·5mm from the dentine horn of the protoconid, (2) the enamel thickness on the cusp tip of the protoconid (ETCP), which is the
Definition Whole enamel volume of M2. Enamel area on the bucco-lingual cross-section through mesial cusps. Maximum distance between two points on the mesial and distal crown surface. Distance between the lowest points of the DEJ on the cervical plane buccal and lingual sides. Minimum distance between the cusp tip of the protoconid and the cervical plane. Angle formed by the intersection of the line which connects the lowest point of the DEJ on the buccal side and the apex of the enamel tip of the protoconid, and the line which passes through the lowest points of the DEJ on the buccal and lingual sides. Linear thickness perpendicular to the DEJ on the buccal slope of the protoconid at 0·5 mm from the dentine horn of the protoconid. Distance between the cusp tip and the dentine horn of the protoconid. Linear thickness from the dentine horn of the protoconid perpendicular to the outer enamel surface (OES) on the lingual slope of the protoconid. Linear thickness perpendicular to DEJ from landmark 2, 3 and 4. Thickness at each landmark with an elevation of 16·2 in P. badius and 12·4 in M fuscata
distance between the cusp tip and the dentine horn of the protoconid, and (3) the enamel thickness on the lingual slope of the protoconid (ETLP), which is a linear thickness from the dentine horn of the protoconid perpendicular to the outer enamel surface on the lingual slope of the protoconid. To evaluate enamel thickness along the lateral face of the tooth, four landmarks were defined on the buccal DEJ surface of the tooth at different levels [Figure 4(a)]. Each landmark was intersected by lines that were parallel with the cervical line (B–L) and the buccal DEJ surface. Landmark 1 (LMK1): at the level of the dentine horn of the protoconid. Landmark 2 (LMK2): through the lowest point of intercuspal fissure on OES. Landmark 3 (LMK3):
611
Figure 3. Cross-section through the mesial cusps of a mandibular molar showing: (a) the angle—BSA; and (b) the six linear measurements—ML, (c) MCW and MBCH; and (d) ETCP, ETBP and ETLP (see Table 1 for list of abbreviations and definitions).
through the lowest level of the DEJ between the buccal and lingual dentine horns (O–P). Landmark 4 (LMK4): at the mid-level between the cervical line B–L and the line O–P. The distribution of enamel on the buccal side was measured in two ways. The enamel thickness (ET) was measured perpendicular to the DEJ from LMK2, LMK3 and LMK4 (Figure 4(b)]. Although the enamel volume is considered by some researchers to be the optimum measurement for comparative purposes (Martin, 1983, 1985), it is meaningless when the patterning of enamel thickness distribution is being discussed. The linear enamel thickness perpendicular to the DEJ allows for comparison of the patterning of enamel thickness distribution (Macho & Thackeray, 1992; Macho & Berner, 1993,
1994; Schwartz, 2000b), but does not provide information about the thickness of the enamel rim after dentine exposure. Benefit (1987) measured thickness of the enamel rim of worn cercopithecine teeth, and discussed its functional implications, so did Kay (1981, Table 1). In order to obtain an estimate of the enamel rim width for evaluation of its functional meaning in different tooth wear stages [Figure 4(c)–(e)], the enamel rim width (ERW) was measured as the thickness at each reference point with an elevation of 16·2 (in P. badius), and 12·4 (in M. fuscata) from the cervical line. These angulations were previously established empirically to be that found in worn teeth of P. badius and M. fuscata respectively (Shimizu, in prep). In this study 39 P. badius and 29 M. fuscata M2’s were used for the
612
.
Figure 4. Cross-section through the mesial cusps showing (a) the reference points for measuring the enamel thickness and the enamel rim width. LMK1 is defined as the dentine horn of the protoconid. And the other points are defined as the intersections on the buccal DEJ surface of tooth crown relative to a line parallel to the crown base (line B–L) through the lowest point of intercuspal fissure (line Q–R) for LMK2, the lowest point of DEJ between cusps (line O–P) for LMK3, and the midpoint of the line O–P and the line B–L (line M–N) for LMK4; (b) three dimensions of the enamel thickness (ET2–4). All measurements are perpendicular to the DEJ from LMK2–4; (c)–(e) the definition of the enamel rim width (ERW). The enamel rim formed after dentine exposure has an angle relative to the cervical plane of: (c) is the angle of the enamel rims relative to the cervical plane; (d) 16·2 for P. badius; (e) 12·4 for M. fuscata.
analysis. These specimens included various stages of tooth wear corresponding to Landmark 1 to 4 in this study. The average
angular measurement for P. badius was 16·2 (S.D. 3·77) and the average angular measurement for M. fuscata was 12·4 (S.D. 2·11).
613
Table 2 Comparison of Procolobus badius and Macaca fuscata in enamel and occlusal dimensions
Species P. badius Mean S.D. CV n M. fuscata Mean S.D. CV n T-value Significance
ML/MCW (%)
MBCH/MCW (%)
BSA
5·1 0·54 10·57 8
147·3 16·15 10·98 8
98·9 13·78 13·94 8
72·0 5·34 — 8
0·2 0·04 14·56 8
6·6 0·43 6·51 6 5·62 ***
131·4 5·29 4·01 6 2·29 *
80·2 5·47 6·81 6 3·31 **
54·7 2·89 — 6 7·12 ***
0·3 0·03 9·85 6 3·11 **
EV (mm3)
EA (mm2)
MCW (mm)
45·9 2·75 5·99 8
8·4 1·48 17·55 8
91·1 10·48 11·51 6
11·5 0·97 8·78 6
VI (degrees)
One asterisk indicates that P<0·05, two asterisks that P<0·01, and three asterisks indicates that P<0·001 using Student’s t-test. Total enamel volume (EV), cross-sectional enamel area (EA), mesial width on the cervical plane (MCW), the relative maximum mesio-distal length (ML), the relative height of the protoconid (MBCH), angle of buccal slope (BSA), angle of lingual slope (LSA) and volume index (VI).
Since there is a difference in molar size between P. badius and M. fuscata all linear measurements were divided by the mesial width (MCW) for normalization. Interspecific differences were tested using the Student’s t-test, and paired t-tests were used to compare enamel thickness (ET) and enamel rim width (ERW) in P. badius and M. fuscata. In order to assess enamel thickness in relation to enamel volume and enamel cross-sectional area, the linear dimensions ETLP, ETCP, ETBP, ET2, ET3, and ET4 are averaged to calculate mean enamel thickness. When enamel thickness is compared to enamel cross-sectional area, the values of enamel thickness should be obtained independently of the enamel cross-sectional area. In order to normalize the enamel volume data, the enamel volume index (EVI) was calculated: EV EVI= CS
(2)
Where EV is enamel volume, and CS is crown size [see Equation (1)].
Results The Student’s t-test was used to examine interspecific difference in enamel volume index (EVI). A significant interspecific difference was found between P. badius (0·2) and M. fuscata (0·3) (Table 2). The enamel volume (EV) of M. fuscata (91·1mm3) was almost twice as large as that of P. badius (45·9 mm3) (Table 2). Moreover, in P. badius the coefficient of variation of the enamel volume (5·99) was small compared with M. fuscata (11·51). Enamel crosssectional area (EA) of P. badius (8·4 mm2) was also smaller than EA of M. fuscata (11·5 mm2) (Table 2). However, the coefficient of variation of cross-sectional enamel area in P. badius (17·55) was larger than it was in M. fuscata (8·78). Descriptive statistics of linear measurements are given in Table 2. The Student’s t-test was used to examine interspecific difference in MCW, ML, MBCH, and BSA. A significant interspecific difference in MCW was observed between P. badius (5·1 mm) and M. fuscata (6·6 mm), so maximum length (ML) and height of the protoconid
.
614
Table 3 Comparison of Procolobus badius and Macaca fuscata in enamel thickness Species P. badius Mean S.D. n M. fuscata Mean S.D. n T-value Significance
ET(LMK4)
ET(LMK3)
ET(LMK2)
ETBP
ETCP
ETLP
9·3 2·10 8
12·6 2·45 8
12·7 1·74 8
11·1 2·05 8
5·7 0·96 8
2·5 0·33 8
8·1 1·09 6
13·1 0·68 6
15·8 1·54 6
15·4 1·02 6
13·0 0·76 6
7·4 0·76 6
1·28 N.S.
0·51 N.S.
3·49 **
4·71 ***
15·29 ***
16·43 ***
Two asterisks indicates that P<0·01, and three asterisks that P<0·001 using Student’s t-test. Enamel thickness on three landmarks along the lateral wall of crown (ET), enamel thickness on the buccal slope of the protoconid (ETBP), enamel thickness on the cusp tip of the protoconid (ETCP), enamel thickness on the lingual slope of the protoconid (ETLP).
(MBCH) were divided by MCW for normalization. In relation to MCW maximum length of P. badius molar (147·3) was significantly larger than the maximum width of the M. fuscata molar (131·4). Height of the P. badius protoconid (98·9) was also significantly larger than the height of the M. fuscata protoconid (80·2). Procolobus badius had a significantly greater BSA (72·0) than M. fuscata (54·7). All of the linear measurements of enamel thickness were divided by mesial width for normalization and compared using the Student’s t-test (Table 3). Three of the relative enamel thicknesses on the protoconid were thinner in P. badius than in M. fuscata. Relative enamel thickness on the buccal side (ETBP) was 11·1 in P. badius and 15·4 in M. fuscata, while the cusp tip (ETCP) was 5·7 in P. badius and 13·0 in M. fuscata. On the lingual side ETLP was 2·5 in P. badius and 7·4 in M. fuscata. Procolobus badius had significantly thinner enamel along the lateral wall of the crown near the cusp tip (LMK2: 12·7) than M. fuscata (15·8). However, relative enamel thickness did not differ significantly at other levels. Interspecific differences in enamel rim width (ERW) were examined using the Student’s t-test (Table 4). The results
Table 4 Comparison of Procolobus badius and Macaca fuscata in the relative enamel rim width Landmarks P. badius Mean S.D. n M. fuscata Mean S.D. n T-value Significance
1
2
3
4
11·6 1·36 8
12·0 1·59 8
11·9 1·16 8
8·9 1·68 8
19·9 2·60 6
19·0 1·77 6
14·6 0·66 6
8·4 1·24 6
7·89 ***
7·54 ***
5·24 ***
0·69 N.S.
Three asterisks indicates that P<0·001 Student’s t-test was used.
indicated that P. badius has a narrower enamel rim than M. fuscata at three of the four levels on the buccal surface of the protoconid. The relative ERWs of P. badius at LMK1 to LMK3 levels were 11·6, 12·0 and 11·9 respectively, while those of M. fuscata were 19·9, 19·0 and 14·6. Intraspecific comparisons between the raw values for enamel thickness (ET) and enamel rim width (ERW) were carried out using the paired t-test and the results are shown in Table 5 and Figure 5. In P. badius, the ERWs had comparable values with the
615
Table 5 Comparison of enamel thickness (ET) and enamel rim width (ERW) (mm) in Procolobus badius and Macaca fuscata P. badius Landmarks ET Mean S.D. n ERW Mean S.D. n T-value Significance
M. fuscata
1/ETBP
2
3
4
1/ETBP
2
3
4
0·6 0·09 8
0·6 0·07 8
0·6 0·09 8
0·5 0·08 8
1·0 0·10 6
1·0 0·10 6
0·9 0·09 6
0·5 0·06 6
0·6 0·06 8
0·6 0·08 8
0·6 0·07 8
0·5 0·06 8
1·3 0·17 6
1·3 0·09 6
1·0 0·03 6
0·6 0·05 6
0·77 N.S.
0·90 N.S.
0·76 N.S.
1·36 N.S.
5·65 **
6·81 **
3·75 *
0·49 N.S.
One asterisk indicates that P<0·05, two asterisks that P<0·01 paired t-test was used.
Figure 5. Distribution profile of generalized enamel thickness (ET) and enamel rim width (ERW) at each reference point. Symbols show mean values and vertical bars indicate one standard deviation. Pb= P. badius; Mf=M. fuscata.
ETs at all levels. On the other hand, in M. fuscata ERWs were significantly larger than the ETs at the two mid levels (at LMK2 and LMK3). Discussion While enamel thickness distribution has featured prominently in palaeoanthropological
studies over the last 20 years (Martin, 1985; Beynon & Wood, 1986; Grine & Martin, 1988; Grine, 1991; Conroy, 1991; Macho & Thackeray, 1992; Schwartz et al., 1998), these studies have been confined to omnivorous species which possess low-crowned and thick-enamelled molars. The present study was designed to test whether enamel distribution in frugivorous primates is
616
.
Figure 6. Correlation between the enamel volume and the cross-sectional enamel area.
comparable to those found in more folivorous species. The species analysed were the folivorous Procolobus badius and the frugivorous Macaca fuscata. Although sample sizes are small, the trends observed in the present study differ between the species and indicate that molars of both species may be adapted to remain functional throughout the individual’s lifetime. Folivorous P. badius has less enamel in proportion to tooth size than frugivorous M. fuscata as determined by linear measurements (Kay, 1981; Benefit, 1987). Also, the relationship between enamel volume and cross-sectional enamel area seems to be different between the two species. Macaca fuscata have almost the same coefficient of variation for enamel volume as for crosssectional enamel area, whereas P. badius have a much larger coefficient for the crosssectional enamel area. This may be due to the fact that P. badius does retain the same amount of enamel irrespective of tooth size (Figure 6). The enamel area correlates positively with the height of the protoconid, but there are no significant relationships with regard to other dimensions. Hence, in P. badius, molars that differ in size also differ in shape and proportions. Conversely, in M.
fuscata, as the volume of the enamel and/or the area of enamel increases, the length and width of the crown, as well as the height of the protoconid become greater (Table 6). In other words, molars which differ in size do not differ in shape. The results show that the molars of P. badius possess thin enamel, uniform distribution of the enamel on buccal crown surface, and less-flared crowns. Conversely, the molars of M. fuscata possess thick enamel and flared crowns. Most of these characters have been discussed in previous work (Delson, 1973; Kay, 1981; Benefit, 1987). The most important finding in this paper is that these characters, which other researchers have discussed separately, are correlated with each other and may relate to the function of the enamel rims that are formed after dentine is exposed. Sharp ridges are required for the efficient shearing of leaves (Lucas & Teaford, 1994). In order for the enamel edges to efficiently shear leaves, the enamel edges need to be thin and sharp. The width of the enamel rims that surround the dentine patches on the cusp tips is determined by the enamel thickness (ET) and relative position between the buccal slope of the protoconid
617
Table 6 Correlation between enamel volume (EV) or cross-sectional enamel area (EA), and length (ML), width (MCW), height (MBCH) of the crown or mean enamel thickness (AET) P. badius ML MCW MBCH AET
EV EA EV EA EV EA EV EA
y= 3·4x+19·6 y= 0·3x+8·4 y= 2·5x+14·1 y=0·2x+4·69 y=1·6x0·6 y=1·3x+1·3 y=0·4x0·8 y=0·1x+0·2
and the enamel rim plane. When the enamel rims meet the buccal slope of the molar crown at a right angle, the enamel rim width has the lowest values. As this angle becomes more acute or obtuse, the enamel rim width becomes greater. Because the angle in P. badius approximates a right angle (88·2) and the enamel is thinner, it maintains narrow enamel rims. Enamel rim width is comparable to enamel thickness at all levels (Table 5 and Figure 5). How soon after molar eruption does a functional enamel rim form in P. badius? The answer to this question lies in the enamel distribution around the cusp tip of the protoconid. The thin enamel distribution on the lingual slope of the protoconid in P. badius contributes not only to rapid exposure of the dentine in this area, thereby forming sharp enamel rims on the buccal side of the protoconid, but also helps to shift leaf processing from the shearing crests to the enamel edges. Dentine exposure in colobine monkeys first appears on the lingual side of the protoconid cusp tip (Janis, 1984; Benefit, 1987; Shimizu, 2001), and correlated with this, P. badius has the thinnest enamel distribution on the buccal side of the crown (Table 3). By exposing dentine on the lingual side of the protoconid cusp tip, P. badius appears to produce sharp enamel edges. After dentine is exposed, the buccal shearing crests, which have an important role in shearing leaves during phase I of the
M. fuscata r2 =0·23 r2 =0·14 r2 =0·11 r2 =0·00 r2 =0·08 r2 =0·67 r2 =0·30 r2 =0·32
y=2·1x0·8 y=1·73x+2·96 y=2·0x2·3 y=1·9x+0·2 y=1·7x2·4 y=2·0x1·3 y=0·3x0·4 y=0·4x0·4
r2 =0·73 r2 =0·35 r2 =0·60 r2 =0·42 r2 =0·49 r2 =0·49 r2 =0·52 r2 =0·65
chewing cycle (Kay, 1975), gradually wear down and eventually disappear. However, the length of the shearing crests decreases through wear, while the length of the enamel edges formed on the buccal side of the dentine patches increases (Shimizu, 2001). It would seem that the enamel edges substitute functionally for the buccal shearing crests. In addition, a less-flared crown and uniform enamel distribution on the buccal surface of P. badius molars produce enamel edges of constant width during the course of wear. In worn P. badius molars the length of the enamel edges is equivalent to the perimeter of the worn surface on the buccal side of crown. The perimeter of the worn surface does not change much, regardless of the wear progress, because of their less-flared crown. These observations indicate that P. badius retains an effective shearing crest/ enamel rim throughout life. In contrast to colobine monkeys, attrition of the lower molars in macaques begins as a round depression at the protoconid tip and, eventually, enamel is removed from the tip of the protoconid to form a flat surface (Gantt, 1979). After that, the dentine is exposed at the centre of the flattened enamel. The enamel thickness on the cusp tip of the protoconid is greater in M. fuscata than it is in P. badius (Table 5 and Figure 5). The thick enamel distribution on the protoconid cusp tip makes dentine exposure difficult. Its importance for retaining a large
618
.
surface for crushing and grinding may be another factor, however. In M. fuscata the angle formed by the enamel rims and the buccal slope of the molar crown is 67·1 and the enamel is thicker, such that M. fuscata possesses wider enamel rims than would be expected from measures of linear enamel thickness alone. In fact, ERWs are 30% greater at LMK2 compared with ET, and 10% greater at LMK3. Macaca fuscata not only possesses thick enamel on the buccal slope of the protoconid, but also possesses pronounced flare of the molar crown. As a result, the enamel rim of M. fuscata is wider than the enamel thickness on the buccal slope of the protoconid. In conclusion, a significant difference in the relative enamel volume is observed between P. badius and M. fuscata, and enamel volume is not correlated with crosssectional enamel area. In P. badius, the enamel rims are formed quickly, as enamel on the lingual slope of the protoconid is considerably thinner than on the buccal slope of the protoconid. On the buccal surface of the crown the thin enamel is distributed uniformly and the buccal flare is poorly developed. As a consequence, a sharp cutting edge is retained throughout the functional lifetime of the tooth. In M. fuscata, enamel on the cusp tip of the protoconid is thick, so that the dentine is difficult to expose and occlusal relief forms flattened crushing surfaces. After dentine exposure, thick enamel and well-developed buccal flare help to retain wide crushing or grinding surfaces. Folivorous red colobus monkeys have thin enamelled molars that represent a different adaptive strategy for mastication compared with frugivorous Japanese macaques, which have thick enamelled molars. Maintaining a large crushing/ grinding area is important for M. fuscata, so their molars have well-developed buccal flare and thick enamel. Retaining a sharp cutting edge is important for P. badius,
however, so their molars have a poorly developed buccal flare and thin, uniformly distributed enamel.
Acknowledgements I thank Professor Hidemi Ishida, Dr Masato Nakatsukasa and other staff of the Laboratory of physical anthropology, Kyoto University for their helpful suggestions and insightful comments. Dr Hironori Takemoto of ATR helped me analyze the CT data. I acknowledge helpful discussions with Dr Yuzuru Hamada, Dr Toshio Mouri, Dr Yutaka Kunimatsu, Mr Takeshi Nishimura and Ms Michika Kondo of the Primates Research Institute, Kyoto University, and Mr Richard Abel of The University of Liverpool. My special thanks are due to Dr Richard Kay of Duke University and Dr Gabriele Macho of The University of Liverpool for valuable advice. This work was supported by a fellowship of the Japan Society for the Promotion of Science for Japanese Junior Scientists. This paper was much improved with the help of comments from three anonymous reviewers.
References Andrews, P. & Tobien, H. (1977). New Miocene locality in Turkey with evidence on the origin of Ramapithecus and Sivapithecus. Nature 268, 699–701. Benefit, B. R. (1987). The molar morphology, natural history, and phylogenetic position of the Middle Miocene monkey Victoriapithecus. Ph.D. Dissertation, New York University. Beynon, A. D. & Wood, B. A. (1986). Variations in enamel thickness and structure in east African hominids. Am. J. phys. Anthrop. 70, 177–193. Conroy, G. C. (1991). Enamel thickness in south African australopithecines: noninvasive evaluation computed tomography. Palaeont. afr. 28, 53–59. Davies, A. G., Oates, J. F. & Dasilva, G. L. (1999). Patterns of frugivory in three West African colobine monkeys. Int. J. Primatol. 20, 327–357. Delson, E. (1973). Fossil colobine monkey of the Circum-Mediterranean region and the evolutionary history of the Cercopithecidae (Primates, Mammalia). Ph.D. Dissertation, Columbia University.
Gantt, D. G. (1979). Patterns of dental wear and role of the canine in Cercopithecinae. Am. J. phys. Anthrop. 51, 353–360. Gantt, D. G. (1998). Evolutionary and functional significance of hominoid tooth enamel. Connect. Tissue Res. 39, 195–206. Gingerich, P. D. (1977). Correlation of tooth size and body size in living hominoid primates, with a note on relative brain size in Aegyptopithecus and Proconsul. Am. J. phys. Anthrop. 47, 395–398. Gingerich, P. D., Smith, B. H. & Rosenberg, K. (1982). Allometric scaling in the dentition of primates and prediction of body weight from tooth size in fossils. Am. J. phys. Anthrop. 58, 81–100. Gould, S. J. (1975). On the scaling of tooth size in mammals. Am. Zool. 15, 351–362. Grine, F. E. (1991). Computed tomography and the measurement of enamel thickness in extant hominoids: implications for its palaeontological application. Palaeont. afr. 28, 61–69. Grine, F. E. & Martin, L. B. (1988). Enamel thickness and development in Australopithecus and Paranthropus. In (F. E. Grine, Ed.) Evolutionary History of the ‘Robust’ Australopithecines, pp. 3–42. New York: Aldine de Gruyter. Hartman, S. E. (1988). A cladistic analysis of hominoid molars. J. hum. Evol. 17, 489–502. Janis, C. M. (1984). Prediction of primate diets from molar wear patterns. In (D. J. Chivers, B. A. Wood & A. Bilsborough, Eds) Food Acquisition and Processing in Primates, pp. 331–340. New York: Plenum Press. Janis, C. M. (1988). An estimation of tooth volume and hypsodonty indices in ungulate mammals, and the correlation of these factors with dietary preference. Me´m. Mus. Natn. Hist. Nat., Paris, (se´rie C). 53, 367–387. Kay, R. F. (1975). The functional adaptation of primate molar teeth. Am. J. phys. Anthrop. 43, 195–216. Kay, R. F. (1978). Molar structure and diet in extant Cercopithecidae. In (P. M. Butler & K. A. Joysey, Eds) Development, Function and Evolution of Teeth, pp. 309–339. London: Academic Press. Kay, R. F. (1981). The nut-crackers—a new theory of the adaptations of the Ramapithecinae. Am. J. phys. Anthrop. 55, 141–151. Kay, R. F. & Hiiemae, W. L. (1974). Jaw movement and tooth use in recent and fossil primates. Am. J. phys. Anthrop. 40, 227–256. Kay, R. F. & Hylander, W. L. (1978). The dental structure of mammalian folivores with special reference to primates and phalangeroids (Marsupialia). In (G. G. Montgomery, Ed.) The Ecology of Arboreal Folivores, pp. 173–191. Washington, DC: Smithsonian Institution Press. Koehler, P. R., Anderson, R. E. & Baxter, B. (1979). The effect of computed tomography viewer controls on anatomical measurements. Radiology 130, 189– 194. Lanyon, J. M. & Sanson, G. D. (1986). Koala (Phascolarctos cinereus) dentition and nutrition. II. Implications of tooth wear in nutrition. J. Zool., Lond. (A) 209, 169–181.
619
Lucas, P. W. & Teaford, M. F. (1994). Functional morphology of colobine teeth. In (A. G. Davies & J. F. Oates, Eds) Colobine Monkeys: Their Ecology, Behaviour and Evolution, pp. 173–203. Cambridge: Cambridge University Press. Lumsden, A. G. S. & Osborn, J. W. (1977). The evolution of chewing: a dentist’s view of palaeontology. J. Dent. 5, 269–287. Macho, G. A. (1994). Variation in enamel thickness and cusp area within human maxillary molars and its bearing on scaling techniques used for studies of enamel thickness between species. Arch. Oral Biol. 39, 783–792. Macho, G. A. & Berner, M. E. (1993). Enamel thickness of human maxillary molars reconsidered. Am. J. phys. Anthrop. 92, 189–200. Macho, G. A. & Berner, M. E. (1994). Enamel thickness and the helicoidal occlusal plane. Am. J. phys. Anthrop. 94, 327–337. Macho, G. A. & Spears, I. R. (1999). Effects of loading on the biomechanical behavior of molars of Homo, Pan, and Pongo. Am. J. phys. Anthrop. 109, 211–227. Macho, G. A. & Thackeray, J. F. (1992). Computed tomography and enamel thickness of maxillary molars of Plio-Pleistocene hominids from Sterkfontein, Swartkrans, and Kromdraai (South Africa): an exploratory study. Am. J. phys. Anthrop. 89, 133–143. Magnusson, A. (1987). Object size determination at computed tomography. Upsala J. Med. Sci. 92, 277–286. Martin, L. M. (1983). The relationships of the later Miocene Hominoidea. Ph.D. Dissertation, University College London. Martin, L. M. (1985). Significance of enamel thickness in hominoid evolution. Nature 314, 260–263. Maruhashi, T. (1980). Feeding behaviour and diet of the Japanese monkeys (Macaca fuscata yakui) on Yakushima Island, Japan. Primates 21, 141–160. McArthur, C. & Sanson, G. D. (1988). Tooth wear in eastern grey kangaroos (Macropus giganteus) and western grey kangaroos (Macropus fuliginosus), and its potential influence on diet selection, digestion and population parameters. J. Zool., Lond. 215, 491–504. Ohman, J. C., Krochta, T. J., Lovejoy, C. O., Mensforth, R. P. & Latimer, B. (1997). Compact bone distribution in the femoral neck of hominoids: implications for the locomotion of Australopithecus afarensis. Am. J. phys. Anthrop. 104, 117–131. Ramirez-Rozzi, F. V. (1993). Tooth development in East African Paranthropus. J. hum. Evol. 24, 429– 454. Schwartz, G. T. (2000a). Enamel thickness and the helicoidal wear plane in modern human mandibular molars. Arch. Oral Biol. 45, 401–409. Schwartz, G. T. (2000b). Taxonomic and functional aspects of the patterning of enamel thickness distribution in extant large-bodied hominoids. Am. J. phys. Anthrop. 111, 221–244. Schwartz, G. T., Thackeray, J. F., Reid, C & van Reenan, J. F. (1998). Enamel thickness and the topography of the enamel-dentine junction in south
620
.
African Plio-Pleistocene hominids with special reference to the carabelli trait. J. hum. Evol. 35, 523–542. Shimizu, D. (2001). Functional morphology of molars of folivorous primates. Ph.D. Dissertation, Kyoto University. Shimizu, D. (in prep.). Relationship between tooth wear and chewing efficiency in red colobus (Procolobus badius). Spears, I. R. & Crompton, R. H. (1996). The mechanical significance of the occlusal geometry of great ape molars in food breakdown. J. hum. Evol. 31, 517–535. Spoor, C. F., Zonneveld, F. W. & Macho, G. A. (1993). Linear measurements of cortical bone and dental enamel by computed tomography: applications and problems. Am. J. phys. Anthrop. 91, 469–484. Struhsaker, T. T. (1978). Interrelations of red colobus monkeys and rain-forest trees in the Kibale Forest, Uganda. In (G. G. Montgomery, Ed.) The Ecology of Arboreal Folivores, pp. 397–422. Washington, D.C.: Smithsonian Institution Press.
Teaford, M. F. (1983). The morphology of the lingual notch in macaques and langurs. Am. J. phys. Anthrop. 60, 7–14. Ullrich, C. G., Binet, E. F. Sanecki, M. G. & Kieffer, S. A. (1980). Quantitative assessment of the lumber spinal canal by computed tomography. Radiology 134, 137–143. Ungar, P. & Williamson, M. (2000). Exploring the effects of tooth wear on functional morphology: A preliminary study using dental topographic analysis. Palaeontologica Electronica 3(1), http://wwwodp.tamu.edu/paleo/2000_1/gorilla/issue1_00.htm. Walker, P. & Murray, P. (1975). An assessment of masticatory efficiency in a series of anthropoid primates with special reference to the Colobinae and Cercopithecinae. In (R. H. Tuttle, Ed.) Primate Funcutional Morphology and Evolution, pp. 151–194. Mouton: The Hague. Yamanaka, A., Shimizu, D. & Takemoto, H. (2001). New technique of modeling using CT images. Anthrop. Sci. (Japanese Series) 108, 81–90.
Functional implications of enamel thickness in the lower molars of red colobus (Procolobus badius) and Japanese macaque (Macaca fuscata) The purpose of this study is to determine whether teeth are likely to retain their functional efficiency throughout an individual’s life time. This was done by comparing the enamel volume, the cross-sectional enamel area and the pattern of enamel distribution on unworn M2s of folivorous (Procolobus badius: red colobus; n=8) and frugivorous (Macaca fuscata: Japanese macaque; n=6) cercopithecids. The enamel volume of M. fuscata is significantly greater than that of P. badius. As the lower molars of colobines become worn, the dentine is exposed on the buccal cusps and narrow enamel rims are formed around the dentine exposures. The buccal enamel rims are especially well-developed and sharp, a pattern that has probably been selected for as being advantageous for shredding fibrous plant materials. The results of this study demonstrate that the enamel on the lingual side of the protoconid, where dentine exposure occurs first, is much thinner in P. badius than it is in M. fuscata. In addition, the dentine is exposed and thin enamel rims are formed faster in P. badius than in M. fuscata. Also, P. badius has significantly thinner and more uniform enamel distribution on the buccal wall of the crown and a higher protoconid. The buccal flare is well-developed in M. fuscata, but poorly developed in P. badius. It is tentatively suggested that the undeveloped flare and thinner enamel of P. badius combine to enable this species to maintain narrow rims, even after dental attrition, while the high cusps may be an adaptation for providing narrow enamel rims throughout life. 2002 Elsevier Science Ltd. All rights reserved.
Journal of Human Evolution (2002) 43, 605–620 doi:10.1053/jhev.2002.0593 Available online at http://www.idealibrary.com on
Introduction Aged monkeys with heavily worn teeth can be found in the wild. This shows that even worn teeth must be able to masticate food effectively. Other folivorous or graminivorous mammals (e.g., selenodont artiodactyls or bilophodont marsupials) produce and maintain functionally viable molars through dental alterations that occur as a result of tooth wear (Lumsden & Osborn, 1977; Lanyon & Sanson, 1986; McArthur & Sanson, 1988). Preliminary research shows Present address: Department of Human Anatomy and Cell Biology, The University of Liverpool, Ashton Street, Liverpool L69 3GE, U.K. 0047–2484/02/110605+16 $35.00/0
that primates might also maintain similar tooth function after wear. For example, the occlusal surfaces of thick enamelled cercopithecine molars wear flat and become better suited for grinding and crushing (Benefit, 1987). The thin enamel of the gorilla might be advantageous for retaining sharp edges necessary for shredding and slicing (Ungar & Williamson, 2000). Previously, researchers have discussed the importance of tooth geometry and enamel distribution on tooth function (Kay & Hiiemae, 1974; Lumsden & Osborn, 1977; Kay, 1978; Kay & Hylander, 1978; Benefit, 1987; Hartman, 1988; Spears & Crompton, 1996). They have concluded that unworn 2002 Elsevier Science Ltd. All rights reserved.
606
.
teeth belonging to folivorous species tend to possess high relief molar crowns with high cusps and long crests (e.g., colobines) (Kay, 1978; Kay & Hylander, 1978; Benefit, 1987), whereas frugivorous or omnivorous species tend to possess low relief molar crowns with low, rounded cusps (e.g., macaques) (Kay & Hiiemae, 1974; Lumsden & Osborn, 1977; Hartman, 1988; Spears & Crompton, 1996). In general, thick enamelled molars are more resistant to wear than thin enamelled molars, and may also be able to resist higher loads (Macho & Spears, 1999): hard object feeders tend to possess thick enamel (Kay, 1981). Conversely, thin enameled molars that possess longer and sharper crests are better able to shear tough and/or fibrous foods (Lumsden & Osborn, 1977): leaf eaters tend to possess thin enamel and long shearing crests (Kay, 1981). If enamel thickness is indeed a good indicator of a species’ diet, predictions could be made regarding enamel thickness distribution in primates with different diets. In the 1980s various methodologies for measuring enamel thickness were discussed: Martin (1983, 1985) pointed out that a single linear measurement of the enamel thickness, especially a linear measurement taken from a worn tooth (Kay, 1981; Benefit, 1987) or naturally fractured tooth (Andrews & Tobien, 1977), was of limited value. He suggested that enamel volume was the optimum measurement for comparative purposes. However, measurements of enamel volume cannot describe the pattern of enamel thickness distribution. Furthermore, it is difficult to calculate enamel volume non-destructively, but recent developments in computing tomography (CT) have made it easier to examine enamel thickness distribution and volume. Linear enamel thickness perpendicular to the dentino-enamel junction (DEJ) on the cross-sectional CT image of a tooth can be easily obtained for comparison of the patterning of enamel
thickness distribution (Macho & Thackeray, 1992; Macho & Berner, 1993, 1994; Schwartz, 2000b). Over the last few years the distribution of enamel over the tooth crown has been studied, and extensively explored in thick-enamelled hominoids (Macho & Thackeray, 1992; Macho & Berner, 1993, 1994; Macho, 1994; Gantt, 1998; Schwartz, 2000a,b), but not in thinenamelled species. Although inspection of unworn teeth should elucidate the function of worn teeth, analyses of worn tooth function have rarely been attempted. To address this deficiency, in this study, unworn lower molars from red colobus (Procolobus badius), one of the most specialized folivores among living primates, and the more frugivorous Japanese macaque (Macaca fuscata) were studied. The aim was to determine whether teeth are likely to retain their functional efficiency as wear progresses in accordance with their dietary adaptation. CT images of unworn teeth were analysed to gain an appreciation of overall tooth geometry and enamel thickness and distribution. Specifically, this study concentrates on the functional buccal side of the lower second molar. Procolobus badius on Tiwai Island in Sierra Leone eats a diet consisting of 20% mature leaf parts, 32% young leaf parts, 16% flowers, and 31% fruits and seeds (Davies et al., 1999). In the Kibale Forest of western Uganda, leaf parts comprise over 70% of the animal’s diet (Struhsaker, 1978), including 14·5% leaf buds, 27·15% young leaf parts, 23·65% mature leaf parts, and 8% leaves of unknown age. Macaca fuscata on Yakushima Island in Japan has a diet consisting of 76% fruits, 18% leaf parts, and 3% invertebrates (Maruhashi, 1980). Hence, the diets of the two species are considered sufficiently different to test the hypothesis that enamel thickness distribution is indicative of a species’ habitual diet. This paper has five objectives: (1) to determine whether enamel volume is related
607
to preferred diet; (2) to determine whether enamel volume correlates with crosssectional enamel area; (3) to determine whether enamel distribution differs between species; (4) to determine what kind of relationship exists between patterning of enamel distribution and tooth geometry; and (5) to determine whether folivorous species are likely to retain a functional cutting edge throughout life. Materials and methods Materials Eight unworn M2s of P. badius and six unworn M2s of M. fuscata were used in this study. The P. badius sample included five females and three specimens of unknown sex. All of the M. fuscata specimens were females. These specimens are housed at the Primate Institute of Kyoto University, Inuyama, Japan. The origin/age of the P. badius specimens is unknown. The M. fuscata specimens were collected from the wild and range between 3·0 and 6·5 years of age. CT scanning The molars were scanned using a pQCT scanner (peripheral quantitative computed tomography: XCT Research SA+, Norland & Stratec Co.) at the Laboratory of Physical Anthropology, Kyoto University. Specimens were mounted on a wooden stage using adhesive tape so that the cervical plane was parallel to the plane of the line of X-ray beams (Figure 1). The specimens were scanned with a tube voltage of 49·9 kV and a tube current of 0·45 mA. The thickness of the slice and the interval of the adjacent slice were both 0·1 mm. The pixel size was 0·1 mm and each CT image had a pixel matrix of 202202. Determination of the boundary between different materials or between the air and solids can be problematic in CT images. CT values change gradually rather than intermittently from the level of one tissue to that of another at such boundaries (Figure 2). In
Figure 1. Diagram of pQCT. X-ray beams irradiated from the X-ray Source are sent through the specimen toward the Detector. The specimens were placed on a wooden stage with the cervical plane parallel to the plane of the X-ray beams.
this study, the criterion known as a half maximum height (HMH) was adopted, in which the boundary of two materials having different density is determined by the median of the CT number levels of the two materials (Koehler et al., 1979; Ullrich et al., 1980; Magnusson, 1987; Spoor et al., 1993; Ohman et al., 1997). For example the CT value of the boundary between air (CT value 0) and dentine (CT value 1800) is 900. In CT images of a tooth, the air–enamel, air–dentine and enamel–dentine boundaries must be determined. However, the lower thresholds of enamel calculated from the HMH values were different at the air– enamel and enamel–dentine boundaries. The HMH value at the air–enamel boundary is always lower than the HMH value at the enamel–dentine boundary. If the former value is adopted as the lower threshold of enamel it results in an overestimation of the enamel thickness at the dentine–enamel junction (DEJ). On the other hand, if the latter value is chosen as the lower threshold of enamel, CT values that are lower than this value and higher than the HMH value at the true air–dentine boundary are recognized as dentine (Figure 2). This results in
608
.
Figure 2. CT image, CT number profile, diagram of cross-sectional CT image of a tooth, and CT number matrix. (a) CT image of M. fuscata sliced parallel to cervix plane at mid-level of the crown. (b) CT number profile in a linear trace between M and N. (c) Diagram of cross-sectional CT image of a tooth, and CT number profile in a linear trace between O and P. The boundaries between enamel, dentine and air were determined by the method of (HMH) respectively. Enamel–air boundary is the median of enamel CT number and air CT number. Enamel–dentine boundary is the median of enamel CT number and dentine CT number. Dentine–air boundary is the median of dentine CT number and air CT number. (d) A part of the CT number matrix including the enamel–dentine boundary and the enamel–air boundary referred to a square on the cross-sectional CT image (a). Filter Program I converts CT number into four values, and filter II removes non-existent ‘‘dentine area’’ on the outer surface of enamel. In diagram (d) solid areas indicate true enamel, dentine or air area, and areas with oblique lines indicate non-existent ‘‘dentine area’’.
the operational formation of non-existent dentine, partly on the outer surface of the enamel. Thus, the lower thresholds of enamel at the air–enamel boundary and dentine–enamel boundary were determined independently (Yamanaka et al., 2001) (Figure 2). In order to determine the true boundaries, two filter programs were used. The first filter program converts CT numbers into one of four values (Figure 2): the value below the air–dentine boundary is 0, the value between the air–dentine and air–enamel boundary is 1, the value between the air–enamel and dentine–enamel boundary is 2 and the value over the dentine–enamel boundary is 3. The second filter program then converts nonexistent ‘‘dentine areas’’ on the outer surface of enamel into air and enamel, respectively. With regard to the converted CT values using the first filter program in the area of non-existent dentine, the value 1 is converted into the air value (0) and the value 2 into the enamel value (3) (Figure 2). In the case shown in Figure 2, the width of the non-existent dentine is less than three pixels. This width depends on the CT scanner and its settings (i.e., the tube voltage or the tube current). The width of true dentine area contact with enamel, however, is considerably wider. By running the second filter program this can be corrected. Measurements For the three dimensional (3-D) reconstruction of the tooth, the Application Visualization System (AVS, Stardent Computer Inc.) was used on a workstation (Silicon Graphics Octane). Using 3-D reconstructed teeth on the AVS, maximum mesio-distal length was measured. Each 3-D reconstructed tooth was resliced on the AVS to obtain linear measurements and enamel area in crosssection in the anatomically determined plane that passes through the apices of the dentine horns of the mesial cusps perpen-
609
dicular to the cervical plane. The cervical plane is determined by the three points on the cervical margin across the mesiobuccal ridge, distobuccal ridge and mesiolingual ridge. Resliced images were converted to bitmap (BMP) image, and linear measurements and cross-sectional enamel area were obtained using a computer. This study concentrates on measurements on the buccal half of the crown, since wear on P. badius teeth occurs mainly in this region (Walker & Murray, 1975; Teaford, 1983; Benefit, 1987). The following measurements were recorded: for enamel volume (EV, Table 1) the number of the pixels that displayed the enamel value (3) was counted on each CT image forming one tooth, and multiplied by 0·001 (volume of a voxel is 0·10·1 0·1 mm). Cross-sectional enamel area was measured on the bucco-lingual cross-section (EA, Table 1). The number of pixels that displayed the enamel volume (3) was counted on the cross-sectional. In order to obtain an overall approximation of tooth size and shape, three measurements were taken: maximum mesio-distal length [ML, Table 1, Figure 3(b)], mesial width along the cervical plane [MCW, Table 1, Figure 3(c)], and height of the mesio-buccal cusp or protoconid [MBCH, Table 1, Figure 3(c)], size (CS) being the product of these measurements: CS=MLMCWMBCH
(1)
Equation 1 is used to obtain on overall approximation of tooth size. Such measurements as ML, MCW and MBCH are also preferred over tooth volume, as it has been used in several previous studies (Gould, 1975; Kay, 1975; Gingerich, 1977; Gingerich et al., 1982; Janis, 1988). Furthermore, tooth roots are not always fully formed when the tooth is unworn, and the cervical margins are not straight (RamirezRozzi, 1993). This confounds calculation of
.
610
Table 1 List of measurements (see Figure 3) Dimension EV EA
Total enamel volume Cross-sectional enamel area
ML
Maximum mesio-distal length
MCW
Mesial width
MBCH
Height of the protoconid
BSA
Buccal surface angle
ETBP
Enamel thickness on the buccal slope of the protoconid
ETCP
Enamel thickness on the cusp tip of the protoconid Enamel thickness on the lingual slope of the protoconid
ETLP
ET ERW
Enamel thickness along the lateral wall of the crown Enamel rim width
dentine area underneath the enamel cap. The flare of the protoconid was measured as the buccal surface angle [BSA, Table 1, Figure 3(a)] formed by the intersection of two lines. One line connects the lowest point of the DEJ on the buccal side and the enamel tip of the protoconid, and the other line passes through the lowest points of the DEJ on the buccal and lingual sides (the cervical line). Three enamel thicknesses of the protoconid are defined [Table 1, Figure 3(d)] as (1) the enamel thickness on the buccal slope of the protoconid (ETBP), which is a linear thickness perpendicular to the DEJ on the buccal slope of the protoconid situated 0·5mm from the dentine horn of the protoconid, (2) the enamel thickness on the cusp tip of the protoconid (ETCP), which is the
Definition Whole enamel volume of M2. Enamel area on the bucco-lingual cross-section through mesial cusps. Maximum distance between two points on the mesial and distal crown surface. Distance between the lowest points of the DEJ on the cervical plane buccal and lingual sides. Minimum distance between the cusp tip of the protoconid and the cervical plane. Angle formed by the intersection of the line which connects the lowest point of the DEJ on the buccal side and the apex of the enamel tip of the protoconid, and the line which passes through the lowest points of the DEJ on the buccal and lingual sides. Linear thickness perpendicular to the DEJ on the buccal slope of the protoconid at 0·5 mm from the dentine horn of the protoconid. Distance between the cusp tip and the dentine horn of the protoconid. Linear thickness from the dentine horn of the protoconid perpendicular to the outer enamel surface (OES) on the lingual slope of the protoconid. Linear thickness perpendicular to DEJ from landmark 2, 3 and 4. Thickness at each landmark with an elevation of 16·2 in P. badius and 12·4 in M fuscata
distance between the cusp tip and the dentine horn of the protoconid, and (3) the enamel thickness on the lingual slope of the protoconid (ETLP), which is a linear thickness from the dentine horn of the protoconid perpendicular to the outer enamel surface on the lingual slope of the protoconid. To evaluate enamel thickness along the lateral face of the tooth, four landmarks were defined on the buccal DEJ surface of the tooth at different levels [Figure 4(a)]. Each landmark was intersected by lines that were parallel with the cervical line (B–L) and the buccal DEJ surface. Landmark 1 (LMK1): at the level of the dentine horn of the protoconid. Landmark 2 (LMK2): through the lowest point of intercuspal fissure on OES. Landmark 3 (LMK3):
611
Figure 3. Cross-section through the mesial cusps of a mandibular molar showing: (a) the angle—BSA; and (b) the six linear measurements—ML, (c) MCW and MBCH; and (d) ETCP, ETBP and ETLP (see Table 1 for list of abbreviations and definitions).
through the lowest level of the DEJ between the buccal and lingual dentine horns (O–P). Landmark 4 (LMK4): at the mid-level between the cervical line B–L and the line O–P. The distribution of enamel on the buccal side was measured in two ways. The enamel thickness (ET) was measured perpendicular to the DEJ from LMK2, LMK3 and LMK4 (Figure 4(b)]. Although the enamel volume is considered by some researchers to be the optimum measurement for comparative purposes (Martin, 1983, 1985), it is meaningless when the patterning of enamel thickness distribution is being discussed. The linear enamel thickness perpendicular to the DEJ allows for comparison of the patterning of enamel thickness distribution (Macho & Thackeray, 1992; Macho & Berner, 1993,
1994; Schwartz, 2000b), but does not provide information about the thickness of the enamel rim after dentine exposure. Benefit (1987) measured thickness of the enamel rim of worn cercopithecine teeth, and discussed its functional implications, so did Kay (1981, Table 1). In order to obtain an estimate of the enamel rim width for evaluation of its functional meaning in different tooth wear stages [Figure 4(c)–(e)], the enamel rim width (ERW) was measured as the thickness at each reference point with an elevation of 16·2 (in P. badius), and 12·4 (in M. fuscata) from the cervical line. These angulations were previously established empirically to be that found in worn teeth of P. badius and M. fuscata respectively (Shimizu, in prep). In this study 39 P. badius and 29 M. fuscata M2’s were used for the
612
.
Figure 4. Cross-section through the mesial cusps showing (a) the reference points for measuring the enamel thickness and the enamel rim width. LMK1 is defined as the dentine horn of the protoconid. And the other points are defined as the intersections on the buccal DEJ surface of tooth crown relative to a line parallel to the crown base (line B–L) through the lowest point of intercuspal fissure (line Q–R) for LMK2, the lowest point of DEJ between cusps (line O–P) for LMK3, and the midpoint of the line O–P and the line B–L (line M–N) for LMK4; (b) three dimensions of the enamel thickness (ET2–4). All measurements are perpendicular to the DEJ from LMK2–4; (c)–(e) the definition of the enamel rim width (ERW). The enamel rim formed after dentine exposure has an angle relative to the cervical plane of: (c) is the angle of the enamel rims relative to the cervical plane; (d) 16·2 for P. badius; (e) 12·4 for M. fuscata.
analysis. These specimens included various stages of tooth wear corresponding to Landmark 1 to 4 in this study. The average
angular measurement for P. badius was 16·2 (S.D. 3·77) and the average angular measurement for M. fuscata was 12·4 (S.D. 2·11).
613
Table 2 Comparison of Procolobus badius and Macaca fuscata in enamel and occlusal dimensions
Species P. badius Mean S.D. CV n M. fuscata Mean S.D. CV n T-value Significance
ML/MCW (%)
MBCH/MCW (%)
BSA
5·1 0·54 10·57 8
147·3 16·15 10·98 8
98·9 13·78 13·94 8
72·0 5·34 — 8
0·2 0·04 14·56 8
6·6 0·43 6·51 6 5·62 ***
131·4 5·29 4·01 6 2·29 *
80·2 5·47 6·81 6 3·31 **
54·7 2·89 — 6 7·12 ***
0·3 0·03 9·85 6 3·11 **
EV (mm3)
EA (mm2)
MCW (mm)
45·9 2·75 5·99 8
8·4 1·48 17·55 8
91·1 10·48 11·51 6
11·5 0·97 8·78 6
VI (degrees)
One asterisk indicates that P<0·05, two asterisks that P<0·01, and three asterisks indicates that P<0·001 using Student’s t-test. Total enamel volume (EV), cross-sectional enamel area (EA), mesial width on the cervical plane (MCW), the relative maximum mesio-distal length (ML), the relative height of the protoconid (MBCH), angle of buccal slope (BSA), angle of lingual slope (LSA) and volume index (VI).
Since there is a difference in molar size between P. badius and M. fuscata all linear measurements were divided by the mesial width (MCW) for normalization. Interspecific differences were tested using the Student’s t-test, and paired t-tests were used to compare enamel thickness (ET) and enamel rim width (ERW) in P. badius and M. fuscata. In order to assess enamel thickness in relation to enamel volume and enamel cross-sectional area, the linear dimensions ETLP, ETCP, ETBP, ET2, ET3, and ET4 are averaged to calculate mean enamel thickness. When enamel thickness is compared to enamel cross-sectional area, the values of enamel thickness should be obtained independently of the enamel cross-sectional area. In order to normalize the enamel volume data, the enamel volume index (EVI) was calculated: EV EVI= CS
(2)
Where EV is enamel volume, and CS is crown size [see Equation (1)].
Results The Student’s t-test was used to examine interspecific difference in enamel volume index (EVI). A significant interspecific difference was found between P. badius (0·2) and M. fuscata (0·3) (Table 2). The enamel volume (EV) of M. fuscata (91·1mm3) was almost twice as large as that of P. badius (45·9 mm3) (Table 2). Moreover, in P. badius the coefficient of variation of the enamel volume (5·99) was small compared with M. fuscata (11·51). Enamel crosssectional area (EA) of P. badius (8·4 mm2) was also smaller than EA of M. fuscata (11·5 mm2) (Table 2). However, the coefficient of variation of cross-sectional enamel area in P. badius (17·55) was larger than it was in M. fuscata (8·78). Descriptive statistics of linear measurements are given in Table 2. The Student’s t-test was used to examine interspecific difference in MCW, ML, MBCH, and BSA. A significant interspecific difference in MCW was observed between P. badius (5·1 mm) and M. fuscata (6·6 mm), so maximum length (ML) and height of the protoconid
.
614
Table 3 Comparison of Procolobus badius and Macaca fuscata in enamel thickness Species P. badius Mean S.D. n M. fuscata Mean S.D. n T-value Significance
ET(LMK4)
ET(LMK3)
ET(LMK2)
ETBP
ETCP
ETLP
9·3 2·10 8
12·6 2·45 8
12·7 1·74 8
11·1 2·05 8
5·7 0·96 8
2·5 0·33 8
8·1 1·09 6
13·1 0·68 6
15·8 1·54 6
15·4 1·02 6
13·0 0·76 6
7·4 0·76 6
1·28 N.S.
0·51 N.S.
3·49 **
4·71 ***
15·29 ***
16·43 ***
Two asterisks indicates that P<0·01, and three asterisks that P<0·001 using Student’s t-test. Enamel thickness on three landmarks along the lateral wall of crown (ET), enamel thickness on the buccal slope of the protoconid (ETBP), enamel thickness on the cusp tip of the protoconid (ETCP), enamel thickness on the lingual slope of the protoconid (ETLP).
(MBCH) were divided by MCW for normalization. In relation to MCW maximum length of P. badius molar (147·3) was significantly larger than the maximum width of the M. fuscata molar (131·4). Height of the P. badius protoconid (98·9) was also significantly larger than the height of the M. fuscata protoconid (80·2). Procolobus badius had a significantly greater BSA (72·0) than M. fuscata (54·7). All of the linear measurements of enamel thickness were divided by mesial width for normalization and compared using the Student’s t-test (Table 3). Three of the relative enamel thicknesses on the protoconid were thinner in P. badius than in M. fuscata. Relative enamel thickness on the buccal side (ETBP) was 11·1 in P. badius and 15·4 in M. fuscata, while the cusp tip (ETCP) was 5·7 in P. badius and 13·0 in M. fuscata. On the lingual side ETLP was 2·5 in P. badius and 7·4 in M. fuscata. Procolobus badius had significantly thinner enamel along the lateral wall of the crown near the cusp tip (LMK2: 12·7) than M. fuscata (15·8). However, relative enamel thickness did not differ significantly at other levels. Interspecific differences in enamel rim width (ERW) were examined using the Student’s t-test (Table 4). The results
Table 4 Comparison of Procolobus badius and Macaca fuscata in the relative enamel rim width Landmarks P. badius Mean S.D. n M. fuscata Mean S.D. n T-value Significance
1
2
3
4
11·6 1·36 8
12·0 1·59 8
11·9 1·16 8
8·9 1·68 8
19·9 2·60 6
19·0 1·77 6
14·6 0·66 6
8·4 1·24 6
7·89 ***
7·54 ***
5·24 ***
0·69 N.S.
Three asterisks indicates that P<0·001 Student’s t-test was used.
indicated that P. badius has a narrower enamel rim than M. fuscata at three of the four levels on the buccal surface of the protoconid. The relative ERWs of P. badius at LMK1 to LMK3 levels were 11·6, 12·0 and 11·9 respectively, while those of M. fuscata were 19·9, 19·0 and 14·6. Intraspecific comparisons between the raw values for enamel thickness (ET) and enamel rim width (ERW) were carried out using the paired t-test and the results are shown in Table 5 and Figure 5. In P. badius, the ERWs had comparable values with the
615
Table 5 Comparison of enamel thickness (ET) and enamel rim width (ERW) (mm) in Procolobus badius and Macaca fuscata P. badius Landmarks ET Mean S.D. n ERW Mean S.D. n T-value Significance
M. fuscata
1/ETBP
2
3
4
1/ETBP
2
3
4
0·6 0·09 8
0·6 0·07 8
0·6 0·09 8
0·5 0·08 8
1·0 0·10 6
1·0 0·10 6
0·9 0·09 6
0·5 0·06 6
0·6 0·06 8
0·6 0·08 8
0·6 0·07 8
0·5 0·06 8
1·3 0·17 6
1·3 0·09 6
1·0 0·03 6
0·6 0·05 6
0·77 N.S.
0·90 N.S.
0·76 N.S.
1·36 N.S.
5·65 **
6·81 **
3·75 *
0·49 N.S.
One asterisk indicates that P<0·05, two asterisks that P<0·01 paired t-test was used.
Figure 5. Distribution profile of generalized enamel thickness (ET) and enamel rim width (ERW) at each reference point. Symbols show mean values and vertical bars indicate one standard deviation. Pb= P. badius; Mf=M. fuscata.
ETs at all levels. On the other hand, in M. fuscata ERWs were significantly larger than the ETs at the two mid levels (at LMK2 and LMK3). Discussion While enamel thickness distribution has featured prominently in palaeoanthropological
studies over the last 20 years (Martin, 1985; Beynon & Wood, 1986; Grine & Martin, 1988; Grine, 1991; Conroy, 1991; Macho & Thackeray, 1992; Schwartz et al., 1998), these studies have been confined to omnivorous species which possess low-crowned and thick-enamelled molars. The present study was designed to test whether enamel distribution in frugivorous primates is
616
.
Figure 6. Correlation between the enamel volume and the cross-sectional enamel area.
comparable to those found in more folivorous species. The species analysed were the folivorous Procolobus badius and the frugivorous Macaca fuscata. Although sample sizes are small, the trends observed in the present study differ between the species and indicate that molars of both species may be adapted to remain functional throughout the individual’s lifetime. Folivorous P. badius has less enamel in proportion to tooth size than frugivorous M. fuscata as determined by linear measurements (Kay, 1981; Benefit, 1987). Also, the relationship between enamel volume and cross-sectional enamel area seems to be different between the two species. Macaca fuscata have almost the same coefficient of variation for enamel volume as for crosssectional enamel area, whereas P. badius have a much larger coefficient for the crosssectional enamel area. This may be due to the fact that P. badius does retain the same amount of enamel irrespective of tooth size (Figure 6). The enamel area correlates positively with the height of the protoconid, but there are no significant relationships with regard to other dimensions. Hence, in P. badius, molars that differ in size also differ in shape and proportions. Conversely, in M.
fuscata, as the volume of the enamel and/or the area of enamel increases, the length and width of the crown, as well as the height of the protoconid become greater (Table 6). In other words, molars which differ in size do not differ in shape. The results show that the molars of P. badius possess thin enamel, uniform distribution of the enamel on buccal crown surface, and less-flared crowns. Conversely, the molars of M. fuscata possess thick enamel and flared crowns. Most of these characters have been discussed in previous work (Delson, 1973; Kay, 1981; Benefit, 1987). The most important finding in this paper is that these characters, which other researchers have discussed separately, are correlated with each other and may relate to the function of the enamel rims that are formed after dentine is exposed. Sharp ridges are required for the efficient shearing of leaves (Lucas & Teaford, 1994). In order for the enamel edges to efficiently shear leaves, the enamel edges need to be thin and sharp. The width of the enamel rims that surround the dentine patches on the cusp tips is determined by the enamel thickness (ET) and relative position between the buccal slope of the protoconid
617
Table 6 Correlation between enamel volume (EV) or cross-sectional enamel area (EA), and length (ML), width (MCW), height (MBCH) of the crown or mean enamel thickness (AET) P. badius ML MCW MBCH AET
EV EA EV EA EV EA EV EA
y= 3·4x+19·6 y= 0·3x+8·4 y= 2·5x+14·1 y=0·2x+4·69 y=1·6x0·6 y=1·3x+1·3 y=0·4x0·8 y=0·1x+0·2
and the enamel rim plane. When the enamel rims meet the buccal slope of the molar crown at a right angle, the enamel rim width has the lowest values. As this angle becomes more acute or obtuse, the enamel rim width becomes greater. Because the angle in P. badius approximates a right angle (88·2) and the enamel is thinner, it maintains narrow enamel rims. Enamel rim width is comparable to enamel thickness at all levels (Table 5 and Figure 5). How soon after molar eruption does a functional enamel rim form in P. badius? The answer to this question lies in the enamel distribution around the cusp tip of the protoconid. The thin enamel distribution on the lingual slope of the protoconid in P. badius contributes not only to rapid exposure of the dentine in this area, thereby forming sharp enamel rims on the buccal side of the protoconid, but also helps to shift leaf processing from the shearing crests to the enamel edges. Dentine exposure in colobine monkeys first appears on the lingual side of the protoconid cusp tip (Janis, 1984; Benefit, 1987; Shimizu, 2001), and correlated with this, P. badius has the thinnest enamel distribution on the buccal side of the crown (Table 3). By exposing dentine on the lingual side of the protoconid cusp tip, P. badius appears to produce sharp enamel edges. After dentine is exposed, the buccal shearing crests, which have an important role in shearing leaves during phase I of the
M. fuscata r2 =0·23 r2 =0·14 r2 =0·11 r2 =0·00 r2 =0·08 r2 =0·67 r2 =0·30 r2 =0·32
y=2·1x0·8 y=1·73x+2·96 y=2·0x2·3 y=1·9x+0·2 y=1·7x2·4 y=2·0x1·3 y=0·3x0·4 y=0·4x0·4
r2 =0·73 r2 =0·35 r2 =0·60 r2 =0·42 r2 =0·49 r2 =0·49 r2 =0·52 r2 =0·65
chewing cycle (Kay, 1975), gradually wear down and eventually disappear. However, the length of the shearing crests decreases through wear, while the length of the enamel edges formed on the buccal side of the dentine patches increases (Shimizu, 2001). It would seem that the enamel edges substitute functionally for the buccal shearing crests. In addition, a less-flared crown and uniform enamel distribution on the buccal surface of P. badius molars produce enamel edges of constant width during the course of wear. In worn P. badius molars the length of the enamel edges is equivalent to the perimeter of the worn surface on the buccal side of crown. The perimeter of the worn surface does not change much, regardless of the wear progress, because of their less-flared crown. These observations indicate that P. badius retains an effective shearing crest/ enamel rim throughout life. In contrast to colobine monkeys, attrition of the lower molars in macaques begins as a round depression at the protoconid tip and, eventually, enamel is removed from the tip of the protoconid to form a flat surface (Gantt, 1979). After that, the dentine is exposed at the centre of the flattened enamel. The enamel thickness on the cusp tip of the protoconid is greater in M. fuscata than it is in P. badius (Table 5 and Figure 5). The thick enamel distribution on the protoconid cusp tip makes dentine exposure difficult. Its importance for retaining a large
618
.
surface for crushing and grinding may be another factor, however. In M. fuscata the angle formed by the enamel rims and the buccal slope of the molar crown is 67·1 and the enamel is thicker, such that M. fuscata possesses wider enamel rims than would be expected from measures of linear enamel thickness alone. In fact, ERWs are 30% greater at LMK2 compared with ET, and 10% greater at LMK3. Macaca fuscata not only possesses thick enamel on the buccal slope of the protoconid, but also possesses pronounced flare of the molar crown. As a result, the enamel rim of M. fuscata is wider than the enamel thickness on the buccal slope of the protoconid. In conclusion, a significant difference in the relative enamel volume is observed between P. badius and M. fuscata, and enamel volume is not correlated with crosssectional enamel area. In P. badius, the enamel rims are formed quickly, as enamel on the lingual slope of the protoconid is considerably thinner than on the buccal slope of the protoconid. On the buccal surface of the crown the thin enamel is distributed uniformly and the buccal flare is poorly developed. As a consequence, a sharp cutting edge is retained throughout the functional lifetime of the tooth. In M. fuscata, enamel on the cusp tip of the protoconid is thick, so that the dentine is difficult to expose and occlusal relief forms flattened crushing surfaces. After dentine exposure, thick enamel and well-developed buccal flare help to retain wide crushing or grinding surfaces. Folivorous red colobus monkeys have thin enamelled molars that represent a different adaptive strategy for mastication compared with frugivorous Japanese macaques, which have thick enamelled molars. Maintaining a large crushing/ grinding area is important for M. fuscata, so their molars have well-developed buccal flare and thick enamel. Retaining a sharp cutting edge is important for P. badius,
however, so their molars have a poorly developed buccal flare and thin, uniformly distributed enamel.
Acknowledgements I thank Professor Hidemi Ishida, Dr Masato Nakatsukasa and other staff of the Laboratory of physical anthropology, Kyoto University for their helpful suggestions and insightful comments. Dr Hironori Takemoto of ATR helped me analyze the CT data. I acknowledge helpful discussions with Dr Yuzuru Hamada, Dr Toshio Mouri, Dr Yutaka Kunimatsu, Mr Takeshi Nishimura and Ms Michika Kondo of the Primates Research Institute, Kyoto University, and Mr Richard Abel of The University of Liverpool. My special thanks are due to Dr Richard Kay of Duke University and Dr Gabriele Macho of The University of Liverpool for valuable advice. This work was supported by a fellowship of the Japan Society for the Promotion of Science for Japanese Junior Scientists. This paper was much improved with the help of comments from three anonymous reviewers.
References Andrews, P. & Tobien, H. (1977). New Miocene locality in Turkey with evidence on the origin of Ramapithecus and Sivapithecus. Nature 268, 699–701. Benefit, B. R. (1987). The molar morphology, natural history, and phylogenetic position of the Middle Miocene monkey Victoriapithecus. Ph.D. Dissertation, New York University. Beynon, A. D. & Wood, B. A. (1986). Variations in enamel thickness and structure in east African hominids. Am. J. phys. Anthrop. 70, 177–193. Conroy, G. C. (1991). Enamel thickness in south African australopithecines: noninvasive evaluation computed tomography. Palaeont. afr. 28, 53–59. Davies, A. G., Oates, J. F. & Dasilva, G. L. (1999). Patterns of frugivory in three West African colobine monkeys. Int. J. Primatol. 20, 327–357. Delson, E. (1973). Fossil colobine monkey of the Circum-Mediterranean region and the evolutionary history of the Cercopithecidae (Primates, Mammalia). Ph.D. Dissertation, Columbia University.
Gantt, D. G. (1979). Patterns of dental wear and role of the canine in Cercopithecinae. Am. J. phys. Anthrop. 51, 353–360. Gantt, D. G. (1998). Evolutionary and functional significance of hominoid tooth enamel. Connect. Tissue Res. 39, 195–206. Gingerich, P. D. (1977). Correlation of tooth size and body size in living hominoid primates, with a note on relative brain size in Aegyptopithecus and Proconsul. Am. J. phys. Anthrop. 47, 395–398. Gingerich, P. D., Smith, B. H. & Rosenberg, K. (1982). Allometric scaling in the dentition of primates and prediction of body weight from tooth size in fossils. Am. J. phys. Anthrop. 58, 81–100. Gould, S. J. (1975). On the scaling of tooth size in mammals. Am. Zool. 15, 351–362. Grine, F. E. (1991). Computed tomography and the measurement of enamel thickness in extant hominoids: implications for its palaeontological application. Palaeont. afr. 28, 61–69. Grine, F. E. & Martin, L. B. (1988). Enamel thickness and development in Australopithecus and Paranthropus. In (F. E. Grine, Ed.) Evolutionary History of the ‘Robust’ Australopithecines, pp. 3–42. New York: Aldine de Gruyter. Hartman, S. E. (1988). A cladistic analysis of hominoid molars. J. hum. Evol. 17, 489–502. Janis, C. M. (1984). Prediction of primate diets from molar wear patterns. In (D. J. Chivers, B. A. Wood & A. Bilsborough, Eds) Food Acquisition and Processing in Primates, pp. 331–340. New York: Plenum Press. Janis, C. M. (1988). An estimation of tooth volume and hypsodonty indices in ungulate mammals, and the correlation of these factors with dietary preference. Me´m. Mus. Natn. Hist. Nat., Paris, (se´rie C). 53, 367–387. Kay, R. F. (1975). The functional adaptation of primate molar teeth. Am. J. phys. Anthrop. 43, 195–216. Kay, R. F. (1978). Molar structure and diet in extant Cercopithecidae. In (P. M. Butler & K. A. Joysey, Eds) Development, Function and Evolution of Teeth, pp. 309–339. London: Academic Press. Kay, R. F. (1981). The nut-crackers—a new theory of the adaptations of the Ramapithecinae. Am. J. phys. Anthrop. 55, 141–151. Kay, R. F. & Hiiemae, W. L. (1974). Jaw movement and tooth use in recent and fossil primates. Am. J. phys. Anthrop. 40, 227–256. Kay, R. F. & Hylander, W. L. (1978). The dental structure of mammalian folivores with special reference to primates and phalangeroids (Marsupialia). In (G. G. Montgomery, Ed.) The Ecology of Arboreal Folivores, pp. 173–191. Washington, DC: Smithsonian Institution Press. Koehler, P. R., Anderson, R. E. & Baxter, B. (1979). The effect of computed tomography viewer controls on anatomical measurements. Radiology 130, 189– 194. Lanyon, J. M. & Sanson, G. D. (1986). Koala (Phascolarctos cinereus) dentition and nutrition. II. Implications of tooth wear in nutrition. J. Zool., Lond. (A) 209, 169–181.
619
Lucas, P. W. & Teaford, M. F. (1994). Functional morphology of colobine teeth. In (A. G. Davies & J. F. Oates, Eds) Colobine Monkeys: Their Ecology, Behaviour and Evolution, pp. 173–203. Cambridge: Cambridge University Press. Lumsden, A. G. S. & Osborn, J. W. (1977). The evolution of chewing: a dentist’s view of palaeontology. J. Dent. 5, 269–287. Macho, G. A. (1994). Variation in enamel thickness and cusp area within human maxillary molars and its bearing on scaling techniques used for studies of enamel thickness between species. Arch. Oral Biol. 39, 783–792. Macho, G. A. & Berner, M. E. (1993). Enamel thickness of human maxillary molars reconsidered. Am. J. phys. Anthrop. 92, 189–200. Macho, G. A. & Berner, M. E. (1994). Enamel thickness and the helicoidal occlusal plane. Am. J. phys. Anthrop. 94, 327–337. Macho, G. A. & Spears, I. R. (1999). Effects of loading on the biomechanical behavior of molars of Homo, Pan, and Pongo. Am. J. phys. Anthrop. 109, 211–227. Macho, G. A. & Thackeray, J. F. (1992). Computed tomography and enamel thickness of maxillary molars of Plio-Pleistocene hominids from Sterkfontein, Swartkrans, and Kromdraai (South Africa): an exploratory study. Am. J. phys. Anthrop. 89, 133–143. Magnusson, A. (1987). Object size determination at computed tomography. Upsala J. Med. Sci. 92, 277–286. Martin, L. M. (1983). The relationships of the later Miocene Hominoidea. Ph.D. Dissertation, University College London. Martin, L. M. (1985). Significance of enamel thickness in hominoid evolution. Nature 314, 260–263. Maruhashi, T. (1980). Feeding behaviour and diet of the Japanese monkeys (Macaca fuscata yakui) on Yakushima Island, Japan. Primates 21, 141–160. McArthur, C. & Sanson, G. D. (1988). Tooth wear in eastern grey kangaroos (Macropus giganteus) and western grey kangaroos (Macropus fuliginosus), and its potential influence on diet selection, digestion and population parameters. J. Zool., Lond. 215, 491–504. Ohman, J. C., Krochta, T. J., Lovejoy, C. O., Mensforth, R. P. & Latimer, B. (1997). Compact bone distribution in the femoral neck of hominoids: implications for the locomotion of Australopithecus afarensis. Am. J. phys. Anthrop. 104, 117–131. Ramirez-Rozzi, F. V. (1993). Tooth development in East African Paranthropus. J. hum. Evol. 24, 429– 454. Schwartz, G. T. (2000a). Enamel thickness and the helicoidal wear plane in modern human mandibular molars. Arch. Oral Biol. 45, 401–409. Schwartz, G. T. (2000b). Taxonomic and functional aspects of the patterning of enamel thickness distribution in extant large-bodied hominoids. Am. J. phys. Anthrop. 111, 221–244. Schwartz, G. T., Thackeray, J. F., Reid, C & van Reenan, J. F. (1998). Enamel thickness and the topography of the enamel-dentine junction in south
620
.
African Plio-Pleistocene hominids with special reference to the carabelli trait. J. hum. Evol. 35, 523–542. Shimizu, D. (2001). Functional morphology of molars of folivorous primates. Ph.D. Dissertation, Kyoto University. Shimizu, D. (in prep.). Relationship between tooth wear and chewing efficiency in red colobus (Procolobus badius). Spears, I. R. & Crompton, R. H. (1996). The mechanical significance of the occlusal geometry of great ape molars in food breakdown. J. hum. Evol. 31, 517–535. Spoor, C. F., Zonneveld, F. W. & Macho, G. A. (1993). Linear measurements of cortical bone and dental enamel by computed tomography: applications and problems. Am. J. phys. Anthrop. 91, 469–484. Struhsaker, T. T. (1978). Interrelations of red colobus monkeys and rain-forest trees in the Kibale Forest, Uganda. In (G. G. Montgomery, Ed.) The Ecology of Arboreal Folivores, pp. 397–422. Washington, D.C.: Smithsonian Institution Press.
Teaford, M. F. (1983). The morphology of the lingual notch in macaques and langurs. Am. J. phys. Anthrop. 60, 7–14. Ullrich, C. G., Binet, E. F. Sanecki, M. G. & Kieffer, S. A. (1980). Quantitative assessment of the lumber spinal canal by computed tomography. Radiology 134, 137–143. Ungar, P. & Williamson, M. (2000). Exploring the effects of tooth wear on functional morphology: A preliminary study using dental topographic analysis. Palaeontologica Electronica 3(1), http://wwwodp.tamu.edu/paleo/2000_1/gorilla/issue1_00.htm. Walker, P. & Murray, P. (1975). An assessment of masticatory efficiency in a series of anthropoid primates with special reference to the Colobinae and Cercopithecinae. In (R. H. Tuttle, Ed.) Primate Funcutional Morphology and Evolution, pp. 151–194. Mouton: The Hague. Yamanaka, A., Shimizu, D. & Takemoto, H. (2001). New technique of modeling using CT images. Anthrop. Sci. (Japanese Series) 108, 81–90.