AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 92389-200 (1993) Enamel Thickness of Human Maxillary Molars Reconsidered GABRIELE A. MACHO AND MARGIT E. BERNER Hominid Palaeontology Research Group, Department of Human Anatomy and Cell Biology, The University of Liuerpool, England (G.A.M.),and Department of Anthropology, Natural History Museum Vienna, Austria (M.E.B.) KEY WORDS Enamel Maxillary molars thickness, Orofacial biomechanics, ABSTRACT Forty-four modern human maxillary molars (M1 = 21, M2 = 12, and M3 = 11) were sectioned through the mesial cusps in a plane perpendicular to the cervical margin of the crown. Eight measurements of enamel thickness as well as bucco-lingual (BL) and mesio-distal (MD) diameters were recorded for each tooth in order to investigate differences in these dimensions between tooth categories. Uni- and multi-variate analyses revealed first maxillary molars to have generally thinner enamel than second or third upper molars, especially with regard to the occlusal basin. Furthermore, the decrease of MD diameters from anterior to posterior is greater than that of BL diameters. Principal Component Analysis using enamel thickness measurements resulted in complete separation of first molars, while second and third maxillary molars showed a certain amount of overlap. This finding casts doubt on using an overall measure of “molar enamel thickness” derived from mixed samples of molars for taxonomic purposes. There appears to be a relationship between bite force and enamel thickness such that posterior molars, where masticatory forces are stronger, have thicker enamel than anterior teeth. It is suggested that the gradient of enamel thickness between (and within) teeth in extant and extinct species may thus provide further information about relative wear resistance as well as the biomechanical constraints of the orofacial skeleton. o 1993 Wiley-Liss, Inc. Recently, enamel thickness has become an important character in taxonomic studies of hominoids. For instance, Homo and Pongo are usually characterised as thick-enamelled hominoids, whereas Gorilla and Pan have relatively thinner enamel (Gantt, 1983; Martin 1983,1985;Molnar and Gantt, 1977). The thick enamel in Pongo has been interpreted as either retention of the primitive condition (Martin, 1985) or as parallelism (Beynon et al., 1991). [In his earlier work Schwartz (e.g., 1984) argued that thick enamel in Pongo and Homo could also be a synapomorphy (but see Schwartz, 199011. All hominids, however, are thick enamelled. The functional importance of enamel has also been noted (Molnar and Gantt, 1977) and different dietary regimes are associated with varying enamel thicknesses (Andrews 0 1993 WILEY-LISS, INC and Martin, 1991; Gantt, 1983; Kay, 1981; Shellis and Hiiemae, 1986; Simons, 1976). It has been established that species that habitually eat fruits and hard objects, such as seeds and nuts, or are adapted to a more varied diet, have thick enamelled molars (e.g., Andrews and Martin, 1991; Kay, 1981). By contrast, folivorous taxa are characterised by thin-enamelled molars but thick-enamelled incisors, in particular lingually (Shellis and Hiiemae, 1986). However, even within an individual tooth differ- Received September 21,1992; accepted April 29, 1993. Address correspondence to Dr. Gabriele A. Macho, Hominid Palaeontology Research Group, Dept. of Human Anatomy and Cell Biology, University of Liverpool, P.O. Box 147, Liverpool L69 3BX, England. 190 G.A. MACHO AND M.E. BERNER ences in enamel thickness are observed and are explained in terms of differential functional demands placed on the individual cusps; in maxillary molars enamel is thicker on lingual cusps than it is on buccal cusps, whereas the opposite holds for lower molars. This unequal distribution appears to be common to all Primates which have been examined (Molnar and Gantt, 1977; Molnar and Ward, 1977). Although it would seem that taxonomic and functional differences are well documented, inspection of the literature led us to conclude that most of these assumptions are based on small sample sizes. The maximum number of either upper or lower molars studied for any one species in the published literature is nine (Grine and Martin, 1988), which is too small to determine differences between various molars. While early studies were broad and innovative and helped resolve questions pertaining to the taxonomic status of the later Miocene hominoids, the paucity of data for individual tooth types leaves open the possibility for further investigation. In order to enlarge sample sizes used to measure hominoid enamel thicknesses, non-destructive techniques, such as Computed Tomography (CT), have recently been employed (Conroy, 1991; Grine, 1991; Macho and Thackeray, 1992; Spoor et al., 1993). Owing to inherent limitations of CT, however, data derived from actual sections of teeth seem essential to establish the true range of variation of enamel thickness among hominoids. This is especially true where small dimensions are concerned, such as enamel thickness in the cervical region in all teeth or in the case of occlusal enamel thickness in thin-enamelled species (Spoor et al., 1993). In this study we report the amount of variation in enamel thickness found among maxillary molars of a modern human population. Moreover, we discuss our findings with regard to recent taxonomic studies that have placed great significance on enamel thickness. Since a n understanding of the relationship between structure and function is essential if we aim to comprehend the adaptive significance of variation observed between extant and extinct taxa or to contribute to arguments about evolutionary trends, emphasis is placed on the functional importance of differences in enamel thickness between molars. MATERIAL AND METHODS Thirty-eight unworn modern human maxillary molars (M1 = 16, M2 = 12, M3 = 10) were used in the present study. I n addition, five slightly worn M‘s and one slightly worn M3 were incorporated for reasons given below, but measurements made in these worn areas were coded a s missing values. The teeth were dissected from subadult skulls which came from the Slavic necropolis near Zwentendorf, Lower Austria (10th century A.D.). They were excavated by F. Hampl between 1953 and 1958 (Ladenbauer-Orel, 1971), and are now housed in the Department of Anthropology, Natural History Museum Vienna, Austria. A detailed anthropological description of the material is still pending. Prior to sectioning, bucco-lingual (BL) and mesio-distal (MD) diameters were recorded and Carabelli’s cusps were scored (Scott, 1980) and each tooth was photographed in occlusal view. Thereafter, teeth were cut through the tips of the mesial cusps and perpendicular to the base of the crown with a n ISOMET 11-1180 low speed annular saw producing a 300 pm wide cut. In cases where the tips of the dentine horns were not cut precisely, the tooth was ground down with a rotating diamond disk to the optimum plane section. Thereafter the cut faces were polished with aluminum oxide paste. Each tooth was photographed under a Wild-Photomakroskop at a magnification of 1% Measurements of enamel thickness were recorded in millimeters from enlarged photographs (1:lO) using a Vernier sliding caliper. Figure 1 lists the eight measurements of enamel thickness taken for each tooth. Measurements 1 and 8 were taken perpendicular to the dentino-enamel junction (DEJ) from a point on the DEJ intersecting with a reference plane through the lowest point of the DEJ between the cusps and parallel to the crown base (Macho and Thackeray, 1992). Measurements 3 and 7 correspond to CT(L) and CT(B) (Grine and Martin, 1988), while variables 4 and 6 are those defined as “i” and “ h by Martin ENAMEL THICKNESS DISTRIBUTION BETWEEN MAXILLARY MOLARS I L B Fig. 1. Cross-section through maxillary molar showing enamel thickness measurements used in this study. (1983). Variable 2 is the maximum thickness of the latero-lingual enamel measured a s perpendicular as possible to both the DEJ and the enamel surface (Spoor et al., 1993). Finally, enamel thickness a t the intercuspal fissure was taken from the deepest point a t the occlusal surface to the lowest point of the DEJ. The amount of sexual dimorphism in enamel thickness of M3s was also analysed for specimens which could be reliably sexed on the basis of sexually dimorphic features of the skull and the postcranium. In order to test for measurement errors five teeth were repeatedly measured by both authors on 3 consecutive days. In none of the instances did the differences exceed 1% of the dimension. RESULTS Table 1 lists the descriptive statistics of all measurements taken. Decrease of buccolingual and mesio-distal diameters from anterior to posterior through the series M1-M2-M3was statistically significant, although reduction of the MD diameter was more marked than that of the BL diameter. Analysis of variance (ANOVA) was also employed to test for differences in enamel thickness between tooth types. Six out of eight measurements yielded statistically significant results a t the 1% probability level. Only the greatest thickness along the lingual slope of the protocone (variable 2) and the thickness at the intercuspal fissure (variable 5) did not differ between teeth. Since enamel thickness of molars with reported Carabelli traits were within the range of values reported for other molars, 191 these teeth were included in all statistical analyses. These similarities in enamel thickness were unexpected but may have been due to the fact that the sections were almost exclusively posterior to the Carabelli cusp; this may account for the fact that the measurements taken were unaffected. The statistically significant trends regarding changes in enamel thickness from M1 to M3 are graphically depicted in Figure 2. Each bar shows the mean value, standard deviation and 95% of the observations. While enamel thickness along the tooth walls (variables 1 and 8) tends to decrease posteriorly, all other enamel thickness values increase markedly from first maxillary molars to the more posterior teeth. Overall, the occlusal basin (variables 3 , 4 , 6 , and 7) is endowed with thicker enamel in posterior molars than it is anteriorly (Table 1,Figure 2). Standard deviations and, hence, the coefficients of variance, decrease in a posterior direction for enamel thicknesses along the cusp tips (variables 3 and 7); the differences are statistically significant at the 5% probability level using Cochrans C'test. In order to test whether these statistically significant increases can be explained by individual variation or by sexual dimorphism within the sample, three approaches were taken. First, we compared measurements taken from slightly worn Mls with those of M'S from the same individual (Table 2 ) . Owing to the delayed maturation in humans, all individuals with fully formed second molar crowns had first molars which had already erupted and were slightly worn. However, differential wear in these specimens was minimal and, more importantly, did not affect all areas where measurements were taken. Hence, comparisons were carried out (Table 2). It should be noted that for all other analyses measurements taken from these worn areas were coded as missing values. In all instances, measurements of the occlusal basin (variables, 3 , 4 , 6, and 7) and measurement 2 were smaller in first molars, although all MD and most of the BL diameters became smaller distally. With regard to enamel thickness at the intercuspal fissure there appears to be a slight trend towards decrease from anterior to posterior. Similarly, almost all dimensions of the tooth sides (variables 1and 8 ) G.A. MACHO AND M.E. BERNER 192 TABLE 1. Descriptive statistics of enamel thickness measurements (in millimeters) taken, analysis of variance (ANOVA), and nrobabilitv levels Measurement MD t Median SE SD Min-max BL P Median SE SD Min-max No. 1 ! ! X Median SE SD Min-max No. 2 8 X Median SE SD Min-max No. 3 E X Median SE SD Min-rnax No. 4 E X Median SE SD Min-max No. 5 P Median SE SD Min-max N No. 6 Ti Median SE SD Min-max No. 7 2 Median SE SD Min-max No. 8 E X Median SE SD Min-max N M’ M2 M3 10.51 10.60 0.15 0.67 9.4-1 1.8 21 11.57 11.70 0.11 0.48 10.5-12.4 21 1.56 1.53 0.06 0.27 1.10-2.15 21 1.80 1.80 0.04 0.20 1.41-2.16 21 1.59 1.65 0.07 0.27 1.07-1.99 16 1.30 1.26 0.04 0.19 0.9G1.66 21 1.18 1.16 0.06 0.25 0.75-1.72 21 1.25 1.24 0.03 0.15 0.9G1.50 21 1.04 1.06 0.07 0.30 0.39-1.64 17 1.34 1.36 0.04 0.17 0.95-1.61 21 9.13 9.30 0.19 0.65 8.0-10.2 12 11.45 11.50 0.19 0.64 10.cL12.8 12 1.41 1.28 0.09 0.31 1.08-2.04 12 1.99 1.92 0.06 0.19 1.77-2.26 12 2.17 2.17 0.10 0.34 1.6G2.74 12 1.62 1.59 0.06 0.19 1.35-1.96 12 1.17 1.24 0.14 0.47 0.31-2.06 12 1.73 1.66 0.07 0.26 1.29-2.20 12 1.62 1.64 0.07 0.24 1.141.97 12 1.01 0.91 0.09 0.30 0.59-1.47 12 8.64 8.40 0.20 0.68 7.7-1 0.0 11 10.45 10.20 0.28 0.92 9.cL11.9 were smaller in M2s than they were in MIS. This pattern of differences between teeth of the same individual clearly match the crosssectional data presented above. 11 1.18 1.20 0.06 0.17 0.84-1.40 10 1.87 1.93 0.11 0.34 1.11-2.24 10 2.09 2.05 0.04 0.13 1.94-2.30 10 1.59 1.54 0.07 0.22 1.3cL20.2 Anova F-value P 34.20 0.0000 11.26 0.0001 7.90 0.0010 2.42 0.1014 19.27 0.0000 13.02 0.0000 0.14 0.8674 30.41 0.0000 26.82 0.0000 7.92 0.0012 11 1.11 1.18 0.12 0.38 0.27-1.69 11 1.78 1.87 0.08 0.27 1.41-2.21 11 1.64 1.66 0.05 0.16 1.3S1.88 11 1.08 1.00 0.10 0.32 0.62-1.48 11 Second, it could be argued that there might be a strong bias in sex-ratios between different age categories due to differential mortality rates. A demographic analysis of ENAMEL THICKNESS DISTRIBUTION BETWEEN MAXILLARY MOLARS 193 - Changes in Enamel Thickness M I M3 3.0 No.1 2.5 I I I I No.3 I I I I I No. 4 I I I 2.0 1.5 1.o .5 2.5 2.0 1.5 1.o .5 Fig. 2. Plot of mean values and standard deviation (stippled) of six enamel thickness measurements, by tooth type. Light bars represent 95%of the individuals sampled. (Units on the right are millimeters.) contemporary skeletal series from the same region, however, does not support this argument. Stloukal and Vyhnanek (1976) found mortality rates of girls and boys between 6 years of age and adulthood mortality rates to be approximately the same. Similarly, in modern human populations, there is only a slight bias towards increased male mortal- ity after the age one (Statistisches Handbuch fur die Republik Osterreich, 1971). It is thus unlikely that one particular age group studied here is exclusively represented by one sex only. Finally, individuals whose M3s were analysed could be sexed on the basis of sexually dimorphic features of the skull and the post- G.A. MACHO AND M.E. BERNER 194 TABLE 2. Enamel thickness of teeth of the same individual as well as differences between M's and w s No. Tooth MD BL No. 1 No. 2 No. 3 No. 4 No. 5 No. 6 No. 7 NO. 8 M' M2 10.0 9.3 1 -0.7 10.2 8.9 -1.3 10.6 9.4 -1.2 9.4 8.1 -1.3 9.9 9.6 -0.3 11.7 11.7 1.10 1.24 0.14 1.83 1.50 -0.33 1.54 1.18 -0.36 1.32 1.08 0.24 1.61 1.57 -0.04 2.02 2.07 0.05 1.83 1.92 0.09 2.13 2.26 0.13 1.59 1.80 0.21 1.64 1.90 0.27 11.571 2.34 [0.77] [1.011 2.12 11.111 L1.661 2.60 10.931 L1.311 1.87 10.561 L0.501 1.81 [1.31] 1.22 1.64 0.42 1.05 1.35 0.30 1.50 1.81 0.31 1.09 1.35 0.26 1.16 1.59 0.43 1.29 1.15 -0.14 0.98 1.17 0.18 1.76 1.34 -0.42 1.31 1.28 -0.03 1.07 0.71 -0.35 1.23 1.78 0.55 1.23 1.56 0.33 1.39 2.20 0.81 1.03 1.29 0.26 1.17 1.56 0.39 0.65 1.70 11.051 L1.161 1.80 L0.641 [l.lOl 1.65 10.551 11.011 1.38 [0.371 L1.641 1.83 [0.191 1.37 0.83 -0.55 1.45 1.13 -0.33 1.59 1.34 -0.25 1.16 0.59 -0.57 1.40 1.24 -0.17 41 ~ 51 77 156 178 2 . ~ M' M" M2-Mi M' MZ M2-M' M' M2 M2-M' M' M" M2-M' o 11.8 11.5 -0.3 12.0 11.5 -0.5 10.5 10.0 -0.5 10.8 11.4 0.6 i J indicates wear. TABLE 3. Differences in crown diameters and enamel thickness between sexes ' Males Measurement MD x SD 8.85 0.83 4 N No.2 x SD No. 3 No. 4 x SD N - X SD 10.65 1.15 4 1.21 0.10 4 1.70 0.41 4 2.01 0.07 3 1.47 0.12 4 1.11 SD SD N 0.34 4 1.81 0.30 4 1.60 0.07 4 1.05 0.37 4 Females 8.51 0.61 7 10.33 0.85 7 1.16 0.22 6 1.98 0.27 6 2.12 0.14 7 1.66 0.24 7 1.11 0.43 7 1.76 0.27 7 1.66 0.19 7 1.10 0.31 7 t-value 0.77 P TABLE 4 . Coefficients of first two Principal Components (PC) based on a Principal Component Analysis using siz enamel thicknesses, which were significantly different between teeth 0.46 Measurement 0.53 0.61 0.38 0.72 -1.33 0.22 -1.23 0.25 -1.43 0.19 -0.03 0.98 0.29 0.78 -0.58 0.58 -0.24 0.82 IT-tests between males and females and probability levels are also given. Only specimens with M3s could be reliably sexed. cranium. There were no statistically significant differences between the sexes in any of the measurements studied when t-tests were employed (Table 3). Although male teeth were on average larger with regard to BL and MD diameters, there were no systematic trends in enamel thicknesses in ei- ~ Principal Component Analysis _. PC I1 PC I _ ~ _ _ No.8 0.23 -0.48 -0.46 -0.49 -0.48 0.20 0.64 0.21 0.31 0.08 -0.00 0.66 B variance 55.10 26.91 No. 1 No.3 No. 4 No.6 No. 7 _ ther sex (although females have an average thicker enamel). On the grounds of these findings any bias towards a particular sex within a specific age category seems unlikely. Principal Component Analysis was then employed using the six enamel thickness measurements which were significantly different between molars (Table 1,Fig. 2). This multivariate technique was chosen in order to explore the relationships between a number of variables and to analyse the patterning of enamel thickness between teeth. Table 4 lists the coefficients of the first two components, which account for 82% of the variance. Interestingly, PCI discriminates on shape, while only PCII can be regarded as a size component. This interpretation is based on the fact that the loadings of the coefficients of the first variate have different signs whereas those ofthe second component are all positive (Bilsborough, 1984). (There is one negative coefficient in PCII, . ~ ENAMEL THICKNESS DISTRIBUTION BETWEEN MAXILLARY MOLARS 195 Principal Component Analysis +4 ::I +1 2 2 2 . : : 1 1 1 1 33 1 I: 1 1 ’ 1 1 ............. -3- I Fig. 3. Plot of first two Principal Components (M’ = 1; M2 = 2; M3 = 3). however, which equals a value of zero.) PCI contrasts enamel thickness along the tooth walls (variables 1 and 8) with that of the occlusal basin (variables 3, 4, 6, and 7). As can be seen in Figure 3 this component entirely separates first maxillary molars from the remaining molars. Of the sexed individuals neither males nor females had exclusively positive or negative coefficients. Thus, the possibility of PCII being a sex component can be discarded. DISCUSSION Previous studies stressed the importance of hominoid enamel thickness for taxonomic purposes, although the range of variation in extant taxa had not yet been adequately established. The present study set out to provide such a yardstick for maxillary molars of Homo supiens. Interestingly, enamel thickness not only displays a great amount of variation but, more importantly, it differs systematically between molars. Figure 4 shows the typical distribution of enamel among upper molars. The marked difference occurs in enamel thickness at the occlusal basin of first maxillary molars as compared with those of second and third molars. By and large, enamel is thinnest in M’s. Fur- Fig. 4. Ground sections through mesial cusps of M’, M2. and M3. thermore, separation is complete when more than one variable is studied simultaneously by Principal Component Analysis (Fig. 3). This complete distinction between molars within a single human population raises questions pertaining to the usefulness of calculating a measure of overall “molar enamel thickness” for taxonomic purposes. At the protocone (variable 3) enamel increases by 36%from anterior to posterior, whereas the increase at the paracone (variable 7) is even more substantial, with a value of some 56% (Table 1,Fig. 2). It is interesting to note that Aiello et al. (1991) also suggest that relative enamel thickness tends to increase distally in extant hominoids with M,s displaying the thickest enamel. Similarly, in their study on Plio-Pleistocenehominids Beynon and Wood 196 G.A. MACHO AND M.E. BERNER TABLE 5 . Total crown area (mrn’), BL, and MD diameters of one first and one second maxillary molar used in this studv‘ Formulae usually employed to calculate “relative”or “average” enamel thickness are those devised by Martin (1983, 1985). MarM’ MZ (M2/M1) * 100 tin maintained that the total area of enamel Total crown area 101.5 81.3 80.1% divided by the length of the DEJ, measured 11.6 BL 11.9 97.5% from a section of a tooth through the dentine MD 11.1 8.8 79.3% horns and perpendicular to the crown base, Relatcuelaverage enamel thickness Area of enamel cap “c” 26.4 31.9 120.8% approximates the ideal measure of enamel Area of dentine “b” 47.3 48.4 97.7% volume over the surface area of the DEJ. In 22.2 20.7 Length of DEJ “e” 93.210 ele 1.2 125.0% 1.5 the present study various “relative”and “av111.9% (6 * 100) 109.5 122.5 erage” enamel thicknesses (sensu Grine and 131.010 17.1 L(c/e)/m * 100 22.4 Martin, 1988) were calculated for two teeth 2.92 143.8% [(var3 + v a r 7 ) / 2 1 / 6 2.03 145.8% 2.05 I(var4 + v a r 6 ) 1 2 1 / 6 2.99 (Table 5). ‘For both teeth various “relative” and “average” enamel thicknesses The values obtained render some of the were calculated (Grine and Martin, 1988) and the percentage differassumptions on which these formulae are ences between both teeth were computed. based questionable. Since teeth can rarely be regarded as square or circular objects the length of the DEJ in one sectioned plane (1986) indicate that there might be a slight cannot provide information about its length trend towards enamel thickness increase in another. As a case in point, the first and from anterior to posterior, i.e., between pre- the second molars in Table 5 have almost molars and molars. However, they refrained equal BL diameters as well as lengths of the from commenting on differences between DEJs in the sectioned plane, but they differ markedly in MD diameters and total crown molars due to small sample sizes. Inspection of Martin’s (1983) data re- area (determined by planimetry). As a convealed enamel thickness at the cusp tips to sequence, although all measures of relative be smallest in M’s in Pan troglodytes, Go- enamel thickness are greater in the second rilla gorilla, Pongo pygmaeus, and Homo sa- maxillary molar, the relative volume over piens; in Pongo both values are greatest in the surface area of the DEJ of the second M3s, whereas in all other taxa either M2 or molar would still be underestimated, priM3 yielded the greatest figure, but the dif- marily because the equation has not taken ferences between these posterior teeth are into account the shorter outline of the DEJ not marked. Overall, the percentage in- (about 20%) in MD direction. Similar probcrease in enamel thickness calculated for lems would also arise when teeth of different these species is comparable, or even greater, shapes, such as upper and lower molars, are to that found in this investigation. Simi- compared. It should also be borne in mind larly, the distribution of enamel along the that within a tooth posterior cusps tend to be occlusal basin (i.e., variables 4 and 6) from lower than anterior cusps both at the crown anterior to posterior in Pan and Pongo par- (e.g., Kanazawa et al., 1983, 1988; Mayhall allel the findings for Homo of the present and Kanazawa, 1989) and the DEJ (see Corstudy. All three taxa exhibit smallest values ruccini and Holt, 1989; Korenhof, 1960, in M’s. However, Grine and Martin (1988) 1982; Sakai and Hanamura, 1971, 1973). using various measures of “relative enamel Furthermore, even when the value of “relathickness” observed only “. . . a slight ten- tive” enamel thickness is regarded as repredency for relative enamel thickness to in- senting only one particular section rather crease from M1 to M3 . . .” (p. 12) within the than the enamel volume of the tooth crown, same set of data. This could imply that the the derived values may not provide sensitive overall amount of enamel over the tooth enough measures to equalize within-species crown may be similar between molars differences. Percentage differences in “relawithin a species, but that the patterning of tive” and “average” enamel thickness found enamel over the tooth crown changes from between M1 and M2 range from 112 to 145% anterior to posterior. This assumption was (Table 5 ) . These figures are substantial and are similar to those obtained for linear meatested (Table 5 ) . ENAMEL THICKNESS DISTRIBUTION BETWEEN MAXILLARY MOLARS surements in this study. Therefore, while measures derived from these formulae tend to obscure the patterning of enamel thickness, which may yield valuable information (Macho and Thackeray, 1992; Macho and Berner, in press), they do not appear to provide a good estimate of “relative” andlor “average” enamel thickness within species.’ Hence, only linear measurements were considered in the present study. Studies on extant and extinct species placed emphasis on developmental processes in order to explain the differences found, since identical developmental pathways would constitute strong evidence for homology (Grine and Martin, 1988). Molar crown formation times are broadly similar in pongids (Dean and Wood, 1981; but see Anemone et al., 1991), hominids (Beynon and Wood, 1987), and modern humans (Shellis, 1984; and references therein). Although Shellis’ (1984) data on crown formation times are slightly different from those presented by other investigators, it is significant that he gives the smallest value for second molars. Bearing in mind the similarities in crown formation times the marked differences in enamel thickness between taxa are striking, as are those between human maxillary molars. Differences in enamel thickness between genera have been explained by varying enamel extension and enamel secretion rates (Beynon and Dean, 1988; Beynon and Wood, 1986, 1987). In modern humans deciduous and permanent teeth follow a different pattern of enamel formation as well, whereby the former exhibit a faster enamel extension and secretion rate than the latter. With regard to our data the question arises whether differences in enamel thicknesses between maxillary molars can be explained by these processes as well. A slowing of enamel formation during the cervical stage of crown formation and a proportional increase of appositional enamel in posterior teeth is common to all hominoids (Beynon and Dean, 1988; Dean, 1987). Re- ‘The scaling factor CORl used by Beynon and Wood (1986) is thus a more useful measure since it takes the tooth crown area of each tooth category into account. 197 cently, evidence has been provided that ameloblasts of posterior maxillary molars may exhibit slightly reduced secretion rates in humans (Beynon et al., 1991). Based on these studies we may infer that the thicker enamel of posterior molars may be attained through an increased extension rate during the cuspal stage and/or through a prolonged ameloblast secretory activity in the cuspal region. Both of these mechanisms could occur without prolonging total crown formation. Either explanation would also explain the thickened enamel along the buccal wall of the protocone (variable 4) and the lingual wall of the paracone (variable 6), and would account for relatively thin enamel approaching the cervix or the intercuspal fissure (variable 5). Investigation of enamel thickness, results of the Principal Component Analysis, and the coefficients of variation all indicate that thickening of enamel in second and third maxillary molars is systematic rather than random. On the other hand, BL and MD diameters decrease from anterior to posterior in this population, while the coefficients of variation of these measurements increase (Table 1).Dental anthropologists usually interpret this phenomenon of greater variation of crown diameters posteriorly as support of the field concept (Butler, 1939; Dahlberg, 1945). According to this theory the genetic influence diminishes with increasing distance from the most stable tooth within the morphogenetic field. An alternative hypothesis is that the impact of environmental factors increases resulting in a reduction and greater variability of posterior teeth. It is worthy of note, therefore, that in the present study the variation for enamel thickness at the cusp tips of the protocone and the paracone are both statistically smallest in M3s (Table 1, Fig. 2). In light of the expectations raised by the field theory this finding is surprising and we surmise here that functional constraints most probably account for this observation. The successive eruption of molars with a spacing of about 6 years in modern humans would logically imply that first molars have the thickest enamel, since they are in functional occlusion for a longer period of time than all other molars. On the other hand, 198 G.A. MACHO AND M.E. BERNER second and, possibly, third maxillary molars may be subjected to stronger masticatory forces than M’s (Benfer and Edwards, 1991; Koolstra et al., 1988; Mansour and Reynick, 1975a, 1975b; Molnar and Ward, 1977; Ward and Molnar, 1980). Molnar and Ward (1977)reported that axial forces tended to be greatest in distal maxillary molars, although mesial cusps were always more loaded than distal ones within a tooth (Ward and Molnar, 1980). Benfer and Edwards (1991) suggested that the greater wear of second molars, as judged by crown height, could be explained by a change towards mastication of coarser foods in adults as well as by a shift of mastication to the posterior, less worn and, hence, more efficient molars. Although this seems to be a likely interpretation, another explanation comes to mind. During ontogeny the mandible experiences a major restructuring, especially after the eruption of first molars (Enlow, 1975a; Ricketts, 1975). The ramus of the mandible becomes increased in vertical dimensions, thereby accommodating the marked downward enlargement of the nasomaxillary complex (Enlow, 1975b).These developmental changes have an effect on the position of the temporo-mandibular joint with relation to the occlusal plane and, moreover, result in a marked shift of the position of the masseter-pterygoid complex with regard t o the dental arcade. Based on analogy with extant mammals one is led to infer that ontogenetic changes of the masticatory complex in humans lead to an efficient system adapted to prolonged, forceful grinding (DuBrul, 1974; Hannam and Wood, 1989; Hylander et al., 1987; Osborn, 1987; Smith, 19781, although this is only expressed to the full after eruption of second molars. The morphology of posterior molars indicates that they are well adapted to sustain enhanced and prolonged masticatory loads, an interpretation which is in keeping with previous findings on changes of cusp areas within complete upper tooth rows (Macho and Moggi Cecchi, 1992). Anthropologists have investigated the causal relationships between trophic adaptations and facial topographies, and they have also attempted to relate diet with enamel thickness. Yet the association between orofacial morphology and enamel thickness remains open to investigation. For instance, among hominid species Australopithecus (Paranthropus) robustus and A.(P.) boisei have relatively orthognathic faces with a retracted, albeit long, palate, a forward position of the anterior root of the zygoma, and molarized premolars when compared to A. africanus and Homo (see, for instance, Rak, 1983; Kimbel et al., 1984); “robust” australopithecine dentognathic morphology is generally thought t o be indicative of a trophic adaptation. On the basis of the present study one would predict that the antero-posterior gradient of enamel thickness in “robust”australopithecines would be different (less) than that inA. africanus and Homo, in spite of differences in absolute values. This would occur because more of the molar teeth are slung under the masseterpterygoid complex in A.(P.) robustus and A.(P.) boisei. Further work on enamel thickness employing Computed Tomography may shed light on this hypothesis of differential gradients of enamel thickness among PlioPleistocene hominids (Macho and Thackeray, in preparation). In conclusion, the present results as well as inspection of published data indicate that differences in distribution of enamel thickness between maxillary molars may be similar in all hominoids, although absolute values differ. This finding casts doubt on the usefulness of calculating “overall molar enamel thickness” for taxonomic purposes. Furthermore, the patterning of enamel distribution between maxillary molars implies that in the course of evolution enamel thickness has responded to prolonged masticatory loads and increased bite force magnitudes in posterior teeth as reported by biomechanical studies. Although there is a clear relationship between orofacial morphology and functional adaptation, more detailed analyses are needed to elucidate the relationship between enamel thickness and orofacial morphology. While enamel thickness per se may be indicative of dietary adaptations, the gradient of enamel distribution between teeth could provide information about biomechanical constraints on the masticatory apparatus. We contend that an understanding of the functional adaptation of enamel thickness within and between ENAMEL THICKNESS DISTRIBUTION BETWEEN MAXILLARY MOLARS hominid taxa will substantially contribute to our understanding of the evolutionary history and functional adaptations of Homo sapiens. ACKNOWLEDGMENTS We are grateful to J. Szilvassy for permission to section the material and to H. Kritscher for advice. G. Kurat and H. Schonmann allowed us to use the technical instruments in their care. Special thanks are due to G. Sverak for his invaluable help with the sections and to G. Franzke for his help with the figures. The constructive criticism and of M.C. Dean on an earlier draft of the manuscript is greatly appreciated. We thank C . Wood and two anonymous referees for their valuable comments. LITERATURE CITED Aiello LC, Montgomery C, and Dean C (1991) The natural history of deciduous tooth attrition in hominoids. J. Hum. Evol. 21:397412. Andrews P, and Martin L (1991) Hominoid dietary evoLond. B 334t199-209. lution. Phil. Trans. R. SOC. Anemone RL, Watts ES, and Swindler DR (1991)Dental development of known-age chimpanzees, Pun troglodytes (Primates, Pongidae). Am. J. Phys. Anthrop. 86t229-241. Benfer RA, and Edwards DS (1991) The principal axis method for measuring rate and amount of dental attrition: Estimating juvenile or adult tooth wear from unaged adult teeth. In MA Kelley and CS Larsen (eds.): Advances in Dental Anthropology. New York: Wiley-Liss, pp. 325-340. Beynon AD, and Dean MC (1988) Distinct dental development patterns in early fossil hominids. Nature 335t509-514 Beynon AD, and Wood BA (1986) Variations in enamel thickness and structure in East African hominids. Am. J. Phys. Anthrop. 70r177-193. Beynon AD, and Wood BA (1987) Patterns and rates of enamel growth in the molar teeth of early hominids. Nature 326:493-496. Beynon AD, Dean MC, and Reid DJ (1991) On thick and thin enamel in hominoids. Am. J. Phys. Anthrop. 86t295-309. Bilsborough A (1984) Multivariate analysis and cranial diversity in Plio-Pleistocene hominids. In GN van Vark and WW Howells (eds.):Multivariate Statistical Methods in Physical Anthropology. Dordrecht: D. Reidel Publishing Company, pp. 351-375. Butler PM (1939) Studies of the mammalian dentition: Differentiation of the postcanine dentition. Proc. Zool. SOC.London 109B;l-36. Conroy GC (1991) Enamel thickness in South African Australopithecines: Noninvasive evaluation by computed tomography. Palaeont. Afr. 28:53-59. 199 Corruccini RS, and Holt BM (1989) The dentinoenamel junction and the hypocone in primates. Hum. Evol. 4t253-262. Dahlberg AA (1945) The changing dentition of man. J. Am. Dent. Assoc. 32r676-690. Dean MC (1987) Growth layers and incremental markings in hard tissues. A review of the literature and some preliminary observations about enamel structure in Purunthropus boisei. J. Hum. Evol. 16t157-172. Dean MC, and Wood BA (1981) Developing pongid dentition and its use for ageing growth studies. Folia Primatol. 36r111-127. DuBrul EL (1974) Origin and evolution of the oral apparatus. Front. Oral Physiol. 1:l-30. Enlow DH (1975a) Rotations of the mandible during growth. In JA McNamara (ed.):Determinants ofMandibular Form and Growth. Monograph No. 4: Craniofacial Growth Series. University of Michigan: Centre for Human Growth and Development, pp. 65-76. Enlow DH (197513) Handbook of Facial Growth. Philadelphia: W.B. Saunders Company. Gantt DG (1983)The enamel of neogene hominoids: Structural and phyletic implications. In RL Ciochon and RS Conuccini (eds.): New Interpretations of Ape and Human Ancestry. New York: Plenum Press, pp. 249-298. Grine FE (1991) Computed tomography and the measurement of enamel thickness in extant hominoids: Implications for its palaeontological application. Palaeont. Afr. 28:61-69. Grine FE, and Martin LB (1988) Enamel thickness and development in Australopithecus and Puranthropus. In FE Grine (ed.): Evolutionary History of the “Robust” Australopithecines. New York: Aldine de Gruyter, pp. 3-42. Hannam AG, and Wood WW (1989) Relationships between the size and spatial morphology of human masseter and medial pterygoid muscles, the craniofacial skeleton, and jaw biomechanics. Am. J . Phys. Anthrop. 80t429-445. Hylander WL, Johnson KR, and Crompton AW (1987) Loading patterns and jaw movements during mastication in Mucaca fusciculuris: A bone-strain, electromyographic, and cineradiographic analysis. Am. J. Phys. Anthrop. 72t287-314. Kanazawa E, Sekikawa M, and Ozaki T (1983) Threedimensional measurements of the occlusal surface of upper first molars in a modern japanese population. Acta Anat. 116:90-96. Kanazawa E, Morris DH, Sekikawa M, and Ozaki T (1988) Three-dimensional measurement of the occlusal surface of the upper first molar in South African samples. J. Anthrop. SOC. Nippon 96:405415. Kay RF (1981) The nut-crackers: A new theory of the adaptations of the ramapithecinae. Am. J. Phys. Anthrop. 55t141-151. Kimbel WH, White TD, and Johanson DC (1984) Cranial morphology ofAustralopithecus afarensis: A comparative study based on a composite reconstruction of the adult skull. Am. J. Phys. Anthrop. 64t337-388. Koolstra JH, van Eijden TMGJ, Weijs WA, and Naeije M (1988) A three-dimensional mathematical model of the human masticatory system predicting maximum possible bite forces. J. Biomech. 21 r563-576. 200 G.A. MACHO AND M.E. BERNER Korenhof CAW (1960) Morphological Aspects of the Human Upper Molar. Uitgeversmaatschappij Neerlandia: Utrecht. Korenhof CAW (1982) Evolutionary trends of the inner enamel anatomy of deciduous molars from Sangiran (Java, Indonesia). In G Kurten (ed.): Teeth: Form, Function and Evolution. New York: Columbia University Press, pp. 350-365. Ladenbauer-Ore1 H. (1971). Fundberichte aus Osterreich. Band 7. Ferdinand Berger & Sohne OHG: Horn. Macho GA, and Berner ME (in press) Enamel thickness and the helicoidal occlusal plane. Am. J . Phys. Anthrop. Macho GA, and Moggi Cecchi J (1992)Reduction of maxillary molars in Homo sapiens sapiens: A different perspective. Am. J. Phys. Anthrop. 87t151-159. Macho GA, and Thackeray JF (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. 89t133-143. Mansour RM, and Reynick R J (1975a) In vivo occlusal forces and moments. I. Forces measured in terminal hinge position and associated moments. J. Dent. Res. 54t114-120. Mansour RM, and Reynick RJ (197513) In vivo occlusal forces and moments. 11. Mathematical analysis and recommendations for instrumentation specifications. J. Dent. Res. 54t121-124. Martin LB (1983)The Relationships of the Late Miocene Hominoidea. Ph.D. thesis, University of London. Martin LB (1985) Significance of enamel thickness in hominoid evolution. Nature 314t260-263. Mayhall JT, and Kanazawa E (1989)Three-dimensional analysis of the maxillary first molar crowns of Canadian Inuit. Ani. J . Phys. Anthrop. 78:73-78. Molnar S, and Gantt DG (1977) Functional implications of primate enamel thickness. Am. J. Phys. Anthrop. 46t447454. Molnar S,and Ward SC (1977) On the hominid masticatory complex: Biomechanical and evolutionary perspectives. J . Hum. Evol. 6t557-568. Osborn JW (1987)Relationship between the mandibular condyle and the occlusal plane during hominid evolution: Some of its effects on jaw mechanics. Am. J. Phys. Anthrop. 73r193-207. Rak Y (1983) The Australopithecine Face. New York: Academic Press. Ricketts RM (1975) Mechanisms of mandibular growth: A series of inquiries on the growth of the mandible. In JA McNamara (ed.): Determinants of Mandibular Form and Growth. Monograph No. 4: Craniofacial Growth Series. University of Michigan: Centre for Human Growth and Development, pp. 77-100. Sakai T, and Hanamura H (1971)A morphological study of enamel-dentin border on the Japanese dentition. V. Maxillary molar. J. Anthrop. SOC.Nippon 79t297-322. Sakai T, and Hanamura H (1973)A morphological study of enamel-dentin border on the Japanese dentition. VI. Mandibular molar. J. Anthrop. SOC.Nippon 81: 2545. Schwartz J H (1984) The evolutionary relationships of man and orang-utans. Nature 308:501-504. Schwartz J H (1990) Lufengpithecus and its potential relationship to an orang-utan clade. J. Hum. Evol. 19.59 1-605. Scott GR (1980)Population variation of Carabelli’s trait. Hum. Biol. 52t63-78. Shellis RP (1984) Variations in growth of the enamel crown in human teeth and a possible relationship between growth and enamel structure. Archs. Oral Biol. 29t697-705. Shellis RP, and Hiiemae KM (1986) Distribution of enamel on the incisors of Old World monkeys. Am. J . Phys. Anthrop. 71:103-113. Simons EL (1976) The nature of the transition in the dental mechanism from pongids to hominids. J. Hum. Evol. 5:511-528. Smith R J (1978) Mandibular biomechanics and temporomandibular joint function in primates. Am. J. Phys. Anthrop. 49t341-350. Spoor CF, Zonneveld FW, and Macho GA (in press) Linear measurements of cortical bone and dental enamel by computed tomography: Applications and problems. Am. J. Phys. Anthrop. Statistisches Handbuch fur the Republik Osterreich, 1971. Osterreichisches Statistisches Zentralamt: Wien. Stloukal M, and Vyhnanek (1976) Slovane z velkomoravskych Mikulcic. Academia: Praha. Ward SC, and Molnar S (1980) Experimental stress analysis of topographic diversity in early hominid gnathic morphology. Am. J. Phys. Anthrop. 53t38S395.